Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 DOI 10.1007/s00167-012-1933-6

Posters Spine P13-96 Scoliosis in immature athletes M. Massada1, A. Pereira2, R. Sousa3, L. Massada4 1 Hospital de Santo Anto´nio, Servic¸o de Ortopedia, Porto, Portugal, 2 Hospital de Santo Anto´nio, Department of Orthopedics, Porto, Portugal, 3Hospital de Santo Anto´nio, Porto, Portugal, 4Faculdade de Desporto, Universidade do Porto, Traumatologia do Desporto, Porto, Portugal Objectives: To test the hypothesis that asymmetricloading of immature spines in young athletes initiates scoliosis. Overview of literature: The etiology of scoliosis remains unknown in most cases despite extensive research. Scoliosis in athletes has been reported in the literature, but its causative factors havenot been investigated. To date, no existing classifications of spinal deformities suggest a separate scoliotic category etiologically related to sports. There have been some rare reports on scoliosis associated with ballet and some other sports such as tennis and javelin throwing, but without etiologic implications. Methods: We compared the incidence, type and magnitude of scoliotic curves between volleyball, basketball, handball and soccer athletes. We included a group of 1,859 adolescent athletes (boys, mean age: 13.4 years). All completed a questionnaire and data including personal and somatometric characteristics, handedness and type, duration and character of daily-performed physical activities were recorded, along with clinical examination. We analyzed the data and compared between sports to determine differences. Results: Soccer and basketball players had a statistically significant increase in the prevalence of scoliotic spinal curves. A moderate percentage (13.3 and 10.a%, respectively) of asymmetry was present on the Adams forward bending test, as compared to the other modalities (volleyball: 10.3%, handball: 8.6%). The curves were either thoracic or thoracolumbar (4.3–4.4%). Playing hand/ foot dominance was related to the curve direction. Cobb angle had no significant correlation with the duration of playing. When considering all modalities we observed a significant increase in the prevalence of scoliotic spinal curves with age (p \ 0.05). Conclusions: There is a increase in the prevalence of mild scoliosis in the analysed modalities. A statistically significant increase in the prevalence among basketball, soccer, volleyball and handball players may point to repetitive asymmetric muscular contraction in the initiation of scoliotic curves.

P13-620 The prevalence of low back pain among former elite cross-country skiers, rowers, orienteerers and nonathletes: a 10-year cohort study I.S. Foss1, I. Holme2, R. Bahr1 1 Oslo Sports Trauma Research Center, Norwegian School of Sports Sciences, Oslo, Norway, 2Oslo Sports Trauma Research Center, Oslo, Norway Objectives: To compare the prevalence of symptoms of low back pain (LBP) among former endurance athletes with different loading characteristics on the lumbar region: cross-country skiing (flexion loading), rowing (extension loading) and orienteering (no specific loading), as well as a non-athletic control group.

Cross-sectional studies have suggested that the prevalence of LBP may be high among endurance athletes with repetitive back loading, but there are no large, prospective cohort studies addressing this issue. Methods: This is a prospective cohort study based on a cross-sectional survey of cross-country skiers, rowers, orienteerers and nonathletic controls from 2000. A self-reported questionnaire on LBP adapted for sports based on standardized Nordic questionnaires for musculoskeletal symptoms was completed by the participants. Responders were 173 rowers, 209 orienteerers, 242 cross-country skiers and 116 control subjects (88% of the original cohort). Results: There were no group differences with regard to the two main outcomes: reported LBP the previous 12 months (P = 0.66) and frequent LBP the past year ([30 days with LBP) (P = 0.14). More rowers reported frequent LBP the past year than orienteerers (OR = 2.32; P = 0.044). Occupational changes were reported more often by rowers compared to skiers and orienteerers. A previous episode with LBP was associated with LBP later in life (P \ 0.001). Conclusions: LBP was not more common among skiers and rowers compared to nonathletes or orienteerers. The results indicate that prolonged and repetitive flexion or extension loading in endurance sports does not lead to more LBP.

P13-1163 Transversus abdominis performance with rehabilitative ultrasound imaging feedback I.B. Almeida1, R. Matias1 1 Health College of Polytechnic Institute of Setu´bal, Physiotherapy Department, Setu´bal, Portugal Objectives: To assess the effectiveness of ultrasound as a means of real time biofeedback, as well as measure the contribution of different types of biofeedack on transversus abdominis (TrA) and internal oblique (IO) performance, in healthy subjects, through changes in thickness and lateral slide of TrA anterior fascia during abdominal hollowing exercise (AHE). Methods: Seventy-five healthy subjects were divided randomly into 3 groups that received: group 1 control group (CG), no feedback; group 2 verbal and palpatory feedback (VPF), and group 3, realtime ultrasound feedback (RUF). The TrA and IO performance of each subject was assessed twice (before and after receiving feedback) when performing the AHE in a supine hook-lying position. Analysis of variance and T test were used for the independent and paired samples, respectively, to determine significant changes in the performance of TrA and IO, based on intra and inter group analysis. Results: Group 1 (CG) had no differences between moments; group 2 (VPF) had significant differences concerning TrA thickness (p = 0.000) with up to 0.84 mm thickness difference; group 3 (RUF) had significant differences concerning TrA thickness (p = 0.000) with up to 1.94 mm difference. The ability to perform the AHE differed only among group 3 and in group 1 (p = 0.056) only for changes in thickness of TrA muscle. No differences were found among groups. Neither for the lateral slide of TrA anterior fascia, nor for the internal oblique thickness. Conclusions: From the results of this study we conclude that real-time ultrasound feedback, when used alone during an AHE, can give rise to larger increase in TrA thickness when compared to verbal and palpatory feedback. The use of real time ultrasound was shown to be

123

S102 effective as a feedback tool to facilitate the performance of the AHE in a supine hook-lying position in healthy subjects. Discussion of the clinical implications and utility of different types of biofeedback strategies will be presented.

P13-1344 The impact of working length of occipital screws on the pullout resistance of the screw-plate construct. A biomechanical cadaver study A. von Keudell1, G. Korn2, M. Mayer2, H. Resch2, J. Zenner3, H. Koller3 1 Brigham and Women’s Hospital/Harvard Medical School, Boston, United States, 2Paracelsus Medical University Salzburg, Traumatology and Sports Injuries, Salzburg, Austria, 3Deutsches Skoliose Zentrum, Bad Wildungen, Germany Objectives: Modern cervical screw-rod constructs enable rigid occipital screw-plate fixation using several screws. Studies observed resistance to pullout (POS) increased with occipital screw length. Insertion of longer screws place risks concerning the cerebellum and venous sinus, while use of shorter screws is clinically safer and might be equivocal biomechanically. But there are no biomechanical data whether increased osseous engagement of several screws significantly increases a plate’s resistance to pullout. Methods: This is a biomechanical study on 12 human occipital bones. Two groups of each 6 BMD matched-pairs were reconstructed and occipital thickness below the level of external occipital protuberance (EOP) measured. Bones were potted in PMMA with a void prepared anterior to the inner cortical bone table. In Group-1, occipital plates (synapse/synthes) were fixed with a clinically safe set of three 4.5 mm-diameter screws (unicortical fixation). Screw length ranged 10–16 mm at the cephalad hole, 8–12 mm at the middle and caudal holes. In Group-2, maximum screw lengths (8–16 mm, bicortical fixation) at the level of the EOP were selected. All screws were evaluated clinically and using fluoroscopy prior to pullout testing. Specimens were mounted in an electromechanical testing machine (ETM) and plates connected to the ETM with screw axis coaxial with the pullout force. Pullout-loading was conducted at a rate of 2 mm/min. Load–displacement data were taken continuously, peak load-to-failure was measured in Newton and reported as POS. Postoperatively all bones were cut sagittaly along the screws to rule out radiographically invisible breaches of the occipital bone or screw fixation into PMMA, accordingly one pair of plates tested was excluded. Results: BMD for specimens in Group-1 was 206.5 ± 26.7, for Group-2 224.1 ± 673 mg Ca-HA/ml. Osseous purchase, defined as [Summed screw lengths-plate thickness (mm) 9 3(screws)] was 20.0 mm for Group-1 and 30.5 ± 2.5 mm for Group-2. Summed screw length was 26.0 mm for Group-1, 36.6 ± 2.5 mm for Group-2, the differences yielded significance (p \ 0.001). The POS in Group-1 was 524.9 ± 260.8 N, 1,867.8 ± 594.4 N in Group-2, difference was significant (p = 0.001). Statistics revealed a strong correlation between summed screw lengths, osseous purchase and POS (p \ 0.001, r = 0.9). Conclusions: Increasing screw length in the merely cortical occipital bone increased POS of a modern occipital screw-plate construct. Differences were striking resembled by a threefold increased POS with a strategy using longest screws where possible as compared to a ‘shorter-and-safe’ strategy. Although clinical studies showed that loosening of occipital plates is rare, our study offers biomechanical data for surgeons intending screw placement at the occiput.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Shoulder I

P14-110 Autoclaved frozen femoral head allograft bone grafting for large bone defects: a review of its cost-effectiveness C. Seng1, P.L. Chin2, S.-L. Chia1, N.N. Lo1, S.J. Yeo1 1 Singapore General Hospital, Orthopedic Surgery, Singapore, Singapore, 2Singapore General Hospital, Singapore, Singapore Objectives: Femoral head allograft is a viable option for bone grafting. With the potential shortage and high cost of processed allograft available in the market, we saw the need to explore alternative cost effective allografts for usage in bone grafting. Methods: We prospectively followed up 11 patients who underwent femoral head allograft bone grafting from July 2008 to October 2010. 2 patients who were lost to follow up were excluded from the study. The femoral head allograft utilized were obtained from patients with neck of femur fractures. The allograft was processed according to our protocol: The bone was first sterilized at 134C through the process of autoclaving. Thereafter, the femoral head bone was sealed in a container containing normal saline with crystalline penicillin and streptomycin. The allograft was stored for a maximum of 6 months in the Sanyo Ultra-low temperature chiller at -80, after which it was discarded if not used. The allograft was thawed at 4C before use and discarded if not utilised after being thawed. Results: There were 4 patients with large Hill–Sachs lesion (greater than 40% humeral head loss) who underwent bone grafting. 2 patients underwent revision total hip replacement with bone grafting of the acetabulum component. Another 2 patients underwent revision total knee replacement with bone grafting of the femoral component. 1 patient underwent right tibia bone grafting after non-union of the primary fracture. All patients except one achieved bone integration and fusion based on radiological imaging. Time taken for fusion ranges from 4 to 9 months. The patient who underwent right tibia bone grafting had non-union post-operatively and eventually underwent re-bone grafting with autogenous iliac crest bone grafting. The rest of the patients had neither infection nor repeat surgery. The cost of our femoral head allograft bone grafting per patient is 33 USD. Conclusions: Most of the conventional allograft available in the market underwent gamma irradiation for sterilization, however this is a costly procedure. Our method of autoclaved deep frozen femoral head allograft bone grafting for large bone defects is viable and cost effective.

P14-130 Arthroscopically assisted Latarjet–Lafosse procedure: how to make it easier in lateral decubitus position through 3 standard portals O. Milenin1, A. Belopolsky1 1 National Medical Surgical Centre, Traumatology and Orthopaedics, Moscow, Russian Federation Objectives: Anteroinferior instability appears to be a significant problem nowadays. There is a variety of techniques, both open and arthroscopic used to deal with it. Latarjet procedure is considered to be effective and reproducible method and has good long-term results. Arthroscopic modifications of Latarjet procedure are becoming increasingly popular, but are rather difficult to perform and have a steep learning curve.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: We describe a new safe and reliable arthroscopically assisted technique. It combines the advantages of the Lafosse (2007) and Boileau (2010) arthroscopic methods and presents the possibility to treat the anteroinferior instability in the lateral decubitus position of the patient through 3 standard portals and 3 cm graft harvest incision. Results: Our procedure was evaluated by CT and X-ray and show optimal positioning of the coracoid graft and the screws. Conclusions: Our surgical technique was created to simplify Arthroscopic Latarjet-Lafosse procedure. This aim was achieved by using 3 standard portals and a 3 cm incision for a graft harvest. We put our patient in a lateral decubitus position, which is widely used within surgeons all over the world. This allows these surgeons to feel themselves comfortable by performing Latarjet–Lafosse procedure. We may reduce damage of m. subscapularis and neuro-vascular structures using palpatory and blunt method. Keywords: Shoulder, Anterior instability, Arthroscopy, Latarjet, Bankart, Lateral decubitus position, Bony defect, Bristow, Reccurent

P14-238 Arthroscopic rotator cuff repair: single to double row M. Fiodorovas1 1 Klaipeda University Hospital, Sports Trauma Department, Klaipeda, Lithuania Objectives: The purpose of this study was to evaluate the results of an arthroscopic rotator cuff repair with single-row and double-row techniques. Methods: Sixty patients with a full-thickness rotator cuff tear underwent arthroscopic repair with suture anchors. They were divided into 2 groups according to repair technique: single row—30 patients and double row—30 patients. Mostly elderly patients with atraumatic rotator cuff lesion. Age 45–67 (mean 56). All patients were operated in 2008 year. Follow-up period 1 year. Examination using Constant, UCLA shoulder scoring systems for strength, pain and function before OP and year after. Results: According UCLA 90% of the patients had good and excellent post-operative scores with 29 excellent (49%), 25 good (41%), 4 fair (6%), 2 poor (4%). The average Constant score improved from a preoperative rating of 47.8 to a postoperative rating of 85.9. The average increase in the Constant score after the operation was 29.1 points. Both the mean Constant score and UCLA score improved following surgical intervention. Differences between single or double row were not significant. Conclusions: Double-row suture anchor fixation will take longer because twice as many anchors need to be placed. Double-row fixation will be more expensive than single-row. And at short-term follow-up, arthroscopic rotator cuff repair with the double-row technique showed no significant difference in clinical outcome compared with single-row repair.

P14-308 Arthroscopic bone block augmentation of the anterior glenoid without violating the subscapularis tendon: a cadaveric study O. Verborgt1, G. Van den Bogaert2, P. Debeer3 1 University of Antwerp, AZ Monica, Orthopaedic Surgery, Deurne, Belgium, 2AZ St Elisabeth, Herentals, Belgium, 3UZ Pellenberg, Leuven, Belgium, Leuven, Belgium Objectives: To evaluate arthroscopic augmentation of the anterior glenoid using a free graft without violation of the subscapularis tendon. Methods: In 7 cadaveric shoulders arthroscopic augmentation of the anterior glenoid was performed. A preshaped, augmentation block of

S103 2.0 9 1.0 9 1.0 cm with an eccentric plug was arthroscopically introduced and fixated with 1 central screw through the rotator interval without desinsertion or split of the subscapularis tendon. Postoperatively a multislice-CT scan was done to analyze the orientation and position of the screw. Macroscopic dissection was then performed to assess portal placement, damage to the axillary and musculocutaneous nerve, the conjoined tendon and the subscapularis tendon. The position of the augmentation block was assessed in the vertical and horizontal plane. Results: Postoperative CT scans showed no intra-articular perforation of the screw. The mean inclination angle was 21.2 (range, 9–48). Post-operative dissection showed no damage to the axillary or musculocutaneous nerve in any specimen. The conjoined tendon was damaged at the lateral border in 3 specimens. The upper border of the subscapularis tendon was frayed in 5 cases, but no tears were noted. In the vertical plane, the augmentation block was correctly (subequatorial) positioned in 5 cases, 2 blocks were at the level of the equator. In the horizontal plane, the augmentation block position was flush with the articular surface in 5 cases and too medial in 2 cases (\5 mm). Conclusions: Arthroscopic augmentation of the anterior glenoid without subscapularis takedown or split would make the procedure technically easier and safer and it could allow easier post-operative recovery. This study showed that it is technically possible to perform a bone block procedure arthroscopically through the rotator interval without compromising the position and fixation of the block. More biomechanical studies are needed to investigate the effect of this technique on post-operative stability of the shoulder.

P14-403 Clinical outcomes of arthroscopic repair of massive cuff tears. Comparative study of immobile versus mobile P. Randelli1, L. Zottarelli2, C. Fossati2, P. Arrigoni2, V. Ragone2 1 Universita` degli Studi di Milano,Policlinico S. Donato IRCCS, Dipartimento di Scienze Medico Chirurgiche, Milan, Italy, 2 Universita` degli Studi di Milano,Policlinico S. Donato IRCCS, Dipartimento di Scienze Medico-Chirurgiche, San Donato Mil., Italy Objectives: The purpose of this study was to compare clinical outcomes of retracted massive cuff tears treated using an interval slide releases technique if immobile versus cuff repair without interval slide if mobile. Methods: 25 patients underwent arthroscopic repair for massive rotator cuff tears were divided in two groups. In the group 1, a single or double interval slide release was performed to achieve an adequate tendon mobilization. In the group 2 massive cuff tears were arthroscopically repaired without this additional release. Patients were retrospectively evaluated with validated outcomes scores: Constant Score (CS), pain score (VAS) and Single Assessment Numeric Evaluation (SANE). Results: The two groups were comparable for age (group 1: 63 ± 6; group 2: 69 ± 7) gender (% male, group 1: 61% group 2: 50%) and operated dominant side (group 1: 92.3%; group 2: 91.6%) (p [ 0.05). The mean follow-up for group 1 and 2 were 31 and 28 months respectively (p = 0.4). The two groups showed no significant difference in SANE and VAS evaluation (group 1: SANE 77%, VAS 1.3; group 2: SANE 88%, VAS 1.6). No significant difference was found between the two groups for the CS (group 1: 66.5 ± 11; group 2: 75 ± 14; p = 0.1). Subjective CS for group 1 and 2 were 31 ± 5 and 30.8 ± 7 respectively (p = 0.9). A statistical significant difference was found for objective CS in the control group (group 1: 35.5 ± 7; group 2: 44 ± 8; p = 0.009).

123

S104 Conclusions: Subjectively clinical outcomes of arthroscopic repair in rotator massive cuff tears, immobile or mobile are comparable and satisfactory. Although objectively immobile cuff tears show inferior results despite interval slide technique.

P14-406 Biological approach to Bankart repair: a prospective study P. Randelli1, P. Spennacchio2, A. Aliprandi3, V. Ragone4, P. Cabitza5 1 Universita` degli Studi di Milano, Policlinico S. Donato IRCC, Dipartimento di Scienze Medico Chirurgiche, Milan, Italy, 2 Universita` degli Studi di Milano, Policlinico S. Donato IRCC, Dipartimento di Scienze Medico-Chirurgiche, San Donato Milanese, Italy, 3Universita` degli Studi di Milano, Policlinico S. Donato IRCCS, Department of Radiology, Milan, Italy, 4Universita` degli Studi di Milano, Policlinico S. Donato IRCCS, Dipartimento di Scienze Medico-Chirurgiche, San Donato Milanese, Italy, 5Universita´ degli Studi di Milano, Policlinico S. Donato IRCCS, San Donato Milanese, Italy Objectives: New bio-absorbable anchors have been created to promote bone ingrowth and to reduce inflammatory reaction at the implant site. The aim of this study is to demonstrate the biological efficacy of a new Tri-Calcium-Phosphate bio-absorbable osteoconductive anchor (Lupine Byocryl Rapide, LBR) in arthroscopic shoulder Bankart repair. Methods: We prospectively evaluated 13 patients treated by arthroscopic Bankart repair from July 2008 to September 2010. Our concept of Biological Approach to Bankart repair included the evaluation, by MRI (1.5-T unit), of: bone ingrowth, absence of cistic reactions and soft tissue healing. The imaging analysis was done by a single radiologist, blinded to the procedure with the same MRI machine. A new radiological protocol was used to study the bone ingrowth with MRI. The expected bone ingrowth following anchor absorption was evaluated comparing signals in the implant site with surrounding cancellous glenoid bone, in different weighted sequences. Results: Average follow-up was 22 months. The patients were 1 females and 12 males with an average age of 33 (±10) years. A total of 28 LBR have been implanted. We observed 1 case of recurrency (7.6%). All the 13 patients (28 implanted anchors) were available for radiological evaluation at follow-up. No clinical evidence of foreign body reaction has been observed. No case of bony cystic lesion and/or fluid collection has been observed. All the patients showed an MRI signals consistent with new bone apposition. In 6 cases (21.4%) neither the anchor nor the bone tunnel were recognizable at a medium radiological follow up of 23.7 months. Conclusions: A comprehensive biological approach to shoulder Bankart repair can be obtained using new tri-calcium-phosphate bioabsorbable osteoconductive anchors. MRI evaluation showed a complete biocompatibility, no adverse reactions and signals consistent with bone tissue formation at the implant site in all cases.

P14-501 A preliminary prospective study of the use of ultrasound in the arthroscopic debridement of calcific rotator cuff tendinitis M. Sabeti1, M. Schmidt2, C. Schueller-Weidekamm3, E. Nemecek4 1 Vienna Medical School, Orthopaedics, Vienna, Austria, 2Vienna Medical School, Vienna, Austria, 3Vienna Medical School, Radiology, Vienna, Austria, 4Vienna Medical School, Traumatology, Vienna, Austria Objectives: Arthroscopic calcific deposit debridement is established in cases of calcifying tendinitis that are refractory to conservative treatment. The localization of the deposit can be demanding and time

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 consuming using frequently ionizing radiation. Intra-operative ultrasound was recently promoted facilitating the deposit localisation and reducing radiation dose. Methods: In this prospective, controlled, clinical observer-blinded trial 20 patients with calcific tendinitis were randomized into two groups and operated following a standardized protocol. In group I, the deposit was localized conventionally. In group II, the deposit was localized using intra-operative ultrasound. During operation, the needle punctures to detect the deposit and operation times were noted. Clinical and radiological examinations were made 2 and 6 weeks and 9 months after surgery. Results: In group II, the needle punctures to detect the deposit were significantly lower than in group I (p \ 0.0001). Operation time to localize the deposit was also significantly less in group II (p \ 0.033). In both groups, patients improved significantly with increased shoulder function (p \ 0.0001) and decreased pain (p \ 0.0001) 2 weeks and 9 months (p \ 0.001) after surgery. The difference between the groups was not significant. The size of the deposits did no significantly affect the localization. Excellent radiological findings were obtained in both groups after 9 months. Conclusions: Intra-operative US significantly facilitates the detection of calcific deposits during the arthroscopic debridement by speeding up surgery and reducing number of needle punctures. We highly recommend this localization modality to reduce or even avoid the use of ionizing radiation of fluoroscopes.

P14-525 Clinical outcomes of arthroscopic subscapularis repair N.N. Verma1, E. Lin1, G. Van Thiel1, G.P. Nicholson1, B.J. Cole1, A.A. Romeo1 1 Rush University Medical Center, Department of Orthopaedic Surgery, Division of Sports Medicine, Chicago, United States Objectives: Isolated rupture of the subscapularis is a relatively rare injury but a functionally debilitating problem. Thus, no paradigm for treatment has been established. Traditionally, the tear has been managed with an open procedure. However, arthroscopic techniques have recently been developed. The purpose of this study is to evaluate the outcomes of isolated subscapularis tears repaired arthroscopically. Methods: A retrospective review of consecutive patients who had undergone arthroscopic subscapularis repair at least 1 year prior to the study date were included. Follow-up examinations included range of motion and strength testing as well as the completion of the SANE, VAS, SST, ASES, and Constant scores. Results: 39 consecutive patients were identified and 29 (74%) were available for follow-up. Mean age was 47.1 ± 8.3 years (range, 29.0–63.8 years). The average follow up was 35.0 months ± 18.2 (range, 12.6–73.4 months). There were 25 males and 4 females, and the dominant arm was involved in 21 cases (72%). Nineteen of the patients reported regular participation in sports ranging from bodybuilding to golf. Acute tears were present in 16 patients and the remaining 13 had chronic tears. Concomitant procedures included subacromial decompression, biceps tenodesis (18), distal clavicle excision (6), and SLAP repair (3). The mean pre- and post-operative scores were, respectively, SANE 55.7 ± 27.4 and 85.6 ± 27.7; ASES 61.5 ± 16.4 and 82.3 ± 22.2; SST 7.4 ± 2.9 and 10.1 ± 3.0; VAS 3.8 ± 2 and 2.0 ± 2.7. Post-operative UCLA and Constant scores were also recorded; 25.4 ± 5.4 and 79.7 ± 21.8. All post-operative scores were significantly improved from the pre-operative values (p \ .05). The average external rotation at final follow-up was 65. Six patients (20.9%) were considered failures: 1 had a revision, 3 had an ASES score lower than 50, and 2 additional patients reported a poor/fair outcome. 79% of the patients stated that they were satisfied with the results and would have the surgery again. Conclusions: The results of this study indicate that arthroscopic subscapularis repair provides pain relief and restores shoulder

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 function in a relatively young patient population at an average followup of 3 years. The success rate was 79% in an active patient population. Longer-term studies are required to determine if similar results are maintained over time.

P14-669 Can we improve the reliability of the Constant–Murley score? D. Blonna1, A. Tellini1, D.E. Bonasia2, R. Rossi3, S. Michele4, F. Castoldi1 1 Mauriziano Hospital, University of Torino, Department of Orthopaedics and Traumatology, Torino, Italy, 2University of Torino, CTO Maria Adelaide, Torino, Italy, 3University of Torino, Mauriziano Umberto I, Torino, Italy, 4Mauriziano Hospital, University of Torino, Torino, Italy Objectives: The Constant–Murley score (CMS) is currently one of the most used scales for shoulder dysfunctions. There is however a general consensus that the CMS could be improved by upgrading the expertise of the observer, enhancing the standardization of the items and correcting the score using either a reference population or a contralateral unaffected side. The aim of this study is to determine if these changes are really effective. Methods: Two consecutive series of 55 patients, with shoulder dysfunction, were enrolled in a test–retest study and examined by two orthopaedic surgeons with different levels of expertise in using the CMS. The following scores were measured: Constant–Murley score in its original description (CMS), Individual Relative-CMS, RelativeCMS and Standardized-CMS. For each variable the intraobserver and interobserver reliability was calculated. Results: The less experienced observer has the worst intraobserver reliability using the CMS. The systematic error between examinations was 4 points (95% upper limit of agreement: 22 points). For the expert, the systematic error was 2.4 points (95% upper limit of agreement: 16 points). As a result of introducing the StandardizedCMS the intraobserver reliability improved significantly for both observers (systematic error: 0.4 points). The correction against the contralateral unaffected side and the reference population determined a worsening of reliability in both the observers. Interobserver reliability showed an improvement similar to that of intraobserver reliability. The systematic error between observers was 4 points (95% upper limit of agreement: 24 points) using the CMS and improved to 1 point (95% upper limit of agreement: 12 points) adopting the Standardized-CMS. Conclusions: This study showed that the standardization of the items significantly improved both the intraobserver and interobserver reliability of the Constant-score. The level of expertise of the observer has less of an effect on reliability when the score is applied with a higher level of standardization. The Individual Relative-CMS and the Relative-CMS should be used only after a proper sample size analysis.

P14-836 Arthroscopic Bankart reconstruction improves EQ-5D index and reduces the bone mineral areal mass in the calcanei L. Ejerhed1, A. Elmlund2, J. Kartus3 1 NU-Hospital Organization, Trollha¨ttan/Uddevalla, Trollha¨ttan, Sweden, 2Danderyds Sjukhus, Department of Orthopaedics, Stockholm, Sweden, 3NU-Hospital Organization, Orthopaedics, Trollha¨ttan, Sweden Objectives: Arthroscopic Bankart reconstruction and its significance for quality of life, the influence on bone areal mass (BMA), and activity level have previously not been thoroughly investigated. Methods: Patients with posttraumatic recurrent anterior shoulder instability scheduled for arthroscopic Bankart reconstruction were

S105 prospectively included in the study. The BMA was measured in both calcanei using the dual energy X-ray absorptiometry (DXA) technique. EQ-5D was used to estimate quality of life and the activity was measured using Tegner activity level and. The patients were assessed before surgery and after six, 18 and 36 months. Results: 39 patients were included in the study and 22 patients (7 females and 15 males), median age 40 years; have been followed for 36 months. The BMA in the calcanei decreased in female and male patients at 18 months with 3.5–4.2% (p = 0.01) and 4–7% respectively (p = 0.02). TheEQ-5D index before surgery was 0.65 and after 36 months 0.87 (p = 0.002). The Tegner activity level was 5 (1–9) before the injury, 3 (0–9) preoperatively and 3.5 after 36 months (n.s. preop vs. 36 months). Conclusions: Arthroscopic Bankart reconstruction rendered a better quality of life index. The patients did not increase their activity level according to Tegner. Both female and male patients had a decrease in BMA of between 3.5 and 7% 18 months after surgery, which was more than the expected age related decrease. The surgical trauma induced the BMA decrease.

P14-907 Augmentation of suture anchor fixation with polymethylmethacrylate bone cement H.-S. Seo1, S.-C. Lee2 1 Cartilage Repair Center affiliated with Himchan Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea, 2Himchan Hospital, Seoul, Republic of Korea Objectives: The objective of this study is to evaluate outcomes of rotator cuff repair with augmentation of suture anchor fixation using polymethylmethacrylate (PMMA) bone cement. Methods: We enrolled 15 consecutive patients (15 shoulders) who underwent a mini-open rotator cuff repair with PMMA augmentation of suture anchor fixation due to osteoporosis or bone cyst in the greater tuberosity. With a minimum duration of follow-up of 2 years, they were evaluated clinically using the American Shoulder and Elbow Surgeons (ASES), Constant, and University of California, Los Angeles (UCLA) score. In addition, ultrasonography (USG) was used to evaluate the structural integrity of the repaired tendon at a mean of 12.6 months (range, 12–16 months). Results: Of the patients, 12 were available for follow-up evaluation. The mean age at the time of surgery was 72.6 years (range, 64.3–81.8 years), and the mean follow-up period was 28 months (range, 25–33 months). The American Shoulder and Elbow Surgeons score improved from 48.4 preoperatively to 88.6 at final follow-up (P \ 0.01). The Constant score increased from 44.1 to 85.7 (P \ 0.01), and the University of California, Los Angeles (UCLA) score also improved from 14.3 to 28.4 (P \ 0.01). Postoperative ultrasonography revealed that retear rate was 25.0% (3 of 12). Conclusions: When performing a rotator cuff repair in the setting of osteoporotic humeral head, PMMA bone cement can be used to augment suture anchor fixation, reducing the risk of anchor pull-out failure, which seemed to contribute to functional improvement after rotator cuff repair.

P14-923 Arthroscopic single or double row repair of isolated full thickness SSP tear: prospective results after 24 months R. Heikenfeld1, R. Listringhaus1, G. Godolias1 1 St. Anna Hospital Herne, Center for Orthopaedics and Traumatology, Herne, Germany Objectives: The purpose of this study was to evaluate the results after arthroscopic reconstruction of isolated SSP lesions. Does double row repair in smaller lesions lead to better results and a lower retear rate?

123

S106 Methods: 80 patients with an isolated full thickness SSP tear were divided into 2 groups. Group 1 (27 men, 13 Frauen, average age 57 y) was repaired using a single row technique with 2 anchors (Mitek Fastin) with arthroscopic Mason-Allen stiches. Group 2 was repaired using a double row technique using 1 medial anchor (Mitek Fastin) and 2 lateral anchors (Versalok). PreOP an ultrasound and MRI was obtained as well as standard X-rays (a.p., outlet-view, transaxial). Included to this study were only patients matching the following criteria: intraoperative cartilage lesions B Outerbridge Grade2, fatty degeneration B Goutallier Grade 2, ap extent of the tear \2.5 cm. Prospective follow up after 6, 12 and 24 months using UCLA and Constant Score as well as MRI at last follow up. Results: 37 patients in group 1 and 36 patients in group 2 were completely evaluated. Both groups showed improvement in Constant Score from 49.3 to 89.4 in group 1 (single row) compared to 47.6 and 90.7 in group 2 (double row). MRI at last follow up showed 4 retears in the single row group and 3 retears in the double row group. 1 shoulder in the double row group needed revision due to a loose anchor. Conclusions: We could not find any significant difference between clinical results and retear rate using a single or double row suture anchor configuration in isolated full thickness SSP tears.

P14-938 Anatomic tunnel positioning in AC joint repair using tendon grafts results in increased stability of graft fixation K. Beitzel1, L.E. Geaney1, J. Apostolakos1, M.P. Cote1, R. Arciero1, A.D. Mazzocca1 1 University of Connecticut Health Center, Department of Orthopedic Surgery, Farmington, United States Objectives: Coracoclavicular (CC) ligament reconstruction for AC Joint stabilization relies on clavicular fixation for successful results. One reason for failed reconstruction may be loosening of the graft secondary to poor fixation. The purpose of this study was to determine the bone density at specific bone tunnel locations in the clavicle and in correlation to this determine ultimate load to failure of a semitendinosus graft fixed with an interference screw in a cadaveric model. We hypothesized that failure is related to the decreasing bone quality at the lateral clavicle and its subsequent lack of secure fixation. Methods: Bone mass densitometry (GE Lunar Prodigy) was tested at 5 mm intervals from the lateral to the medial end of 11 clavicles (mean age 61 + 9.4 years). The BMD ROIs were 5 mm in width and chosen so that density was specific for the proposed tunnel area. Tunnels were drilled starting 7.5 mm from the lateral edge in 10 mm increments for a total of 5 tunnels (5 mm) in each clavicle. 5.5 9 8 mm PEEK Tenodesis Screws (Arthrex, Naples, FL, USA) were used to fix a semi-tendinosis (5 mm) graft in the tunnel. Grafts were cyclically loaded from 5 to 70 N for 3,000 cycles followed by load to failure at a rate of 120 mm/min. Results: The bone mineral density of the cadaveric clavicle increased significantly from lateral to medial at 0.304 ± 0.078 g/cm2 at 10 mm from the lateral 0.760 ± 0.103 g/cm2 at 50 mm (all p values \0.03). Similarly, the load to failure increased from lateral to medial. The load at 7.5 mm was 125.3 ± 42.5 N, the load at 17.5 mm was 189.4 ± 95.3 N (p = 0.69), the load at 27.5 mm was 240.8 + 148.1 N (p = 0.15), the load at 37.5 mm was 291.0 ± 122.1 N (p = 0.01), the load at 47.5 mm was 349.3 ± 120.3 N (p = 0.00). The majority of specimens failed by tendon pullout. The Pearson correlation coefficient between tunnel position and load to failure was 0.653. The correlation between bone density and load to failure was 0.659 and the correlation between tunnel position and bone density was 0.803. These all showed strong correlation. Conclusions: Failure at the lateral bone tunnel in CC ligament reconstruction for AC joint repair may be a result of poor bone

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 quality. BMD showed that optimal bone density was found in the anatomical insertion area of the CC ligaments between 25 and 50 mm from the lateral end of the clavicle. The low bone mineral density more lateral correlated with decreased load to failure. When drilling bone tunnels for this surgery, consideration should be given both to anatomic position as well as bone quality, which is poorer laterally.

P14-971 Arthroscopic management of proximal humerus malunion with tuberoplasty and rotator cuff re-tensioning A. La¨dermann1, P.J. Denard2, S.S. Burkhart2 1 La Tour Hospital, Orthopaedic Surgery and Traumatology, Meyrin, Switzerland, 2The San Antonio Orthopaedic Group, San Antonio, United States Objectives: This report describes the mid- to long-term results of arthroscopic tuberoplasty and rotator cuff retensioning for proximal humerus malunion. The hypothesis was that this approach would lead to significant improvements in functional outcome, thus providing a viable alternative to the traditional open management of proximal humerus malunions. Methods: Between August 2001 and October 2009, nine patients with a mean age of 49 years underwent shoulder arthroscopy tuberoplasty and rotator cuff advancement for malunion of the proximal humerus and were included in this study. The mean delay between the initial fracture and our surgery was 19 months. We developed a systematic technique to take down the rotator cuff over the malunited proximal humerus, to then perform a tuberoplasty, and then to re-tension and repair the rotator cuff by advancing it on the greater tuberosity. The mean follow-up was 50 months (range, 12–108). Results: Patients demonstrated a mean active forward elevation of 164 (range 90–180, gain 43), recovery of a mean active external rotation of 45 (range 30–60, gain 16), and a mean pain score of 1.8 points (range 0–5, reduction 3.8 points). The overall functional results according to the University of California Los Angeles score were excellent in three cases, good in three cases, and fair in three cases. No patient required additional surgery. Eight of the nine patients (89%) were able to return to their previous sports or activities. All patients declared themselves as being satisfied with the result. Conclusions: Arthroscopic tuberoplasty and rotator cuff re-tensioning for proximal humerus malunion is a viable alternative to traditional open methods, particularly in young patients. A comprehensive approach is recommended which addresses stiffness, associated pathology, impingement, and reestablishes rotator cuff function. Although the technique is technically demanding, it allows preservation of the native humeral head, is associated with a very low complication rate, and avoids concerns about long-term prosthetic survival in young patients.

P14-982 Acromioclavicular joint reconstruction: a comparative biomechanical study of three techniques A. La¨dermann1, P. Hoffmeyer2, B. Stimec3, J. Fasel3 1 La Tour Hospital, Orthopaedic Surgery and Traumatology, Meyrin, Switzerland, 2HUG, Orthope´die-Traumatologie, Gene`ve, Switzerland, 3Geneva University Hospitals, Geneva, Switzerland Objectives: Acute acromioclavicular joint dislocations indicated for surgery can be treated with several reconstruction techniques. The purpose of this in vitro study was to evaluate the acromioclavicular joint stability after three types of validated reconstruction as compared to the native situation. Methods: Nine pairs (right–left) of intact cadaveric shoulder specimens were assigned into three study groups with randomly distributed samples according to the coracoclavicular distance. The study groups

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 were instrumented with either acromioclavicular and coracoclavicular cerclages (CE), a Twin Tail TightRope (TR), or a LCP S-A Clavicle plate (CP). Native and instrumented specimens were tested quasistatic non-destructively (superior; 70 N, anteroposterior; ±35 N, 10 mm/min) and cyclically until failure (superior, valley load: 20 N, initial peak load: 70 N, increment: 0.02 N/cycle). Results: The TR study group showed the highest stiffness (superior: 73.77 ± 14.04 N/mm, anteroposterior: 29.58 ± 1.52 N/mm), followed by CE (superior: 59.73 ± 10.33 N/mm, anteroposterior: 24.31 ± 4.14 N/mm) and CP (superior: 24.08 ± 5.29 N/mm). Instrumentation generally led to increased superior and anteroposterior stiffness in each study group but to a significant superior stiffness reduction for CP (p = 0.029). Significantly lower coracoclavicular distance at valley load after 1 and 500 cycles was observed for TR (p = 0.018) and CE (p = 0.041) compared to CP. Cycles to failure in CE (7,298 ± 1,244 cycles) and TR (4,434 ± 727 cycles) were significantly higher compared to CP (1,683 ± 509 cycles), p = 0.011 and p = 0.031, respectively. Conclusions: The TR system provided the highest stability but failed earlier than the CE reconstruction compared to the native situation. The CE reconstruction might mimic the native acromioclavicular joint stiffness better than the other two setups, leading to more physiologic reconstruction.

P14-988 An innovative method of shoulder strength assessment A. La¨dermann1, P. Collin2 1 La Tour Hospital, Orthopaedic Surgery and Traumatology, Meyrin, Switzerland, 2Centre Hospitalier Prive´ de Saint Gre´goire, St Gregoire, France Objectives: Strength testing is an important aspect of shoulder examination. It is essential for diagnosing sports related conditions, assessment of progress and outcome of surgical repairs, fitness to return to work/sports. Strength assessment to date has still many limitations; there is no single standard instrument for measuring and various devices mentioned in the past had problems in terms of reliability, accuracy and cost. The purpose of this study was to compare the values of a method of strength testing with a weighing machine with existing methods. The hypothesis was that the presented method was reliable as a dynamometer. Methods: Eighty shoulders, 60 normal (group 1) and 20 pathologic (group 2) were tested in standard testing position of 90 of elevation in the scapular plane with elbow extended, forearm pronated and resistance given just proximal to the wrist. A weighing machine, an isometric dynamometer and a spring balance were used for strength testing, Each shoulder were tested three times with each device and mean value were noted for all three devices and subjected to statistical analysis. Results: There was a statistically significant difference in groups 1 and 2 between isometric dynamometer and spring balance (p = 0.0172 and p = 0.0218, respectively) but not between weighing machine and isometric dynamometer (p = 0.5713 and p = 0.5582, respectively). Conclusions: The described method with a weighing machine offer an attractive alternative in strength testing as it as accurate as a isometric dynamometer and readily available in all setups.

P14-1125 All-arthroscopic double bundle coracoclavicular ligament reconstruction using autogenous semitendinosus graft. A series of eighteen cases with a new technique J. Ranne1, J. Sarimo1, S. Orava1 1 Hospital NEO, Turku, Finland

S107 Objectives: An acromioclavicular joint (AC-joint) separation typically occurs after falling on the shoulder. Treatment is often conservative. In type III–V dislocations the pronounced lifted position and anterior–posterior translation of the distal clavicle may cause problems among physically active patients and operative treatment may be considered. Methods: We present our new all arthroscopic anatomical double bundle coracoclavicular (CC) reconstruction technique using a semitendinosus tendon autograft. What is new in this technique is that the dorsal limb of the graft goes around the dorsal edge of the clavicle recreating the conoid ligament. The anterior limb goes straight up recreating the trapezoid ligament. There is only one 6 mm drill hole in the clavicle and one 4.5 mm drill hole in the coracoid. The fixation apparatus uses the same drill holes. The solution effectively stabilizes the AC-joint and prevents anterior posterior translation. The material consists of 18 patients, the longest follow up time being 2.5 years. Results: The results have been very promising and the patients have been able to return to high demand sports or heavy labor. This novel all arthroscopic double bundle CC-joint reconstruction is an effective and reliable method in stabilizing the clavicle and neutralizing the anterior–posterior translation. Conclusions: This arthroscopic technique provides good stability of the ac-joint, a good esthetic outcome and we find it to be technically practical for the surgeon.

P14-1187 Arthroscopically assisted repair of acute grade III–V acromioclavicular joint dislocations J. Sarasquete1, L. Bruno1, F. Abat1, C. Alvarez1, A. Vidal2, I. Proubasta1 1 Hospital de la Santa Creu i Sant Pau. UAB, Orthopaedics and Traumatology, Barcelona, Spain, 2Centro Medico Teknon, Orthopaedics and Traumatology, Barcelona, Spain Objectives: To evaluate the clinical and radiological results of the arthroscopic repair of acute grade III–V acromioclavicular (AC) dislocations using a coracoclavicular (CC) suspension system. Methods: Twenty-one patients (mean age 36 years, 24–52) diagnosed with an acute high-grade AC joint dislocation (less than 3 weeks) were consecutively operated under arthroscopic control with a CC suspension system. Four of the dislocations were type III, 3 were type IV and 14 were type V according Rockwood. The clinical evaluation included a Visual Analog Scale (VAS) for pain, the DASH questionnaire and the SF-36. Radiological examination was taken to evaluate vertical and horizontal instability with Zanca and Alexander outlet/axial view. Results: Twenty patients were evaluated after a mean follow-up of 24.2 months (18.5–34.6). The other patient presented a mechanical failure after a traumatism that required revision arthroscopy and CC reconstruction with semitendinous graft. The interval from trauma to surgery was 9.3 days (3–20). The VAS, Dash and SF-36 showed an significative improvement from the pre-operative evaluation 7.83 (6–9), 80.4 (58.2–90.1) and 30.1 (PCS) (18.1–45.2); 45.5 (MCS) (29.6–54.5) to postoperative 1.94 (1–3), 3.5 (0.2–9.2) and 57.2 (PCS) (54.5–62); 55.1 (MCS) (46–58), respectively. In five patients (25%) the radiological results were not satisfactory in vertical and horizontal planes and related with high degree (IV–V) instability and older than 40 years (p \ .05). The clinical results in this residual instability subgroup were lower but not significative (p [ .05). Conclusions: Immediate arthroscopic repair of the acute grade III-V AC dislocations provides short term satisfactory clinical results with low patient morbidity. Patients with high degree of inestability and older than 40 may require higher mechanical support and/or biological contribution.

123

S108 P14-1250 A comparison of the effectiveness of ultrasonography guided subacromial injections and blind injections in patients with subacromial bursitis K. Goshima1 1 National Hospital Organization Ishikawa Hospital, Orthopedic Surgery, Kaga, Japan Objectives: Complaints of shoulder pain are very frequent in clinical practice. Blind injection of subacromial bursa for therapeutic purposes is commonly employed. However, the SAB is a very thin tissue. In the blind procedure, we cannot evaluate whether the injection needle end has been inserted into the SAB. High frequency ultrasonography is an accurate and safe imaging modality for guiding musculoskeletal injections. We prospectively compared the effectiveness of treatment with ultrasonography guided injection versus blind injection in patients with subacromial bursitis. Methods: We studied 50 consecutive patients with impingement sign. Patients with chronic inflammatory arthritis and complete rotator cuff tear were excluded. Patients were randomized to receive either USguided injections (Group U, n = 25) or blind SAB injections (Group B, n = 25) by the same orthopedic surgeon. The effectiveness of the treatment was indicated by a visual analog scale (VAS) for pain (before and 5 min after injection). Blind injection was done by an anterolateral approach. In US-guided injection, patients sat with the ipsilateral arm in extension. A 23G needle was inserted parallel to the transducer in a semi-oblique plane. The needle was advanced under realtime US control until the needle tip entered the SAB. The injection fluid contained 0.5 ml (2.5 mg) Betamethasone and 3 ml 1% Lidocaine. Results: VAS score showed a significantly greater improvement in Group U compared Group B (mean VAS score decrease: 46.7 for Group E vs. 32.8 for Group B, P \ 0.05). Ninety-two percent of patients in Group U had relief of pain of 50% or more, while 16% in Group B. Conclusions: US is an accurate and safe imaging modality for guiding musculoskeletal injections. In addition, we can understand the pathological condition with US examination. In this study, the ultrasonography guided injection technique is more accurate and can result in significant reduction in shoulder pain as compared with the blind injection technique.

P14-1319 Arthroscopic management of posterior shoulder instability M. Antonogiannakis1, E. Mataragas1, V. Tsiampa1, G. Arealis1, N. Tzanakakis1, M.V. Andriopoulou1 1 IASO GENERAL, Centre for Shoulder Arthroscopy, Athens, Greece Objectives: Posterior shoulder dislocation is difficult to diagnose and treat. The purpose of this study was to evaluate the effectiveness of arthroscopic posterior shoulder reconstruction. Methods: From 2005 to 2009 we treated 8 patients with recurrent posterior instability, 5 male and 3 female whose age ranged from 15 to 40 years old (average 27.75). They all presented with traumatic posterior instability, involuntary, unilateral. Two patients had multidirectional instability predominantly posterior. 3 patients had only one episode of posterior dislocation, one of them presented locked posterior dislocation which remained undiagnosed for 9 months. A closed reduction under anesthesia was achieved and shoulder arthroscopy followed. Another patient remained undiagnosed for 2 months. Five patients were into overhead or contact sports at a recreational or amateur level. 2 patients had a reverse Hill–Sachs lesion that was addressed by an arthroscopic tenodesis of the subscapularis and anterior capsular attachment on the bone defect by use of suture anchor. They all underwent posterior labrum and capsule reconstruction and 7 of them had a rotator interval closure. Patients

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 were followed up postoperatively with the Western Ontario Shoulder Instability score, the American Shoulder and Elbow Surgeons score and the Rowe Zarins score. Recurrent subluxation or dislocation was documented. Results: All patients were available for follow up which ranged from 32 to 72 months (average 41 m). At 6, 12 months and at years 1–3 the Rowe, and UCLA scores were measured. Additionally the WOSI score was used during their last evaluation in 2011. There was no recurrence of posterior instability. The mean scores in the latest follow-up were: Rowe score (93.2 ± 6), UCLA score (33 ± 2), WOSI score (62.5 ± 60, 97% of the normal shoulder). The WOSI sub-scores were: physical (23.12 ± 23.07), sports (20 ± 16), lifestyle (9.38 ± 9.1) and emotional (10 ± 7). 4 patients participate actively in sports in recreational and amateur level and are very satisfied, 1 is satisfied but does not participate in sports and 3 have minor complaints of pain and range of motion. Conclusions: Arthroscopic management of posterior shoulder instability can produce sufficient and reliable results. Arthroscopy enables the surgeon to have a clear view of the joint and the variety of pathology that emerges in cases of multidirectional and posterior instability. The evolution of the arthroscopic technique and materials used provide for the surgeon the means to deal with the less common cases of posterior instability.

P14-1443 A musculoskeletal model of the shoulder capable of informing clinical decision-making R. Matias1, A. Seth2, A.P. Veloso3 1 Faculty of Human Kinetics of TUL and Health College of PIS, Setu´bal, Portugal, 2Bioengineering, Stanford University, Stanford, United States, 3Faculty of Human Kinetics of the Technical University, Lisbon, Portugal Objectives: To develop a musculoskeletal shoulder model capable of accurately reproduce scapulothoracic and glenohumeral movements. Methods: A three-dimensional musculoskeletal shoulder model was created using OpenSim software system. The model includes graphical representation of the thorax/spine, clavicle, scapula, humerus, radius and ulna and uses 11 DOF to describe the relative movements of above mentioned segments. The model includes sixteen muscle– tendon actuators, 4 sets of ligaments and a 4 DOF scapulothoracic joint. To assess how accurately the model reconstructs human movements of daily living, kinematic data of the thorax, scapula and humerus was collected using an electromagnetic three-dimensional tracking system (Flock of Birds). Seven daily-activities including arm elevation in the sagittal (flexion) and coronal (abduction) planes, shoulder shrug, 2-point crutch gait, pushups, combing hair, and getting on object from a table, were performed. The model inverse kinematics results were then compared with the recorded data to determine model error by computing the root-mean-square of the Euclidean distance of experimental and model markers. Results: Root-mean-square of the Euclidean distance of the 9 experimental markers (3/segment) and their homologous model markers was computed for each daily-activity, resulting in the following model errors: flexion (4 mm + 2), abduction (6 mm + 3), shoulder shrug (2 mm + 1), 2-point crutch gait (4 mm + 1), pushups (5 mm + 3), combing hair (6 mm + 5), and getting on object from a table (4 mm + 2). Conclusions: From the results it can be concluded that this shoulder model accurately reconstructs scapula and humerus movements for a wide range of daily-activities involving the shoulder from surface markers. Assessment of the motion of the scapula and humerus is applicable to a variety of clinical scenarios. We will discuss the use of this model to assist clinicians in their clinical reasoning and decisionmaking. This model and the OpenSim simulation software are freely available for download from SimTK.org.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P14-1466 Can PEMFs reduce postoperative pain and capsulitis occurrence after arthroscopic rotator cuff repair? A prospective, randomized study L. Osti1, R. Papalia2, A. Del Buono2, N. Maffulli3, V. Denaro2 1 Hesperia Hospital, Modena, Italy, 2Campus Biomedico Roma, Roma, Italy, 3Queen Mary University of London, Barts and The London School of Medicine and Dentistry, Centre for Sports and Exercise Medicine, London, United Kingdom Objectives: The purpose of this study was to compare the clinical outcomes and the ovccurrence of capsulitis after application of Pulsed electromagnetic fields (PEMFS) after arthroscopic single row repair for small to medium rotator cuff tears. Methods: Randomized controlled study; Level of evidence, I. Sixty-five patients who underwent shoulder arthroscopy for the repair of a 2- to 4-cm rotator cuff tear (proven by arthroscopy) were randomly divided into 2 groups, using a block randomization procedure. A treatment group (N = 31) consisted of those who received postoperative PEMPS application for the first 6 weeks post-op. A control group (N = 34) consisted of those who did not receive that treatment. Patients were evaluated for pain, range of motion, strength, and overall satisfaction at baseline, at 3 month, 1 year and 2 year follow ups. The Constant, the University of California at Los Angeles (UCLA) were administered to all the patients. Occurrence of capsulitis was assessed on clinical ground at 3 months. Results: Postoperative clinical scores were significantly improved compared to preoperative status in both groups. No intergroup differences in Constant and University of California, Los Angeles scores were found at any appointment. At 3 month appointment, pain visual analog scale and range of motion (external rotation in abduction) were significantly improved after PEMFS application. No significant differences were reported at 1 and 2 year follow-ups. Capsulitis occurred in patient of the PEMFS treatment group, and in 2 of the control group. However, despite numerical differences, these differences were not statistically different. Conclusions: This study shows that PEMPS application improves pain and external rotation in the short term. The lower occurrence of capsulitis after PEMFS administration had no statistical relevance. Nevertheless, as the study may have been underpowered to detect important differences in occurrence of capsulitis, additional investigations are needed.

P14-1490 Biomechanics of a new technique for minimal-invasive coracoclavicular ligament reconstruction in chronic acromioclavicular joint instability B. Schliemann1, S. Lenschow1, M. Herbort1, M.J. Raschke1 1 University Hospital Mu¨nster, Trauma, Hand and Reconstructive Surgery, Mu¨nster, Germany Objectives: Acromioclavicular (AC) joint separation represents a common injury and accounts for approximately 12% of all shoulder girdle dislocations (Collins 2009). Several surgical procedures have been described to treat high-grade injuries. However, surgical techniques are demanding and often require extensive preparation of the coracoid process. We established a new minimal-invasive technique to easily shuttle a tendon graft through a drill hole into the coracoid process using a flip button and a polyester cord in a lifting block fashion. The purpose of the present biomechanical study was to evaluate the biomechanical properties of this anatomic CC ligament reconstruction compared to a traditional tendon loop and a synthetic double bundle reconstruction. Methods: A porcine metatarsalia model was used to assess superoinferior fixation strength of the different augmentation techniques.

S109 Cyclic loading from 20 to 70 N for 1,000 cycles was performed. Permanent elongation was determined after 1,000 cycles. Finally, a load to failure protocol was applied to all specimens. Results: All specimens of the three different groups survived the cyclic loading protocol. The maximum load to failure under superior loading conditions were 760 ± 78 N for the anatomic reconstruction group, 702 ± 48 N for the conventional tendon loop group and 1,117 ± 91 N for the synthetic double bundle reference group. The double bundle reconstruction group revealed significantly higher maximum loads compared to the other groups (p \ 0.001). The anatomic reconstruction group with a flip button/tendon graft tend to have higher maximum loads than the conventional tendon loop group. However, this difference was not statistically significant (p = 0.143). The permanent elongation was the lowest for the double bundle reconstruction group (0.42 ± 0.14 mm) compared to 1.2 ± 0.37 mm for the anatomic and 1.14 ± 0.31 for the conventional tendon augmentation group. The conventional tendon loop showed the highest stiffness levels (68 ± 13 N/mm). Conclusions: The biomechanical properties of the anatomic coracoclavicular ligament reconstruction using a flip button/polyester cord in a lifting block fashion are comparable to those of the conventional tendon loop. The maximum load of the anatomic reconstruction technique was 760 ± 78 N which is equivalent with the maximum loads described for the intact CC ligaments (Harris et al. 2000, Grutter et al. 2005). Furthermore, it exceeds the maximum loads described for other reconstruction techniques such as the Weaver–Dunn procedure (523 ± 98.6 N) and the GraftRope (646 ± 167.4 N; Thomas et al. 2011). Our results suggest that the described technique is an alternative option to reconstruct the CC ligaments in chronic AC joint instability. The use of a flip button and a polyester cord in a lifting block fashion to shuttle the tendon graft into the coracoid may facilitate the surgical procedure because extensive preparation of the coracoid process is not required.

P14-1519 A comparative and critical analysis of rehabilitation techniques and self-rehabilitation techniques used for the treatment for stiff shoulder: a prospective multicenter study of 148 stiff shoulder cases P. Grosjean1, P. Gleyze2, T. Georges3, E. Laprelle4, P.H. Flurin4, C. Charrousset5 1 UGECAM Alsace, Colmar, France, 2Albert Schweitzer Hospital, Orthopaedic and Arthroscopic Unit, Colmar, France, 3ATOL, CHU, Nancy, France, 4Centre Chirurgie Orthope´dique et Sportive, Merignac, France, 5Institut Oste´o Articulaire Paris Courcelles, Paris, France Objectives: This study evaluates and compares the impact of each exercise used for conventional rehabilitation and for self rehabilitation for treatment of stiff shoulder. Methods: A prospective multicenter study has been performed with the individualisation of three different therapeutic populations (P1: Conventional rehabilitation—58 cases, P2: Self-rehabilitation with stimulation for pain management—59 cases, P3: Self-rehabilitation for pain management in conjunction with a physical therapist—31 cases). The impact of each exercise and its achievable outcome on day and night pain, patient function, moral, goniometric measurements. A constant score was evaluated daily during the first 6-weeks then on a weekly basis until the 3rd month (regression tests). Physio and balneotherapy always gave favorable results. Passive mobilisation, Sohier method, cervico dorsal massages were shown to be harmful during the first weeks then became positive after 1-month. The negative impact was significant for all subjective and objective criteria (p \ 0.05).

123

S110 Results: Self-rehabilitation exercises with stimulation for pain management are immediately beneficial for night pain, function, patient moral in reference to day pain in correlation with the duration of the exercises (p \ 0.05). Pain during the day was a positive criteria for improving the clinical results (p \ 0.05). After 6 weeks, clinical improvement with self-rehabilitation is limited/takes time. The failure rate (AAE \ 140) are equivalent (14–17%). Conclusions: Pushing the limit of the patients pain threshold in conjunction with the ultilisation of a controlled combination of selfrehabilitation and standard rehabilitation exercises for pain management can give a quick, reliable results and good long term clinical outcomes for the treatment of stiff shoulder.

Shoulder II

P15-53 Comparison of active and passive measurements of shoulder proprioception in healthy individual M. Unal1, B. Dilek2, S. Gulbahar2, E. Akalin2, D. Akseki3, H. Pinar4 1 Isparta S¸ ifa Hospital, Orthopaedics and Traumatology, Isparta, Turkey, 2Dokuz Eylu¨l University, Physical Therapy and Rehabilitation, Izmir, Turkey, 3Balikesir University, Orthopaedics and Traumatology, Balikesir, Turkey, 4Dokuz Eylu¨l University, Orthopaedics and Traumatology, Izmir, Turkey Objectives: Importance of proprioception in the treatment and prevention of sports injuries has become increasingly clear. Measurement of proprioception is still controversial and no standard method exists. The purpose of this study was to determine the differences between active and passive measurements of shoulder proprioception in healthy volunteers. Methods: Twenty-two women, 12 men, total 34 healthy volunteers with normal shoulders whose ages were between 22 and 29 (av. 26.18) were included in the study. Proprioceptive level was measured from the dominant extremity with the technique of both active and passive joint position sense by a isokinetic dynamometer. 0 neutral and 30 external rotation were selected as the starting positions, and the tests were done at 10 internal and 10 external rotation positions from these angles. A total of eight comparisons were done according to these parameters. Statistical analyses were done by Wilcoxon Ranked Sign test. Results: Significant differences were found between reproduction errors of target angles between passive and active measurements at all degrees (p \ 0.05). We found significant difference between active and passive measurements at internal and external rotation from 0 position (p \ 0.05) and also there were significant difference found at internal and external rotation from 30 external rotation position (p \ 0.05). Conclusions: Active and passive testings of proprioception seem to give different results. These differences should be kept in mind during proprioception measurements.

P15-112 Suspensory fixation for intraosseus arthroscopic long head biceps tenodesis. Preliminary results J.A. Camacho Chaco´n1, A. Calvo Diaz1, A. Roche Albero1, S. Ferrer Peiron1, J.V. Badiola Vargas1 1 Miguel Servet University Hospital, Orthopaedic Surgery Department, Zaragoza, Spain Objectives: The purpose of this study is to evaluate preliminary clinical and radiogic outcome of patients who underwent a newfound procedure of Suspensory Fixation for Intraosseus Arthroscopic LHB Tenodesis.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: This was a prospective study of 50 patients of mean age 47.8 years (range 30–58 years) who were diagnosed with LHB pathology. Patients underwent a newfound Suspensory Fixation for Intraosseus Arthroscopic LHB Tenodesis Technique with ToggleLoc System (Biomet), the procedure was performed in patients without rotator cuff tears and with a level of work/sport activity of moderate/high performance. Follow-ups at mean of 12 months (range 1–24 months) where held and evaluated with Constant Score, Visual Analog Scale for Pain, and with digital dynamometer strength measurement; additionally radiological evaluation was assessed with X-ray and MRI. Subsequently surveys were facilitated to patients to value level of satisfaction for postoperative results and estimated time to return to work. Results: Clinical parameters were statistically significant and muscle strength improved. Surveys conclude satisfaction of results and decreased average time to return to work. Conclusions: This study shows that Suspensory Fixation for Intraosseus Arthroscopic LHB Tenodesis newfound procedure delivers excellent postoperative results, prompt incorporation to work and patients satisfaction. P15-115 Functional results and anatomic control of an arthroscopic repair technique of large and massive rotator cuff tears by a side-by-side suture T. Rousseau1, S. Bertiaux1, O. Courage2, X. Roussignol1, F. Duparc1, F. Dujardin1 1 Rouen Hospital, Orthopedic Department, Rouen, France, 2Hopital Prive´ de l’Estuaire HPE, Le Havre, France Objectives: For this study an arthroscopic technique was used to repair large tears of the rotator cuff. To limit tension, side-by-side sutures were used. The goal of this study was to evaluate the result of the technique after 2 years by analyzing the functional score and the continuity of the tendon by 3D sonogram. Methods: The retrospective study included a continuous series of 50 patients with an average age of 66.6 years old with a range of 46–80 years old. The patients presented large or massive retracted supraspinatus and infraspinatus tears and were treated between January 2007 and March of 2008. The treatment consisted of a bursectomy, an acromioplasty, and side-by-side suturing of the rotator tendons associate with tuberosity suture using one anchor with 2 sutures passing through the two sides. The weighted Constant score was evaluated preoperatively and after 24 months. The continuity of the rotator cuff was controlled by sonogram. The prognostic factors were studied. Results: 98% patients were reviewed. The average weighted Constant score was significantly better (p \ 0.05), going from 40% (18–67) preoperatively to 91.7% (40–107) postoperatively for the entire field. 56% of the cuffs in this series were continent with a weighted Constant score of 98.4% (40–126). We did not observe the recovery of force for this subgroup. 88% of the patients declared that they were satisfied or very satisfied with this operation. Conclusions: The side-by-side arthroscopic suture without tension technique allows a durable restoration of the continuity of the tendon in over 50% of the cases. Even in the case of a reoccurrence of a rupture and the absence of the recovery of force, the improvement of the functional score continues to be significant after 24 months.

P15-153 The use of a pre-operative Botulinum toxin injection in arthroscopic rotator cuff repair. A 5 year clinical follow up study H. Van der Bracht1, K. van Rooyen2, P. Huijsmans3, J. de Beer2 1 Gent University Hospital, Gent, Belgium, 2Cape Shoulder Institute, Panorama, South Africa, 3HAGA, Hospital, The Hague, The Netherlands

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Advances made in surgical technique and rehabilitation protocols have increased the rehabilitation speed after rotator cuff repair. Our hypothesis was that injecting Botulinum Toxin (Botox) into the rotator cuff muscles prior to surgery would enable the patient to use the upper limb actively in the postoperative period without compromising the rotator cuff repair. Methods: A retrospective clinical follow-up study with a mean follow-up of 5.8 years was performed. 11 Patients received a preoperative Botox injection in the rotator cuff muscles and were treated with an arthroscopic double row rotator cuff repair. Paresis of the injected muscles was observed intra-operatively and active mobilization of the upper limb immediately postoperative was allowed in all patients. Muscle function, clinical evaluation and ultrasound evaluation were performed at regular time points and at latest follow-up. Results: The mean Constant–Murley score improved from 67 to 88. All patients subjectively reported the outcome of the procedure as excellent and no side effects of the Botox were noted. Conclusions: A Botox injection in the rotator cuff muscles prior to a rotator cuff repair is a safe technique which allows for immediate active mobilization of the shoulder without compromising the final outcome.

P15-181 Evaluation of the diagnostic accuracy of preoperative MR arthrography for the subscapularis tendon tears T. Sano1, H. Matsuoka1, K. Nakayama1, T. Saji1 1 Shizuoka General Hospital, Department of Orthopaedic Surgery, Shizuoka, Japan Objectives: The purpose of this study was to evaluate the diagnostic accuracy of preoperative MR arthrography for the detection and grading of subscapularis tendon tears based on only arthroscopic findings. Methods: We examined 113 cases who received primary arthroscopic rotator cuff repair from January 2009 to May 2011. Sixty-eight cases were males, and 45 cases were females. Seventy cases were right shoulders, and 43 cases were left shoulders. Mean age of patients at the time of surgery was 62.7 years (range from 35 to 79 years). The indications of surgery for all patients in this study were rotator cuff tears which include supraspinatus, infraspinatus and/or subscapularis tendon. The time interval from preoperative MR arthrography to the arthroscopic surgery was within less than 3 months. MR imaging was performed with a 1.5 T system (Avanto, Siemens Medical Systems, Erlangen, Germany). The imaging conditions were 18 cm of field of view, 4 mm of slice thickness and 0.4 mm of intersection gap. The subscapularis tendon was evaluated with transverse and parasagittal images. We classified the findings of preoperative MR arthrography and the arthroscopic surgery into Grade M0 to M3 and Grade 0–3 respectively. Grade 0 and M0 were normal subscapularis tendon. Grade 1 and M1 were partial tear of articular side of the subscapularis tendon. Grade 2 and M2 were full thickness tear of superior one-third of the subscapularis tendon. Grade 3 and M3 were full thickness tear of superior two-third or all of the subscapularis tendon. We evaluated the diagnostic accuracy of preoperative MR arthrography based on only arthroscopic findings. Results: For the subscapularis tendon tears (Grade 1–3), sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 92.2, 80.6, 91.0, 82.9 and 77.0% respectively. For partial tear of articular side of the subscapularis tendon (Grade 1), sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 86.7, 80.6, 84.8, 82.9 and 84.0% respectively. Conclusions: The preoperative MR arthrography is useful in detection and grading of the subscapularis tendon tears.

S111 P15-184 Technique and outcomes of endoscopic scapulothoracic bursectomy and partial scapulectomy O.A.J. Van der Meijden1, T.R. Gaskill1, M.P. Horan1, P.J. Millett1 1 Steadman Philippon Research Institute, Vail, CO, United States Objectives: The scapulothoracic articulation is an underappreciated yet critical component of normal glenohumeral function. Although uncommon, abnormalities within this articulation can result in pain or mechanical symptoms. The objective of this study was to assess the efficacy of endoscopic scapulothoracic bursectomy in patients with snapping scapula syndrome. Methods: In this IRB-approved study, 23 shoulders in 21 consecutive patients were identified that had undergone endoscopic treatment of snapping scapula syndrome. Each patient described mechanical symptoms, failed non-surgical modalities, and reported symptomatic relief from a local anesthetic injection prior to surgical intervention. Pre- and postoperative pain and functioning levels were assessed using the ASES, quick-DASH and SANE shoulder scores and patient satisfaction was recorded using a 10 point VAS scale. Univariate and paired t tests were used for data analysis. Significance was established when p B 0.05. Results: The mean age at time of surgery was 33 years (19–58 years). A scapulothoracic bursectomy alone was performed in 2 shoulders and the remaining 21 shoulders underwent both bursectomy and scapuloplasty of the superomedial or inferomedial scapular border. At a mean follow-up of 2.5 years (2–4 years), available for 21 of 23 (91%) shoulders, a significant improvement in the ASES score was noted from 53 points (range 17–83) preoperatively to 75 points (range 32–100) postoperatively (p = 0.001). The mean SANE and quickDASH scores at follow-up were 75 (range 15–100) and 34 (range 0–89) respectively. Overall, patient satisfaction with surgical outcome was 6 out of 10 points. Two shoulders (7%) were revised for persistent scapulothoracic pain and 2 others subsequently underwent additional shoulder surgery for various other pathologies. Conclusions: Although immediate significant pain and functional improvement, in addition to low revision rates, can be expected after endoscopic bursectomy and scapuloplasty, the average postoperative ASES and SANE scores are lower than expected. Further analysis will be necessary to determine how outcomes can be improved and if prognostic risk factors can be identified.

P15-237 Transosseus rotator cuff repair M. Fiodorovas1 1 Klaipeda University Hospital, Sports Trauma Department, Klaipeda, Lithuania Objectives: The goal of rotator cuff repairs is to achieve high initial fixation strength, minimize gap formation, maintain mechanical stability under cyclic loading, and optimize the biology of the tendonbone interface until the cuff heals biologically to the bone. The purpose of this study was to evaluate the results of an arthroscopic transosseus rotator cuff repair. The goal of rotator cuff repairs is to achieve high initial fixation strength, minimize gap formation, maintain mechanical stability under cyclic loading, and optimize the biology of the tendon-bone interface until the cuff heals biologically to the bone. The purpose of this study was to evaluate the results of an arthroscopic transosseus rotator cuff repair. Methods: Forty patients, age 43–69 (mean 54), with a full-thickness rotator cuff tear underwent arthroscopic all inside transosseus technique repair. For this procedure we used Tornier ArthroTunneler system. All patients were operated in 2010 year. Follow-up period 6 months. Examination using Constant, UCLA shoulder scoring systems.

123

S112 Results: According UCLA 86% of the patients had good and excellent post-operative scores. The average Constant score improved from a preoperative rating of 43.7 to a postoperative rating of 84.1. The average increase in the Constant score after the operation was 27.2 points. Conclusions: This repair maximizes the utility of a single-row repair technique by preserving the suture limbs of the medial single-row and bridging these sutures over the footprint insertion with distal-lateral interference screw suture fixation. This technique may be considered a surgical option if an inserted suture anchor on the greater tuberosity is pulled out with a small amount of tension, especially during arthroscopic rotator cuff repair in osteoporotic elderly patients. Such type of fixation compresses the tendon, optimizing tendon-to-tuberosity contact.

P15-266 Reliability of the geometric classification of rotator cuff tears based on magnetic resonance arthrography P. van der Zwaal1, B. Thomassen1, T. Urlings2, T. de Rooij2, J.-W. Swen1, E. van Arkel1 1 Medical Center Haaglanden, Orthopaedic Surgery, The Hague, The Netherlands, 2Medical Center Haaglanden, Radiology, The Hague, The Netherlands Objectives: To determine the reliability of the geometric classification of rotator cuff tears based on magnetic resonance arthrography. Methods: We retrospectively reviewed preoperative MR arthrograms of 73 consecutive patients who were surgically treated for their fullthickness rotator cuff tear. The images were blinded and evaluated by two orthopaedic shoulder surgeons and two musculoskeletal radiologists using the geometric classification of rotator cuff tears and measuring the sagittal/coronal dimensions of the tear. Review was performed twice with an interval of at least 8 weeks. Agreement was calculated using the weighted Kappa coefficient (j) and the intraclass correlation coefficient (ICC). Results: Intraobserver agreement was excellent for both the geometric classification and the sagittal/coronal dimension measurement (j = 0.81–0.92; ICC 0.84–0.98). The ICC for the interobserver reliability was excellent for all sagittal and coronal dimension measurements (ICC = 0.95–0.97). The interobserver agreement of the geometric classification was good for the orthopaedic surgeons (round 1: j = 0.75; round 2: j = 0.73). The interobserver agreement for the radiologists was excellent in observation round 1 (j = 0.82) and good in observation round 2 (j = 0.71). The interobserver agreement between orthopaedic surgeons and radiologists was found to be moderate to moderately good (j = 0.52–0.66). Conclusions: The geometric classification using MR arthrography for rotator cuff tears has good to excellent intra- and interobserver agreement in experienced observers. It meets the criteria of an efficient classification and we believe it is a useful tool in the evaluation of rotator cuff tears with good reliability.

P15-294 Comparison of glenohumeral contact pressures and contact area after glenoid reconstruction with Latarjet or distal tibial osteochondral allograft N.N. Verma1, S. Bhatia1, G. Van Thiel1, D. Gupta1, E. Shewman1, A.A. Romeo1, B.R. Bach1 1 Rush University Medical Center, Department of Orthopaedic Surgery, Division of Sports Medicine, Chicago, United States Objectives: Multiple bone-grafting procedures exist for patients with glenoid bone loss and shoulder instability, with favorable results shown after autologous bone-grafting procedures such as the Latarjet

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 reconstruction; yet many concerns remain. Reconstruction with distal tibial osteochondral allograft can offer improved joint congruity and a cartilaginous articulation for the humeral head. The purpose of this study was to investigate changes in glenohumeral contact area, contact pressure, and peak force after creation of 30% anterior glenoid defect and subsequent reconstruction with a Latarjet coracoid graft or distal tibia osteochondral allograft. It is hypothesized that the distal tibial bone graft will best normalize articular contact area, contact pressure, and peak force within the glenohumeral joint. Methods: Eight cadaveric shoulder specimens (5 right, 3 left) from donors under age 50 were dissected free of all soft tissues and randomly tested in 3 static positions of humeral abduction–30, 60, and 60 abduction with 90 external rotation (ABER)—with a 440 N compressive load. Glenohumeral contact area, contact pressure, and peak force were determined using a digital pressure mapping system (Tekscan, South Boston, MA, USA) for (1) intact glenoid; (2) glenoid with 30% anterior bone defect; (3) glenoid after reconstruction with a distal tibial bone graft or Latarjet bone block. Each glenoid underwent both types of reconstruction, with random assignment to which occurred first. A one-way ANOVA with Tukey’s Post Hoc analysis was performed to compare values between testing conditions. Results: Reconstruction with a distal tibial allograft resulted in significantly higher glenohumeral contact area than the Latarjet in 60 abduction and ABER (p \ 0.05). The distal tibial allograft also demonstrated significantly lower peak forces than the Latarjet in the ABER position (p \ 0.05). In the bone loss model, the distal tibial allograft showed significantly higher contact areas and significantly lower contact pressures and peak forces than the 30% defect model at all positions. The Latarjet also followed this pattern but without statistically significant differences in contact area and peak force between the defect and Latarjet in the ABER position (p [ 0.05).

Glenhumeral contact pressure map with arm in 60 abduction and 90 external rotation (ABER position). Higher pressures are signified by yellow and orange, lower pressures by blue and green (see scale). a Intact glenoid; b glenoid with 30% anterior bone defect; c glenoid after Latarjet bone graft procedure; d glenoid after bone reconstruction with distal tibial allograft. GH Contact Pressure Map with Key Conclusions: Reconstruction of large anterior glenoid bone defects with distal tibial osteochondral allograft provides improved glenohumeral congruity over the Latarjet at 60 abduction and in the ABER position. Lower peak forces in the ABER position may also be seen with a distal tibial allograft over the Latarjet. Though these mechanical properties may translate into clinical differences, further studies are needed to fully understand their effects.

P15-309 Posterior capsulorraphy in addition to arthroscopic Bankart repair for instability without bone loss O. Verborgt1, M. Vanhees2, R. van Riet3, N. Janssen2, G. Declercq2 1 University of Antwerp, AZ Monica, Orthopaedic Surgery, Deurne, Belgium, 2University of Antwerp, AZ Monica, Deurne, Belgium, 3 Wilrijk, Belgium Objectives: To evaluate the effect of adding a posterior capsulorraphy to standard arthroscopic Bankart repair on functional outcome and recurrence rate in patients without significant bone loss.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: 73 patients with recurrent, anterior shoulder instability were reviewed at a mean follow-up of 47.1 months (range, 24–83). Sixtyfour patients were involved in competitive sports, of which 25 were contact sports. In all patients a posterior capsulorraphy was done using minimum 2 capsulo-labral tugs in addition to a standard arthroscopic Bankart repair. Patients were evaluated using the Oxford Shoulder Instability Score (OSIS). Return to previous level of sports and work, survival rate, rate of recurrence and need for revisions were assessed. Results: Overall, the average OSIS was 19.1 points (range, 12–44). Return to previous type of work was possible in 94.5% and return to the same level of sports was possible in 85%. The recurrence rate over a period of maximum 7 years was 10.9%. All recurrences occurred in the period 3–5 years after the index operation, with a peak incidence at 4 years post-operatively. In 6.8% there was a complete re-dislocation, in 4.1% there were only one or more subluxations and 4.1% needed revision surgery. In contact athletes, the average OSIS was 19.4 points (range, 12–34). Return to sports was possible in 80%. The recurrence rate for this group was 20% and the revision rate was 8%. This was significantly higher than in the non-contact athletes group (P \ 0.001). Conclusions: This study demonstrates reliable functional outcome after arthroscopic Bankart repair with posterior capsulorraphy for recurrent, anterior shoulder instability without bony lesions. However, functional outcome and recurrence rate did not improve compared to results after standard arthroscopic Bankart repair alone. Moreover, in contact athletes the results are much less favorable with higher recurrence rates at mid-term follow up even in the absence of bony lesions.

P15-311 Standard or decelerated rehabilitation following arthroscopic rotator cuff repair: a randomized, prospective study O. Verborgt1, S. Defoort2, A.M. Cools3, G. Van den Bogaert4 1 University of Antwerp, AZ Monica, Orthopaedic Surgery, Deurne, Belgium, 2AZ St Jozef, Izegem, Belgium, 3University Hospital Ghent, Faculty of Medicine and Health Sciences, Department of Rehabilitation Sciences and Physiotherapy, Ghent, Belgium, 4 AZ St Elisabeth, Herentals, Belgium Objectives: To compare the effect on functional and structural outcome of standard and decelerated rehabilitation programs after arthroscopic rotator cuff repair. Methods: 40 patients with full-thickness rotator cuff tears were randomly assigned to a standard or decelerated rehabilitation protocol after arthroscopic repair. Standard protocol included immediate physiotherapy with passive mobilisations, while the decelerated group did not get formal therapy the first 6 weeks. All patients were clinically assessed pre-operatively and post-operatively at 6 weeks, 3 months and 1 year using Constant score, VAS and SST. At 1 year postoperatively, the integrity of the repair was assessed with ultrasound examination. Results: Clinically, there was no significant difference between standard and decelerated rehabilitation groups at 6 weeks, 3 months or 1 year post-operatively in terms of Constant score, SST, VAS score and structural integrity on ultrasound. Although not significant, only pain (p = 0.05) and VAS score (p = 0.05) tended to be better at 6 weeks after decelerated rehabilitation compared to standard rehabilitation. Conclusions: Concerns for recalcitrant stiffness and poorer functional outcomes have led to some to favor early postoperative therapy. We found that early restriction of motion did not lead to poorer functional outcome at 6 weeks, 3 months or 1 year postoperatively or difference in structural integrity on ultrasound after 1 year.

S113 P15-374 Refixation of the supraspinatus tendon in the rat model: influence of continuous G-CSF application via phospho-lipid gel on tendon structure and biomechanical properties S. Buchmann1, L. Walz2, T. Reichel3, K. Beitzel4, G. Winter5, A.B. Imhoff3 1 Schulthess Clinic, Department of Orthopaedic Surgery, Zu¨rich, Switzerland, 2University Hospital Basel, Clinical Trial Unit, Basel, Switzerland, 3Technical University Munich, Department of Orthopaedic Sports Medicine, Munich, Germany, 4University of Connecticut Health Center, Department of Orthopaedics, West Hartford, United States, 5Ludwig Maximilian University, Department of Pharmacy, Pharmaceutical Technology and Biopharmaceutics, Munich, Germany Objectives: With phospholipidgels (PG) a continuous in vitro release of growth factors can be performed. In a pilot study a positive influence on tendon remodeling through a continuous G-CSF release via osmotic pump was found. But the pump implantation revealed a high complication rate (tube dislocations up to 30%). The hypothesis was now proposed that this effect can also be shown with a G-CSF application via PG with a reduced potential of complications. Methods: In 60 Sprague–Dawley rats (400 g, 14 weeks) the supraspinatus tendon (SSP) was completely detached, 3 weeks later a transosseous refixation was performed. Footprint and tendon surface were covered with in total 0.2 ml PG. 4 groups (each n = 15) were followed: Control (no PG), Placebo (PG without G-CSF), G-CSF high (10 lg/24 h) and G-CSF low (1 lg/24 h). 6 weeks after refixation histological analysis (n = 9) was performed with HE staining (MOVIN-Score) and Collagen I/III staining. Biomechanically (n = 6) maximal load to failure (N) and stiffness (N/mm) were measured and compared to the healthy contralateral side. Results: The MOVIN sum scores showed no sign. differences but the collagen staining revealed with G-CSF sign. lower Collagen III as Control/Placebo. Biomechanically the GSF tendons showed a higher percentage of restored load to failure (100% = healthy contralateral side) compared to Placebo (G-CSF low 81.6% p = 0.02/G-CSF high 80.4% p = 0.062/Placebo 62.7%). The stiffness values showed no sign. differences. The placebogel showed neither in histology nor in biomechanical properties negative side effects on the healing process. Conclusions: Our results show that a continuous G-CSF release positively influences the tendon healing process. But the results from the pilot study could not have been proven, because the histological analysis (HE staining) showed not the sign. results as the previous study did. There is an improvement of the biomechanical properties (load to failure). The application of the placebo PG showed no negative influence on healing behavior (inflammation, granuloma). Further studies with larger groups and different growth Factors are necessary, also large animal studies might be taken into consideration.

P15-405 Long term quality of life in patients treated for massive retracted immobile cuff tears P. Randelli1, L. Zottarelli2, C. Fossati2, P. Arrigoni2, V. Ragone2 1 Universita` degli Studi di Milano,Policlinico S. Donato IRCCS, Dipartimento di Scienze Medico-Chirurgiche, Milan, Italy, 2 Universita` degli Studi di Milano,Policlinico S. Donato IRCCS, Dipartimento di Scienze Medico-Chirurgiche, San Donato Mil., Italy Objectives: The aim of this study is to evaluate long term clinical outcomes of arthroscopic repair in patients with massive contracted immobile rotator cuff tears. Methods: We included in this study 26 patients with massive contracted rotator cuff tears, either partially or completely, arthroscopically repaired between 2005 and 2009. In 18 patients

123

S114 (70%) the cuff was completely repaired to the bone using an interval slide technique (single or double interval slide in 11 and 7 patients respectively). In the remaining 8 patients (30%) a functional partial rotator cuff repair was performed (in 3 of them the repair was performed using an interval slide technique). Patients were retrospectively evaluated with validated outcomes scores: Disabilities of the Arm, Shoulder and Hand questionnaire (DASH), Visual Analog Scale (VAS) for pain, Simple Shoulder Test (SST) and Single Assessment Numeric Evaluation (SANE). Range of motion (ROM) was determinated by a self-assessment questionnaire. Results: The mean age was 64.1 (range 49–74). The mean postoperative follow-up was 39.6 months (range 19–70). A overall satisfactory quality of life was reported (76% mean SANE). The final mean DASH and SST was respectively 23.8 and 8.8. The residual level of pain was low (mean VAS 1.8). Mean range of motion were 157, 5 in elevation and 55, 3 in extra-rotation. Conclusions: Arthroscopic repair in massive contracted, immobile, rotator cuff tears provide a long term satisfactory quality of life. This minimal invasive surgical approach can be a valid alternative to reverse shoulder arthroplasty.

P15-425 Isokinetic testing in patients with rotator cuff rupture I. Ho¨fling1, L. Thorwesten2, U. Va¨a¨ta¨inen1, P. Sipola3, L. Niemitukia3, H. Kro¨ger1 1 Kuopio University Hospital, Orthopaedics and Traumatology, Kuopio, Finland, 2University of Mu¨nster, Sports Medicine, Mu¨nster, Germany, 3Kuopio University Hospital, Radiology, Kuopio, Finland Objectives: The purpose of the present study was to investigate the association between isokinetic parameters and rotator cuff tear size and location, acromial shape and various anthropometrical and clinical parameters in patients with rotator cuff rupture. Methods: A total of 35 patients (average age 58.2 ± 6.5 years, 16 females and 19 males) with symptomatic rotator cuff rupture were studied. Diagnose was confirmed with ultrasound and MRI. Preoperative parameters included age, BMI, pain intensity and duration, range of motion (ROM), and acromial shape, which was determined from preoperative X-rays using Bigliani’s classification. Isokinetic testing of both shoulders in flexion and extension was performed using a Lido Active isokinetic system. Peak torque (PT), average peak torque (APT) and total work done (TWD) at 80/s and 120/s were evaluated. Tear size and location were recorded during surgery (rotator cuff repair and acromioplasty). Results: Men achieved significantly higher values for all isokinetic parameters, but when calculating strength deficit of the affected shoulder compared to the healthy side no gender differences were noted. Significantly lower values were obtained for the affected shoulder than for the healthy side for all parameters but no significant differences between right and left shoulder were observed. There was a negative correlation between age and isokinetic parameters for the affected shoulder [r = -0.371 (p = 0.037) to r = -0.539 (p = 0.001)] and for the healthy side [r = -0.378 (p = 0.033) to r = -0.565 (p = 0.000)] but not for strength deficit of the affected side. We found no correlation between isokinetic data and rotator cuff tear size, pain duration, ROM or BMI. Isokinetic values did not differ with location of rotator cuff rupture, acromial shape or pain intensity. Conclusions: Isokinetic values differed between men and women and were lower on the affected side. With increasing age there was a decrease in strength. Pre-operative pain intensity and duration, ROM, acromial shape or size and location of rotator cuff rupture did not have any influence on isokinetic parameters.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P15-427 Pre-operative isokinetic testing does not predict long-term outcome of surgery in patients with rotator cuff rupture I. Ho¨fling1, L. Thorwesten2, U. Va¨a¨ta¨inen1, P. Sipola3, L. Niemitukia3, H. Kro¨ger1 1 Kuopio University Hospital, Orthopaedics and Traumatology, Kuopio, Finland, 2University of Mu¨nster, Sports Medicine, Mu¨nster, Germany, 3Kuopio University Hospital, Radiology, Kuopio, Finland Objectives: The purpose of the present study was to investigate the association between pre-operative isokinetic parameters and outcome of surgery in patients with rotator cuff rupture. Methods: A total of 35 patients (average age 58.2 ± 6.5 years, 16 females and 19 males) with symptomatic rotator cuff rupture were studied. Diagnose was confirmed with ultrasound and MRI. Preoperative isokinetic testing of both shoulders in flexion and extension was performed using a Lido Active isokinetic device. Peak torque (PT), average peak torque (APT) and total work done (TWD) at 80/s and 120/s were evaluated. Rotator cuff repair and acromioplasty were performed. At follow-up (5.7 ± 1.1 years after surgery) UCLA-scores, range of motion (ROM) and pain intensity were obtained. Results: We found no correlation between PT, APT, TWD (also when calculated as percentage of non-affected side) and post-operative UCLA-scores. Isokinetic values did not correlate with post-operative ROM or improvement in ROM compared to pre-operative values. There were no significant differences in isokinetic data between patients with no pain at follow-up and patients who still experienced some pain. Conclusions: Pre-operative isokinetic testing does not predict longterm benefit from surgery regarding UCLA-scores, ROM or postoperative pain in patients with rotator cuff rupture.

P15-451 The influence of ultrasound-guidance in the rate of success of acromioclavicular joint injection P. Borbas1, T. Kraus2, H. Clement1, W. Grechenig1, A.M. Weinberg3, N. Heidari4 1 Medical University of Graz, Department of Traumatology, Graz, Austria, 2Medical University of Graz, Department of Pediatric Surgery, Pediatric Orthopaedic Unit, Graz, Austria, 3Univ. Klinik fu¨r Kinderchirurgie, Medizinische Universita¨t Graz, Graz, Austria, 4 Royal London Hospital, Department of Trauma and Orthopaedics, London, United Kingdom Objectives: Injections of the acromioclavicular joint (ACJ) are performed routinely on patients with ACJ arthritis, both diagnostically and therapeutically. The aim of this prospective controlled study was to estimate the frequency of successful intra-articular ACJ injections with the aid of sonographic guidance versus non guided ACJ injections. Methods: A total of 80 cadaveric AC joints were injected with a solution containing methylene blue and subsequently dissected to distinguish intra- from peri-articular injections. In 40 cases the joint was punctured with sonographic guidance, as well as 40 joints were injected in the control group without the aid of ultrasound. Results: The rate of successful intra-articular ACJ injection was 90% (36 of 40) in the guided group and 70% (28 of 40) in the non-guided group. Ultrasound was significantly more accurate for correct intraarticular needle placement (p = 0.025). Conclusions: The use of ultrasound-guidance significantly improves the success rate in ACJ injection and we recommend it for therapeutic ACJ injections in the routine clinical practice.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P15-460 Primary stability of a new ultrasound assisted suture anchor system for rotator cuff reconstruction in comparison to the primary stability an established benchmark suture anchor system in osteopenic and healthy human humeri M.F. Gu¨lecyu¨z1, M. Pietschmann1, C. Schro¨der1, P.E. Mu¨ller1 1 Ludwig-Maximilians University, Campus Grosshadern, Department of Orthopaedic Surgery, Munich, Germany Objectives: The primary stability of a new ultrasound assisted rotator cuff suture anchor system ‘‘Sombrero’’ (diameter: 3.8 mm) (SportWelding GmbH) loaded with USP 2 FiberWire was tested in osteopenic and healthy humeri and compared to the primary stability of the benchmark anchor BioCorkscrew FT 5.5 mm (Arthrex GmbH), also loaded with either a USP 2 FiberWire  or TigerWire suture. Methods: The suture anchors were tested in 10 osteopenic and 4 healthy humeri with an average age of 78 and 76 years respectively and a bone mineral density (BMD) of 54.74 mg calcium hydroxyapatite per milliliter (Ca2+-HA/ml) and 117.9 Ca2+-HA/ml. To simulate postoperative conditions a cyclic loading protocol was performed. The maximum failure load (Fmax), the system displacement and the modes of failure were recorded. Results: The Fmax of the ‘‘Sombrero’’ with 217.5 N (SD ± 78.68) in osteopenic humeri was almost equivalent to the Fmax of the BioCorkscrew FT 5.5 mm with 220.8 N (SD ± 67.85). In healthy humeri, Fmax values of 279.2 N (SD ± 57.92) and 245.8 N (SD ± 60.03) were recorded for the ‘‘Sombrero’’ and Bio-Corkscrew FT 5.5 mm anchor systems respectively. The system displacement of the ‘‘Sombrero’’ and Bio-Corkscrew FT 5.5 mm measured 0.37 mm (SD ± 0.12) and 0.57 mm (SD ± 0.18) respectively in osteopenic humeri and 0.49 mm (SD ± 0.36) and 0.71 mm (SD ± 0.24) in healthy humeri. The modes of failure observed during this testing were anchor dislocations and suture dislocations; suture ruptures did not occur. The ‘‘Sombrero’’ suture anchor system failed solely due to anchor dislocations. Conclusions: This study shows that the ‘‘Sombrero’’ suture anchor system in combination with USP 2 FiberWire retrieves similar maximum failure loads and system displacement values as the established Bio-Corkscrew FT 5.5 mm anchor system. The primary stability of the Sombrero and Bio-Corkscrew FT 5.5 mm anchor systems seems to be independent of the bone mineral quality since there were no significant differences in the maximum failure loads and system displacement values.

P15-467 The clinical and radiographical results of reverse total shoulder arthroplasty with eccentric glenosphere N. Mizuno1, G. Walch1 1 Centre Orthope´dique Santy, Lyon, France Objectives: Scapular notching is a common concerning finding after reverse total shoulder arthroplasty (RSA). Eccentric glenospheres have recently been developed in order to potentially prevent notching. The purpose of this study was to evaluate the clinical and radiologic results of RSA with an eccentric glenosphere and compare the incidence and the severity of scapular notching with a concentric glenosphere. Methods: A prospective evaluation was performed of 57 consecutive RSA performed over a 2 year period. At a minimum of 2 years postoperative, 47 RSAs with a mean 30.4 months follow-up were evaluated clinically and radiographically and compared to a historical control group of concentric glenospheres performed with the same implant by the same surgeon. Results: The mean Constant score significantly increased (from 32.4 to 71.8) postoperatively (p \ 0.0001). Active forward flexion and

S115 external rotation also significantly increased (p \ 0.0001). Overall, scapular notching was present in 19 shoulders (40.4%). Grade 1 notching was observed in 13 shoulders (27.7%), grade 2 in 5 shoulders (10.6%), grade 3 in 1 shoulder (2.1%), and grade 4 in 0 shoulders. There was no significant difference in the incidence (p = 0.289) of notching between the eccentric and concentric glenospheres. However, the severity of notching was significantly decreased (p = 0.011) with an eccentric glenosphere. The postoperative Constant score was not significantly different between patients with and without notching (p = 0.651). Conclusions: A Grammont type RSA with eccentric glenosphere can result in good clinical outcomes. An eccentric glenosphere does not prevent notching, but decreases the severity of scapular notching at early follow-up.

P15-531 The feasibility and results of arthroscopic removal of proximal humeral locking plates and glenohumeral arthrolysis after proximal humeral fractures A. Maqdes1, B. Levy2, T. Bauer3, P. Hardy4 1 West Paris University, Boulogne, France, 2West Paris University, Orthopedics Department, Boulogne, France, 3Ambroise Pare´ University Hospital, Orthopedic Surgery, Boulogne, France, 4 West Paris University, Ambroise Pare Hospital, Orthopaedic Surgery, Boulogne, France Objectives: The objective of this study is to assess the feasibility and results of arthroscopic removal or proximal humeral locking plates and glenohumeral arthrolysis after proximal humeral fractures. Methods: The study is a retrospective non-comparative study that includes eleven patients who benefited from ORIF for a proximal humeral fracture. The average age of the operated patients was 55 (36–74) years. Inclusion criteria were a consolidated fracture at the time of removal, significant passive and glenohumeral stiffness after 6 months of well-guided physiotherapy program, arthrogenic screw(s), and osteonecrosis of the humeral head leading to considerable pain. A pre and postoperative clinical exam and constant score at 6 months were conducted for each patient. A detailed description of the surgical technique is depicted. The mean follow-up was 17.7 months (6–84). Results: The surgery was successfully performed in all attempted cases with an average operative time of 105 min (100–110). Preoperative and postoperative analysis showed a visual analogue pain score decrease of 60% (4.7–2.8). Postoperative mobility testing showed, on average, an improvement of 28 of abduction (77–105), 13 of forward flexion (85–98), 17 of external rotation (15–32) and an internal rotation that improved from the L3 to T12 level. The average change of the Constant score was 18 points. The average hospital stay was 1.2 days (1–2). Sports were restarted at around 6 months postoperatively. Amongst the 11 patients operated, no infections or wound dehiscence occurred. Conclusions: Arthroscopic removable of proximal humeral locking plates is a seductive, minimally invasive technique that shows a lot of promise. This technique permits the achievement of the objective in combination to a glenohumeral diagnostic arthroscopy and a glenohumeral arthrolysis if required. This technique is advantageous because it can be performed in a day case fashion with minimal patient hospitalization and hospital expenses. Cosmetic benefits are also non negligible as only a few keyhole incisions are necessary. This study confirms the feasibility and beneficial results of an arthroscopic technique for the removal of proximal humeral locking plates combined with a glenohumeral arthrolysis in patients with prior proximal humeral fractures.

123

S116 P15-558 Margin convergence revisit: arthroscopic partial repair of irreparable large to massive rotator cuff tears Y.-M. Chun1, S.-J. Kim1 1 Arthroscopy and Joint Research Institute, Yonsei University, Orthopaedic Surgery, Seoul, Republic of Korea Objectives: The aim of the current study is to evaluate the outcome of arthroscopic partial repair and margin convergence of irreparable large to massive rotator cuffs. Methods: Between January 2003 and July 2007, 27 patients that met the inclusion criteria underwent arthroscopic partial repair and margin convergence of irreparable large to massive rotator cuff tears. An irreparable tear was defined as tear with a minimum anterior to posterior width of 3 cm or larger, where it was not feasible to completely cover the humeral head with the cuff at the time of the surgery. Results: The preoperative mean tear size was 42.1 ± 6.2 mm. The mean size of the post-surgery residual defect in the repaired tendon along the medial margin of the greater tuberosity was 12.0 ± 5.5 mm. All shoulder scores showed improvement. The simple shoulder test improved from 5.1 ± 1.2 to 8.8 ± 2.1 (p \ 0.001); the Constant score from 43.6 ± 7.9 to 74.1 ± 10.6 (p \ 0.001); and the UCLA score from 10.5 ± 3.0 to 25.9 ± 5.0 (p \ 0.001). Both Constant and UCLA shoulder scores also showed an inverse correlation with defect size. We compared muscle strength between the affected and contralateral side, and found that the strength of the affected side was not restored to the same level as the contralateral side (p \ 0.001). Conclusions: Arthroscopic partial repair and margin convergence showed satisfactory short-term outcomes in irreparable large to massive rotator cuff tears. Thus, it is suggested that, even in large to massive tears that appear irreparable, attempting to repair it as much as possible to so as to possibly convert it into a functional rotator cuff tear by recreating a balanced forced couple can be helpful in reducing pain, and improving functional outcomes.

P15-572 Evaluation of arthroscopic subacromial decompression using a 2-portal technique M. Marican1, L.W. Lim1, A.H.C. Tan1 1 Singapore General Hospital, Orthopaedic Surgery, Singapore, Singapore Objectives: To assess the feasibility, safety and effectiveness of a 2-portal arthroscopic subacromial decompression (ASD) technique for impingement syndrome. Methods: A prospective study performed over 12 months for patients who had failed at least 6 months of conservative management. Patients with rotator cuff tears, labral tears and calcifying tendonitis were excluded. The ASD was performed by a single surgeon using posterior (camera) and anterior (debrider) arthroscopic portals, obviating the need for a lateral portal. Preoperative and postoperative (mean 2.5 years) evaluation using UCLA Shoulder scores, Modified Simple Shoulder scores and patients’ level of satisfaction with surgery. Results: 30/31 (97%) patients were satisfied with their surgery. There were zero complications or recurrence of symptoms. Mean preoperative (3.52/12) versus post-operative Simple Shoulder score (11.52/ 12) showed a 227% improvement (P \ 0.01). Mean preoperative (12.1/35) versus postoperative UCLA Shoulder scores (31.5/35) showed a 160% improvement (P \ 0.01). We found debriding off the anterior portal allowed better delineation and appreciation of the acromial spur. Conclusions: This newly described 2-portal ASD technique for isolated shoulder impingement has proven to be feasible, safe and effective.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P15-579 Locked plate fixation of proximal humerus fractures with and without the medial humeral calcar: a biomechanical evaluation M. Safran1, G. Abrams1, D.P. Lindsey2, E. Cheung1, N. Giori3 1 Stanford University, Orthopaedic Surgery, Redwood City, United States, 2VA Palo Alto Health Care System, Biomechanics, Bone and Joint Center of Excellence, Palo Alto, United States, 3Stanford University, Orthopaedic Surgery, Bone and Joint Center of Excellence, Palo Alto, United States Objectives: Locked plate fixation of proximal humerus fractures has gained wide-spread acceptance. Recent investigations have shown reduction of the medial calcar or screw support in this region is important for preventing varus collapse and subsequent failure. No studies to date have biomechanically determined whether inferior-medial screw support alone is sufficient to prevent varus collapse. Methods: Eighteen (9 pairs) of fresh frozen cadavers were utilized and underwent dual-emission X-ray absorptiometry (DEXA) scanning prior testing. Each pair was randomly assigned to either the intact medial calcar (+MC) or missing medial calcar (-MC) group. Both groups underwent a surgical neck osteotomy with the -MC group having the additional removal of a triangular wedge of bone (measuring 1 cm at the base) at the medial calcar area. Fractures were then reduced and fixated with a Synthes 3-hole 3.5 mm proxial humerus locking plate. Drilling was performed to the subchondral surface and fluoroscopic imaging was utilized to ensure correct positioning of all screws. Specimens then underwent cyclic varus loading at 7.5 Newton-meters (N-m) at 2 Hz for 5,000 cycles or until failure, with surviving specimens undergoing load to failure. Specimen survival, varus angulation, screw penetration through articular cartilage, and distance from screw tip to subchondral bone (screw subchondral distance—SSD) on fluoroscopic imaging was recorded. Logistic regression, linear regression, Chi-squared, and Student’s t test analyses were utilized with an alpha value of 0.05 as significant. Results: Average specimen age was 71 years (range 54–88) with 4 males and 5 females. There was no significant difference in bone mineral density of the paired specimens (p [ 0.8). Eight of 9 (89%) -MC specimens failed during cyclic loading versus 3 of 9 (33%) +MC specimens (p \ 0.05). Total cycles endured were greater in the +MC group (3,587 vs. 1,606; p = 0.07) while varus angulation was significantly increased in the -MC group (23.4 vs. 11.6; p \ 0.05). There were no cases of screw penetration though the articular surface in any case. Load to failure for the intact +MC specimens was 184.5 N. Regression indicated that there was a trend towards significance in correlating SSD and varus angulation during cyclic loading (p = 0.08) and SSD distance of 5 mm correlated with a 7.8% failure rate in +MC specimens. Conclusions: The -MC specimens has a 2.59 failure rate versus the +MC specimens. This suggests that the inferior screws in the typical locking plate design do not fully compensate for medial calcar comminution or displacement. In these circumstances, additional medial support may be warranted to reduce the risk of varus collapse. Furthermore, decreased SSD distance correlated with improved biomechanical stability without increasing the risk of screw penetration through the articular surface.

P15-599 Outcome of a partial cap resurfacing implant for humeral head defects in patients with shoulder instability N.C. Frisch1, P. Kodali2, A. Miniaci3, M. Jones4 1 Cleveland Clinic, Orthopaedic Surgery, Shaker Heights, OH, United States, 2University of Texas Health Science Center, Orthopaedic

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Surgery, Houston, TX, United States, 3Cleveland Clinic, Department of Orthopaedic Surgery, Cleveland, United States, 4Cleveland Clinic, Cleveland, United States Objectives: Osseous defects of the humeral head play a significant and well-documented role in the pathoanatomy of chronic shoulder instability. While various methods to address these bony lesions have been described, one newer method is the use of the HemiCAP (Arthrosurface, Inc. Franklin, MA, USA) partial humeral head resurfacing arthroplasty. The purpose of this study is to report the early outcomes of HemiCAP implants used for the treatment of Hill–Sachs lesions in anterior shoulder instability and for the treatment of reverse Hill–Sachs lesions in posterior shoulder instability. Methods: We performed a retrospective cohort study using prospectively collected pre-operative data. Our surgical outcomes database was used to identify patients with a diagnosis of shoulder instability that underwent placement of a HemiCAP prosthesis. Patients with less than 6 months follow-up were excluded. Patients were contacted by mail and telephone to complete a brief survey and an outcomes questionnaire. Complication rates were determined including infection, reoperation, and dislocation. Statistical analysis of preoperative and postoperative outcomes scores was performed using paired t test with significance of 0.05. Results: Twenty (20) patients meeting study criteria were identified, with a mean follow-up time of 21.5 months (range 6–49 months). 16/20 patients had undergone HemiCAP placement for a Hill–Sachs lesion in anterior shoulder instability, while 4/20 patients had undergone HemiCAP placement to address a reverse Hill–Sachs lesion. None of the 20 patients had suffered a repeat dislocation since surgery. In the anterior instability patients, the postsurgical outcomes scores revealed a significant decrease in the mean musculoskeletal review of systems score of 4.07 (p \ 0.0001) and a significant increase in the mean SF-12 physical composite score of 10.89 (p = 0.003). Outcomes scores in reverse Hill–Sachs patients revealed similar improvements, but lacked statistical significance due to N = 4. Conclusions: The HemiCAP partial resurfacing implant is an off-theshelf device that results in a near anatomic humeral head reconstruction and proves effective for addressing humeral head defects both in anterior and posterior shoulder instability.

P15-681 The applicability of rigid intramedullary devices for clavicle midshaft fractures fracture fixation A. Bachoura1, S. Kamineni2 1 University of Kentucky, Orthopaedics, Lexington, United States, 2 University of Kentucky, Orthopaedics and Sports Medicine, Lexington, United States Objectives: The purpose of this article is to investigate the morphological safety and applicability of rigid intramedullary fixation of midshaft clavicle fractures, by analyzing the pertinent clavicle anatomy using three-dimensional computer simulation. Methods: Computed tomography scan was used to scan 22 skeletonized clavicles. A computer program was used to replicate middle segment fracture fixation by fitting a cylindrical passage within the clavicle, in the area that intramedullary devices normally cross during surgery. The cylindrical corridor crossed the fracture line on both sides, and the number of cortical diameters that were bypassed was noted. It was assumed that one to two cortical diameters had to be bypassed to achieve adequate fixation. The medial and lateral exit points of the cylindrical passage were then measured and described in relation to the sternoclavicular and acromioclavicular ends respectively. Results: Simulation revealed that 15/22 clavicles could bypass two cortical diameters on either side of the midline fracture, 6/22 clavicles

S117 could make it past one cortical diameter medial to the fracture line, and 1/22 clavicles, could not bypass any cortical diameters medial to the fracture line. The medial exit point of the cylindrical passage was anterior in 20/22 cases and on average 44.2 mm lateral to the sternoclavicular end. The lateral exit point of the cylindrical passage was posterosuperior in 16/22 cases and on average 26.5 mm medial to the acromioclavicular end. Conclusions: In select cases the presented intramedullary technique appears to be a morphologically safe and effective method of fixation. Clinical results will be necessary to validate these findings.

P15-701 Predictors of success of nonoperative treatment for full-thickness rotator cuff tears: a multicenter cohort study W. Dunn1, J.E. Kuhn2, MOON Shoulder Group 1 Vanderbilt University Medical Center, Orthopaedics/Sports Medicine and Internal Med/Public Health, Health Services Research Center, Nashville, United States, 2Vanderbilt University Medical Center, Orthopaedic Surgery and Rehabilitation, Nashville, United States Objectives: Full thickness rotator cuff tears (RCT) are extremely prevalent, yet \5% come to surgical repair. In those that undergo surgical repair and postop rehabilitation, reported healing rates have ranged from 13 to 69%, yet patients who have failed repairs report good outcomes and satisfaction with treatment. These data suggest that physical therapy may be effective in treating the symptoms of some patients with full thickness tears. The purpose of this study is to determine predictors of failure of nonoperative treatment using a multicenter prospective cohort study design. Our hypothesis was that younger, more active patients would be more likely to fail nonoperative treatment and request surgery. Methods: A prospective multicenter cohort study design was used. Inclusion criteria included all patients with full thickness RCTs seen on MRI without other disease states. Baseline data from this cohort was used to examine risk factors for failing a standard rehab protocol. Subjects that ultimately underwent rotator cuff surgery were defined as failing nonoperative treatment. A Cox proportional hazards model was fit to determine what baseline factors predicted surgery, or failure of rehab. The dependent variable was time to surgery; independent variables were tear severity and baseline patient factors (age, activity level, BMI, sex, VAS pain level, education, handedness, comorbidities, duration of symptoms, strength, employment, smoking status, and patient expectations). Results: Of 433 subjects enrolled median age was 62; 49% were female. The dominant shoulder was involved in 69% of the cohort. The median baseline VAS was 4.4. The proportion of subjects with symptoms\1 month was 8%, 1–3 months was 22%, 4–6 months was 20%, 7–12 months was 14%, and [1 year was 37%. Isolated supraspinatus tears were found in 73%, 21% had tears involving the supraspinatus and infraspinatus ± teres minor, and 6% had subscapularis involvement. Tendon retraction was minimal in 48%, midhumeral level in 34%, glenohumeral in 13%, and 5% had retraction to the glenoid. Subjects that had surgery declared themselves early with a median f/u time of 120 days (interquartile range (IQR): 72, 176), while those that responded to rehab contributed a median of 731 days of f/u time (IQR: 366, 739). Multivariable modeling, adjusted for the independent variables listed above, identified patient expectations regarding physical therapy (p \ 0.0001) as the strongest predictor of surgery. Younger age (p = 0.042), higher activity level (p = 0.011), and not smoking (p = 0.023) were also significant predictors of having surgery. Conclusions: Severity of cuff pathology (size of tear, retraction), pain level, and weakness were not associated with failure of rehab. The strongest predictor was low patient expectation about physical

123

S118 therapy. Other factors found to affect having surgery were higher activity level, younger age, and not smoking.

P15-757 Injury of the suprascapular nerve during Latarjet procedure: an anatomic study A. La¨dermann1, P.J. Denard2, S.S. Burkhart2 1 La Tour Hospital, Orthopaedic Surgery and Traumatology, Meyrin, Switzerland, 2The San Antonio Orthopaedic Group, San Antonio, United States Objectives: Although suprascapular nerve palsy following Latarjet procedure has been described, no study to our knowledge has examined the relationship between the specific exit point of screws securing the coracoid graft during the Latarjet procedure and the major and minor branches of the supra scapular nerve. The purpose of this study was to evaluate the relationship between the specific exit point of screws securing the coracoid graft during Latarjet procedure and the suprascapular nerve. Methods: Ten fresh frozen shoulder specimens were dissected after having performed an open Latarjet procedure. Results: The mean distance from the posterior exit site of the superior screw to the suprascapular nerve at the base of the scapular spine was only 4 mm. Two of the superior screws were directly in contact with the major branch of the suprascapular nerve and two screws were also in contact with minor branches of the suprascapular nerve. As for the inferior screw, we noticed contact with the major branch in one case and with minor branches of the suprascapular nerve in six cases. In the axial plane, the screws were not in contact with the suprascapular nerve if the angle relative to the glenoid was less than or equal to 10. Conclusions: The proximity of the suprascapular nerve to the posterior glenoid rim puts this nerve at risk during insertion of the screws used for the Latarjet procedure.

P15-780 Retrospective comparison of arthroscopic, mini-open and open rotator cuff repair by clinical, radiological and biomechanical measures P. Lubiatowski1, P. Kaczmarek2, P. Ogrodowicz3, M. Breborowicz3, M. Dzianach2, L. Romanowski3 1 University of Medical Sciences in Poznan, Department of Orthopaedics and Hand Surgery, Rehasport Clinic, Poznan, Poland, 2 Rehasport Clinic, Poznan, Poland, 3University of Medical Sciences in Poznan, Department of Orthopaedics and Hand Surgery, Poznan, Poland Objectives: Technique of rotator cuff (RC) repair may be important for treatment prognosis. The aim of the study was to review and compare outcome of RC repair with three approaches: open, miniopen and arthroscopic. To our knowledge no such paper comparing all three techniques have been published so far. The advantage of this study is also very detailed evaluation including clinical, ultrasound and bio-mechanical testing. Methods: 124 RC repairs (42 open-OR; 34 arthroscopically assisted mini-open-AAMO and 48 arthroscopic-AR) in 111 patients treated in our department in the years 2002–2008. Average age of 56 (40–80). Follow-up time was 1–7 years. Results obtained from clinical evaluation (ASES, SST and UCLA scores, pain), radiological (ultrasound scan for RC integrity) and biomechanical testing (isometric and isokinetic). Results: Results of clinical and bio-mechanical assessment are included in the Tables 1 and 2. Retear rate by ultrasound scan was 38% in OR group, 29% in AAMO group and 19% in AR. They included both partial and complete tears.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Table 1 Result of RC repair: clinical measures Measure

OR

UCLA

26.9 ± 5.5

SST ASES

AAMO

AR

30.7 ± 4.6

31 ± 5.7

8±4

10 ± 2.3

10.7 ± 2.5

68.7 ± 24.3

81.7 ± 18.6

87 ± 16.3

Pain level

3.5 ± 2.8

2.1 ± 2.4

1.4 ± 1.9

Active abduction range ()

140 ± 44.1

168.9 ± 19.5

174.8 ± 16.6

Active external rotation range ()

44.5 ± 22

56.8 ± 22.4

59.2 ± 14.6

Table 2 Result of RC repair: biomechanical Measure

OR

AAMO

AR

Isometric flexion

38.5 ± 25.3 56.1 ± 24.7 70.6 ± 41.2

Isometric abduction

28.6 ± 22

Isometric external rotation

61.6 ± 31.5 68.3 ± 29.9

External rotation peak torque at 180/s

10.1 ± 8.7

12.2 ± 5.9

15.5 ± 8.3

External rotation peak torque at 90/s

15.8 ± 8.5

18.8 ± 5.6

21.1 ± 8.4

51.8 ± 30.1 69.7 ± 30.7 74 ± 43.2

Results were significantly better when AR was compared to open repair in all measures. No significant difference was found between AR and AAMO. Although AAMO seemed better then OR, statistical significance was shown only in some outcomes (isometric flexion, active external rotation, retear rate). Conclusions: Operative treatment allows for significant improvement of patient satisfaction and shoulder function. Less invasive treatment improves final outcome in multiple respects: patient’s comfort and living functionality, shoulder function and biomechanics and decreased retear rate. This has been particularly significant for arthroscopic repair when compared to open approach.

P15-833 Single row versus double row arthroscopic rotator cuff repair: clinical and 3 T MR arthrography results: a retrospective study C. Tudisco1, S. Bisicchia1, E. Savarese2, R. Fiori3, D.A. Bartolucci3, S. Masala3 1 University of Rome Tor Vergata, Orthopaedic Surgery, Rome, Italy, 2 San Carlo Hospital, Orthopaedic Surgery, Potenza, Italy, 3University of Rome Tor Vergata, Rome, Italy Objectives: To evaluate the clinical and radiological results of two groups of patients operated on for a medium or large rotator cuff tear with two different techniques. Methods: We retrospectively reviewed two groups of patients operated on arthroscopically for a medium or large rotator cuff tear. First group consisted of 20 patients operated on with a single row (SR) technique, second group consisted of 20 patients operated on with a double row (DR) technique. All patients were evaluated at a minimum of 2 years after surgery. Clinical evaluation consisted of Constant and Murley Score (CMS) and Simple Shoulder Test (SST). The structural integrity of the rotator cuff was evaluated with a 3 T MR Artrhography (MRA). MR imaging at 3 T, especially with intraarticular contrast medium (MRA), demonstrated better diagnostic performance than 1.5 T in the musculoskeletal setting, with a consequent improvement in bone and soft tissue detail. Primary end point

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 was re-tear rate at 3 T MRA. Secondary end points were CMS and SST. We also retrospectively compared surgical time and implant expense in both groups. An unpaired t test was used to compare both groups with respect to objective outcomes. Categorical variables were compared using a Chi-square test between both groups. For all statistical tests, the alpha level was set at 0.05. Results: In SR group there were 13 men and 7 women, the average age at follow-up was 66 ± 8 (range 47–78) years. The mean surgical time was 92 (range 73–118) minutes. The mean implant expense was 400 (range 250–600) euro. The mean follow-up was 30.0 ± 5.0 (range 25–40) months. The mean CMS and the mean SST scores were 70 ± 9 (range 58–85) and 9.4 ± 1.7 (range 6–12) points respectively. A re-tear was observed in 12 patients (60%). In DR group there were 12 men and 8 women, the average age at follow-up was 63 ± 7 (range 57–73) years. The mean surgical time was 104 (range 85–136) minutes. The mean implant expense was 600 (range 450–800) euro. The mean follow-up was 28.9 ± 2.3 (range 24–32) months. The mean CMS and the mean SST scores in DR group were 67 ± 15 (range 37–89) and 10.1 ± 2.0 (range 7–12) points respectively. A re-tear was observed in 5 patients (25%). There was a statistically significant difference in surgical time (p \ 0.05) and in implant expense (p \ 0.05) between the two groups. There were no statistical differences between the two groups in mean CMS and SST scores at follow-up (p [ 0.05). There was a statistically significant lower retear rate in DR group (Chi-square = 5.01; p = 0.02). Conclusions: To the best of our knowledge this is the first report on a 3 T MRA in the evaluation of two different techniques of rotator cuff repair. DR repair provided a statistically significant lower re-tear rate, despite no difference in clinical outcomes. We think that leakage of the contrast medium is due to an incomplete tendon-to-bone sealing that is not a re-tear. This phenomenon could have important medicolegal implications.

P15-840 Real-time sonoelastography of the rotator cuff C. Tudisco1, S. Bisicchia1, M. Stefanini2, M. Antonicoli2, G. Simonetti2 1 University of Rome Tor Vergata, Orthopaedic Surgery, Rome, Italy, 2 University of Rome Tor Vergata, Rome, Italy Objectives: To report on radiological findings of Real-time sonoelastography (RTSE) of the rotator cuff in patients affected by unilateral rotator cuff tear, and to determine if an association exists between clinical and radiological findings. Methods: Fifty patients (32 male, 18 female) were prospectively enrolled in the study according to inclusion and exclusion criteria listed in Table 1. All the patients were evaluated clinically with Constant and Murley score (CMS), Simple Shoulder Test (SST), Quick DASH and VAS for pain. Subsequently a RTSE of both shoulders was performed to determine the quality of the tendons. RTSE shows tissues with different elasticity in different colours, from red (soft) to blue (hard). RTSE was graded as grade 1 B 25% red

Table 1 Inclusion and exclusion criteria Inclusion criteria

Exclusion criteria

Small tears (maximum Snyder C2) Massive retracted tears No retraction (Patte A)

Associated subscapularis tendon tears

Degenerative or traumatic tears

Previous infection of the shoulder

LHBB disease

Degenerative osteoarthritis

Written informed consent

Rheumatoid arthritis

S119 colouring, grade 2 = 25–50% red colouring, grade 3 = 50–75% red colouring, and grade 4 C 75% red colouring. Results: The mean age of the patients was 62 (±5 years), the dominant arm was involved in 36 patients. The mean CMS was 61 (±5) points, the mean SST was 7 (±7) points, the mean Quick DASH was 3 (±1) points, the mean VAS for pain was 8 (±1). RTSE showed 15 grade 1 tendons, 24 grade 2, 8 grade 3 and 3 grade 4. There was a statistically significant association between clinical outcome and the grade of tendon quality at RTSE (p \ 0.05). For all statistical tests, the alpha level was set at 0.05. Statistical analyses were performed with SPSS v.15.0 (SPSS Inc., an IBM Company, Chicago, IL, USA). Conclusions: The quality of the rotator cuff tendons evaluated by RTSE strictly correlate with the clinical results of the patients.

P15-871 Subscapularis tear, its modified lafosse classification and its arthroscopic repair V. Sharma1, A. Babhulkar2 1 Bhagat Chandra Hospital, Orthopaedics, New Delhi, India, 2 Dinanath Mangeshkar Hospital, Shoulder & Sports Injuries, Pune, India Objectives: Anterosuperior tears are now more frequently being reported. In this study the arthroscopic repair of the subscapularis tendon, its surgical technique and a modified classification is described with the aim to highlight the advantages to do this procedure arthroscopically. The aim of this study was to classify these tears with a modified Lafosse classification and describe the surgical technique of arthroscopic subscapularis tears with a short term follow-up. Methods: A retrospective study was carried out from Nov 2009 to Sept 2011. A total of 20 patients 13 males and 12 females, with an average age of 53.5 years. Most of the cases were anterosuperior tears with only 6 isolated subscapularis tears. All the patients underwent preoperative X-ray, MRI and ultrasonography. The tears were classified using a modification of the classification proposed by Lafosse et al. which is based on intraoperative arthroscopic findings. This classification was modified to include the tears involving the inferior fibres with intact superior fibres and longitudinal tears observed in our study. Surgical technique for the repair is described for a better footprint repair. Early mobilization was practised with elastic resistive exercises been introduced at 6 weeks for all the patients. Results: All 20 patients were evaluated for the repair at 3 month and 6 months. Mean average follow-up of 9.9 months (range 6–22 months). All patients improved there UCLA scores post operatively and all were satisfied within the short term follow-up (P \ 0.05). Conclusions: Clinical diagnosis of Subscapularis tear is difficult to make and is infrequently reported, and can be missed on routine arthroscopic examination (especially partial tears) due to the high dependency on the MRI reports which are under reported, unless a high degree of clinical suspicion is practised. Arthroscopic repair of the subscapularis tear results in significant improvement of the shoulder function and is the best modality to treat these tears.

P15-900 The possibility of supraspinatus and infraspinatus separation and suture in cases of massive rotator cuff tears T. Komatsu1, Y. Hayashi1, O. Takeda1, M. Akita1, T. Iwasaki1, J. Ito1 1 Aomori Prefectural Central Hospital, Orthopedics, Aomori, Japan Objectives: Suture at the anatomically correct site is important in preserving the function of the shoulder joint in arthroscopic rotator cuff suture. In recent years there have been reports in anatomical studies of the infraspinatus being stretched from back to front and attached to a large portion of the greater tubercle or the supraspinatus

123

S120 being attached to the front of the greater tubercle. There are likely to be tears of two or more tendons including the supraspinatus and infraspinatus in cases of massive rotator cuff tears, although during the actual operation there are many cases in which the supraspinatus and infraspinatus cannot be separated to confirm this. We examined the proportion of massive rotator cuff tear cases involving two or more tendons, where the supraspinatus and infraspinatus could be separated and the tears confirmed and sutured. Methods: We separated cases of massive rotator cuff tears undergoing arthroscopic rotator cuff suture into those where the supraspinatus and infraspinatus could be separated, the tears confirmed and sutured (group A) and cases where the two tendons could not be separated nor confirmed (group B). In group A the supraspinatus was sutured with a single-row or double-row method and the infraspinatus was sutured with a double-row or suture-bridge method. In group B, the entire torn area was sutured with a double-row or suture-bridge method. The Japanese Orthopedic Association Shoulder (JOA) score were examined in both groups for cases where the number of cases, gender and age could be observed for 6 months or longer. Results: Of the 93 cases of arthroscopic rotator cuff suturing performed at this hospital from January 2009 to December 2010, 32 cases were massive tears. There were 12 patients in the A group (7 male, 5 female) with an average age of 68 and average JOA score of 94.8 (at an average of 11.3 months after surgery). In this group there were two cases complicated by tears of the subscapularis tendon and one case where complete suture was impossible. There were 20 patients in the B group (12 male, 8 female) with an average age of 69 and average JOA score of 87.4 (at an average of 11.5 months after surgery). In this group there were three cases complicated by tears of the subscapularis tendon, and one case where complete suture was impossible. There was no significant difference in age and gender between the two groups, however the JOA score of group A was significantly (0.01 [ p) higher. Conclusions: The supraspinatus and infraspinatus could be separately confirmed and sutured in arthroscopic technique in 12 out of the 32 cases of massive rotator cuff tears (37.5%). Although these are only short-term results, JOA scores were significantly higher in the group where separation and suture was possible.

P15-976 Mid-term outcome of arthroscopic revision repair of massive and non-massive rotator cuff tears A. La¨dermann1, P. Denard2, S. Burkhart2 1 La Tour Hospital, Orthopaedic Surgery and Traumatology, Meyrin, Switzerland, 2The San Antonio Orthopaedic Group, San Antonio, United States Objectives: No information is available regarding the outcome of arthroscopic revision of massive rotator cuff tears, at mid-term follow-up. The purpose of the present study was to evaluate the midterm functional outcome of arthroscopic revision rotator cuff repair and compare the outcomes of non-massive and massive rotator tears. Methods: A retrospective review was performed of patients over a 10 year period who underwent an arthroscopic revision rotator cuff repair. The cohort was divided into two groups based upon tear size (non-massive and massive tears). Results: The cohort consisted of 21 non-massive tears and 53 massive tears with a mean follow-up of 63 months. The two groups had similar baseline characteristics. In the overall cohort, following arthroscopic revision repair there was a significant reduction in pain (p \ .001), and increase in active forward elevation (p = .003) and functional outcome by American Shoulder and Elbow Surgeons

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 (p \ .001) and University of California, Los Angeles (UCLA) (p \ .001) scores. The rate of patient satisfaction was 78%. There was no significant difference between the two groups (non-massive vs. massive) in postoperative forward elevation, pain, or functional outcome. A poor functional outcome according to the UCLA score was associated with female sex (p = .005), preoperative active forward flexion below 136 (p = .004), and preoperative pain score [5 (p = .002). Conclusions: The results of the present study suggest that arthroscopic revision rotator cuff repair may be a reasonable treatment option even in cases of massive retears. This technique can yield reliable improvements in both pain and function at an acceptably high rate in this difficult patient population.

P15-1015 Ultrasound-guided percutaneous needle aspiration of rotator cuff calcifications: a new technique P. Ciampi1, C. Scotti2, G. Peretti1, G. Fraschini1 1 San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy, 2University of Milan, Residency Program in Orthopaedic and Traumatology, Milan, Italy Objectives: The US-guided needle percutaneous treatment of rotator cuff calcific tendonitis is still debated and no gold standard procedure has been identified. The purpose of this study is to evaluate the clinical outcome of a new technique for ambulatory treatment of calcific tendonitis of the shoulder. Methods: We treated 50 patients affected by calcific tendinitis (40 females, 10 males, 38 right shoulders and 12 left shoulders) with an average age of 37.6 years, resistant to conservative treatment. Exclusion criteria were as follows: rotator cuff tears, acromion or A– C joint pathologies, humeral pathologies. All patients were evaluated with international scales (UCLA score, SPADI score and VAS scale). X-ray and ultrasound of the affected and contra lateral shoulder were performed before treatment, at 3 months and at 2 years from treatment (medium follow-up 15.5 months). Data were evaluated using T Test. Calcium deposits were classified into small (\0.5 cm), medium (0.5–1.5 cm) and large ([1.5 cm). The procedure was performed by the same orthopaedic operator using a 11.0-MHz linear transducer according to Hedtam technique. The ultrasound-guided procedure was performed with patient placed in Beach Chair decubitus. After the transducer and the skin of the shoulder had been cleansed, the calcification was localized with US, and local anesthesia in the subacromial space was done (5/6 cc.). One 18-gauge modified needle attached to a syringe was inserted into the calcification, with a continuous US monitoring. The calcific deposit was punctured frequently with the needle. The aspiration and injection were performed with mild pressure using a saline solution with another 18-gauge modified needle. The lavage was continued until the aspirate was free of calcic particles. The aspiration of calcific deposits was clearly visible with ultrasound during the procedure. Results: Before the procedure the rotator cuff calcifications had a mean area of 1.22 cm; after the procedure the mean area was 36 mm. Successful aspiration of calcium crystals was achieved in 35 patients. 2 cases required a second procedure which was necessary to remove the calcific deposits. We observed an improvement of SPADI score (29.8%), VAS scale (45.3%) and UCLA score (43.4%). For all patients the procedure was free of pain. No complications such as infection, bleeding or rotator cuff tears were observed after the procedure. Conclusions: Our findings indicate that our technique for the treatment of calcific tendonitis of the shoulder seems to be an effective therapy, less traumatic and less invasive than other treatments.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P15-1025 The importance of early passive motion in shoulder rehabilitation after arthroscopic rotator cuff repair R. Akmes¸ e1, K.I. Yıldız1, O. Tecimel1, Y.G. Bilgetekin1, C¸. Is¸ ık1, M. Bozkurt1 1 Ankara Atatu¨rk Training and Research Hospital, Orthopaedics and Traumatology, Ankara, Turkey Objectives: In this study, we try to show the effects of the early passive motion of the shoulder on pain and functional activity level after arthroscopic rotator cuff repair. Methods: This study included 17 patients (12 women, 5 men) who underwent arthroscopic repair of stage 2 and 3 rotator cuff tears. The mean age was 56 (42–64). No preoperative rehabilitation program was applied to patients. The passive range of motion was initiated at the 1st day after the surgery. At the postoperative 1st week passive range of motion in flexion and abduction was applied, rotations were forbidden. At the postoperative 2nd week passive range of motion in external rotation was added but internal rotation was forbidden. Passive internal rotation was initiated at the postoperative 3rd week. The active range of motion was applied at the end of postoperative 3rd week in full range of motions. Patients were evaluated preoperatively and for 12 weeks postoperatively. Pain was assessed by Visual Analog Scale (VAS), and functional activity level was assessed by The Disabilities of The Arm Shoulder and Hand (DASH) questionnaire. Results: There was significant difference between the preoperative and postoperative VAS and DASH scores in the patients (p \ 0.05). The pain at rest was significantly decreased at the postoperative 4th week. To evacuate the pain at rest the VAS score was decreased from 3.5 in preoperative period to 0.3 in postoperative 12th week. The VAS in activity was decreased from 5.6 in preoperative period to 1.2 at the postoperative 12th week. The postoperative functional activity was significantly better in the postopeartive 4th week which was an early time to gain these functions. DASH score was 72.4 in preoperative period. In the postoperative periods DASH decreased to 40.8 at postoperative 4th week; 24.6 at postoperative 8th week; 20.8 at postoperative 12th week. Conclusions: The rehabilitation protocol of shoulder with early passive motion after arthroscopic rotator cuff repair prevents both the pain and functional limitations by preventing the adhesions of inferior capsules which occurs with restrictions of the motions in the early periods. It is possible for patients to return daily living activities early with the help of early passive rehabilitation protocol without any complications.

P15-1109 Intrateser and intertester reliability of the cuffmeter: new device for rotator cuff thickness measurements F. Franceschi1, R. Papalia2, S. Vasta3, A. Palumbo3, N. Maffulli4, V. Denaro2 1 Campus Biomedico University, Orthopaedic and Trauma Surgery, Rome, Italy, 2Campus Biomedico University, Orthopaedic Surgery, Rome, Italy, 3University Campus Biomedico of Rome, Rome, Italy, 4 Queen Mary University of London, Barts and The London School of Medicine and Dentistry, Centre for Sports and Exercise Medicine, London, United Kingdom Objectives: Partial thickness rotator cuff tears is a high-incidence pathology. Many different treatment options are available but which is the best one is still controversial. Several studies reported that the successful treatment is related to the rate of tendon thickness involved. Ultrasonography (US) and Magnetic Resonance Imaging (MRI) are the most currently used diagnostic tools. However, accurate thickness measurements are difficult using US due to the variability of the slice plane position and orientation, because it is an operator-dependent technique and because of the patient physical

S121 characteristics. MRI is not available for all patients and the MRI tendon thickness measurement is less accurate than an anatomical one. We wished to propose new method to measure intra-operatively rotator cuff thickness. Methods: To validate this new device we measured tendon thickness by both US and cuffmeter, in 50 shoulder undergoing arthroscopy for other pathology (instability). Two testers (one experienced and the other one student surgeon) performed the measurements. Results: The mean thickness values were 7.55 (SD: 0.64) for US series and 7.8 (SD: 0.78) for intra-operative measurement obtained by the device. The ICC for intratester reliability for tester 1 ranged from 0.80 to 0.92 and for tester 2 ranged from 0.77 to 0.90. Intertester reliability between the two testers ranged from 0.93 to 0.99 for all trials. Conclusions: The study concluded that intratetser and intertester reliability of cuffmeter measurements were good to excellent. Scoring the cuffmeter device measurement was consistent between the experienced and student surgeon.

P15-1151 Effectiveness of the hyaluronic acid in the different stages of the evolutive cuff pathology: a perspective study A. Busilacchi1, D. Enea1, S. Cecconi1, A. Gigante1 1 Polytechnic University of Marche, Orthopedics, Ancona, Italy Objectives: Hyaluronic acid (HA) is a widely used molecule in shoulder pathology, administered through the intra-articular/subacromial way. Indications for these injections are: rotator cuff tear, Duplay disease, frozen shoulder, osteoarthritis, etc. Although several studies were published, literature still lacks in details about the pertinence in the indications. Goal of the present study was to point out the correct indication for HA injective therapy through a perspective study: firstly defining the safety and efficacy of HA in the different stages of cuff tears, then evaluating the effectiveness at a long term follow up (90 days) as secondary endpoint. Methods: During the period January 2007 to October 2008, using strict recruitment criteria, 100 patients suffering of cuff pathology were enrolled. The population of study were divided into 4 groups according to Neer classification with a fourth added group (cuff-teararthropathy). Each patient underwent a cycle of 3 US-guided injections of HA (Sinovial-IBSA 0.8%-16 mg/2 ml), every 15 days. To perform the injection, antero-lateral way was used. Follow up was planned every 15 days (t0, t15, t30, t45 e t90), using VAS, OxfordShoulder-Score (OSS) and Constant–Murley. Results: Primary endpoint. In stage I and II, at day 30 of FU, a significant reduction of VAS and increase of Constant–Murley and OSS was recorded. In 4 stage benefits were recorded for the first 45 days, while OSS and Constant did not show any improvement. Secondary endpoint was satisfied for stage I, II and IV. All data are statistically significant (ANOVA). Adverse or side effects had were in line with the percentages reported in literature. Conclusions: HA in cuff pathology represents a valid and safe alternative to other conservative treatments, in particular to corticosteroids. HA prescription should be made in appropriate case: the best results were reached in patient with bursitis (grade 1) or partial tear (grade 2). Not a big benefit is reported in patients with complete cuff tear (grade 3), while patients with osteoarthritis (grade 4) feel few and short-lasting benefits. These data allow to consider HA therapy particularly useful for bursitis or partial cuff tear, while in presence of cuff tear arthropathy it must be considered just to delay surgery as well as a temporary solution looking forward to arthroplasty. In complete cuff tear, HA is not effective and surgical reconstruction still remains the best option.

123

S122 P15-1156 Early recovery from rotator cuff surgery may incite an increased likelihood of re-tear: compliance may be the critical issue M. Haber1 1 Southern Orthopaedics, University of Wollongong, Wollongong, Australia Objectives: Reducing the incidence of re-tear is of significant clinical importance. Developing a greater understanding of post-operative factor which increase the risk of re-tear can further enhance rehabilitation protocols. Methods: A group of 149 patients who had undergone rotator cuff repair surgery were evaluated in this study. This group is comprised of 93 males and 56 females with an average age of 59.05 years (SD ± 9.64). Preoperative, 6, 12, and 26 week Constant Shoulder Scores (CSS), Western Ontario Rotator Cuff Index (WORC), Oxford Shoulder Score (OSS) and Short Form 12 (SF-12) were assessed. A two-tailed, unequal variance paired samples t test was used to compare preoperative and postoperative outcomes. An outcome p value of p \ 0.05 from the t test was regarded to be statistically significant. Relationships between patients with and without re-tear at 6, 12 and 26 weeks were investigated. Results: Average patient 6 week CSS, WORC, OSS and SF-12 scores for patients re-tearing at 12 weeks and patient 12 week scores retearing at 26 weeks were more favourable than those patients without a re-tear. CSS, WORC, OSS and SF-12 scores for patients with a retear were significantly different (p \ 0.001) from patients that did not re-tear before the re-tear occurred. Conclusions: Correlation of cuff integrity and clinical outcomes has remained controversial. Pre-operative and intra-operative factors which may correlate with re-tear or failure to heal have been well documented. However there is little information on postoperative factors which correlate with sound cuff healing. Of concern is the importance of compliance with post-operative rehabilitation protocols. These results show that clinical outcome scores in patients prior to retearing are better than patients that re-tear. Patients that re-tear at 12 weeks have more favourable clinical outcomes at 6 weeks than non-tearing patients; this is similar for patient scores at 12 weeks tearing at 26 weeks. These outcomes suggest that patients with better early recovery from surgery are more likely to re-tear. One explanation is there may be a compromise in compliance. The patients with the least pain post-operatively are the ones most likely to sustain a re-tear. This may be due to poor compliance due to less pain. These findings are helpful in stressing the importance of rehabilitation compliance in the early post-operative period.

P15-1195 The postoperative results of arthroscopic HAGL repairs for anterior shoulder instabilities T. Kobayashi1 1 KKR Hokuriku Hospital, Department of Oprhopaedic Surgery, Kanazawa, Japan Objectives: The most frequent essential lesion of traumatic anterior instability is Bankart lesion, and humeral avulsion of glenohumeral ligament lesion (HAGL lesion) is rare. We investigated the postoperative result of HAGL repair for anterior shoulder instability. Methods: We performed one-hundred sixty-two surgeries for traumatic anterior shoulder instabilities for past 10 years. The essential lesion for 145 shoulders was Bankart lesion, that for 2 capsular rupture, that for 7 rotator interval lesion, and that for 8 HAGL lesion (4.8%). These eight included 4 males, 4 females, 5 right side, and 3 left side, the average age at operation was 44.8 (range from 23 to 66) years old, and the average follow-up period was 18.2 (range from 12

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 to 30) months. One revision HAGL repair after arthroscopic Bankart repair and seven primary HAGL repairs were included. We investigated preoperative features that were age at first dislocation, first traumatic episode, glenoid shape by Sugaya’s classification, associated lesion, type of Hill–Sachs lesion, and the number of used anchors, and evaluated postoperative recurrent dislocation, external rotation limitation, the level of return to sports activity, and the postoperative clinical evaluation by Japanese orthopaedic shoulder instability score. Results: The average age at first dislocation was 33.3 (range from 20 to 48) years old. The first traumatic episodes were falling down at skiing and snowboarding in four, spike at volleyball in one, serve at tennis in one, direct blow at judo in one, rowing boat in one. Seven had normal shaped glenoids except one erosion in revision case by Sugaya’s classification. No Hill–Sachs lesion were observed in four, narrow and shallow in three, wide and deep in one with Bankart lesion. The repair procedure were soft tissue repair in one, one anchor-two sutures in five, two anchors in one, and three anchors in one. No associated Bankaer lesion and SLAP lesion was observed, and one shoulder had recurrent subluxation after direct blow at ski. Postoperative 10 external rotation limitation was observer in one who was performed previous Bankart procedure. Seven cases except one who had recurrent subluxation returned to previous activity level. Average JSS shoulder instability score was recovered from 71.8 to 97 points. Conclusions: Clinical features of traumatic anterior shoulder instability caused by HAGL lesion were relatively high age at first traumatic episode, rare glenoid deformity, subtle Hill–Sachs lesion and rare associated SLAP lesion. The postoperative results were successful without apparent external rotation limitation. One anchortwo sutures procedure was preferable and sufficient for HAGL repair.

P15-1201 Minimally invasive coracoclavicular stabilization with double augmentation for acute acromioclavicular dislocation S.W. Choi1, M.K. Kim2, K.W. Nam1, H.S. Kang1 1 Jeju national university, Jeju, Republic of Korea, 2Inha University Hospital, Incheon, Republic of Korea Objectives: The purpose of treatment of acute AC joint dislocation should be to return the patient to the level of function before injury, with a pain-free, strong and mobile shoulder. But Ideal treatment for AC joint dislocation is currently controversial. The Objectives of this study were to introduce an operative technique for AC dislocation that utilizes suture anchors or double flip button, and to evaluate the clinical usefulness of our surgical results. Methods: Thirty-five patients (32 males and 3 females) with Rockwood types IV or V acromioclavicular joint dislocation were studied. All the patient were operated on using suture anchors or double flip button placed between the clavicle and the coracoid process. Postoperative evaluation were analyzed retrospectively, clinically, and radiographically. Results: After a mean follow-up of 46.7 months, the mean Constant score for the 35 patients was 90.3 (range, 74–100). In spite of excellent clinical results and patient satisfaction, one patients required revision surgery. Slight loss of reduction was noted in 8 cases, but their functional outcomes were good. Conclusions: The principles of treatment for acute acromioclavicular joint dislocation should be reduced anatomically and maintained stabilization until the recovery of damaged coracoclavicular ligament. Indirect reconstruction of coracoclavicular ligament using suture anchor or double flip button is minimally invasive technique, easily performed, and does not require surgery for removal. And this surgical technique is considered a useful way early return to normal activities because it enables early shoulder joint motion.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P15-1204 Evaluation of current biologic carriers for mesenchymal stem cell application in rotator cuff surgery K. Beitzel1, M.B. McCarthy1, R. Russell1, C. Edgar1, R.A. Arciero1, A.D. Mazzocca1 1 University of Connecticut Health Center, Department of Orthopedic Surgery, Farmington, United States Objectives: Application of human mesenchymal stem cells (MSCs) is considered one of the future key factors for improving the healing environment in rotator cuff repair (RCR). MSC’s potential for selfrenewal and ability to differentiate into bone, tendon and cartilage may enhance tendon to bone healing. However a practicable biological carrier (extracellular matrix—ECM) is needed to localize and maintain a high concentration of cells at the site of the repair zone. Purpose of this study was to examine the reaction of human MSCs in culture to currently available scaffold materials compared to native tendon tissue as a control. Hypothesis was that currently available ECMs would be suitable for cell application but demonstrate significant difference in MSCs adhesion, proliferation, and viability. Methods: Human bone marrow aspirated from the proximal humerus was used to isolate and culture MSCs. MSCs were defined by (1) their potential of forming colonies; (2) their ability to differentiate into tendon, cartilage, bone and fat tissue; and (3) by FACS analysis (CD: 73,90,45). ECM-Samples (5 9 5 mm2) were taken from fresh frozen human RC tendon, human highly cross-linked collagen membrane (Flexigraft), porcine non-crosslinked collagen membrane (Mucograft), a human platelet rich fibrin matrix (PRF-M) and a fibrin matrix based on platelet rich plasma (ViscoGel). Each sample was soaked for 30 min in the MSC solution (450 9 103 cells/0.1 ml) and then transferred into control media. Cells were counted for adhesion (24 h); thymidine assay (disintegrations per minute) was obtained to examine cell proliferation (96 h) and live/dead stain (calcein-AM/ethidium homodimer-1) was evaluated with confocal-microscopy (Zeiss LSM510) for viability (168 h). Histology (H&E) was performed after 21 days and the unloaded scaffolds were scanned with electron microscopy. ANOVA & Tukey post hoc test were performed for statistical analysis. Results: A significantly greater number of cells adhered to both Mucograft and PRF-M. Cell activity (proliferation) was significant higher in the Mucograft compared to PRF-M and Vicogel. There were no significant differences found in the results of the Life/Dead assay.

Assay

RC-

Flexigraft

Muco-

PRF-M

graft

Tendon

Visco-

Signifi-

gel

cance p (ANOVA)

Adhesion (cells) Proliferation (dpm)

5,800 ±

7,213 ±

24,425 ±

38,500 ±

5,375 ±

4,107

4,781

7,521

10,663

892

1,869 ±

1,507 ±

1,594

1,881

Life/dead (%)

77 ± 38

72 ±

3,159 ±

0.000

0±0

0±0

0.004

100 ± 0

100 ± 0

0.269

74 100 ± 0

37

Conclusions: Differences comparing the various types of ECMs existed and surgeons should be aware of the particular properties of

S123 each ECM. This allows selection of ECMs based on their specific features (e.g. highly crosslinked for more biomechanical properties vs. not crosslinked for their biological properties). Results of this study may be utilized as a basis for further animal or clinical studies on the application of mesenchymal stem cells for improved healing in RC-repair.

P15-1215 Functional effectiveness of arthroscopic rotator cuff repair in the elderly E. Antonogiannakis1, V. Tsiampa1, G. Arealis2, E. Mataragas2, G. Stamatakis2 1 IASO General, 2nd Orthopaedic Department, Center of Shoulder Arthroscopy, Athens, Greece, 2IASO General, Athens, Greece Objectives: The increased desire of elderly individuals for physical activities in our days, has led to surgical treatment of rotator cuff lesions. According full-thickness tears management, many operative interventions have been advocated. In the following study are presented the functional outcomes of arthroscopic repair of massive rotator cuff tears in the elderly. Methods: Between July 2003 and May 2011, 71 elderly patients underwent arthroscopic rotator cuff tear repair. 31 patients aged over 70 years old, with a follow-up ranged from 32.5 to 59.2 months (2–5.5 years) were reviewed. Fifteen were men and sixteen women, their mean age was 74.58 years old (70–82), and their activity level was high in 80%, medium in 27% and low in 3%. All patients suffered from symptomatic full thickness rotator cuff tears, and underwent arthroscopic rotator cuff repair with suture anchors. The preoperative and postoperative functional outcome evaluation included the ASES, CONSTANT according Boehm, and UCLA scores. The tear size was measured in the sagittal plane at its insertion into its respective anatomic footprint, and the classification of DeOrio and Cofield was recorded medium (1–3 cm) in most of the cases (13), massive ([5 cm) in 11 cases, large (3–5 cm) in 6 cases, and small in one. All repairs were performed using a single row fixation, in order to avoid overtensioning and failure of the repair. The anchors used per shoulder were 3.2 in total (range 1–4, SD ± 1.5). Postoperatively the patients in all groups performed a similar patient controlled rehabilitation program that included early passive and active exercises starting from the 2nd postoperative day. Results: The mean scores at the final follow up are: CONSTANT 78.52 (SD ± 17.24), UCLA 30.55 (SD ± 3.44) and ASES 80.94 (SD ± 17.69). 28 of the patients were satisfied and would opt for arthroscopy if given the choice. Two suffered from postoperative intra-articular shoulder infection that was treated with arthroscopic lavage and antibiotics. Both suffered from uncontrolled diabetes and both had bad final results (CONSTANT 29, UCLA 11 and ASES 40). One sustained a heart attack 3 months postoperatively and couldn’t follow the rehabilitation program. Most patients returned to the desired activity level, even the ones with higher demands. The patients with massive tears were painless and had good range of motion but their strength did not return to normal. Conclusions: Arthroscopy in the elderly is effective. It can successfully restore adequate and painless function. Muscle strength may not return to normal and the patients should be informed before the procedure. The limit of shoulder arthroscopy should be the extend of tendon lesions and arthritis coexistence, rather than the patient’s age. Comorbidities, especially diabetes should be managed before the arthroscopic procedure.

123

S124 P15-1221 Folding–unfolding of the anterior shoulder capsule as an indicator of glenohumeral ligament tension: analysis of stable, unstable and frozen shoulders N. Pouliart1 1 Universitair Ziekenhuis Brussel, Orthopaedics and Traumatology, Vrije Universiteit Brussel, Brussels, Belgium Objectives: Judging adequate tension of the glenohumeral ligaments can be difficult as no definite measurements exist. The MGHL and ABIGHL are arthroscopically visible as folds in the anterior shoulder capsule. In a cadaveric study, these folds became more prominent in ADIR, whereas they were smoothed out upon ABER. The present study was set up to determine whether this folding– unfolding mechanism (FUM) is useful as an indicator of capsular tension. Methods: The indication for surgery and examination under anaesthesia were correlated with observation of the FUM and capsuloligamentous lesions in 550 consecutive shoulder arthroscopies in lateral decubitus with 30 of abduction in the scapular plane. One hundred and thirty-two shoulders underwent arthroscopy for instability, 54 for a frozen shoulder, and 183 for rotator cuff tears—10 involving the subscapularis tendon—and 181 for various other pathologies. Results: Stable shoulders: The ABIGHL and the MGHL were each marked by a fold in N(eutral) R(otation) that became more prominent in I(nternal) R(otation). When progressing to full ER both folds were smoothed out to form a sheet-like capsule. Additional differentiation between the FUM for the MGHL and ABIGHL was possible by changing AB to 15 for the MGHL and to 45 for the ABIGHL. Frozen shoulders: The anterior capsule was smooth without visible folds in any degree of rotation, limited by the adhesive capsulitis. Releasing the capsule from the glenoid rim did not change this appearance. Unstable shoulders: In 92 shoulders with at least anteroinferior instability, the MGHL and ABIGHL still formed folds in IR. However, the folds were not as prominent when the capsule was detached from the labrum. Full ER, increased up to 90 in some patients, did not result in smoothing of the folds, not even with increased abduction. In 39 shoulders with isolated anterosuperior instability, the FUM of the ABIGHL had the same appearance as in stable shoulders. The FUM of the MGHL was deficient is these cases. In addition, the rotator cuff interval did not narrow with ER. With a deficient FUM in the presence of an intact labrum, it was possible to correlate midsubstance and HAGL lesions of the ABIGHL and MGHL with increased translation in the load-and-shift tests. After repair of the labroligamentous lesion and associated capsular shift, the FUM reappeared with an ER that was reduced to 45. Conclusions: These observations suggest that smoothing of the anteroinferior capsule at a maximum of 45ER and 45AB could be used as an indication of normal tension in the MGHL and IGHL. When the FUM does not occur within this range, these ligaments are probably insufficient, be it torn or stretched. During capsular shift, especially when confronted with elongation or midsubstance lesions, a reappearing FUM could be used to evaluate achievement of adequate capsular tension. When no folds at all are visible, even with full IR, this indicates a very tight capsule and likely a frozen shoulder, especially when rotation is decreased.

P15-1230 Correlation of clinical and structural outcomes after arthroscopic rotator cuff repair with suture bridge technique S.W. Choi1, G.M. Kim1, K.W. Nam1, M.K. Kim2

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Jeju National University, Jeju, Republic of Korea, 2Inha University Hospital, Incheon, Republic of Korea Objectives: To evaluate the clinical outcomes and the maintenance of repair integrity or retear rate of arthroscopic rotator cuff repair using a suture bridge technique for patients with over the medium sized rotator cuff tears. And to evaluate retear patterns in the cases with structural failure after arthroscopic suture-bridge repair with use of follow up magnetic resonance imaging (MRI). Methods: Fifty-three patients (26 males, 27 females; mean age 59 years; range 44–74) underwent arthroscopic rotator cuff repair with suture bridge repair technique from May, 2008 to September, 2010. Clinical and functional evaluations were made according to the Constant score, the score for the visual analogue scale for pain, the UCLA score, as well as a full physical examination of the shoulder. All patients were judged to reveal healed tendon on magnetic resonance imaging at least 12 months after postoperatively. And we evaluated repair integrity of the cuff muscles using the classification of Sugaya et al. All statistics were analyzed using the SPSS software package (Version 12.0; SPSS Inc., Chicago, IL, USA). Results: The average clinical outcome scores and strength were all improved significantly at the time of the last follow-up (P \ 0.01). The rotator cuff wad healed in 41 (77.4%) of the 53 cases, and there was a recurrent tear in 12 cases (22.6%). The larger the intraoperative tear size, the higher the rate of retear (P \ 0.009). The incidence of retear tended to increase with age older than 60 years (P \ 0.031). When the severity of preoperative fatty degeneration of the cuff muscles was higher, there was a greater chance of a recurrent tear (P \ 0.001). Conclusions: arthroscopic suture bridge rotator cuff repair have higher healing rate, improvement in the pain relief and the ability to perform the activities. Tendon healing is affected various factors. Healing rate is affected by patient age, preoperative tear size, and preoperative fatty degeneration. The retear in cases with a suture bridge technique tended to be more frequently at the musculotendinous junction. Interestingly, almost patients were satisfied with the results of surgery although repair integrity is not maintained.

1

P15-1285 Simultaneous Bankart and rotator cuff repair for anterior shoulder instability with Bankart lesion and rotator cuff tear T. Saji1, T. Sano1, H. Matsuoka1 1 Shizuoka General Hospital, Department of Orthopaedic Surgery, Shizuoka, Japan Objectives: To evaluate the outcomes of simultaneous Bankart and rotator cuff repairs under arthroscopy. Methods: Four patients with 4 shoulders who underwent arthroscopic Bankart repair with rotator cuff repair for anterior shoulder instability were included in this study. The clinical charts including surgical records were investigated retrospectively. Videos of each operation were also reviewed. Results: The periods from initial injury to surgery were 2, 10 months, 16 and 45 years. All had Bankart lesion. Two patients who were operated within 1 year from the first dislocation were more than 75 years old and had large rotator cuff tear including subscapularis tendon tear. The other two, who had the first dislocation at young age, had partial supraspinatus tendon tear. All had Bankart repair with bioabsorbable anchors. Partial supraspinatus tendon tears were repaired by trans tendon technique with bioabsorbable anchors. Full thickness rotator cuff tears were repaired with metal anchors. One with subscapularis tendon tear did not have subscapularis tendon repair and sustained a recurrent dislocation even after surgery. The other three did not experience recurrent anterior shoulder instability at latest follow-up of 9–50 months after surgery.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: The patients in the present study were divided into two groups; one with partial supraspinatus tendon tear and one with large full thickness rotator cuff tear including subscapularis tendon tear. The former had initial injury at young age and the latter, at old. In the treatment with anterior shoulder instability with subscapularis tendon tear, at least subscapularis tendon should be repaired.

P15-1292 Surgical treatment of chronic acromionclavicular dislocations: comparison of two surgical techniques P. Ciampi1, C. Scotti1, G. Peretti1, G. Fraschini1 1 San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy Objectives: Surgical treatment of chronic complete acromioclavicular (AC) joint dislocation is still debated and no gold standard surgical procedure has been identified. The aim of this study is to compare the outcome of two surgical procedures of coracoclavicular reconstruction with conservative treatment by reporting our experience in chronic AC dislocation types IV-V according to Rockwood. Methods: A retrospective series of 90 patients, 84 men and 6 women with a mean age of 31.5 years treated for AC dislocations from 1999 to 2009 is reported. The dominant extremity was involved in 66 injuries (74%). The time from injury averaged 3 months (1–6). The injury was the result of a sport trauma in 67 patients (75%). Patients were divided into three groups: group 1 receiving AC reconstruction with Dacron vascular prosthesis, group 2 receiving AC reconstruction with LARS artificial ligament, group 3 receiving conservative treatment. The groups were homogeneous with respect to age, sex, type of trauma and duration of symptoms. Pre-operative assessment included bilateral static plain radiographs and bilateral dynamic radiographs taken with the patients standing and holding a 5 kg weight in each hand. Inclusion criteria were as follows: type III or higher AC dislocation according to Rockwood, minimum 1 month after trauma, no previous surgery performed. Follow-up was performed after 1, 6 and 15 months after surgery. The modified UCLA acromioclavicular rating scale was used in all patients after 15 months in order to evaluate the overall success of the procedure. In order to verify effectiveness of the procedure, the UCLA score, and the Shoulder Pain and Disability Index (SPADI) score were utilised in those patients who did not experience any complication. The various complications were separately recorded in order to highlight the safety of the procedure. Data were statistically analysed by use of paired Mann–Whitney tests and one-way analysis of variance tests. Results: Patients treated surgically presented significant better functional outcome compared to patients treated conservatively with overall positive results in 93.3% of patients for group 2 and 53.3% of patients of group 1. However, reconstruction with Dacron vascular prosthesis presented an unacceptable high complications rate (43.3%) compared to LARS group characterized by low complication rate (6.6%). Conclusions: Our results show that anatomic AC reconstruction with LARS artificial ligament resulted in both satisfactory functional outcome and low complication rate. Therefore, we recommend this procedure for the treatment of chronic complete AC dislocation.

P15-1322 Cost-effectiveness of rotator cuff repair surgery A. von Keudell1, L.D. Higgins1, J.P. Warner2, N.B. Jain1 1 Brigham and Women’s Hospital/Harvard Medical School, Orthopaedic Surgery, Boston, United States, 2Massachusetts General

S125 Hospital/Harvard Medical School, Orthopaedic Surgery, Boston, United States Objectives: Shoulder arthroscopy is the 2nd most common performed procedure in orthopaedic surgery with more than 75,000 cuff repairs per anno. However, there is scarce data regarding the cost-effectiveness of this procedure. This study tries to elucidate cost utility of rotator cuff surgery taking complications and re-tear rates associated with the procedure into account. Methods: We developed a cost effectiveness analysis model to examine the impact of the different health utilities according to complication probabilities and associated costs. The model input parameters were derived from published literature and included a mean risk of failure rate of 15% (3.9–32%), infection of 1.8% (1.7–1.9%), nerve injury of 1.0% and stiffness rate of 10% (2–20%). A minimum of 2-year follow-up studies were used for health utility scores using SF36 Short Form. Average life expectancies were applied to generate cost-effectiveness ratios. Overall hospital costs ($10,605.20) were retrieved from literature and cost for additional physical therapy ($150), physician appointments ($250) and additional procedures for revision ($15,605.20) or for complications ($5,302.60) for complications, were estimated. To examine variability of results to uncertainty in parameter estimates we conducted a wide set of sensitivity analyses. Results: The mean gain of QALY after rotator cuff surgery was 1.56 for an uncomplicated rotator cuff repair. The estimated lifetime gain in QALY from the revision surgery was 0.50 and from complications 1.02 was noted if complications or a re-tear occurred after the index surgery. This yielded a cost-effectiveness ratio of $6,793.20/QALY by the use of a minimum of 2-year follow-up of the SF-36 for uncomplicated rotator cuff repair. In revision rotator cuff surgery due to a complication or a re-tear the ratio increased to $20,907.80/QALY and $62,420.80/QALY, respectively. Sensitivity analysis demonstrated the robustness of the model. Conclusions: This study demonstrates the cost-effectiveness of rotator cuff surgery, even taken complications into account. However in case of a re-tear after the index surgery the cost-effectiveness is higher when compared to other surgical interventions in orthopaedic surgery.

P15-1329 Inter-observer reliability of the Sugaya’s classification to assess structural integrity of rotator cuff after arthroscopic repair M.F. Saccomanno1, L. Deriu1, S. Careri1, C. De Ieso1, C. Fabbriciani1, G. Milano1 1 Catholic University, Orthopaedics, Rome, Italy Objectives: To evaluate the inter-observer reliability of the Sugaya’s classification to assess structural integrity of rotator cuff after arthroscopic repair on magnetic resonance imaging (MRI). The hypothesis of the study was that Sugays’s classification has a very good inter-observer reliability. Methods: A consecutive series of 50 patients who underwent arthroscopic repair of full-thickness rotator cuff tear were enrolled for the present study. Exclusion criteria were: partial-thickness and irreparable cuff tears, isolated or combined subscapularis tears, and previous surgery to the same shoulder. Each patient underwent an MRI of the operated shoulder 12 months after surgery. Structural integrity of rotator cuff was evaluated on T2-weighted images and classified into 5 types according to Sugaya’s classification (type I, sufficient thickness with homogenously low intensity; type II, sufficient thickness with partial high intensity; type III, insufficient thickness without discontinuity; type IV, presence of a minor discontinuity; type V, presence of a major discontinuity). All the MRI exams were evaluated once by two observers (one orthopaedic surgeon and one radiologist well-experienced in musculoskeletal imaging). Each observer was unaware of the other observer’s results.

123

S126 Inter-observer agreement in classifying structural integrity of rotator cuff was analyzed by calculating overall percent agreement and kappa statistics with quadratic weighting. Significance was set at p \ 0.05. Ninety-five confidence intervals (95% CIs) were calculated for percent agreement and weighted k coefficient. Results: Percentage agreement between the two observers was 60% (95% CI: 45–73%). weighted k coefficient was 0.78 (95% CI: 0.64–0.92). Conclusions: Sugaya’s classification for evaluation of structural integrity of rotator cuff tendons after arthroscopic repair showed a good inter-observer reliability.

P15-1335 Functional outcome after arthroscopic rotator cuff repair in the elderly M. Antonogiannakis1, V. Tsiampa1, G. Arealis1, E. Mataragas1, I. Hiotis1 1 IASO GENERAL, Centre for Shoulder Arthroscopy, Athens, Greece Objectives: The increased desire of elderly individuals for physical activities in our days, has led to surgical treatment of rotator cuff lesions. Different operative interventions for full-thickness tears have been proposed. In the following study are presented the functional outcome of arthroscopic repair of rotator cuff tears in the elderly (aged over 70). Methods: Between July 2003 and May 2009, 31 elderly patients underwent arthroscopic rotator cuff tear repair, with a follow-up ranging from 32.5 to 59.2 months (2–5.5 years). Fifteen were men and sixteen women, their mean age was 74.58 years old (70–82), and their activity level (normalized for age) was high in 80%, medium in 27% and low in 3%. All patients suffered from symptomatic full thickness rotator cuff tears, and underwent arthroscopic rotator cuff repair with suture anchors. The preoperative and postoperative functional outcome evaluation included the ASES, CONSTANT according to Boehm, and UCLA scores. The tear size was measured in the sagittal plane, at its insertion in the anatomic footprint, and was recorded according to the classification of Cofield and De Orio. There were 13 Medium size (1–3 cm) tears, 11 massive ([5 cm), 6 large (3–5 cm), and one small. All repairs were performed using a single row fixation as there was concern of overtensioning the repair due to loss of tissue elasticity. The anchors used per shoulder were 3.2 in total (range 1–4, SD ± 1.5). Postoperatively the patients in all groups performed a similar patient controlled rehabilitation program that included early passive exercising starting from the 2nd postoperative day and delayed active (after 2 months). Results: The mean scores at the final follow up were: CONSTANT 78.52 (SD ± 17.24), UCLA 30.55 (SD ± 3.44) and ASES 80.94 (SD ± 17.69). 28 of the patients were satisfied and would opt for arthroscopy if given the choice. There were 2 postoperative infections treated with arthroscopic lavage and antibiotics. Both suffered from uncontrolled diabetes and both had bad final results (CONSTANT 29, UCLA 11 and ASES 40). One sustained a heart attack 3 months postoperatively and couldn’t follow the rehabilitation program. Most patients returned to the desired activity level, even the ones with higher demands. The patients with massive tears were painless and had good range of motion but their strength did not return to normal. Conclusions: Arthroscopy in the elderly is effective. It can successfully restore adequate and painless function. Muscle strength may not return to normal and the patients should be informed before the procedure. The limit of shoulder arthroscopy should be the chronicity of the lesions and the coexistence of arthritis (rot cuff arthropathy), rather than the patient’s age. Comorbidities,

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 especially diabetes should be managed before the arthroscopic procedure.

P15-1399 Throwing sport improves shoulder proprioception P. Lubiatowski1, P. Kaczmarek2, M. Wojtaszek3, W. Dudzinski2, M. Nowak4, L. Romanowski5 1 Rehasport Clinic, University of Medical Sciences, Orthopaedics and Hand Surgery, Poznan, Poland, 2Rehasport Clinic, Poznan, Poland, 3 University of Medical Sciences, Orthopaedics and Hand Surgery, Poznan, Poland, 4Rehasport Clinic, Dynasplint, Poznan, Poland, 5 University of Medical Sciences, Poznan, Poland Objectives: Neuromuscular control is important mechanism in shoulder function. It has been found that proprioceptive abilities deteriorate in shoulder instability. Proper treatment restores that function. It is controversial whether throwing sports and sports training effects shoulder proprioception. The aim of the study was to analyze the shoulder proprioceptive abilities of professional handball players. Methods: Material consisted of 70 professional handball players from top national league and 25 healthy volunteers as control. Shoulder proprioception was evaluated by measuring error of active reproduction of joint position (EARJP). Own construction electronic goniometer (Propriometer) with accuracy of 0.1 was used for measurements. Both dominant and non-dominant shoulders were evaluated. EARJP was measured in 4 directions and 3 positions for each direction: flexion and abduction (60, 90, 120), external and internal rotation (30, 45, 60). Three repetitions for each position were performed to obtain an average EARJP for every subject and position. Some correlations have been made regarding to limb dominance, external/internal rotation in abduction and isokinetic shoulder profiles. All subjects had also measurements of rotational motion (GIRD) and isokinetic testing. Results: Results of EARJP have been depicted in Tables 1 and 2. Correlations. Based on statistical analysis (parametric, nonparametric) there was significantly better proprioceptive control (lower EARJP) in throwing versus non throwing shoulder and throwing versus control group shoulder. There was no difference in proprioception when comparing dominant and non-dominant shoulders in control group. Higher reference angles correlated with better proprioception.

Table 1 Results for abduction and flexion 60 reference Dominant shoulder

90 reference

120 reference

Nondominant

Control

Dominant shoulder

Nondominant shoulder

Control

Dominant sholder

Non-dominant shoulder

Control

Abduction

5±3

5.5 ± 3

5.1 ± 2.7

3.4 ± 2.2

4.5 ± 2.3

4.3 ± 2.3

3.6 ± 2.5

4.1 ± 2.4

4.2 ± 2

Flexion

4.9 ± 3.3

6.7 ± 5

6.1 ± 2.6

3.2 ± 1.9

4.4 ± 4.3

3.6 ± 1.9

3.1 ± 1.6

4.1 ± 2.6

3.7 ± 2.2

Table 2 Results for external and internal rotation 30 reference

45

Dominant shoulder

Nondominant shoulder

Control

Internal rotation

3.2 ± 1.8

3.1 ± 1.8

3.4 ± 1.3

External rotation

2.9 ± 1.5

3.9 ± 2.2

3.9 ± 1.7

60

Dominant shoulder

Nondominant shoulder

Control

3 ± 1.2

3.1 ± 1.8

3±1

3.6 ± 1.7

3.3 ± 1.4

3.4 ± 2

Dominant shoulder

Nondominant shoulder

Control

2.6 ± 1.4

3.1 ± 1.2

3 ± 1.3

3 ± 1.8

3.1 ± 1.7

3 ± 1.5

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Increased external rotation correlated with better shoulder control in mid range of external rotation. Conclusions: Throwing sport (as handball) affects neuromuscular shoulder control. Handball players show significantly better proprioceptive abilities when compared to control normal population. Throwing shoulder has better neuromuscular control then the opposite one. There may other factors also improving shoulder proprioception (higher external rotation range, more extreme shoulder position).

P15-1436 The efficacy of two different patches for rotator cuff repair P. Ciampi1, C. Scotti1, G. Peretti1, M. Vitali2, G. Fraschini1 1 San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy, 2University of Milan, Residency Program in Orthopaedics and Traumatology I, Milan, Italy Objectives: The treatment of massive rotator cuff tears represents a challenging problem in shoulder surgery. Traditional repair techniques are associated with high rupture rates due to excessive tension on the suture and the presence of degenerated tendon tissue. For these reasons, surgeons have attempted to solve this problem by the use of grafts, synthetic materials or biologic tissues. The purpose of this study was to compare the efficacy of the use of pericardium patch with the use of prolene patch in the repair of extensive rotator cuff tears. Methods: A retrospective series of 180 patients, 115 men and 65 women with a mean age of 66.8 years treated for a massive rotator cuff tear from 1997 to 2008 is reported. The inclusion criteria were: symptomatic patients with pain, deficit of elevation, not responsive to the physiotherapy, tear size (2 or more tendons), minimum follow-up of 2 years since surgery. Patients were divided into three groups according to the type of treatment received: group 1 was treated with pericardium patch, group 2 with prolene patch, group 3 with simple suture. The groups were homogeneous with respect to age, sex, type of lesion and duration of symptoms. Plain radiographs, ultrasound and MRI of the shoulder were performed preoperatively and at 3 years. Patients were clinically evaluated using the UCLA score before surgery and at 2 months and 3 years after surgery. Pain was assessed by use of VAS scale. Strength was evaluated with a digital dynamometer. The surgical procedure was similar in all three groups. An anterior shoulder incision was made. Minimal acromioplasty was performed. The edges of the rotator cuff tear were identified, derided, mobilized and sutured. In group 1, a pericardium patch was used as an augmentation graft, in group 2, a prolene patch was placed, while in group 3 the simple suture was performed. The change in UCLA scores, VAS scale were analyzed with the paired Student’s test, assuming a normal distribution of the total score. Results: Patients treated with patch presented a significant better functional outcome compared to the patients treated without patch with overall positive results in 96% of patients for group 2 and 59.3% of patients of group 1. Average UCLA scores improved from 10.9 preoperatively to 24.6 at final follow-up for group 2, from 10.7 to 14.8 for group 1 and from 10.7 to 14.7 for control group. The average abduction strength was 8.8 kg in group 1, 13.6 in group 2 and 8.5 kg in control group with a statistically significant difference between the two patch groups. Incidence of retears at third year was significantly high for group 1 (17%) and for control group (14%) compared to group 2 (5%). No adverse side effects were reported during the study period. Conclusions: The use of prolene patch as an augmentation graft in the treatment of massive rotator cuff tears is safe and, in most patients, can give a significant pain relief and improvement of range of motion and strength with few complications.

S127 P15-1453 Do microfractures influence recovery after rotator cuff repair? A randomized controlled study L. Osti1, R. Papalia2, A. Del Buono2, N. Maffulli3, V. Denaro2 1 Hesperia Hospital, Modena, Italy, 2Campus Biomedico Roma, Roma, Italy, 3Queen Mary University of London, Barts and The London School of Medicine and Dentistry, Centre for Sports and Exercise Medicine, London, United Kingdom Objectives: Microfractures induce the release of growth factors and stimulate musculoskeletal healing process, but the evidence on its effects at the interface between tendon and bone after rotator cuff repair is still lacking. Hypothesis: In medium rotator cuff tears, microfractures accelerate recovery in pain relief, functional outcome, overall satisfaction. Methods: Fifty patients who underwent repair of small to medium rotator cuff tears were randomly divided into 2 groups, using a block randomization procedure. The treatment group included 25 patients who underwent microfracturing in the context of rotator cuff repairing, the control group (N = 25) included 25 patients who had undergone rotator cuff repair alone. All the patients underwent traditional single row repair. Microfractures were performed in a standard technique over the footprint. Outcomes were assessed preoperatively and at 2, 4, 6 weeks, at 3, 6, and a minimum of 24 months from the index surgery with respect to pain, range of motion, strength, and overall satisfaction. The Constant, the University of California at Los Angeles (UCLA) and ASES scores were administered to all the patients. Results: In the context of cuff repair, microfractures significantly improve pain, range of motion, strength, and overall satisfaction when compared with cuff repair alone at 2, 4, 6 weeks and 3 months. Constant and UCLA scores were significantly higher in the treatment group than the control group up to 3 months from surgery UCLA: (27. ± 3 vs. 24.1 ± 2.5; Constant: 67 ± 10 vs. 58.8 ± 11; P \ .05). Comparable outcomes and functional scores were observed after 3 months in both the groups. Conclusions: In small to medium size cuff tears, microfracturing improves pain, range of motion, strength, and overall satisfaction in the short term. Nevertheless, additional investigations and larger randomized studies powered for healing rate, are needed to further determine the effect of this procedure.

P15-1457 Glenohumeral osteoarthritis after arthroscopic Bankart repair for anterior instability F. Franceschi1, R. Papalia2, A. Del Buono2, G. Rizzello3, N. Maffulli4, V. Denaro2 1 Campus Biomedico University, Orthopaedic and Trauma Surgery, Rome, Italy, 2Campus Biomedico University, Orthopaedic Surgery, Rome, Italy, 3University Campus Biomedico of Rome, Rome, Italy, 4 Queen Mary University of London, Barts and The London School of Medicine and Dentistry, Centre for Sports and Exercise Medicine, London, United Kingdom Objectives: Little work has been published around shoulder arthropathy after arthroscopic repair for glenohumeral instability. Arthroscopic stabilization of Bankart lesions does not prevent the development of postoperative glenohumeral osteoarthritis. Methods: We compared clinical (Rowe and Constant scores) and radiographic preoperative and postoperative data of 60 patients who underwent arthroscopic Bankart repair. Osteoarthritis was graded preoperatively and postoperatively by using the Buscayret classification grading system. The average age at surgery was 27.6 years, and follow-up averaged 8.0 years. Results: The postoperative incidence of osteoarthritis in patients with no preoperative degenerative changes was 21.8% (12 of 55 patients).

123

S128 The incidence of degenerative joint disease of the glenohumeral joint was significantly associated with older age at first dislocation and at surgery, increased interval time from the first dislocation to surgery, increased number of preoperative episodes of dislocation, increased number of anchors used at surgery, and status of the labrum at surgery. A higher number of preoperative dislocations, a greater length of follow-up, and reduced external rotation in abduction influenced Rowe and Constant scores. Conclusions: The number of anchors used and the state of the labrum are the most important factors predisposing to the development of radiographic degenerative changes. Longer follow-up investigations are needed to draw meaningful conclusions.

P15-1495 The arthroscopic bone needle. A new, safe and cost-effective technique for rotator cuff repair H. Frick1, M. Volz1, M. Haag1, J. Stehle1 1 Sportklinik Ravensburg, Ravensburg, Germany Objectives: Reconstruction of a rotator cuff tendon tear using transosseous sutures has been time proven when surgery was done open or mini-open and has the advantage of no implants and cost effectiveness. Arthroscopic rotator cuff repair is less invasive, but suture anchors are expensive. This abstract will introduce a novel technique for treatment of arthroscopic rotator cuff repair combining the advantages of an arthroscopic procedure and transosseous sutures. The purpose of the study was to evaluate the clinical results, patient satisfaction, re-rupture rate and hardware costs of this procedure. Methods: 66 patients with a tear of the supraspinatus tendon were treated with the Arthroscopic Bone Needle from 08/2008 to 11/2009. 60 patients were evaluated about 1 year after surgery with the Constant Score (CS). Additionally, patient satisfaction and complications were evaluated. The hardware costs of a supraspinatus reconstruction using the Arthroscopic Bone Needle were documented and compared in each case to the hardware costs using the estimated number of suture anchors that would have been necessary. A separate study was performed to evaluate the re-rupture rate after supraspinatus reconstruction: 20 consecutive patients (operated from 07/2010 to 01/2011) had an MRI 3 months postoperatively and were evaluated by independent radiologists. Results: The average CS at follow up was 73 (SD 12) which equals a normalized CS for age and gender of 92% (SD 15). This represents a very good clinical result. 56 patients (93%) were satisfied or very satisfied with the surgery. One adhesive capsulitis occurred in this series but no axillary nerve injury or fracture of the greater tuberosity occurred. 2 patients (10%) had a re-rupture of the reconstructed tendon in the MRIs. The hardware costs of a supraspinatus reconstruction were reduced by 80% by using the Arthroscopic Bone Needle compared to suture anchors (€ 121 vs. € 600). Conclusions: The Arthroscopic Bone Needle technique proved to be a new, safe and cost-effective method with good clinical results and low re-rupture rate for the repair of rotator cuff ruptures.

P15-1501 Translation and cultural adaptation of the American shoulder and elbow surgeons standardised shoulder assessment form (ASES) for evaluation of shoulder function to the Turkish language D. Celik1, A. Atalar2, M. Demirhan2, A. Dirican3 1 Istanbul University, Faculty of Istanbul Medicine, Department of Orthopedics and Traumatology, Istanbul, Turkey, 2Istanbul University, Faculty of Medicine, Department of Orthopedics, Istanbul, Turkey, 3Istanbul University, Faculty of Medicine, Department of Biostatistics, Istanbul, Turkey

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: The American Shoulder and Elbow Surgeons Standardised Shoulder Assessment Form (ASES) is a standard shoulder assessment form, comprised of objective and subjective sections, prepared by shoulder and elbow surgeons. The purpose of this study is to translate the subjective part of the American Shoulder and Elbow Surgeons Standardised Shoulder Assessment Form (ASES) into Turkish, and set forth its cultural adaptation and validity. Methods: The original version of the ASES score was translated into Turkish by in accordance with stages recommended by Guillemin. Sixty-three patients (average age: 48.2 ± 13.4, range: 18–74 years) suffering from different shoulder complaints were included in the study. The ASES form was completed twice at 3–7 day intervals for test–retest reliability. The interrater correlation coefficient (ICC) and the internal consistency coefficient (Cronbach Alpha) were used to calculate test–retest reliability. Patients were asked to answer Short Form 36 (SF-36) and the Shoulder Pain and Disability Index (SPADI) for correlation; results were analysed using Pearson’s Correlation test. Results: Test–retest reliability ASES pain, function and total score were 0.95, 0.86 and 0.94 respectively. The Cronbach Alpha Coefficient for the total ASES score was 0.88. The correlation between the total ASES score and the total SPADI score was -0.82, the correlation coefficient between ASES pain and SPADI pain was -0.79 (p \ 0.000), and the correlation between ASES function and SPADI function was -0.53 (p \ 0.000). The highest correlation was between ASES and SF-36 bodily pain, and ASES and SF-36 mental health (r = 0.64, r = 0.56 p \ 0.000), and the lowest correlation was between ASES and SF-36 physical component score and ASES and SF-36 social function (r = 0.28, r = 0.33 p \ 0.000). Conclusions: The Turkish version of the ASES form is a valid and reliable shoulder assessment form that can be used for numerous shoulder disorders. Keywords: Subjective shoulder scores, ASES, Turkish validation

P15-1517 Management of stiff shoulder: a prospective multicenter comparative study of 6 major techniques used, about 235 cases P. Gleyze1, P. Clavert2, T. Benkalfate3, C. Charousset4, P.-H. Flurin5, T. Georges6 1 Albert Schweitzer Hospital, Orthopaedic and Arthroscopic Unit, Colmar, France, 2Hoˆpitaux Universitaires De Strasbourg, Strasbourg, France, 3Clinique De La Sagesse, Rennes, France, 4Institut Osteo Articulaire Paris Courcelles, Paris, France, 5Centre Chirurgie Orthope´dique Et Sportive, Merignac, France, 6ATOL, CHU, Nancy, France Objectives: The authors wanted to evaluate and compare the therapeutic power of different techniques used for the treatment of stiff shoulder and specially the place of arthroscopic capsulotomy and to propose a standard therapeutic algorithm. Methods: A comparative multicentric prospective study (passive FF reduced of more than 30) has been realized: T1: Conventional rehabilitation with respect of pain limit T2: Self exclusive rehabilitation over pain control T3: Self rehabilitation controlled by a physiotherapist T4: T1 + Loco-regional anesthesia block T5 T1 + Capsular distention T6 Endoscopic capsulotomy The therapeutic power of each technique was evaluated every day during a 6-weeks period, and then at 3-, 6-months and a final review with evaluation of the subjective and objective criteria i.e. constant score, goniometric measurements. Results: Total 235 cases (T1: 58 cases, T2:59 cases, T3: 31 cases, T4: 11 cases, T5: 31 cases, T6: 45 cases) had patient follow up during 18-months (3–21).

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conventional rehabilitation; (T1) improved the pain and passive motion. Overall the pain techniques (T2 & T3) give a quicker improvement of both day and night pain (S6-p \ 0.05) during the first 6 weeks. Loco-regional anesthesia (T4) and distention (T5) can accelerate the improvement. A capsulotomy cannot give faster pain relief but can improve passive motion at between 3- and 6-months (p \ 0.05). If we consider T1 as the gold standard, the relief of the pain barrier improves the final result by 12%, loco-regional anesthetic block or distention of 3–6% and capsulotomy 15% this means 3% more than with self-rehabilitation. Conclusions: The authors propose, for all symptomatic stages no matter what stage of the evolution: • 1- to 3-months: self-rehabilitation cominated with conventional rehab • 3- to 6-months: in case of failure of treatment, loco-regional anesthetic block or distention can sometimes be considered • 6-months: in case of failure, an endoscopic capsulotomy can be considered.

P15-1518 Radiologic comparative analysis of anatomical bony reconstruction after total shoulder arthroplasty; interest of a neck osteotomy with a fixed angulation of 135: a series of 64 cases P. Gleyze1, K. Elkothi2, J. Kany3, D. Katz4, P. Sauziere5, P. Valenti5 1 Albert Schweitzer Hospital, Orthopaedic and Arthroscopic Unit, Colmar, France, 2Avenue Mermoz, Lyon, France, 3Clinique de l’Union, Toulouse, France, 4Clinique du Ter, Ploemeur, France, 5 Institut de la Main, Clinique Jouvenet, Paris, France Objectives: The variability of the anatomy of the proximal humerus, has lead to develop new designs of shoulder arthroplasty which take into account the inclination, the retroversion and posterior and medial offset. We hypothesized that a fixed humeral neck osteotomy of 135 with a prosthesis which offered variable diameter, thickness and excentric humeral head would be able to restore the anatomy. (This work compares the anatomical reconstruction obtained by 2 types of glenoid implants with a fixed humeral neck osteotomy of 135). Methods: A radiographic comparative analysis between normal and operated shoulder was performed on 64 patients who sustained an anatomical shoulder arthroplasty. An anteroposterior view in neutral rotation and an axillary view under fluoroscopy were analysed by an independant operator. Results: Humeral stem was 1.6 in varus, with a 16 of retroversion of the head. The glenoid implant was well centered (overloaded of 8%). Compared analysis with controlateral intact side. Significant differences were pointed out on retroversion (op side: 16/cl: 32, p \ 0.05), and on the distance between the top of the head and the top of the great tuberosity (op side: 2.8, cl: 4.7 mm, p \ 0.05). No differences for: cervicodiaphyseal angle (op: 135, 134.6), distance between the center of rotation of the head and the glenoid surface (op: 23.15/cl: 24.6 mm), lateral aspect of the great tuberosity/ center of the glenoid (op side: 29 mm/cl: 28 mm) (the) center of humeral head/axis of the humerus (op side: 4.4/cl: 4.9 mm), humeral axis/lateral border of the acromion (op side: 8.3 mm, cl: 8.2 mm) and sub acromial space (op side: 8.2, cl: 8.4 mm). No differences between the two models of glenoid implant. The stability of all measurements was correlated to the low variation of the neck osteotomy angle imposed by the ancillary system at 135. The variation coefficient was effectively inferior to 2. Conclusions: This work demonstrates the importance of a fixed angle neck osteotomy to fully use the potential of adaptation of the implants for anatomical reconstruction.

S129 P15-1520 Lack of active external rotation and massive cuff tear: assessment and therapeutic options: retrospective review of 20 cases P. Valenti1, P. Gleyze2, O. Boughebri3, A. Kilinc4 1 Institut de la Main, Shoulder Department, Paris, France, 2Albert Schweitzer Hospital, Orthopaedic and Arthroscopic Unit, Colmar, France, 3Orthope´die, Hoˆpital Prive´ Armand Brillard, Paris, France, 4 Institut de la Main, Orthope´die, Paris, France Objectives: In massive posterosuperior cuff tear, the extension of the lesion to the teres minor is rare but can explain a lack of active external rotation. In Erb’s palsy, Episcopo in 1934, described a rerouted of teres major and latissimus dorsi around the humerus to restore active external rotation. The goal of this retrospective study was to report the result of Episcopo procedure performed Isolated or combined with a reverse shoulder arthroplasty to restore active external rotation. Methods: 20 patients (15 female, 5 men), average age 70 years olds (54–82) were unable to maintain the arm at the side in neutral rotation. 4 patients (group 1) had a complete active anterior elevation and 16 (group 2) had a pseudoparalysis shoulder with an antero superior escape of the humeral head. MRI imaging showed an atrophy of the teres minor with a fatty degeneration as stage 3 or greater (according Goutallier et al. classification score) for supra and infraspinatus muscles. Episcopo procedure through a deltopectoral approach was performed isolated for 2 cases and combined with a reverse shoulder arthroplasty in 16 cases. 2 patients refused any surgical procedure. Results: The mean follow up was 29 months (12–72). For the 2 patients operated in group1, External rotation the arm at the side improved from -20 to 30 and the horn blower sign disappeared without any change of active anterior elevation. For the 16 patients (group 2) Forward elevation improved from 67 pre operatively to 116 post operatively (p \ 0.0003); external rotation the arm at side improved from -5 to 25 (p \ 0.001); external rotation in 90 of abduction improved from -10 to 36 (p \ 0.00005) Medial rotation score improved from 4.9 to 5.5 points but was no significant. Pain increased from 5.4 to 14.1 (p \ 0.00006); Daily activity score progressed from 4.7 to 11 points (p \ 0.0004).Strength improved from 1.58 to 6.33 (p \ 0.0002). The Constant score increased from 25 to 56 points (p \ 0.00004). 10 patients were very satisfied, 6 were satisfied and two were disappointed. Conclusions: Clinical examination (active and passive ROM) should be correlated with a CT scan or MRI to analyse systematically a lack of external rotation. The approach, the direction, the final positioning of the tendon transfer should be discussed. Isolated Episcopo procedure through a short approach at the superior medial part of the arm is sufficient to restore active external rotation the arm at the side and the arm in 90 of abduction. Reverse shoulder arthroplasty should be combined with an Episcopo procedure to restore both (anterior elevation and external rotation) in pseudoparalytic shoulder with an external lag sign.

Elbow/hand

P16-29 Surgical treatment with suture-bridge or double row in triceps avulsion H. Valencia-Garcı´a1, A. Lopez-Hualda2, R.M. Egea-Gamez1, P. Gonza´lez-Onandı´a2, H. Fahandezh-Saddi2, F. Moreno-Coronas2 1 Hospital Universitario Fundacio´n Alcorcon, Traumatology, Alcorcon, Spain, 2Hospital Universitario Fundacio´n Alcorcon, Alcorcon, Spain Objectives: The triceps tendon ruptures are rare (2% of all tendon injuries and less than 1% of the breaks in the upper limb). The most common fracture is the avulsion of the bone-tendon insertion.

123

S130 Conservative treatment is reserved for partial avulsions. Fixation and double row anchors used in rotator cuff repair provides better bonetendon junction, supports biological healing and restore normal anatomy. We present a case of distal triceps avulsion treated with a reintegration with anchors at ‘‘double line’’ explaining the technique and excellent results. Methods: Patient 34 years old, goes to hospital after accidental fall left arm retraction and full extension of elbow. He has pain and swelling over olecranon region and inability to extend the elbow against gravity. On examination, palpating a discontinuity at 2 cm proximal to the olecranon tip. Campbell-Thompson test positive amended (elbow at 90 and forearm relaxed, does not cause compression of the triceps elbow extension). Radiological study can be seen a fragment of cortical signs suggestive of the scale and the ultrasound confirmed complete avulsion. Under axillary plexus block and ischemia following the preventive arm, we proceed by cleaning the posterior longitudinal fracture focus. Reintegration is performed with 4 anchors (Panalok) with absorbable suture placed as a double row stitched on themselves and then sutured back uncut crossed over the other row. Finally, the interval is closed with stitches side, we proceed to hemostasis without ischemia and wound closure in layers, ending elbow immobilized at 90 with a brace for 3 weeks. After removal of the splint proceeded to manipulation-assisted and 5 weeks was initiated active muscular work. Results: After 4 months of intervention, the patient has a range of motion of 4–135, no pain in daily activities and previous work with a force of 5/5 on manual testing and a result regarded as excellent. It does not require physical removal. Conclusions: Avulsion of the triceps can go unnoticed. Differential diagnosis sprain, bursitis, olecranon fractures or radial head. The mechanism of injury may be a slowdown superimposed on an eccentric contraction, with or without tendon injury envelope. May be associated with systemic disorders such as chronic renal failure with secondary hyperparathyroidism, hypocalcemic tetany, rheumatoid arthritis, osteogenesis imperfecta, anabolic steroid use, diabetes insulin dependent or local factors such as local steroid injections or bursitis. The treatment is surgical repair anatomic within the first 3 weeks. The setting provides a solid anchor, shortens the operating time and suture-bridge distribution increases the contact surface for tendonbone healing. The reconstruction of the footprint of the double row insertion provides greater rigidity, stability and strength, reduces the gap of the track and increases the contact area and contact pressure.

P16-307 Complications associated with single-incision distal biceps tendon repair L.J. Bisson1, Y. Gawai1, W. Wind1, M. Fineberg1 1 University of Buffalo, Orthopaedics and Sports Medicine, Buffalo, United States Objectives: Distal biceps tendon repair is typically either performed through a single anterior incision, or using two incisions (anterior and posterior). No study has focused on the complications associated with the single incision technique. Methods: Sixty-seven consecutive distal biceps repairs using the single incision anterior approach performed from 2003 to 2010 were retrospectively reviewed to determine the incidence of surgical complications. Patients with persistent complications at last clinical follow-up were interviewed to determine the status of the complication. Results: Thirty-nine of 67 (58%) patients experienced at least one post-operative complication, which included transient sensory nerve dysfunction in 34 (51%), infection in four patients (3 superficial and 1 deep, 6%), loss of motion in three (4.5%), re-rupture in two (3%), and

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 hardware complications in 3 (4.5%). Six of seven (86%) workers compensation patients had complications. At final follow-up, 32 of 34 patients with sensory nerve changes had complete resolution and two patients could not be contacted. The incidence of complications for patients having surgery from 0 to 15 days post injury was 41% (11/ 29), from 16 to 42 days was 72% (16/22), and greater than 42 days was 75% (12/16) (p = 0.035) Tourniquet use and type of hardware used were not associated with a statistically significant difference in complications. Conclusions: The most common complication following distal biceps tendon repair using an anterior approach is transient sensory nerve dysfunction, which usually resolves within 6 months of surgery. This surgery should, when possible, be performed within 2 weeks of injury to decrease the incidence of a post-operative complication.

P16-356 Effect of hot and cold applications on elbow proprioception H. Kaynak1, M. Altun1, M. Unal2, M. Ozer1, D. Akseki3 1 Celal Bayar University, Physical Education and Sports, Manisa, Turkey, 2Isparta Sifa Hospital, Orthopaedics and Traumatology, Isparta, Turkey, 3Balikesir University, Orthopaedics and Traumatology, Balikesir, Turkey Objectives: Importance of proprioception in the treatment and prevention of sports injuries has become increasingly clear. Outcome of treatments are increased through proprioceptive rehabilitation, and re-injury risk is reduced. It is also believed that incidence of injury may be reduced by using pre-injury proprioceptive education. Bracing, muscle fatigue, exercises and surgery has been shown to affect proprioception level. Little is known about the effect of heat and cold application on proprioceptive capability which is the most commonly used treatment modalities on sportive rehabilitation. Purpose of this study was to investigate the effects of cold and heat application on elbow Proprioception in healthy volunteers whose activity levels are between Tegner activity levels between five and eight. Methods: A total of 82 healthy volunteers (35 women, 47 men) whose ages were ranged between 18 and 28 years (av: 20.93) were included in the study. Dominant extremity was right side in all subjects and all the subjects were student in High School of Physical Education and Sports with a Tegner activity levels between five and eight. Proprioception of the dominant elbow (right) was measured by using a digital goniometer with the technique of active joint position sense at the beginning of the study, following cold application and hot application with 1 week interval, respectively. Two target angles were defined 30 and 60 of elbow flexion. Statistical analyses were done by using GLM Repeated Measures and Paired Samples t tests, Pearson Correlation. Results: Reproduction errors were significantly decreased following hot application in both target angles (p \ 0.05), which means increased proprioceptive acuity of the elbow. Following cold application, significantly increased reproduction error were found in one of the two target angles (p \ 0.05) as a result of proprioceptive deterioration. Conclusions: We obtained increased elbow Proprioception following hot application but, a decrease following cold in healthy volunteers in whom Tegner activity levels between five and eight. These findings are seems to be promising and may be useful in prevention and treatment of sports injuries, if supported by future studies.

P16-586 Investigation of radioulnocarpal joint load distribution effects of radial shortening osteotomy for the treatment of Kienbo¨ck disease: a biomechanical study A. Turan1, B. Uzun2, M. Unal3, B. Uyulgan2, B. Unver2, I. Gunal2 1 Siverek Government Hospital, Sanliurfa, Turkey, 2Dokuz Eylu¨l University, Izmir, Turkey, 3Isparta Sifa Hospital, Isparta, Turkey

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: We evaluated the effects of radial shortening osteotomy on the radiolunate joint load distribution. Methods: This biomechanical study used standard left wrist models made of solid foam (Sawbones, Malmo¨, Sweden). The radioulnocarpal joint load distribution was investigated in the normal wrist model and 2 mm and 4 mm radial shortening osteotomy models under 14 and 25 kgf loads when the wrist position was neutral. Results: In the normal wrist group, the comparison of the average pressure distributions under the 14 and 25 kgf loads shows that the increase at the ulnocarpal joint loading and the decrease at the radiolunate joint loading were statistically significant (p = 0.012, p = 0.036, respectively). When we compared the respective average pressure distributions after 14 and 25 kgf loading, there were no significant differences between the normal wrist and the osteotomy groups. Conclusions: Consequently, radial shortening osteotomy is not effective in decreasing radiolunate joint loading when the wrist is in neutral position.

P16-611 Intraoperative carpal tunnel pressure in carpal tunnel release: comparison of endoscopic and open methods S. Ejiri1, S. Konno1 1 Fukushima Medical University School of Medicine, Orthopaedic Surgery, Fukushima, Japan Objectives: In the previous conference, we reported the short-term outcomes of endoscopic (ECTR) and open (OCTR) carpal tunnel release for idiopathic carpal tunnel syndrome (CTS). In our report, we showed that symptoms and electrophysiological findings worsened postoperatively only in some patients in the ECTR group, and that intraoperative carpal tunnel pressure may be responsible for these poor outcomes. The objective of the present study was to compare changes in intraoperative carpal tunnel pressure between ECTR and OCTR. Methods: subjects: A total of 18 hands in 18 patients were investigated. Patients were randomly allocated into the ECTR (Okutsu method) and OCTR groups. The ECTR group included 10 hands in nine patients (all women) aged 51–74 years (mean, 63 years), while the OCTR group included eight hands in eight patients (one man, seven women) aged 44–76 years (mean, 66 years). No significant differences were observed in age or sex between the two groups. Carpal tunnel pressure (ICP) was measured 2 cm peripheral to the distal wrist crease on the radial side of the carpal tunnel using a catheter pressure transducer (Millar Instruments, Inc.). ICP was measured continuously during surgery from catheter insertion to completion of decompression. Results: The mean ICP before and after decompression were 28.6 and 6.3 mmHg, respectively, in the ECTR group and 37.8 and 0 mmHg, respectively, in the OCTR group, and there were no significant intergroup differences. Intraoperative ICP peaked at sheath insertion in the ECTR group and at insertion of the raspatory in the OCTR group. In five of the 10 hands in the ECTR group, the maximum ICP was higher than the measurement limit of the measuring device. In the other five hands, the mean maximum ICP was 181.0 mmHg, while in the OCTR group the mean maximum ICP was 61.3 mmHg, indicating a significant intergroup difference (P \ 0.01). Conclusions: Although there have been previous studies that measured ICP before and after carpal tunnel release, few studies have continuously measured ICP during surgery. Application of high pressure on the median nerve during surgery may worsen the shortterm outcome following ECTR.

S131 P16-693 Three dimensional analysis of the insertional footprints of the triceps brachii A. Bachoura1, K. Sasaki1, S. Kamineni2 1 University of Kentucky, Orthopaedics, Lexington, United States, 2 University of Kentucky, Orthopaedics and Sports Medicine, Lexington, United States Objectives: To accurately and objectively determine the dimensions and the shape of the triceps brachii footprint. This information may provide valuable knowledge for surgeons during triceps repair procedures. Methods: Twenty-one elbows in eleven cadavers were dissected and the triceps preserved. A Faro laser arm device was utilized to probe and digitize the proximal ulna and its respective triceps insertional footprint. The digitized footprints were then analyzed using a threedimensional computer inspection software. A center point was determined as a reference point on the footprint half way between the proximal to distal height in line with the trochlear notch of the ulna. The medial to lateral width, the width from the central point to the medial end, the width from the center point to the lateral end, the proximal to distal length and the area of the footprint were measured and the shape of the footprint qualitatively described. An unpaired T test was used to determine whether a difference in means existed between males and females. A paired T test was used to determine whether a difference in means existed between the left and right sides. A p value less than 0.05 was considered statistically significant. Results: There were six male (11 triceps) and five female cadavers (10 triceps). Eleven triceps were right sided and ten were left sided. The mean total width, lateral extension width, medial extension width, height and area were 53.7, 27.7, 25.7, 16.3 mm and 519.6 mm2 respectively. Significant differences were found between male and female specimens for the overall width of the triceps (p = 0.033), the width of the lateral extension from the midpoint (p = 0.021) and the area of the lateral extension from the midpoint (p = 0.029). There were no statistically significant differences in any dimensions when comparing the left and right sides. In all cases, the shape of the triceps footprint was described as ‘‘dome-shaped’’ with bands extending medially and laterally along the olecranon. Conclusions: The size difference between male and female triceps appears to be dictated by the medial to lateral width of the footprint and not the proximal to distal length. When the center of the triceps is referenced along the plane of the trochlear notch, the width of the lateral extension appears to be responsible for the observed difference.

P16-1104 Positive effects of autologous conditioned plasma (ACP) on the healing of human epicondylitis O. Pu¨tz1, P. Klein1, H. Dewitz1, P. Scha¨ferhoff1 1 Media Park Clinic, Department of Orthopaedics and Sports Medicine, Cologne, Germany Objectives: Recent research has indicated that autologous platelet rich plasma (PRP) plays a central role in orthopaedic procedures. There are several techniques and systems to produce platelet rich plasma. Consequential there are variable compositions with different effects. Because of their anti-inflammatory, regenerative and modulating capacity due to their content of growth factors and mediators we made a prospective study to assess the effect of autologous conditioned plasma (ACP) on the clinical outcomes in the treatment of human epicondylitis. Methods: 20 consecutive patients with a human epicondylitis were treated with autologous conditioned plasma (ACP). The inclusion criteria were at least 6 month of unsuccessful conservative treatment (shockwave therapy, corticosteroid injections and physiotherapy) and

123

S132 a pain on the visual analogue scale (VAS [ 6). The treatment regimen was 5 injections per case with an injection rate of 1 per week. Validated Outcome measures included: VAS and DASH score. Patients were evaluated at baseline (before the injection) as well as 3 and 6 months after treatment. Additionally the VAS was evaluated before every injection. Validated Radiological Follow-up: a magnetic resonance imaging (MRI) examination was performed pre- and posttreatment to evaluate tendon regeneration. With reference to the reproducibility the physiotherapy were standardized and realized in the same rehabilitation-center. Results: The results of our study with autologous conditioned plasma (ACP) in patients with epicondylitis confirm the trend of international studies with PRP. After a short increase on the pain scale, the patients describe a significant reduction in pain and activities of daily living (ADL) at the end of the injection series. A further improvement was observed after the 3 and 6 months follow-up (VAS/DASH pre: 9/71.7 to post 0.5/11.4). The MRI follow-up with an average of 6,7 weeks after first injection showed no significant enhancement regarding the tendon regeneration. Conclusions: Clinically, patients after ACP therapy for human epicondylitis improve significantly in VAS and DASH score. Better results were found in patients with partial tears at the insertion of the tendon. Although the patient described an enhancement the MRI follow-up could not detect an significant improvement. Further studies with control groups are essential and need to be done in blinded randomized prospective studies to proof these preliminary results.

P16-1326 Histological and biochemical analysis of collagen types I and III in transverse carpal ligament in patients with carpal tunnel syndrome (CTS) A. Kunamneni1, K. Sasaki1, A. Sinai2, S. Kamineni1 1 University of Kentucky, Orthopaedics and Sports Medicine, Lexington, United States, 2University of Kentucky, Microbiology Immunology and Molecular Genetics, Lexington, United States Objectives: The objective of the present study was to biochemically and histologically examine the transverse carpal ligament (TCL) harvested during open carpal tunnel release in order to identify the roles of collagen types I and III in patients associated with carpal tunnel syndrome (CTS) between women and men with increasing age. Methods: This study was approved by the ethical committee of UK and Good Samaritan Hospital, Lexington and informed consent was obtained from all participating patients before specimen collection. Biopsy samples of TCL were collected from 10 patients [these subjects were divided into two groups (A & B) of five, with equal numbers of men and women in each group according to gender] undergoing surgery for idiopathic CTS and these samples were divided into 2 pieces. One piece was fixed in 10% formalin immediately for histological analysis and the second was used for biochemical analysis. Specimens were embedded in paraffin, cut into 10 lm-thick sections, and stained with hematoxylin and eosin (HE) and Picrosirius Red (PSR). To visualize the birefringent collagen in PSR staining, a Nikon ECLIPSE EP200 microscope was fitted with a polarizing filter, Slides were then captured with a Spot RT slider cooled CCD camera as digital images. The tissues were homogenized in 15-fold excess of extraction buffer, and centrifuged, and preserved at -80C. Total protein content in the sample was determined using BCA Protein Assay Kit. Constant total protein for each sample was resolved by 7% SDSPAGE and transferred to nitrocellulose membrane using a semi-dry

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 transfer cell apparatus. Western blotting (WB) was done with anticollagen I and III antibodies. Goat anti-rabbit IgG-perixodase was used as the secondary antibody with ECL as the detection system. For each independent sample, WB was done at least in duplicate. For semi quantification of Western signals, the densities of specific antibodies were measured with Image J. Results: H&E stained TCL sections of CTS patients showed a dense connective tissue with aligned fibroblasts intermingled between collagen fibers and few areas of fibrosis in both women and men. PSR staining of histologic sections was used to assess collagen deposition in the TCL. Image analysis quantification of PSR staining of the ligament revealed a reduced total renewal rate of collagen types I and III with increasing age in women and men. In TCL, men had significantly higher collagen I (P \ 0.01) and less collagen III (P \ 0.05) than women by WB analysis. The ratios of collagen types III & I was significantly increased (P \ 0.05) in women than men. Also, WB analysis demonstrated that the collagen types I and III levels decreased almost linearly with increasing age in both women and men, similar to PSR staining-polarization microscopic data. Conclusions: Further accumulation of this kind of information would eventually be helpful in understanding the intrinsic causes of agedependent degenerative alteration of TCL.

P16-1354 Treatment of mild and advanced cases of elbow OA with arthroscopic debridement and intra-articular hyaluronic acid injections S. Kamineni1, D. Patten2, Z. Wani3, R. Yoshida3 1 University of Kentucky, Orthopaedics and Sports Medicine, Lexington, United States, 2Imperial College London, Milton Keynes, United Kingdom, 3University of Kentucky, Lexington, United States Objectives: Middle aged as well as elderly patients are often affected by elbow arthritis, primary degenerative or post-traumatic. This study investigates the efficacy of arthroscopic debridement plus/minus intra-articular hyaluronic acid (HA) injections with respect to pain relief, arc of movement, and functional improvement in 30 elbows with osteoarthritis. Methods: 30 elbows were treated for posttraumatic (n = 12) or primary degenerative (n = 18) osteoarthritis of the elbow by arthroscopic debridement. HA injection protocol was either preoperative (6 cases), postoperative (n = 4), combined pre- and postoperative (n = 5) intraarticular HA (Synvisc) injections, or without additional Synvisc injections (n = 9). A clinical examination and Mayo elbow performance score was conducted at an average of 15 months (range 12–18 months) post-operation. The results were statistically analysed with the Mann–Whitney and Wilcoxon tests. Results: Intra-articular cartilage changes were observed to be mild fraying (n = 7), significant fraying/fibrillation (n = 9), and significant fibrillation with areas of bare bone (n = 14). The treatment resulted in statistically significant pain reduction for both posttraumatic and primary degenerative OA groups. Pain relief was significantly better in the group with exposed bony areas following debridement alone, compared to the group without visible bone (p = 0.005). In patients with exposed bone, pain relief was significantly better without additional intra-articular HA (p = 0.039). The Mayo Elbow Performance Score (MEPS) improved significantly by 30 points (p \ 0.0001 Wilcoxon test), with p = 0.008 in the post traumatic group and p = 0.0005 in the primary degenerative group. There was a greater improvement in the group with exposed bone without additional HA, and an improved trend in the group with mild cartilage fraying with additional HA.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: Hyaluronic acid (HA) is known to stimulate chondrocyte metabolism and have protective effects on cartilage. We combined this potential beneficial property with elbow debridement, as documented in the literature and corroborated in this study, to treat patients with different stages of elbow osteoarthritis. Our findings reveal a trend toward symptomatic and functional benefit when HA is combined with debridement in osteoarthritic elbow joints without exposed bone. There is a symptomatic detriment associated with HA in osteoarthritic joints with exposed bony areas. Our results support the use of HA in combination with elbow debridement in earlier stages of osteoarthritis, but not in advanced cases with exposed bone.

P16-1362 A normalised parameter for the localisation if the radial nerve (PIN) in the proximal forearm C. Norgren1, E. Davidson1, Y. Ruriko1, A. deane1, S. Kamineni2 1 University of Kentucky, Lexington, United States, 2University of Kentucky, Orthopaedics and Sports Medicine, Lexington, United States Objectives: Accurately localising the radial nerve in the proximal forearm has diagnostic and therapeutic implications. However, there are significant variations in anatomy which have traditionally posed problems for defining useable parameters. We attempt to provide a patient normalised localising parameter. Methods: Thirty-five fresh, soft embalmed whole cadaver were studied, with minimal disruptive dissection techniques. We measured the transepicondylar distance (TED), radial nerve distance in different positions of forearm rotation (neutral, pronation, supination) from the lateral epicondyle, and the radial nerve width. Measurements were taken with a digital caliper, along the inter-connecting line between the apex of the lateral epicondyle and tip of the ulna styloid. The measurements were performed by two individuals, on two separate occasions, with inter-observer and inter-occasion blinding. Results: In neutral forearm rotation the radial nerve was located in all specimens, between 65% (4.1 cm) and 105% (6.6 cm) of the TED, measured from the lateral epicondyle. Average point of location is 85% TED from lateral epicondyle. In supination it was located between 50% (3.15 cm) and 90% (5.7 cm) TED. Average point of location is 72% TED from lateral epicondyle. In pronation between 70% (4.4 cm) and 120% (7.6 cm) TED. Average point of location is 101% TED from lateral epicondyle. Conclusions: The location of the radial nerve in the proximal forearm may aid in diagnosis, injections, surgical approaches, and understanding neurological symptoms after injuries to the forearm. A noninvasive localising parameter, normalised to individual patients’ dimensions, as described in the study, may be of benefit in all these aspects radial nerve care. We present an easy to use normalised parameter that can help to localise the PIN in the proximal forearm In neutral forearm rotation, a point located at the 85% TED, along the longitudinal forearm axis line (lateral epicondyle to radial styloid) localises the PIN in the supinator tunnel.

P16-1369 Rupture of the distal triceps brachii tendon: case report N. Geada1, B. Canilho1, N. Lanc¸a1, B. Mota1, R. Barroso1 1 Hospital Nossa Senhora do Rosa´rio, Servic¸o de Ortopedia, Barreiro, Portugal Objectives: Distal triceps brachii tendon rupture is a rare event that represents less than 2% of all tendon injuries. For this reason the diagnosis is often missed in the acute injury.

S133 This lesion is usually caused by a fall on outstretched hand with incomplete extension of the elbow, direct blow or indirect eccentric loading of the triceps. Disruption of the tendon can occur at the tendon attachment to the bone or in the musculotendinous junction, in decreasing order of frequency. Although it’s a rare event, some conditions like chronic renal failure, hyperparathyroidism and anabolic steroids abuse can predispose tendons to rupture. We describe a case of a 32-year-old male, car mechanic and recreational bodybuilder, with a previous history of anabolic steroid abuse (intramuscular injections of trenbulone, testosterone and nandrolone decanoate), otherwise healthy. He presented to the emergency room after a workout session of weight lifting in which he felt a snap and pain in his right elbow during eccentric loading of the triceps. Methods: The patient had visible swelling, local pain, a palpable gap proximal to the olecranon and inability of active extension of the elbow. Radiographs of the elbow didn’t revealed any fracture or avulsion. Surgery exposure accomplished at the same day revealed a complete rupture of the tendon near its insertion point into the olecranon. Repair was done with a nonabsorbable suture passed through a transverse hole in the olecranon and Kessler-like knot in the triceps tendon. The elbow was immobilised with a long arm cast at 60 from full extension for 5 weeks. Progressive active strengthening began at 12 weeks. Results: At his latest evaluation (16 weeks) he presented a range of motion between 10 and 130, almost symmetric strength, a Disabilities of the Arm, Shoulder and Hand score of 9.5, a Mayo Elbow Score of 100 and a Oxford Elbow Score of 45. He resumed his job as a car mechanic. These results are similar to previous case reports and studies, even in those associated with drugs abuse. Conclusions: Although this is a rare lesion and often missed the patient present and past clinical history should increase our level of suspicious. Triceps tendon ruptures usually results in good to excellent outcome when the condition is promptly recognised and treated within the first 3 weeks.

P16-1480 Evaluation of biochemical alterations of collagen types I and III expression in elbow capsule following trauma A. Kunamneni1, K. Sasaki1, A. Sinai2, S. Kamineni1 1 University of Kentucky, Orthopaedics and Sports Medicine, Lexington, United States, 2University of Kentucky, Microbiology Immunology and Molecular Genetics, Lexington, United States Objectives: The objective of the study is to evaluate the biochemical alterations of the elbow capsule following trauma through microscopy and western blot (WB) analysis. Methods: Anterior elbow capsules were collected, with institutional IRB approval at the time of joint release for post-traumatic contracture with trauma duration of 1–18 months. Patients ranged in age from 19 to 58 with a mean of 33 years. There were 2 females and 6 males. The elbow joint capsule in patients with lateral epicondylitis was defined as healthy control. The case of healthy control was 23-years old male. During surgery, resected specimens were obtained from anterior capsules, and these were divided into 2 pieces. One piece was fixed in 10% formalin immediately, and the second was used for WB analysis. Specimens were embedded in paraffin, cut into 8 lm-thick sections, and stained with hematoxylin and eosin (HE) and Picrosirius Red (PSR). To visualize the birefringent collagen in PSR staining, a

123

S134 Nikon ECLIPSE EP200 microscope was fitted with a polarizing filter, Slides were then captured with a Spot RT slider cooled CCD camera as digital images. Constant total protein for each sample was resolved by 7% SDSPAGE and transferred to nitrocellulose membrane using a semi-dry transfer cell apparatus. WB was done with anti-collagen I and III antibodies. Goat anti-rabbit IgG-perixodase was used as the secondary antibody with ECL as the detection system. For each independent sample, WB-blotting was done at least in duplicate. For semi quantification of Western signals, the densities of specific antibodies and calnexin as positive control were measured with Image J. Results: HE-stained sections and polarizing light microscopy (PSR) of control capsule revealed a well organized, parallel arrangement of collagen fibers with intervening fibroblasts. The red to yellow fibers (typical of type I collagen) and the green fibers (typical of type III collagen) were seen in all specimens. All contracture capsules revealed extensive disorganization of the collagen fiber bundle arrangement as assessed with polarized light microscopy. WB analysis for collagen types I and III showed greater presence in the control capsule compared to contracture capsules. We also found that the levels of collagen types I and III were maximum expressed in contracture specimens around 4 months trauma duration. The ratio of collagen I to III demonstrate a relative downregulation of collagen III, which progresses with higher time of contracture formation. Conclusions: The results may be useful for the definition of elbow contracture at the molecular level. Although it is difficult to obtain elbow capsules from individual suffering trauma without a subsequent joint contracture as control, it might be more helpful in understanding the normal and pathological process.

Pelvis-groin

P17-60 The role of intra-articular infusion of long lasting steroids and hyaluronic acid in conservative treatment of degenerative labral tears of the hip joint G.N. Tzoanos1, N. Manidakis1, I. Koutroumpas1, N. Tsavalas2, P. Katonis1, A. Karantanas2 1 University Hospital of Heraklion, Crete, Orthopedic, Heraklion, Greece, 2University Hospital of Heraklion, Crete, Radiology, Heraklion, Greece Objectives: To evaluate the functional outcome of degenerative tears of the hip labrum treated by a series of intra-artcular hip joint infusions. Methods: 14 patients, 8 males and 6 females with average age 38 years with hip pain, clicking or catching sensation of no acute precipitating aetiology. The Flexion Adduction Internal Rotation (FADDIR) test, the Stinchfield and the Flexion Abduction External Rotation (FABER) test were positive. Exacerbation of pain with twisting or pivoting motion on clinical examination. Plain radiographs and magnetic resonance arthrography were used for confirmation of the diagnosis. A protocol of three consecutive fluoroscopic guided hip infusions were performed at 0, 2 and 4 weeks. Each infusion

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 contained a combination of Triamcinolone Hexacetonide (Kenacort) and Hyaluronic Acid (Hyalart). Hip movement at the extremes of motion, squatting, kicking, jumping, twisting, pivoting etc. was discouraged for the duration of treatment. Results: The average follow up was 12 months. Plain films includes an antero-posterior (AP) and true lateral X-rays. In 10 patients demonstrated cam-type FAI, pincer-type FAI or combination of cam and pincer impingement. Magnetic resonance arthrography demonstrated degenerative tears in different portions of the labrum. All patients complied with the treatment protocol. The Visual Analogue Scale (VAS) pain scores were used for evaluation. 12 patients had a good to excellent result with symptomatic relief and return to previous level of activity at the final follow-up visit. Conclusions: Serial intra-articular hip infusions using a combination of long lasting steroid and hyaluronic acid can provide early symptomatic relief in active patients with degenerative labral tears. This can be attributed to the control of synovitis which co-exists with the acute exacerbation of a degenerative labral tear. The long lasting efficacy of this protocol remains to be seen.

P17-95 Outcome of adductor tenotomy as treatment for chronic groin pain in professional male soccer players M. Massada1, R. Sousa1, A. Pereira1, L. Massada2 1 Hospital de Santo Anto´nio, Porto, Portugal, 2Faculdade de Desporto, Universidade do Porto, Traumatologia do Desporto, Porto, Portugal Objectives: Sports-related chronic groin injury is a common clinical condition in the Portuguese Professional Soccer League (PSL). Chronic adductor-related groin pain has been reported to account for between 5 and 15% of all injuries in PSL, resulting insubstantial absences from both practice and competition. This condition remains a clinical challenge because there is little consensus on the diagnosis, investigation, and treatment. Little is published on the surgical treatment when conservative measures fail. Accordingly, the purpose of our study was to determine the outcome of adductor tenotomy as treatment for chronic groin pain in professional male soccer players. Single outcome measures were used to serially assess athletes over a subsequent 2-year period. Methods: We reviewed the outcomes of 17 male soccer players (34 groins), aged 20–38, who underwent adductor tenotomy for chronic groin pain. The criteria for surgery was a history of long-standing (range, 6–30 months) adductor-related pain, refractory to conservative treatment. Questionnaire assessments were made preoperatively and at an average follow-up of 25 months. Results: At follow-up 25 months (range, 4–48) after surgery, all patients regained or increased their pre-injury Tegner activity scores. The mean return to sports was at 7.2 weeks postoperatively, and 12 of 17 returned to competition within a mean of 12 weeks after surgery. All patients returned to full athletic activity. Groin outcome scores improved from a mean of 428.4 ± 48.1 to 454.4 ± 29.0 post-operatively. One patient developed recurrent symptoms following reinjury 19 months post-surgery, and fully recovered following conservative management. Conclusions: In conclusion, adductor tenotomy provides good longterm outcome in the treatment of chronic adductor-related groin pain refractory to conservative treatment.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P17-194 Return to sport following total hip arthroplasty (THA): do we all agree? Y. Kaplan1 1 Hebrew University, Lerner Sports Center, Jerusalem, Israel Objectives: As younger, healthier patients are having THA, a greater emphasis is being placed on postoperative function, activity, and exercise. With recent advances in implant technology and surgical technique, the survival rates for modern prosthetic designs and patients with these high demands are promising. There is no current consensus on the safety of resuming to sport. The purpose of this narrative literature review is to provide a summary of the literature relating to returning to sporting activity following a THA. Methods: An electronic search was conducted up to September 2011, using medical subject headings and free-text words. Subject-specific search was based on the terms ‘‘total hip arthroplasty’’, ‘‘return to sport’’ and ‘‘exercise’’. Results: Ten articles were found to be suitable. Most of them were narrative reviews and expert opinion. One guideline consensus paper was published. Substantial limitations were observed in most of the publications, including small sample size, patient selection, trial quality, heterogeneity of outcome assessments, and potential sources of confounding variables not investigated. Conclusions: The age group that has demonstrated the greatest increase in THA was patients between the ages of 45 and 64. Conflicts emerge with some studies that report lower survival rates for hip and knee arthroplasty in patients participating in high-impact sports. Each sport should be evaluated on its potential risk to a joint replacement, whether from the force of repetitive injury or the possibility of catastrophic failure. Likewise, a surgeon can use techniques, biomaterials, and implants that will maximize an athlete’s chance of success over time. There is a need for a long-term, high-quality, prospective randomized control trial that will compare low versus high impact sports and their effect on the prostheses. Until then, definitive recommendations should be made based on each patient’s expectations, goals and the surgeon’s past experience.

P17-532 Nerve injury after hip arthroscopy with labral repair, a prospective cohort study: is there a learning curve? C. Dippmann1, O. Kraemer2, S. Winge3, P. Ho¨lmich4 1 Hvidovre University Hospital, Orthopedic Department, Hvidovre, Denmark, 2Amager Hospital, University of Copenhagen, Copenhagen, Denmark, 3Copenhagen Private Hospital, Orthopedic Department, Lyngby, Denmark, 4Amager Hospital, University of Copenhagen, Department of Orthopaedic Surgery, Copenhagen, Denmark Objectives: Hip arthroscopy requires distraction of the hip joint. Nerve injury is reported to be the most common postoperative complication and occurs in 0.5–10%. The purpose of the present study was to evaluate the influence of surgical experience on the complication rate. Methods: We performed hip arthroscopy with labral repair, rim trimming and cheilectomy on 2 cohorts of patients: Group A including 49 patients from April 2007 to January 2009, and Group B including 50 patients from March to October 2010 (group B). All patients had follow-up after 8, 26 and 52 weeks. Patients with nerve injury at follow-up examinations received a separate questionnaire at the time of data collection. Results: Traction time was reduced by 25% from 137 min in group A to 103 min in group B. The number of nerve injuries caused by

S135 traction was reduced from 65% in group A to 28% in group B and traction related nerve injury ([52 weeks) had decreased from 4 to 2%. Direct lesions by portal placement were increased from 2 to 16%. One patient (2%) in group A had a lesion of the lateral femoral cutaneous nerve. In group B however 8 patients (16%) had nerve injury related to the portals or affections of branches of the lateral femoral cutaneous nerve. No patient in group A but 1 in group B suffered from transient erectile dysfunction. There was no statistical significant correlation between traction time and nerve injury neither in group A nor group B. Conclusions: Hip arthroscopy with labral repair has a steep learning curve, but with longer experience the rate of nerve injuries caused by traction seems to be minimized. There was no correlation between traction time in subgroups with and without nerve injuries. Because of the proximity of the midanterior portal to the lateral femoral cutaneous nerve or branches of it, there is a risk of nerve injury during portal placement.

P17-597 Active and passive pain coping strategies before hip school: a descriptive cross sectional study M. Hansson1, M. Kla¨ssbo2 1 Hela Kroppen Sjukgymnastik AB, Stockholm, Sweden, 2Centre for Clinical Research, Va¨rmland County Council, Karlstad, Sweden Objectives: Hip disability is most often caused by hip osteoarthritis (OA) and pain is the major reason for seeking health care. Pain coping strategies influence behaviour and the progress of disabilities. The aim of this study was to assess the use of active and passive pain coping strategies with the Swedish version of the Pain Coping Inventory (PCI-S) in persons with hip disability before participating in hip school and analyse, differences between and correlations with gender and other background factors and further to analyse the test– retest reliability of the PCI-S. Methods: Fifty-two persons [41/11 women/men, mean age 63 (standard deviation 8.6)], scheduled for hip school in three different physiotherapy departments, filled in the PCI-S together with a background form and Hip disability and Osteoarthritis Outcome Score (HOOS). The background form included data on the participants’ age, gender, length, weight, other eventually diseases, how long they have had hip disability, if they have both or only one hip affected and how they came into contact with Hip school. All instruments were put in an envelope and sent to the author. To test the reliability ten other persons filled in PCI-S twice, a week apart. A non-parametric descriptive statistics was used to analyse data. To compare background factors, the results from PCI-S and HOOS and difference between men and women the Mann–Whitney U test was used. Mann–Whitney U test was also used to compare gender difference on item-level in PCI-S. To evaluate if there were correlations between pain coping strategies and background factors Spearman rank correlation was used. Test–retest reliability were analysed with intraclass correlation coefficient (ICC 2.1) for kappa and with analyses of variance (ANOVA). The ANOVA model was based on a repeated measures analysis of variance with test occasions as the independent variable. The statistical analyses were conducted using SPSS, SAS and STATISTICA for Windows. Results: The reported use of pain coping strategies varied a lot with no strategy used very often, but the most common used strategies were distraction (active) and resting (passive). Overall slightly more active

123

S136

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

than passive strategies were used, active significantly more often by women (p \ 0.003). The PCI-S showed good test–retest reliability (ICC 0.95 for active strategies in total and 0.88 for passive). Conclusions: The results from this study show that pain coping strategies vary a lot from person to person with no strategy in general used very often. Before hip school slightly more often active compared to passive strategies were used and more often by women. PCIS can be a reliable instrument to assess pain coping strategies in order to learn more about how people cope with pain and improve the quality of rehabilitation.

P17-787 Endoscopic treatment of the external snapping hip syndrome: new surgical technique N. Kunac1, N. Medancˇic´1, D. Starcˇevic´1, D. Trsˇek1, M. Hasˇpl1 1 Special Hospital for Orthopaedic Surgery Akromion, Krapinske Toplice, Croatia Objectives: To present a new endoscopic technique in the treatment of external snapping hip. Methods: In this prospective consecutive study we present two patients with longstanding hip pain and snapping and failure of extensive conservative treatment. The diagnosis was clinical in all cases and anteroposterior pelvis radiographs and magnetic resonance imaging were taken to evaluate the hip joint. Endoscopic release of the iliotibial band was performed with the patient in the lateral decubitus position using 2 portals, the superior and inferior trochanteric (Fig. 1). After establishing the portals, we resected the subcutaneous tissue with radiofrequency (RF) probes and a shaver, and then identified the iliotibial band. We created one longitudinal and three perpendicular cuts in the band with a hooked RF probe (Fig. 2). Snapping was tested at different times during the operation as a guide to adequate release of the iliotibial band. Results: At an average 2-year follow-up, our two patients had no complications and there were no signs of snapping or pain in the hip. Conclusions: We present an endoscopic technique for the treatment of external snapping hip. It is a minimally invasive procedure with little discomfort for the patient. There is no need for extensive

Fig. 1 a Clinical photograph showing the positions of the greater trochanter and the superior and inferior trochanteric portal (left hip). b The inferior trochanteric portal is created first and the arthroscopic cannula with the blunt obturator is introduced and directed to the superior trochanteric portal where a skin incision is made. c A shaving blade is introduced through the superior trochanteric portal to clear the space superficial to the iliotibial band. d The arthroscopic cannula with a camera is moved to the superior portal and a hooked electrocautery probe is introduced at the inferior portal

123

Fig. 2 a The arthroscopic image showing the great trochanter after iliotibial band release and trochanteric bursectomy. b The final defect in the iliotibial band made by a 10 cm longitudinal and three 3 cm perpendicular cuts

postoperative physical rehabilitation and patients can be discharged home on the first postoperative day. Our results are comparable to those reported for open procedures, although longer follow-up and larger series of patients are necessary in making some definitive conclusions.

P17-844 Arthroscopic repair versus debridement in Asian hip labrum tear S.E. Park1, S.K. Kim1 1 Dongguk University International Hospital, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: The purpose of this study was to compare the outcomes of arthroscopic labral debridement with those of labral refixation. And the radiologic parameter was compared within both groups and other patients without hip pain. Methods: From Jan 2010 to May 2011, hip arthroscopy has been practiced. The modified Harris hip score (HHS) was compared with labral repair group and debridement group. Also the modified Harris hip score (HHS), before and after the scope was analysed. The clinical improvement and pain relief was analysed. Independent T test was used for statistical analysis at probability level of 95%. SPSS for Windows (Version 12, Chicago, IL, USA) was used. The radiologic parameter, alpha angle, CE angle, extrusion index was compared with other patients without hip pain. The patients was followed up, and checked the modified Harris hip score (HHS), NRS pain scoring system and subjective clinical satisfaction. Results: From January, 2010 to May, 2011 we prospectively enrolled 48 patients who underwent arthroscopic surgery of the hip for hip labral pathology and met the inclusion criteria for this study. The mean Harris hip score (HHS) improved from 67.68 to 77.90 and on abrasion group, the mean Harris hip score improve from 66.68 to 83.29, on repair group, 68.53–85.69. The clinical results of labral repair was superior to simple abrasion (p = 0.035). Conclusions: The clinical result of labral repair group was superior to that of debridement group. Labral tear should be repaired as much as one can. The radiologic parameter was not correlated with arthroscopic findings between the repair, and debridement group, and was not predictive factor of hip pathology.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P17-942 Septic arthritis of the adult hip: arthroscopic management J. Schro¨der1, M. Hufeland1, N.P. Haas1 1 Charite´, Universita¨tsmedizin Berlin, Center for Musculoskeletal Surgery, Berlin, Germany Objectives: Arthroscopic surgery for primary septic arthritis of the hip is an established procedure for the treatment of pediatric hip infections. In the case of septic arthritis of the adult hip, open surgery remains the treatment of choice. We present a case series of 4 adult patients where arthroscopic management was successful in treating primary septic hip arthritis. Methods: Admission records from 2007 to 2011 were reviewed, and 8 patients treated for septic hip arthritis were identified. Four showed advanced joint destruction due to delayed presentation (average 22 days). They were therefore treated with primary joint resection (Girdlestone’s procedure). Four patients (average 46 years of age) presented in our emergency department with a short duration of acute hip pain (average 2 days) and elevated WBC and CRP. Septic arthritis was confirmed by diagnostic joint aspiration showing purulent fluid with abundant granulocytes. In each patient, standard orthogonal radiography confirmed a well-preserved joint space, and MRI excluded extraarticular abscess formations. Immediate arthroscopic management was performed with 4 portals. In each case, debridement, partial synovectomy, intraoperative high-volume irrigation, postoperative drainage and intravenous antibiotic treatment was performed. All patients underwent a scheduled second-look arthroscopy. Results: The average time of in-patient treatment was 13 days. WBC and CRP levels decreased significantly during the treatment course. Each patient had an unimpaired joint without remarkable destruction of the cartilage when evaluated surgically. The pain level reduced markedly during the postoperative course. Partial weight bearing for 3 weeks was advised in each case. During follow-up (average 33 months), none of the patients showed recurrence or progressive joint degeneration. Conclusions: Septic arthritis of the adult hip is a rare but serious cause of acute hip pain. In patients with preserved joint space and without extraarticular abscess formations, arthroscopic management is a promising minimally invasive option of treatment with low rates of post-surgical morbidity.

P17-987 Reachabilitiy of extraarticular pathologies of iliopsoas tendon and bursitis of greater trochanter in hip arthroscopy: a cadaver study F. Thorey1, S. Budde2, M. Ettinger2, U.-V. Albrecht3, M. Ezechieli2 1 ATOS Clinic Heidelberg, Center for Hip, Knee and Foot Surgery, Heidelberg, Germany, 2Hannover Medical School, Department of Orthopaedic Surgery, Hannover, Germany, 3Hannover Medical School, Institute for Legal Medicine, Hannover, Germany Objectives: Different pathologies leading to psoas tendon pain and chronic bursitis of the grater trochanter are well known. Looking at the literature there is no study known to us describing the reachability and distance to anatomical landmarks. Methods: 12 hips of 6 human cadavers (mean age 49.3) were arthroscopied. The reachability of the iliopsoas tendon at the lesser trochanter and the bursa at the greater trochanter were documented with the camera. Also more ventral portals than the commonly used were tried out to reach the lesser trochanter. After performing arthroscopy on each hip, needles where placed along the extraarticular portals followed by a careful dissection of the area. The distance of the portals to the important anatomical landmarks was measured and documented.

S137 Results: Both the reachabiity of the iliopsoas tendon at the lesser trochanter and the bursa at the greater trochanter throughout the conventional portals could be underlined and a sufficient distance to the important anatomical landmarks was recorded. The mean distance of the distal ventro lateral (DVL) and the wide distal ventro lateral portal (DDVL) to the nervous cutaneous femoris lateralis was 26.8 and 32.2 mm. The mean distance from the both more ventral located portals to the nervous arteria and vena femoral is was 28 mm. Conclusions: This is the first study known to us which describes in detail the reachability of the extraarticular structures underlined by anatomical preparation. In addition to that we could show that more ventral located portals than the usually used DVL and DDVL for tenotomy of the iliopsoas tendon at the lesser trochanter have still enough distance to the important neurovascular structures of the ventral femur and can also be used in addition if necessary. These findings will help orthopaedic surgeons to recognise which structures are being addressed during arthroscopic surgery and may facilitate the development of future hip procedures.

P17-994 Clinical results after arthroscopical treatment of cam impingement F. Thorey1, M. Lerch2, M. Ettinger2, M. Ezechieli2, S. Budde2 1 ATOS Clinic Heidelberg, Center for Hip, Knee and Foot Surgery, Heidelberg, Germany, 2Hannover Medical School, Department of Orthopaedic Surgery, Hannover, Germany Objectives: The clinical diagnosis of a Cam impingement in young adults is often delayed and can lead to osteoarthritis and total hip arthroplasty. Standard diagnostic for this pathology are conventional X-rays, arthro-MRI and clinical examination. In the last years an arthroscopical treatment was even possible in cases with Cam impingement with a large expansion. The purpose of this study was to analyse the outcome after arthroscopical treatment of Cam impingement in our clinic. Methods: We reviewed 262 patients who were treated arthroscopically between January 2007 and December 2008 who had different hip pathologies. Demographic factors, physical examination findings, symptoms, previous treatments, and other surgical procedures that these patients had undergone were analysed. Magnetic resonance arthrography findings and radiographic abnormalities were also recorded. Results: Out of 262 patient eight-seven patients were treated because of a cam impingement. Thirty-one female patients (36%) and fifty-six male patients (64%) with a mean age of 33 years were included. Activity related pain was reported in 83 patients (95%). In 87 patients (100%) a positive impingement sign could be found, and 10 patients (8%) showed a limping. The mean time from the beginning of symptoms to the diagnosis of a labral tear was 9 months. Six patients (7%) had undergone an unsuccessful operative procedure prior to the diagnosis of a Cam impingement. Surgery on another anatomical site has been recommended for seven patients (8%). At an average of 13.4 months after hip arthroscopy seventy-one patients (82%) reported clinical improvement compared to the preoperative status (modified Harris Hip Score). Conclusions: The good results of this study with the clinical improvement of hip pain showed the potency of arthroscopic treatment of cam impingement. However, patients often have a long history of hip pain before the diagnosis of a Cam impingement is confirmed. Therefore, the diagnosis of a Cam impingement should be suspected as the history and radiographs may be non-specific for this diagnosis.

123

S138 P17-998 Hip arthroscopy in the dysplastic hip B. Lund1, S.E. Christiansen1, T.G. Nielsen1, M.U. Vinter2, K. Søballe2, M. Lind1 1 University Hospital of Aarhus, Department of Sports Traumatology, Aarhus C, Denmark, 2University Hospital of Aarhus, Department of Hip Surgery, Aarhus C, Denmark Objectives: Patients with developmental dysplasia of the hip (DDH) can have pathologic changes of labrum and cartilage. After treatment with peri-acetabular osteotomy (PAO) impingement can become symptomatic. The purpose of this study was to describe the arthroscopic findings and the short-term clinical outcome after hip arthroscopy in patients with DDH both prior to and after PAO. Methods: Data is based on a retrospective case series of 27 DDH patients out of 139 hip arthroscopies from 2008 to July 2010. Twenty-seven hip arthroscopies were performed (one bilateral) in 26 consecutive patients (6 male, 20 female; average age, 35 years). Eight of these patients had hip arthroscopy performed prior to a PAO and 19 had previously been operated with PAO. Mean time after PAO was 58 months (3–108). All presented with mechanical hip symptoms and groin related pain. Patients were evaluated by Hip Outcome Score (HOS), mHarris Hip Score (mHHS), and registration of complications and reoperations. Results: For the patients who had arthroscopy prior to PAO the CEangles were 17 (2–24) and in the post PAO group the CE-angles were 35 (27–45). Radiologic signs of acetabular retroversion by prominent ischial spine and ‘‘cross-over’’ sign of the supero-lateral portion of the acetabular rim were seen in 37% patients. Alpha-angle was mean 87. The proximal joint space width was measured on a standing pelvic X-ray was 4.2 mm (2.4–5.6 mm). Eight patients had previous surgery because of Calves–Legg–Perthes (CLP) or slipped femoral head. Labral tears was found in all cases with varying degrees of cartilage damage in the acetabulum or on the femoral head. The labrum was reattached to the acetabular rim in all cases, cartilage lesions were debrided, and micro fracture was performed in three patients with grade IV changes. Labral width was measured and most patients had a large hypertrophic labrum, mean width 8 mm (3–12 mm). All but three patients had resection of the head neck junction for CAM-like deformities. All patients were seen for postoperative follow-up (mean 21 months, range 7–36). The mean post operative HOS score was 77 (51–100). Preoperative mHHS was 49 21–81) and postoperative mHHS in the patients not receiving THR was 71 (49–87). Six patients or 22% went on to total hip replacement within the follow-up period. It was patients with femoral head deformities and joint space\3 mm that went on to total hip arthroplasty. Conclusions: In the dysplastic hip patient, hip arthroscopy can improve hip function, but careful patient selection is necessary. Hip arthroscopy is not recommended in patients with severe deformities of the femoral head or joint space loss indicating progressive arthritis of the joint. Studies with longer follow-up are necessary to investigate whether hip arthroscopy can reduce the risk of osteoarthritis development in DDH patients.

P17-1001 Arthroscopical treatment of acetabular labral tear: an analysis of 262 patients F. Thorey1, S. Budde2, M. Ezechieli2, M. Ettinger2, M. Lerch2 1 ATOS Clinic Heidelberg, Center for Hip, Knee and Foot Surgery, Heidelberg, Germany, 2Hannover Medical School, Department of Orthopaedic Surgery, Hannover, Germany Objectives: The clinical diagnosis of a labral tear of the acetabulum in young adults is often delayed. Standard diagnostic for this pathology

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 are arthro-MRI and clinical examination. We analysed 262 patients with symptomatic acetabular labral tears retrospectively, that have undergone an arthroscopic treatment in our department. Methods: We retrospectively reviewed 262 patients who had a arthroscopic confirmed labral tear of the acetabulum. Demographic factors, physical examination findings, symptoms, previous treatments, and other surgical procedures that these patients had undergone were analysed. Magnetic resonance arthrography findings and radiographic abnormalities were also recorded. Results: In this study 152 female patients (58%) and 110 male patients (42%) with a mean age of 43 years were included. Activity related pain was reported in 227 patients (87%), and 191 patients (73%) had night pain. In 246 patients (94%) a positive impingement sign could be found, and 55 patients (21%) showed a limping. The mean time from the beginning of symptoms to the diagnosis of a labral tear was 11 months. Thirteen patients (5%) had undergone an unsuccessful operative procedure prior to the diagnosis of the labral tear of the acetabulum. Surgery on another anatomical site has been recommended for thirty-nine patients (15%). At an average of 13.4 months after hip arthroscopy 217 patients (82%) reported clinical improvement compared to the preoperative status (modified Harris Hip Score). Conclusions: The good results of this study with the clinical improvement of hip pain showed the potency of arthroscopic treatment of labral tears. However, patients often have a long history of hip pain before the diagnosis of a labral tear of the acetabulum is confirmed. Therefore, the diagnosis of a labral tear should be suspected as the history and radiographs may be non-specific for this diagnosis.

P17-1121 The bleeding pelvic fracture: predictors for angiographic evaluation A. Vellasamy1, F. Hussein1, B. Vaarun1, R. Zhu1, I. Aris1 1 Changi General Hospital, Orthopaedic Surgery, Singapore, Singapore Objectives: A single centre retrospective study to evaluate the characteristics of patients with the bleeding pelvic ring fracture so as to delineate significant predictive parameters for angiography and embolization in the Singaporean/Asian population; to formulate a guideline for the management of the bleeding pelvic fracture and to determine if early angiographic evaluation is the way to go for majority of these fractures with hemodynamic instability. Methods: 121 patients with traumatic pelvic ring injuries were identified over 5 years, from June 2005 till June 2010, from a singlecentre National University Hospital. Out of these 121 patients, 15 patients who underwent angiographic evaluation were identified. 11 out of the 15 had angiography and embolization done, while the remaining 4 only had angiography done. Another group of 29 patients who had not undergone angiography were identified from the main population via age-matched criteria and found representative of the main population via 2 samples T test (Mann–Whitney U test). Systolic blood pressure on arrival in the emergency room, base excess, 24 h fluid and blood requirements, computed tomography findings, injury severity scores (ISS), fracture classifications based on Tile’s and Young’s and Burgess classification systems, haemoglobin levels, coagulation profiles were compared between the 15 angiography patients and the 29 non-angiography group of patients. Case notes and computer records of all patients were analysed. Analysis of data was made using the SPSS Ver. 17. Results: Angiography versus non-angiography Angiography group had lower mean haemoglobin levels (mean = 9.28), had a larger proportion of patients (80%) with contrast blush noted on contrast-enhanced CT scan (CECT), a higher proportion with unstable pelvic fracture patterns as classified by Tile

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 (B and C fractures) (80%), and Young and Burgess (APC II, III, LCII, III, VS, CM) (92.4%) (p \ 0.05). Embolization versus non-embolization Embolized group had higher proportion (81.8%) with hematoma noted on CECT scan, higher proportion with blush on CECT (100%), and higher proportion with unstable fracture patterns (UFPs) (72.7%) as classified by Tile (p \ 0.05). Positive predictive value for embolisation Tile (B and C)—47%, Blush—73% Hematoma on CECT—39% Combined—75%. Conclusions: It is significantly noted that in the bleeding pelvic ring fracture, lower mean initial haemoglobin results, CT findings of hematoma and blush and unstable pelvic fracture patterns seem to herald the need for angiographic evaluation and subsequent embolization.

P17-1174 Do we need a new method to assess femoral head deformity on plain radiographs? J.I. Erquicia1, P.E. Gelber2, X. Pelfort3, M. Tey1, J.C. Monllau4, I. Ormazabal5 1 Instituto Universitario Dexeus, Arthroscopic Unit, ICATME, Barcelona, Spain, 2Hospital de la Santa Creu i Sant Pau, Department of Orthopaedic and Traumatology, Barcelona, Spain, 3Parc de Salut Mar. Universitat Auto`noma de Barcelona, Orthopedic Surgery, Barcelona, Spain, 4Conmed Linvatec Spain, Barcelona, Spain, 5 Institut Universitari Dexeus, Barcelona, Spain Objectives: Cam-type femoroacetabular impingement is a morphologic deformity, that can be assessed by plain radiographs and MRI. Even though MRI study proposed by Notzli is the gold standard, to quantify the deformity, plain radiographs are still a concern. The purpose of this study is to investigate the accuracy and reproducibility of a new method to quantify cam deformity on plain radiographies in the diagnosis of cam-type deformity. Methods: Between 2007 and 2009 350 surgeries had been performed at our institution for FAI due to isolated cam type impingement. We selected twenty cases, according to inclusion criteria (all image studies performed at our institution with the same protocol). For every case we collected a full preoperative study with X-ray study (AP pelvis, Dunn view) and Arthro-MRI study and postoperative with X-ray. Five different observers quantified the alpha angle on MRI as described by Notzli, and on plain radiographs as described in the literature. Then deformity was quantified on axial of Dunn view with a new method, in which pathological angle results from the conjunction of two lines in the center of the femoral head, the first from the point where femoral head lost sphericity, the second from the point where the base of the great trochanter cross the femoral metaphysis. Results: Anesphericity assessment has the best correlation at MRI studies. The new method demonstrated best interobservers correlation than the standard method on plain radiographics. Conclusions: The value of the alpha angle on a plain radiographs presents a very high variability. The system proposed by the authors seems to minimize the inter-observer difference.

P17-1179 Risk of sciatic nerve traction injury during hip arthroscopy: is it the amount or duration? An intra-operative nerve monitoring study M. Safran1, J. Telleria2, J.M. Glick3 1 Stanford University, Orthopaedic Surgery, Redwood City, United States, 2University of Washington, Orthopaedic Surgery, Seattle,

S139 United States, 3University of California San Francisco, Orthopaedic Surgery, San Francisco, United States Objectives: Most surgeons advocate limiting the traction weight and time during hip arthroscopy to 50 pounds and 2 h, respectively. However these parameters are not based on experimental evidence. We prospectively studied nerve injury using intra-operative nerve monitoring to identify the incidence, pattern and pre-disposing factors for nerve traction injury during hip arthroscopy. Methods: During 1998–2001, the motor (MEP) and/or somatosensory (SSEP) evoked potentials were recorded in 76 patients undergoing routine hip arthroscopy. 16 subjects were excluded due to incomplete data. Changes in the posterior tibial and common peroneal nerves were evaluated to assess the effects of traction intensity and time on nerve dysfunction. Baseline electromyographic (EMG) values were recorded prior to the start of the procedure, and the contra-lateral nonoperative limb was continuously monitored to serve as a control. Nerve dysfunction was defined as a 50% reduction in amplitude of SSEPs or MEPs or a 10% increase in latency of the SSEPs; nerve injury was defined as a clinically apparent neurologic deficit in sensory or motor function. When the EMG indicated nerve dysfunction the surgeon was notified, and vital signs, depth of anesthesia, patient position and technical troubleshooting were evaluated. Traction time and weight were continuously monitored with a custom footplate tensiometer. Results: Of the 60 patients (31 females, average age 37 years, range 16–61), 35 (58.3%) experienced intra-operative nerve dysfunction and 4 (6.7%) sustained a clinical nerve injury—none were permanent. The average maximum traction weight in patients who did and did not have nerve dysfunction was 84.0 lbs (range, 50–125) and 72.6 lbs (range, 50–100), respectively (age/sex adjusted; p = 0.04; 95% CI, 1.01–1.08). The odds of nerve dysfunction increased 4% with every one-pound increase in the traction intensity (odds ratio = 1.04). The average total traction time in patients who did and did not have nerve dysfunction was 96 min (range, 42–240) and 82 min (range, 38–160), respectively (age/sex adjusted; p = 0.20). Age and sex were not statistically significant risk factors. Following the removal of traction 6 of the 35 patients with nerve dysfunction returned to baseline EMG levels within 5 min (17.1%), 19 patients by 15 min (54.3%) and 21 patients by 30 min (60.0%); in 14 (40.0%) patients removal of traction did not result in full return of EMG amplitude or latency to baseline range by the end of nerve monitoring. Conclusions: The incidence of SSEP or MEP changes are far more common than clinically identified. The maximum traction intensity is the greatest risk factor for nerve dysfunction during hip arthroscopy. This study did not identify a discrete threshold of traction force or traction time that put patients at a higher risk for nerve dysfunction. Surgeons should attempt to minimize traction time and intensity during hip arthroscopy.

P17-1236 All-arthroscopic labral reconstruction in the hip by use of collagen scaffold - operative technique and early results T. Piontek1, K. Ciemniewska-Gorzela1, A. Szulc2, J. Naczk3 1 Rehasport Clinic, Medical University of Poznan´, Department of Pediatric Orthopedics and Traumatology, Poznan´, Poland, 2University of Medical Sciences, Clinic of Pediatric Orthopedic Surgery, Poznan, Poland, 3Rehasport Clinic, Poznan´, Poland Objectives: It has been shown that the acetabular labrum plays a crucial role in hip joint mechanics and overall physiologic function. Main procedure for labral tear are removing damage tissue and suturing. Same authors describe autograft and allograft labral reconstruction. Autogenous grafts have disadvantages including an additional harvest site with its associated pain and morbidity and,

123

S140 sometimes, poor quality and limited amount of the graft. Porcine collagen matrices have the potential to be helpful for grafting of soft tissue defects. Collagen membranes have been routinely used intraorally for wound dressings as well as for replacement of missing tissue. The aim of this study is to present the operative technique of arthroscopic labral reconstruction in the hip by use of Chondro Gide graft and early results after 1 years. Methods: We performed 9 labral reconstructions using a Chondro Gide graft in patients with advanced labral degeneration or deficiency. We control patients before and after 6, 12, 24 months after operation using MRI scans. The modified Harris Hip Score (MHHS), Athletic Hip Score (AHS) and analogous patient satisfaction score were used to measure outcomes postoperatively. Results: The mean MHHS improved from 67 preoperatively to 90 postoperatively (P = 0.001). The mean AHS improved from 51 preoperatively to 83 postoperatively (P = 0.001). Median patient satisfaction was 8 out of 10. Magnetic Resonance Imaging showed integration of the labrum. One year after Chondro Guide implantation, a nonhomogeneous signal is still evident at the implant site. Conclusions: This study showed that patients who have labral deficiency or advanced labral degeneration had good outcomes and high patient satisfaction after arthroscopic intervention with acetabular labral reconstruction.

P17-1241 Modified Thomas’ test for measuring the fixed flexion deformity of the hip M. Ling1 1 Ministry of Health Holdings, Singapore, Singapore Objectives: The Thomas’ test is a clinical examination technique widely used today to measure the fixed flexion deformity of the hip joint. It has, however, been cited as being inaccurate and unreliable in the estimation of the fixed flexion deformity, lately. The purpose of this study is to prove that by applying the Thomas’ test as originally described, on a group of normal subjects, a pelvic tilt results leading to an over reading of a fixed flexion deformity, should it be present. Methods: 50 normal subjects with no underlying hip or spine pathology were recruited for this study. The Thomas’ test was carried out to demonstrate that a pelvic tilt results. Results: The Thomas’ test resulted in a posterior tilt of the pelvis in all 50 normal subjects. The average degree of tilt was 15.4. Conclusions: The Thomas’ test is an inaccurate way of measuring the fixed flexion deformity of the hip joint. By hyperflexing the contralateral hip, as originally described by Huge Owen Thomas, and obliterating the lumbar lordosis a pelvic tilt is produced. This is likely to lead to an overestimation of a fixed flexion deformity of the hip. We describe a modification where the examiner places one hand under the lumbar spine while flexing the contralateral hip. The flexion deformity of the ipsilateral hip is measured as soon as the lumbar spine contacts the examining hand. Flattening or obliteration of the lumbar spine is carefully avoided.

P17-1365 The effect of hydrotherapy on the return to sports following hip arthroscopy in athletes E. Schilders1, A. Dimitrakopoulou1 1 The Wellington Hospital, The London Hip Arthroscopy Centre, London, United Kingdom Objectives: In the world of high-level athletes, a lot of pressure is put on the medical team to allow a fast return to sports following hip arthroscopy. It is currently unknown which factors can accelerate

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 the return to sports following surgery. This study assesses the effect of a standardized hydrotherapy program on the return to sports. Methods: A mixed group of athletes were assessed retrospectively following hip arthroscopy. The level and type of sports was recorded and categorized in predictable or unpredictable movements. All the athletes were scored prospectively and at 2 year follow up using the visual analogue pain scale (VAS), The Modified Harris Hip Score, The sports score of the Hip Osteoarthritis Outcome score and patient satisfaction. Operative findings and the type of treatment were recorded. Postoperatively all patients received a standard rehabilitation protocol and were divided in 2 groups; group 1 received a standard rehabilitation program + hydrotherapy, group 2 a standard rehabilitation program without hydrotherapy. For statistical analyses an independent samples t test, A Kolmogorov–Smirnov and Shapiro– Wilk test were used. Results: Retrospective study of 82 patients (35 elite athletes and 47 recreational athletes) undergoing 98 hip arthroscopies with a mean followup of 24.7 months. The distribution of pathology was as follows; Femoroacetabular impingement (n = 90), instability (n = 4), isolated labral tear (n = 2), dysplasia and labral tear (n = 2). The patients who received hydrotherapy returned to sports after a mean of 18.8 weeks (SD = 12.4), the athletes who had a standard rehabilitation program without hydrotherapy returned to sports after 22.3 weeks (SD = 13.8) A very significant difference was found between the 2 rehab groups in relation to the return to sports (p = 0.000). High level athletes returned to sports at a mean of 16.3 weeks (SD = 9.63), and recreational athletes at 20.8 weeks (SD = 13.9). In sports with predictable movements the mean return to sports is 18.5 weeks, unpredictable movements 19.0 weeks. The mean patient satisfaction was 8.44, The VAS decreased from 6.90 preop to 1.33 postop. The preoperative MHHS was 63.8, postoperative 93.5. The sportscore increased from 55.9 preoperatively to 92.4 postoperatively. Conclusions: In our assessment of variables that influence the return to sports we found that when hydrotherapy is available during the rehabilitation a significantly faster return to sports can be achieved. Other variables such as age, level of sports, type of sports, the preoperative Visual Analogue pain score, Modified Harris Hip Score or the HOOS sports score or intra operative findings did not affect the speed with which athletes were able to return to sports following hip arthroscopy.

P17-1475 Hip arthroscopy for labral tears in acetabular dysplasia; the role of the tonnis angle A. Dimitrakopoulou1, E. Schilders1 1 The Wellington Hospital, The London Hip Arthroscopy Centre, London, United Kingdom Objectives: Labral tears are a common finding in patients with acetabular dysplasia. Currently, the role of arthroscopic treatment and the factors that might influence the outcome of labral treatment are still uncertain. The aim of this study is to assess if the size of the Tonnis angle influences the clinical outcomes in patients with acetabular dysplasia receiving arthroscopic treatment for labral tears. Methods: Thirty consecutive patients with acetabular dysplasia and associated labral tear underwent hip arthroscopy. They were reviewed retrospectively; data was collected prospectively. Inclusion criteria were centre edge (CE) angle \25, Tonnis grade 0, lateral sourcil height C2 mm. The Tonnis angle was measured. Intraoperative findings were recorded and the type of labral treatment: resection or repair. Patients were assessed using the Modified Harris Hip Score

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 (MHHS) pre- and post-operatively after an average of 2 years. A linear regression model was used to assess the relation of the difference in MHHS (preop-postop) and the Tonnis angle (explanatory variable). A linear regression model was also fitted within each of the treatment groups separately. Results: 30 patients; 23 females and 7 males (mean age 37 years, range 23–51) were assessed pre-operatively and at 2 year follow-up. The average CE angle was 21 (range 15–24). The average Tonnis angle was 13.8 (6–25). Eleven patients had a labral debridement and 19 patients a labral repair. A significant (p = 0.0013) relation between the reduction in MHHS score difference (preop-postop) as the Tonnis angle increases was found. Conclusions: This study demonstrates that in patients with dysplasia who received arthroscopic treatment for labral tears, the surgical outcome (MHHS) is depending on the Tonnis angle. Patients with an increased Tonnis angle [19, independent of the type of treatment, have a poorer outcome and possibly, a different treatment option should be considered.

Knee-ACL I

P18-73 Anterior cruciate ligament reconstruction surgery and osteoarthritis; primary reconstructions versus revision reconstructions using allograft material A.J. Kievit1, F.J. Jonkers2, J.H. Barentsz2, L. Blankevoort1 1 Academic Medical Center, Orthopaedic Surgery, Amsterdam, The Netherlands, 2Medical Center Alkmaar, Orthopaedic Surgery, Alkmaar, The Netherlands Objectives: The primary objective of this study is to assess if there is a difference in the amount of knee osteoarthritis in revision ACL reconstructed patients using an allograft material as compared to primary ACL reconstructed patients. Secondary questions are if there are differences with respect to stability and activity between the two groups and if the need for (revision) ACL reconstruction has an effect on the quality of life in the long term. It is hypothesized that there will be more osteoarthritic changes in revision reconstructed patients using allografts than in primary reconstructed patients when controlling for stability outcomes. Methods: For this retrospective cohort study twenty-five patients who have undergone revision anterior cruciate ligament reconstruction using allografts were identified and compared to twenty-seven randomly selected primary anterior cruciate ligament reconstruction patients operated in the same hospital in the same period using the same technique. Main outcome measures were IKDC radiographic osteoarthritis sum score, KOOS, IKDC functional outcome measures, anterior laxity and EQ-5D at follow-up. Results: Median follow-up was 4.7 years for revision reconstructed patients as compared to 5.1 years for primary reconstructed patients. Radiographic IKDC sum scores of osteoarthritis were found to be significantly worse in revision patients with a median of 4 compared to primary patients with a median of 1 (p = 0.016). Significantly worse outcomes were found in the Sport (median 50 VS 85, p = 0.006) and QOL (median 56 vs. 81, p = 0.001) subscores of the KOOS. IKDC functional outcome measures were the same in both groups except for the Pivot shift test (p = 0.037). No differences were found in anterior drawer, Lachman or KT-1000 arthrometer testing. Present day health scores of EQ-5D were worse for revision reconstructed patients (median 70 vs. 80, p = 0.007).

S141 Conclusions: Revision reconstructed patients have more signs of osteoarthritis and worse quality of life than primary reconstructed patients even though they have comparable IKDC success rates and KT-1000 arthrometer laxity test results.

P18-90 Anatomical evaluation of quadriceps tendon: application for the ACL graft T. Iriuchishima1, K. Shirakura1, F.H. Fu2 1 Gunma University Hospital, Division of Rehabilitation Medicine, Maebashi, Gunma, Japan, 2University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: In the anterior cruciate ligament (ACL) reconstruction, hamstrings tendon or patella tendon are normally used. However, the incidence of graft elongation or anterior knee pain have been reported and therefore, use of quadriceps tendon as ACL graft is looked again. The purpose of this study was to evaluate anatomical structure of quadriceps tendon on the detail to use the tendon as ACL graft. Methods: Fourteen cadaveric knees with an average age of 78.8 were used. After careful dissection of skin and fascia, each muscle structure of quadriceps was evaluated. Especially, rectus femoris tendon was measured its widest and narrowest width and distance from proximal end of patella, length and thickness. In some cases, quadriceps was rolled over as one block, and evaluated from knee joint side. Results: Rectus femoris tendon was independently existed, and other muscles (vastus lateralis, vastus medialis and vastus intermedius) were combined each other. The narrowest width of rectus femoris was 15.3 mm, and the narrowest point existed 4.8 mm proximal from the upper end of patella. The average length of rectus femoris was 27.3 cm. The biggest tendon of quadriceps was consisted with vastus lateralis and vastus intermedius tendons. Tendon of vastus lateralis and vastus intermedius were overlapped and strongly connected. The tendon of vastus medialis was normally thin and weak. Conclusions: If only the rectus femoris tendon was used as ACL graft, there is a risk of rupture of the tendon because narrowest point exists close to the patella. Over all quadriceps tendon is big enough to be used as ACL graft. However, the direction of each tendon is different, and therefore, the strength of quadriceps tendon is questionable.

P18-137 Analysis of tunnel widening after double-bundle ACL reconstruction with an outside-in femoral tunnel formation: a study using a remodeled MRI with a tunnel direction Y.S. Lee1, J.A. Sim1, J.H. Kwak1, S.W. Nam1, B.K. Lee1 1 Gachon University School of Medicine, Department of orthopedic surgery, Incheon, Republic of Korea Objectives: This study examined the tunnel widening using a remodeled MRI (Magnetic Resonance Imaging) with a tunnel direction. Methods: Forty double bundle ACL reconstructed patients who underwent post-operative MRI were enrolled in this study. The tunnel widening was examined using a remodeled MRI (Magnetic Resonance Imaging) with a tunnel direction. Site specific analysis was performed according to the depth (the entrance, mid, and exit portion) and wall (anterior, posterior, medial, and lateral walls). The correlation between MRI and widening was also evaluated. Results: The mean tunnel widening of the femoral AM (Anteromedial), femoral PL (Posterolateral), tibial AM, and tibial PL in terms of the most widened diameter were 1.93 (25.4), 2.14 (30.8), 2.56 (32.8), and 3.16 mm (44.5%), respectively. The tibial PL tunnel showed significant widening than the other tunnels. At the entrance, tunnel widening occurred mostly, followed by the order of the mid and exit

123

S142 portion. Correlation analysis of the time interval of MRI acquisition and tunnel widening showed little association. Conclusions: Tunnel widening after a double bundle ACL reconstruction using an outside-in technique with press-fitting of the graft was acceptable compared to previously published studies. The tibial PL tunnel showed the most widening among the 4 tunnels examined with the tunnel entrance being most widened area. Correlation analysis of the time interval of MRI acquisition and tunnel widening showed little association.

P18-222 Anatomy of the anterior cruciate ligament related to hamstrings tendons grafts. A cadaveric study N. Pujol1, P. Boisrenoult2, P. Beaufils3 1 Hopital Andre Mignot, Orthopaedic Department, Le Chesnay, France, 2Hopital Andre Mignot, Chirurgie Orthope´dique, Versailles, France, 3Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France Objectives: The aim of modern techniques for anatomic reconstruction of the ACL is to reproduce ACL footprints, in order to restore anatomy and therefore biomechanics. An oversizing of the graft is in theory possible, disturbing these principles. Hypothesis: There is an oversizing of the hamstrings grafts related to ACL dimensions (footprints and ligament itself). Methods: 22 paired cadaver knees were dissected. ACL dimensions at mid-portion were measured after removing the synovial membrane. Measurements were also taken for ACL footprints. Hamstrings were harvested and prepared in a quadruple strands graft in order to measure the mean circumference. Results: The average tibial insertion site area of the ACL of all cadaver knees was 117.9 mm2 (range, 90–130 mm). The average femoral insertion site area of the ACL was 96.8 mm2 (range, 80–121 mm). The average diameter of the ACL tendon at mid-substance was 6.1 mm (range, 5–7 mm). The average cross-sectional area of the ACL at mid-substance was 29.2 mm2 (range, 20–38.9). The average diameter of 4 stranded hamstring tendons was 6.7 (range, 5–8). The cross-sectional area of 4 stranded hamstrings tendon grafts was on average 35.3 mm2 (range, 20–50). There is a correlation between 4-stranded hamstrings grafts and native ACL dimensions (footprints, ligament at mid substance, P \ 0.01). The cross sectional area of hamstrings tendons is significantly larger than the ACL area at mid-substance (mean 20.9%, p \ 0.05). Conclusions: With current ACL reconstruction techniques, the graft itself is oversized at a mean of 20%, despite a good correlation between the ACL and quadrupled hamstrings tendons, especially among small subjects and women. The question arises whether the anatomic reconstruction of the ACL should fill ACL footprints or mimic the ligament itself. So, future studies are needed to prove the benefits or disadvantages of a narrow or thick graft, placed in an anatomic position.

P18-313 A prospective study of the association between bone bruising and intra-articular injuries associated with acute ACL tear L. Bisson1, J. Marzo1, M. Kluczynski1, C. Buyea1 1 University of Buffalo, Orthopaedics and Sports Medicine, Buffalo, United States Objectives: Bone bruising of the lateral femoral condyle (LFC), lateral tibial plateau (LTP), and medial tibial plateau (MTP) is

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 commonly found on MRI following ACL injury, and may be associated with intra-articular injuries. Little is known about the association between bone bruising and these injuries, as well as other demographic factors associated with bone bruising. Methods: Two-hundred forty-eight consecutive ACL-injured patients who had an MRI within 6 weeks of injury and arthroscopy within 3 months of injury were prospectively studied to determine the association (using adjusted odds ratio (aOR) obtained from logistic regression) between demographic and injury factors, and the presence and degree of bone bruising (categorized as none/minimal, mild, moderate, and severe) and associated intra-articular abnormalities. Results: The average age at injury was 25 ± 11 years. Fifty-three percent of subjects were male, and 74% of injuries were non-contact. Eighty-nine percent occurred during sports, with 59% being level III sports. Ninety-two percent of subjects had bone bruising (76% LFC, 85% LTP, 5% medial femoral condyle (MFC), 27% MTP), 63% had meniscal tears, and 14% had chondral injuries. LFC bone bruising was less likely to have occurred among subjects aged 18–28 [aOR 0.27 (95% CI 0.09, 0.82)] and [28 (0.18 (0.05, 0.61) years compared to subjects \18 years-old. Males had a greater incidence of severe LFC bruising [aOR 15.66 (3.19, 76.92)], as well as a greater overall volume of bone bruising in all areas combined [aOR 6.21 (1.57, 24.49)]. Lateral meniscal tears were associated with the presence of LFC bruising [aOR 2.57 (1.04, 6.32)] and LTP bruising [aOR 3.13 (1.06, 9.23)], but not with lateral bruising severity. Medial meniscus tears were associated with moderate [aOR 8.14 (1.93, 34.27)] and severe LTP bruising [aOR 15.30 (2.34, 100.10)]. Conclusions: Bone bruising is more common and severe in young males, and on the lateral side is associated with lateral meniscal tears. Medial meniscal tears are associated with increasing severity of LTP bruising. The greater combined volume of bone bruising in males could in part be explained by ACL tears occurring with lower energy in females.

P18-410 ACL double bundle linked cortical-aperture tibial fixation: An in vitro biomechanical study J. Nyland1, M. McGinnis1, D. Lee Yee Han1, S. Kibbe1, R. Burden1, D.N.M. Caborn1 1 University of Louisville, Orthopaedic Surgery, Louisville, United States Objectives: This in vitro biomechanical porcine model study attempted to determine the soft tissue tendon graft tibia fixation characteristics of the ACL DB Tightrope SystemTM (Arthrex, Naples, FL, USA). Methods: Eight tibiae with apparent bone mineral density that simulated that of young, human bone and 11.5 mm diameter profundus tendon grafts were used in this study. With an anterior cruciate ligament (ACL) drill guide set at 55 and a 42 mm intraosseous ‘‘length’’ guidewire aligned with the center of the tibial ACL insertion footprint, a FlipcutterTM was used to create 11.5 mm diameter, 27 mm deep tibial sockets. The composite tibial fixation site consisted of a 27 mm socket and a 15 mm guidewire channel. Grafts were positioned in the tibial socket and sutures were tensioned to properly situate Peak implant-graft constructs and optimize extra-cortical button fixation. Tibiae were potted in 7.62 cm diameter, 17.78 cm long tubes and loaded into a 6 of freedom clamp with the servohydraulic device loading vector aligned directly with the tibial socket providing a ‘‘worst case’’ loading scenario. After pre-loading to 25 N, constructs underwent pre-conditioning for 10 cycles (0–50 N, 0.5 Hz), 500 submaximal loading cycles between 50 and 250 N (1 Hz), and load to failure testing at 20 mm/min with load (N) and displacement (mm) data recorded at 10 Hz. Descriptive statistical analysis was performed.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: Displacement during submaximal cyclic loading was (mean ± standard deviation) 2.79 ± 0.88 mm (95% confidence interval = 2.12–3.45 mm). Yield load at failure was 475.2 ± 35.5 N (95% confidence interval = 445.6–504.9 N). Displacement at yield load was 5.96 ± 0.86 mm (95% confidence interval = 5.23–6.68 mm). Ultimate load at failure was 671.4 ± 97.5 N (95% confidence interval = 589.9–752.9 N). Stiffness during load to failure testing was 127.5 ± 14.6 N/mm (95% confidence interval = 115.3–139.8 N/mm). Displacement at failure load was 16.3 ± 5.9 mm (95% confidence interval = 11.4–19.4 mm). Fixation failure during load to failure testing was ‘‘button of device pulling through the tibia cortex’’ (n = 1), ‘‘suture pulling through the graft’’ (n = 3), ‘‘graft loop failure’’ (n = 1), ‘‘Peak implant-graft construct failure’’ (n = 2) and ‘‘combined tibia cortex button slippage-implant failure’’ (n = 1). Conclusions: Tibial fixation using the ACL DB Tightrope SystemTM provides submaximal cyclic loading displacement, ultimate failure loads, and stiffness that is comparable or superior to other established tibial soft tissue tendon graft fixation methods. Clinical studies with long-term follow-up are needed to confirm results.

P18-435 Analysis of anterior tibial subluxation to femur at maximum extension in anterior cruciate ligament-deficient knees R. Kuroda1, D. Araki1, H. Sasaki1, T. Matsushita1, S. Kubo1, M. Kurosaka1 1 Kobe University Graduate School of Medicine, Orthopaedic Surgery, Kobe, Japan Objectives: There has been previously reported abnormal kinematics in the anterior cruciate ligament (ACL) deficient knee. It has been reported that reduction of the tibia relative to the femur with the knee in maximum extension is required to correctly position the tibial tunnel in the sagittal plane in ACL deficient knee. However, irreducible anterior tibial subluxation in chronic ACL deficient knee was also reported. Therefore, we hypothesized that knee kinematics and alignment might change in ACL deficient knee. Antero-posterior translation and anterior tibial subluxation to the femur were measured in ACL deficient knees and compared to contralateral intact knees, and also compared between acute and chronic ACL deficient knees. Methods: Forty-two ACL deficient patients (24 male, 18 female, average ages: 26.5, average time since initial injury: 255 days) were evaluated for this study. No contralateral ACL injury or osteoarthritic change was in all the patients. All patients had a full range of knee motion without flexion contractures. Anterior tibial subluxation to the femur was measured in maximum extension position using True lateral view of fluoroscopy taken at the time of surgery under a general anesthesia in a supine position. Also tibial alignment was compared with associate meniscal injury and posterior tibial slope. Results: Tibial anterior displace significantly increased in chronic cases (p \ 0.05). Time since initial injury has strong correlation to the tibial anterior subluxation (y = 0.0065x + 2.121, R = 0.58, p \ 0.01). Medial meniscus injury also has strong correlation to the tibial anterior subluxation. Posterior tibial slope has no correlation to the tibial anterior subluxation. Conclusions: In chronic group, tibial anterior subluxation to the femur in maximum extension position significantly increased. Our study suggested that sagittal knee alignment might be changed in ACLdeficient knees and that chronic ACL-deficiency may leads to irreducible tibial subluxation. It might be difficult to reduce tibial anterior subluxation completely in such cases. Therefore, caution should be needed especially to position the tibial tunnel correctly in ACL reconstructions for chronic ACL deficient knees.

S143 P18-483 A biomechanical analysis of suture to tendon attachment techniques used during ligament reconstruction B. Bradley1, R. Walter1, D. Isaac1, M. Hockings1 1 South Devon Hospitals NHS Trust, Trauma Orthopaedics, Torquay, United Kingdom Objectives: When performing arthroscopic anterior cruciate ligament reconstruction using hamstrings autograft, suture material is tied to the distal end of the hamstrings tendon in order to control its passage through the tibial tunnel and to allow tensioning of the graft prior to its fixation within the tibia. The aim of this study was to assess the speed and strength of three different techniques of securing the suture to the graft using an animal model. Methods: Suture material was tied to the distal end of fresh porcine digital flexor tendons using three techniques: (1) whip stitch; (2) modified Prussik knot, also known as the double luggage tag knot; (3) a proprietary suture attachment device (WhipknotTM Smith & Nephew). The tendons were then loaded. The load to failure and time taken to attach the suture was recorded. Results: The mean load at failure was 112 N (SD 9.7) for the whip stich technique, 111 N (26.5) for the modified Prussik knot and 136 N (15.9) for the WhipknotTM. Applying the WhipknotTM was significantly faster than securing the tendon end with the modified Prussik technique (8.8 vs. 15.8 s p \ 0.01) which in turn was faster than whipstitching (15.8 vs. 121.5 s p \ 0.01). Conclusions: The load to failure for all three techniques exceeded the calculated loading requirements for intra-operative tensioning of a four tendon hamstrings autograft. Whipstitching is the most commonly employed technique within current clinical practice. This study demonstrates that the modified Prussik technique and WhipknotTM have a significant speed advantage over whipstitching. However the WhipknotTM device incurs a higher financial cost. We would now advocate use of the modified Prussik technique for securing the distal ends of the hamstrings autograft when performing arthroscopic anterior cruciate ligament reconstruction.

P18-530 Anatomic transtibial ACL reconstruction: effect of tunnel placement, size and reamer characteristics N.N. Verma1, S. Bhatia1, K. Korth1, D. Gupta1, B.J. Cole1, B.R. Bach1 1 Rush University Medical Center, Department of Orthopaedic Surgery, Division of Sports Medicine, Chicago, United States Objectives: Dependence of femoral tunnel positioning on tibial tunnel orientation in ACL reconstruction (ACLR) using a transtibial technique has been well established. The purpose of this study is: 1. to identify the impact of tibial reamer size and placement on the creation of anatomic femoral tunnels via a transtibial approach and 2. to identify the reamer design that best preserves tibial tunnel integrity while maintaining anatomic femoral tunnel placement during transtibial drilling. Methods: 8 cadaveric knee specimens were fixed at 90 flexion and neutral rotation. After removing the anterior capsule and patella, native joint anatomy was recorded with the MicroScribeTM digitizer. Tibial and femoral tunnels were drilled in a transtibial ACLR manner using the optimal tibial starting point described by Piasecki et al. Tibial tunnels were drilled progressively with 6, 7, 8, 9, 10, 11 mm reamers. After each reaming, a beath pin was placed in the posterior aspect of the tibial tunnel and digitized. After drilling the 11 mm tibial tunnel, femoral tunnels were drilled with a 10 mm half fluted reamer, followed by a 10 mm full reamer. Each tibial tunnel’s location and geometry relative to the native anatomy were digitized.

123

S144

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Results: In 6 knees, the center of the femoral ACL footprint was first reached by the 9 mm tibial tunnel. In 2 knees, the center was reached with an 8 mm tunnel. The 6 and 7 mm tibial tunnels did not allow for anatomic positioning in any specimen and had errors of 4.75 ± 1.95 mm and 2.94 ± 0.54 mm, respectively. After use of 11 mm tibial reamer, tibial tunnel length was 32.07 ± 2.62 mm. Femoral reaming with a 10 mm full reamer posteriorized the articular aperture of the tibial tunnel 5.44 ± 1.84 mm, resulting in a 59.62 ± 28.1% expansion of tunnel area at the posterior surface. The half fluted reamer preserved the aperture of the tibial tunnel created during tibial reaming. The center of the native ACL tibial footprint was 2.0 ± 0.49 mm (range 1.1–2.7 mm) anterior to the posterior aspect of the lateral mensicus’ anterior horn.

Fig. 1 Activity causing ACL injury and sports ha

ACL footprints Conclusions: Use of the transtibial ACLR technique may result in nonanatomic femoral tunnel placement with tibial tunnel holes smaller than 8 or 9 mm. Creating at least a 9 mm tibial tunnel, however, allowed for anatomic femoral placement. Half fluted reamers may be more advantageous for femoral tunnel reaming, resulting in less posterior tibial tunnel expansion than full reamers and possibly leading to improved graft mechanics. The center of the tibial attachment site was found to be anterior to the posterior aspect of the lateral mensicus’ anterior horn, which has been traditionally described as the anatomic center.

P18-550 Anterior cruciate ligament injury related to physical education S. Sasaki1, Y. Ishibashi1, E. Tsuda1, Y. Yamamoto1, Y. Kimura1, Y. Fujita1 1 Hirosaki University Graduate School of Medicine, Orthopaedic Surgery, Hirosaki, Japan Objectives: Most anterior cruciate ligament (ACL) injuries occur during sports activities. Although the majority of the reports have been focused on athletes, a large number of patients who had ACL injury during physical education classes or recreation are also seen in daily medical practice, and the injury mechanism and the prevention strategy for non-athletes should be studied too. The purpose of this study was to determine the characteristics of ACL injury in physical education classes. Methods: Three hundred sixty-one patients who had primary ACL reconstruction between 2005 and 2009 in our institution were studied. All data were collected through interviews and retrospective review of the medical records and we surveyed age at time of injury, gender, activity causing ACL injury, and sports habits. Results: Eighteen physical education related ACL injuries were recorded during the period, representing 5% of all registered ACL injuries. Five male and 13 female students were injured and all of them visited our outpatient clinic within 1 month after the injury. The mean age of injured patients was 16.3 ± 1.9 (mean ± SD)

123

Fig. 2 Inciting event of ACL injury years, and they included 1 elementary school student, 4 junior high school students, 11 high school students and 2 college students. The most common activity causing the injury was vaulting in 9 patients (50%), followed by basketball in 4 patients (22%), and volleyball, soccer, baseball, dodge ball, and gymnastics in 1 patient each (5.6%). Seven of these 18 patients had some sports habits, while 11 did not. Eight of 9 patients (89%) who were injured when vaulting had no regular sports habits, while 3 of 4 patients (75%) who were injured in basketball had regular sports habits. In other ball games, the majority of patients had some regular sports habits (Fig. 1). All ACL injuries were non-contact injuries and the most common inciting event for ACL injury was a landing (78%) (Fig. 2). Conclusions: The current results indicate that vaulting is a high risk activity of ACL injury in physical education, particularly for students who do not regularly engage in sports.

P18-591 Anatomic anterior cruciate ligament reconstruction with curved guides and flexible reamers. Surgical technique M.A. Ruiz-Iban1, J. Diaz-Heredia1, H. Gomez Santos1, I. Cebreiro1, P. Garcia Cabezuela1, I. Moreno Quintanilla1 1 Hospital Ramon y Cajal, Dep. de Cirugı´a Ortope´dica y Traumatologı´a, Madrid, Spain

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Anterior cruciate ligament (ACL) reconstruction is a very common intervention in sports surgery. Traditional reconstruction with a single tunnel in both femur and tibia with either bone-tendonBone or semitendinous-gracilis grafts has produced excellent functional results in the last 20 years. Although the outcomes have been satisfactory some authors have searched for a more anatomic way of reconstructing the ACL with the hope of improving the clinical outcomes, diminishing the chances of appearance of secondary osteoarthritis and limiting the presence of residual rotatory instability. The objective of this video is to present the surgical technique of an anatomic reconstruction of the ACL using a set of curved guides. Methods: Surgical technique consisted in extraction of the semitendinous and gracilis grafts and preparation of both grafts separately. One tibial tunnel was made using a standard tibial guide and the femoral tunnel was performed using an specific set of curved guides and nitinol wires (the Clancy system) passed through the standard anteromedial portal and flexible drills. This system allows for precise insertion control of the femoral tunnel wires, allowing to anatomically reproduce the exact femoral footprint of the native ACL. The grafts were passed through the tibial and femoral tunnels and fixated proximally with Endobutton and distally with bioabsorbable interference screws. The surgery was performed initially in young patients (younger than 30 years old) with high functional requirements but this technique is used now routinely in all our ACL reconstruction cases. Results: We have performed this procedure in more than 50 patients. All patients had satisfactory early postoperative results with discharge next day and no postoperative wound complications or knee effusion. After 1 year follow up the functional scores were good. Preliminary clinical evaluation did not demonstrate differences in range of movement or clinical instability between previous technique and this new technique. Conclusions: Anatomic single bundle ACL reconstruction with the Clancy guides is a good treatment option in patients with high functional requirements.

P18-622 Arthroscopic treatment for avulsion fracture of the anterior cruciate ligament in seven children, with the arthroscopic pull-out fixation Y. Ohashi1, T. Toratani1, M. Kosaka1, J. Nakase2 1 Kanazawa University, Orthopaedic Surgery, Kanazawa, Japan, 2 Kanazawa University, Kanazawa, Japan Objectives: It is well known that, in case we left untreated displaced Anterior Cruciate Ligament (ACL) avulsion fracture, it set up instability of the knee joint. The purpose of this study was to assess the clinical results of arthroscopic pull-out fixation for the ACL avulsion fractures in 7 children. Methods: Seven patients with ACL avulsion fracture, one girl and 6 boys at an average age of 10.9 years (range 8–15 years), were included in this study. All of them were indicated for surgical treatment on the basis of radiographic evidence of McKeever type III (4 patients) and IV (3 patients) fractures. Five patients injured by skiing, two patients injured by playing basketball. They were underwent arthroscopic pull-out fixation with ENDOBUTTON (Smith & Nephew). Average of follow-up duration is 39 months (range 7–75 months). Postoperative therapy underwent external fixation and non-weight bearing for 2 weeks. Results: All patients achieved normal knee extension and flexion, and bone union at an anatomical position. Lysholm score was 100 in all patients. The average of operative time was 106 min (range 60–145 min). All of them returned to their sport activity.

S145 Conclusions: There are some fixation methods for ACL avulsion fracture. For example, cannulated screw fixation, pull-out fixation and tension band wiring. The cannulated screw fixation has the strongest initial fixed power than other fixations. But, it has some demerits. In case the bony fragment is small or comminuted, fixed power will decrease. There is a fear of bony fragment damage during the fixation. And protruding the screw head from the ACL, it set up intercondylar notch impingement. Furthermore, there is a fear of injury of the epiphyseal plate. There is a report about 21 patient’s bony fragments. According to the paper, it was evaluated by CT that the bony fragment had broken in the case of 13. But 10 cases couldn’t be evaluated the bone condition at X-ray. In this study, 3 patients had comminuted bony fragments. For this reason, we choose the arthroscopic pull-out fixation. To prevent re-dislocate from fixed position, we added the figure-of-eight suture fixation technique. Fibers crossing in the front of ACL, passing through the inside of ACL body was pressing the bony fragment to tibia plateau. We experienced the good results by arthroscopic treatment for ACL avulsion fracture in 7 children. It is consider that an arthroscopic pull-out fixation for ACL avulsion fracture is useful treatment for strong fixation in the damaged condition of the bony fragment.

P18-708 Bespoke anatomic ACL reconstruction. Individualized femoral tunnel placement guided by intercondylar notch ridge anatomy: a 3D CT study J. Robinson1, A. Porteous2 1 The Bristol Knee Group, Orthopaedic Surgery, Avon Orthopaedic Centre, Bristol, United Kingdom, 2Avon Orthropaedic Centre, Southmead Hospital, Bristol, United Kingdom Objectives: Every individual’s ACL femoral attachment may be uniquely defined by the topography of the lateral intercondylar and bifurcate ridges. The purpose of this study was to determine whether intra-operative identification of these osseous ridges leads to reliable reconstruction of the native femoral ACL attachment in both chronic single-bundle (SB) and double-bundle (DB) cases and to measure the variability in ridge morphology amongst individuals. Methods: Pre- and post-operative high-definition 3D, surface-rendered, CT reconstructions of the lateral intercondylar notch were obtained for 25 patients undergoing ACL reconstruction (17 SB, and 8 DB or isolated PL bundle augmentations). Native ACL femoral attachment morphology was defined from ridge topography on the pre-operative CT scans. The centers of the ACL attachment, AM bundle and PL bundle were recorded using the Bernard grid. During ACL reconstruction, soft tissue was carefully removed from the lateral notch wall with RF coblation to preserve and visualize osseous ridge anatomy. For SB reconstructions the femoral tunnel was sited centrally on the lateral bifurcate ridge, equidistant between the lateral intercondylar ridge and posterior cartilage margin. For DB reconstructions, tunnels were located either side of the bifurcate ridge leaving a 2 mm bony bridge. For isolated PL bundle augmentation, the tunnel was sited equidistant between the intercondylar ridge and articular cartilage margin, shallow to the AM bundle. Post-operative 3D CTs were obtained within 6 weeks to correlate tunnel positions with pre-op native ridge topography. Results: The mean position of the centre of the native ACL femoral attachment, defined on the pre-op 3D CTs, was similar to previous in vitro studies: 27% along Blumensaat’s line and 38% of the width of the lateral femoral condyle. However, we noted considerable variability amongst individuals: range 21–33% along Blumensaat’s line and range 31–43% of the width of the femoral condyle. Despite the variability of individual’s ACL attachments, there was a very close

123

S146 correlation between the pre-operative CT localization of the native ACL attachment center and the position of the SB ACL reconstruction tunnel (R [ 0.92): \1.4% difference along Blumensaat’s line and \1.4% difference along width of Femoral condyle. Similar results were observed for DB reconstructions and isolated bundle augmentations. Conclusions: ACL attachment site morphology varies between individuals. Intra-operative localization of the osseous landmarks (lateral intercondylar and bifurcate ridges) in chronic cases appears to lead to accurate anatomical tunnel placement whether using single or doublebundle reconstruction techniques. Due to the variability of attachment site morphology, use of the ridge topography to determine tunnel positions is likely to lead to a more individualized reconstruction compared with use of an o’clock position, offset aimer or image intensifier.

P18-766 Autologous platelet-rich plasma promoted proliferation of synovial fluid mesenchymal stem cells in alginate hydrogel H. Tang1, C.-H. Chen1 1 Keelung Chang Gung Memorial Hospital, Keelung, Taiwan, Republic of China Objectives: Platelet-rich plasma (PRP) has been harvested from blood for tissue regeneration because of the many factors in it which activate osteoid production, collagen synthesis, and cell proliferation. In recent decades, orthopedic surgeon and sports physicians often use autologous PRP combined with fibrin to achieve tissue healing in clinical studies. This study uses that autologous synovial fluid mesenchymal stem cells (SFMSCs) and PRP from black small ear pig, and mixed with alginate. The PRP affect on proliferation of SFMSCs in alginate hydrogel is evaluated. These results could use to alginate hydrogel wrap up injury ligament for repairing in future. Methods: Synovial fluid was harvested from female black small ear pigs (weight about 30 kg). The synovial fluid was centrifuged to remove the supernatant, and the SFMSCs were obtained. For determined epitope profile of the SFMSCs, one million cells were suspended in antibody (CD29, CD34, CD90, CD45, CD44 and CD105). PRP was collected by drawing the whole blood of black small ear pig via neck vessel penetration. The whole blood was centrifuged to form the PRP. The same alginate concentration (1%), different concentrations of PRP 0, 20, 50, 70 and 90% solution was prepared. The 5 9 104 cells/ml of SFMSCs was prepared in above five kinds of the PRP alginate solution. The SFMSCs-PRP-alginate solution dropped into 0.1 M of calcium chloride solution for 2 min to form cell beads-like alginate hydrogel. The cell alginate hydrogel were cultured for 0, 3, 7, 14 and 28 days. At each time point, 10 cell alginate hydrogel were took out, and dissolved with lysis buffer to determine DNA content. Cell viability in alginate hydrogel was observed by confocal assay. Results: The spindle shape SFMSCs were observed. The SFMSCs expressed CD34 (-), CD90 (+), CD29 (+), CD45 (-), CD44 (+) and CD105 (+) was found. SFMSCs from black small-eared pig express as normal stem cell was found. The DNA content increased with culture time when concentration of PRP greater than 50% in alginate hydrogel. The DNA content of 0% PRP in alginate hydrogel decreased with time was observed. The DNA content of 20% PRP in alginate hydrogel decreased before 7 days and increased after 7 days was observed. The DNA content increased with the concentration of PRP in alginate hydrogel was found. Therefore, 90% PRP in alginate hydrogen was the best concentration for cell proliferation. Almost live cells in five kind of alginate hydrogel were observed. The most dense cell number was observed in 90% PRP. Excepted 0% PRP, cell number increased with time in other concentration of PRP was

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 obtained. The PRP enhanced SFMSCs proliferation in alginate hydrogel was proved. Conclusions: In this study, SFMSCs-PRP-alginate complex hydrogel provided good proliferation and viability. It is feasibility for healing partial tear of cruciate ligament.

P18-806 Arthroscopic primary anatomic anterior cruciate ligament reconstruction using patellar tendon autograft: a retrospective study with more than 15 years of follow-up S. Mogos1, J.-M. Fayard2, B. Sonnery-Cottet2, P. Chambat2 1 Orthopaedic Clinical Hospital Foisor, Bucharest, Romania, 2 Orthopaedic Center Santy, Lyon, France Objectives: The aim of this study is to retrospectively evaluate the long term results after arthroscopic primary anatomic anterior cruciate ligament reconstruction using patellar tendon autograft, in patients with at least 15 years of follow-up. Methods: Fifty-seven patients were retrospectively examinated. All patients underwent a similar arthroscopic procedure, identical postoperative rehabilitation and all operations were performed by a single, experienced surgeon. All patients were evaluated by an independent examinator both clinically and radiographically. The range of motion testing, the complete set of clinical tests for the anterior cruciate ligament function and complete radiographic evaluation for assessment of arthritis were all performed. The Subjective and Objective IKDC Forms were filled-in. Results: The mean follow-up period was 182 months (more than 15 years). None of the examined patients had a range of motion deficit superior to 5. 95% of the patients had a negative Lachman Test, with only 5% having a positive testing. The Pivot Shift Test was negative for 68% of the patients, while 25% had a grade I, 5% a grade II and 2% a grade III pivot shift testing. For the 40 patients available for the differential laxity measurement (after excluding 17 patients who had a bilateral ACL lesion) there was a mean preoperative difference of 5.5 mm and 10.1 mm versus 1.8 mm and 3.6 mm postoperatively at the latest follow-up as evaluated respectively by Rolimeter testing at maximal manual force and by TELOS testing. Radiographically, 49 patients (86%) had a normal examination, 5 (8.8%) presented with mild arthritis defined by articular remodeling and 3 (5.2%) had moderate or severe arthritis with detection of joint space reduction. As for the IKDC Objective Score, 25 patients (43.9%) were classified as grade A, 23 (40.4%) as grade B, 6 (10.5%) as grade C and 3 (5.2%) as grade D. The mean IKDC Subjective Score was 85.8 out of 100. Conclusions: The most important finding of this study was that arthroscopic primary anatomic anterior cruciate ligament reconstruction using patellar tendon autograft provides good results in the long term, both in terms of knee laxity and prevention of arthritis. The incidence of arthritis in our study was less than in other published studies and this was possible due to the reduced interval between ACL lesion and the moment of the surgical procedure, and consequent reduced incidence of associated meniscal lesions. We believe that with use of accurate surgical and rehabilitation techniques, this represents a reliable surgical procedure in the short, as well as in the long term.

P18-821 Anterior cruciate ligament femoral tunnel drilling anteromedial portal: 3-dimensional plane drill angle affects tunnel length relative to notchplasty S.C. Hwang1, F. Fu2 1 Gyeongsang National University Hospital, Orthopaedic Surgery, Associate Professor, Jinju, Republic of Korea, 2University

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 of Pittsburgh Medical Center, Orthopaedic Surgery, Pittsburgh, United States Objectives: The purpose of this study was to correlate femoral tunnel length with 3-dimensional drilling angle through the AM portal with and without notchplasty. Methods: Taking the computer data only from a previous anatomic study was performed using 16 cadaveric knees. First, the ACL femoral insertion was dissected and outlined for gross anatomical observation and the knees were scanned by computed tomography. Three-dimensional (3D) measurements were calculated using software (Geomagic, Inc.) and included center of ACL footprint and the area of the femoral ACL insertion. The femoral tunnel aperture centers were measured in anatomic posterior-to-anterior and proximal-to-distal directions and with the quadrant method (relative to the femoral notch). Software (Solidwork, Inc.) is used for create an ACL tunnel between an anatomical center of ACL and 3-Dimensional plane. Then, It made the diameter 8 mm cylinder. The cylinder was rested upon the anatomical center of ACL footprint and placed in three different position: coronal plane, sagittal plane, and axial plane. Finally, with a femoral tunnel through the center of ACL femoral insertion, the effect of notchplasty on the femoral tunnel length were measured. Results: With the quadrant method, the mean distances of the anatomical femoral tunnel center locations parallel to the Blumensaat line were 31.98 ± 2.7% (range, 21.7–34.23%) and 27.45 ± 2.1% (range, 22.75–30.2%). Length of the femoral tunnel versus 3-dimensional drilling angles with 3 mm notchplasty were as follows. 45/45/45 and 35 mm. 20/70/45 and 31 mm. 30/60/45 and 33 mm. 60/30/45 and 42 mm. The differences in tunnel length between 3-dimensional entrance drilling angle and notchplasty were statistically significant. Conclusions: Our finding suggest that 3-dimensional drilling femoral tunnel through the accessory AM portal at an low axial angle and high coronal angle without notchplasty produce longer femoral tunnel. Also, this would indicate that as little amount of bone as possible should be removed from the posterior portion of the intercondylar notch in anatomical ACL reconstruction.

P18-859 A new design femoral fixation device for anterior cruciate ligament reconstruction: mechanical evaluation and comparison L.-Y. Chen1, C.-H. Chen2 1 Keelung Chang Gung Memorial Hospital, Orthopaedic Department, Keelung City, Taiwan, Republic of China, 2Keelung Chang Gung Memorial Hospital, Keelung, Taiwan, Republic of China Objectives: A lot of suitable fixation devices for ligament reconstruction have been commonly and widely used nowadays. Although each type of fixation device shows its own advantages and good clinical outcome, it still has unexpected problems or unavoidable defects. The objective of this study is to develop a new femoral fixation device for ACL reconstruction which is able to improve and solve the problems and defects and evaluate its mechanical properties by following the ASTM standards for safety concern and compare its initial fixation properties with biodegradable interference screw by in vitro animal study. Methods: A new femoral fixation device was developed with five concerns—convenient technique, anatomic site fixation, small incision, no graft tear nor twist, and circumferential healing. Totally 12 sample products were made for further experiment. The axial pullout test was conducted to evaluate the mechanical properties of the new device and the interference screw by following the ASTM F543 standards with the use of totally ten Solid Rigid Polyurethane Foams. Afterward these two fixation devices were tested with the use of totally fourteen porcine femora to compare the initial conditions.

S147 Results: Comparing the maximum pullout force with pcf 20 polyurethane foams, the interference screw group (722.05 ± 130.49 N) was significant higher (P \ .01) than the Endoscrew group (440.79 ± 26.54 N). With pcf 10 polyurethane foams, the maximum pullout force for the interference screw group (242.61 ± 37.36 N) was found to be significant higher (P \ .001) than the Endoscrew group (99.33 ± 30.01 N). In animal study, the Endoscrew (646.39 ± 72.38 N) produced a significantlyhigher load (P \ .05) when compared with the interference screw (489.72 ± 138.64 N) in ultimate failure loads. In regard to pullout stiffness, there was no statistically significant difference (P \ .13) between the Endoscrew group (99.15 ± 12.16 N/mm) and the interference screw group (87.96 ± 11.12 N/mm). The cyclic stiffness was also not significantly different (P \ .44) between the Endoscrew group (93.09 ± 16.07 N/ mm) and the interference screw group (85.78 ± 14.76 N/mm). Conclusions: The Endoscrew is able to avoid some problems by basic concepts, general mechanisms, and literature comparisons. The axial pullout test showed that the strength of the Endoscrew was close to the strength required for activities of daily living as previous study mentioned, but it is inappropriate to use it on osteoporosis patients. By animal study, we concluded that the Endoscrew has a superior initial mechanical condition when comparing with the biodegradable interference screw.

P18-913 Anterior cruciate ligament (ACL) reconstruction using a quadriceps tendon autograft and press-fit fixation has equivalent results compared to a standard technique using semitendinosus graft: a prospective matched-pair analysis after 1 year J. Hoeher1, M. Balke2, M. Albers1, P. Helm2, B. Bouillon2, R. Akoto2 1 Arthro Sports Clinic, Cologne Merheim Medical Center, Cologne, Germany, 2University of Witten/Herdecke, Department of Trauma and Orthopedic Surgery, Cologne Merheim Medical Center, Cologne, Germany Objectives: To evaluate clinical outcome after primary ACL reconstruction using a quadriceps tendon graft QTG compared to a standard technique using a quadruple semitendionosus graft (STG) in a prospective matched pair analysis over 1 year. Methods: Fifty patients underwent primary ACL reconstruction using a quadriceps tendon autograft (QTG) and press-fit fixation in a anteromedial portal technique by a single surgeon and were followed prospectively. In all cases a 20 mm cylindrical bone block (9.4 mm diameter) was retrieved from the patella with an oscillating saw and was wedged into the femoral bone tunnel (9.0 mm diameter) to achieve press fit fixation. On the tibial side a bone cylinder harvested from the tibial bone tunnel was wedged alongside the soft tissue graft and additional fixation was achieved by tying the sutures over a bone bridge. All patients were treated with an identical rehabilitation program. Patients with revision ACL surgery, reconstructive cartilage repair and concomitant PCL injuries were excluded. As a control group 50 matched-pairs patients who underwent ACL reconstruction using a standardized technique with quadruple semitendinosus graft (STG) over an anteromedial portal and interference screw fixation were selected from a pool of 230 patients being treated over the same time period by the same surgeon (matched-pairs criteria: age, sex, meniscus and cartilage lesions). All patients were reevaluated after 1 year using subjective and objective IKDC scores and instrumented laxity measurements. A unpaired t test was used to detect differences between study groups (p \ 0.05). Results: Yet, 26 pairs of patients were evaluated at 1 year. The mean age at operation was 31.5 ± 10.0 (17–49) years for QTG and 28.9 ± 9.7 (17–49) for STG (p [ 0.05). Secure graft fixation using the press fit technique could be achieved in all cases. The mean surgical time was 111.3 ± 8.8 min for QTG and 85.7 ± 12.7 min for HTG. Perioperatively, no major complications were observed in both groups.

123

S148 Conclusions: ACL reconstruction using QTG with press fit fixation provides equivalent clinical results compared to standard STG 1 year after surgery. Advantages of using a quadriceps tendon graft may include, less bone loss in the femur, the lack of need for implants and saving the hamstring function.

P18-914 Anatomical ACL reconstruction: single bundle quadriceps versus double bundle hamstring M.C. Lee1, J.K. Lee1, S. Lee1, J. Jang1, S.H. Chun1, S.C. Seong1 1 Seoul National University College of Medicine, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: The objective of this study was to compare the clinical and functional outcomes of single bundle ACL reconstruction using quadriceps autograft and double bundle ACL reconstruction using hamstring autograft. Methods: In this retrospective matched paired study, we compared 32 patients who underwent unilateral ACL reconstruction using double bundle hamstring autograft, with 32 patients who underwent anatomical single bundle ACL reconstruction using quadriceps autograft. All patients were followed up for more than 2 years (range, 24–49 months). Clinical evaluations included manual laxity test, KT1000 arthrometry, IKDC subjective evaluation, Tegner, Lysholm scoring, questionnaire regarding the anterior knee pain and Cybex II isokinetic testing. Comparisons of tunnel locations on the femur and tibia were evaluated by the quadrant method on the 3D reconstructed CT images. Results: Anterior drawer, Lachman and pivot shift tests were grade 0 or 1 in over 95% of patients in both groups after surgery. The maximum side-to-side differences in KT-1000 arthrometry improved from 3.2 to 2.1 mm in quadriceps group and 3.5–2.4 mm in hamstring group. Lysholm score improved from 68.9 to 92.7 and 70.6–90.8, respectively. IKDC subjective evaluation improved from 58.1 to 82.5 in quadriceps group and 57.1–81.3 in hamstring group. Tegner activity score was 4.9 and 4.8 at the final follow up, respectively. There were no significant differences between the groups (all p [ 0.05). Questionnaires revealed no differences between the groups on postoperative anterior knee pain during strenuous work, climbing stairs, long period sitting, kneeling and normal daily activities (all p [ 0.05). Concerning the recovered muscle strengths measured by Cybex II isokinetic testing at the final follow-up, flexor strengths were significantly better in quadriceps group than in hamstring group at 180/s (p = 0.042). Regarding the location of tunnels on the femur and tibia, there were no significant differences (all p [ 0.05) except the high-low location on the femur. The high-low location of the femoral tunnels were higher in quadriceps group (p = 0.004), but the actual mean difference was only 1.5 mm. Conclusions: In anatomical single bundle ACL reconstruction with the quadriceps autograft, we could achieve similar knee stability, functional results and postoperative anterior knee pain compared to the double bundle ACL reconstruction with hamstring autograft. Moreover, using the quadriceps autograft produced better results in flexor muscle strength recovery than using the hamstring autograft.

P18-953 Anterior cruciate ligament reconstruction: soft tissue versus BTB V. Predescu1, V. Georgeanu2, C. Prescura2, S. Cristea2 1 St Pantelimon Clinical Emergency Hospital, Orthopaedic Surgery, Bucharest, Romania, 2St Pantelimon Clinical Emergency Hospital, Bucharest, Romania Objectives: This study aims to compare two arthroscopic surgical techniques in reconstruction ACL based on fixation and grafting different: the intra-articular autograft bone-patellar tendon-bone

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 (BTB) ligamentoplasty and quadruple hamstring autograft (semitendinos-gracilis) ligamentoplasty. Methods: A number of 149 patients underwent ACL reconstruction between 2006 and 2010, with a minimum follow-up of 1 year. The inclusion criteria were presence of single leg chronic ACL insufficiency, age between 15 and 45 years, no previous ligament surgery, no insufficiency of the posterior cruciate ligament (PCL), absence of osteoarthritis and no sign of posterolateral corner injury. 71 patients received a hamstring autograft and 78 patients received a bonepatellar ligament-bone autograft. All patients were enrolled in an exercise program to facilitate motion of the knee immediately after the operation, and all patients returned for postoperative evaluation. All patients were subjected to a clinical evaluation, with assessment of anterior drawer, Lachman’s and the pivot-shift tests. They also completed the International Knee Documentation Committee, Knee Society Score and functional knee score for anterior knee pain. Lachman testing was done using the KT-1000 arthrometer. Results: In the postoperative course the Lachman test was negative in 114 patients, 1+ in 26 patients, and 2+ in 9 patients. All patients with ACL reconstruction using the anteromedial portal had Lachman testing negative after 1 year. Pivot shift test was moderately positive in 25 cases postoperatively (all these 25 cases were done with the trans tibial technique). In term of knee stability there was no difference between BTB group and hamstrings group and also in return to sport activities. In post operative period there was an increase in pain score in BTB group, and we have recorded also kneeling discomfort and an increased area of decreased skin sensitivity. There were also 2 cases with decreased ROM due to patellar tendon shortening. Conclusions: There was no difference in clinical results between BTB group and hamstrings group. The most important advantage of the hamstring autograft is that avoids the disturbance of the extensor mechanism of the knee. In general, patellar tendon autografts are preferable for high-performance athletes, and hamstring autografts and allografts have some relative advantages for lower-demand individuals. In our opinion there is no indications for synthetic ligaments. Functional performance is compromised in patients who undergo a patellar tendon graft compared with a semitendinosus graft, possibly due to an altered activation of the quadriceps and hamstring muscle.

P18-974 Anatomical double bundle ACL reconstruction: transtibial versus anteromedial targeting of the femoral tunnels P. Zeman1, P. Nepras1, K. Koudela Jr.1 1 Charles University Prague, Clinic of Orthopaedics and Traumatology, University Hospital, Pilsen, Czech Republic Objectives: The aim of our study was comparison of two possible options of targeting the anatomical position of the femoral tunnels during anatomical double bundle anterior cruciate ligament reconstruction using transtibial or anteromedial portal technique. Methods: We used transtibial and anteromedial portal technique of anterior cruciate ligament anatomical double bundle reconstruction in 36 patients (29 males, 7 females). The surgeries were performed since October 2009 till December 2010 in our clinic. According to the position of the tip of the guide wire, we have divided our patients in 3 subgroups. The first was where the tip of the wire was in the middle of the centre of the anatomical insertion site of anteromedial or posterolateral bundles, the second out of the centre but still inside the original anatomical insertion and the third out of insertion site. Results: Using the transtibial anteromedial tunnels we have targeted centre of the femoral anteromedial insertion site with the tip of the guide wire just in 1 patient (2.8%), out of the centre but inside of the anteromedial insertion site in 4 cases (11.1%) and out of the insertion site in remaining 31 cases (86.1%). Targeting of the posterolateral femoral tunnels using anteromedial tibial tunnels was in all of cases (36 patients) out of the native posterolateral insertion site.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Using the posterolateral transtibial targeting of anteromedial femoral tunnels the tip of the guide wire was in the centre of anteromedial insertion in 11 cases (30.5%), out of the centre but still inside of the anteromedial insertion sites in 16 cases (44.5%) and out of the anteromedial original insertion site in 9 cases (25%). Targeting of the posterolateral femoral insertion sites throughout posterolateral tibial tunnel the tip of the guide wire was out of the insertion site in all cases. Using the technique, during which the guide wire was inserted through accessory anteromedial portal, we were able to achieve the centres of anteromedial and posteromedial original anatomical insertion sites in all 36 cases (100%). Conclusions: During the anatomical double bundle anterior cruciate ligament reconstruction we achieved native anatomical position of all posterolateral tunnels and in most cases also anteromedial tunnels using accessory anteromedial portal. Transtibial targeting was unsatisfactory.

P18-979 Biomechanical comparison of ACL reconstruction femoral anchors using a novel test method D.W. Hohman Jr.1, M. Ehrensburger1, L. Bisson2 1 State University of New York at Buffalo, Orthopaedic Surgery, Buffalo, United States, 2University of Buffalo, Orthopaedics and Sports Medicine, Buffalo, United States Objectives: Describe a novel test method for evaluation of ACL fixation, and use this method to compare the ‘‘time-zero’’ biomechanics of two aperture fixation femoral anchors. Methods: 20 fresh-frozen porcine femurs and bovine extensor tendons were used in this study (10 per device). Tendon grafts were sized (7.5 mm) and devices (9 mm) (AperFixTM, Cayenne Medical/AppianFxTM, KFx Medical) deployed. Each femur was cut, inverted so the bone tunnel was oriented vertically, and placed in an x–y grip within a load frame (MTS 858 Mini Bionix). Tendons were secured in a thermal electric cooled clamp (Bose). Exposed tendon gauge length was 39 mm. Infrared position sensors (Optotrak 3020) were attached to a custom T-bar secured to the anchors, and all setup components. This novel setup allowed for measurement of the MTS actuator and femoral anchor motion with high position resolution (0.01 mm). Mechanical testing consisted of three phases. (1) Construct was conditioned for 20 cycles (5–50 N at 0.25 Hz) (2) cyclic loading for 1,500 cycles (50–200 N at 1 Hz) (3) load to failure at 150 mm/min. All loads were in line with the longitudinal axis of the bone tunnel. Outcomes for cyclic loading included tendon elongation and anchor displacement. Reported outcomes for load to failure included construct stiffness, maximum load, yield load, and load at 5 mm of displacement. 5 mm of anchor was considered clinical failure. Independent measures t tests compared the performance of the devices (P \ 0.05). Results: Figure 1 shows construct subsidence during cycling. Figure 2 displays calculated parameters during the load to failure testing.

Fig. 1 Representative data illustrating the trend in the displacement over time data collected during the cyclic loading for both the AppianFx and Aperfix

S149

Fig. 2 Plot of load and stiffness data calculated in the load to failure test phase. The bar plots display mean values ±1SD. *Significant difference between Aperfix and AppianFx

Fig. 3 Figure 3 displays the tabulated tendon elongation and anchor displacement during the cyclic phase testing indicating no significant differences when comparing the anchors. Conclusions: Mechanical performance of the anchors were comparable during cyclic loading. Construct subsidence during cyclic loading is a combination of creep and motion of the anchor as it is ‘‘seating’’ into the bone. The majority of subsidence occurred early in cycling. Similar maximum loads are reached for both devices during load to failure.

P18-992 A comparison of patellatendon versus quadricepstendon in anterior cruciate ligament reconstruction: a prospective and randomized study B. Lund1, S.E. Christiansen1, T.G. Nielsen1, P. Faunoe1, M. Lind1 1 University Hospital of Aarhus, Department of Sportstraumatology, Aarhus C, Denmark Objectives: Anterior cruciate ligament (ACL) with bone-patella-bone (BPB) had previously been shown to be associated with post-operative anterior knee pain. It is unknown whether ACL reconstruction with quadriceps-tendon-bone (QTB) has similar complications and if clinical results are otherwise comparable to ACL reconstruction with BTB grafts. The purpose of this study was to compare clinical results of ACL reconstructions using BTB graft and QTB graft in a prospective randomized study. We hypothesized more anterior knee pain for patients reconstructed with BTB grafts compared to QTB grafts. Whereas subjective outcome scores and knee laxity was hypothesized to be similar between groups. Methods: From 2005 to 2009, a total of 60 patients were enrolled in this study. Inclusion criteria were isolated ACL injuries in adults. 51 patients were available for follow-up. Of the patients available for follow-up 25 patients were randomized to BPTB grafts and 26 to QTB grafts. Follow-up examinations were performed by an independent examiner at 1 and 2 years postoperatively, and anteroposterior knee laxity measured with a KT-1000 arthrometer and

123

S150 objective IKDC score. Patient evaluated outcome was performed by Knee Osteoarthritis Outcome Score (KOOS), and subjective IKDC score. Anterior knee pain was assessed with Knee walking ability test in which 2 m of knee walk was graded by the patients as problemfree, uncomfortable, difficult or impossible. Results: The baseline characteristics of the patients, did not differ between the 2 groups. The knee walking ability test demonstrated significantly less anterior knee pain in the QTB group, with only 7% of patients grading knee walking difficult or impossible compared to 34% in the BPTB group. At 1-year follow-up there was no difference the two groups in subjective IKDC and KOOS scores, but after 2 years there was a tendency to better subjective scores in the QTB-group. Anterior knee laxity was equal between the two groups. Conclusions: The use of the Quadriceps Tendon Bone graft results in less anterior pain than BTB grafts but has otherwise similar subjective and objective knee outcome scores. The QTB graft could be a better graft alternative for ACL reconstruction than BTB grafts in patients not eligible for hamstring graft and for patients with kneeling activities.

P18-1018 ACL reconstruction: clinicals results using hamstring versus allograft E. Furlan1, S. Baldan1, A. Rovini1, R. Nardacchione1 1 Policlinico Abano Terme, Knee’s Surgery, Abano Terme, Italy Objectives: This is a retrospective clinical study designed to assess the safety and functional short-term clinical outcome using allografts in isolated ACL reconstruction compared with hamstrings. Methods: From January 2008 to January 2010 we have performed 280 ACL reconstructions using allografts tendons and 670 using Hamstrigs tendon. The surgical technique has always been the same. The femoral fixation was performed with Rigid Fix (Mitek) or TTS system (Ireva), the tibial fixation was always performed with interference screw Biorci HA (Smith & Nephew). Forty primary reconstruction performed with allograft (31 males, 9 females, mean age 33.58 years), for a mean of 37.64 months, and 40 with hamstring (35 males, 5 females, mean age 32.90 years) for a mean of 28.76 months were evaluated in this study by using the Lysholm Knee Scoring Scale, the Tegner score and scale KOOS. Statistical analysis is performed using T Student test. Results: The mean KOOS score is 89.11 (SD 18.23) for allograft e 97.94 (SD 2.22) for hamstring (p = 0.0066). The mean Tegner score is 6.97 (SD 1.72) for allograft e 5.90 (SD 1.70) for hamstring (p = 0.0058). The mean Lysholm knee scoring scale is 92.51 (SD 13.41) for allograft and 98.05 (SD 2.99) for hamstring (p = 0.0063): excellent in 33, good in 1, fair in 4 and poor in 2 cases using the allograft; using hamstring the score is excellent in all the cases. Conclusions: In conclusion we can say that, in our experience, the use of hamstring tendon is to prefer for reconstruction in isolated ACL tear. Longer follow up, especially in high level athlete, is necessary in the future.

P18-1055 A prospective study of 100 double bundle ACL reconstructions: our technique with hybrid fixation and its results A. Goyal1, D. Chaudhary1, D. Joshi1, V. Bahl1 1 Sports Injury Centre, Delhi, India Objectives: Single Bundle ACL (SB ACL) reconstruction has been the gold standard for decades. However there remained a subset of patients having instability and dissatisfaction as compared to their pre injury function. This led to the evolution of double bundle ACL reconstruction which restores the anatomy of ACL. There have been some inherent problems in this technique though as creation of two

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 more tunnels limits the various available options of rigid fixation devices. We hereby present our technique of ‘‘hybrid fixation’’, with our results with it. This allows for a rigid fixation and a near aperture fixation and limits the risk of confluence of femoral tunnels. Methods: 100 patients of isolated ACL injury were treated prospectively using ipsilateral hamstring autografts. Anteromedial (AM) bundle comprised of tripled semitendinosus graft and was fixed using transfix (Arthrex, Naples, FL, USA) on the femoral side and Biointerference (bioscrews) screws on the tibial side. Posterolateral (PL) bundle comprises of tripled gracilis graft fixed on femoral and tibial side with bioscrews. This is thus a ‘‘Hybrid Fixation’’ of soft tissue graft during DB ACL reconstruction. The patients were then evaluated for functional and objective results using KT-1000 arthrometer, Lysholm score, IKDC scores and Isokinetic muscle strength testing postoperatively for a period of at least 2 years. Results: The KT-1000 results were evaluated using paired t test with the p value set at 0.001. At the end of 2 years, the total anteroposterior translation (KT-1000 manual maximum) showed improvement of mean of 6.4 mm (range 3–9 mm) (p value \0.001) postoperatively compared to the normal side. The maximal manual displacement improved preoperatively from mean value of 7.4 mm (range, 3–14 mm) to a postoperative range of mean 2.1 mm (range, 1–7 mm) which was statistically significant. The Lysholm score too showed statistically significant improvement from 52.4 (12–76) pre-operatively to a post-operative score of 89.1 ± 3.2 (range, 47–100). According to the IKDC score maximal patients had normal (group I-54%) or nearly normal (group II-36%) results. We had no incidence of nerve or vascular injuries or any confluence of tunnels intraoperatively with our technique. The isokinetic strength measurements showed that quadriceps strength lagged behind the normal side (34.12% deficit on concentric contraction) at 3 months which improved to 16.8% at the end of 6 months. There was no statistically significant difference in hamstring muscle strength at 3 and 6 months. Conclusions: In our study it was noted that Double bundle ACL reconstruction had good functional outcomes using hybrid fixation method using cross-pin and bioscrews as described. It also provides a way out to the surgeon in case of eventuality of a confluence of tunnels or a posterior femoral blowout.

P18-1084 Anatomic anterior cruciate ligament reconstruction: 3D CT analysis and clinical results of mid bundle tunnel positions determined by the ‘ruler technique’ A. Getgood1, S. Spencer1, J. Bird1, M. Carmont2, P. Thompson1, T. Spalding1 1 University Hospitals Coventry and Warwickshire, Coventry, United Kingdom, 2Princess Royal Hospital, Orthopaedic Dept, Telford, United Kingdom Objectives: Anatomic reconstruction of the Anterior Cruciate Ligament is now considered the optimal technique to restore functional stability to the ACL injured knee. Single bundle mid footprint techniques have been described, which may be more preferable to more complicated double bundle techniques. As yet there have been few series reporting clinical results. The purpose of this study is to report the clinical results of a consecutive series of patients undergoing anatomic single bundle, mid footprint ACL reconstruction, determined by the ‘ruler technique’ and assessed post operatively with 3D CT and clinical evaluation. Methods: This study represents a case series of anatomic ACL reconstructions performed at our institution by two surgeons using the same technique. ACL femoral tunnel position was determined using a new measuring ‘ruler technique’ based on measuring 50% along a line from the proximal border of the articular cartilage and the distal articular margin, using a specific ruler parallel to the femoral axis. Patients were reviewed at 3 and 6 months, then annually. Outcome measures included

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 KT-1000 arthrometry, IKDC, Lysholm, Tegner and KOOS scores. Accuracy of this technique was validated by measuring post-operative 3D CT scans to determine tunnel position in relation to previously published radiographic quadrant descriptions. Results: Between July 2009 and June 2011 150 anatomic ACL reconstructions were performed, of which 130 met the inclusion/exclusion criteria. The mean age was 29 (range 16–52) with male:female ratio of 98:32. The mean position of the femoral tunnel on the Bernard-Hertel grid was 29%/33% with a target of 27/34 based on the mean of previously published data. This confirmed a satisfactory tunnel position demonstrating the mean centre of the femoral tunnel to be within 2 mm of the optimal position. Three patients encountered a complication perioperatively including one DVT, one deep infection requiring washout and antibiotics, and one non-fatal pulmonary embolism. At 1 year, 4 failures (4%) were reported (3 traumatic graft ruptures following too early a return to sport; 1 alcohol associated fall). KT-1000 arthrometry showed a mean side-to-side difference of 1.29 mm (range -2 to +5 mm). A statistically significant improvement in all outcome measures from baseline was observed at 1 year (p \ 0.05). Conclusions: The use of the ruler technique produced femoral tunnels very close to the anatomic centre of the femoral attachment, as determined by 3D CT compared to published radiographic criteria. Clinical results are very encouraging, however long term comparative studies are required to determine whether this technique will produce objective clinical gains in the future.

P18-1157 Selective posterolateral bundle augmentation in ACL tears: MRI and clinical evaluation of a series of 40 patients with a mean 24 months follow-up B. Sonnery-Cottet1, R. Zayni2, P. Archbold1, Y. Carillon1, J. Clechet1, P. Chambat1 1 Centre Orthope´dique Santy, Lyon, France, 2Centre Orthope´dique Santy, Orthopeadic surgery, Lyon, France Objectives: To evaluate the clinical outcome and graft survival following reconstruction of the PL bundle with preservation of the AM remnant. Methods: In this prospective study, 40 patients in which ACL remnant was found to bridge the AM femoral and tibial insertion at arthroscopic surgery underwent an isolated PL bundle reconstruction using a doubled or tripled semitendinosous graft and an outside in technique. Clinical evaluations including instrumented knee testing, IKDC evaluation, and knee function evaluation (Lysholm and Tegner score), were performed preoperatively and at a mean follow up of 2.02 years (min 1.2; max 3.33). Demographic and clinical data of the series are presented in Table 1. All patients underwent MR scanning at the last control using sagittal and coronal proton density fast spin echo (FSE) sequences, and T2 fatsuppressed FSE sequences. Graft continuity and signal intensity were evaluated by two independant radiologists which were blinded to the patients clinical datas. Tunnel enlargement was determined by digitally measuring the widths perpendicular to the long axis of the posterolateral tunnels on an oblique coronal and axial plane and compared with the Table 1 Demographic and clinical data Number of patients

40

Men/women

23/17

Age at surgery (years)

30 (min 14; max 56)

Non contact sports

20 (50%)

Contact sports

20 (50%)

Time interval from injury to surgery (months)

5.7 (min 0.8; max 40)

Follow-up time (years)

2.02 (min 1.2; max 3.3)

S151 Table 2 Results of the series Pre-injury Subjective IKDC (points)

Before surgery

Final follow-up

p

43.5 (min 14.9; max 77)

89.9 (min 72.5; max 100)

\0.01

Objective IKDC (patients) A

0

25

B

23

15

C

15

0

D

2

0

Lysholm score

60.8 (min 17; 94.2 (min 70; \0.001 max 89) max 100)

Tegner scale Rolimeter test (mm)

6.8 (min 3; 2.3 (min 1; max 10) max 4)

6.4 (min 3; max 10)

5 (min 4; max 1.6 (min 2 10) max 4)

\0.001 \0.01

intraoperative drill diameter. To quantitatively determine normalized signal intensity of the graft the signal/noise quotient (SNQ) was calculated. Results: The clinical results of the series are presented in Table 2. On MRI, all grafts but one were continuous. The mean tibial and femoral diameters were respectively 7.9 mm (min 7; max 9) and 7.8 (min 7; max 9) at surgery and measured at 9 mm (min 7, max 13) and 8.5 mm (min 7; max 12) at last follow up. The mean SNQ was 4.75 (min 1.2; max 14.6). The preserved AM bundle was visualised in 88% of the patients. Statistical workup demonstrated no correlation between SNQ and instrumented differential laxity (p = 0.28), objective IKDC (p = 0.5), subjective IKDC (p = 0.9). Conclusions: The results of this 40 selective PL bundle reconstruction are encouraging and demonstrate a low side to side laxity. At 2 years follow-up, all graft but one were viable and no dramatic tunnel enlargement was observed with this more anysometric graft.

P18-1271 ACL RI-reconstruction with fresh-frozen tendon allograft: clinical outcomes at 6 years follow-up C. Corradini1, P.F. Bottiglia Amici Grossi1, A. Ventura2, V. Macchi3, E. Malagoli3, C. Verdoia3 1 Gaetano Pini Orthopaedic Institute, Sports Trauma Center, State University of Milan, Milan, Italy, 2Gaetano Pini Orthopaedic Institute, CAM, Minimally Invasive Surgery Unit, Milano, Italy, 3 Gaetano Pini Orthopaedic Institute, Orthopaedic and Traumatologic Clinic, State University of Milan, Milan, Italy Objectives: The improved clinical outcomes of ACL reconstruction have permitted in sportsmen population a successfully return to sport related activities but also the risk for new injuries. The increasing number of neo ACL ruptures have risen the need for revision surgery. In the recent past the failing grafts, and in particular those irradiated tendon from cadaver, have produced a limited utilization of allografts. The renewed soft tissue bank with fresh-frozen allograftsfrom donors may have opened new operative strategies. But little data about clinical outcomes of revision ACL surgery with fresh-frozen allografts have been published. Methods: Between 2005 and 2011 forty revision ACL reconstructions were performed. All were performed using double-looped peroneal or posterior tibial or semitendinosus or gracilis tendon allografts fixed

123

S152 with bio-absorbable pins and screws. The details of the technique varied according to the original graft choice and the abnormality encountered. A second group of patients undergone in the same period by the same equip to primary ACL reconstruction with unharvested ipsilateral double-looped semitendinosus and gracilis autograft were also considered. The clinical outcomes were assessed by both objective (dynamometric and stabilometric) and subjective measures entered into a computerized database. Clinical and stability results between primary and revision ACL reconstruction were compared. Results: Two groups between 26 and 49 year-old resulted homogeneous for number, age, sport level and length of follow up. The stabilometric results of revision surgery were inferior to primary reconstruction, but as regards stability, the difference between primary and revision cases was not significant. Any difference within primary reconstruction with BPTB or ST + G autograft reconstruction was revealed. There were significant improvements in the scores of clinical outcomes after revision surgery but they remain inferior in confront to primary surgery. There were no difference in clinical and stability results in different tendon allografts. Till 6 years follow-up no infection was observed and no progressive laxity was detected. No decreased sport activity level was related to clinical outcomes. Conclusions: Revision ACL reconstruction with fresh-frozen tendon from donor doesn’t seem inferior to those with another autograft and of primary. ACL ri-reconstruction with allograft is more than an opportunity of choice in sportsmen.

P18-1318 Biomechanic testing of the porcine triple bundle anterior cruciate ligament M.P. Pinto1, T.R. Protta1, G. Tantisricharoenkul1, P. Smolinski2, F. Fu3 1 UPMC Center for Sports Medicine, Pittsburgh, United States, 2 University of Pittsburgh, Department of Mechanical Engineering, Pittsburgh, United States, 3University of Pittsburgh Medical Center, Orthopaedic Surgery, Pittsburgh, United States Objectives: The purpose of this study is to investigate the biomechanics of the antero-medial (AM), intermediate (IM) and posterolateral (PL) bundles in porcine anterior cruciate ligament (ACL). Methods: Fourteen fresh frozen, mature porcine knees were tested using a robotic/universal force moment sensor (UFS) system. This system applied an anterior loading of 89 N at 30, 60 and 90 of knee flexion to measure anterior tibial translation (ATT) and a 4 Nm

Fig. 1 Anterior tibial translation (ATT; mm) during the simulated KT-1000 test group 1 (AM sectioned, IM sectioned and PL sectioned)

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 2 Anterior tibial translation (ATT; mm) during the simulated KT-1000 test group 2 (AM sectioned, PL sectioned and IM sectioned)

Fig. 3 Anterior tibial translation (ATT; mm) during the simulated KT-1000 test group 3 (IM sectioned, PL sectioned and AM sectioned)

internal rotation (IR) and external rotation (ER) tibial torque at 30 and 60 of knee flexion before and after each bundle was selectively cut. The testing protocol was divided into 3 different groups according to the sequence of bundles sectioned: group 1—AM sectioned, IM sectioned and PL sectioned; group 2—AM sectioned, PL sectioned and IM sectioned; group 3—IM sectioned, PL sectioned and AM sectioned. Results: For this study, the Sample Size was calculated to be, twentyfour specimens (Figs. 1, 2, 3). Conclusions: The AM bundle stabilized the porcine knee against anterior tibial translation (ATT) and rotatory loads. At high knee flexion angles, the IM bundle supports the AM bundle in preventing ATT. Also, at low knee flexion angles, the IM bundle supports the PL bundle in preventing ATT. The role of the IM bundle supports both the AM and PL bundles through all flexion angles, especially at 30 (full extension in the porcine knee). The PL bundle stabilized the porcine knee against rotatory load especially near full extension. The IM bundle has an important role in porcine knee stability.

P18-1356 A pilot study on the ability of an unloading brace to reduce the risk of ACL injury M. Nicholls1, K. Briem1 1 University of Iceland, Reykjavik, Iceland Objectives: Hewett et al. [1], showed that peak knee abduction moments and angles (peak and at initical contact) during a drop

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Table 1 Mean knee abduction angle

Initial contact Peak

No brace

Brace

Difference

p

-7.44

-8.26

0.82

0.18

6.45

3.94

2.51

0.02

Table 2 Mean knee internal abduction moment No brace

Brace

Difference

p

25.08

31.95

-6.87

0.09

Initial contact Peak

vertical jump (DVJ) are predictive of ACL injury. Unloading braces, typically used for unicompartmental knee OA, are designed to unload the affected compartment by abducting or adducting the knee. The objective of this study was to see if an unloading knee brace designed to adduct the knee could alter knee abduction angles and moments in a way that might reduce risk of ACL injury. Methods: This controlled laboratory trial included 8 healthy males between the ages of 18 and 40. Lower limb kinetics and kinematics were obtained for each subject over a series of DVJs during a single session. Subjects stood on top of a 31 cm high box and were instructed to drop directly down off the box, land on both feet and immediately jump as high as they can [1]. Subjects performed three successful DVJs for the no brace and braced condition. Data were captured using two force plates sampling at 1,200 Hz and eight digital cameras at 125 Hz. Eighteen retroreflective markers were placed on specific anatomical locations on the pelvis and the lower limbs to define the joint centres and segments, and marker clusters were used to track each segment. Raw force data and marker data were filtered through a low pass Butterworth digital filter at 20 Hz. Angles and moments at each joint were calculated using Visual3D. Paired t tests were performed to assess outcome measures between conditions. Results: The mean peak abduction angle was significantly less for the brace than for no brace condition (p \ 0.05; Table 1), whereas no difference was found at initial contact (IC). Although not statistically significant, the mean peak abduction moment was slightly greater for the braced condition (p = 0.09; Table 2). Conclusions: Despite low statistical power, a significant reduction in mean peak abduction angle was shown while wearing an unloader type brace. While this maybe a positive change, the average reduction of 2.5 reflects our study population and is considerably less than the 7.6 difference that Hewett el al. found between females who subsequently injured their ACL and subjects who didn’t. Although the increase in peak internal abduction moment was non significant a greater internal abduction moment is also indicative of a reduction in ACL injury risk. An unloader brace may offer a method of reducing ACL injury risk. However further study is required with an at risk population. Furthermore in order to fully understand any change in moments at the knee, load sharing between the brace and the knee needs to be understood. Reference 1. Hewett TE et al. The American Journal of Sports Medicine. 2005

P18-1392 Augmentation of anterior cruciate ligament after partial tears M. Iosifidis1, I. Melas2, T. Kyriakidis2, A. Kyriakidis1 1 Papageorgiou G.H., Orthopaedic Surgery, Thessaloniki, Greece, 2 Papageorgiou G.H., Thessaloniki, Greece

S153 Objectives: Conservative treatment of chronic partial ACL tears can be successful in non active patients. On the contrary, active patients require ACL reconstruction, which usually sacrifices the intact bundle of the ligament. The purpose of this study is to present the short-term results of the augmentation technique for partial ACL tears. Methods: During the period 2008–2011 we operated 32 patients with partial ACL tear. They were 17 men and 15 women with a mean age of 27.9 years (23–39). All of them had knee trauma history. Anterior drawer and Lachman tests were positive to all of them, while the pivot shift test was negative in 23 patients, trace negative in 4 and positive in 5 patients. They all had knee MRI, where the suspicion of an ACL tear was set. During arthroscopy we found lesion of the antero-medial bundle in 20 patients, and the rest 12 had a postero-lateral bundle tear. The patients underwent an ACL reconstruction using hamstring tendons autograft (semitendinosus and gracilis) and with simultaneous preservation of the not affected bundle. The graft was stabilized either with an endobutton or an absorbable interference screw in the femur and with an absorbable interference screw and post-fixation with staple in the tibia. Results: The mean follow-up time was 15.3 months. At the last follow-up, none of the patients experienced any pain or instability. The Lyscholm score improved from an average of 72.4 preoperatively to 91.6 postoperatively. According to the Tegner Activity Level the mean pre-injury level was 6.4 and became 6.2 at the time of last follow-up. All patients returned to their preinjury occupational and sports activity. Conclusions: Preserving the residual bundle of the ACL has been reported to improve the results of its reconstruction, due to increased mechanical strength and blood flow to the graft but also by maintaining some propioceptive innervation. Our experience with this technique, confirmed, in the short term, its promising results.

P18-1422 Arthoscopic-assisted fixation of anterior cruciate ligament avulsion (tibial eminence fractures) in adults K. Fehske1, C. Weißer1, P. Ziai2, R.H. Meffert1 1 Klinik fu¨r Unfallchirurgie, Universita¨tsklinikum Wu¨rzburg, Wu¨rzburg, Germany, 2Medizinische Universita¨t Wien, Orthopa¨die und orthopa¨dische Chirurgie, Wien, Austria Objectives: Tibial avulsion of the anterior cruciate ligament (ACL) are fractures of the tibial eminence and AO-classified as A 1 injuries of the tibial head. Those injuries are mostly seen in sekeletally immature patients. The aim in operative treatment should be the refixation of the acl to avoid an acl reconstruction and to restore knee stability. Methods: From April 2009 until November 2010 9 patients with acl avulsion have been treated in our department. Pre-operatively all patients underwent conventional X-ray and MRI to verify the diagnosis and to look for further pathologies. The acl refixation was done arthroscopically-assisted. With the tibial-aming device (acl-reconstruction) two tibial tunnels were drilled, one medial and one lateral of the eminence fracture. The acl avulsion was distally armed with a fiber-wire suture and adapted tibially. The suture was pulled out tibially through the tunnels and tied together distally. Follow-up examinations have been after 1, 3, 6 and 12 months including X-ray and subjective knee scores. Results: Our patients have been 27.8 ± 12.7 years old, at time of operation one patient was older than 30 years of age, one was under 20 years of age. 44.4% our patients have been female. Seven patients were injured within a sports-accident, two in a motor-cycle accident. Time-interval from trauma to surgery was 9 ± 2.9 days. In one

123

S154 patient we needed to change intraoperatively to open refixation. At the last examination after 11.2 ± 1.2 months 88% of our patients had no subjective instability, the Lachman test grade was 1+, the Rolimeter measurement showed a side to side difference considering the anterior drawer of +1.8 ± 0.2 mm. One patient had a 10 loss of flexion after 10 months. One patient needed to undergo acl-reconstruction 13 months postoperatively due to another trauma. The radiological follow-ups showed good fragment-adaptation with no secondary dislocation. Conclusions: This operative treatment of acl avulsion is a good method to keep the original ligamental structures with its proprioceptive function and to avoid an acl reconstruction. Mid-term and long-term results have to proof if the objectively and subjectively restored knee stability will prolong. The current literature lets us presume that clinical results after acl-refixation are superior to the results after acl-reconstruction.

P18-1446 Anterior cruciate ligament revision reconstruction with MacIntosh procedure using the fascia lata J. Arndt1, G. Ducrot1, A. Londero1, J.-H. Jaeger1 1 Centre de Chirurgie Orthope´dique et de la Main, Strasbourg, France Objectives: Failed anterior cruciate ligament reconstructions follow the increase of primary reconstructions. Preoperative planning of the revision surgery asks the question of graft selection and type of procedure to use. The purpose of this retrospective study was to assess the functional outcomes after ACL revision surgery with MacIntosh procedure using the fascia lata. Methods: The series included 73 patients (61 males and 12 females), performed for ACL revision surgery with MacIntosh procedure using the fascia lata, between 2001 and 2006. The mean age at the revision surgery was 30.2 years. Were excluded patients with posterior or postero-lateral laxity associated, controlateral knee laxity or instability, prior valgus tibial osteotomy or postero-lateral complex reconstruction. The primary reconstruction was performed with 68 autografts (38 patellar tendons, 19 hamstring tendons, 10 fascia lata, 1 patellar and quadriceps tendons), and 5 synthetic grafts. An extra-articular supplementation was made for 7 intra-articular reconstructions. Functional outcomes were assessed at an average follow-up of 53 months and minimum of 30 months. All patients were interviewed to determine the subjective evaluation with IKDC Score, Lysholm Knee Score and Tegner Activity Scale. Thirty-nine patients underwent an objective evaluation by clinical examination according to the IKDC protocol, and antero-posterior laxity measurement with a KT2000 arthrometer. Results: At follow-up, the mean IKDC subjective score was 79.5 ± 20% (range, 21–100%), with a mean current function at 7.1 out of 10. Mean Lysholm Knee Score was 82.8 ± 17%, distributed in 21 excellent results (29%) 27 good (37%), 15 fair (20%) and 10 poor (14%). The Tegner Activity Scale was 7.8 out of 10 before initial ACL injury, and 6 at follow-up. Mean differential laxity measured with KT-2000 (maximal manual) was 2.2 ± 2.2 mm. The rate of normal or nearly normal knees (IKDC A or B) at objective evaluation was 79% for the global evaluation (26% A, 54% B, 20% C, 0% D), including 95% for range of motion and 85% for laxity. The pivot-shift test was negative in 37 out of 39 knees (95%). At follow-up, 63% of patients underwent a meniscectomy. Conclusions: ACL revision surgery can provide good functional outcomes, though they are less satisfactory than after a primary reconstruction. The results we found when using the fascia lata with MacIntosh procedure are comparable to previous series reported in the literature. This procedure allows good function improvement and knee stability restitution, in antero-posterior direction with its intra-

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 articular part, but also rotatory with its automatic extra-articular supplementation. It can therefore provide a good alternative procedure for ACL revision surgery.

Knee-ACL II

P19-46 Double bundle anterior cruciate ligament reconstruction; not for all; an MRI study G.N. Tzoanos1, A. Ntailiani2, N. Manidakis1, N. Tsavalas2, P. Katonis1, A. Karantanas2 1 University Hospital of Heraklion, Crete, Orthopedic, Heraklion, Greece, 2University Hospital of Heraklion, Crete, Radiology, Heraklion, Greece Objectives: The purpose of this LRI study was to estimate the incidence of anatomic restrictions which may prevent performing a double-bundle reconstruction of a torn anterior cruciate ligament (ACL). When the femoral notch width (FNW) is less than 12 mm and the tibial insertion site (TIS) is smaller than 14 mm in diameter, double-bundle reconstruction cannot be performed due to technical difficulties. The hypothesis was that significant rate of patients do not meet the criteria for double bundle ACL reconstruction. Methods: We retrospectively studied 658 knee MRI examinations, which corresponded to 635 patients. All MRI examinations were performed on a 1.5T scanner using a synergy coil. Both the coronal and sagittal images were evaluated for the measurements. The femoral notch width (FNW) and the ACL tibial insertion site (TIS) length were digitally measured using a computer workstation. We measured FNW on coronal T1-w spin echo images at the level of decussation of ACL and posterior cruciate ligament and the ACL TIS length on sagittal fat suppressed proton density turbo-spin-echo images at the point of maximum length. Exclusion criteria included patients with open physes, severe osteoarthritic changes (Grade III or greater) and multi-ligamentous injuries. Results: We found 8.51% of patients to have open physes, 9.27% severe osteoarthritic changes and 0.759% multi-ligamentous injuries. The above were excluded from the analysis. In our study group, 14.741% of subjects proved to be non-proper candidates for doublebundle ACL reconstruction technique depending on the results of measurements. Of them, 13.677% had short TIS length, 1.367% had narrow femoral notch width (FNW) and 0.3% had both. Conclusions: There is a significant rate of patients who do not meet the criteria for double-bundle ACL reconstruction technique. MRI is able to depict this subgroup of patients and thus is important preoperative tool to decide if double-bundle ACL reconstruction can be performed.

P19-59 Double bundle anterior cruciate ligament reconstruction using bone-patellar tendon-bone and gracilis composite graft: minimum 2 years follow-up T. Tajima1, E. Chosa1, K. Yamamoto1, K. Kawahara1, N. Yamaguchi1 1 University of Miyazaki, The Division of Orthopedic Surgery, Miyazaki, Japan Objectives: In cases of anterior cruciate ligament (ACL) injuries, bone-patellar tendon-bone (BTB) graft and hamstring graft are frequency used. An altered rotational axis resulting in significantly greater translation of the lateral compartment in the single bundle

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 compared with double bundle ACL reconstruction was reported. The present study was modified surgical procedure that the double bundle ACL reconstruction using BTB and gracilis graft, and performed to evaluate the clinical and functional outcome of present new concept technique. From 2007, sixteen patients with a mean age of 26.3 ± 8.5 years underwent double bundle ACL reconstruction using BTB and gracilis composite graft with an arthroscopic technique (Thirteen males and three females. Six cases for primary ACL reconstruction, seven cases for revision surgery, and three cases for ACL reconstruction in multiple ligament injuries). The mean follow-up period was 33.7 ± 13.4 months after surgery. Methods: A reconstruction procedure was modified for the ACL using a double bundle that is the combination of BTB and gracilis tendon composite autograft. Two tibial and two femoral bone tunnels are used to reconstruct two bundles of ACL; an anteromedial bundle (AMB) and a posterolateral bundle (PLB). The femoral bone tunnels are created just posterior to the resident‘s ridge. The tibial bone tunnels are created at the center of AM and PL tibial attachment, respectively. BTB is fixed in the AM tunnels produced on the anatomical points of tibia and femur. The gracilis graft is fixed in an anatomical PL tunnel produced. The mean width of BTB is 7.0–7.5 mm, since 10 mm graft is sometimes not suitable for patients, who have the small or medium size of patella tendon width such as Asian people and females. For these patients, 10 mm graft is bigger than one-third of patella tendon width. Therefore, BTB graft was harvested at 7.0–7.5 mm for AMB, and gracilis graft was adjustment at 5.0–5.5 mm for PLB. A clinical and functional evaluation using Lysholm score was performed. An anterior tibial load of 134-N by Kneelax arthrometer measurement and Biodex isokinetic strength testing at 60 by experienced physician were made. Results: The mean score on the Lysholm score was improved from 57.8 ± 20.9 before surgery to 91.9 ± 11.6 after surgery. In response to a 134-N anterior tibial load, the use of BTB and gracilis composite graft could restore anterior tibial translation to within 1.0 ± 3.5 mm of the intact contralateral knee. The mean muscle strength after surgery compared to contralateral knee was 81.6 ± 11.9% in knee extension, and 93.5 ± 19.5% in knee flexion. Conclusions: Although this study is a small case series, the double bundle ACL reconstruction using BTB and gracilis composite graft provides satisfactory clinical and functional outcome at a minimum 2 years follow up.

P19-62 Chlorhexidine gluconate cleansing has no effect on the structural properties of human patellar tendon allografts A.D. Sobel1, L. Bisson1, D.W. Hohman1, J.L. Jones1 1 SUNY University at Buffalo, Department of Orthopaedics, Amherst, United States Objectives: Although instances of patellar tendon graft (PTG) contamination during anterior cruciate ligament (ACL) reconstruction are rare, surgeons must have a plan for this situation. Options include cleansing the graft, harvesting another PTG from the contralateral knee or an ipsilateral hamstring autograft, using an allograft, or stopping the procedure and completing it at another time. Except for cleansing the contaminated graft, all of the aforementioned options result in greatly increased morbidity or cost to the patient making graft cleansing a favorable choice. The purpose of this study was to evaluate the effects of an antimicrobial chlorhexidine soak on the structural properties of human patellar tendon anterior cruciate ligament (ACL) allografts.

S155 Methods: Sixteen human patellar tendon allografts obtained from Allosource (Centennial, CO) were randomly split into two groups of eight. Grafts in one group were soaked in 4% chlorhexidine gluconate for 30 min and the others were kept moist in saline-soaked gauze. All grafts were potted in custom made molds with Bondo (3 M, St. Paul, MN) putty and loaded into a Bionix Mini MTS machine (MTS Systems Corp, Eden Prairie, MN) for mechanical testing (Fig. 1). Mechanical testing involved the use of a combined cyclic axial loading and load-to-failure protocol modeled after previously published studies evaluating PTG structural properties. Data on preload width, preload thickness, cyclic elongation, ultimate tensile load (UTL), and stiffness were obtained through measurement and mechanical testing of the grafts. Statistical comparison of the mean values for each structural property of the testing groups was performed using SPSS version 18 (SPSS Inc., Chicago, IL, USA). Significant differences were considered if p values were \0.05. Results: Graft donor ages ranged from 29 to 43 years old. There was no difference in the mean values of graft dimensions of the chlorhexidine versus normal saline exposed groups before mechanical testing (width 9.48 vs. 9.56 mm, p = 0.89; thickness 4.01 vs. 4.57 mm, p = 0.34). Graft elongation was not statistically different between the groups (2.52 vs. 1.43 mm, p = 0.27). No statistically significant difference was noted between the ultimate tensile load (2,219 vs. 1,878 N, p = 0.36) or stiffness (274.3 vs. 297.0 N/mm, p = 0.63) of the grafts in both groups. Conclusions: Structural properties of human patellar tendon allografts are not significantly affected when sterilized with 4% chlorhexidine gluconate for 30 min. Surgeons wishing to utilize a contaminated graft intraoperatively may do so without concern that sterilization of the graft with chlorhexidine will impact graft strength.

P19-75 Clinical and radiological accuracy in the diagnosis of reattached ACL ruptures O. Al-Dadah1, A. Dhillon1, C.T.J. Servant1 1 Ipswich Hospital, Trauma and Orthopaedic Surgery, Ipswich, Suffolk, United Kingdom

Fig. 1

123

S156 Objectives: The clinical and radiological diagnosis of ACL tears can be difficult, particularly in chronic injuries. The objective of this study was to assess the clinical and MRI diagnostic accuracy of chronic complete ACL ruptures which are free ended and unattached compared to those that have abnormally reattached to neighbouring structures. Methods: A total of 182 consecutive patients who underwent ACL reconstruction were retrospectively analysed. All subjects had arthroscopic confirmation of the precise morphology of ACL tear pattern. Group 1 consisted of patients were the ACL was free ended and unattached. Group 2 patients displayed an ACL tear pattern were the ligament was found to be abnormally reattached to a neighbouring anatomical structure. The pre-operative MRI findings and the clinical findings on examination under anaesthesia (in particular that of the Lachman and Pivot shift tests) were recorded. Results: There were 78 patients (42.9%) assigned to Group 1 and 104 patients (57.1%) assigned to Group 2 were the PCL was noted to be the commonest site of abnormal reattachment. MRI scans correctly reported a complete ACL tear in 45 patients (86.5%) in Group 1 and 58 patients (74.4%) in Group 2 (p = 0.123). On Lachman testing, 77 patients (98.7%) in Group 1 had a side-to-side increase in laxity of more than 5 mm, compared with 93 patients (89.4%) in Group 2 (p = 0.014). Similarly, 72 patients (92.3%) in Group 1 had an absent endpoint, compared with 82 patients (76.9%) in Group 2 (p = 0.008). All subjects in both groups had a positive pivot shift test, with a trend to more patients demonstrating a convincing pivot ‘clunk’ in Group 1 (p = 0.050). Conclusions: A chronic complete ACL tear with an abnormal reattachment to a neighbouring anatomical structure (most often by healing to the PCL) is a relatively common phenomenon. This particular ACL tear morphology appears to be more difficult to clinically and radiologically diagnose than free ended and unattached tear patterns. A reattached ACL tear is more likely to produce an end-point on Lachman testing and a reduced degree of anterior translation thereby increasing the probability of false negative findings on physical examination. However, the abnormal vertical orientation may still give rise to rotational (pivoting) instability that is similar to a completely ruptured but free and unattached ACL tear pattern. Reattached ACL tears were also found to be less likely to produce a convincing ‘clunk’ on pivot shift testing and indeed can mimic or be confused with a partially injured or even intact ligament on MRI scans, although these latter variables did not reach statistical significance. These findings should be borne in mind when assessing chronic knee injuries.

P19-93 Biomechanical comparison of tibial eminence fracture fixation with high-strength suture, endobutton, and suture anchor O. Hapa1, A. Barber2, R. Ozden3, E. Bozdag4, E. Sunbuloglu5, Y. Dogramaci3 1 Mustafa Kemal University, Orthopedics and Traumatology, Ankara, Turkey, 2Plano Sports Medicine Center, Dallas, United States, 3 Mustafa Kemal University, Orthopedics and Traumatology, Hatay, Turkey, 4Istanbul Technical University, Mechanical Engineering, Istanbul, Turkey, 5Istanbul Technical University, Istanbul, Turkey Objectives: Purpose was to biomechanically compare anterior cruciate ligament tibial bony avulsion fixation by a suture anchor, Endobutton, and high strength sutures subjected to cyclic loading. Methods: Using 49 ovine knees type III tibial eminence fractures were created and 7 different types of repairs were performed. Each repair groups contained 7 specimens. The repair groups were: No. 2 FiberWire (Arthrex, Naples, FL, USA); No. 2 UltraBraid (Smith &

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Nephew, Andover, MA, USA); No. 2 MaxBraid (Arthrotek, Warsaw, IN, USA); No. 2 Hi-Fi (ConMed Linvatec, Largo, FL, USA); No. 2 OrthoCord (DePuy-Mitek, Norwood, MA, USA), Ti-Screw suture anchor (Arthrotek, Warsaw, IN, USA), and the titanium Endobutton (Smith & Nephew, Andover, MA, USA). These constructs were cyclically loaded (500 cycles, 0–100 N, 1 Hz) in the direction of the native ACL and destructively tested (100 mm/min). Endpoints included ultimate failure load (N), pull-outstiffness (N/mm), cyclic displacement (mm) after 100 cycles, between 300 and 500 cycles, and after 500 cycles, and mode of failure. Bone density testing was performed in all knees. Results: Bone density was not different among the groups. The Endobutton had a higher ultimate failure load than FiberWire, UltraBraid, Hi-Fi, and the suture anchor group (P \ .05). MaxBraid and OrthoCord had higher failure loads than the suture anchor group (P \ .05). MaxBraid also had a higher failure load than the Hi-Fi group (P \ .05). Stiffness was not statistically different for the various tested constructs. After 100 cycles, the Endobutton had less displacement that FiberWire, UltraBraid, MaxBraid, and Hi-Fi groups (P \ .05). The suture anchor group had less displacement than Hi-Fi and FiberWire groups (P \ .05). The displacements of the different tested constructs between 100 and 500 cycles and total displacements after 500 cycles were not statistically different. The predominant failure mode was suture rupture. Conclusions: Under cyclic loading conditions in an ovine model, Endobutton fixation of tibial eminence fractures provided greater initial fixation strength than suture anchor fixation or fixation with various high strength sutures except for OrthoCord. Initial cyclic loading of ACL tibial eminence fractures should consider the strength of the repair construct because conventional suture repair even with UHMWPE sutures may not provide enough strength.

P19-114 Comparison of tunnel placements and clinical results of singlebundle anterior cruciate ligament reconstruction before and after starting the use of double-bundle technique P. Suomalainen1, A.-S. Moisala1, A. Paakkala2, P. Kannus3, T. Ja¨rvela¨4 1 Tampere University Hospital (TAYS), Tampere, Finland, 2Tampere University Hospital (TAYS), Radiology, Tampere, Finland, 3UKK Institute, Tampere, Finland, 4Sports Clinic and Hospital Mehila¨inen, Tampere, Finland Objectives: Investigate whether the locations of the grafts in single bundle (SB) anterior cruciate ligament (ACL) reconstruction have changed to more anatomical as the double bundle (DB) method has become more familiar. Methods: Both groups were operated using anteromedial (not transtibial) portal and free-hand technique (Group A (N = 25) in 2003, Group B (N = 25) in 2007). The evaluation methods preoperatively and at the 2-year follow-up (two blinded examiners): clinical examination, stability measurement (KT-1000 arthrometer), the International Knee Documentation Committee (IKDC) and the Lysholm knee scores. A musculoskeletal radiologist made tunnel measurements from the magnetic resonance imaging (MRI). Results: The average tunnel placement in the femoral side: from Blumensaat’s line 27% (Group A), 26% (Group B), from the posterior edge of the femur 32% (Group A), 29% (Group B). The average tunnel placement in the tibial side: from the anterior edge 45% (Group A), 45% (Group B), from the lateral side 57% (Group A) and 54% (Group B) (P = 0.024). Graft failures ending up to revision ACL surgery: 4 (Group A), 0 (Group B) (P = 0.045). Operation time reduced from 2003 to 2007 19 min (P = 0.001). Conclusions: Tunnel placement at the femoral side was already very low (anatomical) among patients operated in 2003. No significant

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 difference was found when comparing to the patients operated in 2007. There were significantly more graft failures in the former group suggesting that the use of the DB method in ACL surgery in 2007 may have also improved the technique and results of the SB ACL reconstruction.

P19-117 Correlation between tibial guidewire position on intraoperative X-ray and intraoperative tibial bone tunnel communication in double-bundle anterior cruciate ligament reconstruction Y. Hara1, T. Tomihara1, G. Yoshida1, Y. Hashimoto2, N. Shimada1 1 Shimada Hospital, Orthopaedic Surgery, Habikino, Japan, 2Osaka City University Graduate School of Medicine, Department of Orthopaedic Surgery, Osaka, Japan Objectives: Accurate and anatomic tunnel placements are essential to perform the anterior cruciate ligament (ACL) reconstruction successfully. If the position of the bone tunnel is too near in anatomic double-bundle (DB) ACL reconstruction, there is a risk of bone tunnel communication. Intraoperative bone tunnel communication may affect intraoperative graft passage as well as postoperative outcomes. In tibial side, there is a possibility that soft-tissue remnant of the tibial ACL footprint makes arthroscopic control difficult. Therefore, to avoid the tibial tunnel communication, we have taken intraoperative X-ray after the tibial guidewire insertion. The aim of this study was to evaluate the correlation between intraoperative tibial guidewire position on intraoperative X-ray and bone tunnel communication in DB ACL reconstruction. Methods: Forty-four patients who underwent primary anatomical double-bundle ACL reconstruction using autogenous hamstring graft were included in this study. In all patients, intraoperative lateral X-ray was taken photograph to adjust a position of anteromedial (AM) and posterolateral (PL) tibial guidewire, and postoperative 3-dimensional CT (3-D CT) was taken at 1 week postoperatively to evaluate the tunnel communication. The distance between the midpoints of AM and PL tibial guidewire width on medial tibial plateau at the lateral X-ray photograph (W) and the amount of anteromedial bundle (AMB) and posterolateral bundle (PLB) graft radius (G) were measured. The difference between W and G was calculated (W minus G, W - G). Patients were divided into two groups depending on whether two tibial bone tunnels were separated or not (Group S: tibial tunnels were separated, Group C: tibial tunnels were communicated). Results: Intraoperative tibial bone tunnel communication between AM and PL bone tunnels occurred in 25.0% of patients (11 patients) at tibial plateau on 3-D CT. There were 33 patients in Group S and 11 patients in Group C. The mean age was 23.0 (range 14–39) years in Group S and 24.1 (range 16–39) years in Group C. W in Group S (6.7 mm, range 4.9–8.9 mm) was significantly longer than that in Group C (5.9 mm, range 4.6–6.8 mm) (p \ 0.05). There was no significant difference between G in Group S (5.8 mm, range 5.3–6.5 mm) and that in Group C (5.9 mm, range 5.3–6.5 mm) (p [ 0.5). W-G in Group S (1.0 mm, range -0.9–3.2 mm) was significantly longer than that in Group C (0 mm, range -0.9–1.1 mm) (p \ 0.01). Conclusions: The influence of bone tunnel communication in DB ACL reconstruction on postoperative clinical results is still unknown. However, to avoid intraoperative tibial bone tunnel communication, the distance between AMB and PLB guidewire on intraoperative lateral X-ray should be at least 1.1 mm longer than the amount of AMB and PLB graft radius.

S157 P19-132 Comparison of rotatory stability after the double bundle and single bundle ACL reconstruction K. Okazaki1, T. Izawa1, Y. Tashiro1, S. Matsuda1, Y. Iwamoto1 1 Kyushu University Graduate School of Medical Sciences, Orthopaedic Surgery, Fukuoka, Japan Objectives: Anterior cruciate ligament (ACL)—deficient knee causes rotatory instability as well as anterior instability. Recent kinematic studies of ACL suggest that the double bundle ACL reconstruction is expected to improve the rotatory stability than the single bundle reconstruction. However, the clinical advantage of double bundle reconstruction is still controversial. We previously reported the method to assess the rotatory stability quantitatively using an open MRI. In this study, we compare the clinical results of double bundle and single bundle ACL reconstruction using this method. Methods: The study includes 50 patients with an isolated ACL injury. Single bundle (SB) reconstruction using quadruped hamstring tendon was performed for 23 patients. Double bundle (DB) reconstruction using two doubled hamstring tendons was performed for 27 patients. The patients were evaluated 1 year after the surgery. Slocum’s anterolateral rotatory instability (ALRI) test was performed under the open MRI gantry. ALRI was evaluated by subtracting the anterior tibial translation at the medial compartment from that at the lateral compartments in the sagittal images. Anterior stability was evaluated by KT-2000 arthrometer. Lysholm score and Tegner score was recorded. Results: ALRI was significantly smaller in DB group than SB group (1.3 ± 1.8 vs. 4.6 ± 3.2 mm). Anterior instability measured by KT2000 was significantly smaller in DB group than SB group (1.1 ± 2.2 vs. 2.6 ± 2.2 mm). There were no significant differences between two groups for Lysholm score and Tegner score.

Clinical results Group S (n = 23)

Group D (n = 27)

Significance p

Lysholm knee score/preoperative

64.8 ± 9.6

67.0 ± 8.1

.452

Lysholm knee score/postoperative

96.0 ± 4.3

97.4 ± 2.3

.751

Tegner activity score/preinjury

6.3 ± 1.0

6.2 ± 0.7

.895

Tegner activity score/postoperative 5.7 ± 1.3

5.8 ± 1.3

.956

Pivot-shift test/(-)

13 patients (57%)

23 patients (85%)

.058

Pivot-shift test/(+)

9 patients (39%)

4 patients (15%)

Pivot-shift test/(++)

1 patient (4%) 0 patient (0%)

Anterior side-to-side difference

2.6 ± 2.2 mm 1.1 ± 2.2 mm

ALRI values sideto-side difference

4.6 ± 3.2 mm 1.3 ± 1.8 mm \.001

.013

Conclusions: Both reconstruction methods showed satisfactory results in Lysholm score and Tegner score. However, the DB group showed significantly better stabilization for rotatory and anterior tibial translation than the SB group.

123

S158 P19-134 Comparison of tunnel locations of double bundle ACL reconstruction created by conventional transtibial technique with anatomic tunnel locations using a 3-D CT model Y.S. Lee1, B.K. Lee1, J.A. Sim1, J.H. Kwak1, S.W. Nam1 1 Gachon University School of Medicine, Department of orthopedic surgery, Incheon, Republic of Korea Objectives: The purposes of this study were: (1) to compare the tunnel locations created by conventional transtibial technique with reference data and (2) to evaluate the affecting factors that make it difficult to position the correct femoral tunnel or break the bone bridge between tibial two tunnels. Methods: A prospective study was performed with 28 patients that underwent double bundle ACL reconstruction. Locations of each tunnel using an anatomic coordinate axes method (ACAM) were measured. Measurements included: bone bridge between tibial two tunnels (BB), height from the union (HU), body mass index (BMI), the ratio between the length of Blumensaat’s line and the anterior– posterior length of lateral femoral condyle (RBL), and the ratio between anterior–posterior and proximal–distal length in the medial wall of lateral femoral condyle (RAPPD). Results: The posterior-anterior direction of femoral AM tunnel, the proximal–distal direction of femoral PL tunnel, and the posterioranterior direction of femoral PL tunnel were statistically significant when compared with reference data. In the correlation analysis between BB or HU and other variables, the AM tunnel of tibia and RBL showed moderate negative correlation. The cutoff point of tunnel breakage in the RLB was 1.14. Conclusions: Conventional transtibial drilling technique of double bundle ACL reconstruction does not reproduce correct tunnel locations when compared with reference data. This problem is more prominent if the anterior–posterior length of lateral femoral condyle is lessened when compared to the length of Blumensaat’s line. Therefore, we must consider the limitation of this technique and should selectively approach according to the characteristics of each patient.

P19-136 Comparative analysis of femoral tunnels between outside-in and transtibial double bundle ACL reconstruction: a 3-dimensional computed tomography study Y.S. Lee1, J.A. Sim1, J.H. Kwak1, S.W. Nam1, B.K. Lee1 1 Gachon University School of Medicine, Department of orthopedic surgery, Incheon, Republic of Korea Objectives: The Objectives of this study were (1) to compare locations of the femoral tunnels created by an outside-in, transtibial technique, and the reference data and (2) to compare the diameter of the tunnel entrance based on the real reaming size. Methods: A prospective comparative study was performed with 20 outside-in and 22 transtibial double bundle ACL patients. CT scans of the operated knees of the outside-in and transtibial group were performed at 1.25 days (ranging from 1 to 3) and 2.7 weeks (ranging 3 day to 4 weeks), respectively. 3D surface models were then produced. For all 3 plane data sets, the positions of the femoral tunnels were measured using an anatomic coordinate axes method (ACAM). For a comparison of the tunnel diameter, the difference between the real reaming and measured diameter were determined first in the CT. Subsequently, the differences in the outside-in and transtibial technique were compared. Results: In the comparison between outside-in and reference data, the posterior-anterior direction of PL tunnel showed an anterior position compared to reference data, even though it was positioned more posteriorly compared to that of transtibial technique (p = 0.003). In the comparison between transtibial and reference data, the posterior-

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 anterior direction of the AM and PL tunnels showed an anterior position compared to reference data (p = 0.019 and 0.005). The transtibial technique showed significant different diameters in both AM and PL tunnels (p = 0.000 and 0.000). Conclusions: The outside-in technique is better for restoring the normal insertional anatomy than the transtibial technique, particularly the AM tunnel of the femur. On the other hand, it is difficult to restore the normal insertion of the PL bundle regardless of the reconstruction technique. The transtibial technique showed an oblique tunnel configuration, which suggests that eccentric reaming is unavoidable because the reaming angle is determined by the tibial tunnel.

P19-159 Determining factors influencing the angle between the intraarticular part of an ACL graft and the tibial tunnel: a pilot study H. Van der Bracht1, B. Stuyts2, B. Page3, J. Bellemans4, P. Verdonk5 1 Gent University Hospital, Gent, Belgium, 2Sintt-Augustinus, GZA Ziekenhuizen, Wilrijk, Belgium, 3Stellenbosch University, Tygerberg, South Africa, 4University Hospitals Leuven, Catholic University Leuven, Orthopaedic Department, Pellenberg, Belgium, 5 Gent University Hospital, Orthopaedics, Gent, Belgium Objectives: Recently a new technique for ACL revision surgery was described. In this technique a tibial tunnel is drilled from the lateral side of the tibia towards the centre of the ACL footprint to deal with tibial bone stock problems often present in ACL revision surgery. Because the tibial tunnel is drilled from the lateral side of the tibia towards the ACL footprint one can expect a ‘‘killer angle’’ created between the lateral tibial tunnel (LTT) and the intra-articular part of the ACL graft. This pilot study was performed to evaluate the different determining factors influencing the angle between the intraarticular part of an ACL graft and the tibial tunnel (AIAT). Methods: One human cadaveric knee was used for this study. The skin and the fat were stripped around the knee as was the anterior knee capsule but all ligamentous structures were kept intact. The center of the tibial ACL footprint and the center of the femoral ACL footprint were marked. A tibial ACL drill-guide was used and K-wires were drilled 2 cm medial and 2 cm lateral of the center of the tibial tubercle with drill-guide angles of 45, 55 and 65. For each of these 8 situations a picture was taken at a right angle to the surface created by the center of the femoral ACL footprint, the center of the tibial ACL footprint and the cortical entry point at the tibia of the K-wire with the knee in extension and at 30, 60, 90 and 120 of knee flexion. This was repeated three times. AIAT were measured digitally from the pictures. Results: The more the knee is flexed, the bigger the AIAT for the medial tibial tunnel (MTT) (24–44) and LTT (49–84). For the MTT the AIAT is increased with bigger drill-guide angles but for LTT the AIAT is decreased with bigger drill-guide angles. The AIAT was bigger for the LTT compared to the MTT for the same drill-guide angle and the same amount of knee flexion in each situation varies from 10 to 20 with a 65 drill-guide angle and up to 30–50 difference with a 45 drill-guide angle. The difference between both techniques decreased with bigger drill-guide angles but was not influenced by the amount of knee flexion. Conclusions: The AIAT is bigger for the LTT compared to the MTT and is influenced by the amount of knee flexion and the drill-guide angle. The clinical consequence of the bigger AIAT needs to be further determined.

P19-176 Double bundle reconstruction of anterior cruciate ligament with a femoral fixation device T. Hiranaka1 1 Takatsuki General Hospital, Department of Orthopaedic Surgery and Joint Surgery Centre, Takatsuki, Osaka, Japan

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Double bundle reconstruction of anterior cruciate ligament (ACL) has been focused on. However, there is virtually no device but the Endobutton for double bundle ACL reconstruction. EZLoc is a new femoral fixation device which has metal loop for graft suspension and lever arm for fixation on the lateral femoral cortex. The aim of this study is to compare clinical results of double bundle ACL reconstruction with the EZLoc and with the Endobutton. Methods: Forty patients underwent double bundle ACL reconstruction in our institution were included in this study. There were twenty patients with the Endobutton and twenty with EZLoc. All operation were carried out under fluoroscopic-based navigation system and clinical outcome, radiographic findings and complication were compared between the groups. Results: Satisfactory clinical results were found in both groups. Conclusions: We conclude that the EZLoc is to be an alternative femoral fixating device to the Endobutton representing an equivalent clinical outcomes. No lift off of the device and less frequent of tunnel enlargement is an advantages of the EZLoc. Although our first impression of the EZLoc is encouraging, further investigation is required because this study is not a case–control study and follow up duration was short.

P19-246 Comparison of femoral bone tunnel communication after ACL double bundle reconstruction: Outside-in versus inside-out technique T. Tomihara1, G. Yoshida1, Y. Hara1, M. Taniuchi1, N. Shimada1 1 Shimada Hospital, Orthopaedic Surgery, Habikino, Japan Objectives: To make femoral tunnels in ACL double bundle reconstruction (ACL-DBR), several techniques have been reported. However, little is known concerning the relationship between the methods of drilling and postoperative femoral tunnel communication. The purpose of this study is to evaluate the femoral bone tunnels after ACL-DBR on CT, and to compare the femoral tunnel communication with outside-in technique to that with inside-out technique. Methods: Forty-five patients who underwent ACL-DBR with autogenous hamstring graft were included in this study. In Group A, femoral tunnel was drilled by outside-in technique. There were 21 patients (11 men and 10 women) with a mean age of 24.1 years (14–56). In Group B, femoral tunnel was drilled by inside-out technique through far anteromedial portal. There were 24 patients (12 men and 12 women) with a mean age of 27.8 years (14–63). CT scan was taken at 1 week and 6 months postoperatively. On reconstructed CT images, tunnel divergency (Angle F) at 1 week and distance between the antero-medial bundle (AMB) and postero-lateral bundle (PLB) apertures at 1 week (D-1) and 6 months (D-6) were measured. In addition, the patients in Group A and B were divided into subgroups depending on whether the femoral bone bridge between the tunnels remained or not at 6 months. Group A-R: with the bone bridge in Group A, Group A–C: without the bone bridge in Group A, Group B-R: with the bone bridge in Group B, Group B–C: without the bone bridge in Group B. Results: Average Angle F was 11.8 (3.6–23.8) in Group A and 10.1 (5.5–16.4) in Group B (p = 0.14). Average D-1 was 1.9 mm (0.6–3.9) in Group A and 1.7 mm (0.4–3.7) in Group B (p = 0.47). Average D-6 was 0.6 mm (0–2.5) in Group A and 0.5 mm (0–3.0) in Group B (p = 0.61). The femoral bone bridge remained at 6 months in 9 patients (42.9%) of Group A and in 8 patients (33.3%) of Group B (p = 0.51). In subgroups of Group A, D-1 was 2.6 mm (1.5–3.6) in Group A-R and 1.4 mm (0.6–2.6) in Group A–C (p = 0.002). Angle F was 11.6 (5.3–18.3) in Group A-R and 12.2 (3.6–23.8) in Group A–C (p = 0.82). In Group B, D-1 was 2.6 mm (1.8–3.7) in Group B-R and 1.3 mm (0.3–2.4) in Group A–C (p = 0.002). Angle F was 11.2

S159 (6.5–16.4) in Group B-R and 9.5 (5.5–14.7) in Group B–C (p = 0.24). Conclusions: Since femoral tunnel divergency and distance between AMB and PLB of outside-in technique were similar to those of insideout technique, the drilling technique did not influence the postoperative bone tunnel communication in ACL-DBR. In Group A-R and B-R, average D-1 was more than 2 mm. Thus, to avoid the postoperative femoral bone tunnel communication, more than 2 mm distance between apertures of AMB and PLB should be preserved at the surgery in both techniques.

P19-268 Complex regional pain syndrome type I post reconstruction of anterior cruciate ligament: treatment with percutaneous radiofrequency N. Santos Assis de Melo1, F. dias Assis2, I.D. de Oliveira Neri1, H. Pidner Neto3, J.F. Santos Duarte Lana4 1 Hospital 22 de Outubro, Mogi Mirim, Brazil, 2Clı´nica Singular, Campinas, Brazil, 3Hospital da Beneficeˆncia Portuguesa de Bele´m (PA), Bele´m, Brazil, 4IMOR, Uberaba (MG) e Clı´nica Singular, Campinas, Brazil Objectives: The authors report the case of a 42-year old male who had submitted to reconstruction of the anterior cruciate ligament of the right knee with hamstrings and evolved with complex regional pain syndrome Type I. The therapeutic modality instituted was percutaneous lumbar sympathectomy conventional radiofrequency. Methods: M.F.B., male, 42 years, submitted to reconstruction of the primary right anterior cruciate ligament with hamstrings. Normal results were presented on physical examination, laboratory study, cardiologic and pre-anesthetic evaluation. Patient received epidural anesthesia and a catheter was placed within the epidural space to provide post-operative analgesia for the duration of his hospital stay, standard anesthesia procedure adopted at that hospital. On the 3rd post-operation day, the patient reported hyperalgesia and swelling, pallor and a decrease in the temperature of the operated joint compared to the contra lateral one. Patient reported that opioid and non-opioid painkillers prescribed were inefficacious to control his severe pain. There was no stroke of any significance, nor flogistic signs in the operated knee. The patient had no fever and laboratory studies revealed nothing abnormal. On examination by a vascular surgeon, nothing noteworthy was observed. It was decided that the patient would be admitted in a pain management center, on the suspicion of sympathetic hyperactivity, suggesting an acute case of complex regional pain syndrome Type I (CRPS I). A pain specialist was requested to evaluate the patient, who recommended a course of 25 mg amitriptiline every 8 h, 300 mg gabapentine once a day and 5 mg of methadone every 8 h. Pain diminished significantly. On the following day, a diagnostic block of the right L2, L3 and L4 sympathetic ganglia was carried out under fluoroscopic guidance, using 0.5 ml of lidocaine 2% in each ganglion. Pain and autonomic signs were completely resolved for 4 h. Aftermost the patient was submitted to conventional RF treatment of the sympathetic ganglia, mentioned above (Fig. 1). Three 145 mm length, 10 mm active tip 22G needles were used for the procedure. Each ganglion received RF lesioning for 60 s at 80C (Fig. 2) with Neurotherm radiofrequency machine. Radiological guidance was used with a tunnel vision technique and the procedure had occurred without any events. Results: Diagnosis of complex regional pain syndrome Type I was confirmed and the therapeutic modality instituted was percutaneous lumbar sympathectomy conventional RF, culminating in full resolution of the clinical features. Conclusions: The authors conclude that RF lesioning of the lumbar sympathetic chain could be an adequate technique for patients with

123

S160 CRPS type I of the lower limbs, mainly when associated with a multidisciplinary approach. Further studies with large samples are needed to evaluate the efficacy of this method.

P19-279 CT and MRI controlled prospective evaluation of bovine bone interference screws for ACL-reconstruction using BPTB-graft K. Tecklenburg1, P. Burkart2, N.F. Friederich3, C. Fink4 1 Orthopaedic Surgery Munich, Munich, Germany, 2St. Anna Klinik, Luzern, Switzerland, 3Kantonsspital Bruderholz, Klinik f. Orthop. Chirurgie & Traumatologie, Bruderholz, Switzerland, 4Sportsclinic Austria, Innsbruck, Austria Objectives: Metallic interference screws have been continuously replaced by other screw materials such as bioresorbable screws or allograft screws. Allograft bone screws prove to be biologically safe, provide adequate graft fixation in ACL reconstruction and seem to be completely remodelled by human bone. However, due to limited availability and high production cost, a bovine xenograft screw has now been introduced. We evaluated this new interference screw made from bovine bone regarding clinical results, screw resorption and new bone formation. Methods: 20 consecutive patients who underwent ACL reconstruction have been prospectively evaluated in two surgical centres. A bovine xenograft interference screw has been used for tibial fixation of a patellar tendon autograft in ACL reconstruction. Subjective and clinical IKDC were evaluated preoperatively and for follow up until 24 months post op. CT scans were performed preoperatively and at 3, 12 and 24 months. Additionally, MRI scans were obtained at 3 and 24 months postoperatively. CT and MRI scans were used to determine screw resorption, bony remodeling around the screw and possible inflammatory reaction of the surrounding bone. Results: Screw breakage during insertion occurred in 2 cases. Subjective and clinical results at any time of follow up were satisfying and comparable to other clinical results of ACL reconstruction using tibial interference screw fixation. No inflammatory response or other adverse effects could be detected on MRI after 3 and 24 months in all cases. CT scans: bone block incorporation was completed at 3 months in all patients. Signs of screw degradation with bony remodeling at the implant site were clearly visible in 8 cases. In these cases first signs of degradation were observed after 12 months with almost complete bony remodeling after 24 months. However, in 12 cases the bovine screw seemed to be inert without any alteration, degradation or bony replacement even after 24 months. Conclusions: Bovine interference screws provided adequate graft fixation in all cases and graft healing was successful with good clinical results. No inflammatory response could be detected on MRI. However, after 24 months screw degradation and bony replacement had only taken place in 8 out of 20 cases. Compared to an allograft bone screw that showed complete bony remodeling in all cases in an earlier study, the bovine interference screw seems to have a more variable degradation behavior.

P19-317 Effect of gamma radiation and freeze drying on the maximum tensile strength of the demineralised cortical bone S. Elnikety1, C. Bendegrass1, G. Blunn1 1 UCL, Biomedical Engineering Lab, London, United Kingdom Objectives: Demineralised Bone Matrix (DBM) is widely used in Orthopaedics and dentistry as a bone graft substitute and may be used to augment bone formation in load bearing applications. In this study we examine the effect of gamma irradiation and freeze drying on the tensile strength of demineralised cortical bone (DCB).

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: Tibias were harvested from skeletally mature ewes and cut into bony strips. Demineralisation was done using 0.6 M HCl and confirmed by X-ray. Specimens were washed in phosphate buffer solution (PBS) until a pH of 7.0 ± 0.2 was achieved in the washing solutions. Specimens were allocated into 4 groups; group (A) non freeze dried non gamma irradiated, group (B) freeze dried non gamma irradiated, group (C) non freeze dried gamma irradiated mention the level of gamma irradiation and group (D) freeze dried and gamma irradiated. On the day of testing, the specimens were rehydrated in phosphate buffered saline for 1 h and then trimmed into dog bone shape. The maximum tensile force and stress were measured. Statistical analysis using the Mann–Whitney U test was carried out. Results: The Median of maximum tensile force for group (A) was 218 N (95% CI: 147.9–284.7 N), group (B) was 306 N (95% CI: 154.1–488.6 N), group (C) was 263 N (95% CI: 227.8–315.6 N), group (D) was 676 N (95% CI: 127–1,094.9 N). The median stress for group (A) was 15 N/mm2 (95% CI: 8.8–21.5 N/ mm2), group (B) was 41 N/mm2 (95% CI: 31.4–50.8 N/mm2), group (C) was 18.7 N/mm2 (95% CI: 12.8–22.4 N/mm2), group (D) was 54.6 N/mm2 (95% CI: 7.9–90.6 N/mm2). Group (D) results were statistically higher (p = \0.05) compared to group (A) and (C), while there was no statistical significance compared to group (B). Conclusions: Previously published studies suggested the possibility of using DCB as ACL graft substitute. We examined the effect of gamma radiation as the most common sterilisation technique in medical field and the freeze drying as a possible technique for long term storage on the tensile strength of the DCB. Freeze drying significantly increases the tensile strength of the DCB while gamma irradiation has no significant effect. Our results indicate that freeze dried gamma irradiated DCB can be used as a ligament substitute.

P19-323 Computer-assisted navigation for acl reconstruction: a randomized comparative study between the performance of the femoral tunnel in a transtibial or anteromedial way J. Minguell1, E. Castellet2, J. Cortina3, J. Nardi3, E. Caceres3, L. Carrera3 1 Hospital Vall d’Hebron, Knee Unit, Orthopaedic Department, Barcelona, Spain, 2Hospital Vall d’Hebron, Orthopaedic Surgery, Barcelona, Spain, 3Hospital Vall d’Hebron, Barcelona, Spain Objectives: The correct positioning of the femoral and tibial tunnel is an important requirement to achieve a proper stability in the reconstruction of the anterior cruciate ligament (ACL). The aim of our study is to compare anteroposterior stability and control of internal and external rotation in ACL reconstruction using a navigation system, by means of two common femoral tunneling techniques, the anteromedial and the transtibial. Methods: Between July 2009 and July 2011, a prospective randomized study of 53 patients was carried out, comparing the realization of the femoral tunnel through an anteromedial or transtibial way in the ACL reconstruction by use of the OrthoPilot navigation system (B. Braun-Aesculap, Tuttlingen, Germany). 26 patients were assigned to the transtibial (TT) group and another 27 to the anteromedial (AM) group. The presence of differences between groups was assessed in terms of improvement in the control of the anteroposterior translation, internal rotation and external rotation, between before and after the surgery. We have used statistical program SPSS 17 to analize data.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: The improvement in the postoperative anteroposterior translation with respect to preoperative was on average 9.6 mm in the TT group, which represents a 64.4%, and 8.6 mm in the AM group representing a 61.5%. There were no statistically significant differences between the two groups (p [ 0.05). As for the internal rotation, the improvement was 4 (23.7%) in the TT group and 4.7 (27.5%) in the AM group, not being significant the differences (p [ 0.05). The external rotation was controlled on average 2.5 (14.1%) in the TT group and, 1.9 (10%) in the AM group (p [ 0.05). Conclusions: There were no statistically significant differences in terms of improving the anteroposterior stability, or to control internal and external rotation, in ACL reconstruction by use of a navigation system, making the femoral tunnel either through an anteromedial or transtibial way. The two methods of reconstruction of ACL are useful to achieve rotational and translational stability.

P19-330 Comparison of clinical results according to amount of remnant in anterior cruciate ligament reconstruction using anterior tibialis tendon allograft S.J. Lee1, J.J. Lee2, C.H. Choi3 1 Yonsei University College of Medicine, Seoul, Republic of Korea 2 Yonsei University College of Medicine, Orthopedic Surgery, Seoul, Republic of Korea 3Yonsei University College of Medicine, Gangnam Severance Hospital, Orthopedic Surgery, Seoul, Republic of Korea Objectives: We analyze the clinical results of anterior cruciate ligament (ACL) reconstruction with the remnant-preserving technique by use of a tibialis anterior tendon allograft according to the amount of the tibial remnant of the ACL. Methods: This study included 107 patients who enabled at least 12 month follow-up among patients who had received ACL reconstruction using tibialis anterior tendon allograft from March 2008 to March 2010. Through a retrospective review on their operative videos, we divided the subjects into two groups. Group I comprised 62 patients with an ACL remnant of more than 7 mm, and group II comprised 45 patients with less than 7 mm. At final follow-up, the objective stability was evaluated with the pivot-shift test, side to side difference in KT-2000 arthrometer and Telos stress radiograph. The subjective clinical result was assessed using IKDC subjective score, Lysholm score and Tegner activity score. For the functional test, the single-legged hop test, single limb standing test and modified reproduction of passive positioning test were conducted. Results: There were no significant differences between two groups with respect to the objective stability with the KT-2000 arthrometer, stress radiograph and pivot-shift test (P = .317, P = .209 and P = .390, respectively). No significant differences in subjective clinical scores were observed with the IKDC subjective score, Lysholm score and Tegner activity score (P = .317, P = .078 and P = .280, respectively). There were also no significant differences between the groups in terms of functional test. In the single-legged hop test, it was measured as 93% (group I) and 91% (group II) comparing with the contralateral healthy knee (P = .220). Regarding the single limb standing test (P = .943) and modified reproduction of passive positioning test (P = .255), there were no significant differences between the groups. Conclusions: Although it was expected that the group having more preserved remnant in the ACL reconstruction surgery would show better results in their stabilities, subjective clinical results and functional results, there were no significant differences.

S161 P19-337 CT scan evaluation of the tunnel positioning in ACL reconstruction using an ‘‘outside in’’ femoral guide with reversed arthroscopic portals and an ACL tibial stump preserving procedure F. Buscayret1, B. Sonnery-Cottet2, J. Barth3, N. Graveleau4, P. Chambat5 1 Polyclinique Saint-Jean, Montpellier, France, 2Orthopaedic Center Santy, Lyon, France, 3Clinique des Ce`dres, Echirolle, France, 4CMC Paris V, Paris, France, 5Centre Orthope´dique Santy, Lyon, France Objectives: To evaluate the femoral tunnel placement using an outside/in femoral guide and reversed arthroscopic portal and the tibial tunnel placement when the ACL tibial stump is preserved. Methods: 63 patients who had an ACL reconstruction with a tripled semitendinosous graft (STG) or a doubled STG graft were included in the series. After minimal debridement of the lateral wall of the notch, the femoral tunnel was performed from outside to inside using a specific guide introduced through the anterolateral portal and the scope through the anteromediaI portal. For the tibial tunnel, the ring shaped target of the guide was positioned in the anteromedial region of the ACL footprint. A 3D CT Scan of the reconstructed knee was obtained for all the patients. The tibial tunnel placement was evaluated on an axial view with the tibial quadrant method. For the femoral tunnel, the scan was oriented in a true lateral view by superimposing the posterior curvature of the femoral condyles. Using the transparency mode, the radiologic Blumensaat line was identified and marked. Using the 3D mode, the medial femoral condyle was progressively removed until the roof of the intercondylar notch was superimposed on the radiologic blumensaat line. The position of the femoral tunnel was determined using the quadrant method (Fig. 1).

Fig. 1 Radiologic Blumensaat line identification using the transparency mode (a). Using the 3D mode (b), the medial condyle is removed until the roof is superimposed on the marked Blumensaat line (c). Quadrant is then applied (d)

123

S162

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Table 1 Reported values in previous anatomic and r Authors

Type of study

AM bundle femoral

AM bundle tibial

% of notch % of % of AP % of ML depth lateral length width wall height Zantop

Radiologic 18.5

Colombet Anatomic

26.4

Tsukada

Anatomic

25.9

Lorentz

Radiologic 21

Our study Radiologic 19.2

22.3

36

–

25.3

36

–

17.8

37.6

53.5

22

41

52

26

37.6

48

Results: The center of the tibial tunnel was positioned at 37.6% (SD 3.3%) of the anteropostrior length and at 48% (SD 2.4%,) of the mediolateral width of the tibial plateau. The femoral tunnel was positioned at 19.2% (SD 2.36) of the depth of the notch and 26% (SD 3.9) of the lateral wall height. No posterior tunnel wall blow out have been observed. These results are similar to the values reported in previous anatomic and radiologic studies of femoral and tibial bundle insertion (Table 1). Conclusions: The 3D CT scan evaluation demonstrated a precise and reliable femoral and tibial tunnel placement. The outside in technique with reversed arthroscopic portals allows to perform femoral anatomic tunnel placement with minimal notch debridement. Preserving the ACL tibial stump did not compromise tibial tunnel placement.

P19-385 Double-bundle versus single-bundle ACL reconstruction: prospective, randomized study with 5-year clinical results P. Suomalainen1, T. Ja¨rvela¨2, M. Ja¨rvinen1 1 Tampere University Hospital (TAYS), Tampere, Finland, 2Sports Clinic and Hospital Mehila¨inen, Tampere, Finland Objectives: Although various different techniques to reconstruct the torn ACL are available today, the main focus seems to be in the single-bundle versus double-bundle debate. The purpose of this study is to compare the clinical results of the patients with double-bundle or single-bundle ACL reconstruction in a prospective, randomized clinical trial with a minimum of 5-year follow-up. Methods: 90 patients were randomized into three groups of ACL reconstruction with hamstring autografts using aperture interference screw fixation: double-bundle ACL with bioabsorbable screw fixation (DB-Group) (N = 30), single-bundle ACL with bioabsorbable screw fixation (SBB-Group) (N = 30), and single-bundle ACL with metallic screw fixation (SBM-Group) (N = 30). The evaluation methods were clinical examination, KT-1000 arthrometric measurement, the International Knee Documentation Committee (IKDC) and the Lysholm knee scores. Radiographic evaluation was made by a musculoskeletal radiologist, who was unaware of the patients’ clinical data. All the operations were performed by one experienced orthopaedic surgeon, and all clinical assessments were made by one blinded and independent examiner. Results: Preoperatively there were no differences between the groups. 76 patients (84%) (21 in the DB-Group, 28 in the SBB-Group, and 27 in the SBM-Group) could be examined at the minimum of 5-year follow-up (range, 60–75 months). 11 patients (7 in the SBB-Group, 3 in the SBM-Group, and 1 in the DB-Group) had graft failure during the follow-up and went to ACL revision surgery (P \ 0.05). At 5 years, the DB-Group had the best anterior stability measured by the KT-1000 arthrometer (1.6 mm in the DB-Group, 2.2 mm in the SBBGroup, and 2.3 mm in the SBM-Group), although the difference was

123

not statistically significant. In the DB-Group, 20% of the patients had degenerative changes in the medial femorotibial compartment and 10% in the lateral femorotibial compartment, while the corresponding figures were 33 and 18% in the SB-Groups (NS). No significant group-differences were found in the knee scores. Conclusions: Our randomized controlled study showed that the double-bundle ACL resulted in significantly less graft failures leading to revision ACL surgery than the single-bundle ACL during the 5-year follow-up. Also, the double-bundle technique tended to result in better anterior stability and less degenerative changes to the knee than the single-bundle techniques, although this was not a statistical significant finding.

P19-493 Comparison of femoral graft bending angle and tunnel length between transtibial technique and transportal technique in anterior cruciate ligament reconstruction: an in vivo Imaging analysis using three dimensional-computed tomography (3D-CT) J.H. Wang1, J.G. Kim2, H.C. Lim3, J.H. Ahn4, H.J. Kim3, J.H. Bae5 1 Sungkyunkwan University School of Medicine, Samsung Medical, Seoul, Republic of Korea 2College of Medicine, Korea University, Guro Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea 3Korea University College of Medicine, Guro Hospital, Orthopedic surgery, Seoul, Republic of Korea 4Samsung Medical Center, Sungkyunkwan University, Orthopaedic Surgery, Seoul, Republic of Korea, 5Korea University Ansan Hospital, Orthopedic surgery, Ansan, Republic of Korea Objectives: To investigate which technique would reduce bending stress at the femoral tunnel aperture and make short tunnel length after ACL reconstruction by comparing the femoral graft bending angle and tunnel length between the single bundle (SB) transtibial (TT) and double bundle (DB) transportal (TP) technique. Methods: 53 consecutive patients underwent an ACL reconstruction using an auto-hamstring tendon graft. Four patients who underwent a revision ACL reconstruction and had multiple ligament injuries were excluded. Forty-nine patients underwent an ACL reconstruction using a SB TT (Group I, 20 patients) and DB TP (Group II, 29 patients) technique. Twenty patients with a less than 6 months interval from the injury time to the operation time were classified as group I and underwent an ACL reconstruction using the remnant preservation single bundle (SB) with the TT femoral tunnel technique. Twentynine patients with over 6 month interval from the injury time to the operation time were classified as group II and underwent an ACL reconstruction using DB with transportal (TP) technique without remnant preservation. All procedures were performed by a single surgeon. Femoral graft bending angle and tunnel length were measured by CT image using OsiriX imaging software (Fig. 1). Groups I and II were compared and statistical analysis was performed using SPSS software. Results: The mean anteromedial (AM) and posterolateral (PL) femoral graft bending angle of group II [111.5 ± 8.8 (range, 130.7–93.6) and 118.9 ± 9.8(range, 135.2–91.7), respectively] was significantly more acute than that of group I [125.3 ± 11.1 (range, 106.9–143.1)] (P \ 0.001 for AM, P = 0.04 for PL). The mean difference of femoral graft bending angle between the TT and TP technique was 13.8 for AM and 6.4 for PL. The mean femoral tunnel length of group I was 41.6 ± 8.9 mm (range, 33.7–60.5 mm), and the mean femoral AM and PL tunnel lengths of group II were 33.9 ± 3.2 mm (range, 26.4–40.0 mm) and 34.2 ± 3.4 mm (range, 28.7–42.4 mm), respectively. The mean femoral tunnel length of group I was significantly longer than that of group II (P = 0.001). The mean difference of femoral tunnel length between TT and TP technique was 7.7 mm for AM and 7.4 mm for PL. The number of

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1 cases with a femoral tunnel length \30 mm of the AM and PL tunnel in group II was 3 and 3, respectively. Conclusions: The femoral graft bending angle and the femoral tunnel length of the TP technique performed in the maximally flexed knee position was more acute and shorter than those of the TT technique after ACL reconstruction. This might increase the bending stress at the femoral tunnel aperture and shorter graft length in the tunnel after an ACL reconstruction using TP technique compared to the TT technique.

P19-511 Development of a virtual three-dimensional kinematic model of the human knee joint G. Rochcongar1, L. Geais2, J.-F. Potel3, P. Beaufils4, C. Hulet1, R. Seil5 1 Caen University Hospital, Orthopaedic Department, Caen, France, 2 Laboratory of Mechanics, Conservatoire National des Arts et, Paris, France, 3Medipole Garonne, Toulouse, France, 4Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France, 5Centre Hospitalier Luxemburg, Clinique d’Eich, Olympic Medical Center, Orthopedic Department, Luxemburg, Luxembourg Objectives: Computerized kinematic analysis of the human knee joint is currently limited to finite element models which do not allow simulating fast and wide range of knee motion patterns. The objective was to develop a 3-dimensional model of the human knee joint during a passive flexion–extension and an internal–external rotation motion between 0 and 130 of flexion by using parametric rigid body modeling. Methods: Modelization was performed by using 3D software (Solid Works, Dassault Systems SolidWorks Corp., Concord, MA, USA). The bony structures were modeled according to CT scan data of an adult healthy human volunteer. Ligament and tendon insertions were placed according to 9 cadaveric knee dissections. One to 3 bundles were used to simulate each ligamentous structure. The biomechanical properties of the ligaments and tendons were simulated with a mathematical algorithm. Three muscles (quadriceps, popliteus and fascia lata) were modeled into passive structures. Once the computerized model was established an iterative experimentation method with more than 1,000 repetitions was chosen by applying standardized variations to ligament strain to adapt it to the natural kinematics. To compare those computer results, kinematics were recorded with a navigation system—eNact Knee Software (Stryker, Howmedica, Ruthefort, USA) on 9 cadaver knee specimens. A passive motion range was applied between 0 and 130 of flexion in neutral rotation. Results: A complete physiologic passive range of motion of the knee could be reproduced between 0 and 130 of flexion with a typical motion pattern (at 130 of flexion: 16 of internal tibial rotation,

S163 posterior motion of the medial femoral condyle of 11 mm and of 24 mm of the lateral femoral condyle) and a physiologic behavior of the ligaments (shortening of the distance between the insertions of the posterolateral bundle of the ACL by 5 mm; lengthening of the anterolateral bundle of the PCL by 4 mm). Furthermore the model allowed simulating the relative influence of each ligament at each degree of flexion/extension and internal/external rotation on knee motion. Conclusions: Current computerized modeling techniques allow for an adequate simulation of passive human knee motion. The developed model allowed for a simulation of a passive motion pattern between 0 and 130 of knee flexion which was close to normal kinematics. Furthermore the relative influence of each ligament on normal knee motion could be recorded. The model is promising for further kinematic studies under physiologic and pathologic conditions and an individualized treatment approach of each knee.

P19-524 Clinical outcome after oval revision ACL reconstruction surgery using autologous hamstring tendon grafts P. Scha¨ferhoff1, P. Klein1, B.-G. Bo¨ttenberg1, M. Sa¨ugling1, H. Dewitz1 1 MediaPark Clinic, Department of Orthopaedics and Sports Medicine, Cologne, Germany Objectives: Anatomical oval ACL reconstruction showed accurate recreation of the femoral and tibial footprint, good clinical outcomes and satisfactory anteroposterior and rotational stabilities. But what about the results after revision surgery? The aim of this study was to evaluate short- and middle-term clinical results of patients after oval revision ACL reconstruction (contralateral hamstring autograft). Methods: In a retrospective study 29 patients who sustained a rerupture of the ACL were followed up after oval revision ACL reconstruction. All patients described at baseline a subjective instability. The mean follow up were 18.2 months. Clinical results, such as, International Knee Documentation Committee (IKDC), Lysholm knee and Tegner scores, and manual knee laxity and instrumented anterior laxity test findings (KT-1000) were evaluated. Results: At the time of the FU the average of the Lysholm knee score was 92.1 and the IKDC Subjective Evaluation was 82.2. Postoperative 84.7% of the patients were graded as Group A or B using the objective IKDC. Activity levels, according to the Tegner activity scale, improved from 1.9 preoperatively to 5.8 at the time of the FU. The Lachman test, at last follow-up, showed normal laxity in 25 patients and grade I laxity in 4. According to the pivot-shift test, 25 patients had normal laxity and 4 grade I laxity. By instrumented anterior laxity testing (KT-1000), mean side-to-side difference was 4.4 mm at last follow-up. Conclusions: ACL revision surgeries performed with the oval technique allow an accurate recreation of the femoral and tibial footprint and offer good clinical outcomes, satisfactory anteroposterior and good rotational stabilities. All patients had documented healing and/or full/painless knee motion and had full return to ADL.

P19-527 Comparison of tunnel widening after single-bundle and anatomic double-bundle ACL reconstruction A. Achtnich1, P. Forkel1, S. Metzlaff1, E. Lopez Ha¨nninen2, W. Petersen1 1 Martin-Luther-Krankenhaus, Orthopaedic and Trauma Surgery, Berlin, Germany, 2Martin-Luther-Krankenhaus, Radiology, Berlin, Germany Objectives: The purpose of this prospective study was to compare postoperative bone tunnel enlargement after single-bundle versus

123

S164 double-bundle anterior cruciate ligament (ACL) reconstruction with semitendinosus graft. Methods: Twenty-one consecutive patients underwent anatomic arthroscopic ACL reconstruction using double-bundle (db) technique and twenty-four patients underwent single-bundle technique (sb). Magnetic resonance imaging scans were performed on the second postoperative day and at mean follow-up of 8 months. Clinical evaluation was performed using IKDC form, the Lysholm knee score, KT-1000 arthrometer (side to side) and subjective satisfaction of patients. Results: There were no significant differences in the clinical measures between single- and double-bundle reconstructions. The postoperative anterior laxity measured with the KT-1000 arthrometer was not significant different. Using double-bundle technique bone tunnel communication intraoperatively was observed in two patients on the tibia and in no case on the femur. At 8 month postoperatively there was no significant difference in tunnel enlargement between single and double bundle technique. In each group, there were no significant relationships between tunnel enlargement and clinical outcome. Conclusions: We conclude that there is no difference between bone tunnel widening of anatomic anterior cruciate reconstruction using single-bundle or double-bundle technique.

P19-541 Clinical and functional outcomes after an accelerated rehabilitation program following single-bundle ACL reconstruction with hamstring autograft M.A. Hernan Prado1, M.F. Diez Ramos2, O.S. Perez Moro2, R. Llopis Miro´1 1 Hospital Universitario Santa Cristina, Unidad Aparato Locomotor, Traumatologı´a, Madrid, Spain, 2Hospital Universitario Santa Cristina, Unidad Aparato Locomotor, Rehabilitacio´n, Madrid, Spain Objectives: The aim of this study was to evaluate the clinical and functional outcomes in patients who received an accelerated rehabilitation program after single-bundle ACL reconstruction with a quadruple-stand hamstring autograft. Methods: We performed a retrospective review of 250 patients underwent LCA reconstruction with autogenous hamstring tendons. They didn’t use brace and immediately after surgery patients had progressive increase in range of motion and full assisted weightbearing. Patients were evaluated clinically and radiographically at 3, 6 and 12 months postop. Activity levels and subjective functional results were evaluated with the Tegner activity scale and Lysholm knee score, respectively. Knee stability was assessed with Lachman test, anterior drawer test and pivot shift test, and knee range of motion was measured. Data collection also include the incidence of associated injuries, the length of rehabilitation program and complications. Results: The average time of rehabilitation under supervision was 25 days. After 12 months, the mean increase of anterior laxity from immediately after surgery was not clinically significantly. Clinical assessments found gradual improvements in flexion–extension and activity levels (Lysholm preop: 59.4/Lysholm 12 months: 95.5). Conclusions: The accelerated rehabilitation program after singlebundle ACL promotes functional recovery with an excellent clinical outcome.

P19-562 Does pressfit technique reduce femoral tunnel enlargement in ACL reconstruction using autologous hamstring tendons? A prospective randomized CT study K. Nha1, J.I. Kim2, J.H. Kwon2, H.D. Lee1, B.H. Jang2

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 1

Inje University, Ilsanpaik Hospital, Koyangsi, Republic of Korea, Inje University, Ilsanpaik Hospital, Koyang, Republic of Korea Objectives: To compare the incidence of tunnel widening (TW) in patients undergoing anterior cruciate ligament (ACL) reconstruction with autohamstring graft either hard to pass the graft (group A: pressfit) or easy to pass the graft (group B: control) in the femur tunnel using Rigid fix implant (Depuy, Mitek) and intrafix bioscrew and spike washer screw distal to the tibia tunnel. Methods: 85 patients were included and randomized and 69 patients were assessed. The evaluation included CT scan, KT-2000, International Knee Documentation Committee (IKDC) ratings and Lysholm score. The diameter of the CT of the femoral tunnel at the minimum 1 year follow-up was compared with the tunnel diameter of the CT from 1 week postoperatively. A more than 2 mm enlargement was considered tunnel widening. Results: Pressfit group was associated with lesser femoral tunnel widening, but this was not found to be statistically significant. In group A in which pressfit graft was used, 21(60%) of 35 patients had developed femoral TW. In group B, 24(71%) of 34 patients had TW in the femur. There was no statistical difference in two groups and TW near the joint is significantly less TW compare to the middle and end part in group A. No significant difference was found with respect to Lysholm score, IKDC, and KT 2000 evaluation. Conclusions: There was no significant reduction to TW in group A compared the group B. But there was a little differences of shape in two groups.

2

P19-571 Bundle specific injury patterns of anterior cruciate ligament: value of MR imaging J.-Y. Choi1, C.B. Chang2, M.J. Chang2, T.K. Kim2 1 Seoul National University College of Medicine, Radiology, Seoul, Republic of Korea, 2Seoul National University Bundang Hospital, Orthopaedic Surgery, Seongnam-si, Republic of Korea Objectives: Coupled with double bundle concept of anterior cruciate ligament (ACL), selective bundle ACL reconstruction has received attention for the patients with preserved anteromedial (AM) or posterolateral (PL) bundle. However, reported prevalence of intact bundle in the patients underwent ACL reconstruction widely varied. In addition, the value of MRI for prediction of bundle specific injury patterns has not been well established. This study aimed to document the bundle specific injury patterns of ACL in the patients with ACL injury undergoing reconstruction, and to examine the efficacy of MRI for evaluation of the bundle specific injury patterns of ACL. Methods: We evaluated 212 knees in 212 patients underwent primary ACL reconstruction. During arthroscopy, condition of each bundle was assessed by gross observation via AM and AL portal, and by

Fig. 1 Arthroscopic findings of isolate PL bundle injury

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 2 Oblique coronal MR images of isolated PL bundle injury evaluation of bundle tension in various range of knee motion and rotation of the tibia (Fig. 1). Then, we classified patterns of AM and PL bundle injuries into 3 categories including complete injury, continuous but nonfunctional (stretched out and no functional tension), and intact. On MR images including oblique coronal images (Fig. 2), patterns of AM and PL bundle injuries were evaluated and classified into 3 categories including complete injury, partial injury (partial disruption or attenuated), and intact. The agreements between the MR and the arthroscopic assessments on the injury patterns of each bundle were examined by the kappa statistics and cross-table analysis. Results: In the study cohort 21 knees (10%) had an intact AM or PL bundle, and among them, 5 knees had an intact AM bundle while 16 knees had an intact PL bundle. In 68 knees (32%), a continuous AM or PL bundle was found, but the bundle was stretched out and nonfunctional. Complete injuries of both the AM and PL bundles was seen in the remaining 58%. The overall agreement between the MR and the arthroscopic assessments was satisfactory, but agreement for PL bundle injury was poorer than that for AM bundle injury (kappa coefficient = 0.610 for AM bundle, 0.522 for PL bundle). The accuracies of MR prediction for arthroscopic findings were 93% for AM bundle and 80% for PL bundle. Compared to the arthroscopic findings, PL bundle injury was more likely to be overestimated on MRI. Conclusions: In the study cohort, 10% of patients with ACL injury warranting reconstruction had an intact bundle, mostly intact PL bundle. Our results found that MRI would be a valuable tool for preoperative assessment of the bundle specific injury patterns of ACL. In particular, AM bundle injury can be estimated on MRI with high accuracy.

P19-602 Changes in bone tunnels in flexion MR imaging during the early postoperative period after anterior cruciate ligament reconstruction J. Nakase1, K. Kitaoka2, T. Toratani1, M. Kosaka1, Y. Ohashi1, H. Tsuchiya1 1 Kanazawa University, Orthopaedic Surgery, Kanazawa, Japan, 2 Kijima Hospital, Orthopaedic Surgery, Kanazawa, Japan Objectives: A tendon graft must be securely fixed in a bone tunnel to obtain satisfactory knee stability after anterior cruciate ligament (ACL) reconstruction. Fibrous interzone (FIZ) forms between the tendon graft and the bone tunnel wall after ACL reconstruction. This zone plays an important role in tendon graft fixation to the bone tunnel. Only a few reports have examined changes in bone tunnels observed over time on MRI during the early postoperative period after ACL reconstruction. In this study, we performed flexion MR imaging at 1, 3, 6, 9, 12 and 24 weeks after ACL reconstruction and examined changes in the FIZ using two-dimensional sagittal and coronal images.

S165 Methods: The subjects were 10 patients who had undergone ACL reconstruction between June and August 2010. There were 5 men and 5 women and their mean age was 29.4 years. ACL reconstruction was performed by the method of Howell. Double-folded semitendinosus tendon and gracilis tendon were used. Fixation was achieved by cross pins on the femoral side and by a washer plate and screw on the tibial side. In addition, bone grafting was performed using a special device in the proximal femoral bone tunnel and in the anteromedial tibial bone tunnel. A 0.4 T APERTO MRI system (HITACHI Medical) was used. T2-weighted FSE MRI was performed in the sagittal and coronal planes at 1, 3, 6, 9, 12 and 24 weeks postoperatively. Slice thickness was 2.5 mm, and imaging was performed so that the knee flexion angle was 60 and the tendon graft was straight. The maturity of the FIZ was semiquantitatively evaluated by comparing signal intensities of the FIZ and other tissues in the same image. If the signal intensity of the FIZ was the same as that of the patellar tendon or iliotibial band, then a score of 3 was given. If it was the same as that of muscle tissue, then a score of 2 was given. If it was between that of muscle tissue and synovial fluid, then a score of 1 was given. If it was the same as that of synovial fluid, then a score of 0 was given. Results: In the femur, the tendon-bone junction matured gradually up to 9 weeks postoperatively, and the lateral and medial aspects began to mature by 12 weeks postoperatively. In the tibia, all regions matured over time and finished until 12 weeks postoperatively. The lateral aspect matured the fastest among the regions, followed by the posterior, medial, and anterior aspects. The FIZ of the tibia matured earlier than that of the femur. Conclusions: Compressive force was thought to be important in tendon graft fixation to the bone tunnel. The FIZ of the tibia matured faster than that of the femur, and the reasons were thought to be that a larger amount of ACL remnant remained on the tibia and the movement of tendon graft was less in the tibia than in the femur. This study is the first one to evaluate in detail the tendon-bone junction in the early postoperative period after ACL reconstruction using flexion MRI. Thus, this report is considered to have great clinical significance.

P19-715 Bone density comparison at the femoral attachment in the acute and chronic anterior cruciate ligament injured knee S. Yamasaki1, Y. Hashimoto1, J. Takigami2, H. Nakamura1 1 Osaka City University Graduate School of Medicine, Orthopaedic Surgery, Osaka, Japan, 2Osaka City University Graduate School of Medicine, Orthopaedics, Osaka, Japan Objectives: To investigate bone density at the femoral attachment of anterior cruciate ligament (ACL) post-injury and compare to normal healthy knees. There are a few published reports that the overall bone mineral density in ACL injured knees decrease compared to normal knees, but there are no published reports on post-injury bone quality of the local femoral attachment of ACL. Methods: We examined 10 patients who received unilateral ACL reconstruction at our department. We divided into two groups, acute group and chronic group. Acute group had 5 patients (4 males, 1 female. Mean age: 21.6) who received reconstructive surgery within 3 months from the injury (a median of 66.4 days after injury). The chronic group also had 5 patients (4 males, 1 female. Mean age: 22.0) who received the same reconstructive surgery more than 1 year after their injury (a median of 764 days).All patients received Computed Tomography (CT) of their bilateral knees the day before surgery. Under the reconstructed images using three-dimensional (3D) software, we defined 6 zones (AM bundle-anterior: AA, AM bundle on the ridge: AR, AM bundle-posterior: AP, PL bundleanterior: PA, PL bundle on the ridge: PR, PL bundle-posterior: PP) around the residents’ ridge on lateral view of 3D-CT image in the

123

S166 knee. We measured CT density scale (Hounsfield Unit, HU) on the bony surface of 10 points in each zone of the injured and normal knees. We also measured 10 points at the bony surface of the patella as a control. Results: In the acute group, the HUs of around the attachment in injured/normal knees were 610/730, 898/983, 682/738, 405/428, 525/565, 471/505 at AA, AR, AP, PA, PR and PP, respectively. In the chronic group, the HUs values of that were 441/579, 649/1,014, 556/916, 377/447, 466/725, 402/635 at AA, AR, AP, PA, PR and PP, respectively. In the normal knee, the HUs on the ridge was higher than other zones in both groups. The HUs of patella as a control in injured/normal knees were 1,238/1,337 in the acute group, 1,249/ 1,333 in the chronic group. In both groups the HUs at the patella decreased by 10% post-injury. In the acute group, the HUs in injured knee at all zones was lower than normal knee, and the degree of decrease was the same as the control. While in the chronic group, the HUs in injured knee decreased significantly lower than normal knee at AR, AP, PR, PP, and the degree of decrease at that site was significantly much more than the control. Conclusions: This is the first report that bone density of the femoral attachment of ACL is higher than any other area in living knees. In the chronic group the HUs on the ridge decreased much more due to several factors that are shared with the acute group including limited weight bearing and prolonged disuse of the injured knee. Furthermore, we would like to emphasize that this phenomenon may be explained by the relationship of decreased tensile force of the ligament at the attachment site and the prolonged period of time.

P19-782 Defining chronicity in an ACL deficient knee: when is a knee with an acutely torn ACL no longer ‘‘acute’’? C.C. Kaeding1, A. Pedroza2, D. Flanigan1 1 The Ohio State University College of Medicine, Department of Orthopaedic Surgery, OSU Sports Medicine, Columbus, Ohio, United States, 2The Ohio State University College of Medicine, Sports Medicine, Columbus, Ohio, United States Objectives: A chronic ACL deficient knee is more likely to have meniscus and chondral injuries than the acute ACL injured knee. The purpose of this study is to define in the ACL injured knee the point in time in which the associated intra-articular injury status statistically differs from that of an acutely injured knee. Methods: In a prospectively collected ACL database, the time interval from ACL injury to ACL reconstruction was collected for each of 311 patients. Intra-operative findings were then recorded, including the presence of medial or lateral meniscal tears, as well as chondral damage in the medial, lateral, or patellofemoral compartments. Criteria for having chondral damage were: any grade 2 lesion involving 50% or more of the condylar width or any grade 3 or 4 lesion. Data was statistically analyzed to detect differences in meniscal tears or articular cartilage damage with respect to time using a Wilcoxon rank sum test. Those incidences where statistical significance was found were examined to determine the criteria for acute/chronic classification. Results: The prevalence of medial meniscal injury increased with increasing time from ACL injury to surgery. The prevalence of medial meniscal injury became significantly higher in chronic injuries when the cutoff point for chronic ACL injuries reached 8 weeks. At this point the prevalence of a medial meniscus tear was 25% in the acute group and 34% in the chronic group (p = 0.026). The prevalence of lateral meniscal injury did not increase with increasing time ACL injury to surgery. Lateral meniscus status was therefore not evaluated in regard to a cutoff point for a chromic injury. The prevalence of chondral injury in the medial, lateral, and patellofemoral compartments increased with increasing time from ACL

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 injury to surgery. The prevalence of medial compartment chondral injury became significantly higher in chronic injuries when the cutoff point for chronic ACL injuries reached 4 weeks. At this point the prevalence of medial compartment chondral damage was 14% in the acute group and 26% in the chronic group (p = 0.017). The prevalence of lateral compartment chondral damage became significantly higher when the cutoff point for chronic injuries reached 6 weeks. At this point the prevalence of lateral compartment chondral damage was 19% in the acute group and 28% in the chronic group (p = 0.05). The prevalence of patellofemoral compartment chondral damage became significantly higher when the cutoff point for chronic ACL injuries reached 8 weeks. At this point the prevalence of patellofemoral compartment chondral damage was 16% in the acute group and 23% in the chronic group (p = 0.045). Conclusions: ACL deficient knees that had reconstruction performed 8 or more weeks after injury had statistically greater incidence of medial meniscus tears and chondral damage in all 3 compartments. ACL injuries undergoing reconstruction 8 or more weeks after index injury are no longer ‘‘acute’’.

P19-842 Bone tunnel enlargement following anterior cruciate ligament reconstruction with hamstrings tendons. A comparative study of three different techniques I. Apostolopoulos1, I. Tsolos1, A. Andreakos1, S. Tsiplakou2, I. Staikidou3, G. Giannikouris3 1 Sismanoglion Hospital, Orthopaedic Department, Athens, Greece, 2 KAT Hospital, Athens, Greece, 3Sismanoglion Hospital, MRI Department, Athens, Greece Objectives: The purpose of the present study was to compare the bone tunnel changes using hamstring tendon grafts with different fixation methods. Methods: Since 2006, 83 patients underwent single band ACL reconstruction with: (1) femoral extracortical or non anatomical device (RetroButton, Arthrex) Group A, (2) Femoral transfixation pin fixation device (BioTransfix, Arthrex) Group B, and (3) Aperture fixation (AperFix, Cayenne medical) Group C. Tibial fixation with absorbable screws and post fixation with a staple was used in the first two techniques and a non absorbable screw in the last one. The average age of the patients was 26 years (19–45) and the mean time interval from injury to reconstruction was 9 months (3/52–29/52). The evaluation methods were clinical examination, knee scores and instrumented laxity measurements at 6, 12 and 24 months. We also performed MRI at 6 months, 1 and 2 years postoperatively. Results: Eleven patients were completely lost to follow up, leaving 72 patients for analysis. The mean increase of the femoral tunnel widening on the coronal plane in MRI in 6/12/24 months was for group A 29.8%/36.5%/35.9 and for the sagittal plane 27.3%/31.5%/ 31.2% respectively. In group B it was 19.2%/24.6%/25.1% in the coronal plane and 21.1%/24.8%/24.6% in the sagittal plane respectively and in group C it was 17.2%/20.6%/20.5% in the coronal plane and 17.8%/19.5%/19.7% for the sagittal plane respectively. For the tibial side the tunnel widening was in group A, 25%/28.2%/28% for the coronal plane and 24%/27.8%/28.3% for the sagittal view. In group B, 22.3%/24.9%/23.6% and 21.6%/ 22.9%/22.1% for the sagittal view respectively and finally for the groupC 24.2%/25.9%/25.6% for the coronal view and 27.8%/29.8%/ 29.1% for the sagittal view respectively. Tunnel widening did not correlate with increased laxity, poor IKDC, Lysholm and Tegner knee scores. Conclusions: Tunnel widening occurs with both elastic and rigid fixation methods. Graft tunnel motion is not the sole cause of tunnel widening in ACL reconstruction. Tunnel widening does not correlate with poor outcome in the short term.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P19-881 Differences between awake and anesthetized patients in knee kinematics during the pivot-shift test for anterior cruciate ligament deficiency S. Oka1, R. Kuroda1, T. Matsushita1, S. Kubo1, K. Nagamune2, M. Kurosaka1 1 Kobe University Graduate School of Medicine, Orthopaedic Surgery, Kobe, Japan, 2Fukui University Graduate School of Engineering, Human and Artificial Intelligent System, Fukui, Japan Objectives: The pivot-shift test is commonly performed in a clinical setting and recognized as one of the most reliable clinical test in diagnosis for the anterior cruciate ligament (ACL) deficiency. In a previous study, we quantified the knee kinematics during the pivotshift test using an electromagnetic measurement system (EMS). However the study was conducted under anesthesia and it still remains to be tested whether the pivot-shift test can be performed accurately when the subjects were awake for daily practice. In this study, we assessed the acceleration of reduction during the pivot-shift test using the EMS and compared the differences between the under awake and anesthesia conditions. Methods: Fifty patients with unilateral ACL ruptures were assigned for this study (mean age: 26.6 years). The acceleration of reduction during the pivot-shift test was measured by the EMS on the day before surgery in the outpatient clinic room when the patients were awake and on the operative day when the patients were under general anesthesia before surgery in the operating room. The tests were performed by a single experienced surgeon and the clinical grading of the pivot-shift was also evaluated by the examiner. The statistical difference was evaluated was by one-way analysis of variance. Results: The clinical grades of the pivot shift for the ACL-deficient knees were evaluated as none (-) in 15 knees, glide (+) in 23 knees, clunk (++) in 12 knees when the patients were awake, and none (-) in 3 knees, glide (+) in 28 knees, clunk (++) in 16 knees, and gross (+++) in 3 knees when the patients were under anesthesia. All the contralateral intact knees were evaluated as none (-). The mean acceleration of reduction were 0.8 ± 0.3 m/s2 in the intact knees and 1.1 ± 0.4 m/s2 in the ACL-deficient knees when the patients were awake, and 0.7 ± 0.2 m/s2 in the intact knees and 1.7 ± 1.0 m/s2 in the ACL-deficient knees when the patents were under anesthesia. The mean acceleration in the ACL-deficient knees was significantly larger than that in the intact knees when the patients were awake (p \ 0.01) and under anesthesia (p \ 0.001). In the ACL deficient knees, the acceleration was significantly larger when the patients were under anesthesia than awake (p \ 0.001). Conclusions: Although the mean acceleration of reduction in the ACL-deficient knee was significantly larger than in the intact knees, the clinical grades and the acceleration of reduction tended to be smaller when the patients were awake compared with the condition in which patients were under anesthesia. Our observations suggested that the knee kinematics during the pivot-shift test in the ACL-deficient knees tends to be affected by the patients’ consciousness and it needs a caution if quantifying the pivot-shift movement when the patients are awake.

P19-917 Degenerative changes after ACL reconstruction in relation to tunnel position R. van Dijck1, D. Saris2, J.W. van Ommeren3, T. Enneking4, A. Fievez5 1 Medinova Clinic, Rotterdam, The Netherlands, 2UMC Utrecht, Orthopaedics, Utrecht, The Netherlands, 3Kliniek Zestienhoven Rotterdam, Rotterdam, The Netherlands, 4Franciscus Hospital Roosendaal, Orthopedics, Roosendaal, The Netherlands, 5Medinova Clinic, Orthopedic Surgery, Zestienhoven, The Netherlands

S167 Objectives: The results of ACL reconstruction with bone-patella-bone and semitendinosus/gracilis autograft have been well documented in literature. However the relation of the ACL reconstruction and the occurence of degenerative changes remains controversial. There is less known about the development of degenerative changes in relation to tunnel position. In this study we evaluate the degenerative changes after ACL reconstruction in relation to the tunnel position. Methods: Between 1988 and 1998, 305 patients underwent an ACL reconstruction with a bone-patella-bone autograft (n = 210) or a semitendinosis/gracilis autograft (n = 95). Of these patients, 276 patients were available for evaluation with a mean follow-up time of 8.4 years. Evaluation included: detailed history, physical examination, functional knee ligament testing, KT-1000 arthrometer testing, one-leg hop testing, Lysholm score, Tegner score and International Knee Documentation Committee standard evaluation form. X-rays were used for scoring degenerative changes according the Ahlback/ Fairbank’s classification. Position of the graft was measured radiographically using the Amis circle and Taylor score. Results: At the final follow-up: 108 patients (39%) had an IKDC score grade A, 121 (44%) an IKDC score grade B, 33 patients (12%) grade C and 14 patients (5%) grade D. According to the Ahlback and Fairbank’s classification 65 patients (23%) developed degenerative changes between ACL reconstruction and final follow-up. Between ACL injury and ACL reconstruction 24 patients (9%) developed degenerative changes. A longer delay between ACL injury and ACL reconstruction in patients with gross instability, leads to more secondary damage (cartilage damage and meniscal lesions) and these patients had significantly more degenerative changes (p \ 0.05). Patients with a worse Amis score (\60%) had significantly (p \ 0.05) more often a positive pivot shift after ACL reconstruction. Patients with a positive pivot shift after ACL injury and/or after ACL reconstruction had significantly (p \ 0.05) more degenerative changes in comparison with patients without a positive pivot shift. Of the 20 patients with a Taylor II score, 9 patients (45%) developed degenerative changes between ACL reconstruction and final followup. Also patients with a higher Tegner score over years (p \ 0.01) and patients with a meniscectomy during ACL reconstruction (p \ 0.05) developed significantly more osteoarthritis. Conclusions: After ACL reconstruction with a semitendinosus/gracilis autograft or a bone-patella-bone autograft 23% of the patients developed degenerative changes. Patients with a non-optimal tunnel position had more often a positive pivot shift and showed more frequently degenerative changes.

P19-984 Comparisons of static and dynamic tibial translation between children and adults and between sexes S. Tagesson1, E. Witvrouw2, N.G. Hesar2, J. Kvist1 1 Department of Medical and Health Sciences, Division of Physiotherapy, Linko¨ping, Sweden, 2Rehabilitation Sciences and Physiotherapy, Ghent University, Ghent, Belgium Objectives: To compare the static and dynamic tibial translation between children and adults. To compare the static and dynamic tibial translation between women and men and between girls and boys. Methods: Sixty-seven skeletally immature children (age 8–13 years) and 63 adults (age 18–30 years) without previous knee injuries participated. Sagittal tibial translation was measured with the CA-4000 electrogoniometer (OSI Inc., Hayward CA, USA). Static translation was measured during the instrumented Lachman test with the subject strapped to a special seat with the knee flexed to 20–30. Using a force handle the proximal tibia was pushed posterior and pulled anterior with a controlled force. The total static tibial translation

123

S168

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Table 1 Tibial translation (mm) (mean ± SD) Test

Children

Adults

P value

Girls

Boys

P value

Women

Men

P value

La

9.1 ± 2.9

7.3 ± 2.7

\ 0.001

8.6 ± 3.3

9.6 ± 2.4

0.178

7.7 ± 2.3

6.8 ± 3.1

0.179

(n = 67)

(n = 63)

(n = 33)

(n = 34)

(n = 33)

(n = 30)

10.1 ± 2.8

7.9 ± 3.0

9.0 ± 2.6

11.3 ± 2.7

8.8 ± 3.3

6.8 ± 2.3

(n = 22)

(n = 30)

(n = 11)

(n = 11)

(n = 17)

(n = 13)

5.9 ± 2.8

6.8 ± 3.0

5.9 ± 2.7

6.0 ± 2.8

7.8 ± 2.7

5.7 ± 3.0

(n = 63)

(n = 61)

(n = 32)

(n = 31)

(n = 33)

(n = 28)

90 N La 134 N Gait

0.010

0.089

0.055

0.848

0.074

0.005

(TSTT) were assessed at 90 N and 134 N forces. Dynamic translation was measured as the maximal anterior tibial translation (MATT) during gait. An independent samples t test was used to compare the tibial translation between groups. Pearson correlation coefficient was used for calculation of correlation between static and dynamic knee stability. Results: The TSTT was greater in the children compared to the adults. The MATT during gait did not differ between the children and the adults (Table 1). In the children there was no correlation between the TSTT and the MATT (r = 0.231, p = 0.069). In the adults there was a weak correlation between the TSTT and the MATT (r = 0.348, p = 0.006). In the child group there were no differences between the girls and the boys in TSTT or MATT during gait (Table 1). In the adult group no difference in TSTT between the women and the men could be detected. The women had greater MATT during gait compared to the men (Table 1). Conclusions: The static tibial translation was greater in the children compared to the adults, whereas the dynamic tibial translation did not differ. There was no difference in static tibial translation between the women and the men. The women had greater dynamic tibial translation compared to the men. A hypothesis raised from the present study is that one factor for the decreased risk for ACL-injury in children compared to adults and in men compared to women, may be the ability to utilize less amount of the possible joint play during activity (i.e. gait) in non-injured knees, as found for the children and adult male.

P19-1004 Comparative study of double bundle techniques used in ACL reconstruction A. Maestro1, J. Fernandez Lombardia2, A. Guerrero3, J. Jorge3, L. Rodriguez4 1 FREMAP, Sports Medicine, Orthopaedic Surgery Department, Gijon, Spain, 2Hospital Arriondas, Arriondas, Spain, 3ITMA, Llanera, Spain, 4Hospityal Cabuenes, Gijon, Spain Objectives: To demonstrate that the mechanical strength of the bone allows intensive rehabilitation of the knee, independently to use one or two femoral tunnels. Methods: The study by finite element method the stresses to which is subject to the distal femur in the following cases and fixing the following systems: two tibial and two femoral tunnels (Endobutton CL for AM bundle and Endobutton direct for PL), only one tibial tunnel and two femoral (same fixation system that previous case in a course, and PL Biosure screw for PL in the other course). In all cases, the tibial fixation was performed with one interference screw. The study was carried out with the minimum load required for an intensive rehabilitation (500 N). Results: In the case of two tibial tunnel, the fixation systems create a force on the cortical bone of 92 MPa for the AM tunnel and 71 MPa for the PL. With one tibial tunnel and two femoral tunnels, the

123

Endobutton CL (AM) yields 82 MPa on AM femoral cortex and 48 MPa Endobutton Direct on the PL. In the latter case, the Endobutton generates a force of 120 MPa in the cortex, and the screw on the PL tunnel generates 9 MPa. Conclusions: The results show that the studied techniques involve no substantial differences from the mechanical point of view. The distribution of forces in the femoral tunnel is different, because the screw to fix the mechanical responsibility lies on an area much larger but less hard, and in all cases the stresses are kept below the elastic limit of the oseus material (120 MPa, 9 MPa) and therefore would be able to keep intensive rehabilitation (500 N).

P19-1008 Comparative study of femoral fragility after SB or DB techniques used in ACL reconstruction A. Maestro1, J. Fernandez Lombardia2, A. Guerrero3, J. Jorge3, L. Rodriguez4 1 FREMAP, Sports Medicine, Orthopaedic Surgery Department, Gijon, Spain, 2Hospital Arriondas, Arriondas, Spain, 3ITMA, Llanera, Spain, 4Hospital Cabuen˜es, Orthopaedic Surgery, Gijon, Spain Objectives: To know the possibility of femoral fracture after DB technique. Methods: The study was made by finite element method the stresses to which is subject to the distal femur in the following cases and fixing the following systems: only one femoral tunnel (SB) fixed by mean of Endobuttom CL system and two femoral tunnels (Endobutton CL for AM bundle and Endobutton direct for PL), The study was carried out with the minimum load required for an intensive rehabilitation (500 N). Results: The ma´ximum load peak showed 120 MPa on the femoral SB tunnel and 109 MPa and 53 MPa respectively on the AM and PL tunnels for DB. These results show that the use of DB techniques in ACL reconstruction, causes an increase in femoral fragility, but it’s below the yield strength of cortical bone material (120 MPa) and therefore no risk of fracture. Conclusions: The choice of DB technique, will be determined by the choice of surgeon or patients, but there is no risk of femoral fracture.

P19-1019 Clinical outcome measures in anterior cruciate ligament surgery: clinician versus patient completed knee scores O. Al-Dadah1, L. Shepstone2, S.T. Donell1 1 Norfolk and Norwich University Hospital, Trauma and Orthopaedic Surgery, Norwich, United Kingdom, 2University of East Anglia, Medical Statistics, Norwich, United Kingdom Objectives: Clinical outcome measures are important in both the conduct of clinical research and in the evaluation of knee surgery in every day clinical practice. They can be categorised as being generic or disease-specific and clinician-completed or patient-completed. A wide variety of validated outcome scores are available in the literature. The objective of this study was to investigate if there is a difference between clinician-completed and patient-completed outcome scores in detecting improvement following ACL reconstruction. Methods: Fifty patients with an ACL rupture were prospectively evaluated using a total of nine clinical outcome measures. The five clinician-completed knee scores included Tegner Activity Score, Lysholm Knee Score, Cincinnati Knee Score, International Knee Documentation Committee (IKDC) Objective Knee Score and Tapper and Hoover Meniscal Grading Score. The Four patient-completed knee scores included IKDC Subjective Knee Score, Knee Outcome Survey—Activities of Daily Living Scale (KOS-ADLS), Short Form12 Item Health Survey (SF-12) and Knee Injury and Osteoarthritis Outcome Score (KOOS). Thirty-four of the 50 patients underwent an

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 ACL reconstruction and were reassessed with all nine outcome scores upon their follow-up review 3 months post-operatively. Results: Statistical analysis of pre-operative findings compared to post-operative results demonstrated a significant longitudinal improvement of all five clinician-completed knee scores including Tegner Activity Score from 3.3 to 4.1 (p = 0.006), Lysholm Knee Score from 71.7 to 85.3 (p \ 0.001), Cincinnati Knee Score from 62.6 to 75.9 (p \ 0.001), IKDC Objective Knee Score from mode average grade Abnormal to Nearly Normal (p = 0.001) and Tapper and Hoover Meniscal Grading Score from mode average grade Fair to Good (p \ 0.001). However none of the four patient-completed knee scores revealed a statistically significant improvement postoperatively. Conclusions: The longitudinal analyses of the clinician-completed outcome measures were found to be inconsistent with those of the patient-completed instruments. Therefore the mode of administering and collecting data either for research purposes or use in clinical practice is an important factor to consider when implementing the use of outcome measures in patients with ACL injuries as it can have a significant influence on the end results. Both formats have their advantages and limitations. As a disparity exists between the two categories, the use of both a clinician-completed and a patient-completed clinical outcome measure maybe the most prudent approach to assessing and quantifying ACL injuries and the outcome following surgery.

P19-1139 Carioca and co-contraction tests can be effective predictors of returning to sports after anterior cruciate ligament reconstruction S.H. Jang1, J.G. Kim1, J.K. Ha1 1 Inje University, Seoul Paik Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea Objectives: There is a lack of standardized objective criteria to accurately assess the ability of a patient to progress through the end stages of rehabilitation and safely return to their previous level of athletic activity after anterior cruciate ligament (ACL) reconstruction. The objective of this study was to determine objective factors involved in returning to sports following ACL reconstruction. Methods: Based on our inclusion criteria of a minimum 2-year followup, pre-injury sports activity level of Tegner 5 or greater, males aged 16–29 at the time of surgery, and having completed all the required tests and questionnaires, we retrospectively evaluated 67 patients among the 588 patients who underwent ACL reconstruction from March 2002 to January 2010. The patients were divided into ‘‘returnto-sports’’ (n = 51) and ‘‘non-return’’ groups (n = 16) by periodically surveying participants using a questionnaire. Comparisons between the two groups were made using pre-operative and postoperative International Knee Documentation Committee questionnaires (IKDC), Lysholm score, and KT-2000 arthrometer. Flexor and extensor muscle strength, using the Biodex system, one-leg-hop test, and three functional tests (co-contraction, shuttle run, and carioca test) were used for assessment. Results: Overall clinical results, including IKDC score, Lysholm score, and KT-2000 arthrometer, improved in all patients post-operatively and no significant difference was seen between the two groups (p [ 0.05). Although there was no significant difference in flexor or extensor deficits, one-leg-hop test, or shuttle run test, there were statistically significant differences in the co-contraction and carioca tests (p \ 0.05).

S169 Conclusions: Tests that assess ability to perform task that involves rotational moments showed statistically significant differences between the two groups. Co-contraction and carioca tests can be used as effective tools to assess a patient’s readiness to return to a previous level of sport activity. A step-wise rehabilitation program, stressing rotational functional stability, should be used when preparing a patient to return to a previous level of sport activity after ACL reconstruction.

P19-1146 Correlation between remnant type of the ACL and proprioception in the knee T. Mammoto1, A. Hirano2 1 Mito Kyodo General Hospital, Mito, Japan, 2Mito Kyodo General Hospital, Orthopaedic Surgery, Mito, Japan Objectives: Proprioception of the knee perceives joint movement and joint position, and it is important in their muscular control. ACL injured knee reduces joint position sense (JPS) compared to the uninjured knee. ACL contributes not only stability of the knee joint but also proprioception including JPS. No previous study has reported about the relationship between a remnant type of the ACL and JPS. The purpose of this study is to assess the relationship between the remnant type of the ACL and JPS in the knee. Methods: Ten patients, hospitalized for ACL surgery, were recruited. The remnant type of the ACL was evaluated under arthroscopy and divided by two groups; Scarring to the PCL and the ACL remnant disappeared. JPS was measured before their ACL surgery, according to Skinner‘s method. Foot was placed in a foot/ankle apparatus to secure the foot at 90 to the shank. Digital protractor was put on the apparatus. Ten measurements into extension/flexion directions were measured both injured and uninjured leg. JPS was expressed as the differences between the target angle and the reproduced angle. An average was used as the value of JPS. Paired t test or Mann–Whitney U test was carried with a level of significance of p \ 0.05. Results: Five participants showed disappearance of ACL remnant (ACL-D) and 5 remnants were scarring to PCL (ACL-P). In the ACLD, JPS in the injured was significantly higher than that in the uninjured (into extension; the injured 7.7 ± 3.7 vs. the uninjured 4.1 ± 3.1, p \ 0.05. into flexion; the injured 17.2 ± 12.4 vs. the uninjured 8.3 ± 5.3, p \ 0.05). In the ACL-P, JPS was significantly higher into flexion (the injured 4.4 ± 2.0 vs. the uninjured 2.9 ± 2.6, p \ 0.05). The angle difference in the ACL-D was significantly higher than that in the ACL-P into flexion (17.2 ± 12.4 vs. 4.4 ± 2.0, p \ 0.05). Differences between the injured and the uninjured in the ACL-D were significantly higher than that in the ACL-P (8.9 ± 7.3 vs. 1.5 ± 1.4, p \ 0.05). Conclusions: As previously reported, JPS was deteriorated in the ACL injured than in the uninjured. Number of mechanoreceptor in the ACL correlates joint position sense. The ACL remnant scarring to the PCL may help to collect information in joint position sense, and that may contribute to recognize joint position. JPS was deteriorated in flexion than that in extension. Proprioception influences the ACL-hamstring reflex, and that participate joint stability in the knee joint. Deterioration in feedback to the hamstrings might arise error into flexion than into extension.

123

S170 Joint position sense was deteriorated in the ACL injured knee than in the uninjured. Knee with disappeared ACL remnant was deteriorated joint position sense than that with scarring to PCL.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Table 1 RC for both the limbs and preferred/instructed manoeuvres RC (m/s2)

Left Max

P19-1155 Bone-patella-tendon-bone autografts versus iliotibial band autograft for ACL-reconstruction: a 15-year follow-up of a prospective randomized controlled trial P. Ho¨lmich1, F. Stensbirk1, K. Thorborg1, L. Konradsen2 1 Amager Hospital, University of Copenhagen, Department of Orthopaedic Surgery, Copenhagen, Denmark, 2Bispebjerg Hospital, University of Copenhagen, Section of Sports Surgery, Copenhagen, Denmark Objectives: Bone-patella-tendon-bone (BPTB) and hamstring tendons are the most frequently used autografts for reconstruction of the anterior cruciate ligament (ACL). The long-term results after using the iliotibial band autograft (ITB) are not fully known. If equal in quality the ITB graft could be a useful alternative as a primary graft, and in revision surgery or multiligament reconstruction. The hypothesis was that the ITB-reconstructed knees would show similar re-rupture rates as those reconstructed with the BPTB autograft. Methods: From 1995 to 1996 sixty subjects scheduled for primary ACL-reconstruction were included in a prospective randomized controlled trial. Three senior knee-surgeons, experienced in both types of ACL-surgery, performed all operations. Blinded observers involved in neither operation nor rehabilitation performed the observations at follow-up. A standardized and supervised rehabilitation program was used for both groups for 6 months. 30 patients received the ITB-reconstruction, and 30 received the BPTB-reconstruction. Primary outcomes were failure-rate (n = 49) and KOOS (Sport/Rec, QOL) (n = 38). Secondary outcomes were KOOS (pain, symptoms, ADL), Tegner activity scale, anterior-knee-pain (AKPscore), Lysholm Score, rolimeter laxity, extension deficit, single-hop and crossover-hop for distance. The statistical calculations were performed in SPSS statistics. Rerupture rates were compared using the Chi-square test. For comparisons between groups both parametric and non-parametric statistics were used. Results: At 15-year follow-up no differences existed between the groups. Graft-failure occurred in 3 BPTB-subjects (12.5%) and 4 ITB-subjects (16%) (P = 0.53). Sport/Rec score for the BPTB-group was 73(22), and 75(30) for the ITB-group (P = 0.82). The QOL score was 68(21) and 72(24) for the BPTB-group and ITB-group, respectively (P = 0.58). Conclusions: We found similar graft-failure-rates and KOOS-scores when comparing BPTB- and ITB-operated individuals, at 15-year follow-up. The ITB graft had equal long-term results compared to the BPTB graft and is recommended as a reliable alternative autograft for ACL-reconstruction.

P19-1162 Biomechanical insights of a standardized pivot-shift maneuver: an in vitro multi-tester study N. Lopomo1, V. Musahl2, Y. Hoshino3, M. Ahlde´n3, F.H. Fu2, S. Zaffagnini1 1 Istituto Ortopedico Rizzoli, Laboratorio di Biomeccanica e Innovazione Tecnologica, Bologna, Italy, 2University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States, 3University of Pittsburgh, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: Pivot-shift (PS) test is commonly used to diagnose ACL insufficiency but still remains surgeon specific. Several methods have been proposed to quantify the PS phenomenon. The aim of this study

123

Intra-tester

Inter-tester

Right Min

Range

Slope

Max

Min

Range

Slope

Preferred

1.6

0.5

1.8

19.6

2.9

0.7

3.3

74.2

Instructed

1.3

0.6

1.6

15.4

1.5

1.0

2.4

47.6

Preferred

2.0

0.5

2.3

21.0

3.0

0.7

3.4

74.2

Instructed

1.3

0.7

1.7

17.6

1.5

1.0

2.3

44.2

Fig. 1 Bar-chart of the averaged values of acceleration parameters

Fig. 2 Normalized acceleration curves obtained for each tester; in red the average normalized curve was to evaluate intra- and inter-tester variability in performing both surgeon-specific techniques and a standardized technique, quantified by means of an acceleration sensor. Methods: A complete lower body specimen was used for the study. Two different instability grades were created on left and right knee (ACL and ACL plus lateral meniscus respectively). 12 expert surgeons performed 3 repetitions of PS test based on their preferred technique followed by an instructed one. An acceleration sensor skinfixed to tibia was used to quantify acceleration parameters (max, min, range, slope). Repeatability Coefficient (RC) analysis was used to assess tester repeatability, Student t test was used to evaluate the differences between limbs and test conditions, Pearson Correlation Coefficients (R) were evaluated considering the acceleration curves for each tester and each repetition. Statistical significance was set at 95%.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: The RCs are reported in Table 1. Max acceleration was significantly higher in the left knee during instructed technique (p \ 0.05, Fig. 1); moreover the standard deviation decreased once introduce the instructed manoeuvre. Rs and averaged curves are reported in Fig. 2. Conclusions: All the acceleration parameters obtained were more consistent using the instructed PS technique (reduction of RCs and standard deviation between testers). The trends of the acceleration curves was more coherent in standardized PS. The parameters seem to be sensitive enough to allow discrimination of instability grades in the instructed PS, even if future studies are needed to investigate the advantages of a standard technique in comparing instability between testers. P19-1259 Comparison of ST/G and BPTB grafts in ACL reconstruction, results from the Danish registry of knee ligament reconstruction L. Wagner1, T. Thillemann2, A. Pedersen3, M. Lind2 1 Aarhus University Hospital, Division of Sportstrauma and Institute of Clinical Epidemiology, Aarhus, Denmark, 2Aarhus University Hospital, Division of Sportstrauma, Aarhus C, Denmark, 3Institute of Clinical Epidemiology, Aarhus University, Aarhus, Denmark Objectives: The choice of graft for Anterior Cruciate Ligament (ACL) reconstruction remains controversial, though the two most widely used grafts for ACL reconstruction are semitendinosus/gracilis (ST/G) and bonepatellar tendon-bone (BPTB). Despite numerous studies comparing ACL with these two grafts, there is still controversial about difference in outcome. Since 2005 the Danish registry of Knee ligament reconstruction (DKRR) has monitored the quality and development in ACL reconstruction. This database contains data from all clinics in Denmark performing ACL reconstruction. The objective of this study was to report the revision rate and patient related outcome measures when comparing the use of BPTB and ST/G graft in primary ACL reconstruction. Methods: This prospective cohort study evaluates 11089 primary ACL procedures form DKRR registered in the period from 2005 to 2010 using 9254 ST/G grafts and 1835 BPTB grafts, respectively. We calculated the ACL survival for both graft types using revision ACL reconstruction as primary endpoint. Further, we assessed the 1 year Knee Osteoarthristis Outcome score (KOOS) and Tegner score, according to graft type. Results: The use of ST/G graft in ACL reconstruction increased form 70% of all grafts in 2005 to 92% of all grafts in 2010. The revision rate after 4 years when using ST/G graft and BPTB graft was 3.6% and 2.8%, respectively. KOOS score preoperatively was comparable for the two groups. One year postoperatively the KOOS score was comparable between the two groups for pain, symptoms, ADL and quality of life. However, the ST/G group had a slightly higher score for sports and recreation with 62.8 compared to 58.6 for the BPTB group. Conclusions: The use of ST/G graft in ACL reconstruction have increased considerably over the last years although the literature remains controversial concerning outcome. A meta-analysis from 2003 states that there is significant higher graft failure when using ST/G compared to BPTB graft, which correlates with our results. However there is consistent data from several studies demonstrating that BPTB results in significantly more anterior knee pain than ST/G graft. The KOOS score was comparable in the two groups in our study. This is the first nationwide population based study reporting the results with different surgical techniques after primary ACL reconstruction. Our study suggests limited impact of graft choice for the outcome after primary ACL reconstruction. P19-1317 Can the single leg hop test identify neuromuscular and biomechanical asymmetries in ACL reconstructed patients? S. Xergia1, E. Pappas2, F. Zampeli1, S. Georgiou1, A.D. Georgoulis1

S171 1 Orthopaedic Sports Medicine Center, University of Ioannina, Department of Orthopaedic Surgery, Ioannina, Greece, 2Division of Physical Therapy, Long Island University-Brookly, Brooklyn, United States Objectives: To test if the single hop test identifies biomechanical and neuromuscular asymmetries in the return to sports phase after anterior cruciate ligament reconstruction (ACLR). Methods: Twenty-two physically active men with ACLR (mean age ± SD, 28.8, 11.2; height 1.77 ± 0.04; body mass 76.75 ± 10.5) at 6–9 (mean time after ACLR ± SD, 7.01, 0.93) months after surgery participated in this study. All patients completed isokinetic testing at three velocities (120/s, 180/s, 300/s) and single hop test for distance with simultaneous kinematic and kinetic evaluation. Linear regression analysis was performed between the asymmetry index (LSI) of the distance of the hop tests, the peak torque values, the peak flexion angles and the peak flexion moments, in order to investigate the correlation between them. Results: Linear regression analysis revealed significant correlation between the LSI of the single hop test for distance and the LSI of isokinetic extension at 120/s (p = 0.044), extension at 180/s (p = 0.042). Conclusions: The single leg hop for distance is a valuable functional tool that besides performance can be used to assess knee extension deficits. However, it cannot predict kinetic and kinematic asymmetries possibly due to effective neuromuscular compensation strategies. The last phase of rehabilitation should emphasize the quality of the hopping movement, which cannot be directly estimated by the functional measures of the single hop for distance or the isokinetic evaluation.

P19-1364 Double bundle three tunnels ACL reconstruction A. Maestro1, J. Fernandez Lombardia2, L. Rodriguez3, P. Garcia4, A.R. Meana4 1 FREMAP, Sports Medicine, Orthopaedic Surgery Department, Gijon, Spain, 2Hospital Arriondas, Arriondas, Spain, 3Hospital Cabuen˜es, Orthopaedic Surgery, Gijon, Spain, 4Hospital Cabuen˜es, Gijon, Spain Objectives: To know the results after DB-3T (one tibial and two femoral) respect to joint stability and clinical behavior compared with SB. Methods: Consecutive 35 patients with a unilateral rupture of the ACL treated with the DB-3T technique, and a control group of 36 non-concurrent patients, with similar conditions, treated with the SB technique were followed during 2 years. Anterior/posterior translation (APT), pivot-shift test (PS), one-leg hop and IKDC subjective evaluations were performed. An statistical analysis was made. Results: There were no significant differences in the reduction of APT (p = 0.798). However, there was a significant improvement in the remaining parameters in the DB-3T group: greater reduction in rotational instability (p = 0.005), greater one-leg hop capacity (p \ 0.0005), and greater subjective evaluation (p = 0.047). Conclusions: After 2-year follow-up, patients having undergone ACL reconstruction using hamstring tendons by means of a double bundle technique with a single tibial tunnel, showed greater rotational stability, greater one-legged hop capacity and a greater subjective evaluation than those having undergone single-bundle reconstruction.

P19-1421 Double-bundle anterior cruciate ligament reconstruction: a comparative cadaver study of femoral tunnels performed with in–out and out–in techniques M. Ronga1, P. Punzetto1, L. Callegari2, E. Genovese2, F.H. Fu3, L. Pederzini4

123

S172 1

University of Insubria, Dep. of Orthopaedics and Traumatology, Varese, Italy, 2University of Insubria, Dep. of Radiology, Varese, Italy, 3University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States, 4Sassuolo Hospital, Department of Orthopaedics and Arthrocopic Surgery, Sassuolo, Italy Objectives: Double bundle (DB) anterior cruciate ligament (ACL) reconstruction is nowadays a common procedure to manage ACL rupture. Many authors have reported in several short term follow-up prospective randomized studies a greater knee stability in DB reconstruction compared to single bundle. Despite several techniques reported in literature, as yet no study has demonstrated which DB technique has better outcomes. There are several factors that can influence biological integration of grafts and the bundle’s (anteromedial AM; posterolateral PL) mechanical behavior. The difference in diameter between tunnels on the intra-articular femoral side and graft is one of the most important. The aim of this study is to compare the diameter at the femoral tunnel entrance between two different techniques of DB ACL reconstruction in a cadaver model: the insideout and the outside-in techniques. The hypothesis is that a difference exists, in terms of diameter at the femoral tunnel entrance, between the abovementioned techniques. Methods: The study included 8 knees from 8 different cadavers divided into 2 groups. Group A: femoral tunnels were performed using an in–out technique: the PL tunnel from the AM portal and the AM tunnel from the transtibial PL tunnel. Group B: both tunnels were performed using an out-in technique with an out-in anatomical ACL guide system. All tunnels were drilled with a 7 mm acorn reamer. Upon cadaver dissection, the samples were evaluated by CT-scan on coronal and axial planes. The diameters of the two tunnels were measured on both planes. Results: Tunnel convergence was never observed in the two planes. In group A, AM tunnel measured 7.07 mm (range 7–7.1) on axial plane and 7.02 mm (range 7–7.1) on coronal plane. In group B, AM tunnel measured 7.1 mm (range 7–7.2) on axial plane, and 7.15 mm (range 7–7.3) on coronal plane (p [ 0.05). In group A, PL tunnel measured 8.32 mm (range 8.2–8.4) on axial plane and 8.45 mm (range 8.4–8.5) on coronal plane. In group B, PL tunnel measured 7.15 mm (range 7–7.3) on axial plane and 7.02 mm (range 7–7.1) on coronal plane. (p \ 0.05). Conclusions: Double bundle ACL reconstruction is a promising technique in terms of clinical results and knee stability. However, there has been some criticism concerning the procedure. Indeed, tunnel enlargement and tunnel communication could jeopardize a revision ACL surgery. Our study showed that the PL femoral tunnel entrance diameter is significantly larger in the inside-out than outsidein technique. Theoretically, the out-in technique should determine an inferior PL tunnel enlargement after ACL reconstruction. Several other variables such as type of graft, fixation and rehabilitation program should also be considered. Biomechanical and prospective randomized control studies between in–out and out-in techniques could confirm this hypothesis.

P19-1461 Can bone dowel prevent tibial tunnel enlargement in ACL reconstruction? A. Silva1, R. Sampaio2 1 D. Pedro V Military Hospital, Hospital Militar Regional no. 1, Porto, Portugal, 2HPP, Porto, Portugal Objectives: To evaluate prospectively if the impaction of a bone dowel in the tibial tunnel prevents the tunnels from enlarging beyond their original diameter. Methods: Seventeen patients that underwent arthroscopically assisted ACL reconstruction with hamstring autologous graft with impaction of autologous bone dowel in the tibial tunnel were evaluated

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 prospectively. All patients underwent CT of the knee on the day of surgery, at 3-months and 12-months post-op. The cross-sectional area of the tunnels was measured on 2-mm thick transversal images at 15 mm and 20 mm from the tip of the posterior wall of the tunnel. Statistical analysis was performed with the SPSS, version 18.0. Intra-class correlation coefficients were computed to measure the intra- and inter-observer agreement. Due to the sample size, median and interquartile range (percentiles 25 and 75) was used to describe continuous variables. Results: On the day of surgery, the median cross-sectional areas of the tunnels were 77.0 and 79.0 mm2, respectively at 15 mm and 20 mm. At 3-months, the median cross-sectional areas of the tunnels were 70.0 and 65.0 mm2, at 15 mm and 20 mm. At 12-months post-op, the median cross-sectional areas of the tunnels were 69.0 and 69.0 mm2. The median enlargement of the tunnels between 3- and 12-months post-op was 0.0 mm2 at 15 mm and -2.0 mm2 at 20 mm. Conclusions: The impaction of an autologous bone dowel in the tibial tunnel during hamstring ACL reconstruction keeps the tunnels from enlarging beyond their original diameter and that there is no further enlargement of the tunnels after 3 months post-op.

P19-1507 Clinical and radiological outcome of a simple and innovative technique for arthroscopic tibial spine avulsion fixation V. Pandey1, K.K.V. Acharya1, P. Sripathi Rao1, S.K. Rao1 1 Kasturba Medical College, Orthopedics Surgery, Udupi, India Objectives: Tibial spine avulsion is not an uncommon injury with motor vehicle accidents being most common mechanism of injury. Meyers and McKeever’s classified it into three radiological types. Type I (undisplaced), type II (partially displaced, and type III as completely displaced. Zaricznyj proposed a fourth category (Type IV) for comminuted avulsed fragment. Type I is managed conservatively whereas II, III and IV are currently managed operatively. Any neglected injury of tibial spine avulsion can lead to disastrous consequences of malunion or nonunion leading to flexion deformity, loss of extension and instability. Presently, various fixation methods are in practice such as staples, cancellous screw, sutures, wires and anchors. In the following series, we describe an arthroscopic technique which is simple, inexpensive and very effective suitable for type II to type IV tibial eminence avulsion fractures, using high strength sutures through the Anterior cruciate ligament (ACL)-bone junction and tying over a fixation button adjacent to tibial tuberosity, without using any specialized equipment. Methods: 11 patients with Tibial spine avulsion were operated using this simple technique since March, 2009. Clinical diagnosis was confirmed using plain radiographs and magnetic resonance imaging of the knee. Tibial spine was fixed using intravenous cannula needle and high strength sutures (Ultrabraid, Smith and Nephew, USA) which were passed through two tunnels from the bed of fracture exiting near tibial tuberosity and tied adjacent to tibial spine over a suture button (Smith and Nephew, USA). Postoperatively, all were immobilized in an extension brace for 2 weeks. Gentle knee mobilization was started after 2 weeks. Gradual weight bearing was permitted after 4 weeks with brace. Full weight bearing was permitted after complete radiological union. Results: Nine patients were male and two were female. All presented within 2 weeks of injury. Mean age was 26 years. Radiologically, nine cases were type 3 and other two were type 4. Mean follow up was 14.8 months (range; 6–22 months). Mean time for fracture union was 12.72 weeks. At the final follow up, nine patients had grade 0 Anterior drawer and Lachman whereas two had grade 1 anterior drawer and Lachman. None demonstrated pivot shift. Postoperative mean Lysholm score at the final follow-up was 98.18. On IKDC grading at the final follow up, nine had IKDC A whereas two had

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 IKDC B grade. At the final follow-up, none demonstrated any instability, extension loss or loss of movements as compared to unaffected knee. Conclusions: This is a simple, reproducible and reliable arthroscopic technique of fixing an avulsed tibial spine using inexpensive tools. It avoids usage of costly instruments, implants and complex maneuvers. It is Magnetic resonance friendly in postoperative period. The fixation is rigid and permits early mobilisation avoiding knee stiffness. It also completely obviates the need for implant removal later. It has the further advantage of its potential to be used in paediatric population as the size of two tunnels is less than 6%, avoiding damage to the growth plate. Keywords: Tibial spine, Avulsion fracture, Arthroscopy, Rigid fixation Level of evidence, IV

Knee-ACL III

P20-42 Femoral tunnel widening after hamstring anterior cruciate ligament reconstruction with bioabsorbable Cross-Pin N.-H. Choi1, Y.-H. Jung2, J.-S. Oh1 1 Eulji Medical Center, Department of Orthopaedic Surgery, Seoul, Republic of Korea, 2Daejin Medical Center, Department of Orthopedic Surgery, Seongnam, Republic of Korea Objectives: Previous reports have shown that bio-Transfix implants (Arthrex, Naples, FL) have the possibility of fracture during the early postoperative period. However, to date, there exists no study reporting the radiological and clinical significance of broken bio-Transfix implants. The purpose of this study was to compare radiological and clinical outcomes between intact and broken implant groups after hamstring ACL reconstructions fixed with bio-Transfix. Methods: Hamstring ACL reconstructions using bio-Transfix in 50 patients with a minimum of 2 years’ postoperative follow-up were reviewed. On the anteroposterior (AP) and lateral radiographs, the diameter of the femoral tunnel was measured at the widest diameter and compared to the diameter of the reamer used at surgery. Magnetic resonance imaging (MRI) scans were obtained 6 months postoperatively for all patients to evaluate the condition of bio-Transfix implants. The patients were divided into intact and broken bioTransfix groups. Postoperative stability evaluations were performed using the Lachman and pivot-shift tests, and instrumented laxity testing using the KT-1000 arthrometer. Functional evaluations were performed using the Lysholm score and Tegner activity scale. Results: Follow-up MRI scans revealed that the bio-Transfix was broken in 11 and intact in 39 patients. 5 implants were broken within the femoral tunnel and 6 were broken outside the femoral tunnel. 2 bio-Transfix in the intact group were bent. On the AP and lateral radiographs at 24 months postoperatively, the average diameter of the femoral tunnel in the intact group increased by 13.1 and 17.1%, respectively. In broken group at 24 months postoperatively, the average diameter of the femoral tunnel increased by 33.6 and 26.5%, respectively. There were significant differences between the two groups in the average diameters of the femoral tunnel on the AP radiographs at 24 months postoperatively (p = .000). However, on the lateral radiographs, there were no significant differences between the two groups. Postoperative knee stability tests and functional evaluations showed no significant differences between the two groups. Conclusions: Broken bio-Transfix implants resulted in significant femoral tunnel widening especially in the coronal plane. Surgeons

S173 should be aware of this phenomenon and take caution when selecting a fixation device for hamstring grafts.

P20-89 Intercondylar roof impingement in the anatomical double bundle ACL reconstruction: in vitro and in vivo studies T. Iriuchishima1, K. Shirakura1, F.H. Fu2 1 Gunma University Hospital, Division of Rehabilitation Medicine, Maebashi, Gunma, Japan, 2University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: The purpose of this study was to evaluate the graft roof impingement in the anatomical double bundle ACL reconstruction. Methods: For the in vitro evaluation, fresh frozen cadaver knees were used (n = 15). With the careful dissection of intact ACL, anatomical double bundle ACL reconstruction was performed using hamstrings auto graft. Before and after ACL reconstruction, pressure between intercondylar roof and ACL was measured using pressure sensitive films (Prescale: Fuji film Co., Ltd., Tokyo, Japan). Knees were then moved with 40 N of force and from full flexion to full extension, and the pressure pattern on the film was analyzed. For the in vivo evaluation, post-operative graft visualized 3D-CT (n = 24) and MRI (n = 20) was performed with full knee extension. The relationship between the graft and the intercondylar roof was evaluated using an axial view at the most distal slice of the intercondylar roof. Qualitative evaluation of the ACL graft was performed using MRI with a sagittal view of the T2 image. Results: In the cadaver study, no significant difference of impingement pressure was observed between intact ACL and anatomical double bundle ACL reconstruction. In the clinical studies, no positive impingement case was observed in the axial view of evaluation. However, saggital view of MRI showed some impingement suspected cases. In 1 case, ACL graft was bowed posteriorly, and in 3 cases, intensity alteration of the graft was observed. In these impingement suspected cases, femoral tunnels were created outside of anatomical ACL footprint. Conclusions: When the grafts were placed accurately in the ACL footprint, no impingement will occur in the anatomical double bundle ACL reconstruction.

P20-92 Kinematic characteristics of the chronical anterior cruciate ligament-deficient knee during in gait J. Zhou1, G. Li1, F. Li1 1 National Institute of Sports Medicine, Beijing, China Objectives: To determine how kinematic parameters of lower extremity joint may change as a result of anterior cruciate ligament (ACL) deficiency. Methods: Two groups were compared: 30 ACL deficiency patients (mean 14 months after injury) and 30 matched controls. A threedimensional motion analysis was used to determine kinematics data of the lower limb during comfortable-speed walking. The variables examined in the present study were the lower limb kinematic in special phase as well as time-distance parameters during gait. Oneway analyses of variance were performed on the subject means on the listed parameters. Results: The maximum knee flexion angle knee significantly decreased during the stance and swing phrase, while ankle dorsiflexion angle increased in injuried knee when compared control group. A significant difference was found in maximum tibial rotation angle when compared with the control group.

123

S174 Conclusions: Gait kinematic characteristic of ACL deficiency subjects showed tendency of compensating which is due to the anterior instability for ACL rupture.

P20-126 Immunohistological evaluation of the proprioceptive potential of the residual stump of an injured anterior cruciate ligament K. Bali1 1 PGIMER, Orthopaedics, Chandigarh, India Objectives: To evaluate the presence of proprioceptive potential in the residual remnants of the ruptured ACL, by indirectly picking up receptors in ACL residual tissue. Methods: Tissue from the ACL remnants was harvested during arthroscopy prior to ACL reconstruction in 63 consecutive patients. This was evaluated for evidence of residual proprioceptive fibers immune-histologically using H&E, and monoclonal antibodies to S-100 and NFP (Neurofilament protein). Results: Histological examination of harvested ACLs showed good subsynovial and intra-fascicular vascularity with free nerve endings in most of the injured stumps. Morphologically normal mechanoreceptors (as seen in H&E) and proprioceptive fibres (positivity with monoclonal antibody for NFP) were found in many of the injured stumps (46 and 52.4%). A statistically significant correlation between injury duration and persistence of mechanoreceptors and proprioceptive fibres was noted. More fibers were seen in patients with ACL remnant adherent to the PCL, and this difference was also statistically significant. Conclusions: Since proprioception is now understood to be major factor to maintain knee stability, mechanical reconstruction alone may not give the best results in ACL deficient knees. We have indirectly demonstrated persistent residual proprioceptive fibers in injured ACLs, more so in the early injury phase and with PCL adherence. We conclude that leaving the ACL remnants, if surgically possible, may be of potential benefit during ACL reconstruction, as some re-innervation and regaining of proprioceptive potential maybe possible. This would improve clinical outcomes.

P20-127 Evaluation of outcomes in conservatively managed concomitant Type A and B posterolateral corner injuries in ACL deficient patients undergoing ACL reconstruction K. Bali1 1 PGIMER, Orthopaedics, Chandigarh, India Objectives: An unrecognized (and therefore untreated) concomitant PLC injury is being increasingly recognized as one of the principle causes of poor outcomes after a mechanically sound ACL reconstruction in ACL deficient knees. There is enough evidence to suggest operative management of Type C PLC injuries in knees with ACL tear. However there is paucity of literature regarding the outcomes of ACL reconstruction in ACL deficient knees with concomitant Type A and Type B PLC injuries. Methods: We prospectively evaluated all the patients who underwent arthroscopic ACL reconstruction over a period of 3 years from January 2007 to December 2009. Patients with multi-ligament injury, Type C PLC injury, associated bony/chondral/meniscal injury or those undergoing revision ACL surgery were excluded from the study. A total of 102 patients who completed a minimum follow up of at least 1 year were ultimately included in the study. The patients with divided into three groups: group A with isolated ACL injury, group B1 with concomitant Type A PLC injury and group B2 with concomitant Type B PLC injury. The associated PLC injury in all these

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 patients was managed conservatively. Outcome assessment was based on IKDC scores measured preoperatively and at last follow up visits. Results: The mean age of the patients was 25.33 years (16–38 years) with 95 males and 7 females. The average follow up was almost 2.5 years (13–46 months). Group A had 88 patients while group B1 and B2 had 6 and 8 patients respectively. The preoperative IKDC scores were comparable for all the groups. The follow up IKDC scores were similar (statistically insignificant, p value: 0.421) for group A and group B1. Group B2 had poorer follow up IKDC scores as compared to group A and this result was found to be statistically significant (p value: 0.0001). Conclusions: We believe that conservative management of a concomitant Type B PLC injury adversely affects the outcomes of ACL reconstruction in these patients and should be avoided. Type A PLC injuries, on the other, do well without surgery and can be left as such even when associated with a concomitant ACL tear.

P20-174 Failure of a polyethylene terephthalate ligament used in cruciate ligament reconstruction: a case series J.T.K. Melton1, D. Wood2, J. Roe2, B. Caldwell2, M. Cross2, L. Pinczewski2 1 Australian Institute of Musculoskeletal Research, Sports Medicine, Sydney, Australia, 2Australian Institute of Musculoskeletal Research, Sydney, Australia Objectives: The search for readily available grafts which reduce morbidity and lead to more rapid recovery following reconstruction of the cruciate ligaments of the knee has driven the development of multiple artificial ligaments made from a variety of materials. The ligament advanced reconstruction system (LARS) is one such ligament which is currently in regular usage. The authors work in a large sports medicine practise which is often referred patients with failed cruciate reconstructions. They report the presentation and mechanisms of failure seen in a series of failed LARS cruciate ligament reconstructions referred to the unit. Methods: Clinical and operative records between January 2008 and August 2011 from a large orthopaedic and sports medicine clinic were reviewed and all cases of revision surgery following failed LARS reconstruction of the cruciate identified. The clinical presentation, imaging findings, operative records, histology reports and postoperative course were assessed. Results: 12 patients with failed LARS were identified. Mean age 33.75 years (22–55 years). 11 cases were LARS ACL reconstruction and 1 for PCL reconstruction. Median time to failure was 16 months (2 months to 8.5 years). Reasons for presentation including pain, swelling (5 cases), recurrent instability (6 cases), crepitus (1 case) and stiffness (2 cases). Imaging and operative findings included, effusion, synovitis (2 cases), ruptures of the graft (6 cases), failure of fixation (3 cases), free graft fibres within the joint (5 cases), cavitation of femoral tunnel (1 case) and evidence of abrasive chondral damage (1 case). One case (cavitation of tunnel) required two stage revision and all other cases underwent single stage revision. Histology shows multinucleated giant cell infiltrate consistent with foreign body reaction. Patient progress following revision is satisfactory. Conclusions: The authors report a series of failed LARS grafts. They identify problems with recurrent instability, synovitis, rupture of the graft, PET material debris within the knee and abrasive chondral wear. These mechanisms of LARS failure have been described in relation to previously used artificial ligaments. We suggest that medium term outcome studies and ligament reconstruction registries that demonstrate efficacy and safety are required before this artificial graft is accepted into common usage.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P20-190 Identifying individuals with an anterior cruciate ligament deficient knee as copers and non-copers: a narrative literature review Y. Kaplan1 1 Hebrew University, Lerner Sports Center, Jerusalem, Israel Objectives: ACL rupture may result in increased tibiofemoral laxity and impaired neuromuscular function, which ultimately may lead to knee instability and dysfunction. Individuals who opt to choose surgery, due to these changes, may be defined as ‘‘non-copers’’. Conversely, ‘‘copers’’ may be defined as individuals with an ACL deficient knee who do not have functional impairment and instability and who successfully resume pre-injury activity levels without surgical intervention. This narrative literature review is designed to explore the differences and outcomes between individuals who have had anterior cruciate ligament (ACL) reconstruction and those who did not undergo surgical intervention following a tear of the ACL. Second, to review the evidence related to the ability to identify individuals who may or may not need surgery after an ACL rupture. Finally, to describe the differences between copers and non-copers. Methods: An electronic search was conducted up to April 2011, using medical subject headings and free-text words. Subject-specific search was based on the terms ‘‘anterior cruciate ligament reconstruction versus conservative treatment’’, ‘‘copers’’, ‘‘non-copers’’. Results: A similar percentage of copers and non-copers return to sporting activity. Three papers used an algorithm and screening examination involving individuals with ACL injuries. Evidence exists that, as opposed to copers, non-copers have: deficits in quadriceps strength, vastus lateralis atrophy, quadriceps activation deficits, altered knee movement patterns, reduced knee flexion moment, and greater quadriceps/hamstring co-contraction. Conclusions: ACL screening examination shows preliminary evidence for detecting potential copers. Objective differences exist between copers and non-copers. Individuals with ACL injury should be informed of the possibility of good knee function following a nonoperative rehabilitation program.

P20-245 Knee proprioception in the ACL injury risk position S.M. Mir1, S. Talebian1, N. Nasseri1 1 Tehran University of Medical Sciences, Tehran, Islamic Republic of Iran Objectives: Knee joint proprioception combines sensory input from a variety of afferent receptors that encompasses the sensations of joint position and motion. Most studies have favored testing knee proprioception with in the sagittal plane in a non-weightbearing (NWB) position. No study has measured proprioception in a manner relevant either to the mechanism of injury. Therefore, the aim of this study is to measure knee joint position sense (JPS) in non-contact ACL injury risk and normal weight-bearing (WB) condition in healthy subjects using the active reproduction of the angles. Thirty male healthy volunteer athletes participated in the study. Methods: Thirty male healthy volunteer athletes (age, 24.19 ± 5.27 years; height, 177.9 ± 7.7 cm; weight, 78.03 ± 11.7 kg) participated in the study. The dominant knees of subjects were tested. JPS was evaluated by reproduction of the angles in the ACL injury risk and normal WB conditions. Electrogoniometer was used to measure JPS. Absolute angular error (AAE) was considered as a dependent variable. Results: The results showed less accurate knee JPS in non-contact ACL injury risk position, rather than normal WB. Significant difference was found in ACL injury risk and normal positions (p \ 0.05). Conclusions: The poorer JPS in non-contact ACL injury risk positions compared with normal WB condition identified in this study may contribute to the increased incidence of ACL injury. One of the most

S175 common mechanisms of non-contact ACL injury is knee dynamic knee valgus. It is possible that people with less accurate proprioception in non-contact ACL injury risk positions may be at greater risk of ACL injury.

P20-409 Interference screw fixation of soft tissue tendon grafts in the tibial tunnel: an in situ comparison of composite interference screws J. Nyland1, R. Krupp1, J. Greene1, R. Bowles1, R. Burden1, D.N.M. Caborn1 1 University of Louisville, Orthopaedic Surgery, Louisville, United States Objectives: Soft tissue tendon graft slippage is problematic over the initial months following anterior cruciate ligament (ACL) reconstruction due to the variable cancellous bone density in the tunnel region of interest. This in vitro porcine model study compared ACL soft tissue tendon graft tibial tunnel fixation using 4 different composite interference screws. Methods: Based on an a priori statistical power analysis 48 porcine profundus tendons and tibiae were divided into four groups of 12 matched specimens. Tibiae were pre-screened using DEXA scans to only include specimens that simulated young human tibial bone mineral density (BMD). Equivalent diameter soft tissue tendon grafts were assigned to each group. Tibial bone tunnels were drilled at 0.5 mm [ graft diameter. Depuy MilagroTM (Group 1), Arthrex BiocompositeTM (Group 2), Stryker BiosteonTM (Group 3) and Smith & Nephew Biosure HATM (Group 4) interference screws were compared. The same fellowship trained surgeon performed all graft implantations and interference screw insertions. Maximal insertion torque and perceived within group biomechanical testing outcome predictions (0–10 visual analog scale, end range descriptors 0 = Extremely poor, 10 = Excellent) were recorded. Tibiae were potted and loaded into a 6 of freedom clamp with the servohydraulic device tensile loading vector aligned directly with the bone tunnel providing a ‘‘worst case’’ loading scenario. Constructs were pre-loaded to 25 N, followed by pre-conditioning (0–50 N, 0.5 Hz, 10 cycles), 500 submaximal loading cycles between 50 and 250 N (1 Hz) and load to failure testing at 20 mm/ min. Results: Tibial BMD, graft diameter, graft length, screw insertion torque, graft loop distance, tibial tunnel length, and perceived within group biomechanical test outcome did not differ between groups. Mean displacement during submaximal cyclic loading did not display statistically significant group differences (P = 0.15). Mean stiffness during submaximal cyclic loading (P = 0.15), mean yield load at failure (P = 0.38), mean displacement at yield load (P = 0.45), mean stiffness during load to failure testing (P = 0.69), mean load at failure (P = 0.53) and mean displacement at ultimate failure load (P = 0.08) did not display statistically significant group differences. Conclusions: Under controlled surgical and biomechanical test conditions the four composite interference screw groups displayed comparable soft tissue tendon graft tibial tunnel fixation. Since biomechanical test performance was comparable between groups the influence of composite screw use on soft tissue tendon graft-tunnel integration during the remodeling and ligamentization phases takes on greater importance.

P20-472 Knee function following non-operative treatment after anterior cruciate ligament injury in skeletally immature children. A prospective cohort study H. Moksnes1, L. Engebretsen2, M.A. Risberg3

123

S176

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

1

Norwegian School of Sport Sciences and Hjelp24 NIMI, Sports Medicine, NAR, Oslo University Hospital, Oslo, Norway, 2Oslo University Hospital, Orthopaedic Center, Oslo Sportstrauma Research Centre, Oslo, Norway, 3Oslo University Hospital, NAR Orthopedic Center, Oslo, Norway Objectives: To prospectively evaluate the functional outcome from non-operative treatment of anterior cruciate ligament injuries in skeletally immature children. Methods: The inclusion criterion was traumatic intrasubstance anterior cruciate ligament rupture sustained at age 12 years or younger. Open femoral and tibial growth plates were confirmed by MRI. Knee function was evaluated at baseline and at a 1 year follow-up using four single-legged hop tests (single hop, triple crossover hop, triple hop, and 6 m timed hop), isokinetic muscle strength tests (peak torque at 60/s), registration of activity level, and self-reported questionnaires (IKDC 2000, KOS-ADLS, and KOOS). Results: Forty-six consecutive children with an average age of 11.0 years (7.0–12.9 years) at time of injury were included. Baseline examination was performed at median 28 weeks (min 11 weeks, max 211 weeks) after injury, and follow-up examination at median 53 weeks (min 44 weeks, max 59 weeks) after baseline. Twentyseven (68%) of the children had returned to their pre-injury activity level at follow-up. Three of the non-reconstructed children have had a repair of the medial meniscus. Four children underwent anterior cruciate ligament reconstruction during the follow-up period due to recurrent giving way episodes (n = 3), or meniscus injury (n = 1, medial meniscus repair). Results at follow-up demonstrated excellent leg symmetry indexes for hop tests ([94%) and for muscle strength measurements ([94%). Additionally, paired-sample t tests demonstrated significant improvement in knee function: self-reported questionnaires (IKDC 2000 p = .04, KOS-ADLS p = .02, KOOS pain p = .03, KOOS ADL p = .02, KOOS QoL p = .002), and the

Table 1 Functional outcome paired samples Baseline (SD) Follow-up (SD) p value KOS-ADLS

87.6 (12.4)

90.9 (10.4)

.02

IKDC 2000

79.1 (14.8)

83.6 (13.4)

.04

Single hop (%)

90.5 (14.7)

97.3 (13.3)

.004

Triple crossover hop (%) 92.2 (11.3)

94.6 (11.9)

.24

Triple hop (%)

92.2 (9.6)

95.0 (10.3)

.10

6 m timed hop (%)

93.7 (9.2)

96.3 (9.5)

.09

Quadriceps strength (%)

89.1 (13.0)

94.4 (10.2)

.07

Hamstring strength (%)

95.3 (18.5)

94.2 (13.2)

.75

Fig. 1 Results KOOS paired samples

123

single-legged single hop test (p \ .01) from baseline to follow-up (Table 1; Fig. 1). Conclusions: Skeletally immature children demonstrate excellent knee function with limb symmetry index of [94%, and high selfreported function with the majority returning to pre-injury activity levels. Additionally, a low occurrence of secondary injuries following non-operative treatment after anterior cruciate ligament injury was noticed. Hence, studies with more than 1 year follow-up are needed.

P20-477 Improved morphological outcomes after ACL reconstruction with quadrupled hamstrings grafts and additional bone plug augmentation at 5-year follow-up S. Nebelung1, G. Deitmer2, R. Gebing3, F. Reichwein1, W. Nebelung1 1 Marienkrankenhaus Du¨sseldorf-Kaiserswerth, Du¨sseldorf, Germany, 2 Crossklinik, Basel, Switzerland, 3Department of Diagnostic Radiology, St. Vinzenz-Krankenhaus, Du¨sseldorf, Germany Objectives: Hybrid fixation has been hypothesized to improve outcomes of anterior cruciate ligament (ACL) reconstructions. The purpose of this study was to evaluate mid-term clinical and morphological outcomes after transfemoral graft fixation using either a conventional (Group A, n = 37) or a modified technique with additional bone plug augmentation (BPA) of the femoral tunnel aperture (Group B, n = 22). Methods: Sixty-one months (range, 52–69 months) after ACL reconstruction using a quadrupled hamstring autograft with femoral BioTransfix fixation (Arthrex, Naples, FL, USA) and tibial PLLA interference screw fixation (Deltascrew, Arthrex) with or without additional BPA, 59 patients were followed up by clinical and MRI examination. A-p-laxity measurements and IKDC-, Lysholm and Tegner activity-scoring were performed. Imaging included assessment of bone tunnel diameters, graft condition and graft filling at the femoral bone tunnel aperture. Results: Patients having undergone additional BPA had a significantly higher degree of graft filling at the femoral bone tunnel aperture (p = .0073). Though not significant, they also tended to have ‘healthier’ grafts (p = .0644) and less a-p-laxity-difference (p = .1372), both in average scores and total patient numbers. Lysholm-, IKDC- and Tegner activity index scores were not found to be significantly different. Equally, no differences were found in bone tunnel diameters. Conclusions: Additional BPA is an easy-to-perform, cheap and safe manoeuvre, which has the capacity to improve morphological and clinical outcomes at 5-year follow-up. However, femoral tunnel widening is unaffected by additional BPA.

P20-488 Is femoral tunnel length correlated with geometry of the intercondylar notch and femoral condyle after double bundle ACL reconstruction using the transportal technique? An in vivo imaging analysis using computed tomography J.G. Kim1, J.H. Wang2, H.C. Lim3, J.H. Ahn4, H.J. Kim3, J.H. Bae5 1 College of Medicine, Korea University, Guro Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea, 2Sungkyunkwan University School of Medicine, Samsung Medical, Seoul, Republic of Korea, 3Korea University College of Medicine, Guro Hospital, Orthopedic surgery, Seoul, Republic of Korea, 4Kang buk Samsung Medical Center, Sungkyunkwan University, Orthopaedic Surgery, Seoul, Republic of Korea, 5College of Medicine, Korea University, Ansan Hospital, Orthopedic surgery, Ansan, Republic of Korea Objectives: To analyze femoral tunnel geometry via computed tomography (CT) imaging and evaluate the distal femoral anatomic

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 factors affecting femoral tunnel length after anterior cruciate ligament (ACL) reconstruction using the transportal technique. Methods: Twenty-nine patients underwent an anatomic double-bundle ACL reconstruction with the femoral tunnel drill via a transportal technique. CT imaging with OsiriX imaging software was used to measure femoral tunnel length [anteromedial (AM), posterolateral (PL)]. In order to measure the length of virtual femoral tunnel made in the center of ACL femoral insertion site simulating anatomic single bundle ACL reconstruction (central femoral tunnel), we drew a line that passed the half point of the line connecting the center of the intraarticular aperture of the AM and PL tunnels (Fig. 1). Femoral tunnel divergent angle, femoral condyle parameters [medial/lateral femoral condyle anteroposterior (MFC/LFC AP) distance, medial/lateral posterior condyle anteroposterior size (MPC/LPC offset distance) and medial to lateral epicondylar distance (M–L epicondylar distance)] and intercondylar notch size parameters (maximum notch width, middle notch width, notch height and notch area) was measured Correlations between femoral tunnel length and femoral condyle size and intercondylar notch size parameters were analyzed. Results: The mean AM, PL, and central femoral tunnel lengths were 33.3 ± 3.9 mm (range, 26.4–40.0 mm), 33.6 ± 3.6 mm (range, 28.7–42.4 mm), and 34.3 ± 3.2 mm (range, 26.1–39.3), respectively. The femoral tunnel length \30 mm developed in 7 (24.1%) cases for AM and 4 (13.8%) cases for PL. The mean femoral tunnel divergent angle was 14.4 ± 4.1 (range, 7.4–23.7). A positive correlation was found between AM femoral tunnel length and MFC/LFC AP distance (p = .01, r = .46/p = .01, r = .43, respectively), M-L epicondylar distance (p = .03, r = .39), middle notch width (p = .009, r = .47), notch height (p = .001, r = .57), and notch area (p \ .001, r = .58). However, PL and central femoral tunnel length were not correlated with femoral condyle size or intercondylar notch size. Conclusions: After double-bundle ACL reconstruction using the transportal technique through the AAM portal, the AM and PL femoral tunnels showed mean tunnel length [30 mm and a divergent angle. However, the femoral tunnel length \30 mm developed in some cases. AM femoral tunnel length was correlated with femoral condyle size (MFC AP distance, LFC AP distance and M–L epicondylar distance) and intercondylar notch size (notch width, notch height and notch area).

P20-489 Inaccuracy of intraoperative femoral tunnel length measurement: comparison between transportal and outside-in techniques after double bundle anterior cruciate ligament reconstruction with an in vivo imaging analysis using 3D-CT J.H. Wang1, J.G. Kim2, J.H. Ahn3, H.C. Lim4, H.J. Kim4, J.H. Bae5 1 Sungkyunkwan University School of Medicine, Samsung Medical, Seoul, Republic of Korea, 2College of Medicine, Korea University, Guro Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea, 3Kangbuk Samsung Hospital, Sungkyunkwan University, Orthopaedic Surgery, Seoul, Republic of Korea, 4Korea University College of Medicine, Guro Hospital, Orthopedic Surgery, Seoul, Republic of Korea, 5Korea University Ansan Hospital, Orthopedic Surgery, Seoul, Republic of Korea Objectives: To evaluate the accuracy of intraoperative femoral tunnel length measurement and to compare it between the two techniques after anterior cruciate ligament (ACL) reconstruction using the transportal (TP) and outside-in (OI) techniques. Methods: 62 consecutive patients underwent ACL reconstruction using an auto-hamstring tendon graft at our center. Of the 62 patients, thirty-nine patients underwent primary unilateral DB ACL reconstruction using either the TP (using Bullseye) or the OI technique (using FlipCutter). The participants were randomized on the day of surgery to either a TP technique group (Group I, 21 cases) or an OI

S177 technique group (Group II, 18 cases). The intraoperatively measured femoral tunnel length was recorded, and the postoperative femoral tunnel length was measured using computed tomography (CT) with OsiriX imaging software. We calculated the femoral tunnel length difference (femoral tunnel length measured intraoperatively—femoral tunnel length measured postoperatively) and graft length in the femoral tunnel (fixed graft length = femoral tunnel length measured postoperatively—Endobutton or Retrobutton size). In each group, we compared femoral tunnel length measured between intraoperatively and postoperatively. Tunnel length difference between intraoperatively and postoperatively measured femoral tunnel lengths was compared between the two groups. Results: There was no significant difference on patient demographics between group I and II. The mean intraoperatively measured femoral tunnel length was significantly longer than that measured postoperatively in group I (p \ .001 for anteromedial (AM) and posterolateral (PL) femoral tunnel length) and II (p = .007 for AM and p \ .001 for PL).The mean AM/PL femoral tunnel length difference in Group II were larger than those of Group I (p = .04 for AM, p = .003 for PL). The mean PL fixed graft length of Group II was shorter than that of Group I (p = .05). The mean AM fixed graft length of Group II was shorter than that of Group I; however, there was no significant difference between the groups. The numbers of cases with an AM fixed graft length \15 mm in Groups I and II were 4 (4/21) and 7 (7/18), respectively (p = .17), and the incidences of PL fixed graft length in Groups I and II were 4 (4/21) and 8 (8/18), respectively (p = .08). Conclusions: After anatomic DB ACL reconstruction, intraoperatively measured femoral tunnel length was longer than that measured postoperatively in both TP and OI technique. However, the intraoperative femoral tunnel length measurement of the OI technique obtained using a FlipCutter was less accurate than that of the TP technique using Bullseye.

FP20-533 Soft tissue healing in an in vivo anterior cruciate ligament (ACL) allograft model N.N. Verma1, S. Bhatia1, R. Bell1, R. Frank1, B.R. Bach1, S. Chubinskaya1 1 Rush University Medical Center, Department of Orthopaedic Surgery, Division of Sports Medicine, Chicago, United States Objectives: The effect of low-dose gamma irradiation on healing of soft tissue allografts remains largely unknown. The purpose of this study was to compare soft tissue healing in a bone tunnel using three types of ACL grafts: non-irradiated allografts, low-level (1.2 MRad) gamma irradiated allografts, and autograft controls. We hypothesized that soft tissue allograft healing to bone would be delayed compared to that of autograft tissue and that low-dose (1.2 MRad) gamma irradiation would not affect the healing response of allograft tissue after ACL reconstruction. Methods: Surgery: 48 New Zealand white rabbits underwent bilateral ACL reconstructions with semitendinosus tendon graft. Sixteen rabbits were reconstructed with autografts, the remainder with allografts. The 32 allograft rabbits each received one irradiated allograft (1.2 Mrad), with the contralateral leg receiving a non-irradiated allograft. Animals were euthanized at 2 or 8 weeks post-operatively. Biomechanics: Tensile stiffness, maximum load, and displacement at maximum load were measured using custom designed grips on an electro-mechanical materials testing system. Histology: Tibial and femoral segments were sectioned perpendicular to the tunnel axis (at the mid-portion of the tunnel length) allowing for histologic and histomorphometric analyses at the tendon-bone interface. Image J software (NIH, Bethesda, MD) was used to compute the percentage new growth within the bone tunnels.

123

S178 Results: There were no significant differences between the maximum load or stiffness values among all groups at 8 weeks. At 2 weeks, autograft exhibited significantly (p \ 0.01) lower maximum load compared to the non-irradiated grafts. There was no statistical difference between autograft and irradiated allografts, or between irradiated allografts and non-irradiated allografts at 2 weeks. Regarding histology, at both 2 and 8 week time points, autograft tendon displayed more advanced degenerative and remodeling processes in comparison with irradiated allograft and non-irradiated allograft. Histomorphometric analyses demonstrated no differences between the grafts at any time point. Conclusions: The maximum load and stiffness of a healing tendon graft in ACL reconstruction appears to be independent of low-dose (1.2 Mrad) irradiation. At 8 weeks, there was no difference with regard to biomechanical analysis of bone-tendon healing in allografts versus autograft controls. Histologic analyses suggested a faster remodeling response in autograft specimens in comparison with allografts at all time points.

P20-552 Histological analysis on the tendon-to-bone healing utilizing bone marrow-derived MSCs in ACL reconstruction model without a tibial bone tunnel T. Kanazawa1, T. Soejima1, K. Noguchi1, K. Tabuchi1, M. Noyama1, K. Nagata1 1 Kurume University, School of Medicine, Orthopaedic Surgery, Kurume, Japan Objectives: To verify whether a structure identical to the normal ligament-bone insertion could be regenerated at the tendon-bone interface without a bone tunnel, we designed an original ACL reconstruction model without a tibial bone tunnel and investigated the histological changes occurring between the bottom of the grafted tendon and the tibial bone. Moreover, to enhance the anchoring process in this model, we transplanted bone marrow-derived MSCs (bMSCs) between the grafted tendon and the bone. Our first hypothesis was that the grafted tendon would be anchored via chondroid tissue at part of the joint-aperture site even if a bone tunnel was not created. Second hypothesis was that the application of bMSCs at tendon-bone interface would yield results histologically superior to those for controls in ACL reconstruction model without a tibial bone tunnel. Methods: 14 skeletally mature Japanese white rabbits were used. Autologous bMSCs were harvested from the bone marrow 2–4 weeks before surgery. For bilateral ACL reconstruction in our originally designed model, bMSCs in collagen sponge or fibrin sealant carrier were transplanted between the grafted tendon and the bone in the experimental limb, whereas the control limb received the carrier only. The animals were euthanized at 4 and 8 weeks after the operation, and 7 rabbits at each time point were used for histological analysis. Results: At 4 weeks, poorly organized fibrovascular tissue consisted of type 3 collagen was evident between the grafted tendon and the bone. In the bMSCs group, a chondroid cell layer stained with safranin-O and type 2 collagen was observed at the posterior interface in two of seven specimens, compared with only one of seven specimens in the control group. No chondroid cell layer was observed at the anterior interface in either group. At 8 weeks, the interface was less cellular and had become progressively organized comparison with 4 weeks. In the bMSCs group, a chondroid cell layer was more obvious at the posterior interface in five of seven specimens compared with 4 weeks. In addition, a chondroid cell layer was observed at both the anterior and posterior interfaces in only one specimen. Conclusions: Even in our present ACL reconstruction model without a tibial bone tunnel, integration via chondroid tissue was seen at part of the joint-aperture site. Our first hypothesis was acknowledged. Thus, the bone tunnel was not essential factor as viewed in light of this study. On

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 the other hand, there were no appreciable differences between the groups, and integration between the grafted tendon and the bone was more markedly localized at the posterior interface than at the anterior interface. Thus our second hypothesis was disproved, at least under the specific conditions of this study. However, although only 1 case, an active chondroid cell layer was observed between the grafted tendon and the bone throughout in bMSCs group. Further studies are needed to determine the potential utility of bMSCs application.

P20-585 Knee stability, athletic performance and sport-specific tasks in non-professional soccer players after ACL reconstruction: comparing trans-tibial and antero-medial portal techniques C. Tudisco1, S. Bisicchia1, F. Chiozzi2, M. Piva2 1 University of Rome Tor Vergata, Orthopaedic Surgery, Rome, Italy, 2 Isokinetic Group, Rome, Italy Objectives: To retrospectively evaluate two groups of non-professional soccer players (Tegner score = 7 points) operated on with two different surgical techniques for an ACL reconstruction in terms of stability, muscle strength, endurance, sport-specific tasks and well validated knee scores. Methods: We retrospectively evaluated two groups of patients operated on for an ACL reconstruction drilling the femoral tunnel either through the antero-medial portal (AMP) or with a trans-tibial technique (TT). At follow-up patients were evaluated with KT-1000 arthrometer, manual pivot shift test, isokinetic test, Incremental Anaerobic Threshold Test and sport specific tasks. The evaluation was integrated with well validated knee functioning scales (Knee score modified by Insall and IKDC score. In all cases KT-1000 measures and pivot shift tests were performed by the same author. Anterior tibial translation was expressed as mean ± standard deviation of the difference between operated and contralateral knee at 67, 134 N, under manual maximum translation and under quadriceps contraction. Pivot shift test was classified according to the IKDC scoring system as: normal (same anterior tibial translation as the contralateral knee), nearly normal (glide), abnormal (clunk) and severely abnormal (gross instability). An unpaired t test was used to compare both groups with respect to objective outcomes. Categorical variables were compared using a Chi-square test between both groups. For all statistical tests, the alpha level was set at 0.05. Statistical analyses were performed with SPSS v.15.0 (SPSS Inc., an IBM Company, Chicago, IL, USA). Results: At KT-1000 evaluation the two groups obtained similar results without statistically significant differences (p [ 0.05). At manual pivot shift test, a glide was present in all patients of TT group, otherwise none of the patients in AMP group showed a positive pivot shift test (p \ 0.05). At isokinetic test both groups showed only a minimal deficit of flexor and extensor muscles strength and a low flexor/extensor muscles ratio (valore normale 66%). In the TT group the flexor deficit at 90/s was mild (-10.5%). Both TT and AMP groups showed comparable results at isokinetic test, without any statistically significant difference (p [ 0.05). At Incremental Threshold Test both groups ran at comparable speed at S2 and S4 without any statistically significant difference (p [ 0.05), but patients in AMP group showed significantly higher heart rates both at S2 and S4 (p = 0.001 and 0.02 respectively). Conclusions: The group of patients in which femoral tunnel was drilled through the antero-medial portal obtained statistically significant better results in considered outcomes when compared to the other group. The better rotational stability of the knee achieved drilling the femoral tunnel through the AMP lead to better clinical and functional result, that are very important for active patients involved in sports.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P20-603 Incidence and risk factors of graft failure in double-bundle anterior cruciate ligament reconstruction E. Tsuda1, Y. Ishibashi1, Y. Yamamoto1, S. Maeda1, Y. Kimura1, S. Sasaki1 1 Hirosaki University Graduate School of Medicine, Orthopaedic Surgery, Hirosaki, Japan Objectives: Recent laboratory and clinical studies demonstrated the superiority of double-bundle anterior cruciate ligament reconstruction (DB-ACLR) to closely restore the normal knee kinematics under mechanical and manual testing. It is expected that this biomechanical advantage of DB-ACLR contributes to reduce the incidence of postoperative graft failure, however, has not been established. The objective of this study was to investigate the incidence and risk factors of graft failure after DB-ACLR. Methods: Two hundred seventy knees of 154 female and 113 male patients (average age of 24.3 ± 11.6 years) who received primary DB-ACLR from 2003 to 2009 were studied. The semitendinosus tendon with or without the gracilis tendon was transplanted using a arthroscopic transtibial technic. The patients who had a graft failure with recurrent anterior knee instability were collected, and the patient profiles were investigated focusing on age at the primary DB-ACLR, gender, time period from the primary DB-ACLR to graft failure, sports activities and episodes of trauma. Knee laxity in the clinical tests and knee muscle strength immediate before graft failure were also investigated. Results: Eleven knees (4.1%) of 8 female and 3 male patients had graft failure. The female ratio of 73% was significantly higher and the age at the primary DB-ACLR of 14.8 ± 2.7 years was significantly lower in the patients experienced graft failure compared to the patients without graft failure. The time period from the primary DBACLR to graft failure was 13.5 ± 9.2 months. All patients except one experienced traumatic re-injury in sports activities, and 7 of 10 patients were injured in playing basketball. All knees were clinically stable under KT-1000 measurements and recovered to 80% of quadriceps and hamstrings torque before re-injury. Conclusions: DB-ACLR might provide much more chance to return to sports, and consequently increase traumatic graft failure. Younger age at the surgery, female gender and playing basketball were risk factors of graft failure after DB-ACLR.

P20-647 Hip acetabular dysplasia and joint laxity of female anterior cruciate ligament-injured patients J. Yamazaki1, T. Muneta2, I. Sekiya3, Y.-J. Ju4, H. Koga4 1 Tokyo Medical and Dental University, Tokyo, Japan, 2Tokyo Medical and Dental University Hospital, Department of Orthopaedic Surgery, Tokyo, Japan, 3Tokyo Medical and Dental University Hospital, Section of Cartilage Regeneration, Graduate School, Tokyo, Japan, 4Tokyo Medical and Dental University, Orthopedic Surgery, Tokyo, Japan Objectives: It has been noted that some female anterior cruciate ligament (ACL)—injured patients have complaints of both coxalgia and joint laxity. The purpose of this study is to compare the hip characteristics of ACL-injured female patients with that of an ACLuninjured female control group and to evaluate the relationship between acetabular dysplasia and generalized joint laxity in ACLinjured female patients. Methods: Hip radiographs of 100 female ACL-injured patients and 40 female athletes without any hip joint complaints or history of ACL injury were evaluated by measuring their center-edge angle (CEA). In addition, generalized joint laxity tests using 8 items were performed for ACL-injured patients. Anterior–posterior (A–P) tibiofemoral translation of the uninjured knee was measured using a KT-1000 knee

S179 arthrometer to evaluate joint laxity under anesthesia before ACL reconstruction. Results: The average CEA of female ACL-injured patients was 25.5 ± 5.3 (uninjured side) and 25.8 ± 4.8 (injured side), and that of the control group was 28.2 ± 4.2 (right side) and 29.2 ± 5.7 (left side), both P \ .05. Among the 100 patients with ACL tears, both the generalized joint laxity score and A-P tibiofemoral translation of the group with acetabular dysplasia (CEA of \25, n = 37) were significantly greater than that of the normal group (CEA of C25, n = 63). There was a negative correlation between the CEA of female ACL-injured patients and both the generalized joint laxity score and A-P tibiofemoral translation. Conclusions: The CEA of female ACL-injured patients was significantly smaller than that of control group. Statistical analysis showed a moderate negative correlation between the CEA and generalized joint laxity score. Female athletes with an ACL injury had an increased prevalence of acetabular dysplasia and generalized joint laxity.

P20-658 Factors contributing to quadriceps muscle weakness after ACL reconstruction J.F. Item1, N.C. Casartelli1, M. Bizzini1, N.A. Maffiuletti1 1 Schulthess Clinic, Neuromuscular Research Laboratory, Zurich, Switzerland Objectives: Quadriceps muscle function plays an important role in patients following ACL reconstruction. These patients present a considerable deficit in quadriceps muscle strength (weakness), which has been identified as mediator in the development of knee osteoarthritis. The purpose of this retrospective pilot study was to investigate some of the neural and muscular mechanisms possibly underlying quadriceps muscle weakness after ACL reconstruction. Methods: Twenty-one subjects (17 men, 4 women; mean age ± SD: 31 ± 8 years; mass: 74 ± 10 kg; height: 175 ± 7 cm) were tested 6.5 month after unilateral ACL reconstruction with bone-patellartendon-bone (N = 10) or hamstring tendon (N = 11) graft. In each group, four patients had an isolated ACL rupture while the remaining patients had an ACL rupture with concomitant meniscal injury. The main exclusion criterion was previous surgery on the uninvolved lower extremity. Subjects were asked to perform quadriceps maximal voluntary contractions (to assess muscle strength), with superimposed electrical stimuli delivered to the femoral nerve both during (to assess the completeness of muscle activation) and immediately after the contraction (to assess muscle contractility). All measurements were conducted in isometric conditions with 70 of knee flexion. Vastus lateralis muscle thickness was also investigated using B-mode ultrasonography. Student’s paired t tests (one-tailed) were used to detect eventual differences between the involved and uninvolved side (p \ 0.05) for muscle strength, activation, contractily and thickness. The percent difference between the two sides (bilateral deficit) was also calculated for these four main outcomes. No significant difference was observed between the two operating techniques, and therefore results were collapsed together. Results: The mean bilateral deficit for quadriceps muscle strength was 26% (p \ 0.001). Muscle activation did not differ significantly between the two sides (80% for the involved vs. 84% for the uninvolved), but was quite low for both quadriceps. The involved quadriceps showed significant bilateral deficits of 10 and 7% (p \ 0.01) for muscle contractility and thickness, respectively. Conclusions: Quadriceps muscle weakness 6 months following ACL reconstruction is due to both muscular (impaired muscle contractility and reduced muscle mass) and neural factors (impaired muscle activation). Contrary to the muscular deficits that are mainly restricted to the involved quadriceps, neural impairments seem to affect the two sides. These findings may help clinicians ‘‘understanding of their

123

S180

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

patients’’ neuromuscular characteristics following ACL reconstruction and improve postoperative rehabilitation programs.

P20-689 Functional outcome after transphyseal ACL reconstruction in children C. Holwein1, S. Hinterwimmer1, H.O. Mayr2, P. Brucker1, E.-O. Mu¨nch2, A.B. Imhoff1 1 Technische Universita¨t Mu¨nchen, Abteilung fu¨r Sportorthopa¨die Klinikum rechts der Isar, Mu¨nchen, Germany, 2OCM Klinik, Orthopa¨dische Chirurgie, Mu¨nchen, Germany Objectives: The treatment algorithm for skeletally immature patients with intraligamentous ACL-tears has changed over the last years. Conservative treatment options with the intention to protect the open physes are at risk of resulting in secondary meniscal and cartilage damage. Furthermore patients’ sporty expectations till skeletal maturity are forcing us to develop new strategies. Aim of the current study was to proof that, compared with the healthy knee, transphyseal single-bundle hamstring reconstruction of intraligamentous ACL-tears in skeletally immature patients shows equal stability, no growth disturbances and good results in return to sports. Methods: From 42 patients (28 m, 14 w) with open physes, who received transphyseal single-bundle hamstring ACL-reconstruction between 10/2006 and 04/2010, we retrospectively examined 37 patients with a follow-up of 24.9 ± 11.1 months. Mean age at time of surgery was 13.2 ± 1.6 years in boys and 13.1 ± 1.0 years in girls. Five patients could not be included because of surgery at the contralateral knee (2) or ACL-rerupture (3). Femoral fixation was performed with endo-button (21), transfix-pin (17) or bioabsorbable interference screws (4), tibial fixation each with bioabsorbable interference screws (42). Knee function was determined with the IKDC 2000 form. Knee joint stability was assessed with the KT-1000 in neutral position and in a defined internal and external rotation using a validated instrument. Body-height, leg-length and leg-axis were determined. Sports activity was evaluated with a questionnaire including type and intensity of sport. Results: IKDC subjective score reached 92.9% in mean. IKDC objective was grade A or B in 75.7% and grade C in 24.3%. Compared to the healthy knee tibial translation of the affected knee in neutral position resulted in 1.2 mm (-1.5 to 5.0 mm), in 2 Nm lower leg internal rotation in 0.7 mm (-3.5 to 5.0 mm) and in 2 Nm external rotation in 0.8 mm (-1.9 to 5.8 mm). No leg length discrepancy more than 1 cm (-0.2 ± 0.5 cm) could be found. There was no valgus or varus deformity more than 3 in 34 patients. 3 patients showed varus (1) or valgus (2) malalignement between 3 and 4.5. 35 Patients indicated that their sports ability at least remained the same. Patients were doing sports on average 5.7 h (0–17.5 h) per week, 57% of them in competitive sports. Conclusions: Subjective and objective knee stability in children after transphyseal ACL-reconstruction is convincing. Rerupture rate is low. Abnormal growth disturbances were not seen. Return to competitive sport is possible.

P20-741 Finite element analysis of graft positions in ACL reconstruction P. Smolinski1, S. Kramer2, F.H. Fu3 1 University of Pittsburgh, Department of Mechanical Engineering, Pittsburgh, United States, 2University of Pittsburgh, Pittsburgh, United States, 3University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: In ACL reconstruction, it is important to understand the forces in the graft under normal knee kinematics. Finite element

123

Fig. 1

Fig. 2

analysis was used to calculate the forces in various ACL replacement graft positions under healthy knee kinematics. Methods: Using a robotic system, the passive path of a fresh frozen knee was found from 0 to 90 of flexion. Anterior tibial translation (ATT) was measured under an 89 N load at 0, 30, 60 and 90. This data would be used in the computational model. The ACL was transected and tunnels used to mark the anterio-medial (AM) and posterior-lateral (PL) bundle insertions. A transtibial technique was used to drill tunnels in a PL to High AM position. Five tunnel configurations were evaluated: (1) AM–AM, (2) PL–PL, (3) Mid–Mid geometric mid-point between the centers of the AM and PL insertion sites), (4) PL-High AM and (5) Double bundle. Knee geometry was obtained from a CT image. Six mm grafts were created connecting the insertion sites. A previously validated hyperelastic material model with a 3rd order strain energy function was used for the graft model. The graft was assumed to be at zero stress at 0 flexion. The force in the graft was calculated during all loading conditions. Results: The tension in all graft models increases as a function of flexion angle (Fig. 1). Double bundle and AM–AM initiate tension at 24 flexion. Mid–Mid and PL–PL grafts initiate tension at 60 and 63. The PL-HighAM graft model initiates tension at 36 and increases to 311.3 N at 90. Figure 2 shows graft tensions under anterior tibial translation at different flexion angles. Double bundle and AM–AM exhibit higher graft tension in lower flexion angles, while PL-HighAM exhibits low graft tension at lower flexion angles. At 90 of flexion, PL-HighAM

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 has the largest tension due to the already high residual tension caused by flexion. Conclusions: Graft position greatly effects graft tension. The PLHighAM graft model has the largest tension. Zero graft tension at full extension is a limitation of the study. The choice of graft position in ACL reconstruction is highly debated and recent attention has been focused on the importance of anatomical tunnel placement. The current study provides insight into the effect that graft position and orientation has on graft tension, when subject to healthy knee kinematics.

P20-769 Flexible instruments outperform rigid instruments to place anatomic ACL femoral tunnels without hyperflexion M. Steiner1, R. Smart2 1 Harvard Medical School, Section Chief Sports, New England Baptist Hospital, Boston, United States, 2New England Baptist Hospital, Sport Medicine Section, Boston, United States Objectives: This study evaluated the ability of flexible instruments compared to rigid instruments to place anatomic femoral tunnels in ACL reconstructions using both transtibial and anteromedial drilling without hyperflexion. Methods: Rigid and flexible pins were placed in twelve matched pairs of cadaver knees with transtibial drilling (6 pairs) and anteromedial drilling (6 pairs) at 110 flexion. Intraosseous pin lengths, femoral exit locations, and tunnel alignment were measured. Results: Transtibial drilling with rigid pins placed relatively vertical femoral tunnels 5.8 ± 1.0 mm superior to the central ACL insertion Transtibial drilling with flexible pins placed tunnels in the center of the femoral attachment, but the tunnels were relatively close to the posterior femoral cortex, 8.0 ± 5.9 mm (p \ .05), compared to transtibial drilling with rigid pins. Anteromedial drilling resulted in central anatomic pin placements with rigid and flexible instruments. Tunnel lengths with flexible pins were longer, 42.0 ± 7.2 mm compared to tunnel lengths with rigid pins, 32.5 ± 7.1 (p \ .01). Flexible pins exited farther from the posterior cortex compared to rigid pins (p \ .01). In three of six knees with rigid pins the exit point was at the posterior border of the femoral cortex. All flexible pins exited a safe distance from the posterior femoral cortex. Conclusions: Transtibial drilling with rigid instruments did not produce anatomic femoral tunnels. Transtibial drilling with flexible pins produced anatomic tunnels but the tunnels were close to the posterior femoral cortex. Anteromedial drilling without hyperflexion produced anatomic tunnels using rigid and flexible instruments but with flexible instruments the tunnels were longer and were farther from the posterior femoral cortex. Anteromedial drilling with flexible pins produced tunnels with good length and the best position. Flexible instruments compared to rigid instruments can facilitate the creation of anatomic femoral tunnels using anteromedial drilling without hyperflexion.

P20-790 Effects of stiff and soft landing techniques on knee loading during a modified single-leg cross-over hop A. Benjaminse1, A. Gokeler2, B. Otten1 1 University Medical Center Groningen, University of Groningen, Center for Human Movement Sciences, Groningen, The Netherlands, 2 University Medical Center Groningen, Center for Rehabilitation, Groningen, The Netherlands Objectives: Anterior cruciate ligament (ACL) ruptures frequently occur in non-contact athletic maneuvers during rapid decelerations

S181 with the knee close to extension. Vertical ground reaction force (vGRF) magnitude is a key component to the knee load. We investigated the influence of stiff and soft landing on knee loading and the potential role of instructions on improving technique. Methods: Healthy recreationally athletes (8 males (M); 23 ± 2.2 years, 1.85 ± 0.09 m, 74.6 ± 8.8 kg; 8 females (F); 21 ± 0.9 years, 1.76 ± 0.0 m, 66.1 ± 8.8 kg) performed a modified single-leg cross-over hop for (sub maximal) distance across a 20 cm width path. Subjects were asked to perform five natural (i.e. using their own preferred technique), ten soft and ten stiff landings (instructed with respectively ‘land as soft or as hard as you can’). The third of four landings was captured on the force plate and subjects were required to keep balance for 3 s after the fourth landing. Landing techniques were counterbalanced between subjects. External moments and angles were analyzed at peak vGRF. Multivariate ANOVA and post hoc Bonferroni was conducted with a set at 0.05. Results: For peak vGRF (N/kg) a main effect for technique was attained between the natural (M = 25.38 ± 11.62, F = 29.65 ± 4.36) and stiff (M = 46.55 ± 10.95, F = 42.47 ± 11.10) and the soft (M = 14.13 ± 8.13, F = 15.42 ± 8.47) and stiff landing techniques (p \ 0.001). Knee valgus moments (Nm/kg) were significantly different between the natural (M = -0.04 ± 1.16, F = 0.29 ± 0.74) and stiff (M = 1.01 ± 2.41, F = 1.01 ± 2.41) and the soft (M = 0.05 ± 0.83, F = 0.40 ± 0.57) and stiff landing techniques (p \ 0.001). Knee flexion moments were significantly lower in females, regardless of technique (p = 0.011). Significant differences were found between soft (M = -0.94 ± 0.97, F = -0.68 ± 1.90) and stiff (M = -0.55 ± 1.78, F = -0.21 ± 1.99) landings (p = 0.029). For flexion/extension angles (), a significant difference between the natural (18.75 ± 14.09) and stiff (9.85 ± 13.09) landing techniques for females was found (p = 0.004). For varus/valgus angles (), the natural landing technique was significantly different between gender (M = -0.18 ± 5.56, F = -4.70 ± 10.98, p = 0.019). Furthermore, both males (-4.50 ± 10.14) and females (-4.51 ± 7.97) showed greater knee valgus angles during stiff landing techniques (p \ 0.001). Conclusions: Regardless of gender, natural kinematics and kinetics did not change during soft landing, indicating that a relatively safe landing technique was used. Females had a more extended knee than males during stiff landing, which might indicate that females are less able to cope with knee flexion during greater impact landings that requires more quadriceps activity. This study shows that females produced stiffer landing differently than males, possibly causing or due to the difference in knee flexion moment at peak vGRF. Landing with less flexion and therefore less shock attenuation is not optimal for meniscus loading and might reflect a ‘ligament dominant’ landing profile.

P20-791 Graft failure after primary ACL reconstruction: an analysis of time to failure C. Kaeding1, M. Schroeder2, A. Pedroza3 1 The Ohio State University College of Medicine, Department of Orthopaedic Surgery, OSU Sports Medicine, Columbus, Ohio, United States, 2The Ohio State University College of Medicine, College of Public Health, OSU Sports Medicine, Columbus, Ohio, United States, 3The Ohio State University College of Medicine, Sports Medicine, Columbus, Ohio, United States Objectives: Many studies have examined risk factors for graft failure after ACL reconstruction (ACLR), but very little has been done evaluating when graft failure occurs. This study has 2 Objectives. In ACL grafts that fail: 1. when do they fail and

123

S182

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

2. what factors influence this time to failure. This study is not evaluating risk of failure, but rather the timing of graft failure. Methods: An analysis of a prospective ACL reconstruction (ACLR) database was done to find revisions of primary ACLRs. 116 of these had a clear time interval from primary ACLR to graft failure determined by chart review. Kaplan/Meier survival analysis was performed to evaluate time to failure (TTF) and the influence of age, activity, sex, graft type, meniscus status and BMI on TTF. Wilcoxen test for equilivancy of survival functions was performed to determine statistically significant differences in TTF. Activity was recorded as a Marx Activity Score (0–16). Graft type was evaluated as hamstring versus patella tendon as well as allograft versus autograft. Median TTF (mTTF) was the time at which 50% of the failures had occurred. Results: For the entire cohort the mTTF was 22 months. Physical activity had the strongest influence on TTF. High activity (Marx C12) patients had mTTF of 17 versus 51 months for low activity (Marx \12) patients (p \ 0.002). With respect to sex, there was no difference in the low activity group, but in the high activity group females had mTTF of 12 versus 25 months for the males (p = 0.05). In allograft versus autograft analysis, regardless of activity level, there was no difference in TTF. In hamstring versus patella tendon graft analysis, regardless of activity level there was no difference in TTF. Presence of a medial meniscus tear resulted in a longer mTTF in low activity patients (p \ 0.014). Presence of a lateral meniscus tear did not influence TTF. Normal BMI patients had a shorter mTTF compared to overweight patients even after adjusting for activity (p \ 0.018). Conclusions: In this cohort of first time revision ACLRs, the median TTF was 22 mos. Factors that did not influence TTF included lateral meniscus status, auto versus allograft use, or hamstring versus patella tendon use. Factors that did influenced TTF included activity level, sex, presence of a medial meniscus tear, and BMI. Activity had the strongest influence with high activity patients having significantly shorter time to failure after primary ACLR than the low activity group. Fig. 1 P20-795 Kinematics of single-bundle and double-bundle ACL reconstruction in medial meniscus-deficient knees using a porcine model P. Smolinski1, J. Zhou2, G. Tantisricharoenkul3, L. Chen3, M. Linde-Rosen4, F.H. Fu5 1 University of Pittsburgh, Department of Mechanical Engineering, Pittsburgh, United States, 2National Institute of Sports Medicine, China, Beijing, China, 3University of Pittsburgh, Department of Orthpaedic Surgery, Pittsburgh, United States, 4University of Pittsburgh, Pittsburgh, United States, 5University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: Medial meniscal injuries and meniscetomy are common in anterior cruciate ligament (ACL) injured patients. The purpose of this study was to investigate the kinematics of single-bundle and double-bundle ACL reconstruction in the medial meniscus-deficient knee. Methods: Sixteen fresh-frozen mature porcine knees were tested using a robotic testing system. An 89 N anterior tibial load and a 4 N-m internal tibial torque were applied to knees in two groups: the anatomic single-bundle reconstruction group (SB) and the anatomic double-bundle reconstruction group (DB) in four states: 1. intact medial meniscus (IMM), 2. medial meniscus deficiency (MMD), 3. medial meniscus and ACL deficiency (MM_ACLD) and

123

Table 1 Relative reduction (RR) in ATT after ACLR compared with MM_ACLD Flexion angle

Group

N

RR rate Mean ± SD (%)

30

SB

8

63.7 ± 9.5

DB

8

72.6 ± 7.6

SB

8

50.9 ± 14.1

DB

8

57.0 ± 7.2*

SB

8

45.7 ± 13.7

DB

8

48.8 ± 9.7

60 90

* Compare with SB group P \ 0.05 4. ACL reconstruction (ACLR), either SB ACL reconstruction or DB ACL reconstruction. All the ACL reconstructions were done by three-portal technique. For DB group, the AM and PL tunnel positions were selected in the middle of AM and PL insertion sites on both femur and tibia. For the SB group, the tibial and femoral tunnels were drilled between the center of AM and PL insertion sites of the tibia and femur. The graft

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 was secured with an Endobutton (Smith & Nephew Endoscopy) on the femur and with washer screw on the tibia. The fixation of the SB graft was done with an 80 N load at 30 of knee flexion (full extension of the porcine knee). For DB reconstruction, the PL bundle was tensioned at 40 N at 30 of knee flexion and the AM bundle tension was 40 N at 60 of knee flexion. Results: After ACL transection, ATT increased significantly. Only at 30 of knee flexion was ATT restored to the intact value (DB reconstruction) in the meniscus deficient knee (Fig. 1). Furthermore, the ATT relative reduction rate was reduced in the DB ACL reconstruction group more than the SB group at 60 knee flexion. (Table 1, P \ 0.05). There is no difference in internal rotation (IR) between SB ACL reconstruction and DB ACL reconstruction. There is no significant difference for the relative reduction rate between the groups as well. The relative reduction is the difference between in the quantity before and after the reconstruction. Conclusions: Anatomic DB reconstruction can better restore the intact knee’s kinematics than SB reconstruction in the meniscus injured porcine model. Double-bundle ACL reconstruction may be preferred in the knees with high grade instability such as ACL injury with concomitant meniscus deficiency.

P20-797 Femoral tunnel drilling during ACL reconstruction: evaluation of the tunnel characteristics using either flexible- or straight reamers through an anteromedial portal P. Christel1, W.G. Clancy Jr2, F. Anne3, D. Appleby4 1 Habib Medical Center Olaya, Sports Medicine, Riyadh, Saudi Arabia, 2The University of Wisconsin-Madison, Sports Medicine, Madison, United States, 3Hopital Prive´ Paul d’Egine, Imaging, Champigny s/Marne, France, 4Smith & Nephew, Inc., Endoscopy, Biostatistical Services, Andover, United States Objectives: It has been shown the ACL femoral foot print can be hardly reached with transtibial drilling. As an option, medial portal (MP) drilling has been developed to solve this issue. However, MP drilling with straight drill includes pitfalls: bending the knee at 120 of flexion, damages to the cartilage of the medial femoral condyle and, shorter femoral tunnel with oval aperture. Flexible drill systems have been introduced to overcome these drawbacks but, little is known on the femoral tunnel characteristics when using flexible— compared to straight drills. Hypothesis: Compared to straight drills, flexible drills allow obtaining a more horizontal tunnel orientation, a more circular tunnel aperture, a shorter intra bony length of the guide wire. Methods: Nine pairs of human fresh knee cadavers were used. Under arthroscopic control, 10 mm femoral sockets were drilled through a MP portal using either straight or flexible drills in the center of the ACL foot print. Distances of the pin exit to the lateral structures were measured and, after removal of all soft tissues, direct measurements on the specimens and, 3D CT were used to measure socket aperture diameters (‘‘D’’ & ‘‘d’’), socket orientation and, distal femoral bone morphometry: bicondylar diameter, lateral femoral condyle width, intercondylar notch width and, divergence angle of the condyles. Statistical analysis evaluated the influence of gender on the results. Results: The mean distance to the LCL femoral attachment was 30.5 mm for straight pins (SP) and 17.3 mm for flexible pins (FP) (p \ 0.0001). The mean distance to the popliteus tendon attachment was 40.7 mm for SP and 27.2 mm for FP (p \ 0.0003). The mean distance to the peroneal nerve was always superior to 40 mm for both pins. There was no significant difference in the bone morphometry parameters for both groups either with direct measurements or 3DCT. Only ‘‘D’’ was significantly larger for rigid drills compared to flexible ones (p \ 0.001 after direct measurements, p \ 0.005 for 3D-

S183 CT); Thus, flexible drills lead to a more circular aperture. The surface area of the aperture was significantly larger for rigid drills (p \ 0.001 after direct measurements, p \ 0.02 for 3D-CT). With both methods of measurement the mean interosseous length was not significantly different either with FP or SP. In the coronal plane, the socket was more horizontal for flexible drills (p \ 0.001). In the axial plane socket orientation was not significantly different. There was no gender influence on the socket characteristics; only the drill’s type was significant. Conclusions: The results are in accordance with the initial hypothesis: drilling the femoral socket using flexible drills through a MP leads to a more circular aperture than with rigid drills with a significantly lower surface area. The drilled socket is more horizontal than with rigid drills, but the interosseous lengths of the pins are not significantly different. Graft stability at the tunnel aperture might increase with flexible drills.

P20-805 Effects of anatomic and non anatomic partial anterior cruciate ligament augmentation on knee rotational stability in a porcine model P. Smolinski1, C. Yapici2, K. Keklikci3, D. Kim3, M. Linde-Rosen4, F.H. Fu2 1 University of Pittsburgh, Department of Mechanical Engineering, Pittsburgh, United States, 2University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States, 3University of Pittsburgh, Department of Orthpaedic Surgery, Pittsburgh, United States, 4University of Pittsburgh, Pittsburgh, United States Objectives: Many studies have clarified that the anterior cruciate ligament (ACL) consists of two functionally and distinct bundles: the posterolateral (PL) and the anteromedial (AM). Each bundle has a different size, position and function. The reciprocal tension pattern of the PL and AM bundles cause each to be subjected to different injury mechanisms, depending on the knee flexion angle. A partial ACL injury may leave one bundle intact. The purpose of this study is to compare knee kinematics between anatomic and non-anatomic partial PL ACL augmentation. The hypothesis is that partial PL anatomic ACL augmentation provides better restores knee kinematics than nonanatomical partial PL ACL augmentation. Methods: Eight fresh frozen, mature porcine knees were tested using a robotic/universal force-moment sensor (UFS) testing system1. All the surgical reconstructions were done using a three portal technique via arthroscopy. Four groups were compared: 1. Intact ACL, 2. Deficient IM and PL, 3. Anatomic PL augmentation (Fig. 1), 4. Non Anatomic PL augmentation (Fig. 2). An 89-N anterior load was applied, and anterior tibial translation (ATT) was measured at knee flexion angles of 30, 45, 60, and 90 and a rotatory tibial load of 4 N m, external rotation (ER) and internal rotation (IR) were applied at 30, 45 and 60 of knee flexion. Data obtained from the different knee conditions was analyzed with software (Prism GraphPad Version 5.0a) with two-tailed Paired t test with the level of significance set at p \ 0.05. Results: There was no significant difference between in ATT (p [ 0.05) in all flexion angles. There was a significant difference between anatomic PL and non anatomic high PL group in IR at 30 (p \ 0.05) and 60 (p \ 0.05), and in ER at 60 (p \ 0.05) of knee flexion. Conclusions: The ACL of the porcine knee consist of three bundles; AM, PL and IM (intermediate). The IM bundle has a minor role in

123

S184 porcine knee stability. For this reason in order to protect the AM bundle in a knee with a PL bundle tear, this study investigated the rotational stability before and after PL bundle reconstruction. The most important finding in this study is non-anatomic PL ACL augmentation showed significant rotational laxity compared with the intact knee. A high tunnel location is performed in single bundle and double-bundle reconstructions as well as in ACL augmentation surgery. There has been much interest in symptomatic partial ACL tears selected individualized augmentation of the AM or PL bundle may provide many advantages. Protected by the intact bundle, it may allow earlier rehabilitation and return to sports, vascularization and joint position sense may be increased.

P20-822 Efficacy of the bone tunnel evaluation by CT for ACL revision surgery Y. Hashimoto1, S. Yamasaki1, Y. Hara2, J. Takigami3, T. Tomihara4, H. Nakamura1 1 Osaka City University Graduate School of Medicine, Orthopaedic Surgery, Osaka, Japan, 2Osaka, Japan, 3Japan Osaka Yodogawa Christian Hospital, Orthopaedics, Osaka-shi, Japan, 4Shimada Hospital, Orthopaedic Surgery, Habikino, Japan Objectives: Outcome of revision ACL surgery were reported inferior to primary ACL reconstruction. Many reasons for example graft selection, tunnel enlargement, meniscus damage and cartilage defect were described previously. One reason of poor outcome was inadequate location of recreated the femoral bone tunnel. The purpose of this study was to evaluate efficacy of the bone tunnel evaluation by CT for ACL revision surgery. Methods: Three patients underwent ACL revision surgery at 2010. The patients were all females. One is 19 years old who had underwent single ACL reconstruction using hamstring tendons 2 years ago, the others is 42 years old of single reconstruction using artificial graft 22 years ago and 43 years old with BTB graft 18 years ago. CT examination of the all patients was performed with a helical CT machine preoperatively and postoperative 1 week. Images of 1 mm slice thickness were reconstructed using three-dimensional (3D) reconstruction software. The location of the femoral bone tunnel were evaluated with Quadrant method [(Shallow-deep)%/(high-low)%] and anticipated location of femoral bone tunnel were prepared for preoperative planning. We underwent double bundle ACL reconstruction using Hamstring tendons for all the revision surgery. We compared the location of the anticipated bone tunnels and actual bone tunnels. Results: The location of bone tunnel of 19 years old had been 26%/ 20% preoperatively, anticipated location was AM: 26%/40%, PL: 35%/63% and the actual location was AM: 24%/46%, PL: 35%/65% without tunnel communication. Actual AM tunnel was created slightly lower than anticipated bone tunnel but almost the supposed position. The location of bone tunnel of 42 years old with artificial graft had been 34%/24% preoperatively, anticipated location was AM:21%/45%, PL: 37%/62% and the actual location was AM:24%/ 50%, PL: 44%/67% without tunnel communication. Actual AM and PL tunnels was created lower and shallower than anticipated bone tunnels. The location of the interference screw of 43 years old with BTB graft had been 49%/25%, anticipated location was AM:23%/ 24%, PL:34%/50% and the actual location was AM:27%/25%, PL: 39%/53% without tunnel communication. Actual AM and PL tunnels was created slightly shallower than anticipated bone tunnels. Conclusions: Preoperative CT evaluation is effective to understand the location of created bone tunnels. 3D-CT clearly shows whether the created tunnel is correct or incorrect position and is easily planned the anticipated bone tunnels. Our experiment is useful method to recreate the precious bone tunnels without tunnel communication.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P20-824 Higher bone mineral density of femoral insertion site of ACL than that of non-insertion site Y. Hashimoto1, S. Yamasaki1, J. Takigami1, Y. Hara2, T. Tomihara3, N. Hiroaki1 1 Osaka City University Graduate School of Medicine, Orthopaedic Surgery, Osaka, Japan, 2Osaka, Japan, 3Shimada Hospital, Orthopaedic Surgery, Habikino, Japan Objectives: The tunnel of femoral side was recently created referring the ‘‘resident’s ridge’’ under the reconstruction of the anterior cruciate ligament (ACL). The anatomy of ‘‘resident’s ridge,’’ was described the raised bony landmark commonly visualized just anterior to the femoral attachment of the ACL. In performing an ‘‘anatomic’’ ACL reconstruction, the location of the anatomy of these fiber bundles is critical. The objective of this study was to define the morphology and bone property of the femoral attachment areas of ACL by radiograph and DXA prepared the slices vertically to ACL insertion using cadaveric knees. We evaluate the distance from articular cartilage edge to anterior edge of ligament insertion and bony ridge and bone mineral density around the femoral attachment by DXA analysis. Methods: 12 male cadavers from the anatomical course at Osaka city university medical school of medicine were used in this study. The cadaver’s knee had no ACL rupture, or significant osteoarthritis. The average age was 82.3 years old. Distal femur was split in the sagittal plane using an bone trimmer through the mid line of the intercondylar notch and cross-sections 3-mm thick of lateral condyle was made parallel to Blumensaat’s line. Four slices were prepared and named group1, 2, 3 and 4 from deep to shallow. We classified the shape of the bony ridge by X-ray for each specimen. The distance from articular cartilage edge to attachment of ACL and anterior ridge were measured. Bone surface of the each cut specimens at the lateral femoral condyle were divided into 3 areas (A: anterior to the attachment, C: center of attachment, P: posterior part of attachment) and bone mineral density (BMD) of each areas were measured by DXA analyzer. Results: The shape of the medial wall of the lateral condyle at ACL insertion were identified as the three types, one ridge type, trapezial type and flat type. One ridge type were 58%, 58%, 75% and 33%, trapezial type were 42%, 42%, 0% and 0% and flat were 0%, 0%, 25% and 67% in group 1, 2, 3 and 4, respectively. The average distance from articular cartilage edge to insertion of ACL and anterior ridge were 11.6 and 9.9 mm, respectively. The average of the BMD was 68.7, 99.9 and 85.9 mg/cm2 in A,C and P of Slice1, 63.2, 92.9 and 79.0 mg/cm2 of Slice2, and 62.7, 81.1 and 78.5 mg/cm2 of Slice3 and 68.2, 67.2 and 63.7 mg/cm2 of Slice4. The BMD of center area at Slice 1 and 2 were higher than that of other zones. Conclusions: In this study, we showed the bony ridge was similar location to anterior attachment of ACL. The reason why the BMD of center area at Slice 1 and 2 were higher than that of other zones is thought that mechanical tensile stress to attachment of tendon to bone were concerned to higher BMD.

P20-825 Effect of notchplasty in anterior cruciate ligament reconstruction: a biomechanical study in porcine knees K. Keklikci1, C. Yapici2, D. Kim3, M. Linde-Rosen4, P. Smolinski5, F.H. Fu6 1 Gulhane Military Medical Faculty Haydarpasa Teaching Hospita, Department of Orthopaedic Surgery, Istanbul, Turkey, 2University of Pittsburgh Medical Center, Department of Orthopaedic Surgery, Pittsburgh, United States, 3University of Pittsburgh Medical Center, Pittsburgh, United States, 4University of Pittsburgh, Pittsburgh, United States, 5University of Pittsburgh, Department of Mechanical

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1 The effect of notchplasty on ATT

Engineering, Pittsburgh, United States, 6University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: In anterior cruciate ligament (ACL) reconstruction, notchplasty is frequently performed by many surgeons, and it is not well understood how notchplasty may effect tunnel placement and knee kinematics. The purpose of this study is to investigate the biomechanical effects of notchplasty on the ACL reconstructed knee. The hypothesis is that notchplasty will affect knee biomechanics in ACL reconstructed knee. Methods: Twelve (n = 12) mature, fresh frozen porcine knees were tested using a robotic testing system. All ACL reconstructions were performed via arthroscopy using a three-portal technique. Four knee states were compared: (1) intact ACL (2) ACL deficient, (3) anatomic single bundle (SB) ACL reconstruction and (4) anatomic SB ACL reconstruction with a 5 mm notchplasty. The same tunnel was used for states 3 and 4. The graft was tensioned at 80 N and fixed at 60 (full extension of the porcine knee is 30) of knee flexion with an Endobutton on the femur and a post-screw on the tibia. All knees were subjected to the following loading conditions: an 89-N anterior tibial load to test anterior tibial translation (ATT) and a 4 N-m internal rotation (IR) and external rotation (ER) tibial torques. The knee kinematics and in situ force obtained from the different knee conditions were compared. The data (mean ± standard deviation), was analyzed using the statistical software (Prism GraphPad Version 5.0a) using a two-tailed Paired t test with the level of significance set at p \ 0.05. Results: There was no significant differences between pre and post notchplasty in ER at 30 (full extension in porcine knee) and 60 of flexion (p [ 0.05). However, a significant difference was found between pre and post notchplasty in ATT at 30 and 60 of flexion (p \ 0.05) (Fig. 1). In situ force in anatomic single bundle reconstruction with notchplasty was significant lower than intact and anatomic reconstructed ACL pre-notchplasty at 30, 60 and 90 flexion of the knee (p \ 0.05). In response to an IR tibial torque, there were not significant differences between pre and post notchplasty at 30 of knee flexion (p [ 0.05). However, a significant difference was found between pre and post notchplasty in IR at 60 (p \ 0.05) of knee flexion. Conclusions: The results of this study showed that notchplasty did not result in an increase in ER at 30 and 60 of knee flexion and IR at 30 degree of knee flexion, but lead to an increase in ATT at 30 and 60 of knee flexion and in IR at 60 of knee flexion. Notchplasty had more effect on anterior stability than rotational stability. This change in knee kinematics could be detrimental to a bone healing and graft ligamentization and could lead to graft failure.

S185 P20-862 Histological study of the posterolateral bundle after a partial tear of the anterior cruciate ligament N. Pujol1, C. Bazille2, C. Hulet3, P. Colombet4, P. Djian5, French Arthroscopic Society 1 Hopital Andre Mignot, Orthopaedic Department, Le Chesnay, France, 2Caen University Hospital, Department of Pathology, Caen, France, 3Caen University Hospital, Orthopaedic Department, Caen, France, 4Centre de Chirurgie Orthope´dique et Sportive de, BordeauxMe´rignac, Bordeaux Me´rignac, France, 5Cabinet Goethe, Paris, France Objectives: Anatomic double-bundle reconstruction in anterior cruciate ligament (ACL) tears has been developed during the last few years. The ACL consists of two major fiber bundles, the anteromedial (AM) and posterolateral (PL) bundles. Although disagreement exists among arthroscopic surgeons about the occurrence of isolated ruptures of the AM or PL bundle, some arthroscopic studies have documented their occurrence but without any histological analysis of the intact bundle. The aim of the study was to evaluate the healing response, innervation and femoral attachment of an intact remnant PL bundle after partial tear. Methods: Twenty-six PL bundles were harvested from patients having a partial ACL tear, before undergoing a standard single-bundle reconstruction. Formalin-fixed and paraffin-embedded sections were analyzed after hematoxylin an eosin stain and using immunohistochemistry techniques with actin and PS100 antibodies. Results: In 12 cases (46%), no significantly healing response was detected. Fourteen PL bundles were characterized by increasing cell number density and neovascularization of 20–80% of bundle surface with numerous actin-immunostained cells. A normal femoral attachment was present in 5 cases (19%). Among these cases, healing response was variable from no significant to major. Numerous free nerve endings were present in repaired-area in all cases. Conclusions: Histological findings of partial tear of ACL retrieved 46% of normal bundles, 19% with normal femoral attachment. This analysis confirms that partial tear does really exist. However 54% of bundles showed variably healing responses suggesting a mechanism of elongation without rupture of the PL bundle. Moreover, neural elements are still present, encouraging surgical reconstruction of the ruptured bundle while preserving the intact bundle, when possible. This technique has potential advantages over current ACL reconstruction techniques, including the preservation of the complex attachment sites and innervations of these structures.

P20-901 Elongations and yield loads of four tibial fixation systems. Comparative analysis in hamstring ACL reconstruction H.E. Robert1, M. Collette2, H. Lanternier3, X. Cassard4, M. Bowen5, T. De Polignac6 1 CH North Mayenne, Mayenne, France, 2Clinique Edith Cavell, Bruxelles, Belgium, 3Polyclinique de l’Europe, Saint Nazaire, France, 4 Clinique des Ce`dres, Cornebarrieu, France, 5Northwestern Memorial Hospital, Chicago, United States, 6Clinique Ge´ne´rale, Annecy, France Objectives: The biomechanical strength of hamstring tendon autograft anterior cruciate ligament (ACL) reconstruction is one possible determinant of outcome. The relaxation of tension in the tendon to bone fixation can result in residual knee laxity. The aim of the study was to measure the elongations and yield strengths of four commonly used hamstring graft tibial fixation devices. Methods: Eight adult porcine tibias for each of four devices were used to fix quadrupled human semitendinosus tendons. The four tibial devices tested were: bioabsorbable Delta screws (diameter 9 mm,

123

S186

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

length 40 mm) (Arthrex), WasherLoc device (Biomet), TightRope Reverse (Arthrex), and TLS PEEK screws (FH Orthopedics). Grafts fixed with TLS screws and Tightrope Reverse were pretensioned with 500 and 90 N, respectively. The free end of each graft was connected to a tensile testing machine (MTS Bionix II) by a pulley. Each construct was subjected to cyclic loading (1,000 cycles from 70 to 220 N at a speed of 80 cycles/min) and then loaded to ultimate failure, according to the Coleridge and Amis protocol (Knee Surg Sports Traumatol Arthrosc 2004; 12: 391–397). The elongations (graft slippage + tendon elongation) and the yield loads were recorded. Groups were compared by the Kruskal–Wallis one-way analysis of variance non parametric method. The result was considered statistically with the p value for significance set at 0.05.

Elongation (mm) Mean ± SD

Yield load (N) Mean ± SD

Delta screw

3.31 ± 11.2

841 ± 392

WasherLoc

3.59 ± 2.6

511 ± 95

TightRope reverse

3.91 ± 1.39

561 ± 113

TLS PEEK screw

0.25 ± 0.36

1,015 ± 129

Results: The elongation was significantly lower for the TLS system (p \ 0.05). Elongation was the summation of the viscoelastic behavior of non preconditioned tendons, the graft slippage at the fixation and the tensioning of the interfaces. TLS provided significantly higher yield load (p \ 0.05) than WasherLoc and TightRope Reverse but not for the Delta screw (p [ 0.05). Conclusions: The combination of preconditioning and rigid fixation is necessary to maximize biomechanical strength at time zero for hamstring ACL reconstruction.

P20-922 Histological basis for improving outcome after anatomic acl reconstruction by preserving the remnant stump R. Prejbeanu1, H. Haragus2, D. Vermesan1, D. Crisan1 1 University of Medicine and Pharmacy Timisoara, Timisoara, Romania, 2Emergency Clinical County Hospital, I-st Clinic of Orthopedics and Trauma, Timisoara, Romania Objectives: The lengthy period of ligamentisation and revitalization of the tendinous neoligament is responsible for the very long period needed to complete recovery—minimum of 6 months. Numerous attempts are made in the direction of improving healing after ACL reconstruction. We aimed to investigate if there is a histological basis for improving outcome after ACL reconstruction by preservation of the remnant stump on the tibial insertion. Methods: We prospectively followed 116 knees with primary anatomical single bundle hamstrings ACL reconstructions. One group (56 knees) had the entire injured ACL removed from the tibial insertion and for the other we kept as much as possible from the remnant stump and shaped only to prevent impingement. From the remnant stump (60 knees) we selected a subgroup of which 11 patients together with 8 matched for age and gender from the stump removal group agreed to participate in a substudy with second look arthroscopy at 9–12 months and histological examination of the biopsied neoligaments. Two biopsies (proximal and distal halfs) were collected from each anteromedial fascicles of the neoligaments that were histologicaly analyzed for nerve endings and receptors based on

123

immunohistological tests using monoclonal antibodies to S-100 and NFP. Additional we collected The Cincinnati Knee Rating Score, KT1000 measurements, knee extension and return to recreational activities were assessed at 1, 6 and 12 months postoperatively for all patients. Results: Data analysis found differences between the two groups for the Cincinnati Knee Rating Score (paired t test p = 0.0001) and return to recreational activity favoring preservation of the stump at 6 and 12 months which were supported by result from subgroup data which show increased number of nerve endings and receptors in the histological analysis (mean increase of 29% per field). Conclusions: The ACL is, as are all important ligaments, important in defining proprioception and body spatial positioning using mechanoreceptors and neural endings. The role of this structures have only recently been credited to their full potential and importance in recovery, prevention of reinjury and full return of body kinematics. Our data supports the preservation of the remnant stump on the tibial insertion to surround the base of the neoligament, however we need more patients for statistical significance and improving accuracy in analyzing of biopsies.

P20-952 Immuno-histo-chemical properties of the ruptured ACL graft: lacking re-innervation, a cause of graft failure? P.M. Tscholl1, R.M. Biedert2, M. Moschopulos3, S. Edouard4 1 Kantonsspital Winterthur, Chirurgische Klinik, Winterthur, Switzerland, 2Sportclinic Villa Linde, Biel, Switzerland, 3Pathologie La¨nggasse Bern, Bern, Switzerland, 4Promed SA, Laboratoire me´dical, Marly, Switzerland Objectives: ACL re-tear is multifactorial in its origin. Trauma is nearly as frequent as surgical causes due to tunnel mal-positioning, graft type and tensioning or failure of fixation devices. Failure due to biological factors are described in the ligamentization process or osseous graft integration and are only partially understood. Innervation of the graft has been shown in several animal and human studies to be re-established in form of sensibility and the ACL-hamstring reflex. However, its role in graft rupture is unknown. Methods: 20 ruptured ACL grafts from 17 patients (28.8 ± 8.2 years) where analyzed immuno-histo-chemically (HE and S-100) for re-vascularisation, collagen orientation and neural structures. ACL re-rupture occurred on average 5.6 years (range 0.3–16.3) after primary, anatomical ACL reconstruction. All re-tear were intraligamentous. Trauma mechanism was categorized in trauma energy and contact versus non-contact injury. Results: Nerve fibres were inexistent in 17 grafts, 3 grafts showed only little signs of re-innervation. Only one patient without graft re-innervation suffered low energy contact injury, which however occurred 4 months after reconstruction. All others suffered low energy, noncontact injury while landing on jump tasks or running for example. Two out of the three patients with graft re-innervation suffered highenergy contact injury. The third graft with re-innervation was reconstructed more than 16 years earlier. No significant difference was found for time between primary surgery and re-rupture, graft choice, and time between re-rupture and biopsy with regard to re-innervation. Conclusions: In this cohort, patients without having re-established innervation of their ACL graft suffered mainly low-energy and noncontact injuries, whereas the patients with re-innervation suffered high-energy contact traumas. Lacking re-innervation of the ACL-graft may therefore be a leading factor in ACL re-tear with inadequate trauma due to missing sensomotory feed-back mechanism such as the ACL-hamstring reflex. Future studies will have to show, whether this factor may be influenced by either surgery or rehabilitation methods.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P20-977 Flexible drilling of femoral tunnels for ACL reconstruction: must we still hyperflex the knee? Y.H.D. Lee1, J. Nyland2, D. Caborn2 1 Changi General Hospital, Singapore, Singapore, 2University of Louisville, Orthopaedic Surgery, Louisville, United States Objectives: The purpose of this study was to understand differences in the use of standard wires and flexible wires for femoral tunnel drilling via the anteromedial portal in ACL Reconstruction. We wanted to study the tunnel length and exit point at 90 and 120 of knee flexion. We wanted to study if hyperflexion of the knee is required with the use of flexible wires. Methods: We used eight cadaveric knees for this study. We drilled straight and flexible guide wires sequentially at 90 and 120 knee flexion—4 measurements for each cadaver specimen. The intraosseous lengths and exit points of the wire on the lateral femur were documented. Results: We found that knee flexion at 120 produces a longer intraosseous femoral tunnel length when drilling with both the straight or flexible wire. (both p \ 0.05) The use of a flexible wire helped achieve a longer average length than a straight wire at both 90 and 120. However this value did not reach statistical significance. (p = 0.15, p = 0.81 respectively). Using a flexible wire, the exit point is brought away from the posterior border of the femur, reducing the risk of posterior wall blowout. This was seen at both 90 and 120 of knee flexion (both p \ 0.05). It is important to note that for the group with the flexible drilled at 90, the exit point was 8 mm or more above the posterior cortex in all 8 specimens. Therefore the use of flexible wires at 90 has a low risk of posterior wall blowout. Increasing knee flexion and using flexible wires, the exit point is brought higher from the posterior border, average of 19 mm away (p \ 0.05) This creates a larger margin of safety in preventing posterior wall blowout and damage to the lateral structures of the knee as the wires exits. Conclusions: The use of flexible reamers and curved guides can reduce the risk of posterior wall blowout when drilling through the anteromedial portal at 90 of knee flexion but, we feel that increasing knee flexion to 120 with flexible wire use increases the margin of safety in preventing blowouts and injury to posterolateral knee structures.

P20-1027 Knee medial collateral ligament injuries R.H. Alonso1, H. Taddeo2 1 ARST, Wilde, Argentina, 2ARST, Banfield, Argentina Objectives: The goal of this study was to document the frequency and location of Workers’ knee medial collateral ligament injuries which in our view were surgical patients, and show our technique to solve them. In common practice grade III MCL rupture can be treated nonoperatively with some residual laxity tolerated by the patient, but in Workers’ Compensation claimers patients is not because of their more demanding. For this reason when in valgo stress testing we find not the ‘‘end point’’ present in the contra lateral knee we perform a surgical repair. In our first 15 surgical cases despite the improvement in the instability, the end point result was not desired, since technically we worked especially in the superficial leaf (sMCL). The arthroscopic finds in where we saw disruption of the deep leaf (dMCL) give us the idea of attack that point in surgery. Methods: 2,500 Knee surgeries were performed from October 1996 to October 2009. 162 Patients in which was operated the MCL were included. Technique: The first step is Arthroscopic to assess and treat associated injuries and identify if the dMCL injury is meniscofemoral (MF) or

S187 meniscotibial (MT). In acute injuries physical examination and MRI oriented about where is the superficial injury and towards this sector extends the surgical approach. If deep rupture is at that level is exposed and if it does not match, the superficial layer must be opened to see the wound of the deep region and the meniscal ligament junction. In chronic injuries the superficial leaf is opened proximal or distal where the arthroscopy indicates is the deep leaf injury. The key is the first knot. Then tied it must obtain the ‘‘end point’’ at the physical exam. It starts on the outside of the sMCL far (proximal or distal as appropriate) of the meniscus level, then takes the dMCL, the meniscus, and turns in the opposite direction, and must leave in the superficial layer in the lip opposite to where it entered. If the joint is not stable with it, it must be done again. In chronic cases the superficial layer healed elongated, we interweave one lip below the other tensing the superficial leaf and the posterior oblique ligament (POL), and then carried out several knots of strengthening between edges. Results: In 2,500 workers’ knee surgeries we have practiced 162 grade III MCL repairs, with a mean age of 34.7 years (range 19–67), 101 right and 61 left; 155 men and 7 women; 48 above, 111 below, and 3 above-below the meniscus; 125 acute and 37 chronic, the average of treatment was 157.63 days, 15.82 days from first assist to surgery and 141.8 days from surgery at discharge. Conclusions: Our results with conservative treatment in Workers’ Compensation claimers patients were not satisfactory. With surgical treatment, and then developing a technique that recovers the tension in the dMCL, we have achieved good objective results in terms of stability with rigid end point, and patients subjectively feel the knee stable.

P20-1053 Evaluation of cellularity and cell viability of the anterior cruciate ligament remnant tissue according to the time of rupture C. Dominguez1, M. Ekdahl1, M. Acevedo1, R. Hernandez1, B. Morales2, J.C. Bustos2 1 Hospital Dipreca. Universidad de Valparaiso, Department of Orthopaedic Surgery, Santiago, Chile, 2Hospital Dipreca. Universidad de Valparaiso, Department of Pathology and Cell Biology, Santiago, Chile Objectives: Previous studies have shown that the importance of the anterior cruciate ligament (ACL) it is not only mechanical, it also has influence over the proprioception of the knee and the remnants

Fig. 1 No. of cells in tibial, femoral and control cultures (p [ 0.05)

123

S188

Fig. 2

Fig. 3 contribute to the revascularization process of the graft. In the last time many studies have reported benefits on preserving the remnant tissue during ACL reconstruction surgery. Our purpose was to evaluate the cellularity and cell viability of the ACL remnant tissue, analyzing differences of these parameters according to the time of the rupture and the location were the ACL sample was taken: femoral or tibial footprint. Methods: We conducted a histological study. Standardized tissue samples of femoral and tibial ACL remnants were harvested from 11 patients undergoing ACL reconstruction, 1 unrupted ACL was taken from a total knee arthroplasty (TKA) patient as a control. The patients were placed into four groups. (1) 0–6 weeks, (2) 6 weeks to 6 months, (3) more than 6 months, and (4) control. In accordance with the time following ACL rupture. For histology and immunohistochemistry, tissue samples were incubated for 8–12 h in a 0.1% collagenase solution at 37C. After this enzymatic digestion, the cells

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 were isolated by centrifugation and tainted with 0.4% Blue Triptan to make the initial cell count. Finally this pellet was cultured at 37C with 5% CO2 in a Dulbecco’s Modified Eagle Medium (DMEM) with 10% fetal bovine serum solution. To evaluate the cell viability we performed an MTT assay measured on day 5 and 10 making growth charts for each sample. The t test for unpaired samples was used to compare mean values and measure correlation. A p value of \0.05 was considered significant. Results: The presence of fusiform cells (fibroblasts) were noticed at the initial count, regardless of the time of the rupture without significant differences in between groups and between these and the control (p [ 0.05) (Fig. 1). The cell viability measured with the MTT assay showed significant proliferation of the cultures with an increase of the cell population in all samples, achieving almost seven times the number of the initial count, again there were no significant differences between groups (p [ 0.05) (Fig. 1). Conclusions: The presence of viable cells were noticed in all cultures independently of the time of the rupture and there were strong cell replication showing that the ACL remnants have viable tissue through time. These results supports the concept of a more biologic ACL reconstruction where preserving native tissue may have benefits on the graft revascularization and healing (Figs. 2, 3).

P20-1061 Evaluation of femoral tunnel exit points in three different tunnel drilling techniques for anatomical double bundle ACL reconstruction M. Nozaki1, M. Kobayashi1, H. Goto1, Y. Nishimori1, A. Murase2, T. Otsuka1 1 Nagoya City University, Orthopaedic Surgery, Nagoya, Japan, 2 Nagoya City University, Nagoya, Japan Objectives: Several studies have shown the biomechanical advantage of anatomical double-bundle (DB) anterior cruciate ligament (ACL) reconstruction. Although the trans-tibial drilling technique is the most common technique in ACL reconstruction, cadaveric studies using this technique have shown that the femoral tunnel tends to be placed high in the notch, outside the anatomical insertion site. Therefore, many surgeons have attempted to create femoral tunnels through the far anteromedial (FM) portal. However, this FM portal drilling technique has the potential risk of posterior blowout of the lateral femoral condyle. The purpose of this study was to evaluate the femoral tunnel exit point in three different drilling techniques for anatomical DB ACL reconstruction. Methods: 10 cadaveric knees (3 male and 7 female, mean age, 86.1 years) without ligament injury or significant arthritis were used for this study. The ACL was identified and dissected from both tibial and femoral insertions, leaving 1- to 2-mm of soft tissue at the footprints. The central points of each anteromedial bundle (AMB) and posterolateral bundle (PLB) in the tibial and femoral footprints were marked. After marking the center of each bundle, a guide pin was drilled through the center of tibial AMB and PLB insertion site and FM portal, in an attempt to reach the center of the femoral AMB insertion site. The exit point of the guide pin on the lateral side of lateral femoral condyle was documented at 90 and 120 knee flexion. The same technique was used for the PLB. Statistical analysis was performed using ANOVA. Statistical significance was defined as P \ .05. Results: The exit point of the guide pin drilled through the FM portal to the AMB femoral insertion site was in a significantly more posterior position at 90 and 120 knee flexion in comparison to the guide pin drilled through the AMB and PLB tibial insertion site to the AMB femoral insertion site. The guide pin drilled through the FM portal to the PLB femoral insertion site was also in a significantly more

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 posterior position compared with the guide pin drilled through the AMB and PLB tibial insertion site (Table 1). Conclusions: Recent studies have demonstrated the advantage of FM portal drilling in comparison with trans-tibial drilling regarding accuracy of anatomical femoral tunnel placement. Despite the accuracy of the intra-articular femoral tunnel starting point in FM drilling, several studies have reported the potential risk of peroneal nerve injury and posterior blowout. We evaluated the femoral tunnel exit points in three different drilling techniques for anatomical DB ACL reconstruction. Our results demonstrated that the exit point of the femoral tunnel drilled using the FM portal drilling technique tends to be in a posterior position compared to guide pins drilled using the trans-tibial technique. These findings indicate the potential risk of posterior blowout and articular cartilage damage in FM portal drilling.

S189 Table 1 CASE

ACL graft

1…………

Hamstring

Microorganism

Antibiotic

Patellar T.

Staph. Aureus MRSA………

Levo + Rifa

41…………

1…………

8…………

9…………

2…………

6…………

21…………

1…………

6…………

14…………

3…………

6…………

Hamstring

4………… Amox + ClavAc Linez + Rifa 5…………

Hamstring

Hamstring

6…………

7…………

Patellar T.

Hamstring

8…………

Femoral

90

9…………

120

Drilling

FM

AM tibia

through

PL

FM

AM tibia

tibia

Ratio (%) Femoral

Hamstring

AM

PL

tibia

tibia

41.8 ± 11.6

49.5 ± 9.2

FM

AM

PL

tibia

tibia

65.1 ± 16.7

67.2 ± 15.0

15…………

through Ratio (%)

24.2* ± 14.2

28.6* ± 9.2

Trim/

5…………

Hamstring

SCN………

Hamstring

Cipro + Rifa

33…………

1…………

6…………

16…………

1…………

4…………

23…………

1…………

6…………

7…………

1…………

6…………

Unknown…… SCN………

Hamstring

Hamstring

SCN………

Levo + Rifa

SCN……… Levo + Rifa

FM

6…………

SCN………

6…………

12…………

angle Drilling

6…………

1…………

2…………

14…………

120

1…………

13…………

Staph. Aureus

Femoral PL

90

27…………

2…………

Ciprofloxacine

13…………

Levo + Rifa

15…………

Amox + ClavAc Levo + Rifa

tunnel Knee flex

Hamstring

PL

26.8* ± 15.3 36.7 ± 15.6 49.3 ± 13.6 38.2* ± 9.4 71.7 ± 15.8 66.2 ± 10.3

SCN………

22…………

tibia 11…………

Propinebacterium SCN………

Sulf + Rifa Levo + Rifa

10…………

flex angle

Levo + Rifa

SCN………

Femoral AM

tunnel Knee

Antibiotic therapy duration (weeks)

SCN……… Linezolida

Table 1 The exit point of the guide pin on the lateral femoral condyle

Number of lavages

Hamstring

2…………

3…………

Days to first lavage

Hamstring

SCN………

43…………

1…………

50…………

1…………

6………… 6…………

Levofloxacin

49…………

1…………

6……………

Table 2 * Significant difference compared with AM tibia and PL tibia In the value of ratio, the most anterior position on the lateral femoral condyle was defined as 100% and the most posterior position was defined as 0%

P20-1103 Knee joint infection after acl reconstruction; management and mid-term functional outcomes R. Torres-Claramunt1, J. Erquicia2, X. Pelfort1, P.E. Gelber3, P. Hinarejos1, J.C. Monllau3 1 Parc de Salut Mar. Universitat Auto`noma de Barcelona, Orthopedic Surgery, Barcelona, Spain, 2Institut Universitari Dexeus, ICATME, Barcelona, Spain, 3Hospital de la Santa Creu i Sant Pau. UAB, Orthopaedic Department, Barcelona, Spain Objectives: Septic arthritis following ACL reconstruction is a rare but serious complication. Our aim is to report the prevalence and management of this condition as well as to assess its mid-term functional outcomes. Methods: A retrospective analysis of a consecutive series 810 patients who underwent arthroscopically assisted ACL reconstruction of knee joint infections between 2006 and 2009 in two hospitals by the same surgical team occurring was performed. Criteria of septic arthritis were patients with signs and symptoms of joint infection along with positive blood test (ESR, CPR) and synovial positive culture. All the patients were treated with parenteral antibiotics followed by oral antibiotics, according to antibiogram, and at least one arthroscopic debridement. The mean follow-up was 39.3 (range 15–58) months. Follow-up included X-ray, KT-1000 arthrometric evaluation, the IKDC forms and Lysholm score.

CASE

Follow-up (months)

Range of motion

Lachman

Pivot Shift

Lysholm

IKDC

KT-1000 (difference)

1…………

49…………

0/0/130

+……………

-……………

74…………

69…………

1…………

2…………

51…………

0/0/135

-……………

-……………

85…………

77…………

1…………

3…………

54…………

0/0/140

-……………

+……………

45…………

20.7………

-0.5………

4…………

36…………

0/0/130

-……………

-……………

79…………

83…………

-1…………

5…………

15…………

5/0/120

+……………

-……………

62…………

56.3………

-3.5………

6…………

58…………

0/0/145

-……………

-……………

98…………

86.2………

1.5…………

7…………

30…………

0/0/135

-……………

+……………

80…………

84…………

0.5…………

8…………

22…………

0/0/130

+……………

+……………

85…………

88.5………

1.5…………

9…………

35…………

0/0/130

-……………

-……………

92…………

86…………

3…………

10…………

48…………

0/0/130

+……………

-……………

83…………

75…………

3…………

11…………

33…………

0/0/125

-……………

+……………

68…………

61…………

2…………

12…………

34…………

0/0/120

-……………

-……………

100…………

96…………

2…………

13…………

39…………

0/0/120

-……………

-……………

85…………

52…………

1…………

14…………

27…………

0/0/130

-……………

-……………

76…………

71…………

5…………

15…………

20…………

0/0/100

-……………

-……………

54…………

51…………

3…………

Results: The prevalence of postoperative septic arthritis in this series was 1.85% (15 patients). The treatment of infection was successful in all cases. All but one patients healed with the treatment protocol and retained their reconstructed ACL. Microbiology showed that

123

S190 Coagulase-Negative Staphylococcus was present in 10 cases, Staphylococcus Aureus in 2 cases, Meticiline-Resistent Staphylococcus Aureus and Propinebacterium in 1 case each. Finally, in one case the micro-organism could not be identified (Table 1). The time elapsed from ACL reconstruction to the first lavage was 24.9 days (14 SD), and the mean lavages performed were 1.33 (SD 0.6). Regarding to functional outcomes, at the final follow-up, the mean Lysholm score was 77.73 (15.34 SD), the IKDC score was 70.45 (19.56 SD) and the mean difference between the KT-1000 compared to the non-injured contralateral knee was 1.3 mm (2 SD) (Table 2). Conclusions: Early arthroscopic lavage, repeated if necessary, along with adequate antibiotic regimen was a reliable treatment and permitted to retain most of the grafts. There were no recurrences of septic arthritis or bone infection afterwards. Patients with retained ligament showed good mid-term functional outcomes similar to those who do not suffered a joint infection.

P20-1105 Events leading to anterior cruciate ligament injury in world cup alpine skiing: a systematic video analysis of 20 cases T. Bere1, T. Flørenes1, T. Krosshaug2, L. Nordsletten1, R. Bahr1 1 Oslo Sports Trauma Research Center, Norwegian School of Sports Sciences, Oslo, Norway, 2Norwegian School of Sport Sciences, Oslo Sports Trauma Research Center, Oslo, Norway Objectives: We have recently identified 3 main mechanisms for anterior cruciate ligament injuries among World Cup alpine skiers, termed ‘‘the slip-catch’’, ‘‘landing back-weighted’’ and ‘‘the dynamic snowplow’’ mechanisms. However, for a more complete understanding of how these injuries occur, a description of the events leading to the injury situations is also needed. Therefore, the objective of this study was to describe the skiing situation leading to anterior cruciate ligament injuries in World Cup alpine skiing. Methods: Twenty cases of anterior cruciate ligament injuries reported through the International Ski Federation Injury Surveillance System for 3 consecutive World Cup seasons (2006–2009) were obtained on video. Ten experts (9 World Cup coaches, 1 former World Cup athlete) performed visual analyses of each case to describe in their own words factors they thought may have contributed to the injury situation related to different pre-defined categories: (1) skier technique, (2) skier strategy, (3) equipment, (4) speed & course setting, (5) visibility, snow & piste conditions and (6) any other factors. Results: Factors related to 3 of the categories, skier technique, skier strategy, and visibility, snow & piste conditions, were assumed to be the main contributors to the injury situations. These categories accounted for 37, 23 and 21% of the total number of statements made by the coaches, respectively. Skier errors, technical mistakes and inappropriate tactical choices, were the dominant factors. In addition, bumpy conditions, aggressive snow, reduced visibility and course difficulties were assumed to contribute. Factors identified related to the skier technique and skier strategy differed between each of the injury mechanisms, the slip-catch, landing backweighted and the dynamic snowplow. For the slip-catch mechanism (n = 10), the skier did not manage to absorb terrain changes prior to injury (n = 5), did not absorb changes in the rhythm of the course set (n = 2) or initiated the turn too early (n = 3). In all these cases, the skier came to be out of balance backward/inward, losing pressure on the outer ski which caught the snow surface abruptly at the time of injury. For the landing back-weighted mechanism (n = 4), the skier was in a backward position into a jump and made incorrect tactical decisions at take-off, which resulted in an uncontrolled flight with subsequent landing on the ski tails. For the dynamic snowplow mechanism (n = 3), the skier had too straight a line into a downhill

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 turn (n = 2) or initiated the turn too early (n = 1), leading to an unbalances position. Thus, the skier was not able to react to sudden changes in snow and piste conditions and ended up in a snowplow position with inappropriate pressure on the inside edge at the time of injury. Conclusions: Based on this systematic video analysis of 20 injury cases, specific skier errors and factors related to race conditions were identified as main contributor leading to anterior cruciate ligament injuries.

P20-1225 Gait analysis after anterior cruciate ligament (ACL) reconstruction in patients aged 50 years and older S.S. Sarmah1, T. Fayad1, J.M. Pammer1, R. Patel1, F.S. Haddad1 1 University College London Hospitals, Trauma and Orthopaedics, London, United Kingdom Objectives: The objective of this study was to evaluate different gait parameters i.e. cadence, speed, stride length, single and double leg support (SLS/DLS) and energy expenditure (EE) to assess functional outcome in middle-aged group of patients with ACL reconstruction. Methods: We analyzed gait parameters in 40 patients aged over 50 years and older prospectively with the Intelligent Device for Energy Expenditure and Activity (IDEEA, MiniSun, CA) accelerometer as part of our long term follow up assessment. Patients performed activities such as walking in different speed, ascending and descending stairs for 10 min in the follow up clinic. Speed, cadence, stride length and SLS/DLS were compared with the non-operative leg of the same patient and EE was compared with 25 healthy volunteers (control group) who performed the same routine. IKDC, TegnerLysholm and SF 12 scores were collected simultaneously. Results: The mean age of the patients and the control groups were 54.7 years and 29.4 years respectively and the mean follow up period 3.8 years (2.4–6 years). We found no statistically significant difference between cadence, stride length, speed and SLS/DLS between the operated and non-operated leg; p. 0.36, p. 0.62, p. 0.81 and p. 0.06 respectively on unpaired t test. Similarly, no difference was found comparing the overall EE during the gait cycle between the patients and the control group (p. 0.55). The mean IKDC, Tegner-Lysholm and SF 12 score were 86.1 and 87.5 and 55.7/49.5 (PCS/MCS) respectively. Conclusions: We found good functional outcome post ACL reconstruction in middle-aged patients by performing gait analysis. Our results also show that ACL reconstruction surgery is beneficial in patients with high activity level irrespective of their age.

P20-1281 Functional hop test, kinematic and isokinetic asymmetries persist 6–9 months after ACL reconstruction: mechanisms and clinical implications S.A. Xergia1, E. Pappas2, F. Zampeli1, S. Georgiou1, A.D. Georgoulis1 1 Orthopaedic Sports Medicine Center, University of Ioannina, Department of Orthopaedic Surgery, Ioannina, Greece, 2Division of Physical Therapy, Long Island University-Brookly, Brooklyn, United States Objectives: To determine asymmetry in hop tests and possible muscular and kinematic mechanisms responsible for it at the return to sports phase in anterior cruciate ligament reconstructed (ACLR) patients. Methods: 22 men with ACLR (mean age ± SD, 28.8 ± 11.2 years) at 6–9 months (mean time after ACLR ± SD, 7.01 ± 0.93) after surgery and 22 matched healthy men (mean age ± SD, 24.8 ± 9.1 years) participated. All participants completed self-assessment of

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 knee function and performance tools (subjective form of IKDC, Lysholm & Tegner) isokinetic testing and functional assessment (single, triple, and crossover hop tests). MANOVAs were used to test for differences between the ACLR group and the control group and between the two legs of the ACLR group. Results: Compared to the control group, the ACLR group had higher isokinetic asymmetry for knee extension at all speeds (p \ 0.001) and higher distance asymmetry for all three hop tests (p \ 0.001). The ACLR group also exhibited lower extremity joint angle flexion at propulsion (triple, p = 0.003; crossover, p = 0.027,) and landing (single hop, p = 0.032; triple hop, p = 0.002; crossover, p = 0.003). Similar differences were also found in the propulsion phase of the single (p = 0.005) and crossover hop (p = 0.017) between the control reconstructed legs of the ACLR group. Conclusions: ACLR patients demonstrate clear functional hop and isokinetic knee extension deficits as well as kinematic deficits with the lower extremity joints not flexing enough during the propulsion and landing phases of the hop tests at 6–9 months after surgery although the Lysholm and IKDC scores are largely normal.

Knee-ACL III

P20-1334 Femoral fixation strength of device-free isoanatomical bone-patellar tendon-bone single-bundle ACL reconstruction: a biomechanical study L.D. Burger1, M. Arnold1, P. Meyer1, B. Goepfert2, D. Wirz2, M.T. Hirschmann1 1 Kantonsspital Bruderholz, Klinik f. Orthop. Chirurgie & Traumatologie, Bruderholz, Switzerland, 2University of Basel, Laboratory of Biomechanics & Biocalorimetry, Basel, Switzerland Objectives: Implant-free fixation methods such as the A3B-technique are increasingly recognized. In this technique the femoral fixation is achieved using a conoid bone block into a similar shaped tunnel. The purpose of this study was to investigate the graft slippage and strength of the femoral press-fit fixation under cyclic loading and ultimate pull out force. We hypothesized that the press fit fixation will show at least equal or higher ultimate strength of fixation as the published results for fixation with interference screws. Methods: Nine fresh frozen knees were obtained from a tissue bank (Life Legacy Foundation, Tuscon, AZ). After thawing, the skin of the specimens was removed and harvesting of the B-PT-B graft was performed in standardized manner. The knees were stripped of all soft tissue, the femora cemented into a steel rod and fixed against rotational motion using a screw. The femoral tunnel was obtained in anteromedial bundle position. The grafts were then inserted into the femoral tunnel and the steel rods fixed in a MTS 858 Bionix testing machine (MTS Corp, Minneapolis, MN). The orientation of the femoral tunnel was approximately 30 to the horizontal plane, representing a natural angulation of the knee during walking. The tibial end of the graft was fixed by a steel-rod-rope construction and frozen by a cryocuff. Optical tracking markers were adjusted to the femora at both ends of the tendon as well as on the bone-block itself. A motion capture system, consisting of 7 Vicon cameras, was used to assess micro-motion of the bone, plastic deformation and slippage of the graft. The specimens were then put under cyclic tensile loading (1,000 cycles alternating between 70 and 220 N of force at 1 Hz). Finally the ultimate load to failure of the bonetendon-bone graft complex was measured at 1 m/min. The graft was either torn out of the femoral tunnel or the patellar tendon ruptured. The sample size was calculated according to existing biomechanical studies.

S191 Results: The implantation and testing was feasible in all nine specimens. There was no movement of the specimens during cyclic loading. The mean pull out force of all grafts was 806 N ± 295 N, (Median 852 N, Range 448 N–1,349 N). The knees which showed elongation of the tendon before graft failure, had higher terminal pull out force than the knees which showed no elongation (Graft failure n = 6, 860 N ± 319 N, vs. block pullout n = 3, 684 N ± 137 N). A correlation of maximal pull out force and size as well as gender of the specimens was observed. The failure pattern differed between the larger and smaller knees- while the bone-blocks in the smaller specimens tended to slip out of the tunnel at lower loads, in the larger specimens the tendon ruptured at higher loads. Conclusions: The investigated femoral press-fit implant free fixation method showed excellent primary stability with pull out force at least equal to the published results for interference-screw fixations, which are around 500–600 N.

P20-1351 Impact of meniscal and cartilage status on long term results of ACL reconstruction. 15–20 year follow-up results P. Lukasik1, W. Widuchowski1, J. Widuchowski1, R. Faltus1 1 District Hospital of Orthopaedics and Trauma Surgery, Piekary Slaskie, Poland Objectives: The purpose of our study was to review the long term results of ACL reconstruction using the BTB autograft with reference to time of surgery, meniscus and cartilage status at the time of surgery. Methods: Between 1990 and 1995 we performed 627 ACL reconstructions using BTB graft. 608 patients (364 m. and 244 f., the average age 24.6 years) were included into the study. With a mean follow-up of 17.4 years all 608 patients were examined clinically, radiologically and evaluated with internationally accepted knee scales (Lysholm, IKDC, Tegner). The degree of osteoarthritis was classified according to Kellgren and Lawrence scale. We categorized all patients into 4 groups, based on the status of the menisci at the time of surgery: M0—both menisci present, MM or ML—medial or lateral meniscus partially removed and MB—both menisci partially removed. The data were further analyzed within these 4 primary groups according to the articular cartilage status at the time of surgery. We divided all patients into two groups: ‘‘normal’’—2 or less grade articular cartilage (according to Outerbridge’s scale) and ‘‘abnormal’’—grade 3 or 4 in any compartment. Results: 1. Lysholm. According to Lysholm scale we achieved the best results in MO group. The cartilage defects did not affect the final results (mean result 87 p. in ‘‘normal’’ group and 85 p. in ‘‘abnormal’’ group). The worst score was observed in MB group (72 p. in ‘‘normal’’ and 70 p. in ‘‘abnormal’’ group). 2. IKDC. The IKDC scale indicated the best results in MO group. The cartilage defects also did not affect the final results (85% normal or nearly normal results in ‘‘normal’’ group and 83% in ‘‘abnormal’’ group). The worst score was observed in MB group. 3. Tegner. The sports activity assessed with Tegner scale was the best in MO group (mean score 5.4 in comparison to 6.9 before the injury). The cartilage defects and time of surgery did not affect the final results as well. 4. Osteoarthritis. Preoperatively all patients were classified as 0 (89%) or I (11%) Postoperatively best results were seen in MO and ‘‘normal’’ group. 100% patients had radiological changes classified as 1 or 2. Worse results were seen in ‘‘abnormal’’ group, particularly in MB group (92% grade 2 or more). Conclusions: 1. The cartilage status at the time of surgery did not affect, to a considerable extent, the objective results (according to Lysholm

123

S192 and IKDC scales) in long term observation, but had an influence on the radiological changes in the target knee. 2. The final results after ACL reconstruction, in long term study, mainly depended on meniscal status at the time of surgery. Meniscal lesions treated with meniscectomy at the time of ACL surgery led to poor subjective, objective and radiological results.

P20-1441 Isolated anteromedial and posterolateral ACL bundle augmentation: patient outcome at 1 years J. Robinson1, A. Porteous2, J. Murray3, R. Halliday4, D. Piper4 1 The Bristol Knee Group, Orthopaedic Surgery, Avon Orthopaedic centre, Bristol, United Kingdom, 2Avon Orthropaedic Centre, Southmead Hospital, Bristol, United Kingdom, 3The Bristol Knee Group, Bristol, United Kingdom, 4University of Bristol, Southmead Hospital, Bristol, United Kingdom Objectives: Separation of the ACL into anteromedial (AM) and posterolateral (PL) fibre bundles has been widely accepted. The bundles act synergistically to restrain anterior laxity throughout knee flexion, with the PL bundle providing the more important restraint near extension and its obliquity better restraining tibial rotational laxity. 10% of ACL injuries involve isolated rupture to one of these bundles causing patients to present with instability symptoms or pain. As knowledge about the influence of the ACL bundles on knee kinematics has increased, isolated reconstruction of either PL or AM bundle has been advocated. However only one cohort study of 17 patients has been presented in the clinical literature. Methods: KOOS (Knee Injury and Osteoarthritis Outcome Score) and IKDC (International Knee Documentation Committee Form) scores at 1 years post op were obtained for 12 patients who had undergone isolated ACL augmentation between 2007 and 2009. These were compared with previously published outcome scores for standard ACL reconstruction procedures. In addition examination under anaesthesia (EUA) assessments were analysed to see if a pattern of laxity for isolated AM and PL rupture could be determined. Results: There were 5 patients with isolated AM bundle rupture and 7 with isolated PL bundle rupture. EUA analysis demonstrated that patients with isolated PL bundle rupture had increased pivot shift and Lachman test laxity, whereas the AM bundle rupture group had increased laxity with the anterior drawer test. Compared to previously published IKDC scores, there were no difference between isolated bundle augmentation and standard ACL reconstruction. However the KOOS scores showed significantly increased Sports function scores which was significantly better in the isolated bundle augmentations (93/100 vs. 74/100). Differences between isolated AM and PL bundle reconstructions were not distinguishable. Conclusions: Isolated ACL bundle tears make up a significant proportion ACL injuries. Although technically more difficult than standard ACL reconstruction, isolated bundle augmentation appears to result in improved sports function when compared to standard ACL reconstruction.

Knee-ACL IV

P21-2 Proprioceptive deficits after ACL injury. Are they clinically relevant? A systematic review A. Gokeler1, A. Benjaminse2, T.E. Hewett3, L. Engebretsen4, E. Ageberg5, M.P. Arnold6

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 1

University Medical Center Groningen, Center for Rehabilitation, Groningen, Netherlands, 2University Medical Center Groningen, University of Groningen, Center for Human Movement Sciences, Groningen, Netherlands, 3Sports Medicine Biodynamics Center, Human Performance Laboratory, Cincinnati Children’s Hospital Medical Center, Cincinnati, United States, 4Ullevaal University Hospital, Orthopaedic Center, Oslo Sportstrauma Research Centre, Oslo, Norway, 5Lund University, Health Sciences, Lund, Sweden, 6 Kantonspital Bruderholz, Orthopedic Surgery and Skeletal Traumatology, Basel, Switzerland Objectives: To establish the clinical relevance of proprioceptive deficits reported after anterior cruciate ligament injury (ACL). Methods: A literature search was done in electronic databases from January 1990 to June 2009. Inclusion criteria for studies were ACL-D and ACL-R, articles written in English, Dutch or German and calculation of correlation(s) between proprioception tests and clinical outcome measures. Clinical outcome measures were muscle strength, laxity, hop test, balance, patient reported outcome, objective knee score rating, patient satisfaction or return to sports. Studies included in the review were assessed on their methodological quality. Results: In total 1,161 studies were identified of which 24 met the inclusion criteria. Pooling of all data was not possible due to substantial differences in measurement techniques and data analysis. Most studies failed to perform reliability measurements of the test device used. In general the correlation between proprioception and laxity, balance, hop tests and patient outcome was low. Four studies reported a moderate correlation between proprioception, strength, balance or hop test. Conclusions: There is limited evidence that proprioceptive deficits as detected by commonly used tests adversely affect function in ACL-D and ACL-R patients. Development of new tests to determine the relevant role of the sensorimotor system are needed. These tests should ideally be used as screening test for primary and secondary prevention of ACL injury.

P21-28 Pre-operative factors predicting good outcome in terms of health-related quality of life after anterior cruciate ligament reconstruction O. Ma˚nsson1, J. Kartus1, N. Sernert2 1 NU-Hospital Organization, Orthopedic Department, Uddevalla, Sweden, 2NU-Hospital Organization, Department of Development and Research, Trollha¨ttan, Sweden Objectives: The life situation of many patients changes after an anterior cruciate ligament (ACL) rupture and subsequent reconstruction and this may affect their health-related quality of life in many ways. It is well known that the overall clinical results after ACL reconstruction are considered good, but pre-operative predictive factors for a good post-operative clinical outcome after ACL reconstruction have not been studied in as much detail. The purpose of this study was to identify pre-operative factors that predict a good postoperative outcome as measured by the Short Form 36 (SF-36) and Knee Osteoarthritis Outcome Score (KOOS) 3–6 years after ACL reconstruction. The main hypothesis was that one or more pre-operative factors could predict a good post-operative outcome in terms of health-related quality of life. Methods: Seventy-three patients scheduled for ACL reconstruction were clinically examined pre-operatively. Three to six years after reconstruction, the SF-36 and KOOS questionnaires were sent by mail to these patients. Predictive factors for health-related quality of life were investigated using a stepwise regression analysis. Results: The pre-operative factors; Pivot shift, manual Lachman and type of graft, one-leg-hop test, the Tegner activity level pre-injury and

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 flexion deficit explained up to 25% of the outcome variance in both physical and mental health status regarding health-related quality of life measured with SF-36. Using KOOS femuro-patellar pain, Tegner activity level pre-operatively, Pivot shift test and flexion deficit explained up to 23% of the outcome variance of the symptoms and function. Conclusions: In conclusion, pre-operative factors, such as knee laxity, knee function and range of motion, may predict a good post-operative outcome and explain up to 25% in terms of health-related quality of life after ACL reconstruction. It appears that the patients’ pre-injury and pre-operative Tegner activity level are important predictors of post-operative health-related quality of life.

P21-58 MRI comparison of fibular collateral ligament and patellar tendon length M. Milankov1, R. Semnic2, V. Njagulj3, S. Ninkovic4, V. Harhaji4 1 University of Novi Sad, Clinical Centre Vojvodina, Orthopaedic Surgery and Traumatology, Novi Sad, Serbia, 2University of Novi Sad, Imaging Diagnostic Center, Institute of Oncology, Sremska Kamenica, Serbia, 3University of Novi Sad, Center of Radiology, Clinical Center Vojvodina, Novi Sad, Serbia, 4Clinical Center Vojvodina, Medical School, Department of Orthopaedic Surgery and Traumatology, Novi Sad, Serbia Objectives: The problem of using patellar tendon (PT) auto or allografts for fibular collateral ligament (FCL) reconstruction is the occurrence of PT–FCL mismatch. The length of patellar tendon does not correlate to the length of fibular collateral ligament. Methods: Data collected from magnetic resonance imaging (MRI) included PT and FCL length. A series of 151 patients, 102 men— mean age 30 years (18–54), and 49 women—mean age 34 (18–55), who underwent knee MRI for knee pain evaluation were enrolled in the study. Both patellar tendon and fibular collateral ligament were measured using the same—0.6 mm thickness and 0.6 mm spacing Three-dimensional T2 true FISP (fast imaging with steady state precession) sagittal sequence with the knee in slight flexion. The length of the PT was measured from the patellar apex to the tibial tubercle insertion site, on the system console. In order to visualize FCL insertions precisely, sagittal images were reformatted on the system console according to anatomical—oblique ligament position, in anteriorly tilted, paracoronal plane. Results: The mean PT length was 52.88 ± 7.56 mm (37–75) with a significant difference between men and women. The mean FCL length was 61.21 ± 5.77 mm (46–80) with a significant difference between genders. Average differences between FCL and PT length is 8.38 ± 7.23 mm (-9 to 26) without any significant difference between the genders. In 18 (11.92%) patients PT was longer than FCL; in 7 (4.63%) patients it was equal; and in 126 (83.44%) patients PT was shorter than FCL. Conclusions: The length of patellar tendon does not correlate to the length of fibular collateral ligament. If patellar tendon auto or allograft is used for fibular collateral ligament reconstruction, patellar tendon and fibular collateral ligament lengths must be determined preoperatively in order to avoid ligament length mismatch.

P21-161 Ten year results after anterior cruciate ligament reconstruction with hamstring tendons and accelerated rehabilitation: A prospective clinical and radiological study R.P.A. Janssen1, A. duMe´e1, J. van Valkenburg1, H.A. Sala1, C.M. Tseng2

S193 1

Ma´xima Medical Center, Orthopaedic Surgery and Traumatology, Veldhoven, The Netherlands, 2Ma´xima Medical Center, Radiology, Veldhoven, The Netherlands Objectives: Analysis of long-term clinical and radiological outcome after anterior cruciate ligament reconstruction (ACLR) with special attention to knee osteoarthritis and its predictors. Methods: A prospective, consecutive case series of 100 patients. Arthroscopic transtibial ACLR was performed using quadrupled hamstring tendon autografts with a standardised accelerated rehabilitation protocol. Analysis was preoperatively and 9–11 years postoperatively. Clinical examination included Lysholm and Tegner scores, IKDC, KT-1,000 testing (MEDmetric Co., San Diego, CA) and leg circumference measurements. Radiological evaluation included AP weight bearing X-rays, 45 flexion PA weight-bearing X-rays as well as lateral knee and sky views). Radiological classifications were according to Ahlba¨ck and Kellgren & Lawrence. Statistical analysis included univariable and multivariable logistic regressions. Results: Clinical outcome: A significant improvement (p \ 0.001) between preoperative and postoperative measurements could be demonstrated for the Lysholm and Tegner scores, IKDC patient subjective assessment, KT-1,000 measurements, pivot shift test, IKDC score and one leg hop test. A pivot shift phenomenon (glide) was still present in 50% of patients and correlated with lower levels of activity (p \ 0.022). Radiological outcome: At follow-up, 53.5% of patients had signs of osteoarthritis (OA). In this group, 72% had chondral lesions (Cgrade 2) at time of ACLR. A history of medial meniscectomy before or at time of ACLR increased the risk of knee OA 4 times (95% CI 1.41–11.5). A ICRS grade 3 at time of ACLR increased the risk of knee OA by 5.2 times (95% CI 1.09–24.8). There was no correlation between OA and activity level (Tegner score C 6) nor between OA and a positive pivot shift test. Conclusions: ACLR restored anteroposterior knee stability. Most clinical parameters and patient satisfaction improved significantly. Transtibial hamstring ACLR results in radiological signs of OA in 53.5% of the subjects. Risk factors for OA were meniscectomy prior to, or at time of ACLR and chondral lesions at time of ACLR.

P21-182 Management of postoperative synovitis after ACL reconstructions Z. Knoll1 1 Castlepark Clinic, Tata, Hungary Objectives: We analyzed the reasons of intraarticular bleeding and inflammation following anterior cruciate ligament replacement and demonstrate the results of the routinely performed arthroscopic lavage and those of the postoperative management. Methods: We performed 1,538 anterior cruciate ligament replacements in the period from January 01. 2007 to December 31. 2010. Repeated arthroscopy and lavage was required in 30 cases due to articular swelling and fever; also synovectomy was necessary in some cases. We analyzed the data of these patients. Clinical symptoms and complaints commenced 7–14 (10) days after primary surgery. There were 25 men and 5 women among the patients with an age distribution of 19–41 (34.2) years. Repeated surgery was performed in 3 acute and in 27 chronic cases. In our clinic, we apply several methods for ligament fixation. We analyzed the results of the following implants: 1. bio crosspin (absorbable–absorbable) (Stryker) HA/PLLA 19 cases 2. Rigidfix (absorbable- absorbable) (J,J) (PLA) 1 case 3. Slingshot (J,J) (titanium–absorbable) 3 cases

123

S194 4. 5. 6. 7.

Endobutton-staple (Arthrex) (titanium–titanium) 0 cases Endobutton–absorbable screw (Arthrex) (PLLA) 0 cases Endobutton–KFI screw fixation 0 cases Biocrosspin–Cross-screw (J,J-Stryker) (absorbable–absorbable) (PLLA) 8 cases 8. Endobutton-Millagro (Mitek) (TCP-PLGA) (titanium–absorbable) 0 cases. We analysed our results with IKDC score. Results: During the course of repeated surgery, we used more than 6,000 ml solution for rinsing the joint. Swaps were taken for microbiological evaluation. There were infective agents present in 4 cases out of 30. Following repeated surgery, we applied intravenous Dalacin or Klimycin therapy in each case, which has been modified according to the microbiogram. The operated limb was fixed in a brace applied in extended position for 3 weeks. Repeated surgery (synovectomy performed in a closed manner) was necessary in 3 cases. Conclusions: Based on the data collected from numerous patients who underwent anterior cruciate ligament replacement, the ratio of complications is low in our clinic. There is no correlation between this ratio and sex, age or acute/chronic cases, respectively. The ratio was higher after the use of some specific ligament fixation implants. The postoperative therapy protocol applied in our clinic is suitable for the management of complications. Repeated surgery performed as soon as possible after the onset of the complaints reduces the proliferation of infective agents in the knee joint. Repeated surgery performed as soon as possible is more effective when compared to primarily applied antibiotics. Application of some specific ligament fixation implants poses a higher risk of complication.

P21-224 Partial anterior cruciate ligament tears: anatomic reconstruction versus non anatomic augmentation surgery R. Buda1, F. Vannini1, A. Ruffilli1, M. Cavallo1, A. Parma1, S. Giannini1 1 Istituto Ortopedico Rizzoli, Bologna University, VI Department of Orthopaedics and Traumatology, Bologna, Italy Objectives: Treatment of partial anterior cruciate ligament (ACL) tear requires ACL remnant preservation. Aim of this study is to compare the outcome of an anatomic reconstruction (AR) of the torn bundle with a non anatomic augmentation (NAA) technique using the over the top (OTT) femoral route. Methods: 52 athletes with ACL partial lesion (mean age: 23.3 years) underwent AR (26 patients) or NAA (26 patients). We experienced 2 intra-operative damaging of the healthy bundle (AR group) that required a standard ACL reconstruction. IKDC score, Tegner score and KT evaluation were used pre-operatively and at established FU up to 5 years. Results: IKDC subjective score at FU was 88.2 ± 5.7 (AR group) and 90.2 ± 4.7 (NAA group) (p = 0.186). According to the IKDC objective score at final FU we observed 100% of normal knees in the NAA group against the 87.5% in the AR group (p = 0.179). Anteromedial bundle reconstruction showed significantly lower subjective and objective outcomes compared with postero-lateral bundle reconstruction (p = 0.017). Conclusions: The treatment of ACL partial tears is a demanding surgery. Adapted portals, perfect control of the tunnel drilling process and intercondylar space management are required in AR. The NAA technique showed to be simpler providing better results durable over time with a lower complication rate. Antero-medial bundle reconstruction is associated to a poorer outcome especially when performed with AR.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P21-228 Osteoarthritis after anterior cruciate ligament reconstruction: a meta-analysis S. Claes1, L. Hermie2, R. Verdonk2, J. Bellemans1, P. Verdonk2 1 University Hospitals Leuven, Orthopedic Surgery, Pellenberg, Belgium, 2University Hospital Gent, Orthopedic Surgery, Gent, Belgium Objectives: Osteoarthritis (OA) of the knee is commonly thought to be the near-inevitable long-term consequence of anterior cruciate ligament (ACL) reconstruction, although its exact prevalence remains unknown. As current controversy on the development of OA after ACL reconstruction has been fuelled by rather small and heterogeneous case series and expert opinions, the goal of this study was to systematically review the current literature on long-term radiographic outcome after autologous ACL reconstruction and subsequently perform a meta-analysis to obtain evidence-based prevalences. In addition, this manuscript aims at unravelling the relationship between meniscal status and the occurrence of OA in the ACL reconstructed knee. Methods: A systematic review of the literature was performed in PubMed MEDLINE, EMBASE, and Cochrane Library databases to identify all studies concerning radiographic outcome after autologous ACL reconstruction with a mean follow-up of minimum 10 years. Meta-analyses were performed to obtain the average prevalence of OA and the difference between patients with and without meniscectomy. Considered study estimates were the log-transformed odds and odds ratios, the latter expressing the effect of meniscectomy on OA. Between-study heterogeneity was quantified by the I2 statistic and Cochran’s v2-test and taken into account with a random-effects approach. Results: A total of 16 studies could be withheld for meta-analysis, accounting for 1,554 ACL reconstructions performed between 1978 and 1997. Of these knees, 453 (28%) showed radiological signs of osteoarthritis (IKDC grade C or D). Furthermore, 41.9% of the patients with meniscectomy had osteoarthritis, compared with 18.7% of the patients without meniscectomy. The combined odds ratio for meniscectomy equals 3.54 (95%CI: 2.56–4.91). Conclusions: The main finding of this meta-analysis is that the prevalence of knee OA after ACL reconstruction is lower than commonly perceived. However, associated meniscal resection dramatically increases the risk for developing OA. As reconstruction of the ruptured ACL has been shown to reduce the need for secondary meniscectomy, these results indirectly support operative intervention to reduce the occurrence of early OA after an ACL tear.

P21-318 Standardised transtibial acl reconstruction technique can result in high reproducibility rate of the optimal femoral tunnel position: radiographic review S. Elnikety1, M. El-Husseiny1, T. Kamal2, M. Gregoras2, G. Talawadekar2, P. Jeer2 1 UCL, Biomedical Engineering Lab, London, United Kingdom, 2East Kent Hospitals, Trauma and Orthopaedics, Margate, United Kingdom Objectives: The transtibial approach is widely used for femoral tunnel positioning in ACL reconstruction. Controversy exists over the superiority of this approach over others. Few studies reflected on the reproducibility rates of the femoral tunnel position in relation to the approach used. In this study we examine the reproducibility rate of the femoral tunnel position using transtibial approach. Methods: We reviewed AP and Lat X-ray radiographs post isolated ACL reconstruction for 180 patients for femoral tunnel position, tibial tunnel position and graft inclination angle. All patients had their operations performed by one surgeon in one hospital between March

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 2006 and Sep 2010. All operations were performed using one standard technique using transtibial approach for femoral tunnel positioning. Two orthopaedic fellows, with similar experiences, reviewed blinded radiographs. A second reading was done 8 weeks later. Pearson interobserver, intra-observer correlation and Bland–Altman agreements plots statistical analyses were done. Results: Mean age was 29 years (range 16–54), Pearson intra-observer correlation shows substantial to perfect agreement while Pearson’s inter-observer correlation shows moderate to substantial agreement. Previous literature proved that optimal femoral tunnel position for the best clinical and biomechanical outcome is for the centre of the tunnel to be at 43% from the lateral end of the width of the femoral condyles on the AP view and at 86% from the anterior end of the Blumensaat’s line on the lateral view. In our study 85% of the femoral tunnels were within ± 5% of the optimal tunnel position on the AP views, and more than 70% of the femoral tunnels were within ± 5% of the optimal tunnel position on the Lateral view. Interobserver and interobserver correlations show moderate to substantial agreement, Bland– Altman agreement plots show substantial agreements for interobserver and interobserver measurements. These results were found to be statistically significant (p = 0.01). Conclusions: Based on our results we conclude that using one standardised transtibial technique for ACL reconstruction can result in high reproducibility rates of optimal femoral tunnel position.

P21-331 Single-stage revision anterior cruciate ligament reconstruction: 3–10 year follow-up S.J. Lee1, J.J. Lee2, C.H. Choi3 1 Yonsei University College of Medicine, Seoul, Republic of Korea, 2 Yonsei University College of Medicine, Orthopedic Surgery, Seoul, Republic of Korea, 3Yonsei University College of Medicine, Gangnam Severance Hospital, Orthopedic Surgery, Seoul, Republic of Korea Objectives: This study aimed to compare the long-term clinical results of patients who had received revision anterior cruciate ligament (ACL) reconstruction with that of patients who received primary ACL reconstruction. Methods: A retrospective study was performed against 41 patients for whom the revision ACL reconstruction was performed from March 2001 to March 2008. Among 472 patients who had received the primary ACL reconstruction during the same period, another 41 patient with pairing records in age, sex, preoperative measurements in KT 2000, types of graft, and follow-up period were selected as control group. Clinical results in the side to side difference using KT-2000 arthrometer, Lysholm score, Tegner activity score, IKDC score and single-legged hop test were obtained in the last follow-up and their damage of cartilage and meniscus was analyzed. Results: The average patient age was 31.3 years old; the average time to revision was 62 months. The tibialis anterior tendon allograft was used in 26 cases, the achilles tendon allograft was used in 13 cases and the bone patella tendon bone allograft was used in 2 cases. No significant difference was found on KT-2000 arthrometer (2.76 ± 1.57 mm for revision group, 2.91 mm ± 1.50 for control group, p = .781). 3 patients in the revision group and 1 patient in the control group showed positive results in pivot-shift test. It was shown that in the Lysholm score, Tegner activity score, and IKDC score, the mean score obtained by the revision group was 86.5, 6.5, and 87.5 respectively and the mean score of the control group was 94.3, 6.9, and 92.3. It was found that the patients in the revision group showed significant low score particularly in their pain score. In the single-

S195 legged hop test, there was no difference between the groups. Cartilage defect grade of the patients with revision group increased significantly comparing that of the patients with control group and it had some significant correlation with the clinical test. Conclusions: It was identified that the revision ACL reconstruction surgery showed no difference in anterior stability and comeback to sports activity in comparison with results of the primary reconstruction surgery. In clinical results, it was found that relative low score was obtained in the pain score, and it is considered that it had some correlation with the large damage of cartilage in the revision reconstruction surgery.

P21-433 Out in techinique in revision anterior cruciate ligament reconstruction with doubled semitendinosus and gracilis tendons and lateral extra-articular reconstruction: a long term follow-up study E. Monaco1, L. Caperna1, T. Palma1, F. Conteduca1, A. Ferretti1 1 St. Andrea Hospital, University of Rome ‘La Sapienza’, Rome, Italy Objectives: With this study are reported the results of ACL revision surgery using autologous doubled semitendinosus and gracilis tendon (DGST) graft with the addition of an extraarticular reconstruction at a mid to long term follow-up to assess the clinical and radiological outcome and the patients satisfaction. Methods: We performed 46 reconstruction of a previously reconstructed torn anterior cruciate ligament with use of a DGST graft. We used a two incisions technique with an out-in technique for femoral drilling. 1 patient of the 46 patients of this series had a failure and was treated with a repeat revision elsewhere and we considered it as a failure. Clinical and radiological evaluation was performed for all patients at a mean follow up of 5 years (min 2 years, max 12 years). Results: At the subjective evaluation the mean Tegner activity scale improved from apreoperative value of 3.5–5.8 (p = 0.00013) and the mean Lysholm score improved from a preoperative value of 65–95 (p = 0.003). According to the IKDC subjective evaluation the preoperative mean score was 74, none of the patients were in group A or B, 19 were in Group C and 26 in Group D. At the time of follow up the IKDC 2000 mean score was 93 and 26 patients were rated as Group A (58%) 11 patients were rated as group B (24%), and 8 patient as Group C (18%) (p = 0.003). Preoperatively at physical examination all patients had a positive Lachmann test and a positive pivot shift tests (14 patients with a + and 31 patients with a 2+). At the time of follow-up 34 patients had a negative Lachmann test while 11 patient had a + Lachmann test with a firm end-point. There were 3 patients with a positive pivot shift test (1 patient with a + and 2 with a 2+). At preoperative arthrometric evaluation with the Manual Maximum KT-1000 the mean side to side difference was 7.8 mm (range: 5–11). At follow-up 2 patients showed a Side to side difference of more than 5 mm at Kt-100 evaluation. There were signs of ostheoarthrosis in 45% of the patients. Considering as failure a difference on arthrometric test [5 mm and/or the pivot-shift test 2+ or 3+ our failure rate is 3 out of 46 patients (6.5%). Conclusions: The two-incision technique that we use seems like a good choice, especially when a half tunnelled technique was used in the primary reconstruction or when a blow-out of the posterior cortical wall was encountered. In fact, the outside-in technique allows the surgeon to orient the femoral tunnel in such a way that the new graft can be fixed in a previously undrilled area reducing the need of 2 stage procedures. Revision anterior cruciate ligament reconstruction with use of a DGST graft combined with an extra-articular procedure provided satisfactory functional outcomes.

123

S196

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

P21-449 Normative references and minimum detectable change in anterior and rotational knee joint laxity measurements C. Mouton1, D. Theisen1, D. Pape2, H. Agostinis1, R. Krecke2, R. Seil2 1 CRP-Sante´, Sports Medicine Research Laboratory, Luxembourg, Luxembourg, 2Centre Hospitalier Luxemburg, Clinique d’Eich, Luxembourg, Luxembourg Objectives: To establish normative references and measurement error when evaluating anterior and rotational knee joint laxity (KJL). Methods: Healthy participants (age: 37 ± 11) with no history of knee injury were tested prospectively by one rater for anterior (N = 60) and rotational KJL (N = 50) using the GNRB and a new Rotameter prototype, respectively. Participants were tested the same day by another rater to calculate the minimum detectable change (MDC). Anterior tibial displacement (ATD) was determined at 134 and 200 N. Primary (PCa) and secondary (SCa) compliance were calculated as the slope of the displacement-force curve between 30 and 50 N, respectively 100 and 200 N. Rotation angle was determined at 5 and 10 Nm for both internal (IR) and external (ER) rotation, as well for total range (TR). PC and SC were determined between 2 to 5 and 5 to 10 Nm applied torque, respectively. Normality was checked using the Kolmogorov–Smirnov test. Left–right differences were assessed using independent t tests. MDC was calculated according to Weir (2005) based on a repeated measures analysis of variance. Results are presented as mean ± standard deviation and 95% confidence intervals (CI) for left and right pooled values and differences. Results: There were no significant differences between left and right knees for any variable. Pooled results (95% CI and MDCs) are presented in Tables 1 and 2 for anterior and rotational KJL, respectively. The same results are displayed for left–right differences. It can be observed that PCa displays a high measurement variability which is unsuitable for patient follow-up. Conclusions: Normative references can help define in how far the measurements of a patient can be considered pathological. The results from the present study will allow for objective patient follow-up by comparing KJL measurements performed prior to and following

anterior cruciate ligament surgery. Changes in KJL should be greater than MDC to be considered a true modification of the patient laxity. Both absolute laxity measurements and left–right differences should be considered in this respect. Future studies will be directed towards profiling patients throughout the rehabilitation process with special focus on individual with and without a successful clinical outcome.

P21-498 Potential risk of the articular cartilage damage of the medial femoral condyle in accessory anteromedial portal drilling for anatomical double bundle ACL reconstruction A. Murase1, M. Nozaki1, M. Kobayashi1, H. Goto1, Y. Nishimori1, T. Otsuka1 1 Nagoya City University, Orthopaedic Surgery, Nagoya, Japan Objectives: Anatomical double bundle (DB) Anterior cruciate ligament (ACL) reconstruction is of critical importance for restoration of native knee kinematics and improved postoperative function. Some authors have reported that the accessory anteromedial portal (AAP) technique for femoral bone tunnel drilling provides more accurate anatomic positioning compared with the transtibial technique. However, the AAP technique causes a high risk for intraarticular cartilage damage to the medial femoral condyle. The articular cartilage damage may significantly influence postoperative symptoms such as pain and swelling, as well as the development of osteoarthritis. The purpose of this study was to investigate the risk of articular cartilage damage of the medial femoral condyle when performing anatomical DB ACL reconstruction using the AAP technique. Methods: 35 patients who underwent anatomical DB ACL reconstruction in this study. Intraoperatively, the guide pin was advanced to the femoral footprint of the anteromedial bundle (AM group) and posterolateral bundle (PL group) through the AAP at 90 and 120 knee flexion respectively. The distance between the guide pin and the articular cartilage of the medial femoral condyle was measured from the lateral portal at both knee flexion angles.

Table 1 Measure

95%CI of Left–right MDC 95% CI Absolute left–right of abso- (absolute difference values (mean ± SD) difference values) (mean ± SD) lute values

MDC (left–right difference)

ATD 134 (mm)

3.2 ± 0.5

2.2–4.3

0.9

-0.1 ± 0.5

-1.2 to 0.9

1.4

ATD 200 (mm)

4.5 ± 0.6

3.3–5.8

1.0

-0.1 ± 0.6

-1.2 to 1.0

1.6

PCa 29.8 ± 8.5 (lm/N)

13.0–46.5 16.1

0.3 ± 12.9

-24.9 to 25.6

SCa 20.9 ± 2.1 (lm/N)

16.9–25.0

0.6 ± 2.7

-4.7 to 6.0

3.6

26.2 6.3

Table 2 IR at 5 Nm ()

20 ± 5

10–30

3

1 ± 2.5

-4 to 5

4

ER at 5 Nm ()

29 ± 7

14–45

4

0±3

-5 to 6

6

TR at 5 Nm ()

49 ± 12

26–74

6

1±4

-7 to 9

7

IR at 10 Nm ()

36 ± 6

23–48

4

0±3

-6 to 7

5

ER at 10 Nm ()

45 ± 9

27–65

5

0±4

-7 to 7

7

TR at 10 Nm ()

81 ± 14

52–112

7

0±5

-9 to 10

9

PCir (/Nm)

4.4 ± 0.8

2.7–6.1

0.7

0.0 ± 0.6

-1.2 to 1.3

0.9

PCer (/Nm)

5.8 ± 1.2

3.4–8.3

0.8

0.1 ± 0.7

-1.3 to 1.5

1.0

SCir (/Nm)

3.1 ± 0.3

2.4–3.8

0.3

0.0 ± 0.3

-0.6 to 0.5

0.5

SCer (/Nm)

3.2 ± 0.4

2.2–4.2

0.3

-0.1 ± 0.3

-0.6 to 0.5

0.6

123

Through the accessory anteromedial portal

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: The mean distance to the articular cartilage Knee flextion () AM group PL group

Mean (mm)

SD

90

1.5

1.7

120

1.1

1.4

90

3.8

1.7

120

3.0

1.9

The distances to the articular cartilage in the AM group were significantly shorter than in the PL group at both 90 and 120 knee flexion. In addition, there were significant differences between 90 and 120 knee flexion in either group, the guide pin had proved to be closer to the articular cartilage of the medial femoral condyle at 120 knee flexion. Conclusions: Our results showed that AM tunnel drilling through the AAP had a relative higher risk of articular cartilage injury of the medial femoral condyle compared with PL tunnel drilling. In anatomical DB ACL reconstruction, higher knee flexion angles are recommended to avoid peroneal nerve injury and posterior cortex blow out when drilling femoral tunnels through the AAP. However, this study indicated that the risk of articular cartilage injury increased at higher degrees of knee flexion. Our study suggests that surgeons should be particularly careful to avoid articular cartilage injury of the medial femoral condyle during femoral tunnel drilling through the AAP, especially AM tunnel drilling.

P21-628 Patellofemoral osteoarthritis after anterior cruciate ligament reconstruction B.E. Øiestad1, I. Holm2, A.K. Aune3, L. Engebretsen4, R. Gunderson5, M.A. Risberg6 1 NAR/Ortopedic Center, Oslo University Hospital Ullevaal, Oslo, Norway, 2Oslo University Hospital Rikshospitalet, University of Oslo, Division of Rehabilitation, Oslo, Norway, 3Drammen Private Hospital, Drammen, Norway, 4Ullevaal University Hospital, Orthopaedic Center, Oslo Sportstrauma Research Centre, Oslo, Norway, 5Oslo University Hospital Rikshospitalet, Oslo, Norway, 6 Norwegian Research Center for Active Rehabilitation, Department of Orthopaedics, Norwegian School of Sports Sciences, Oslo, Norway Objectives: To investigate the prevalence of patellofemoral osteoarthritis (OA) 10–15 years after anterior cruciate ligament (ACL) reconstruction. Furthermore, the aim was to evaluate the association between knee function and symptoms and patellofemoral OA. Methods: In this prospective cohort study 258 ACL injured subjects were included at the time of reconstruction. Bone-patellar-tendon bone graft was used in 221 (86%) patients and hamstrings tendon graft in 37 subjects (14%). Performance based tests (isokinetic quadriceps strength, triple jump test, and stair hop test) were included at 6 months, 1 year, 2 years, and 10–15 years after the ACL reconstruction. Self-reported questionnaires (Visual Analogue Scale for pain and the Knee injury and Osteoarthritis Outcome Score, KOOS) and radiographic examination including skyline and lateral pictures were included at the 10–15 year follow-up. Radiographs were graded after the Kellgren and Lawrence classification (doubtful, mild, moderate, or severe OA) for both patellofemoral and tibiofemoral OA. Statistical analyses included binary logistic regression to analyze the association between patellofemoral OA and the KOOS scores, quadriceps strength, the triple jump test, and the stair hop test adjusted

S197 for gender, additional injuries, and age. Odds ratio (OR) and 95% confidence intervals (CI) are presented for the regression analyses. Repeated measures analysis of variance (ANOVA) was used to analyze knee function over time for the performance-based tests (quadriceps strength, triple jump test, and stair hop test). Results: Two hundred and ten subjects (81%) were evaluated at the 10–15 year follow-up, including 90 females (43%) and 120 males (57%) with a mean age of 39.1 ± 8.7 years. Additional meniscal or chondral injuries reported at the ACL reconstruction or sustained during the follow-up were seen in 61% of the subjects. Isolated patellofemoral OA was detected in only three subjects (1.5%), and 48 subjects (23%) had patellofemoral and tibiofemoral OA. None had severe patellofemoral OA. Isolated tibiofemoral OA was found in 101 subjects (48%). Subjects with patellofemoral and tibiofemoral OA had significantly worse self-reported KOOS subscores than those without OA (symptoms: OR 0.958, 95% CI 0.924, 0.994, pain: OR 0.947, 95% CI 0.905, 0.990, sport and recreation: OR 0.972, 95% CI 0.950, 0.995). No group differences were detected for muscle strength and hop tests either over time, or at the 10–15 year follow-up. Conclusions: A low prevalence of isolated patellofemoral OA was seen 10–15 years after ACL reconstruction (1.5%), but 23% had a combination of patellofemoral and tibiofemoral OA. Subjects with both tibiofemoral and patellofemoral OA showed significantly lower self-reported knee function and more pain and symptoms compared to those without OA.

P21-638 Sagittal alignment of the knee and its relationship to postoperative knee laxity after anterior cruciate ligament reconstruction M. Terauchi1, K. Hatayama2, K. Saito3, S. Yanagisawa3 1 Social Insurance Gunma Chuo General Hospital, Orthopaedic Surgery, Maebashi, Japan, 2Social Insurance Gunma Chuo General Hospital, Maebashi, Japan, 3Gunma University Faculty of Medicine, Department of Orthopaedic Surgery, Maebashi, Japan Objectives: It is widely acknowledged that knee hyperextension and tibial posterior slope influence knee kinematics and kinetics. The sagittal alignment of the knee in full extension was studied to determine the influence on stability following ACL reconstruction. Methods: We retrospectively studied 38 male and 40 female patients with an average age of 27. One year after anatomical double-bundle ACL reconstruction, a true lateral view of the knee in full extension was acquired using a fluoroscope, and three angles were measured. The angle between the femoral axis and the tibial axis was measured and designated the extension angle (EX). The roof plateau angle (RP) was the angle between the intercondylar roof and a line tangent to the concave profile of the medial tibial plateau (Fig. 1). The tibial posterior slope angle (PS) was defined as 90 minus the angle made by the intersection of tibial axis and the line tangent to the concave

Fig. 1

123

S198

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Table 1 Return to play

Fig. 2

High School (n = 68) [%]

College (n = 28) [%]

TOTAL Return to Play

62

68

Return to SAME performance

42

42

Return but NOT to same performance Did NOT Return to play

29

26

29

34

Reason for NOT Returning to Play due to Fear

63

44

Play at the NEXT level

12

14

Table 1 Saggital alignment measurement by angle Stable group

Unstable group

p

Female group EX

9.8 ± 4.9

14.4 ± 3.5

0.004

RP

69.3 ± 5.0

75.8 ± 5.4

0.01

PS

6.5 ± 2.3

7.6 ± 1.0

0.61

7.8 ± 3.9

9.8 ± 0.9

0.10

RP

69.1 ± 5.7

70.0 ± 3.6

0.63

PS

8.2 ± 3.6

6.7 ± 3.2

0.23

Male group EX

profile of the medial tibial plateau (Fig. 2). Anterior laxity was evaluated for side-to side difference (SSD) measured on anterior stress radiographs using a telos device. Results: Residual anterior laxity determined by telos device 1 year postoperatively was 2.6 mm in females and 2.4 mm in males. In females, there were 18 knees with a residual anterior laxity of 3 mm or more (unstable group) and 22 knees with residual anterior laxity \3 mm (stable group). The unstable group showed larger EX and larger RP than the stable group. There was no significant difference in PS (Table 1). EX correlated with RP (R = 0.52), indicating that knees with hyperextension also showed a vertical position of the femoral roof in relation to the tibial plateau. In males, there were 15 knees in the unstable group, and 23 knees in the stable group. The differences in EX, PS, and RP were not significant between the 2 groups (Table 1). Conclusions: In females, knees with large EX and large RP showed increased knee laxity after anterior cruciate ligament reconstruction. If the graft was placed parallel to the intercondylar roof, a vertical position of the intercondylar roof produced a relatively vertical graft placement and may reduce postoperative stability. These relationships were not seen in males.

P21-679 Return to high school and college level American Football following ACL reconstruction: a MOON cohort study K. Spindler1, K. McCullough2, K.D. Phelps2, E.K. Reinke2, W. Dunn3 1 Vanderbilt University Medical Center, Orthopaedics/Sports Medicine, Nashville, United States, 2Vanderbilt University Medical Center, Orthopaedic Surgery and Rehabilitation, Nashville, United States, 3Vanderbilt University Medical Center, Orthopaedics/Sports Medicine and Internal Med/Public Health, Health Services Research Center, Nashville, United States Objectives: ACL injuries are common career threatening injuries in American football. However, there is limited information on specific

123

return to play (RTP) and no published studies exist on RTP for high school (HS) and college (C) football or on the ability to play at the next level after ACLR. First, quantify the percentage of football players who RTP. Second, determine player opinion on their performance if RTP as well as reason(s) for not RTP. Third, determine the risk factors for not RTP or RTP not at the same level. Methods: Retrospective identification of all football players enrolled in 2002 and 2003 of the MOON cohort. Players were contacted and underwent a structured phone interview regarding participation in football surrounding the injury and factors involved with RTP. Descriptive data were analyzed and presented. Multivariable analysis was not available for football players who did not RTP due to inadequate sample. Results: 145 football players were contacted from the 2002/2003 MOON cohort. There were a total of 96 freshman-junior football players at either the high school or college level that had a competitive football program to RTP (68 HS, 28 C). The results are summarized in Table 1 below. The RTP rates were similar for high school and collegiate athletes with *64% RTP. Based on player perception, there was a 42% overall RTP at the same performance level with *30% who felt they did not perform at a level attained prior to their ACL tear. At both levels, a main reason players did not RTP was fear. As a result of the relatively low numbers at both levels of players who did not RTP (HS = 8, C = 20), multivariable analysis to determine potential risk factors (i.e., player position, meniscus and articular cartilage injuries, and graft type) could not be performed. Additionally, 12% of high school and 14% of college athletes report playing at the next level. Conclusions: RTP after ACLR in competitive high school and college American football is only * 64%. To our knowledge, this is the first report of football-specific RTP for amateur athletes and to identify fear as a modifiable risk factor to improve RTP in competitive football at these levels of play. Further ongoing recruitment is required for a detailed, multivariable analysis of multiple risk factors. The psychological component of RTP is all too frequently underestimated and warrants further investigation.

P21-740 Revision anterior cruciate ligament reconstruction using an anatomic approach for malpositioned tunnels. Clinical and three dimensional computed tomography results M.E. Hantes1, G. Tsougias2, M. Vlychou3, A. Tsarouhas2, K. Malizos4 1 University Hospital of Larisa, Orthopaedics, Larisa, Greece, 2 University Hospital of Larisa, Larisa, Greece, 3University Hospital of Larisa, Radiologic Department, Larisa, Greece, 4University Hospital of Larisa, Orthopaedic Surgery Department, School of Medicine, Larisa, Greece

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: The aim of this study was to evaluate the clinical and computed tomography results of one stage revision anterior cruciate ligament (ACL) reconstruction in previously malpositioned tunnels using an anatomic approach. Methods: Twenty-seven patients who underwent revision ACL reconstruction between 2002 and 2007 were included in this study. Inclusion criteria were knee instability symptoms, a positive pivot shift test, a non-anatomical placement of the femoral tunnel and a completely incorrect or correct tibial tunnel placement (in order to perform one stage revision) as determined by radiographs, magnetic resonance imaging scans and arthroscopic findings. Assessment included IKDC evaluation, Lysholm knee score, KT-1000 arthrometer, and radiologic examination. In addition, a three dimensional computed tomography (3D-CT) was performed in all patients postoperatively to evaluate new tunnel placement. Results: The mean follow-up was 6.5 (3–8) years. An anatomic tunnel placement was obtained in all cases according to 3D-CT. A significantly more oblique graft placement was obtained postoperatively in comparison to vertical orientation of the old tunnels, in the coronal plane (p \ .01). Postoperatively, the mean side to side difference on manual maximum testing improved significantly (2.2 mm vs. 11.2 p \ .01) as well as the mean Lysholm knee score (82 vs. 53 p \ .01). However, according to IKDC evaluation only 15 out of 27 (55%) patients were graded as normal or nearly normal. According to Fairbank scale 18 out of 27 (66%) patients had grade II and III degenerative changes on radiographs. Conclusions: Excellent results in terms of knee stability can be obtained after revision ACL reconstruction using an anatomic approach for previously malpositioned tunnels. However, the knee function is less predictable and it is directly associated with cartilage damage and knee arthritis.

P21-765 Return to skiing and re-injury after anterior cruciate ligament reconstruction in skiers Y. Kimura1, Y. Ishibashi1, E. Tsuda1, Y. Yamamoto1, S. Maeda1, S. Sasaki1 1 Hirosaki University Graduate School of Medicine, Orthopaedic Surgery, Hirosaki, Japan Objectives: Anterior cruciate ligament (ACL) injury is a common severe injury in skiing. It was reported that more than 10% of high level skiers were going to finish their career with multiple ACL injuries. So, it was important to investigate the prevalence of re-injury after ACL reconstruction. The purpose of this study was to determine the relationship between the competitive level, and rates of return to skiing and incidence of re-injury of reconstructed ACL and contralateral ACL injury after ACL reconstruction in skiers. Methods: Forty-nine female and 42 male skiers (age range of 12–60 years) who had undergone primary ACL reconstruction in our institute were studied. The skiers were classified into nationally competitive, regionally competitive and recreational groups based on the skill in skiing at the time of injury, and followed to determine the rate of return to skiing and incidence of re-injury of reconstructed ACL and contralateral ACL injury. Results: There were 33 skiers (36.3%) in the nationally competitive group, 19 (20.8%) in the regionally competitive group, and 39 (42.9%) in the recreational group. The rate of return to skiing was 100% in the nationally competitive group, 83.3% in the regionally competitive group and 50.0% in the recreational group. Five skiers (5.5%) re-injured the reconstructed ACL, and 4 of them were nationally competitive female skiers and 1 was a recreational male skier. Six skiers (6.6%) had contralateral ACL injuries, and they were 3 female and 2 male skiers in nationally competitive group and 1 male skier in recreational group. Mean time period from the primary ACL

S199 reconstruction to re-injury of the reconstructed ACL was 18.2 months and to the contralateral ACL injury was 38.2 months. Conclusions: Nationally competitive skiers returned to skiing in the next season, and they require high speed and high level skills. Higher level skiers had much more chance to return to the pre-injury level of competition and therefore greater risk to be exposed to inciting events of re-injury after ACL reconstruction. Young female high level skiers are especially at risk of re-injury and contralateral injury.

P21-775 Probability of meniscal tears in relation to the time after anterior cruciate ligament injury V. Tutkus1, J. Tutkuviene2 1 Vilnius University, Centre of Sports medicine and Rehabilitation, Faculty of Medicine, Vilnius, Lithuania, 2Vilnius University, Department of Anatomy, Histology and Anthropology, Faculty of Medicine, Vilnius, Lithuania Objectives: The timing of ACL reconstruction after injury is still under the discussion, and the data on meniscal damage after the ACL lesion varies as well. The recommended timing for ACL surgery varies between 6 weeks and 12 months. The aim of present study was to establish the probability of meniscal tears in relation to the time after the ACL injury. Methods: Data on 793 patients (75% males and 25% females) after the ACL injury were analysed. All patients underwent arthroscopic surgery, and meniscal lesions were detected in 561 patients. The meniscal tears were evaluated and ranked by the classification of ISAKOS (Anderson, 2010). The following time periods since the ACL injury to the surgery were composed: 0–6 moths, 6–12 months, 12–24 months, more than 24 months. The logistic regression was applied, and the odds ratios (OR) were calculated to establish the dependence of meniscal lesions on the time after the ACL injury. Results: The number of patients with the intact meniscus decreased with the time since the ACL injury to the surgery. There were 48% of patients without meniscal injuries in the group up to 6 months after trauma, while after 24 months—only 13%. The logistic regression showed the increasing likelihood of meniscal tears with the time from ACL injury to surgery: if more than 2 years passed after trauma, the probability of injury at the medial meniscus increased nearly four times (OR = 3.7), while at the lateral meniscus—very slightly (OR = 0.68). The medial meniscus was torn in 46%, and the lateral one—in 29% of all the patients with ACL injury. The traumatic types—longitudinal and flap tears of medial meniscus dominated and accounted for about 90% of all the tear types. The degenerative types of tears at medial meniscus were a rare phenomenon in the ACL deficient knee. The tears of traumatic origin dominated in the lateral meniscus as well: the longitudinal and flap tears together accounted for about 80%. The posterior horn of medial meniscus was mostly often damaged, and the tears of the anterior horn were very rare. The tears were more or less evenly distributed in the posterior horn and the middle portion of the lateral meniscus, and the anterior horn was damaged more often than at the medial meniscus in the ACL deficient knee. Conclusions: 1. If more than 2 years passed after the ACL trauma, the probability of injury at the medial meniscus increased nearly four times, while at the lateral meniscus—very slightly. 2. The medial meniscus injuries were two times more frequent than the injuries at the lateral meniscus, however, the distribution of tear types was approximately the same in both menisci at the ACL deficient knee. 3. We recommend perform ACL reconstruction within the first 6 months after the injury as the probability for meniscal lesions increases dramatically after the 6 month since the trauma.

123

S200 P21-814 Prospective randomized comparison of transtibial versus transportal techniques for clinical results in the double-bundle ACL reconstruction Y. Takeda1, T. Iwame1, T. Koizumi1, K. Miyatake1, K. Kondo1, T. Takasago1 1 Tokushima Red Cross Hospital, Orthopaedic Surgery, Komatsushima, Japan Objectives: The best technique for creating the femoral tunnel in double-bundle ACL reconstruction is still controversial. The purpose of this study was to compare the transtibial and the transportal techniques with respect to clinical results in double-bundle ACL reconstruction as the randomized clinical trial. Methods: 62 patients were randomly assigned to the two groups according to the blocked randomization. Of these patients, 31 patients underwent femoral tunnel creation with transtibial drilling and 31 patients underwent it with transportal drilling. Demographic data were not significantly different. In the both groups, same anatomic double-bundle ACL reconstruction using hamstring tendons was performed, and the same postoperative rehabilitation program was applied. Length of the femoral tunnel measured during surgery was recorded. Femoral and tibial tunnel placement was evaluated on the volume-rendering 3D-CT images at 7th day from the surgery. Clinical outcome at minimum 1 year (range 12–43 months) after the surgery was assessed by Lachman test, pivot shift test, KT-2000 arthrometer measurement, Lysholm score, the IKDC form, Tegner activity level, and isokinetic muscle strength. Statistical analyses were performed by use of the unpaired t test. A value of p \ 0.05 was considered statistically significant. Results: Five patients in the transtibial group were excluded from this study because the drill could not be positioned within the anatomical footprint through the tibial tunnel. AMB and PLB femoral tunnels created by the transtibial technique were placed in significantly higher and shallower position compared with those by the transportal technique, ant the transportal technique could place the both tunnels closer to the anatomical footprints. Length of the AMB and PLB femoral tunnels with the transportal technique was significantly shorter than that with the transtibial technique, and 10 of 62 tunnels (16%) in the transportal group was shorter than 30 mm. Negative rates of the Lachman test and pivot shift test were not significantly different between the two groups. The average side-to-side differences in the KT2000 knee ligament arthrometer values were 1.3 mm in the transtibial group and also 1.3 mm in the transportal group. There were no significant differences in Lysholm score, IKDC evaluation and Tenger activity score between the transtibial and transportal groups. There were no significant differences in the isokinetic peak torque of knee extension or flexion between the transtibial and transportal groups. Conclusions: In 5 of 31 patients (16%) in the transtibial group, a drill could be positioned only far from the anatomical footprint, and we had to change the procedure to the transportal technique. In addition, more anatomical placement of the femoral tunnels was achieved with the transportal technique. However, there was no significant difference with respect to the clinical outcome between the transtibial and transportal groups.

P21-829 Primary stability of a new technique for tibial fixation of a free tendon graft in ACL reconstruction: The retrograde shim technique S. Lenschow1, J.C. Fu¨tterer1, B. Schliemann1, M. Herbort1, M.J. Raschke1 1 Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: ACL reconstruction using a free tendon graft is the gold standard in the young, active patient. While extracortical fixation provides high fixation strength but low stiffness, aperture fixation by interference screw leads to laceration and uncontrolled rotation of the graft. For this reason, a wedge-shaped implant was developed which is pushed between the tendon graft and the tunnel wall (Megashim, Karl Storz, Tuttlingen, Germany). Our hypothesis was that sufficient tibial fixation strength is achieved if the shim is applied from extraarticular. Methods: Porcine tibia and flexor tendons were used for this study. An 8 mm tunnel was drilled. An 8 mm looped tendon graft was pulled into the tunnel. In the first group, the shim was applied from extraarticular (retrograde). In the second group, the shim was inserted from the articular side (antegrade). In the third group, an 8 mm interference screw was used. 10 specimens were tested in each group. The tibia was fixed to the base of a material testing machine. Maximum load, elongation, stiffness and failure mode were recorded. Cyclic loading was performed between 5 and 250 N for 1,000 cycles. Finally load to failure testing was performed. Statistical evaluation was performed using the Mann–Whitney-U Test. Results: The mean maximum load-to-failure was 774.5 N in group 1, 660.5 N in group 2 and 714.0 N in group 3. There was no significant difference between the groups (p [ 0.05). Stiffness varied between 201.5 N/mm in group 1, 199.4 N/mm in group 2 and 204.7 N/mm in group 3. No significant differences were found between retrograde shim and interference screw fixation (p [ 0.05). Three types of failure mode could be observed. The retrograde shim failed mainly by slippage of the graft (5/10), the antegrade shim failed by rupture of the tendon (7/10). For the interference screw, a rupture of the tendon was the main reason for failure (7/10). Conclusions: Aim of this study was to evaluate the biomechanical properties of the tibial fixation of a free tendon graft with a biodegradable wedge-shaped implant and compare them to a fixation by an interference screw. The results of our study show that there is no significant difference in stiffness, load-to-failure and elongation between the two fixation techniques. Furthermore the shim prevents a rotation and causes less laceration of the graft. However, some limitations apply to this study. No tendon to bone healing was taken into account. Load was applied in line with the bone tunnel which does not reflect the situation in vivo where the angle between the graft and the bone tunnel serves as a pulley, reducing the forces the fixation is subjected to under most loading conditions. The retrograde shim is an alternative to interference screw for tibial fixation of a free tendon graft. Because it causes no rotation, placement of the graft becomes more predictable. To prevent slippage of the graft, a hybrid fixation with an extracortical fixation device should be considered.

P21-916 Primary ACL Reconstruction using a quadriceps tendon graft with press-fit fixation leads to significant less bone loss in the bone tunnels compared to a hamstring graft with interference screw fixation—a CT scan analysis 3 months postoperatively R. Akoto1, J. Mu¨ller-Hu¨benthal2, M. Albers3, P. Helm4, B. Bouillon1, J. Hoeher3 1 University of Witten/Herdecke, Department of Trauma and Orthopedic Surgery, Cologne Merheim Medical Center, Cologne, Germany, 2Clinic for Radiology, Cologne-Triangle, Cologne, Germany, 3Arthro Sports Clinic, Cologne Merheim Medical Center, Cologne, Germany, 4University of Witten/Herdecke, Cologne Merheim Medical Center, Department of Trauma and Orthopedic Surgery, Cologne, Germany

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: To analyse the degree of bone loss in the femoral and tibial bone tunnel after ACL reconstruction by means of a multiplanar reconstruction technique of CT scan images and comparing two techniques of ACL reconstruction. Methods: 12 male patients underwent a primary anterior cruciate ligament (ACL) reconstruction with a quadriceps tendon autograft in a press-fit fixation technique (QTG). Three months after surgery all patients underwent a CT scan (approval by local ethic committee). The diameter of the femoral and tibial bone tunnel were analyzed by a mulitplanar reconstruction technique (MPR) by an independent radiologist. The tunnel size was analyzed in the coronal and sagittal planes. Femural and tibial three measurements were performed, starting at the level of the joint entrance (A) a well as 1 cm (B) and 2 cm (C) apart from the joint entrance. 12 patients who underwent ACL reconstruction using a standardized quadruple semitendinosus technique and interference screw fixation (HTG) over an anteromedial portal served as a control group. A unpaired t test was used for statistical analysis (p \ 0.05). Results: The tunnel diameter both for the femoral and tibial bone tunnel were significantly smaller for QTG compared to STG (yet, n = 6 per group). Results for femoral bone tunnel were (m ± sd): Femoral tunnel diameter sagittal plane (m ± sd): A: QTG 6.8 ± 2.7 mm; STG 11.5 ± 1.1 mm (p = 0.29); B: QTG 4.4 ± 2.5 mm; STG 10.5 ± 0.4 mm (p = 0.04); C: QTG 0.7 ± 0.7 mm; STG 10.0 ± 0.1 mm (p = 0.01). Femoral tunnel diameter coronal plane A: QTG 6.55 ± 3.2 mm; STG 10.9 ± 1.4 mm (p = 0.45); B: QTG 2.2 ± 1.3 mm; STG 10.6 ± 0.6 mm (p = 0.01); C: QTG 0.1 ± 0.1 mm; STG 9.3 ± 0.2 mm (p = 0.01). Tibial tunnel diameter sagittal plane A: QTG 8.7 ± 0.9 mm; STG 12.3 ± 2.4 mm (p = 0.08); B: QTG 10.7 ± 0.8 mm, STG 13.5 ± 1.04 mm (p = 0.01); C: QTG 7.5 ± 2.0 mm; STG 12.1 ± 1.0 mm (p = 0.01). Tibial tunnel diameter coronal plane A: QTG 7.3 ± 2.19 mm; STG 10.45 ± 0.52 mm (p = 0.07); B: QTG 6.25 ± 2.02 mm; STG 11.2 ± 0.8 mm (p = 0.01); C: QTG 5.95 ± 2.37 mm; STG 10.4 ± 1.62 mm (p = 0.04. Conclusions: CT scan analysis with multiplanar reconstruction technique is suitable to evaluate the amount of bone loss in the femur and the tibia after ACL reconstruction. The study demonstrates that ACL reconstruction using QTG and press fit fixation leads to significant less bone loss when compared to a standard hamstring technique.

P21-955 Pain evaluation after ACL ligamentoplasty: ‘‘All-inside’’ versus classical technique. A prospective randomized comparative study H.R.C. Benea1, S. Klouche1, T. Bauer1, P. Hardy1 1 Ambroise Pare Hospital, Orthopaedics Surgery and Traumatology, Boulogne Billancourt, France Objectives: The ‘‘All-inside’’ method is a new minimally invasive procedure for anterior cruciate ligament reconstruction (ACL). It consists of incomplete bone tunnels drilling and cortical fixation. In the so-called ‘‘classical’’ method, the fixation is achieved with interference screws. The hypothesis of this study was that the ‘‘All-inside’’ technique causes less pain than the conventional technique. The main objective was to evaluate the immediate postoperative pain for the first 10 days and at 1 month from the intervention. Methods: This is a prospective randomized comparative trial, run from December 2010 to September 2011, on a total of 44 patients operated for ACL lesion, 22 by the ‘‘All-inside’’ and 22 by the classical method.

S201 The mean age of patients was 31 ± 15 years. No patient was lost to follow up. An ethical committee approved this study. The primary evaluation criterion was daily pain assessment on a Visual Analogical Scale during the first 10 days and at 1 month after surgery. Secondary parameters were analgesics consumption, functional evaluation (IKDC score), delay of discharge of crutches and orthesis and radiographic analysis of tunnels positioning according to Aglietti’s criteria. Results: Two patients were excluded from the analysis because of postoperative complications (a hemarthrosis and a septic arthritis). The mean VAS value for the first 10 days was 14.9 ± 9.8 for the ‘‘All-inside’’ group and 14.5 ± 11.7 for the ‘‘classical’’ group, with no statistical significance. At 1 month, the pain level was 3.7 ± 6.1 for the ‘‘All-inside’’ group and 7 ± 10.5 for the ‘‘classical’’, statistically non-significant. The difference between preoperative and postoperative VAS was highly significant for both ‘‘All-inside’’ (p = 0.001) and classical (p = 0.004) technique. Postoperative analgesic consumption was similar. The clinical 1 month evaluation has showed for all patients a minimal joint effusion, a mobile and stable knee with a negative Lachman and the abandon of crutches before 1 month. The Rolimeter bilateral pre/postoperative anterior laxity evaluation showed a gain of about 1.5 mm in favour of the ‘‘All-inside’’ group. The bone tunnels were correctly positioned. Concerning the evaluation of pain, our study did not show statistically significant differences between the two techniques, probably due to a lack of statistical power. However, the results show a slight advantage for the ‘‘All-inside’’ technique in what concerns the postoperative pain decrement, graft tensioning (knee stability), postoperative range of motion and transplant positioning. Conclusions: The ‘‘All-inside’’ technique can be validated as a reliable procedure with very good results in what concerns pain, stability and knee function. Even if it is not statistically significant, there exists a tendency of pain diminution and better graft tensioning by this procedure when compared to the classical method.

P21-973 Mid term clinical results after augmentation anterior cruciate ligament reconstruction technique F. Abat1, P.E. Gelber2, J. Erquicia3, X. Pelfort4, M. Tey5, J.C. Monllau6 1 University of Barcelona. Hospital de la Santa Creu i Sant Pa, Department of Orthopaedic and Traumatology, Barcelona, Spain, 2 Hospital de la Santa Creu i Sant Pau, Department of Orthopaedic and Traumatology, Barcelona, Spain, 3University Hospital Dexeus. ICATME, Barcelone, Spain, 4Parc de Salut Mar. Universitat Auto`noma de Barcelona, Orthopedic Surgery, Barcelona, Spain, 5 Instituto Universitario Dexeus, Arthroscopic Unit, ICATME, Barcelona, Spain, 6University of Barcelona. Hospital de la Santa Creu i Sant Pa, Barcelona, Spain Objectives: To evaluate at minimum 2-years follow up, the results obtained after anterior cruciate ligament (ACL) augmentation of the anteromedial (AM) or posterolateral (PL) bundles. Methods: Prospective study of 28 patients after augmentation of a single bundle of the ACL. Eight patients were female and 20 male with an average follow up of 31 months (range, 24–37). The mean age of the patients at the time of surgery was 31 years (range, 18–41). There were 12 right knees and 16 left. Clinical and magnetic resonance imaging (MRI) examinations (Fig. 1) raised suspicion to the

123

S202

Fig. 1 MRI evaluation

Fig. 2 Arthroscopically examination

Fig. 3 Selective bundle reconstruction diagnosis. However, the exact injury pattern was established arthroscopically (Fig. 2). In 18 of the cases, the AM bundle was reconstructed. The other 10 cases had a PL bundle reconstruction procedure (Fig. 3). In all cases, the tunnels were drilled through the anteromedial portal. A quadrupled hamstring graft was used in 15 patients, doubled semitendinosus tendon in 7, tripled semitendinosus in 1, doubled gracilis in 4 and tripled in 1. Medium diameter of the graft was 7.2 mm (5.5–9). In 15 cases a Xo-Button (Linvatec) graft fixation device was used, in 12 patients Bio-Crosspin (Stryker) and in 1 an Endo-Button (Smith & Nephew). In 12 of the cases (8 lateral, 4 medial), an associated meniscal tear was found. Functional evaluation was assessed with IKDC, Lysholm and Tegner scores. Stability was evaluated with the pivot shift test and KT-1000 device (MEDmetric, San Diego, CA) at 30 lb. Statistical analysis was performed to compare the preoperative and postoperative evaluation.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: Preoperatively the pivot shift test was positive in 16 patients (57%) and Lachman test in all cases. Both tests were always negative during the follow-up. MRI examination diagnosed a partial ACL bundle lesion in 42.8% of the cases. Anterior tibial displacement with KT-1000 significantly dropped down from a mean of 3.5 ± 2.1 mm preoperatively to a 0.6 ± 2.2 mm postoperatively (p = 0.001). No differences were found comparing side-to-side. Subjective IKDC improved from 56.6 preoperatively to 93.6 postoperatively (p = 0.002). Lysholm score significantly improve from 65.2 (40–86) to 96 (85–100) (p = 0.001) and the Tegner score to 7.2 (5.8–7.8) (p = 0.001). Three patients developed extension deficits. Two cases were due to a cyclops syndrome and 1 was secondary to graft impingement. Conclusions: Selective reconstruction of only the injured ACL bundle, presents excellent functional results in the medium term with low complication rate. Long-term randomized controlled trials are necessary in order to obtain definitive scientific evidence of the superiority of this technique.

P21-999 Reconstruction of anterior and posterior cruciate ligaments in a patient with bilateral congenital absence of booth cruciate ligaments: a case report D. Starcˇevic´1, D. Trsˇek1, N. Medancˇic´1, M. Hasˇpl1 1 Hospital for Orthopaedic Surgery AKROMION, Krapinske Toplice, Croatia Objectives: Bilateral agenesia of the booth cruciate ligaments is a extremely rare congenital disease and this condition is frequently associated with other anomalies of the lower limb. We didn’t found presented one stage surgical ACL and PCL reconstruction. Methods: We are presenting a case of female, age 25, who injured her left knee during walking on uneven ground. Pain and instability of left knee were present despite conservative treatment. She didn0 t have any subjective problems with her right knee. Results: Clinically we found great anterior and posterior sagittal instability of both knees. No other congenital abnormalities we found of lower limb. MRI showed significant closure of femoral intracondylar notch, hypoplasia of tibial eminence and aplasia of both cruciate ligaments. Because conservative treatment fault we decided that surgical treatment is necessary. Arthroscopy showed intact articular cartilage and medial and lateral meniscus. There was no cruciate ligaments and tibial eminence was hypoplastic. Femoral intracondylar notch was closed so we performed significant notch plasty to create space for cruciate ligaments. We reconstructed posterior cruciate ligament with quadrupled hamstring tendons and anterior cruciate ligament with patellar tendon. We follow patient 18 months. On the end of this period knee was painful, no swelling, ROM 0/120 deg., Lachman +, posterior drawer test 0, jerk and pivot shift 0 and she has no subjective giving way of her left knee. Conclusions: We conclude that in case of congenital absence of booth crutiate ligaments, and after presenting subjective instability and knee pain, reconstruction of the ACL and PCL is indicated.

P21-1066 Parcial lesion of lca: what can we expect to the conservative treatment? H. Valencia-Garcı´a1, J.E. Ruiz Zafra2, A. Lopez-Hualda1, J. Martinez-Martı´n3, D. Lopez-Gonza´lez3, F. Panizo-Mota3 1 Hospital Universitario Fundacio´n Alcorco´n, Orthopaedic Surgery and Traumatology, Alcorco´n, Spain, 2Hospital Universitario

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Fundacio´n Alcorco´n, Traumatology, Alcorco´n, Spain, 3Hospital Universitario Fundacio´n Alcorco´n, Madrid, Spain Objectives: Partial ACL injuries account for 10–35% of all injuries. It is estimated that 10% are symptomatic, which means a large percentage without repercussion. In symptomatic patients, the current recommendation is surgery. But the recommendation is not defined in patients with few symptoms or with low physical activity. The natural history of these lesions is unpredictable, in the absence of any tool to measure the vascular damage and therefore the risk of progression to ischemic necrosis of the remaining fibbers (up to 50% of parcial lesions will progress to complete). To review the evolution of 20 patients diagnosed with partial ACL tear on MRI with few symptoms who refused surgical treatment and assess its state with minimum of 1 year follow-up. Methods: 20 patients mean age 29 years (21–42), with a rating of 95 prelesional Lysholm scale (range 90–100) and Tegner 7 (4–9), with no other concomitant lesions in the knee. All suffered minor or major trauma, had symptoms (occasional pain, knee swelling or failures), regardless of mobility and MRI diagnosis of partial rupture (thickened, curved or wavy ACL, signal, and abnormal contour or continued loss of T1 or increased signal intensity in T2), with 60% of PL bundle injuries, 20% of AM and 20% without specification by the radiologist. The diagnosis of partial lesion was made at 4 months of injury (3–5) and all patients were explained his injury, the possible evolution and treatment alternatives. All of them, by personal circumstances, limited employment or functional impairment, they opted for conservative treatment, which consisted of strengthening the quadriceps and hamstrings. None wore braces and none of them limited the activity. Results: Follow-up at least 1 year (1–2). During follow-up, 4 patients (20%) requested intervention failures or limitations in daily activity sports. The rest (80%), 4 patients (20% total, 25% of nonoperated) reported the same level of sport without discomfort and 12 (60% total, 75% of non-operated) reported discomfort results satisfactory but not limiting, residual instability of 4 mm on average maximum tension with KT-1000 (0–6 mm) and a lower fitness level of 1 or 2 points Tegner scale. The average rating follow-up of non-surgery was Lysholm 82 (60–100) and Tegner 5 (3–9) with a degree of overall satisfaction 80% and results IKDC scale of 4 A, 12 B and 1 C. This is a small sample and short follow-up, but with satisfying results in terms similar to those published (Mc Daniel, Odsensten, Kannus, Sommerlath, Noyes, Frunsgaard, Fritschy, Bak, Buckley). The low demand for sports is a non-negligible bias. The fact that most lesion present a PL bundle may influence not showing limitations in daily activities or sports of low demand. Conclusions: The prognosis in patients with partial tears of low demand looks good, accepting a residual laxity and decreased level of activity. Periodic reassessment is recommended in patients who opt for conservative treatment.

P21-1068 Longitudinal tear of medial meniscus posterior horn in anterior cruciate ligament deficient knee significantly influences the anterior stability of the knee S.H. Lee1, J.H. Ahn2 1 Kyung Hee University, Orthopaedic surgery, Seoul, Republic of Korea, 2Kangbuk Samsung Hospital, Sungkyunkwan University, Orthopaedic Surgery, Seoul, Korea, Republic of Korea Objectives: Longitudinal tears of the medial meniscus posterior horn (MMPH) are commonly associated with a chronic anterior cruciate ligament (ACL) deficiency. Many studies have demonstrated the

S203 importance of the medial meniscus (MM) in terms of limiting the amount of anteroposterior (AP) tibial translation in response to anterior tibial loads in anterior cruciate ligament-deficient knees. It was hypothesized that a MMPH tear on an ACL deficient knee would increase the AP tibial translation and rotatory instability. In addition, it was also hypothesized that MMPH repair could restore the tibial translation to the level before the tear. Methods: Ten human cadaveric knees were tested sequentially using a custom testing system under five conditions: intact, ACL deficiency, ACL deficiency with a MMPH peripheral longitudinal tear, ACL deficiency with a MMPH repair, and ACL deficiency with a total medial meniscectomy. The knee kinematics were measured at 0, 15, 30, 60, and 90 of flexion in response to a 134 N anterior and 200 N axial compressive tibial load. The rotatory kinematics were also measured at 15 and 30 of flexion in a combined rotatory load of 5 Nm of internal tibial torque and 10 Nm of valgus torque. Results: MMPH longitudinal tears in ACL-deficient knees resulted in a significant increase in AP tibial translation at all flexion angles except 90 (p \ 0.05). MMPH repair in an ACL-deficient knee showed a significant decrease in AP tibial translation at all flexion angles except 60 compared with the ACL-deficient/MMPH tear state (p \ 0.05). The total AP translation of the ACL-deficient/MMPH repaired knee was not significantly increased compared with the ACL (only) deficient knee but was increased compared with the ACL intact knee (p [ 0.05). A MM total meniscectomy in an ACL deficient knee did not increase the AP tibial translation significantly compared to MMPH peripheral tears in ACL-deficient knees at all flexion angles (p [ 0.05). In a combined rotatory load, tibial rotation after MMPH tears or a total medial meniscectomy in an ACL-deficient knee were not affected significantly at all flexion angles. Conclusions: This study shows that a MMPH longitudinal tear in an ACL-deficient knee alters the knee kinematics, particularly the AP tibial translation. MMPH repair significantly improved AP tibial translation in ACL-deficient knees. These findings may help improve the treatment of patients with ACL and MMPH longitudinal tear by suggesting that the medial meniscus repairs should be performed for greater longevity when combined with an ACL reconstruction.

P21-1176 Simple guidelines for anatomic femoral tunnel placement in ACL reconstruction A.D. Davis1, A. Izaguirre1, C. Brown2, M. Steiner3 1 New England Baptist Hospital, Boston, United States, 2Dubai Orthopaedic Clinic, Wellesley, Massachusetts, United States, 3 Harvard Medical School, Section Chief Sports, New England Baptist Hospital, Boston, United States Objectives: A successful ACL reconstruction requires that a graft or grafts be placed within the anatomic insertion of the native ACL. However, clock face guidelines for placement of grafts can be imprecise and identifying the remnants of the ACL footprint can be difficult. A practical method for use in the operating room to identify the femoral ACL attachment would be helpful. This study used cadaveric knees to measure the height on the lateral wall of the central ACL and its composite bundles and a method was demonstrated to find the ACL center. Methods: Twelve fresh frozen knees were dissected of all soft tissue except the ACL. The tibia and the medial femoral condyle were removed to provide visualization of the ACL femoral insertion. The ACL was divided into its anteromedial (AM) and posterolateral (PL) bundles and their femoral attachments were outlined prior to their removal. The center of each bundle was marked and the center of the ACL was identified. The femur was rotated in the sagittal plane to simulate 90 flexion and the lowest point on the lateral wall of the

123

S204

Fig. 1 notch at the margin of the articular cartilage was identified. A vertical line was drawn starting at this low point on the lateral wall (Fig. 1). Two independent examiners measured the height of the ACL center and its composite bundles on this vertical line and the distance from these center points to the front and back of the notch. To provide a practical method to find the ACL center a 7 mm offset femoral aimer was seated at the vertical height of the central ACL and a pin was placed through the aimer to mark a point on the lateral wall of the notch. The distance from the central ACL to this pin site was measured. To verify the location of the ACL and its bundles the center points were marked with small metal beads and lateral radiographs were analyzed by Bernard’s quadrant method. Results: The height on the lateral wall of the notch of the ACL center was 8.7 ± 0.6 mm. This point was also 1.7 ± 1.7 mm deep in the notch from the vertical line (Fig. 1). The AM bundle height was 9.6 ± 1.1 mm and the PL bundle height was 7.2 ± 1.2 mm. A 7 mm offset femoral aimer seated at the height of the central ACL placed a pin 2.8 ± 0.5 mm deep to central ACL insertion and 2.0 ± 0.5 shallow to the AM bundle insertion. The radiographic measurements of the ACL and its bundles were consistent with published values for these structures. Conclusions: The femoral attachment of the ACL and it bundles can be identified using a simple method based on the height of these structures on the lateral wall of the notch. A 7 mm femoral offset aimer placed at the height of the central ACL point will place a pin between central ACL and the AM bundle attachments. The shallow and deep positions of each bundle can be referenced to a vertical line on the lateral wall of the notch.

P21-1182 Reliability of a semi-automated 3D-CT measuring method for tunnel diameters after anterior cruciate ligament reconstruction S. Claes1, C. Robbrecht2, M. Cromheecke2, P. Mahieu2, J. Bellemans1, P. Verdonk2 1 University Hospitals Leuven, Department of Orthopaedic Surgery and Traumatology, Pellenberg, Belgium, 2University Hospital Gent, Department of Orthopaedic Surgery and Traumatology, Gent, Belgium Objectives: Widening of tibial and femoral bone tunnels is a common complication after anterior cruciate ligament (ACL) reconstruction and can create serious difficulties in case of revision surgery. Although appropriate imaging is key in evaluating bone tunnel enlargement, plain radiographs and MRI have been shown to be

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 unreliable even for simply identifying the presence of a bone tunnel. The goal of this study was to study the reliability of a novel, semiautomated 3-D computed tomography (CT) based image processing protocol for measuring bone tunnel diameters after ACL reconstruction. Methods: Thirty-five ACL-deficient subjects were prospectively included, underwent anatomic single-bundle ACL reconstruction and were evaluated at 1 year after surgery with the use of 3-D CT imaging. Three independent observers semi-automatically delineated femoral and tibial tunnel outlines, after which tunnel ‘‘casts’’ were calculated by the imaging processing software (Mimics v14, Materialise, Haasrode, Belgium) from which a best-fit cylinder was derived by a least-square method and the tunnel diameter was determined. Finally, intra- and inter-observer reliability of this novel measurement protocol was defined. Results: The mean femoral diameters were 10.99 ± 1.74 mm and 11.08 ± 1.74 mm, respectively. The intra-observer ICC was 0.973 (95% CI: 0.922–0.991). The inter-observer ICC was 0.992 (95% CI: 0.982–0.996). The mean tibial diameters among the 3 observers were 10.93 ± 1.61 mm; 11.16 ± 1.56 mm and 11.27 ± 1.54 mm. The intra-observer ICC was 0.955 (95% CI: 0.875–0.985). The interobserver ICC between three observers were 0.986, 0.969 and 0.955 respectively. Conclusions: To our knowledge, this is the first study utilizing 3Dcomputed tomography for the evaluation of tunnel widening after ACL reconstruction. This novel, semi-automated image processing method has shown to yield highly reproducible results for the measurement of bone tunnel diameter, area and volume. Therefore, this technique is a reliable tool for evaluating tunnel widening after ACL reconstruction, especially when planning revision ACL surgery.

P21-1188 Self-reported quality of life, knee function and activity level in persons with bilateral anterior cruciate ligament injury A. Fa¨ltstro¨m1, M. Ha¨gglund2, J. Kvist3 1 Ryhov County Hospital, Linko¨ping University, Physiotherapy, Jo¨nko¨ping, Sweden, 2Linko¨ping University, Physiotherapy, Linko¨ping, Sweden, 3Institute of Medicine and Health Sciences, Physiotherapy Linko¨pings University, Linko¨ping, Sweden Objectives: The aim of this study was to investigate self-reported quality of life, knee function and activity level in patients with bilateral anterior cruciate ligament (ACL) injury. Methods: Patients with bilateral ACL injury were identified through search of hospital records between 2004 and 2009 in three orthopaedic clinics. In total, 147 patients aged 18–45 years were identified, and 83 met the inclusion criteria for the study, having had their first ACL after 1997 and with no other major injuries in the knee joint. Six different questionnaires were used for data collection; the evaluation form Quality of life assessment in ACL deficiency (ACL-QOL), Knee Injury and Osteoarthritis Outcome Score (KOOS), Lysholm knee score, EuroQol (EQ-5D), Tegner activity scale, and a project specific questionnaire. The latter had questions about the level of activity, return to sport, and satisfaction with the level of activity and knee function. Results: Sixty-six patients (80%), of whom 31 (47%) were women, answered the questionnaires. The mean age was 29.1 (SD 7.2) years and the mean follow up times for the first and second ACL injury were 7.9 (SD 2.7) and 4.1 (SD 2.0) years, respectively. Sixty-five percent were reconstructed in both knees, 26% in one knee, and 9% were treated conservatively in both knees. Thirteen patients (20%) had to change their work or education plans because of their ACL injuries. Fifteen (23%) were back to their previous activity and eight (12%) were at the same level as before the ACL injuries. The most common reasons for not returning to the previous activity were

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S205

reduced function of the knee/knees (82%), a sense of not trusting the knees (84%), and fear of re-injury (69%). Sixty-one (92%) had changed their training habits due to their knee injuries. The median activity level according to Tegner activity scale was 9 (range 1–9) before any of the injuries, 7 (range 1–9) after one ACL injury, and 4 (range 1–9) at the time of the study. The median Lysholm score was 82 (range 34–100), 29 patients (46%) had good/excellent results, 24 (38%) fair, and 10 (16%) poor. EQ-5D index of the overall health status was 0.77 (SD 0.22) and EQ-5D VAS 75.5 (SD 17.6). KOOS sub-scores were: symptom 74 (SD 18.8), pain 81 (SD 15.9), ADL 91 (SD 9.9), function in sports and recreation 58 (SD 26.6) and kneerelated quality of life (QOL) 53 (SD 21.5). Mean (SD) of the ACLQOL was 6.0 (1.8) (1–10 scale). Conclusions: Patients with bilateral ACL injuries in our study had worse self-reported quality of life and knee function compared to previously published results for healthy populations, patients with unilateral ACL reconstructed knees, and for patients with unilateral ACL injured knees treated without reconstruction. Fig. 1 P21-1234 Morphology of the medial wall of the lateral femoral condyle S. Kopf1, S. Ingham2, C.A.Q. Martins2, P. Smolinski3, F.H. Fu4 1 Charite´, University Medicine Berlin, Center for Musculoskeletal Surgery, Berlin, Germany, 2University of Pittsburgh, Department of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States, 3 University of Pittsburgh, Department of Mechanical Engineering, Pittsburgh, United States, 4University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: Anatomical anterior cruciate ligament (ACL) has gained more and more attention. Previous anatomical studies have meticulously described the morphology of the femoral and tibial ACL insertion sites. One of the most used techniques to drill the femoral bone tunnel is the transtibial technique, which, however, tends to place the femoral bone tunnel aperture in a non-anatomical position that could impair good clinical outcomes. The purpose of the current study was to evaluate the area of potential non-anatomical placement of the ACL on the medial wall of the lateral femoral condyle, and to compare this non-anatomical area with the area of the native femoral ACL insertion site and its two bundles, anteromedial (AM) and posterolateral (PL). Methods: We dissected six cadaveric knees and marked the medial wall of the lateral condyle including the native ACL insertion site and its two bundles as well as the area anterior to the native ACL insertion site. The lateral condyle was separated from the knee. The entire soft tissue of the lateral femoral condyle was removed macroscopically using surgical instruments and chemically using a 10% sodium hydroxide solution. Afterwards all marked areas were recorded using a three-dimensional (3D) scanner (Faro Arm, Fl, U.S.A.) and the marked areas were calculated using specialized 3D software (Geomagic Studio, Research Triangle Park, NC, USA). Results: The 3D area of the native femoral ACL insertion site was 193 ± 54 mm2 (AM 121 ± 37; PL 72 ± 19) and the area of the entire medial wall of the lateral condyle was 509 ± 61 mm2. Thus, the native femoral ACL insertion site was 37% and the area of potential non-anatomical femoral ACL tunnel placement was 63% of the entire medial wall of the lateral condyle (Fig. 1). Conclusions: It has been known that non-anatomically placed bone tunnels do not restore normal knee kinematics and that impaired knee kinematics cause an early onset of osteoarthritis. The current study shows that the area of potential non-anatomical femoral tunnel placement is about 2/3 of the entire area of the medial wall of the lateral condyle. Nevertheless, thus far it is unknown how much of the

native insertion site should be restored during surgery to yield good long-term results.

P21-1277 Minor continental differences exist between outcome criteria used to determine success after ACL injury by orthopaedic surgeons A. Lynch1, D. Logerstedt1, H. Grindem2, I. Eitzen2, M.A. Risberg2, L. Snyder-Mackler1 1 University of Delaware, Department of Physical Therapy, Newark, United States, 2Oslo University Hospital, Department of Orthopaedics, Norwegian Research Centre for Active Rehabilitation, Hjelp24 NIMI, Oslo, Norway Objectives: This investigation began with a preliminary survey of participants at ESSKA 2010 that developed into a large internet based survey of orthopaedic surgeons. The purpose was to survey expert opinion about criteria these practitioners use to determine success 1 and 2 years after ACL injury between professionals from North America and Europe. Methods: Proposed outcomes were established by an expert panel with consideration of the literature regarding ACL injury with and without reconstruction. Surveys were emailed to the members of ESSKA and AOSSM. Outcomes included: • Return to sport at pre-injury level • Symmetrical performance on a functional test • No more than a mild effusion • No episodes of giving way • No radiographic progression of OA • Laxity difference \3 mm • Pivot shift grade of normal • Symmetrical quadriceps strength • Symmetrical hamstrings strength • Achieve [% on an outcome tool (PRO) A rating scale was developed to obtain opinion on the value of each measure. • Primary Importance, This criterion should be measured in all cases • Secondary Importance, Indicates good progress • Not Important/Do Not Use, Does not affect the outcome or does not matter • Indifferent, No opinion on the value of this measure • Unfamiliar with measure

123

S206 Responses were expressed as a percentage of the total sample. 80% agreement was established as consensus. Responses were compared between orthopedic surgeons from North America and Europe. Results: 493 surgeon responses were obtained (217 European, 276 North American). Regardless of geography, surgeons identified the absence of giving way, patient return to sport, quadriceps strength symmetry and patient reported outcomes as important. European surgeons also identified functional tests as important, while North American surgeons indicated that effusion and a negative pivot-shift as important. Measures of laxity and OA did not meet consensus standards. No consensus was achieved for important pass score criteria or the use of specific individual functional tests or PRO. The IKDC 2000 subjective knee form was the most positively viewed PRO. Conclusions: Outcome criteria surgeons use to identify success after ACL injury are not markedly different based on geography, despite geographical differences in management strategies for patients. Differences exist in testing knee stability with European surgeons preferring dynamic tests where North American surgeons prefer the pivot-shift test. Regardless, both regions identified that symptomatic instability was the chief indication of failure of management after ACL injury with or without reconstruction. These results highlight that despite different management schemes, surgeons typically use similar measures to identify success.

P21-1282 Outcome after re-revision ACL reconstruction in 15 patients from a single institution. 2–15 years follow-up M.J. Strauss1, S.E. Christiansen2, P. Faunoe3, B. Lund3, M. Lind3 1 Aarhus University Hospital, Sportstrauma Division, Department of Orthopedic Surgery, Aarhus, Denmark, 2University of Aarhus, Sportstrauma Division, Orthopedic Department, Skanderborg, Denmark, 3Aarhus University Hospital, Department of Sportstraumatology, Aarhus, Denmark Objectives: Outcome after re-revision anterior cruciate ligament reconstruction (RRACLR) is poorly described due to rare incidence and lack of presented literature. Due to status of the referral center at our clinic we have the possibility to follow-up a reasonable size of ACL re-revisions patients. The present study aims to present epidemiology and clinical outcome after ACL re-revisions with an intermediate follow-up length. Methods: A retrospective study of 17 consecutive patients treated with RRACLR from 2001 to 2009 were included at our clinic. The follow-up study was performed in 2011 and included subjective Knee Osteoarthritis Outcome Scores (KOOS), objective IKDC scores, KT1000 knee laxity measurements, and registration of reoperations and complications. Results: 15 patients were available for follow-up. Mean age was 28 years, 41% were males. All the patients were reconstructed with allograft tendons. Median follow-up was 6 years. KOOS subscores were preoperatively 63, 70, 81, 43 and 28 for Symptoms, Pain, Activity of Daily Living, Sports, and Quality Of Life respectively. At follow-up scores were 59, 69, 67, 33 and 43 for Symptoms, Pain, Activity of Daily Living, Sports, and Quality Of Life respectively. The quality of life score was significantly increased. KT-1000 was 4.9 mm preoperatively and 2.2 mm at follow-up. Conclusions: Subjective outcome scores indicate significant knee impairment with low sport and quality of life scores. This despite the finding of acceptable knee stability at follow-up. Our results indicate limited impact on pain and knee symptoms by RRACLR. This is probably due to accumulated meniscus and cartilage pathology. At mid-term follow-up ability to perform sports have reduced from

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 preoperative levels. This study shows that preoperative patient information of expected outcome after RRACLR is of outmost importance.

P21-1288 Precision of arthroscopic measurements during anterior cruciate ligament reconstruction using a standard arthroscopic ruler K.K. Middleton1, M. Miyawaki1, P. Araujo2, S. Tashman3, J. Irrgang2, F. Fu1 1 University of Pittsburgh Medical Center, Orthopaedic Surgery, Pittsburgh, United States, 2University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States, 3Department of Orthopaedic Surgery, Orthopaedic Biodynamics Laboratory, RiverTech Centre, Pittsburgh, United States Objectives: The concept for anatomical ACL reconstruction requires information on individual patient anatomical characteristics. The size of the insertion site should be known to more adequately reproduce a patient0 s native anatomy. The primary purpose of this study was to determine the precision of intra-operative measurements of the ACL femoral and tibial insertion sites using a standard arthroscopic ruler. Methods: In this cadaveric study, 3 experienced ACL surgeons measured the dimensions of the femoral and tibial insertion sites of 10 cadaveric knees. Using a three portal technique, the ligament was sacrificed prior to the surgeons making the measurements and the respective ACL remnants were carefully dissected using a heating device. The surgeons used an arthroscope and standard arthroscopic ruler to measure the length and width of the femoral and tibial insertion sites with the knee in 90 of flexion. The femoral side was visualized through the central portal (CP) or the accessory medial portal (AMP). The total length of the femoral insertion site in the proximal to distal direction was measured with the arthroscopic ruler bent at a 45 angle and inserted through a high anterolateral (AL) portal. The mid-width of the femoral insertion site was measured perpendicular to the long axis at the widest distance in the anterior to posterior direction. The tibial insertion site was visualized through the AL portal. The standard arthroscopic ruler was inserted through the CP to measure the total length of the tibial insertion site in the anterior to posterior direction. The mid-width was measured at the widest portion perpendicular to the long axis of the insertion site in the medial to lateral direction. Precision of the arthroscopic measurements was estimated based on the standard error of measurement (SEM), which is equal to the square root of the average squared within specimen deviations. Results: For the femoral insertion site, the SEM was 1.27 mm for total length and 1.78 mm for the mid-width. On the tibial side, the SEM was 1.21 mm for the total length and 1.81 mm for the mid-width. Conclusions: Arthroscopic measurement of the length and width of the femoral and ACL insertion sites demonstrated sufficient levels of measurement precision (\2.0 mm). Insertion site measurements are crucial in planning individualized single- or double-bundle ACL reconstruction. Such measurements are particularly important given the goal of individualized anatomic ACL reconstruction to reproduce the patient’s native insertion site dimensions as closely as possible. Further research is needed to demonstrate the accuracy of arthroscopic measurements in comparison to direct measurements of the ACL insertion site.

P21-1295 Minimum 10-year follow-up after anterior cruciate ligament reconstruction using Howells guide: The effect of tibial tunnel placement on clinical and subjective outcome E. Inderhaug1, T. Strand1, E. Solheim1 1 Haraldsplass Deaconess Hospital, Surgical, Bergen, Norway

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: There are no long-term studies evaluating the use of Howells guide for placement of tibial tunnels in ACL reconstructions. We reviewed our first series of patients reconstructed using Howells tibial guide, transtibial drilling of femoral tunnels, double-stranded hamstrings graft, BoneMulch and WasherLoc fixation. The goal was to establish the relationship between the radiographically assessed placement of the tibial tunnel and the long-term clinical and subjective outcome. Methods: 96 ACL reconstructed patients were offered a clinical follow-up at a minimum of 10 years. 79 patients (82%) were available for examination at 10.2 years (SD 1.8). 3 patients had undergone an ACL revision. Patients were assessed with IKDC subjective form, Lysholm, Tegner, clinical examination and standardized radiographs. Tibial tunnel placement was assessed along the AP-diameter of the tibial plateau (Amis-Jakob line) and the inclination of the tunnel was assessed in the coronal plane. Posterior placement of the tibial tunnel was defined as [50% along the AP-diameter of the tibial plateau and a high tunnel inclination was defined as above 75. Results: The mean placement of the tibial tunnel along the APdiameter of tibia was 46% (SD 5). The mean tibial tunnel inclination was 71.1 (SD 4.2, Range (62.6, 81.2)). Four patients were detected to have moderate impingement at the radiographic assessment. One of the revised patients had severe impingement. Nine patients (10.8%) had a posterior tibial tunnel ([50% of the tibial AP diameter). There was no significant difference in Lysholm, IKDC, Lachman or KT1000 values between the posteriorly and anteriorly placed tibial tunnels. There was however a higher incidence of rotational instability in the posteriorly placed tunnels (p = 0.02). Patients with pivot shift grade 2 had a significant lower Lysholm score than those with grade 0 and 1 rotational instability (p = 0.004). No difference in Lysholm, IKDC score, Lachman, KT-1000 values or rotational instability was revealed between the high and low inclination groups. Conclusions: 4 patients were found to have moderate impingement on radiographic evaluation. One of the revised patients had a severe degree of impingement. The use of Howells guide led to a posterior placement ([50% of the tibial AP-diameter) of the tibial tunnel in 10.8% of the patients. A significant higher proportion of patients with grade 1 or 2 rotational instability was found in this group. The group of patients with a grade 2 rotational instability had a significantly lower subjective score than those with only a slight rotational instability. No correlation was found between stability and coronal inclination angle, but in our series only 2 patients had an inclination [80.

P21-1300 No increased occurrence of OA following ACL reconstruction after isolated ACL injury in athletes T. Hoffelner1, H. Resch1, P. Moroder1, J. Atzwanger1, W. Hitzl2, M. Tauber3 1 Paracelsus Medical University Salzburg, Traumatology and Sports Injuries, Salzburg, Austria, 2Paracelsus Medical University Salzburg, Institut for Biostatistics, Salzburg, Austria, 3ATOS Clinic Center, Shoulder and Ellbow Surgery, Mu¨nchen, Germany Objectives: To evaluate the long-term radiographic and clinical results of ACL reconstruction by comparing the injured knee with the contralateral knee in athletes with isolated ACL tear returning to preinjury sports. Methods: Twenty-eight patients, with isolated ACL tears without concomitant injuries at baseline, returning to previous sports were selected. ACL reconstruction was performed using patella or hamstrings tendon graft. Conventional radiographs and a 3-Tesla (3-T) MRI of both knees were obtained, at a mean follow-up of 10 years

S207 after ACL reconstruction and afterwards were compared with each other. The International Knee Documentation Committee (IKDC) score and Tegner activity index were used for clinical evaluation and the Knee Injury and Osteoarthritis Outcome Score (KOOS) for evaluating self-reported knee function. Results: The 3-T MRI revealed positive signs of osteoarthritis (OA) on the operated knee in 33% and on the non-operated knee in 39% (p = 0.64). Conventional X-Ray showed ongoing signs of ROA on the non-injured knee in 14% according to Kellgren and Lawrence, in comparison to 21% on the injured-knee, (p = 0.73). The functional outcomes between the injured knee and the non-injured knee did not show any statistical differences. The mean postoperative IKDC score was 89.2 (±9.3) points and the total KOOS was 92.7 (± 7.8). The median preinjury Tegner score was 8 (±2) corresponding to 7 (± 2) at follow up. In 68% of the patients the Tegner score was unchanged from preinjury to follow-up. Conclusions: Athletes with an isolated ACL rupture showed no increased risk in the development of posttraumatic osteoarthritis in the long-term after ACL replacement when compared to the uninjured contralateral knee. Our findings support the evidence to perform ACL replacement in athletes.

P21-1312 Oral rivaroxaban for the prevention of deep vein thrombosis after arthroscopic ACL reconstruction N. Darabos1, T. Cuti1, D. Car2, V. Rakic-Ersek3, M. Pavic4, E. Elabjer1 1 University Clinic for Traumatology, Sports Trauma Department, Zagreb, Croatia, 2University Clinic for Traumatology, Internal Medicine Department, Zagreb, Croatia, 3University Clinic for Traumatology, Department of Radiology, Zagreb, Croatia, 4 University Clinic for Traumatology, Medical Biochemistry Department, Zagreb, Croatia Objectives: We investigated the efficacy of rivaroxaban, an orally active direct factor Xa inhibitor, in preventing of deep-venous thrombosis (DVT) after anterior cruciate ligament (ACL) arthroscopic reconstruction. Methods: In this randomized trial, 62 patients who were to undergo ACL surgery received either oral rivaroxaban, 10 mg once daily for 28 days, beginning 18 h after surgery (Group A, 31p), or subcutaneous nadroparine calcium 0.4 ml once daily for 7–10 days, beginning 12 h before surgery followed by oral varfarin 3–6 mg once daily (Group B, 31p). Analyses that included filled patient‘s questionary, physical measurement of suprapatelar circumference, biochemical blood tests (creatinin, urea, AST, ALT, GGT, D-dimers, PT/INRI, APTT, fibrinogen, C reactive protein, complete blood count) and Doppler ultrasound of both lower limbs, were done on admission and on 3rd, 6th, 10th day, and after 1 month postoperatively. The primary efficacy outcome was a DVT up to day 28 after surgery. The primary safety outcome was major bleeding. Results: The patients in Group A felt a less inconvenience due to coagulation testing and were more satisfied concerning the oral therapy, according to the data from patient‘s questionary. There was no significant difference in biochemical blood tests between two groups. The suprapatelar circumference measures were consistently more higher in Group B than in group A, without statistically significant difference. There was no difference between two groups regarding the primary efficacy (Doppler ultrasound of both lower limbs) and the primary safety. In both groups there was no postoperative DVT or mayor bleeding. Conclusions: Both procedures revealed a similar efficacy in DVT prevention after ACL surgery, but the administration of rivaroxaban was significantly more convenient for the patients then a combination of nadroparine calcium and varfarine.

123

S208 P21-1350 Revision anterior cruciate ligament reconstruction: A review of 25 cases R. Smigielski1, M. Szymanska1 1 Carolina Medical Center, Warsaw, Poland Objectives: The aim of this retrospective study was to identify the etiology of ACL graft failure and subsequent indication for ACL revision surgery in a group of 24 patients who underwent primary ACL reconstruction (ACLR) at our clinic. Methods: Over a 13 year period, 25 revisional endoscopic ACLR procedures were performed in 24 patients who had all undergone previous primary ACLR at our clinic (those who underwent primary ACLR elsewhere and were admitted to us for an ACL revision were excluded from the cohort). From this retrospective database, we obtained information of principal interest to us on graft failure etiology and additional data on graft source, average time from primary to revision ACLR, average age at time of revision and gender. Results: At the time of revision surgery, 36% of patients had experienced a traumatic event as the cause of graft failure. The next most common cause was inflammation (at 20%) leading to poor graft incorporation. Arthrofibrosis and inadequate graft fixation (incl. notchplasty)/improper graft tensioning were the third most common indications for revision at 17%. 2 out of the 25 procedures were classified to have failed due to ‘‘Inadequate graft incorporation, or failure of the graft to mimic the properties of the original ACL’’ without a single specific suspected etiology. One case required revision due to the formation of a cyst at one of the tunnel openings. With respect to gender, out of the 17 male cases of ACL revision reviewed, 7 had a history of single major trauma. In the female group, no such tendency was observed, with etiological factors being dispersed among the various groups listed above. Our 24-patient cohort displayed an average time from primary to revision ACLR of 2.8 years and an average age of 33.2 years at the time of revision. During revision, BPTB autograft was used 30% of the time, with allograft usage at 20%. The remaining 50% of patients underwent various other corrective procedures addressing specific graft failure mechanisms (cyst removal, osteotomy, fixation repair, bone grafting of tunnels, etc.). Conclusions: Single major trauma was found to be the primary etiological factor behind revisional ACL surgery. It was, however, the male patients who tended to have trauma as the primary cause of ACLR revision, at 41%. Meanwhile, the female population displayed no such etiological majority and graft failure causes were varied. Incidences of inflammation were also common, both acute and those with a later onset. Later onset inflammatory problems were linked with hyperurikemia, while early inflammation required revisional arthroscopic washout. Both complications could be potentially avoided by using appropriate treatment/prophylactic measures. Graft revision surgery is highly varied due mainly to the diverse etiology of ACL graft failures which must be taken into account in preoperative planning.

P21-1363 Single–bundle versus anatomic double-bundle reconstruction of anterior crutiate ligament: A early clinical results M. Hasˇpl1, D. Starcˇevic´1, D. Trsˇek1, N. Medancˇic´1, N. Kunac1 1 Hospital for Orthopaedic Surgery AKROMION, Krapinske Toplice, Croatia Objectives: Many studies support the theory that double-bundle reconstruction controls knee rotation better. Evaluation of the clinical results after reconstruction of the anterior crutiate ligament (ACL) with single-bundle (SB) or double-bundle (DB) surgical technique.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: We conducted a prospective randomized study of a SB group versus DB group ACL reconstruction using the hamstrings tendons. Forty patients, M 30 (75%) and F 10 (25%), aged 28.6 years, with unilateral ACL deficiency were randomized into two groups, by 20 knees each. Used transtibial approach and fixation with pins on femoral side and interferant screw on tibial side. FU 1 year and 7 months. For pre- and post-operative stability assessment used Tegner activity score, Lyscholm score, and IKDC final score. Results: During final examination at the SB group the mean Tegner activity score was 6.85, Lisholm score 90.77 and IKDC score 89.41 (20% group A and 90% group B). At the DB group Tegner activity score was 5.67 (p = 0.09), Lysholm score 94.92 (p = 0.16) and IKDC score 90.13 (p = 0.84) (32% group A and 68% group B). Average Lachman test for SB group 0.62 and for DB group 0.5 (p = 0.58). Average pivot shift test for SB group 0.54 and for DB group 0.25 (p = 0.21). Conclusions: Double-bundle reconstruction does not result in clinically significant advantage when compared with single-bundle. No statistical differences was found between two groups. The results do not clearly support the theory that double-bundle reconstruction controls knee rotation better.

P21-1368 Results of isolated anteromedial or posterolateral bundle ACL augmentation D. Piper1, A. Porteous2, J. Murray1, J. Robinson3, Bristol Knee Group 1 Avon Orthopaedic Centre, Bristol Knee Group, Bristol, United Kingdom, 2Avon Orthropaedic Centre, Southmead Hospital, Bristol, United Kingdom, 3The Bristol Knee Group, Orthopaedic Surgery, Avon Orthopaedic centre, Bristol, United Kingdom Objectives: Functional separation of the ACL into anteromedial (AM) and posterolateral (PL) fibre bundles has been widely accepted. The bundles act synergistically to restrain anterior laxity throughout knee flexion, with the PL bundle providing the more important restraint near extension and its obliquity better restraining tibial rotational laxity. 10% of ACL injuries involve isolated rupture to one of these bundles causing patients to present with instability symptoms or pain. As knowledge about the influence of the ACL bundles on knee kinematics has increased, isolated reconstruction of either PL or AM bundle has been advocated when the other bundle remains intact. This study presents the 1 year patient outcomes of a cohort of symptomatic patients presenting with either isolated PL or AM bundle rupture and treated by ACL augmentation. Methods: KOOS (Knee Injury and Osteoarthritis Outcome Score) and IKDC (International Knee Documentation Committee Form) scores at 1 years post op were obtained for 15 patients who had undergone isolated ACL augmentation between 2007 and 2009. In all patients arthroscopic examination revealed an intact AM or PL bundle. Reconstructions of the ruptured bundle were performed using either tripled semitendinosus or 4-strand semi-tendinous and gracilis grafts., taking care to preserve the intact bundle. Results were compared with scores for standard single bundle (SB) and double bundle (DB) ACL reconstruction procedures. In addition examination under anaesthesia (EUA) assessments were analysed to see if a pattern of laxity for isolated AM and PL rupture could be determined. Results: There were 6 patients with isolated AM bundle rupture and 9 with isolated PL bundle rupture. Pre-op EUA analysis demonstrated that patients with isolated PL bundle rupture had increased pivot shift and Lachman test laxity, whereas the AM bundle rupture group had increased anterior drawer test laxity. There was no difference between isolated bundle augmentation reconstruction and standard SB and DB ACL reconstructions. However the KOOS scores showed

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 significantly increased Sports function scores which was significantly better in the isolated bundle augmentations (93/100 vs. 74/100 SB, 79/100 DB). Differences between isolated AM and PL bundle reconstructions were not distinguishable. Conclusions: Isolated ACL bundle tears make up a significant proportion of ACL injuries. Although technically more difficult than standard ACL reconstruction, isolated bundle augmentation appears to be worthwhile and result in improved sports function when compared to standard ACL reconstruction procedures.

P21-1374 Single bundle ACL reconstruction in medial portal technique: Knee kinematics after femoral tunnel drilling with conventional over the top aimer in comparison to the new ‘‘Medial Portal Aimer’’, ‘‘MPA’’ M. Herbort1, C. Domnick1, S. Lenschow1, B. Schliemann1, W. Petersen2, M.J. Raschke1 1 University of Muenster, Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany, 2Martin Luther Krankenhaus, Klinik fu¨r Orthopa¨die und Unfallchirurgie, Berlin, Germany Objectives: The specific aim of this study was to investigate the influence of the use of the ‘‘medial portal aimer, MPA’’ in comparison to a conventional over the top aimer in medial portal technique on the resulting knee kinematics of the single bundle ACL reconstructed knee. We hypothesized that under a simulated KT 1000 and a simulated pivot shift test, an ACL reconstruction in medial portal technique using the ‘‘MPA’’ for femoral tunnel drilling will restore the intact knee kinematics more closely when compared to ACL reconstruction using the conventional over the top aimer (OTA). We hypothesized that under a simulated KT 1000 and a simulated pivot shift test, an ACL reconstruction in medial portal technique using the ‘‘MPA’’ for femoral tunnel drilling will restore the intact knee kinematics more closely when compared to ACL reconstruction using the conventional over the top aimer (OTA). Methods: In nine fresh-frozen human cadaveric knees (range 51–83 years) the knee kinematics were examined using robotic/UFS testing system (KR 125, KUKA Robots, Augsburg, Germany) by simulated Lachman test with anterior tibial load of 134 N and simulated Pivot Shift Test with combined rotatory load of 10 N-m valgus and 4 Nm internal tibial torque at 0, 15, 30, 60 and 90 of knee flexion. Within the same specimen the knee kinematics (anterior tibial translation) under simulated Pivot shift and Lachman test were determined in different conditions: intact knee, ACL-deficient and single bundle ACL reconstruction using the ‘‘MPA’’ or the over the top guide. Statistical analyses were performed using a Wilcoxon rank test (p \ 0.05). Results: Simulated Lachman Test: Significantly increased anterior tibial translation (ATT) after ACL reconstruction with OTA in comparison to the intact knee in all flexion angles. No significant difference in ATT after reconstruction using the MPA in comparison to intact knee at all flexion angles. Significantly decreased ATT in MPA group in comparison to the OTA group in 0 and 15 of flexion (p \ 0.05). Simulated Pivot Shift Test: Significantly increased anterior tibial translation (ATT) after ACL reconstruction with OTA in comparison to the intact knee in 0–60 of flexion. No significant difference in ATT after reconstruction using the MPA in comparison to intact knee in all flexion angles. Significantly decreased ATT in MPA group in comparison to the OTA group in 15 and 30 of flexion (p \ 0.05). Conclusions: Single bundle ACL reconstruction in medial portal technique using the ‘‘MPA’’ will restore the intact knee kinematics more closely when compared to a reconstruction by using a conventional over the top guide.

S209 P21-1377 Regeneration of hamstring tendons after anterior cruciate ligament reconstruction: A prospective analysis M. van der Velden1, R.P.A. Janssen2 1 Maxima Medical Center, Orthopaedic Surgery, Eindhoven, Netherlands, 2Maxima Medical Center, Orthopaedic Surgery and Traumatology, Veldhoven, The Netherlands Objectives: Primary aim of the study was analysis of hamstring tendon regeneration after anterior cruciate ligament reconstruction (ACLR). Secondary aim was analysis of isokinetic muscle strength in relation to hamstring regeneration. The hypothesis is that regeneration of hamstring tendons after ACLR occurs, and that regenerated hamstring tendons contribute to isokinetic hamstring strength with regeneration distal to the knee joint line. Methods: Twenty-two patients scheduled for ACLR underwent prospective MRI analysis of both legs preoperatively as well as at 2 weeks, 6 and 12 months postoperatively. MRI parameters were tendon regeneration and morphology, muscle retraction and muscle cross sectional area. A double blind, prospective analysis of isokinetic quadriceps and hamstrings strength was performed and at 6 and 12 months postoperatively. Results: Figure 1 (below) shows the results of hamstring regeneration. All 22 patients demonstrated hamstring regeneration after harvest for ACLR. In 14 patients (64%), both semitendinosus and gracilis tendons regenerated. In the remaining 8 patients (36%), one tendon regenerated, the gracilis tendon in all cases. The semitendinosus muscle cross sectional area of the operated leg showed a 32% decrease at 12 months compared to preoperatively (p \ 0.01) as well as 41% decrease compared to the same muscle in the contralateral leg (p \ 0.01). The gracilis muscle cross sectional area decreased 26% (p \ 0.01) and 29% (p \ 0.01) respectively. The median cross sectional area of the semitendinosus muscles without tendon regeneration was smaller after 12 months compared to the semitendinosus muscles with regeneration distal to the joint line (5.98 (±2.05) cm2 vs. 9.99 (±2.61) cm2, p = 0.05). Even though all gracilis tendons regenerated, there is a difference in gracilis muscle cross sectional area in the group of patients with tendon regeneration proximal to the joint line compared to the group of patients with gracilis regeneration distal to the joint line after 12 months (2.76 (±0.69) cm2 vs. 4.80 (±1.09) cm2, p = 0.01).

Fig. 1 Regeneration of hamstring tendons and the insertion level (neo-tendon regenerated tendon, prox. proximal, jl joint line)

123

S210 At 12 months, the group of patients without semitendinosus regeneration showed more semitendinosus muscle retraction compared to the group of patients with semitendinosus regeneration distal to the joint line (13.04 (± 3.35) vs. 3.75 (± 2.02), p = 0.02). There was no significant relationship between isokinetic flexion strength and tendon regeneration. Conclusions: Hamstring tendons regenerated after harvest of both semitendinosus and gracilis tendons for ACLR. There was no relation between isokinetic flexion strength and tendon regeneration.

P21-1398 Morphology and area of ACL femoral insertion site Y. Fujimaki1, M. Miyawaki1, P. Araujo1, J. Irrgang1, S. Tashman1, F.H. Fu1 1 University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: The goal of anterior cruciate ligament (ACL) reconstruction is to restore the native ACL anatomy and normal knee kinematics. Knowledge of ACL insertion site morphology is necessary to best match the bone tunnel with the native ACL footprint. The aims of this study were to evaluate the area of femoral ACL insertion site using a three-dimensional digitizing system, and to compare the femoral insertion area with length, width, and predicted insertion site area based on length and width of insertion site. Methods: Ten fresh-frozen, cadaveric knees with no macroscopic degenerative or traumatic changes were used for this study under a protocol approved by the local research ethics committee. The mean age at death was 60 years. All soft tissue structures were removed to expose the bone, leaving only the ACL and the overlying synovium was carefully removed. The femoral ACL footprint was then outlined using an articulated arm digitization device. These points were projected to 2D as seen from medial–lateral axis and the area of the femoral insertion on this plane was calculated. Length of the femoral insertion site was measured along the longest axis and the width was measured along a line perpendicular the midpoint of the long axis. All measurements were expressed as mean ± standard deviation. To establish the best method to predict the femoral insertion site area, the digitized area was compared with area calculated based on the formula of an ellipse. SPSS version 20 was used for all statistical analyses, with the level of significance set at p \ .05 (Fig. 1). Results: Mean femoral insertion length and width were 17.4 ± 1.4 mm and 9.2 ± 1.2 mm, respectively. The insertion area was 117 ± 20.3 mm2. Regression analysis demonstrated that the digitized area of the femoral insertion was best predicted by an elliptical model (R2 = 0.976, p \ .001). Prediction of the femoral insertion site area based upon width also shows good correlation (R2 = 0.744, p \ .001) but only on length was not as strong (R2 = 0.369,

Fig. 1 Digitized points and area of footprint

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 p = 0.065). The regression equation was as follows: AREAACL = 0.996 9 AREAELLIPSE—7.905. Mean percent difference between the femoral insertion area and predicted area based on an elliptical model was 16.4 ± 3.9%. Conclusions: The femoral insertion area is best predicted by the formula for an ellipse which considers both the length and width of the ACL insertion site. Prediction of the ACL insertion site area based on length alone is insufficient. Estimation of the ACL femoral insertion area will help surgeons optimize surgical parameters (e.g. drill diameter) to create the most anatomical femoral bone tunnel and select the most appropriate method (i.e. single vs. double-bundle) to reconstruct the ACL.

P21-1401 Outcome following infection after ACL reconstruction T. Wood1, N. Smith1, T. Spalding1, P. Thompson1 1 University Hospital Coventry and Warwickshire, Coventry, United Kingdom Objectives: ACL (anterior cruciate ligament) reconstruction is a commonly performed operation. There are a number of series looking at complications, although there is very little evidence on the clinical outcomes following infection. Following a number of deep infections following ACL reconstruction in our department, it was decided to retrospectively review the clinical outcomes following deep infection of all patients undergoing ACL repair since 2007. Methods: A prospective record of all the patients that suffered a deep infection following ACL repair had been kept between January 2007 and April 2011 at our teaching hospital NHS trust, and the two local private hospitals. All patients underwent at least 2 arthroscopic washouts with limited synovectomy if required. Targeted antibiotics were commenced according to the culture results, and following microbiological advice. These patients were reviewed at a minimum of 1 year following eradication of infection (range 12–46 months). There were 7 surgeons performing the ACL reconstructions. The primary outcome measure was graft failure requiring revision. Our secondary outcome measures were a history of ongoing instability, KT 1000TM measurement, Tegner and Lysholm outcome scores. Results: There were 19 patients identified as having suffered infection after ACL infection (mean age 24.3 years, range 15–38 years). Average C Reactive Protein (CRP) was 217 on admission (range 59–397). The most common organism isolated was coagulase negative staphylococcus in 47.3% of cases. No organism could be identified in 31.6% of the cases. There were 3 graft failures within the infection group, 2 of which have subsequently undergone revision ACL reconstruction without complication. Of the remaining 16 patients there were no episodes of ongoing instability in the infection group and mean pivot shift grade was 1.1 in the infection group. Mean KT 1000TM side-to-side difference was +1.8 mm in the infection group. The mean drop in activity on the Tegner score was 1.75 (range 0–6) and mean Lysholm score was 89 (range 56–100). Conclusions: The failure rate is slightly higher than that reported in the literature. Patient reported outcome measuresin the patients are acceptable. We recommend an aggressive approach to the treatment of deep infection following ACL reconstruction. There is very little data in the literature following an infected ACL reconstruction, with this study giving some evidence on clinical outcomes.

P21-1425 Platelet-rich plasma in ACL quadriceps tendon bone reconstruction. Inflammatory response, tunnel widening and proprioception afterwards K. Malinowski1, R. Wiecek1, K. Hermanowicz2, P. Jancewicz2, K. Kuzma2, M. Synder3

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 10

Artromedical’, Belchatow, Poland, 2SPZOZ, Bielsk Podlaski, Poland, 3Medical University of Lodz, Lodz, Poland Objectives: The purpose of the multicenter, blind test, retrospective study was answers for following questions: Would use of PRP in ACL reconstruction improve effectiveness of treatment and/or accelerate biologic transformation of the graft? Would use of the ‘‘biologic weapon’’ allow for starting activity and sport earlier? Additional purpose is to take a note of problems that could occur intra and postoperatively. Methods: In the study we compared two 52 and 54 patients groups that underwent ACL rec. with QTB graft with and without PRP. ACL were reconstructed in anatomic, single bundle way with additional tibial stabilization (bone bridge). We use biologically and chemically neutral titanium screws. We injected PRP on the graft, in tunnels and into harvest site. Mean age was 29 years and follow up 16–28 months. Patients were evaluated by 3 independent surgeons. We had restrictive inclusion criteria: no more than 3 mm additional instabilities, no rotational instability, full ROM, lack of hip and ankle diseases, no more than 1/3 of diameter lateral and 2/3 medial meniscectomy, no more than 2ICRS cartilage damage, no more than 10 malalignement. We performed stage surgeries so finally we did only ACL rec. in last procedure. Results: In PRP group mean postop. fever was higher about 0.5C and it persisted on average 1 day longer. In the group, patients gave painkillers up about 3 days earlier. ROM 0–90 was reached after 40–21 days in PRP group and after 53–25 day in ACL only group adequately. Harvest site pain appeared in the same amount of patients (3) in groups and in 0–3 scale mean score was 1, 3 for PRP and 2, 3 for simple rec. group. In the last one occurred case with complete lack of harvest gap filling. Tibial tunnel widening was on average 1, 4 mm in PRP group and in comparison it was about 30% less Mean side to side instability was 1, 3 mm in PRP comparing to 2 mm in simple group. Patients in PRP group tolerated walking without crutches about 3 days earlier and after 6 months about 10% more of them carried out our sensomotoric test. We noticed no significant differences in KOOS and IKDC scales in follow up, but in simple reconstructions group we found two D and two C (IKDC) scores compared to only one C and no D in PRP one. MRI showed some features of faster graft maturation in PRP group but artifacts connected with titanium screws caused not reliable results. In 3 patients in PRP group that had problems with reaching full ROM after prior procedures we observed reversible but difficult in physiotherapeutic treatment ‘‘stiffness of soft tissues’’. Conclusions: Use of PRP in ACL QTB rec.: • probably causes faster return of propriocepitve answer probably as a result of biologic acceleration of graft maturation and its higher density • probably causes longer time of postoperative inflammatory response but decreases pain, time to crutches-free walking, time to reach of ROM and tunnel widening • makes the procedure more difficult. We suggest caution in cases with prior ROM problems.

P21-1435 Patellar tendon versus hamstrings in ACL reconstruction: a 7 year follow-up prospective randomized study G. Camillieri1, F. Matteo2, V. Calvisi3 1 II Faculty, Roma, Italy, 2Villa Aurora Hospital, Rome, Italy, 3 University of L’Aquila, Rome, Italy Objectives: Single bundle, single tunnel reconstruction of ACL seems to be an ‘‘old procedure’’ after the recent development and knowledge about double-tunnel anatomical reconstruction of ACL. Despite the new trend, prospective long term follow up studies on the single bundle reconstruction are necessary to get more information on a

S211 technique that has occupied the last 20 years and has been executed on millions of patients. Methods: From January 2002 and February 2003, hundred eight patients were recruited for a prospective randomized trial. The exclusion criteria were previous knee operations, chondral lesions, varus deformity more than 10, limb imbalance, meniscal lesion extension more than on third of the meniscus and history of knee tendinopathies. Three groups of 36 patients each were selected by the arthroscopically assisted surgical procedure: A B-PT-B reconstruction, B hamstrings (endobuttonCL-reabsorbable interference screw), and C hamstrings (endobuttonCL—evolgate). A dice method was used for the graft randomization. A standard rehabilitation protocol was used for all the patients, including immediate postoperative mobilisation with a knee brace just for walking (first 3 weeks), protected weight bearing for 2 weeks, and return to full activity at 6–12 months postoperatively. The patients were followed up at 6 months, 1 year and then each year until the seventh year postoperative. Six patients in group A, 8 in group B and 11 in group C were lost at the final follow up. The methods of evaluation included clinical and instrumented laxity testing (Rolimeter), isokinetic muscle torque measurements (Cybex II), International Knee Documentation Committee II ratings, Lysholm score, Tegner activity level. Kujala patellofemoral knee scores, and radiological tunnel enlargement measure. Results: The results revealed no statistically significant differences with respect to IKDC II, Lysholm score, Isokinetic test and Kujala patellofemoral knee scores. There was an enlargement of the tunnels, statistically significant in the hamstring tendon groups, and more evident in the group B. Any widening of the tunnels from 3 to 5 years was recorded in either group. The measurement of laxity using the Rolimeter showed better results for group A without any difference statistically significant. Differently, the Tegner score was statistically favourable for PT group. Radiological narrowing of the joint spaces (IKDC) from 1 to 7 years postoperatively was seen in all the groups, At 7 years follow up, radiological findings were worse per the PT group (p \ 0.05). Conclusions: PT and hamstrings ACL reconstruction seem to show similar functional and clinical results at long term follow-up. Tunnel widening in hamstrings groups, may be a complication in case of revision. The patellar tendon group seems to be more prone to develop osteoarthritis after 5 years postoperatively.

P21-1448 Mini-invasive medial collateral ligament and posterior oblique ligament repair in combined chronic medial and ACL injuries G.L. Canata1 1 Koelliker Hospital, Centre of Sports Traumatology, Torino, Italy Objectives: Residual laxity remains after ACL reconstruction in patients with combined chronic ACL and medial instability. The question arises whether to correct medial capsular and ligament injuries when Grade II and III medial laxity is present. We developed a mini-invasive medial ligament plasty to repair the medial collateral ligament to correct residual medial valgus and rotatory laxity after ACL reconstruction. Methods: We prospectively followed 36 patients with an ACL deficiency combined with chronic Grade II or III valgus and rotatory medial instability. The mean age was 37 years (range, 15–70 years). For all patients, we obtained preoperative and postoperative Knee Injury and Osteoarthritis Outcome, International Knee Documentation Committee, Lysholm, and Tegner Activity Level Scores. The minimum followup was 2 years (median, 3 years; range, 2–7 years). Results: The mean subjective International Knee Documentation Committee score improved from 36 preoperatively to 94 at the last

123

S212 followup. While all patients had an International Knee Documentation Committee score of Grade C or D preoperatively, no patient did postoperatively. The mean Knee Injury and Osteoarthritis Outcome Score improved from 45 preoperatively to 93 postoperatively. Valgus and external rotatory tests were negative in all patients. The mean Tegner activity level decreased from 7 preinjury to 6 postoperatively, and the mean Lysholm score improved from 40 preoperatively to 93 at last followup. Conclusions: This simple technique restored medial stability and knee function to normal or nearly normal in all patients.

P21-1476 Magnetic resonance imaging and clinical evaluation of anterior cruciate ligament reconstruction using the femoral bioabsorbable cross-pin fixation via the anteromedial portal S.H. Lee1, Y.B. Jung2, J.H. Ahn3 1 Kyung Hee University, Orthopaedic Surgery, Seoul, Republic of Korea, 2Hyundae General Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea, 3Kangbuk Samsung Hospital, Sungkyunkwan University, Orthopaedic Surgery, Seoul, Republic of Korea Objectives: The goal of this retrospective study was to investigate the clinical results and graft status according to problems associated with the Rigidfix pin after anterior cruciate ligament (ACL) reconstruction, using the anteromedial (AM) portal technique. Methods: Twenty seven patients who underwent an ACL reconstruction using the AM portal technique with quadruple hamstring tendon autograft were evaluated by MRI at a mean of 6.3 ± 0.7 months after surgery. All patients were evaluated by clinical outcomes and stability tests at mean 29.3 months. The MRI evaluation focused on the status of the Rigidfix pins and signal of the ACL graft. Results: In MRI evaluation of cross-pins, protrusion, breakage, and migration of these pins were observed in 23 (85%), 15 (56%), and 11 (41%) patients, respectively. Cross-pins migrated near the popliteal space in 6 patients. The MRI signals of ACL grafts were classified as grade I, II and III in 17, 9 and one patients, respectively. There were significant differences between the breakage and intact cross-pin groups in terms of anterior stress radiographs (p = 0.032). However, the results of clinical scores, age, range of motion, Lachman test, pivot shift test and ACL graft signal did not significantly differ between the two groups (p [ 0.05). Conclusions: The present study demonstrates that use of the AM portal technique for femoral tunnel positioning can cause a high prevalence of Rigid fix pin problems such as protrusion, breakage, and migration. The broken Rigid fix pins are a factor leading to anterior instability after ACL reconstructions with hamstring tendon.

P21-1482 Midterm follow-up of an Anatomic ‘Hybrid’ ACL reconstruction technique using one femoral and two tibial tunnels G.T. Altman1, S. Jain1, K.P. Patel2, P. Re3 1 Allegheny General Hospital, Department of Orthopaedic Surgery, Pittsburgh, United States, 2Drexel University COM, Philadelphia, United States, 3Emerson Hospital, Department of Orthopaedic Surgery, Concord, United States Objectives: The goal of ACL reconstruction is to accurately restore the anatomy of the native ACL. The Anatomic ‘Hybrid’ ACL reconstruction offers a reproducible, less technically demanding method with which to obtain anatomic ACL reconstruction. The goals of our study were to obtain and measure midterm follow up of the Anatomic ‘Hybrid’ ACL reconstruction technique utilizing one femoral and two tibial tunnels.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: 28 Anatomic ‘Hybrid’ ACL reconstructions using one femoral and two tibial tunnels were done by the senior author. All reconstructions were completed using autograft hamstring. The femoral tunnel was drilled through the anteromedial arthroscopy portal at the center of the femoral attachment site. For the tibial tunnels, two convergent tunnels are drilled in the ACL footprint at the position of the anteromedial and posterolateral bundle insertions. The anteromedial tunnel is referenced off the anterior horn of the lateral meniscus, while the posterolateral tunnel is positioned approximately 2–3 mm from the posterior horn of the lateral meniscus. Using a 6 or 7 mm reamer the two tunnels are created leaving a 2–3 mm osseous bridge intra-articularly and 2.5 cm on the tibial cortex. 2 patients were excluded from the study due to graft failure. Of the remaining 25 patients, 18 have been prospectively followed with Lysholm scores, IKDC scores, KT 1000 measurements, and clinical exam consisting of Lachman, anterior drawer test, pivot shift testing and radiographs. Results: Our graft failure rate was 2/28 (7.1%). 18 patients were followed for an average of 4.9 years (range 3.4–7.5 years), average age at the time of surgery was 24.6 years old (range 14.8 years– 47.4 years). All patients were skeletally mature at the time of surgery. The mean pre-op Lysholm scores were 74.13 (range 28–95), pre-op IKDC scores were 53.4 (range 25.9–75.9) and KT-1000 displacement was 5 mm (range 2–12 mm). At 1 year, mean Lysholm scores had improved to 96.7, IKDC scores were 74.9, and KT-1000 displacement was 1.6 mm. At an average 4.9 years follow up mean Lysholm scores were 89.6 (range 50–100), and IKDC scores were 87.4 (range 50.6–100). One patient who underwent medial and lateral meniscectomies showed radiographic evidence of progressive degenerative changes. The remainder of patients did not show any evidence of radiographic arthritis. Conclusions: Midterm follow up results of the Anatomic ‘Hybrid’ ACL reconstruction are encouraging. The Anatomic ‘Hybrid’ ACL reconstruction using one femoral and two tibial tunnels appears to be a viable, reproducible reconstruction technique.

P21-1486 Quadriceps avoidance gait is more obviously seen in recurrent ACL injured knee than primary ACL injured knee K. Takeda1, Y. Niki1, H. Enomoto1, Y. Suda1, Y. Kiriyama2, T. Nagura2 1 Keio University, Orthopedic Surgery, Tokyo, Japan, 2 Keio University, Clinical Biomechanics, Tokyo, Japan Objectives: The ACL deficiency is considered to lead various instabilities in 6-degrees of freedom. Recent studies have been investigating the influence how the ACL injury affects the knee motion, but they are still controversial. The purpose of this study was to evaluate the change in the three-dimensional gait kinetics of the knees in primary ACL injured patients and recurrent ACL injured patients after reconstruction surgeries. Methods: The subjects were forty-three patients (age: 37.5 ± 9.2 years.) who were diagnosed as unilateral ACL injury by manual tests, MRI and arthroscopy. Six patients of them had recurrent ACL injury after the reconstruction surgery. All patients were going to have ACL reconstruction or re-reconstruction surgery. The average time past injury was 33.0 months with a range of 1–336 months. All the subjects had granted the Institutional Review Board approved informed concent. The measurements were performed using a six-camera system (Qualysis) and force plate (Bertec). Six retro-reflective markers were attached to each subject, at the superolateral aspect of the iliac crest, the lateral aspect of the greater trochanter, the center of ther lateral joint line at the knee, the lateral mallelous, lateral calcaneous and the fifth metatarsal bone. The geometric centers of the hip, knee and ankle were determined from the marker positions and anthropometric measurements, and kinetics of the knee were obtained using an

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 inverse dynamics approach. Each subject requested to walk at their comfort speed. The contralatetal side of the ACL deficient side was also evaluated for comparison. The knee flexion–extension moment, knee adduction-abduction moment and knee rotational moment of the ACL deficient side and healthy side were compared in the primary and recurrent ACL injured patients. An analysis of variance (ANOVA) with a single factor was used to determine the difference (p \ 0.05). Results: ACL deficient knee had significantly less external knee flexion moment than the healthy side during gait in both primary and recurrent ACL injured patients. 67.6% of primary ACL injured group had less external knee flexion moment than the healthy side while 100% of recurrent ACL injured group had less. ACL deficient knee had a tendency to have more external knee abduction moment compared to the healthy side, but the significant difference was not observed. 75.6% of primary group had less or same abduction moment to the healthy side while 100% of recurrent group had less abduction moment. We could not see the tendency in the rotation moment in ACL deficient knee. Conclusions: In this study, the most major gait pattern was knee in less external flexion moment and more external abduction moment, which is known as quadriceps avoidance gait pattern. The pattern was more obviously seen in the recurrent ACL injured group than the primary ACL injured group.

Knee-ACL V

P22-11 Seven to twenty-two years follow-up of Anterior Cruciate Ligament reconstruction: is osteoarthritis rate dependant to approach (trans tibial or trans AM portal)? B. Lee1, J.A. Sim2, J.H. Kwak2 1 Gil Hospital Gachon University, Orthopedic Surgery, Incheon, Republic of Korea, 2Gil Hospital Gachon University, Department of Orthopedic, Incheon, Republic of Korea Objectives: To evaluate the long term outcomes of the ACL reconstruction from the standpoint of osteoarthritis related to femoral tunnel. Methods: We evaluated 31 patients who underwent ACL reconstruction from April 1986 to April 1999 and could be followed-up more than 7 years. Mean follow-up period was 10.1 years (7 * 22 years). In terms of the graft, 11 cases were treated with the ACL reconstruction using a autologous hamstring tendon graft, 20 cases were treated with using a autologous bone patellar tendon bone graft. For femoral tunnel, 11 cases were placed through transtibial tunnel, 20 cases were placed through anteromedial portal using mini-open arthrotomy. Functional and radiographic evaluation was performed. Results: Mean Lysholm score was 89.2 ± 11.7 points. Patients had KT-2000 side-to-side differences were 2.1 ± 1.9 mm. IKDC ligament evaluation showed 38.7% type A, 48.3% type B, 6.5% type C and 6.5% type D. Femoral tunnel were placed at 11 or 1 o’clock position in transtibial technique and placed 10–10:30 or 2–2:30 o’clock position in technique using anteromedial portal. Radiographic analysis for degenerative arthritis revealed that in group using anteromedial tunnel, 50.0% were excellent, 25.0% were good. In group using transtibial tunnel 18.2% were excellent, 18.2% were good. Conclusions: More than 87.1% of cases, long term result of the ACL reconstruction showed good and excellent result in IKDC score. Especially, the group using tunnel through anteromedial portal showed good results for degenerative arthritis than group with transtibial technique.

S213 P22-91 The effect of anatomical single-bundle versus double-bundle ACL reconstruction on medial meniscus-deficient knees: a porcine study J. Zhou1, K. Keklikci2, P. Araujo3, F. Fu4 1 National Institute of Sports Medicine, University, Beijing, China, 2 Gulhane Military Med. Faculty Haydarpasa Teaching Hospital, Department of Orthopaedic Surgery, Istanbul, Turkey, 3University of Pittsburgh, Orthopaedics Surgery, Pittsburgh, United States, 4 University of Pittsburgh Medical Center, Orthopaedic Surgery, Pittsburgh, United States Objectives: The purpose of this study to investigate the effect of single-bundle and double-bundle ACL reconstruction in the medial meniscus-deficient knee. Methods: Sixteen fresh frozen porcine knees were divided into singlebundle and double -bundle reconstructed groups. A robot/universal force moment system (UFS) testing system was used to test anterior tibial translation (ATT), internal rotation (IR) and external rotation (ER). Test was performed on the intact knee, medial meniscus-deficient knee, after ACL resection and single-bundle or double bundle anatomical ACL reconstruction. Results: After ACL transection, ATT increased significantly. Only at 30 of knee flexion was ATT restored to the intact value (DB reconstruction) in the meniscus deficient knee (Fig. 1). Furthermore, the ATT relative reduction rate reduced in DB ACL reconstruction group more than SB group at 60 knee flexion (Table 1; p \ 0.05). The internal rotation (IR) of SB ACL reconstruction group and DB ACL reconstruction group were shown on Table 2. There is no difference for the absolute IR between SB ACL reconstruction group and DB ACL reconstruction group. The IR relative reduction rate after SB ACL reconstruction group were 1.1 ± 2.5% and 2.8 ± 4.0% at 60 and 90 of flexion, while the rate after DB group were 3.6 ± 3.9% and 2.6 ± 2.2% in matched flexion angle. There is no significant difference for the relative reduction rate between the groups. Conclusions: Anatomic DB reconstruction can better restore the intact knee’s ATT than SB reconstruction in the meniscus injured porcine model. Anatomical DB ACL reconstruction showed the trend of reducing IR laxity compared with the SB in medial meniscus deficient model.

P22-98 Tunnel enlargement after anatomic double-bundle anterior cruciate ligament reconstruction using hamstring tendon autografts: a prospective computed tomography study Y. Kawaguchi1, E. Kondo1, J. Onodera1, N. Kitamura1, T. Yagi2, K. Yasuda1 1 Hokkaido University School of Medicine, Sports Medicine and Joint Surgery, Sapporo, Japan, 2Yamanote-dori Yagi Hospital, Department of orthopaedic surgery, Sapporo, Japan Objectives: Bone tunnel enlargement frequently occurs following ACL reconstruction. The presence of large tunnels severely complicates revision ACL surgery. Recently, the authors reported that both the degree and the incidence of femoral tunnel enlargement after anatomic double-bundle ACL reconstruction are significantly less than those after single-bundle reconstruction using a digital radiography. In this study, however, tibial tunnels were not evaluated, because they were overlapped in an X-ray. Therefore, the data of tunnel enlargement after double-bundle reconstruction remain insufficient. The purpose of this study is to clarify the details of the bone tunnel changes including tibial tunnels after double-bundle ACL reconstruction and the relationship between tunnel enlargement and clinical outcomes using a computed tomography (CT).

123

S214 Methods: A prospective cohort study was conducted in patients who underwent double-bundle ACL reconstruction. 35 patients who underwent clinical and CT evaluations at 2 weeks and 1 year after surgery were enrolled in this study. CT of the operated knee was taken in axial, sagittal and coronal views for measuring tunnel changes and evaluating the presence of tunnel coalitions. The tunnel diameter was measured at 1 cm from intra-articular outlet of both femoral and tibial tunnels in each view, perpendicular to the direction of tunnel long axis. In addition, to determine the incidence of tunnel enlargement, the authors defined a tunnel that had a diameter change of more and \10% as ‘enlarged tunnel’ and a ‘reduced tunnel’, respectively. Results: The incidence of femoral AM tunnel enlargement was 23%, 26% and 26% in coronal, sagittal and axial views, respectively, while the incidence of femoral PL tunnel enlargement was 14%, 20% and 20% in coronal, sagittal and axial views, respectively. The incidence of tibal AM tunnel enlargement was 11%, 11% and 9% in coronal, sagittal and axial views, respectively, while the incidence of tibial PL tunnel enlargement was 20%, 6% and 9% in coronal, sagittal and axial views, respectively. The incidence and the degree of femoral tunnel enlargement tended to be more than those of tibial tunnel enlargement. The incidence of tunnel coalition between AM and PL tunnels was 14% in tibia, however, there were no incidences of tunnel coalition in femur. There were no significant relationships between the degree of tunnel enlargement and clinical outcomes. Conclusions: This study showed the details of the tunnel enlargement after anatomic double-bundle ACL reconstruction using CT. According to comparison with previous studies, tunnel enlargement after double-bundle reconstruction tended to be less than that after single-bundle reconstruction. Although tunnel enlargement did not affect clinical outcome, the authors believe that having as little bone enlargement as possible in double-bundle ACL reconstruction greatly minimizes the weakness in bony structure and adverse effects in potential revision surgery.

P22-150 The lateral tibial tunnel in revision ACL surgery. A biomechanical study of a new technique H. Van der Bracht1, L. Verhelst2, Y. Goubau3, P. Verdonk4, J. Bellemans5 1 Gent University Hospital, Gent, Belgium, 2UZ Pellenberg, Leuven, Belgium, Orthopaedic surgery, Pellenberg, Belgium, 3Gent University Hospital, Department Orthopaedic Surgery and Traumatology, Gent, Belgium, 4Gent University Hospital, Orthopaedics, Gent, Belgium, 5University Hospitals Leuven, Catholic University Leuven, Orthopaedic Department, Pellenberg, Belgium Objectives: To evaluate the anatomical possibility to drill a lateral tibial tunnel in ACL surgery. To evaluate the length of a revision lateral tibial tunnel (LTT) in a human cadaver study and to investigate knee stability after a revision ACL reconstruction with a LTT. Methods: 10 human cadaveric knee specimens were used to perform a preliminary investigation to evaluate anatomic considerations. 22 human proximal tibias were used to compare the length of a revision LTT with a classical, medial tibial tunnel (MTT). Another 5 human cadaveric knees were used to investigate knee stability after a revision LTT and to compare it to a primary ACL repair with a MTT performed in the same knees. Translational and rotatory stability were evaluated with computer navigation. Results: A LTT was found to be statistically significant longer (45.0 mm) compared to a MTT (35.2 mm) (p \ 0.001). There was no evidence for a length difference between the intact bone tube length of a revision LTT (36.5 mm) and a MTT. For nearly all measurements, the difference between the ACL repair using a MTT and the revision surgery through a LTT was not only non-significant but also small in magnitude.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: The LTT technique is a new technique for ACL revision surgery. A LTT can easily be drilled and provides a bony tunnel which is statistically significant longer as a MTT. A revision LTT has an intact bone tube as long as a primary MTT. Similar stability is obtained after revision ACL surgery with a lateral tibial tunnel compared to a primary ACL repair with a standard medial tibial tunnel.

P22-151 The consequence of tibial drill-guide angle and drill bit diameter on tibial footprint coverage in ACL surgery H. Van der Bracht1, B.J. Page2, J. Bellemans3, P. Verdonk4 1 Gent University Hospital, Gent, Belgium, 2Stellenbosch University, Tygerberg, South Africa, 3University Hospitals Leuven, Catholic University Leuven, Orthopaedic Department, Pellenberg, Belgium, 4 Gent University Hospital, Orthopaedics, Gent, Belgium Objectives: Different tibial drill-guide angles and drill bit diameters are often used. The purpose of this study was to investigate the consequence of the tibial tunnel diameter and drill-guide angle on tibial ACL footprint coverage. The amount of coverage and the amount of possible overhang of the aperture over the borders of the native tibial ACL footprint were analyzed. Methods: 20 Fresh frozen human knee specimens of \65 years old were used. The ACL was resected at the tibial footprint was coloured with ink. Calibrated digital photographs from the tibial plateau were analysed. 8 Different templates, mimicking the aperture of the tibial tunnel for ACL surgery, were designed. These templates were made for tunnels drilled with drill-guide angles of 45, 55, 65 and 75 and with a tunnel width of 8 and 10 mm. All 8 templates were digitally projected over each of the footprints of the 20 knees. The amount of footprint coverage and the amount of overhang of the templates over the border of the native footprint were calculated. Results: Tibial tunnel inclination and diameter have a significant impact on the amount of footprint coverage. A tibial tunnel drilled with an inclination of \55 will create an aperture with a statistical significant overhang ove the borders of the native tibial ACL footprint. Conclusions: Tibial tunnel inclination and diameter have a significant impact on tibial tunnel footprint coverage. Tibial tunnels which are drilled with an inclination of \55 have a significant risk of hanging over the edges of the tibial ACL footprint and are at risk for harming other structures in the knee.

P22-152 The lateral tibial tunnel in ACL surgery: does it allow for adequate footprint coverage? H. Van der Bracht1, L. Verhelst2, B. Page3, B. Stuyts4, P. Verdonk5, J. Bellemans6 1 Gent University Hospital, Gent, Belgium, 2UZ Pellenberg, Leuven, Belgium, Orthopaedic surgery, Pellenberg, Belgium, 3Stellenbosch University, Tygerberg, South Africa, 4Sintt-Augustinus, GZA Ziekenhuizen, Wilrijk, Belgium, 5Gent University Hospital, Orthopaedics, Gent, Belgium, 6University Hospitals Leuven, Catholic University Leuven, Orthopaedic Department, Pellenberg, Belgium Objectives: The Lateral tibial tunnel (LTT) technique is a new technique for ACL revision surgery. The purpose of this study was to investigate if this technique can safely be performed without harming other structures near the tibial ACL footprint. The hypothesis was that the aperture of a LTT does not cause an increased overhang over the edges of the tibial ACL footprint compared to a MTT when a tunnel with the same width is drilled with the same drill guide angle and drill bit diameter. Methods: 20 human cadaveric proximal tibiae with a mean age of 46 years (21–65 years) were used for this study. Digital pictures of

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 the tibial ACL footprint were taken. Digital templates, mimicking the aperture of a tibial tunnel for ACL surgery drilled with a drill-guide angle of 55 and a width of 8 mm and 10 mm were projected over the footprints of 20 cadaveric knees. The amount of footprint coverage and the amount of overhang of the template over the borders of the ACL footprint were calculated and measured for medial and lateral tibial tunnels. Differences between MTT and LTT were statistically analyzed. Results: No statistically significant differences were found between a LTT and a MTT concerning footprint coverage and possible overhang of the aperture over the edges of the footprint. Conclusions: This study confirms the hypothesis that the LTT technique is safe and feasible concerning tibial footprint coverage because similar footprint coverage and similar overhang of the aperture over the boundaries of the ACL footprint were found when comparing this technique with the standard technique.

P22-326 The effects of different hormonal influences on knee stiffness and proprioception T. Ba´nyai1, G. Cerulli2, J.W.-P. Michael1 1 Clinic for Orthopaedic, Trauma and Spinal Surgery Marienhaus, Bendorf, Germany, 2Let People Move Biomechanical Laboratory, Perugia, Italy Objectives: Women suffer 4–8 times the anterior cruciate ligament (ACL) rupture rate for the same sport as males. Our study group in Perugia (Italy) analyzed the hormonal factor of the possible reasons between the menstrual phases, in the second part of our measurements in Szeged (Hungary) we analyzed pregnant women because they have significantly higher sexual hormone levels. In the third part we measured the effect of oral contraceptives on knee laxity. The aim of our recent paper was the summarize our results. Methods: In the ‘‘Let People Move’’ Biomechanical Laboratory in Perugia, Italy: 9 healthy female volunteers, average age 26.3 (21–32) average BMI (body mass index) 20.08 volt. We measured the parameters in three phases of menstrual cycle. Measuring ACL stiffness: - KT2000 Measuring proprioception: -Stabilometry: active and passive balance control with opened and closed eyes, detecting sway path, splitting to anterior-posterior and lateral axes. -Joint position sense Muscle activity: We used surface EMG combined with Kin-Com isokinetic dynamometry (peek torque and maximal force). In the Department of Obstetrics and Gynecology of SZTE University in Szeged, Hungary: 22 pregnant and 18 non pregnant women, the two groups were comparable (normal distribution according to body mass index and age). In the Department of Traumatology of the Teaching Hospital of SZTE University, Kecskemet, Hungary: We analyzed 19 female volunteers, 8 of them used regularly oral contraceptive (OCP) triphasic pills, 11 never used it. Results: During the menstrual cycle there was no significant difference between the measured parameters. The KT-2000 arthrometer measured significal differences between the elongations of ACL in the pregnant and control groups. The non pregnant group sensed significantly better the joint positions. With opened eyes the pregnant had significantly less total sway, with closed eyes there was no significant difference. Splitted to AP

S215 and lateral axes in pregnancy with closed eyes there was significantly bigger instability in the anterior-posterior axe. The KT-2000 arthrometer measured no significant difference between the groups (knee joint stiffness). During the joint position sense there was no significant difference between the OCP users and the control group. Splitted to AP and lateral axes the OCP users had with closed eyes significantly higher stability in the anterior-posterior axe. In medial–lateral axe there was no significant difference. Loosing the eye control decreased the balance-keeping capability of the non OCP users. Conclusions: The decreased proprioceptive capacity of the knee can increase the vulnerability of ACL. We continue the measurements to clarify this component and prepare the adequate prevention for the sport during pregnancy and state with higher sexual hormone level (oral contraceptives).

P22-361 Tunnel bone density after allogenic bone chip and bone cylinder transplantation in staged revision anterior cruciate ligament reconstruction S. Said1, P. Faunoe2, B. Lund3, S.E. Christiansen4, M. Lind5 1 Aarhus University Hospital, Aarhus, Denmark, 2Aarhus University Hospital, Department of Sports Traumatology, Aarhus, Denmark, 3 University Hospital of Aarhus, Department of Sportstraumatology, Aarhus C, Denmark, 4University of Aarhus, Sports Trauma Division, Orthopedic Department, Skanderborg, Denmark, 5Div of Sportstrauma, Aarhus University Hospital, Aarhus C, Denmark Objectives: Tunnel widening in failed anterior cruciate ligament reconstruction (ACLR) can result in the staged revision procedures with a need for bone transplantation prior to revision reconstruction. Limited knowledge exists regarding to quality of different transplantation methods. The present study used CT-scanning to evaluate tunnel bone density after allogenic bone transplantation using either bone chips or bone cylinders. The hypothesis of the study is bone chips transplantation resulted in higher bone density than bone cylinder transplantation due to possible voids between individual cylinders in the tunnels. Methods: The records of 36 patients operated for 1st stage revision ACLR from January 2003 to August 2010 were included in the study. Twenty-one patients had their tunnels transplanted with bone chips and fifteen patients with bone cylinders from allogenic femoral heads. Allograft bone chips were produced using a bone mill and allograft cylinders were created by 7–8 mm diameter core drilling. Bone density 3–4 months after transplantation were evaluated by CT scanning reconstruction slides with 5 mm intervals throughout the tunnel length using histomorphometry. Results: There were 24 females and 12 males with an average age of 32 years. In femoral bone tunnels, bone density was 66 and 51% when using bone chips and bone cylinders respectively (p = 0.03). In tibial bone tunnels bone density was 60 and 58% when using bone chips and bone cylinders respectively (NS). Conclusions: Bone density after chip allograft bone transplantation in revision ACL reconstruction was superior compared to cylinder allograft bone transplantation.

P22-560 The potential risk of the posterior blowout using far anteromedial portal drilling for anatomical double bundle anterior cruciate ligament reconstruction Y. Nishimori1, M. Nozaki1, M. Kobayashi1, H. Goto1, H. Iguchi1, T. Otsuka1 1 Nagoya City University, Orthopaedic Surgery, Nagoya, Japan

123

S216 Objectives: Recent studies have demonstrated the advantage of far anteromedial portal (FM) drilling in anatomical double bundle anterior cruciate ligament (ACL) reconstruction regarding its accuracy of anatomical tunnel placement. However, many studies have demonstrated the potential risk of FM drilling technique such as short femoral tunnel, peroneal nerve injury, and posterior cartilage damage. Especially the anteromedial femoral tunnel tends to be oblique which may cause the posterior blowout in cases which have narrow intercondylar notch. The purpose of this study was to evaluate the potential risk of the posterior blowout of FM drilling technique for anatomical double bundle ACL reconstruction using a three-dimensional (3D) computed tomography (CT) model. Methods: CT scans were performed on 4 knees (3 male, 1 female, mean age 40.2) with a slice spacing of 0.75 mm. The femoral bone was segmented from the axial CT scan slices using Mimics (Materialise, Leuven, Belgium) and was processed into 3D surface models using Magics (Materialise, Leuven, Belgium). A cylinder with a diameter of 6 mm was created on the 3D model as the femoral anteromedial tunnel. Tunnel placement of the anteromedial bundle in far anteromedial portal drilling is determined using the established anatomic double bundle (anteromedial and posterolateral) tunnel positions. And tunnel angle is determined using the margin which is not touched a medial meniscus and a medial condyle. The plane for measurement was made by passing through the center line of the anteromedial tunnel model tangential to the diameter of the cylinder. We measured the maximum distance between the anteromedial tunnel and the posterior femoral cortex on the measurement surface (Fig. 1). The measurement knee flexion angles were 90 and 120. We compared the distance with 90 and 120 flexion. Statistical analysis was performed using the t test. Statistical significance was defined as p \ 0.05. Results: The average distance for 90 flexion was 1.9 mm, and the average distance with 120 flexion was 4.5 mm. A significant difference was observed in the distance between 90 flexion and 120 flexion. Conclusions: This study suggests that the distance between the femoral anteromedial tunnel and the posterior femoral cortex is short in anatomical double bundle ACL reconstruction using far anteromedial portal drilling. Furthermore, the distance with 90 is shorter than with 120. This suggests that the high flexion angle of the knee may decrease the risk of the posterior femoral cortex damage. In conclusion, FM drilling technique for anatomical double bundle ACL reconstruction with low flexion angle have the potential risk of the posterior blowout.

Fig. 1

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P22-570 Transtibial anatomic ACL reconstruction can lead to short tibial tunnel and medial tibial chondral injury C.B. Chang1, J.-Y. Choi2, J.H. Lee1, T.K. Kim1 1 Seoul National University Bundang Hospital, Orthopaedic Surgery, Seongnam-si, Republic of Korea, 2Seoul National University College of Medicine, Radiology, Seoul, Republic of Korea Objectives: Although independent drilling methods, such as anteromedial portal technique, have received attention for anatomical anterior cruciate ligament (ACL) reconstruction, transtibial technique is still widely used because of its practical aspect. However, for anatomical ACL reconstruction via the transtibial technique, more proximal tibial tunnel starting point would be required which may cause very short tibial tunnel and/or tibial subchondral bone damage due to horizontal nature of the tibial tunnel. This cadaveric study aimed to evaluate the length of the tibial tunnel and risk of the subchondral bone damage in anatomical ACL reconstruction via the transtibial technique, and to investigate the relationship of the native ACL tibial attachment location in the sagittal plane with the length of the tibial tunnel and risk of the subchondral bone damage. Methods: In 20 cadaveric knees, the centers of ACL tibial and femoral attachments were marked with metal pins. Then, full extension anteriorposterior and lateral radiographs, 80 flexion lateral radiographs, and prone kneeling views were obtained. In full extension lateral film, the locations of the ACL tibial attachment center in the sagittal plane were assessed using the line described by Staubli et al. In 80 flexion lateral radiographs and prone kneeling view, we estimated the optimal starting point and angle of tibial tunnel to achieve anatomical femoral tunnel using the transtibial technique. Then, the guide pin was inserted from the estimated tibial staring point and made the tibial tunnel using 8 mm tibial reamer. We measured the length of the tibial tunnel and examined the presence of the subchondral bone damage with area of [1 cm2. The relationship of the native ACL tibial attachment location in the sagittal plane with the length of the tibial tunnel and risk of the subchondral bone damage were analyzed the multivariate regression analysis and the Mann–Whitney U test. Results: The average length of tibial tunnel for anatomical reconstruction was 29 mm (SD: 3.1, range: 20 mm * 35 mm). Among the 20 knees examined, subchondral bone damage [1 cm2 was observed in the 10 knees (50%). The sagittal position of the native ACL tibial attachment was highly correlated with the length of tibial tunnel (regression coefficient = 0.81, p \ 0.001). The knees with subchondral bone damage had significantly more anteriorly placed ACL tibial attachment than the knees without subchondral bone damage (39 vs. 42%, respectively, p \ 0.02). Conclusions: This study found that the tibial tunnel for anatomical ACL reconstruction via transtibial technique could frequently lead to short tunnel length and subchondral bone damage due to horizontal nature of the tunnel. Although the modified transtibial technique is a feasible method in some knees with relatively posterior native ACL tibial attachment, independent drilling methods would be a more reliable option to achieve anatomical ACL reconstruction.

P22-683 The effect of coronal location of the tibial anteromedial tunnel in anatomic double-bundle anterior cruciate ligament reconstruction on knee kinematics S. Asai1, Y. Hoshino2, D. Kim1, V. Pronesti3, P. Smolinski3, F.H. Fu1 1 University of Pittsburgh–School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States, 2Pittsburgh University, Orthopaedics, Pittsburgh, United States, 3University of Pittsburgh, Department of Mechanical Engineering and Materials Science, Pittsburgh, United States

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1 ATT under 89 N anterior tibial load

Objectives: The objective of this study was to evaluate knee kinematics after two different tibial AM placements of medio-lateral (ML) alignment in DB ACL reconstructions. Our hypothesis was that the tibial ML position of AM tunnel would have a significant effect on knee kinematics of the ACL graft. Methods: Three fresh frozen human cadaveric knees were used in this study. To determine knee kinematics, a robotic system (CASPAR Sta¨ubli RX90, OrthoMaquet, Rastatt, Germany) that combines with a universal force-moment sensor (model 4015, JR3 Inc, Woodland, California) was used to measure the 6 degree-of-freedom knee kinematics. An anterior tibial load (89 N) at 0, 30, 60 and 90 of knee flexion was applied to the tibia while measuring the knee kinematics. The loading and simultaneous measurement was repeated in the ACL intact, ACL deficient and two types of reconstructed knees. Two different types of DB ACL reconstructions were performed. In the medial AM reconstruction group the 6 mm tibial AM tunnel was located at the medial part of the AM footprint, while the 6 mm tibial PL tunnel was created on the center of the PL footprint. The 6 mm femoral tunnels for both bundles were then placed in the anatomic positions. In lateral AM reconstruction the same specimen was used after filling the tibial AM tunnel by an epoxy compound. The 6 mm tibial lateral AM tunnel was located at the lateral part of the AM footprint, while the same tibial PL tunnel was used as medial AM reconstructed knee. The femoral tunnels were not changed. The hamstrings grafts were passed through both AM and PL tunnels. On the femoral side, the graft was fixed with an EndoButton CL (Smith & Nephew Endoscopy, Andover, Massachusetts), while on the tibial side with a screw and washer. The AM graft was fixed at 30 of knee flexion and the PL graft was fixed at 0 of knee flexion while applying 20 N tension on each graft. Results: The anterior tibial translation (ATT) of the ACL-deficient knee was greater than that of the intact knee at all selected flexion angles. Both DB reconstructions reduced the ATT from the ACLdeficient condition. After medial AM reconstruction, the ATT was closely restored to the intact level at all flexion angles. After lateral AM reconstruction, the ATT was less than the intact level at 0 and 30 of knee flexion (Fig. 1). Conclusions: The finding of this study was that the different coronal location of the AM bundle on the tibia in the DB ACL reconstruction resulted in different knee kinematics, and that the medially placed AM bundle better restored normal knee kinematics.

S217 P22-702 The pivot shift: a global user guide V. Musahl1, Y. Hoshino2, P. Araujo2, M. Ahlde´n3, S. Zaffagnini4, F.H. Fu1 1 University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States, 2University of Pittsburgh, Department of Orthopaedic Surgery, Pittsburgh, United States, 3Sahlgrenska University Hospital/Mo¨lndal, Mo¨lndal, Sweden, 4 Istituto Ortopedico Rizzoli Bologna, Bologna, Italy Objectives: The purpose of this study was to demonstrate several different techniques for the pivot shift exam. Methods: Twelve international expert surgeons performed the pivot shift test on a whole lower extremity cadaver. The ACL and anterior horn lateral meniscus of the right knee was transected to produce a grade 2 pivot shift (Lane and Warren criteria). Surgeons performed and graded the pivot shift clinically. Video motion analysis was used to capture each individual surgeon’s technique. Motions were analyzed; knee flexion, valgus, internal-external (i.e) rotation, and anterior-posterior (ap) drawer. For analysis, the pivot shift was broken down into 1. fixed rotation versus motion allowing, 2. high force versus low force, and 3. reduction versus dislocation. Results: The average clinical grading of the pivot shift was 2.3 (range 1.5–3). Two surgeons used a dislocation-type maneuver. One surgeon utilized a fixed anterior drawer (Dejour type). The remaining 9 surgeons utilized valgus motion with internal rotation (5), external rotation (1), or motion allowing technique (3). Of the 5 surgeons that utilized fixed internal rotation, 2 used high valgus force with maximum internal rotation and 3 used a low valgus force with small internal rotation. There was no difference in average clinical grading when using high force (2.5 ± 0.6) versus low force (2.3 ± 0.5, n.s.), or using fixed rotation (2.2 ± 0.5) versus motion allowing (2.3 ± 0.6; n.s.). Conclusions: This study showed that there is no consensus on the optimal technique for pivot shift examination. Clinical grading varies amongst surgeons, but not between different techniques. High forces and extremes of rotation are not necessary to produce a clinical detectable pivot shift. In the future, a standardized pivot shift test may be beneficial when comparing outcome following ACL reconstruction.

P22-722 The values of C-reactive protein as predictors of infection after anterior cruciate ligament reconstruction M.A. Ruiz-Iban1, J. Diaz-Heredia1, I. Cebreiro Martinez del Val1, H. Gomez Santos1, F. Aranda1, F. Gonzalez-Liza´n1 1 Hospital Ramo´n y Cajal, Department de Cirugı´a Ortope´dica y Traumatologı´a, Madrid, Spain Objectives: to find out what are the normal PCR values in the first weeks after anterior cruciate ligament reconstruction (ACLR). Methods: A total of 101 ACLR (in 99 patients, 65 males and 34 females, mean age 26.4 ± 5.4 years) were enrolled. Of these 86 were primary ACLR, 14 were reconstructions after a previous failed ACLR, and 45 of them associated meniscal repairs. Surgery was performed by 6 different orthopaedic surgeons with widely variable experience. The surgical technique was autologous T4-isquitibialis tendon reconstruction in 85% of cases. CRP determinations were performed preoperatively and 1, 7, 14, 21 years 28 days after surgery. A range of CRP between 0 and 5 mg/l was considered normal. One patient developed an infection in the second postoperative week and was excluded from the analysis.

123

S218 Results: The mean values obtained were: preoperatively: 1.75 ± 2.4. At 24 h: 7.4 ± 7.6. At 7 days: 6.7 ± 8.2. At 14 days: 3.42 ± 2.7. At 21 days: 3.2 ± 3.4. At 28 days: 3.99 ± 3.8. In 21% of cases the CRP values did not increase over normal values at any time after surgery. Values over 15 mg/l appeared only in 13% of cases at the 24 h determination, in 8% at 7 days and in 2% at 14 days. Values over 30 mg/l appeared only in 4% of cases at the 24 h determination, in 2% at 7 days and in 0% at 14 days. Conclusions: After ACLR a CRP value over 15 mg/l is suggestive of infection after the first week. Values over 30 mg/l are highly suggestive of infection, specially after the first week.

P22-815 Value of oblique coronal and oblique sagittal magnetic resonance images in diagnosis of anterior cruciate ligament tear and evaluation of anterior cruciate ligament remnant M. Kosaka1, J. Nakase1, T. Toratani1, Y. Ohashi1, K. Kitaoka2, H. Tsuchiya1 1 Kanazawa University, Orthopaedic surgery, Kanazawa, Japan, 2 Kijima Hospital, Orthopaedic Surgery, Kanazawa, Japan Objectives: In recent years, anterior cruciate ligament (ACL) remnant has attracted considerable attention in the treatment of ACL tear. The purpose of this study was to investigate the efficacy of additional oblique coronal and oblique sagittal magnetic resonance imaging (MRI) for diagnosis of ACL tear and evaluation of ACL remnant. Methods: We retrospectively reviewed the records of 54 patients (26 male and 28 female, age range: 12–66 years, mean age: 26.9 years) who underwent both MRI of the knee and arthroscopic examination between August 2010 and September 2011. The MR examinations included routine sequences and two sets of oblique images. Five independent readers evaluated the MRI images by the use of three methods: orthogonal sagittal images only (method A); orthogonal sagittal and additional oblique sagittal images (method B); and orthogonal sagittal and oblique coronal images (method C). The status of the ACL was graded from MRI images as intact ACL, complete tear or tear with continuous ACL remnant. The sensitivity, specificity, and accuracy for the diagnosis of an ACL tear and the detection of an ACL remnant with the use of each method were calculated in comparison with arthroscopic findings as the reference standard. Results: The arthroscopic records revealed 27 patients with intact ACLs and 27 with torn ACLs. Among the 27 patients with torn ACLs, 9 didn’t have a continuous remnant (complete tear) and 18 had a certain remnant attaching to the femur or the posterior cruciate ligament. The sensitivities, specificities, and accuracies for the diagnosis of an ACL tear were as follows, respectively: method A, 85.9%/ 76.3%/81.1%; method B: 87.4%/88.1%/87.4%; and method C: 89.6%/90.4%/90%. The sensitivities, specificities, and accuracies of detecting an ACL remnant were as follows, respectively: method A, 51.1%/70.5%/64.1%; method B: 71.1%/81.1%/77.8%; and method C: 67.8%/85.6%/79.6%. Concerning the diagnosis of an ACL tear, the specificities and accuracies for methods B and C were significantly higher than those for method A. In terms of the detection of an ACL remnant, the sensitivities, specificities and accuracies for methods B and C were significantly higher than those for method A. The specificity for method C was higher than that for method B. Conclusions: Additional use of oblique coronal and oblique sagittal MRI of the knee improved the accuracy for the diagnosis of an ACL tear and showed a certain level of efficacy in detecting an ACL remnant. The oblique coronal images, parallel to the femoral intercondylar roof, which revealed ACL as continuous from the femoral origin to tibial insert and clearly depicted ACL especially in the femoral origin area, may provide further improvement in the diagnostic efficacy for the ACL remnant and help in making appropriate decisions for treatment of ACL injuries.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P22-837 The use of an accessory medial portal in anatomic ACL reconstruction. A prospective CT-study V. Chouliaras1, D. Giotis1, C. Christogiannis1, V. Koukounas2, A. Grestas1 1 General Hospital of Arta, Orthopaedic Department, Arta, Greece, 2 General Hospital of Arta, Radiology Department, Arta, Greece Objectives: The last years an accessory medial portal has been proposed for a true anatomic ACL reconstruction. The purpose of this study was to evaluate the position of the femoral tunnel with this new method in comparison with a simple anteromedial portal by the aid of the CT and the 3-D CT. Methods: Twenty-five patients underwent an ACL reconstruction by the aid of an accessory medial portal (Group A) whereas other 25 patients who underwent the same operation without the use of this accessory portal (Group B) was the comparative group. In both groups the femoral tunnel was performed through the anteromedial portal. The position of the tunnel was calculated by the aid of a CTscan, as an angle of the femoral tunnel to a line tangential to the tibial plateau in the coronal level. Results: The statistical analysis was performed with the SPPS system. We used the t test for the two groups. With the accessory portal the mean angle of the femoral tunnel was 34.07 ± 2.02. Without the accessory portal the mean angle of the femoral tunnel was 49.0 ± 4.82. This difference in the femoral tunnel position was statistically significant. In addition by the aid of the 3-D CT we confirmed that in Group A all the femoral tunnels where in an anatomical position while this was not happened in all cases in Group B. Conclusions: We confirmed that by the usage of an accessory medial portal we can perform the femoral tunnel in a more oblique place, more close to the real anatomical position of the natural ACL. Future studies by the aid of gait analysis will reveal a possible clinical superiority of this new method.

P22-889 The influence of the cross sectional area of the ACL graft on the postoperative knee laxity M. Kusano1, Y. Yonetani1, Y. Tanaka1, T. Kanamoto1, S. Horibe1 1 Osaka Rosai Hospital, Orthopaedic Surgery, Sakai, Japan Objectives: Although an initial tension applied to the anterior cruciate ligament (ACL) graft influences the knee laxity, it is unclear whether it should be adjusted according to graft size to restore anterior laxity. The objective of this study was to evaluate a relationship between the cross-sectional area (CSA) of the ACL graft and the anterior laxity after anatomical ACL reconstruction. Methods: Anatomical ACL reconstruction by use of a bone patella tendon bone (BTB) graft with initial tension of 30 N was performed in 21 patients (20 men and 1 woman). Intraoperative CSA was measured with a custom-made area micrometer. After operation, anterior laxities of contralateral normal and postoperative injured knees were measured under general anaesthesia with the Rolimeter (Aircast Europa, Neubeuern, Germany), which was a device designed for measuring knee instability, at 30of knee flexion with maximal manual load. Side to side difference between the normal and the postoperative knees and the ratio of its difference to the normal knee were used for anterior knee laxity parameters. Pearson’s productmoment correlation coefficient was calculated to assess relation between these two values and the CSA of the BTB graft respectively. Results: The mean CSA of the BTB graft was 34.6 ± 4.6 mm2. The mean difference between the contralateral normal and the postoperative knees was 2.86 ± 2.35 mm. The mean ratio of its difference to the normal knee was 38.0 ± 29.8%. Pearson’s product-moment correlation showed moderate negative relationship between these two values and the CSA of the patella tendon. The correlation coefficient

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S219

between the CSA and the side-to-side difference was found r = 0.52 (p = 0.015). The correlation coefficient between the CSA and the ratio of its difference to the normal knee was found r = -0.58 (p = .005). Conclusions: Under the condition of fixed initial tension, the smaller CSA of the BTB graft is, the bigger constraint of anterior laxity is obtained after ACL reconstruction. Not only the initial tension but also the CSA of the ACL graft should be paid to attention to obtain desired anterior laxity.

P22-930 The effects of anatomic and non-anatomic posterolateral bundle augmentation on knee rotational stability: A biomechanical study in porcine knees C. Yapici1, K. Keklikci2, D. Kim1, M. Linde-Rosen3, P. Smolinski4, F. Fu5 1 UPMC, sport medicine, Pittsburgh, United States, 2Gulhane Military Medical Faculty Haydarpasa Teaching Hospita, Department of Orthopaedic Surgery, Istanbul, Turkey, 3University of Pittsburgh, Pittsburgh, United States, 4University of Pittsburgh, Department of Mechanical Engineering, Pittsburgh, United States, 5University of Pittsburgh Medical Center, Orthopaedic Surgery, Pittsburgh, United States Objectives: The partial ACL injury may leave one bundle intact. The PL bundle tear patients may complain of rotational instability with pivoting sports. The anterior cruciate ligament (ACL) of the porcine knee consists of three bundles; AM (antero-medial), PL (posterolateral) and IM (intermediate). However the IM bundle has a minor role in porcine knee stability and the PL bundle has the most important contribution for rotational stability (80.4%) in porcine knee model. The purpose of this study is to compare knee rotational stability between anatomic and non-anatomic PL bundle augmentation. Methods: Nine fresh frozen mature porcine knees were divided into two groups. 1- Intact ACL, 2- Deficient PL and IM, 3- Anatomic PL bundle augmentation, 4- Non-Anatomic PL bundle augmentation (high position) All the surgical reconstructions were done using a three portal technique via arthroscopy and a robotic/universal force sensor (UFS) system was used for testing. An 89-N anterior tibial load was applied, and an anterior tibial translation (ATT) was measured at knee flexion angles of 30, 45, 60, and 90 and a rotatory tibial load of 4-Nm external rotation (ER) and internal rotation (IR) were applied at 30, 45 and 60 of knee flexion. Range of motion was measured after anatomic and non-anatomic PL bundle augmentation. Data (mean ± standard deviation) obtained from the different knee conditions were compared and, was analyzed with (Prism Graph Pad Version 5.0a) for two-tailed Paired t test with the level of significance set at p \ 0.05. Results: IR and ER were increased at all flexion angles after bundles resection and non-anatomic PL bundle augmentation. However, there was a significant difference between intact ACL and partial deficient ACL (AM intact) in IR and ER at all flexion angles (p \ 0.05). There was a significant difference between anatomic and non-anatomic PL bundle augmentation in IR at 30 and 60 of knee flexion (p [ 0.05) (Fig. 1). However there was no significant difference in ER at 45 (p [ 0.05). There was a limited flexion after non-anatomic PL bundle augmentation and there was a significant difference between anatomic and non-anatomic PL bundle augmentation for knee flexion (p \ 0.05). Conclusions: The anatomic PL bundle augmentation augmentation provides knee rotational stability better than non-anatomic PL bundle augmentation. Clinically, a patient with high PL graft may have adequate antero-posterior stability but inadequate rotational stability. The reasonable choice is to protect the intact bundle and anatomic reconstruct the torn bundle.

Fig. 1 The effect of graft position in response internal

P22-1041 The optimum far anteromedial portal position for trans-medial portal technique of ACL reconstruction by pre-operative 3D-CT taken at the knee flexion position A. Eguchi1, N. Adachi1, M. Deie1, M. Nishimori2, A. Nakamae1, M. Ochi1 1 Hiroshima University, Department of Orthopaedic Surgery, Hiroshima, Japan, 2Mitsubishi Mihara Hospital, Mihara, Japan Objectives: In the far anteromedial portal (FAMP) anterior cruciate ligament (ACL) reconstruction technique, there is a risk of damaging the cartilage of the medial femoral condyle during drilling of the femoral tunnels as opposed to the advantage of making circularly holes within the femoral attachments of the anteromedial bundle (AMB) and posterolateral bundle (PLB). Thus, it is very important to make FAMP as vertical as possible to the femoral intercondylar wall but enough to avoid iatrogenic cartilage damage in the medial femoral condyle. The aim of this study is to ascertain the optimum safety location of the FAMP and to examine the related factors which have influence on the location by using preoperative three dimensional-CT (3D-CT) taken at the knee flexion position as simulated ACL reconstruction in the FAMP ACL reconstruction technique. Methods: Thirty-two patients preparing to undergo ACL reconstruction underwent pre-operative CT scans taken at 120 of knee flexion position, which is the same position of intraoperative knee flexion when drilling femoral tunnels. 3D-CT reconstruction of the knee was performed using the volume rendering technique. We marked the anteromedial (AM) and the posterolateral (PL) position of the anatomical ACL attachment of the femur on the 3D-CT sagittal plane, and drew the virtual line that was tangent to the medial femoral condyle from the marked AM and PL position. We termed the intersection of the line to skin surface as optimum FAMP position (OPP), and measured length from this OPP to the medial edge of patella tendon (OPP-PT). Width of transepicondylar axis (TEA) and transverse diameter of intercondylar notch (ICN) were measured on 3D-CT bone model. The relation of those parameters and OPP-PT were calculated from Spearman0 s rank correlation coefficient. Results: The mean length of OPP-PT was 31.0 ± 4.1 mm with a range from 22.6 to 38.9 mm in the all 32 patients. The mean length was 31.9 ± 4.0 mm in men, 29.8 ± 4.2 mm in women. The mean width of TEA was 78.6 ± 5.9 mm, the mean length of ICN was 16.7 ± 2.2 mm, the ratio of TEA to ICN was 21.2 ± 2.5%. There was a significant correlation between the length of OPP-PT to the width of TEA (p \ 0.05). Conclusions: Correlation with height and sex has been shown by our past examination. Combined with this study, it was shown that the length of OPP-PT was related to an individual somatotype. In ACL

123

S220 reconstruction, it is very important to produce the optimum FAMP more safely by using pre-operative 3D-CT.

P22-1048 Volume and contact surface area analysis of bony tunnels between single and double bundle anterior cruciate ligament reconstruction using autograft tendons -In vivo three-dimensional imaging analysisJ.-H. Yang1, J.G. Kim2, J.H. Wang3 1 Seoul Veterans Hospital, Department of Orthopaedic Surgery, Seoul, Republic of Korea, 2College of Medicine, Korea University, Guro Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea, 3Sungkyunkwan University School of Medicine, Samsung Medical, Seoul, Republic of Korea Objectives: The purpose of this study was to analyze the tunnel volume and surface area of transtibial SB versus transportal DB ACL reconstruction with hamstring autograft using a three-dimentional (3D) CT. Methods: Consecutive series of 26 patients who underwent trantibial SB ACL reconstruction and 27 patients with transportal DB ACL reconstruction using hamstring autograft from January 2010 to October 2010 were included in this study. 3D-CTs were taken within 1 week after operation. The bone was segmented with use of Mimics software v14.0. The obtained digital images were then imported in the commercial package Geomagic  Studio v10.0 and Sketchup Pro v8.0 for processing. The tunnel length, diameter, volume and surface area were evaluated. Mann–Whitney U test was used for comparison. A p \ 0.05 was considered significant. Results: Significant difference between the femoral tunnel volume for SB technique (1,496.51 ± 396.72 mm3) and the total femoral tunnel volume for DB technique (1,593.81 ± 469.42 mm3) was not found (p = 0.366). However, total surface area for femoral tunnels were larger in DB technique (p = 0.004). For tibia, the total tunnel volume and the surface area were larger in DB technique (p B 0.001, p B 0.001 respectively). Conclusions: The bone losses for tunnel preparation in DB tend to be larger than the SB ACL reconstruction technique. However, the contact surface areas for healing were larger in DB than the SB tunnels.

P22-1075 Validation of electromagnetic measurement system during the pivot shift test in a cadaveric model: A comparison of cutaneous and transosseous electromagnetic position sensors Y. Nishizawa1, R. Kuroda1, T. Matsushita1, S. Kubo1, K. Nagamune2, M. Kurosaka1 1 Kobe University Graduate School of Medicine, Orthopaedic Surgery, Kobe, Japan, 2University of Fukui, Department of Human and Artificial Intelligent Systems, Graduate School of Engineering, Fukui, Japan Objectives: The pivot shift test is commonly used for assessing dynamic instability in anterior cruciate ligament insufficient knees, which is related to knee function. In past, no method existed which was able to evaluate pivot shift test quantitatively and noninvasively. Recently, we have reported about an electromagnetic measurement system which enables evaluation of the pivot shift test quantitatively and noninvasively. However, in that system we attached the femoral and tibial sensors to skin, using a plastic brace, and measurement accuracy (bone to skin motion) have not fully addressed. So in this

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 study, we have evaluated bone to skin motion during the pivot shift test, comparing the translation and acceleration of reduction obtained simultaneously from two sets of sensors (one attached to the skin braces and another attached to bone) during the pivot shift test. The purpose of this study was to validate electromagnetic measurement system by comparing cutaneous with transosseous electromagnetic sensors during the pivot shift test. Methods: Two fresh-frozen cadaveric knee (whole leg) specimens (left/right) with skin and soft tissues preserved were employed for this study. Preliminarily anterior cruciate ligament was cut arthroscopically. Pivot shift test were performed on these knees by 12 different knee expert surgeons attended Panther Global Summit 2011, Pittsburgh, U.S.A. Tibial translation and acceleration of reduction during the pivot shift test were measured using an electromagnetic measurement system. We measured 166 times of tibial translation and acceleration of reduction. During the measurement, one set of the femoral and tibial sensors were attached to the braces fixed on skin. Simultaneously, another set were rigidly fixed to a transosseous pin. Differences in measurements were assessed and correlation between cutaneous data and transosseous data. Results: Significant differences were found with tibial translation (10.3 mm vs. 20.3 mm, p \ .01) and acceleration (2.5 m/sec2 vs. 3.3 m/sec2, p \ .01) between cutaneous and trnasosseous measurement. However, correlation(R) between cutaneous and transosseous measurements in tibial translation was .705 (p \ .0001), and in acceleration was .890 (p \ .0001). Conclusions: In complex motion as the pivot shift test, tibial translation and acceleration of reduction measured by cutaneous electromagnetic sensors were underestimated when compared to transoseous sensors. However, correlation between cutaneous and transosseous measurements in tibial translation and acceleration of reduction was significantly strong. Therefore, measurements using cutaneous electromagnetic sensors during the pivot shift test were validated and useful for assessing clinical conditions of ACL deficient knees.

P22-1078 The effects of trans-tibial versus anteromedial portal technique on the stress patterns around the femoral tunnel in anatomical single-bundle ACL reconstruction at different knee flexion angles using a finite element analysis Y.-J. Seo1, Y. Yoo2, S.Y. Song3, H.J. Kim4, H.Y. Kim4 1 Hallym University, Orthopaedic Surgery, Hallym Sports Medicine Research Group, Seoul, Republic of Korea, 2 Hallym University, Chuncheon, Republic of Korea, 3 Hallym University, Seoul, Republic of Korea, 4 Gangwon University, Mechanical Engineering, Chuncheon, Republic of Korea Objectives: The purpose of this study was to compare the differences in stress patterns around the femoral tunnel that was created by anteromedial portal or trans-tibial tunnel technique in anatomical ACL reconstruction at different knee flexion angles. Methods: Twelve male subjects’ right knees were scanned with a high-resolution CT scanner with 1 mm slices at four different knee flexion angles (0, 45, 90 and 135). The subjects’ knees were randomized into two groups with a equal number, group A and B. The AM portal was marked with a coin on the knees in group A during CT scanning. The CT images were then processed to obtain the 3D models. And several modeling programs were used to create, manipulate, and analyze the 3D model.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S221

In group A, the femoral tunnel of 10 mm diameter was created in 135 flexion model from the AM portal which was previously marked with a coin. And six of the knees in group B were drilled through trans-tibial tunnel technique with a same diameter at 90 of flexion. The tibial tunnel was made on center point of tibial footprint respectively. The bone and the graft were assumed as isotropic, homogenous. And linear elastic property of the bone was adopted from the data of cancellous bone which was previously reported. The graft which was modelled as a nonlinear hyperelastic material was inserted into corresponding bone tunnel and fixated at the outer aperture, middle and inner orifice of the femoral tunnel. The graft was fixated at the middle in the tibial tunnel with a set of 40 N pretension at full extension. Hence three types of single bundle ACL reconstruction were conducted in each group respectively. Finally stress patterns around the femoral tunnels were determined at different knee flexion angles using a finite element analysis performed with ABAQUS/Explicit code (Figs. 1, 2, 3). Results: In general, the stress occurring around the femoral tunnel was found to increase with increasing knee flexion. Regardless of the Fig. 3 Maximal stress value around the femoral tunnel at different knee flexion angles. (Hole A: femoral tunnel through AM portal technique, hole B: femoral tunnel through trans-tibial tunnel technique)

grafts’ fixation level in the femoral tunnel, the mean maximum stress value of 0.86 (± SD) MPa was monitored at 45 of flexion in group A. And that of group B was 0.6 MPa at same flexion angles with no significant difference. Meanwhile, that of group A was significantly lower than that of group B at 90 of flexion. The mean maximum stresses were 1.5 MPa in group A and 2.0 MPa in group B at 90 of flexion (p \ 0.001). Conclusions: The results of this study demonstrated that anatomical SB ACL reconstruction using a trans-tibial technique developed relatively higher stresses around the femoral tunnel during knee flexion than those in the group using an AM technique.

Fig. 1 Stress patterns around the femoral tunnel during knee flexion

Fig. 2 Simularion of femoral tunnel drilling

P22-1087 The ‘vancomyin wrap: a simple, effective method to reduce infection rates in ACL reconstruction A. Getgood1, T. Wood1, P. Thompson2, T. Spalding1 1 University Hospitals Coventry and Warwickshire, Coventry, United Kingdom, 2University Hospital, Coventry, United Kingdom Objectives: Infection after ACL reconstruction is an uncommon but serious complication. The reported incidence is approximately 0.5%. Wrapping the harvested hamstring graft in gauze soaked in a dilute solution of vancomycin has been proposed as an option to prevent infection. Following a number of deep infections after ACL reconstruction in our University Hospital department, we introduced this technique. The purpose of this study was to re-audit the infection rate after ACL reconstruction, following the introduction of the ‘vancomycin wrap’. Methods: A retrospective audit was performed of all deep infections within the University Hospital department following ACL reconstruction between January 2007 and July 2010 when it was recognized that our infection rate was 4%. Detailed demographic and operative information was collected in order to identify a cause and to assess the effect of the infection on outcome. We then started prospective audit for infection and in October 2010 introduced the use of the ‘vancomycin wrap’—500 mg vancomycin with 100 ml 0.9% saline soaked on gauze and wrapped around the graft following harvest and preparation prior to insertion.

123

S222

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Results: During the 4 year period, 19 patients were identified as having a deep infection, 15 of which were male. The average age was 24.3 years (range 15–38 years). Graft failure occurred in 3 patients, 2 of which have subsequently undergone revision ACL reconstruction without complication. In the remaining 16 patients, all the grafts survived, with good to excellent results observed out to 1 year, with mean side to side difference on KT 1000 arthrometry of 1.8 mm. Following multidisciplinary review, no obvious consistent source or cause of infection was identified. Since the implementation of the vancomycin wrap no further episodes of deep infection following ACL reconstruction have been reported in over 100 ACL reconstructions and other ligament procedures. Conclusions: Deep infection following ACL reconstruction can be a serious complication. The rate of deep infection found within our department was found to be abnormally high compared to that quoted in the literature with no identifiable pattern or remedial cause found. Our policy of early arthroscopic washing out and management however has resulted in a successful outcome in 16 of 19 patients. The implementation of the ‘vancomycin wrap’ has significantly reduced the infection rate in our department. We recommend the widespread use of this step so as to reduce the complication rate and ultimately ensure better outcomes for patients.

Fig. 1 Linear regression for in vitro setup P22-1095 Validation of acceleration parameters in quantifying pivot-shift test: in vitro multi-tester VS. in vivo single-tester analysis N. Lopomo1, Y. Hoshino2, M. Ahlde´n2, K. Nagamune3, V. Musahl4, S. Zaffagnini1 1 Istituto Ortopedico Rizzoli, Laboratorio di Biomeccanica e Innovazione Tecnologica, Bologna, Italy, 2University of Pittsburgh, Department of Orthopaedic Surgery, Pittsburgh, United States, 3 University of Fukui, Department of Human and Artificial Intelligent Systems, Graduate school of engineering, Fukui, Japan, 4University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: Different methods have been developed to quantify the pivot shift (PS) maneuver, specifically tibial acceleration and anterior translation. This study aimed to validate the parameters measured by means of an acceleration sensor both in an in vitro multi-tester setup and in an in vivo single-tester setup. We validated acceleration method using an electromagnetic system and a commercial navigation system. Methods: In vitro Setup: A whole lower body specimen was used. The PS test was performed 3 times on each limb by 12 expert surgeons from world wide. Kinematic data were acquired using 2 different measurement devices: an electromagnetic (EM) system with sensors fixed to the femur and tibia was used as gold standard, and an accelerometer (ACC) system fixed on proximal tibia skin. In vivo Setup: 15 consecutive patients with ACL injury were recruited. A single surgeon performed the PS before ACL reconstruction, acquiring the limb kinematics by means of a commercial navigation system (NAV); a custom-made box containing a commercial accelerometer (ACC) was skin-fixed to the tibia by means of a sterile drape. Statistics: Pearson correlation coefficient (rs) and the linear regression (R2) between the acceleration measured with the EM and NAV systems and the range of acceleration measured with the ACC system were evaluated. Statistical significance was set at the 95%.

123

Fig. 2 Linear regression for in vivo setup

Results: In vitro Setup: ACC system demonstrated positive correlation with the EM system (rs = 0.73 p \ 0.0001). Moreover linear regression (Fig. 1) reported a R2 = 0.53 (p \ 0.0001).

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 In vivo Setup: ACC system demonstrated positive correlation with the EM system (rs = 0.67 p = 0.0064). Moreover linear regression (Fig. 2) reported a R2 = 0.45 (p = 0.0064). Conclusions: Both In vitro multi-tester and in vivo single-tester setup showed good correlation and fair linear regression between the acceleration measured with the ACC system and the results obtained with EM and NAV system. There are several advantages of the accelerometer: it is simple, non-invasive, and can be used for office exams. Future studies will investigate its use for clinical assessment following ACL reconstruction.

P22-1134 3D CT analysis of femoral and tibial tunnel locations after modified transtibial single bundle ACL reconstruction S.D. Cho1, Y.S. Youm1, J. Eo1, K.J. Lee1 1 Ulsan University Hospital, Department of Orthopedic Surgery, Ulsan, Republic of Korea Objectives: To analyze the location of femoral and tibial tunnels by three-dimensional (3D) CT reconstruction images after modified transtibial single bundle (SB) anterior cruciate ligament (ACL) reconstruction, creating a femoral tunnel with varus and internal rotation of the tibia. Methods: Data from 50 patients (50 knees) analyzed by 3D CT after modified transtibial SB ACL reconstructions were evaluated. 3D CT images were analyzed according to the quadrant method by Bernard at the femur and the technique of Forsythe at the tibia. Results: The mean distance of the femoral tunnel center locations parallel to the Blumensaat’s line was 29.6% ± 1.9% along line t measured from the posterior condylar surface. The mean distances perpendicular to the Blumensaat’s line was 37.9% ± 2.5% along line h measured from the Blumensaat’s line. At the tibia, the mean anterior-to-posterior distance for the tunnel center location was 37.8% ± 1.2% and the mean medial-to-lateral distance was 50.4% ± 0.9%. Conclusions: The femoral and tibial tunnels after modified transtibial SB ACL reconstruction were located between the anatomical anteromedial and posterolateral footprints.

P22-1175 The effect of graft selection on in vivo three-dimensional patellar kinematics during robotic tibial rotation using dynamic computed tomography T. Branch1, S.K. Stinton2, T.J. Cunningham2, C. Jacobs2 1 University Orthopaedic Clinic, Decatur, GA, United States, 2 ArthroMetrix, Atlanta, GA, United States Objectives: Increased rates of patellofemoral arthritis, crepitus, and pain following ACL reconstruction performed with a bone-patellar tendon-bone (BTB) autograft when compared to those performed with a semitendinosus/gracilis (ST/G) autograft have been reported. Our purpose was to determine if patients with BTB grafts demonstrated altered patellar kinematics when compared to patients with ST/G grafts using dynamic CT during robotic tibial rotation. Methods: Volunteers who had previously undergone ACL reconstruction with either a single-bundle BTB autograft or a doublebundle ST/G autograft participated in this IRB-approved study. Subjects were positioned supine with knees flexed 20 and passing through the CT scanner gantry, thus allowing their feet to be placed within a robotic tibial rotation system (RKT). The RKT bilaterally cycled the knees into external and internal rotation (ER, IR) at a

S223 velocity of 1.25/s to a torque threshold of 6 N-m. Three preconditioning cycles were followed by 1 test cycle of tibial ER and IR performed by the RKT while repeated axial scans were collected at 1.25 Hz. Kinematic analyses were performed using a free-body assessment of the patella in relation to the femur. Peak ER, IR, yaw rotation (i.e. valgus/varus motion of the patella), and 3 dimensional translation of the patella were calculated for both knees. The length of the patellar tendon and the Insall-Salvati ratio were also measured from the CT images. The BTB and ST/G groups were compared using independent t tests, with p B 0.05 being considered significant. Results: The BTB group demonstrated significantly greater patellar rotation during tibial ER than the ST/G group (9.22 ± 1.23 vs. 5.08 ± 2.31, p = 0.04, d = 1.47). While not significant, the amount of patellar rotation was also greater for the BTB group (2.68 ± 1.95) than the ST/G group (0.01 ± 1.24, p = 0.27, d = 1.46). The patellar tendon length (p = 0.08) and Insall-Salvati ratio (p = 0.02) were both lower in the BTB (tendon length = 38.4 ± 0.9 mm, I–S ratio = 0.69 ± 0.04) group than the ST/G group (tendon length = 52.6 ± 13.6 mm, I–S ratio = 1.16 ± 0.29). Conclusions: Patellofemoral kinematics differed between BTB and ST/G ACL reconstruction patients. Multiple studies have reported that the patellar tendon shortens following the harvest of the graft for ACL reconstruction, with greater shortening associated with greater osteoarthritic changes to the patellofemoral joint. The altered patellar kinematics demonstrated in this study were most likely due to subsequent shortening of the patellar tendon following the BTB graft harvest. The scarring and shortening caused the patella to be more rigidly fixed to the tibial tubercle so that as the tibia rotated, greater patella rotation occurred as opposed to a normal patellar tendon which would be more flexible possibly allowing the patella to have increased yaw motion instead. Studies are underway to determine if the two distinct kinematic patterns demonstrated in this study are related to the risk of patellofemoral arthritis.

P22-1213 Tunnel positioning and graft obliquity in anatomic single bundle and transtibial techniques: an MRI comparison study H. Haragus1, R. Prejbeanu2, B. Deleanu2, A. Popescu2, D. Vermesan2 1 Emergency Clinical County Hospital, I-st Clinic of Orthopedics and Trauma, Timisoara, Romania, 2University of Medicine and Pharmacy Timisoara, Timisoara, Romania Objectives: The introduction of the anatomic single bundle ACL reconstruction technique changed the way surgeons drill the femoral tunnel, by using an anteromedial portal drilling technique (AM) instead of the traditional transtibial technique (TT). This subsequently changed the positioning of the tunnels and the resulting obliquity of the graft, positioning it in a more anatomical fashion. Methods: We used MRI scanning of 20 primary ACL reconstructed knees with hamstrings autogenous graft (10 patients with unilateral ACL reconstruction using AM technique with a, minimum 1 year postop and 10 patients with TT technique, minimum 2 years postop) to determine the difference in graft obliquity and tunnel placement. Results: The position of the tunnels and graft obliquity were found to differ on images in both coronal and sagital planes. Obliquity was determined by measuring the intersection angle of the graft line with the tibial plateau plane on MRI T1 weighted images. The center of the femoral tunnel was determined and the tunnel axis was compared with the long axis of the femur and tibia respectively. Graft obliquity

123

S224

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

avereged 56 (ranging from 51 to 58) using the AM technique and 64 (ranging from 63 to 69) with TT technique. Results were processed with the non parametric Kruskal–Wallis test with p \ .05 defined as significant. We detetmined a statisticaly significant difference in graft obliquity and tunnel angles. Conclusions: The anatomic single bundle recosntruction technique has been found to reproduce more acurately femoral footprint and the orientation of the graft compared to the clasical TT technique where the appropriate tibial tunnel placement resulted in a more vertical than native ACL, but it fails in many cases to reproduce the exact anatomy of the native ACL as normal graft obliquity is not allways restored with either technique.

P22-1325 The influence of applied internal and external rotation on acceleration of the pivot shift momen S. Kopf1, R. Kauert2, J. Halfpaap2, N. Halfpaap2, T.M. Jung3, R. Becker4 1 Charite´, University Medicine Berlin, Center for Musculoskeletal Surgery, Berlin, Germany, 2Otto-von-Guericke University of Magdeburg, Institute of Micro and Sensor Systems, Magdeburg, Germany, 3Virchow-Klinikum/Charite´, Unfallchirurgie & Orthopa¨die, Sektion Sporttraumatologie & Arthroskopie, Berlin, Germany, 4Sta¨dtisches Klinikum Brandenburg, Zentrum fu¨r Orthopa¨die und Unfallchirurgie, Brandenburg, Germany Objectives: The Pivot shift test is frequently used to assess knee instability after ACL injury and reconstruction. In case of a highgrade Pivot shift test an earlier onset of osteoarthritis and a decreased chance of return to the same level of pre-injury sport has been reported. However, the Pivot shift test has been known be subjective and though, there is still a lack in the ability to accurately quantify the Pivot shift test. Improving the Pivot shift test and increasing its accuracy would help to decrease the subjective influence on the grading. Thus, the purpose of our study was to compare the Pivot shift performed in internal and in external rotation between knees with an ACL rupture and contralateral, healthy knees. We hypothesized that 1. the Pivot shift phenomenon increases in external rotation compared to internal rotation during the Pivot shift test in the ACL torn knee, and 2. the Pivot shift phenomenon does not increase in external rotation compared to internal rotation during the Pivot shift test in the contralateral, healthy knee. Methods: Eight patients who presented with an isolated, complete ACL tear were included so far in this study. The Pivot shift was performed on both, non-injured and injured, knees under anesthesia utilizing a tibial and a femoral sensor. The Pivot shift test was repeated ten times on each knee in internal and in external rotation. The linear acceleration and the angular velocity were recorded in three dimensions with 198 Hz. Data were recorded and pre-analyzed at a connected handheld computer. Digital Signal Processing (DSP) began during data acquisition. Frequency filters were used to reduce noise and drift of the signal. To analyze the Pivot shift motion, the Euclidean norm of the acceleration was calculated to separate linear acceleration from turnings without linear acceleration. So, the Pivot shift phenomenon was identified and the amount calculated. Results: An increase of the Euclidean norm of acceleration was observed in internal rotation compared to external rotation in the

123

Fig. 1

knees with torn ACL (p = 0.013) and in the contralateral, healthy knees (p = 0.046) (Fig. 1). Conclusions: External rotation during the Pivot shift tests increases the Pivot shift phenomenon, which approves our first hypothesis. Thus, external rotation might help to easier recognize a positive Pivot shift by increasing the acceleration of the Pivot shift moment. This might be due to the tension of the iliotibial band, which is more relaxed in external knee rotation.2 However, the Pivot shift phenomenon is also increased in external rotation in the healthy, contralateral knee, which disproves our second hypothesis. Therefore, it does not seem to help comparing both, the torn ACL and the contralateral, knee in an external rotation. The present study shows that the Pivot shift phenomenon is increased in external tibial rotation compared to internal rotation.

P22-1341 Transverse femoral fixation of BPTB in ACL reconstruction. Biomechanical study of two different bio-absorbable pins. Controlled animal study G. Camillieri1, C. Vittorio2, E. Adriani3 1 II Faculty, Roma, Italy, 2University of L’Aquila, Rome, Italy, 3Mater Day Hospital, Rome, Italy Objectives: Transverse pin femoral fixation of bone-patella tendonbone (BPTB) in ACL reconstruction has been widely applied during the last decades. Aim of our study is to confront two different system of transverse femoral fixation for BPTB graft: Transfix BTB (Arthrex) and BioTransfix T3 (Arthrex). Methods: 30 fresh-frozen porcine knees (mean age 2.2 years) were assigned to the two groups randomisedly. the patellar bone block and tendon were harvested using the same size in all specimens (10 mm 9 25 mm, 10 mm). The main differences between these two system are the diameter (3.0 mm Transfix BTB and 3.5 mm BioTransfix T3), and section (Transfix BTB is cannulated). Surgical technique adopts the same transverse vectorial guide but different guide sleeves. Zwick-Roell z010 tension/compression device with bone clamps, was used for the study: • Cyclic test (1,000 cycles, 0.5 Hz, 50–250 N/cycle, 100 cycles of preload) • Final pull-out test (1 mm/s) • Failure analysis

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 •

CT scan and densitometry 0

Results: Any implant didn t fail during cyclic test. The elongation average was 1.85 ± 0.63 for Transfix BTB and 1.69 ± 0.87 for BioTransfix T3. Pull-out test showed very similar values in terms of Ultimate Strength Failure (USF), Stiffness at USF and Stiffness: Transfix BTB 818.14 ± 194 N, 70.09 ± 16.87 N/mm, 66.44 ± 15.49 N/mm BioTransfix T3 808.93 ± 196.78 N, 74.04 ± 23.75 N/mm, 70.41 ± 19.36 N/mm The failure mode was bone plug fracture (12 for Transfix BTB and 13 for BioTransfix T3) and tendon failure (3 for Transfix BTB and 2 for BioTransfix T3). The post-test CT scan showed any failure of the fixation devices and the correct position inside the femoral half-tunnel. The mean bone density of porcine femora was comparable to young human femora (1.12 ± 0.31 BMD) Conclusions: The systems showed a similar behaviour in terms of USF, Stiffness, Cyclic load, method of failure and other biomechanical parameters. The reproducibility of surgical technique, the mechanical strength and endurance of the systems suggest two valid options for ACL reconstruction with BPTB even if in vivo studies are necessary to confirm the animal ex vivo biomechanical data.

P22-1342 Use of the ruler technique to determine mid-footprint femoral tunnel position in anatomic anterior cruciate ligament reconstruction in a District General Hospital, analysed with 3-dimensional computed tomography C. McGarvey1, J. Bird1 1 University Hospital Lewisham, Orthopaedic Department, London, United Kingdom Objectives: The recognition that transtibial tunnel drilling results in a non-anatomic vertically oriented femoral tunnel has led to increasing interest in surgical techniques that position the femoral tunnel within the footprint of the native ACL. Various techniques have been described that help the surgeon to identify the mid-footprint for an anatomic single bundle femoral tunnel. We aimed to use the ruler technique and to quantify our results by using 3-dimensional computed tomography scan analysis. Methods: The ruler technique is based on identifying the proximal border of the articular cartilage and using a specific ruler parallel to the femoral axis to locate the origin of the ACL. The accuracy of this technique has previously been validated by measuring tunnel position on postoperative 3D CT scans. This study took the same technique, but utilised it in an independent District General Hospital setting. We also analysed our results using 3D CT analysis. Results: Thirty ACL reconstructions were performed using the ruler technique, with placement of the femoral tunnel at the center of the femoral insertion. The mean position for the center of the femoral tunnel measured by the ruler technique was\1 mm from the theoretic optimal center position. Conclusions: The ruler technique produced femoral tunnels comparable to published radiographic criteria used for tunnel placement. We recommend use of the ruler technique as it is reproducible and accurate when used in District General Hospital setting.

P22-1404 Vascularization of the remnant of ACL and clinical results of ACL reconstructions with remnant preservation J. Cholewin´ski1, K. Ficek2, A. Wiatr2, M. Paja˛k1, A. Kubiak1 1 Bonifratres Hospital, Orthopedics, Katowice, Poland, 2Galen Orthopaedics, Bierun, Poland

S225 Objectives: The purpose of this study was to assess influence vascularization of the remnant of ACL tear on the results of acl reconstructions performed with a single bundle technique, using hamstring tendons. Methods: 71 patients with ACL tear who had ACL reconstructions between 2008 an 2009 were included into the study. Time from injury to the operation was from 2 weeks to 11 months. Transportal femoral tunnel drilling and stabilization with Endobutton CL on femur and Intrafix on tibia were performed. During reconstructions some part of remnant tissue was harvested for microscopical examinations to assess vascularization intensity with 4 grade score. Clinical evaluation 2 years after surgery included Lachman and pivot shift tests, range of motion, arthrometry with using Genu Rob device and IKDC and Lysholm scores. Results: After 2 years follow-up there were statistically significant differences between pre- and post-operative values for Lachman and pivot shift test, IKDC and Lachman scores, and measurements with Genu Rob. 2 patients were re-operated due of ROM limitations with cyclops-like scar formations. 4 patients were qualified to revision reconstruction according to graft failure. Statistically significant correlation between better vascularization and better preservation of the remnants of acl as well as better arthrometry outcomes were found. Shortening the time from injury to the operations improved also arthrometry test. Prolonged time from injury results in limitations of blood supply to the remnant tissue. Conclusions: Reconstructions of ACL with remnant preservation showed good clinical results. Preserving well vascularized ACL remnant and shortening the time from injury to operation improved clinical outcomes.

P22-1408 The femoral insertion area of anterior cruciate ligament Y. Fujimaki1, Y. Fujimaki Yoshimasa1, M. Miyawaki1, S. Tashman1, J. Irrgang1, F.H. Fu1 1 University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States Objectives: Anatomic ACL reconstruction is defined as the functional restoration of the ACL to its native dimensions, collagen orientation, and insertion sites. Knowledge of the ACL footprint morphology and area is necessary to best match the bone tunnel with the native ACL footprint. As one of index, the occupation ratio (% of native insertion site area covered by tunnel aperture) should provide a meaningful tool for selecting appropriate tunnel and graft size. The goals of this study are to determine the relationship between insertion site dimensions (length/width) and insertion site area to establish suitable formula for estimating the femoral insertion area of ACL using arthroscopic measurements. Methods: Fresh-frozen, cadaveric knees (n = 10) with an age range of 57–65 years were used for this study. After disarticuration, the overlying surface membrane was carefully removed to clarify the ACL footprint. The outline of the ACL insertion site was marked, and the medial condyle was cut away to obtain a true lateral view of the medial wall of the lateral femoral condyle. Digital photographs were taken with a reference scale and the area (actual area), major axis (length) and short axis (width, both AM and PL) of the footprint were measured using Image J software (Fig. 1). Since the femoral origin of the ACL has an approximately oval shape, its area was modeled as an ellipse with short axis chosen as the average of AM and PL widths. This estimated area was then compared to actual area using linear regression. Statistical analyses were performed, with significance level set at p \ .05. Results: Mean femoral insertion distances were 15.9 ± 1.3 mm in length, 7.1 ± 1.2 mm in AM width and 7.1 ± 0.8 mm in PL width. In spite of the variety of insertion site size and shape, regression analysis shows highly significant correlation between the real ACL insertion site area and the estimated area using an ellipse formula

123

S226

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: The knee kinematics of 10 patients were measured using a surgical navigation system and described in terms of tibial axial rotation and antero-posterior translation. Results: In the ACL deficient knee, the average maximum tibial rotation during the pivot shift test was 29.0 and the mean maximum translation 17.0 mm. Reconstruction of the AM bundle (which behaves in a biomechanically similar way to a single-bundle reconstruction) reduced the rotational component to 16.4 (p \ 0.0001) and translation to 6 mm (p = 0.0002). Conclusions: Addition of the PL bundle further reduced rotation to 12.6 (p = 0.0007) but had no significant effect on translation. Addition of the PL bundle also significantly reduced translation and coupled tibial internal rotation during the Lachman and Anterior draw tests.

Fig. 1 Medial wall of lateral femoral condyle

(R2 = 0.9758, p \ 0.001). The regression equation was as follows: AREAACL = 1.1407 9 AREAELLIPSE - 5.834. Conclusions: Combined with the ellipse formula and the above regression equation, area of ACL foot print can estimate with arthroscopically measured sizes. Since the area of the tunnel aperture can be easily determined from the width and the length of the oval bone tunnel outlet, the occupation ratio of the reconstructed tunnel and graft relative to the original insertion site can be determined. This study established a method for estimating native ACL insertion site size from arthroscopically available measurements, which could be a valuable tool for intra-operative decision-making for graft choice and tunnel size concerning with occupation ratio.

P22-1418 The control of the knee kinematics achieved by single verses double-bundle Anterior Cruciate Ligament reconstruction: intra-operative measurement using surgical navigation J. Robinson1, J. Murray2, M. Hassaballa1, A. Porteous3, P. Colombet4 1 The Bristol Knee Group, Orthopaedic Surgery, Avon Orthopaedic centre, Bristol, United Kingdom, 2The Bristol Knee Group, Bristol, United Kingdom, 3Avon Orthropaedic Centre, Southmead Hospital, Bristol, United Kingdom, 4Centre de Chirurgie Orthope´dique et Sportive, Me´rignac, France Objectives: Cadaveric experiments using knee testing machines have suggested that anatomical ACL reconstruction, replacing both anteromedial (AM) and postero-lateral (PL) bundles, restores knee rotation kinematics more effectively than does a single-bundle. The aim of this study was to measure intra-operatively the control of the translation and coupled rotations that occur with standard clinical laxity tests (anterior drawer, Lachman and pivot shift).

123

P22-1419 Tibial insertion morphology and area of anterior cruciate ligament S. Maeda1, M. Miyawaki1, Y. Fujimaki1, S. Tashman2, J. Irrgang1, F.H. Fu1 1 University of Pittsburgh, School of Medicine, Department of Orthopaedic Surgery, Pittsburgh, United States, 2Department of Orthopaedic Surgery, Orthopaedic Biodynamics Laboratory, RiverTech Centre, Pittsburgh, United States Objectives: The goal of anterior cruciate ligament (ACL) reconstruction is to restore the native ACL anatomy and normal knee kinematics. The extent to which the bone tunnel fills the native ACL footprint is one metric for assessing anatomical graft placement. Knowledge of the ACL footprint morphology and area is necessary to best match the bone tunnel with the native ACL footprint. The aims of this study were to evaluate the relationship between the ACL tibial insertion area and measurements of length and width, and to evaluate regression equations for estimating the ACL tibial insertion site area from arthroscopically feasible measurements. Methods: Ten cadaveric knees were used for an evaluation of the ACL tibial insertion site area. The average age of the specimens was 60 years. All soft tissue structures were removed to expose the bone, leaving only the ACL. The overlying synovium around the ACL was carefully removed to expose the ACL surface fibers, and the ACL footprint was marked. After all ACL fibers were removed, the specimens were positioned with the tibial plateau perpendicular to the floor and digital photographs were taken with a reference scale. The photographs were uploaded to a PC and the tibial insertion length (LACL), width (WACL), and area (SACL) were measured by image J software. The length was measured along the line from medial edge of tibial tuberosity to posterior portion of ACL. The width was measured perpendicularly to length, and maximum width of insertion site was measured (Fig. 1). Regression analysis was performed for comparison between the tibial insertion area and value of LACL 9 WACL. All results were expressed as mean ± standard deviation. SPSS software was used for all statistical analyses, with significance level set at p \ .05. Results: In 6 knees, the tibial insertion approximated an oval. While in the remaining 4 knees, it was closer to triangular in shape. Mean tibial insertion length and width were 15.7 ± 2.4 mm and 10.2 ± 1.1 mm, respectively. The tibial insertion area was 119.4 ± 26.3 mm2. Linear regression revealed a strong correlation between tibial insertion area and rectangle area (r = 0.95, p \ .0001). The regression equation was as follows: SACL = 1.5338 +0.7296 9 LACL 9 WACLv. Conclusions: The tibial insertion area significantly correlated with the value of length 9 width, and measurements of tibial insertion and this regression equation might be useful in estimating the tibial insertion area. Determining the ACL tibial insertion area will help surgeons optimize surgical parameters (e.g. drill diameter) to create the most

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S227 without cartilage lesions (p = 0.97). At 5 year imaging assessment, the presence of degenerative changes was statistically associated in a positive fashion with age at surgery (p = 0.002) and time period from trauma to surgery (p \ 0.001); in negative measure with postoperative Lysholm (p \ 0.001). 8 of 14 patients who had received additional meniscectomy, with a5.8 OR for development of degenerative changes (95% CI, 1.2–27.8, p = 0.03) compared with patients with no meniscal tears. The presence of cartilage lesions at surgery was not significantly predisposing to further degeneration (p = 0.45). Conclusions: This 2 stage arthroscopic ACL revision after traumatic rupture of the reconstructed ACL is a safe procedure, which allows to address large tunnel defects, with no donor site co-morbidities.

Fig. 1 Photograph of tibial plateau anatomical tibial bone tunnel and select the most appropriate method (i.e. single vs. double-bundle) to reconstruct the ACL.

P22-1451 Two stage procedure in anterior cruciate ligament revision surger. A 5 year prospective study F. Franceschi1, R. Papalia2, A. Del Buono2, G. Rizzello3, N. Maffulli4, V. Denaro2 1 University Campus Biomedico of Rome, Orthopaedic and Trauma Surgery, Rome, Italy, 2University Campus Biomedico of Rome, Orthopaedic Surgery, Rome, Italy, 3University Campus Biomedico of Rome, Rome, Italy, 4Queen Mary University of London, Barts and The London School of Medicine and Dentistry, Centre for Sports and Exercise Medicine, London, United Kingdom Objectives: In USA, 3,000–10,000 patients undergo anterior cruciate (ACL) revision surgery every year. We reported the minimum 5 year clinical, functional and imaging outcomes of 30 patients who had undergone a previously described OATS tube harvester grafting before (Arthrex, Naples, FL) ACL revision procedure. We postulated that this two stage procedure could restore stability in the operated patients and return them to sport activity. We also wished to define whether predisposing factors correlate with the development of longterm degenerative changes. Methods: This is a prospective analysis on 30 patients (19 males and 11 women; average age 29.1 years) who underwent a 2 stage arthroscopic ACL revision procedure after traumatic rupture of the reconstructed ACL. The clinical diagnosis of mechanical instability of the knee (Lachman test, Pivot shift, arthrometer KT-1000 side to side difference) was confirmed at MRI and arthroscopic assessment. Information on age, gender, time from ACL reconstruction to failure, etiology of failure, time from diagnosis of ACL failure to surgery, and type of trauma were collected. Preoperatively and at a minimum follow up of 5 years from the index operation, the objective International Knee Documentation Committee (IKDC) evaluation form, Lysholm score and were administered to all the patients, Lachman and pivot shift were tested. At the last follow-up, radiographic Fairbank changes were compared with pre-operative ones. Results: Postoperatively, all reported clinical and functional outcomes were significantly improved compared to preoperative status. The subgroup of patients who had received additional meniscectomy had significantly lower Lysholm scores than the others (p = 0.002), whereas comparable results were observed in patients with and

P22-1463 The Tegner Activity Scale index as a measure of early regained activity following knee arthroscopy K. Briggs1, W.G. Rodkey2, L. Matheny1, J.R. Steadman3 1 Steadman Philippon Research Institute, Clinical Research, Vail, United States, 2Steadman Philippon Research Institute, Vail, United States, 3The Steadman Clinic and, Steadman Philippon Research Institute, Vail, United States Objectives: In the athletic population is it important to determine if patients regain the activity level they lost due to injury. The Tegner Activity Scale is a single item scale which is based on type of activities and level of activities performed. The purpose of this study was to determine if the Tegner Activity Scale Index (TASI) was a valid measure of recovered activity after knee arthroscopy. Methods: This study was institutional review board approved. The Tegner Activity Scale Index (TASI) is calculated as follows: ((Follow-up Tegner Activity scale—pre-operative Tegner Activity scale)/ (Pre-Injury Tegner Activity scale—preoperative Tegner Activity scale))*100. This calculation provides the percentage of lost activity due to original injury that has been regained as a result of the treatment. One-hundred and fifty-eight patients completed the questions needed for the Tegner Activity Index Scale between November 2009 and January 2011. There were 77 females and 81 males. The average age was 43 years (range: 18–83 years). Average follow-up was 10 months (range: 6–18 months). Patients with follow-up \10 months were not expected to regain full activity. Results: The median TASI was 20% [95%CI: 19.9–34.7]. Nine percent of patients had improved over 100% and were considered outliers. Nine percent of patients who did not improve were over 100% and were considered outliers. These scores were still considered in final analyses. The median TASI for patients with follow-up \10 months was 14% [95%CI: -6.8–26.1%] and the medial TASI for patients with follow-up 10 months or greater was 30%[95%CI: 16.9–58.6%]. The TASI was correlated with follow-up Lysholm (rho = 0.474; p = 0.001) and patient satisfaction (rho = 0.252; p = 0.001). Conclusions: Tegner Activity Scale Index showed construct validity. This score is able to distinguish between patients who have been released to full activity and those who have not. TASI also correlates to function and satisfaction. Eighteen percent of scores were considered outliers. This may be a weakness of the score. This score could be an effective means to determine if the treatment goal of return to activity is really achieved.

P22-1515 Wird nachgereicht S. Plaweski1, N. Mercier2, M. Grimaldi3, D. Saragaglia3 1 Grenoble South Teaching Hospital, Department of Orthopaedic Surgery and Sport Traumatology, E´chirolles, France, 2Centre Hospitalier Universitaire, Hoˆpital Sud, Orthopedie, Echirolles,

123

S228 France, 3Grenoble South Teaching Hospital, Orthopaedic and Sport Traumatology, E´chirolles, France Objectives: Based on biomechanical anatomical studies, doublebundle reconstruction of the anterior cruciate ligament (ACL) was introduced to achieve better stability in the knee, particularly in respect of rotatory loads. Purpose of the study: It was to evaluate and compare per-operative anteroposterior (AP) and rotational knee stability between single and double- bundle anterior cruciate ligament (ACL) reconstruction using navigation. Hypothesis: An in vivo, computer assisted, double bundle ACL reconstruction (DB) is superior than a single bundle (SB) ACL reconstruction at reducing rotatory, and AP laxities of the tibia at 20 of knee flexion and also during the pivot shift test. Study design: In vivo prospective comparative study. Methods: The data of 63 patients who had ACL reconstruction were prospectively collected. Thirty-two patients had single-bundle reconstruction (SB group) and 31 received double-bundle reconstruction (DB group). The per-operative navigation system (Praxim ACL surgetics System) helped to search for a minimal anisometry profile of the grafts which was favorable (graft loosened with flexion) in the anatomic area of ACL insertion and preventing any conflict between the graft and the femoral notch. The system also evaluated Antero Posterior (AP) rotational stabilities and pivot shift. The value of the pivot shift was calculated from the values of the maximum rotation and AP translation obtained when performing the manoeuver before and after ACL reconstruction, comparing SB and DB reconstruction. Results: The post-operative AP displacement of the lateral compartment during the Lachman test was statistically reduced in DB group in comparison to SB group (5.1 ± 4.4 mm vs. 7.1 ± 3.2 mm, p = 0.046), whereas the AP displacements of the medial compartment were also reduced (3.4 ± 3.7 mm vs. 4.5 ± 2.6 mm, p = 0.152) but with no statistical significance. Internal and external rotations at 20 of knee flexion were lower in the DB group than in SB group with statistical significance (respectively 13.2 ± 4.9 vs. 17.5 ± 4.0, p \ 0.001 and 9.1 ± 3.6 vs. 11.5 ± 3.5, p = 0.009). During the pivot shift test the postoperative AP maximal translation was statistically different in both groups: 4.5 ± 2.1 mm in DB group and 6.3 ± 2.7 mm in SB group (p = 0.012)) whereas the maximal rotation was not statistically different: 3.8 ± 2.5 in DB group and 3.4 ± 1.2 in SB group (p = 0.468). Therefore, Colombet’s index was similar in DB group and SB group (respectively 0.21 ± 0.16 and 0.17 ± 0.06, p = 0.279). Conclusions: Our study shows a significant intraoperative advantage in anterior and rotational stability for four-tunnel DB ACL reconstruction compared to SB ACL reconstruction Level of evidence: II Keywords: Anterior cruciate ligament, Double-bundle reconstruction, computer-assisted-navigation, Biomechanics, hamstring tendons

Knee-PCL

P23-133 Re-consideration of measurement methods of stress radiographs for the improvement of reproducibility during the evaluation of knee anterior and posterior instability: a comparison of 5 different methods Y.S. Lee1, K. Nha2 1 Gachon University School of Medicine, Department of Orthopaedic Surgery, Incheon, Republic of Korea, 2Inje University, Ilsanpaik Hospital, Koyangsi, Republic of Korea

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: The purposes of this study were 1) to evaluate the efficacy of the various methods of stress radiographs by means of intra- and inter-observer reliability and test–retest reproducibility, and 2) to prove the effectiveness of our novel stress radiograph, namely BAT method (Blumensaats0 line-Anterior Tibia) method in practical usage. Methods: Stress radiographs of 40 anterior cruciate ligament (ACL) and 40 posterior cruciate ligament (PCL) patients were taken using the Telos device. The values of 4 conventional methods (MM: medial–medial, LL: lateral–lateral, Mid–Mid: middle–middle, and PC: peripheral-central), and 1 newly devised method (BAT: Blumensaats0 line-Anterior Tibia) were compared. The intra-class correlation coefficients (ICCs) were examined to assess the intra- and inter-observer reliability of the measurements. For an evaluation of the reproducibility of each method, stress radiographs were taken twice (before- and after- the examination at the outpatient clinic) on the same day and the values from the 1st and 2nd stress radiograph were compared. Results: In the anterior drawer test, as to measurement reliability, the ranges of ICCs were [0.713, 0.889] for MM, [0.624, 0.812] for LL, [0.834, 0.932] for Mid–Mid, [0.722, 0.893] for PC, and [0.891, 0.963] for BAT methods. As to test–retest reproducibility, the mean differences (SD) of displacement between the first and second radiograph was 1.0 (0.8) for MM, 2.4 (2.3) for LL, 1.7 (1.6) for Mid–Mid, 1.2 (0.6) for PC, and 0.5 (0.7) for BAT method. In the posterior drawer test, as to measurement reliability, the ranges of ICCs were [0.859, 0.958] for MM, [0.773, 0.915] for LL, [0.859, 0.951] for Mid–Mid, [0.852, 0.958] for PC, and [0.893, 0.961] for BAT. As to test–retest reproducibility, the mean differences (SD) of displacement between the first and second radiographs were 1.6 (1.3) for MM, 1.8 (1.7) for LL, 1.7 (1.5) for Mid–Mid, 1.4 (1.1) for PC, and 1.1 (1.2) for BAT. Conclusions: Different methods of measuring stress radiographs resulted in different levels of reliability and reproducibility. In the anterior drawer test, the BAT method showed best measurement reliability and test–retest reproducibility. In the posterior drawer test, the BAT method showed favorable measurement reliability and reproducibility, but the superiority could not be demonstrated.

P23-167 Anatomical study and morphometric analyses of the posterior cruciate ligament0 s femoral insertions J.C. Gali1, H.C. Oliveira1, A.B. Camargo1, C.R. Martins1, P.A. Silva1, E.B. Caetano1 1 Catholic University of Sa˜o Paulo, Sorocaba, Brazil Objectives: To provide anatomical and morphometric basis of the posterior cruciate ligament’s femoral insertions in order to assist the creation of anatomical femoral tunnels, in the ligament surgical reconstruction. Methods: Macroscopic details of the posterior cruciate ligament0 s anterolateral and posteromedial bundles’ femoral insertions were analyzed in 24 anatomical knee pieces. The pieces were photographed by a digital camera and the images obtained were studied by the software ImageJ, where the bundles’ insertion areas were measured in square millimeters, and the length of structures and the distances between significant points were measured in millimeters. Results: The posterior cruciate ligament0 s femoral insertion average total area was 87.29 ± 31.42 mm2. The mean insertion’s areas of the anterolateral and posteromedial bundles were, respectively, 47.13 ± 19.14 and 40.67 ± 16.19 mm2. In 95.8% of the examined knees was verified the presence of the medial intercondylar ridge and in 83.3% of the pieces was noted the medial bifurcated ridge. The

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 average length of the medial intercondylar ridge was 20.54 ± 2.26 mm and the medial bifurcated ridge’s average length was 7.62 ± 2.35 mm. Conclusions: The anterolateral bundle has a femoral insertion area larger than the posteromedial bundle; these bundles’ insertion areas were lower than those previously described in the literature. There were important individual variations related to the area of the bundles in our sample, suggesting that there should be an individual indication for anatomical reconstructions of the posterior cruciate ligament with single or double femoral tunnel.

P23-216 Fixation strength of tibial inlay technique versus posterior slope onlay technique in arthroscopic PCL reconstruction M. Osti1, K.P. Benedetto1 1 Academic Hospital Feldkirch, Feldkirch, Austria Objectives: PCL inlay reconstruction has been widely accepted over the last years. Experimental studies have proven less graft thinning and less elongation using quadriceps tendon compared to transtibial tunnel technique. Direct graft fixation of the graft bone block by screw fixation in open techniques has demonstrated sufficient stability for functional rehabilitation protocol. The objective of this study was to evaluate the fixation strength of two arthroscopic PCL reconstruction techniques. Methods: Ten paired fresh-frozen human cadaveric knee specimen (age 65–85 years) were used. The PCL tibial insertions were marked by leaving the remnants of the anatomical PCL fibrous attachments. Quadriceps tendon-bone grafts were harvested and prepared according to Noyes et al. with a bone block measuring 12 9 10 mm. One tibia of each pair was reconstructed using a standardized arthroscopic tibial inlay technique placing the replicated tibial PCL footprint onto the remnants of the anatomical PCL fibrous attachments (specimen T1, technique P1). The remaining tibias were reconstructed using a standardized arthroscopuic onlay technique placing the replicated footprint center at the most distal edge of the anatomical PCL fibrous attachments (previous physis line, Specimen T2, technique P2). Utilizing parameters similar to Campbell et al. the relative graft-tibia motion and failure properties of the tibia-graft fixation were measured using a servohydraulic material testing device (Instron Corp.). Each tendon was cycled from 50 to 250 N at 1 Hz for 1,000 cycles. Post cycling pull out failure was conducted at 20 mm/min. Failure load, cyclic displacement at 5, 500 and 1,000 cycles and stiffness were calculated. Results: All reconstructions completed the cycling before failure. The displacement at 5, 500 and 1,000 cycles measured consistently more in T1 specimen at 4. 11, 7.73 and 8.18 mm than in T2 specimen at 2.81, 6.01 and 6.46 mm respectively. Mean ultimate load to failure for T1 was 338 ± 130 N compared to 564 ± 212 N for T2 specimen. The calculated mean stiffness for T1 was 73.84 ± 16 N/mm and 55.87 ± 13 N/mm for T2. Failure mode was a teared graft which tended to be pulled out of the bone tunnel in all tests.

S229 Conclusions: PCL onlay technique (T2, P2) yielded higher pull-out strength and less cycling loading displacement compared to PCL inlay technique (T1, P1) and is recommended for arthroscopic PCL reconstruction using quadriceps tendon.

P23-232 How to safely perform the femoral tunnels of the popliteus tendon and fibular collateral ligament in multiligamentary knee reconstructions. Computed tomography evaluation in a cadaveric model P.E. Gelber1, J. Erquicia2, G.L. Sosa2, G. Ferrer2, F. Abat3, J.C. Monllau3 1 Hospital de la Santa Creu i Sant Pau. UAB, Department of Orthopaedic and Traumatology, Barcelona, Spain, 2ICATME- Institut Universitari Dexeus. UAB, Arthroscopic Unit, Barcelona, Spain, 3 Hospital de la Santa Creu i Sant Pau. UAB, Barcelona, Spain Objectives: To determine the best angle to drill the femoral tunnels of the popliteus tendon (PT) and fibular collateral ligament (FCL) in combined reconstructive procedures, to avoid either short tunnels or tunnel collisions with anterior cruciate ligament (ACL) and posterior cruciate ligament (PCL) femoral tunnels. Methods: Eight cadaveric knees were studied. ACL/PCL femoral tunnels were arthroscopically drilled (Fig. 1). PT and FCL tunnels were simulated with a 2.4 mm k-wire drilled through the same entry point at 0 and 30 axial angulations (in reference to a line perpendicular to the femoral anatomic axis), and at 0 and 30 coronal angulations (in reference to the transepicondylar axis), in both cases (Fig. 2). They were scanned by computed tomography and reconstructed 3-dimensionally. Volume-rendering software was used to document relationships of PT and FCL tunnels with the intercondylar notch and ACL/PCL tunnels (Fig. 3). A minimum tunnel length of 25 mm was required. Each measurement was performed by two independents observers and averaged. Intraclass Correlation Coefficient (ICC) was calculated. Results: When the PT tunnel was drilled at 0 axial angulation, there was an increased risk of tunnel collision with the ACL (p \ 0.001). Interference with the PCL tunnel can only be avoided if the k-wire guiding the TP tunnel is drilled at 30 axial and 30 coronal angulations (p \ 0.001). Minimum tunnel length of the PT could only be obtained with both axial and coronal angulations of 30 (p = 0.003). Sufficient tunnel lengths of the FCL were obtained at any of the evaluated angulations (p = 0.036). However, only tunnels drilled at 30 axial and 0 coronal angulations did not collapsed with the ACL tunnels (p \ 0.001). ICC was considered excellent (0.92). Conclusions: When performing posterolateral reconstructions in combination with concomitant anterior and posterior cruciate procedures, PT tunnels should be drilled at 30 of axial and 30 of coronal angulations. FCL tunnels should be drilled at 30 axial and 0 coronal angulations. Different angulations might lead to short tunnels or tunnel collision with the cruciates tunnels.

123

S230

Fig. 1

Fig. 2

Fig. 3

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P23-251 Clinical and radiological long-term outcome after posterior cruciate reconstruction and popliteus bypass surgery–what could we expect? T. Adler1, N.F. Friederich1, F. Amsler2, M.T. Hirschmann1 1 Kantonsspital Bruderholz, Klinik f. Orthop. Chirurgie & Traumatologie, Bruderholz, Switzerland, 2Amsler Consulting, Basel, Switzerland Objectives: The primary purpose of our study was to analyze the longterm outcome of patients treated for combined posterior cruciate ligament and posterolateral corner injuries by posterior cruciate ligament reconstruction and popliteus bypass according to Mueller or refixation of the popliteus tendon. Methods: 16 patients (male:female = 10:6, mean age at surgery 32 ± 14 years) treated by combined posterior cruciate ligament reconstruction and popliteus bypass according to Mueller (n = 7) or a refixation of the popliteus tendon (n = 9). The patients were evaluated with a mean follow-up of 24 ± 3 years using IKDC2000, the SF36, Lysholm and Tegner score and KOOS score. Anterior and posterior KT-1000 arthrometer measurements were performed. Bilateral anterior-posterior stress radiographs (in 30 and 90 flexion) were performed as well as varus stress radiographs. The degree of osteoarthritis was assessed with regards to the Kellgren Lawrence Score. Pearson correlations of predictive factors for worse outcome were performed. The level of statistical significance was p \ 0.05. Results: The total IKDC2000 was B (nearly normal) in 5 (31%), C (abnormal) in 7 (44%) and D (severely abnormal) in 4 (25%) of patients. The mean total SF-36 health survery was 79 ± 20, the Lysholm score 68 ± 22, the KOOS-symptoms 40 ± 13, KOOS-pain 26 ± 24, KOOS-activity 18 ± 18, KOOS-sport 51 ± 32, KOOS-LQ 44 ± 26. The Tegner score decreased from preinjury (7, 4–10) to (4, 2–10) at follow-up. The side to side differences of anterior KT-1,000 arthrometer measurements 134 N were 5 ± 5 mm. The side to side difference of bilateral posterior stress radiographs in 30 was 4 ± 5 mm and 6 ± 3 mm in 90 flexion. The side to side difference of bilateral varus stress radiographs was 3 ± 2 mm. The Kellgren Lawrence score showed minimal osteoarthritis in 7 (44%), moderate osteoarthritis in 7 (44%), and severe osteoarthritis in 1 patient (12%). One of the patients had undergone a total knee replacement. Conclusions: The challenging group of patients with combined posterior cruciate ligament injury and posterolateral corner injuries treated by posterior cruciate ligament reconstruction and popliteus bypass according to Mueller et al. or popliteus refixation showed only moderate clinical and radiological outcome, even when the age of the patients at surgery and the long-term follow-up is acknowledged. The type of posterolateral corner reconstruction using these two methods did not show any statistical difference in outcomes.

P23-287 Results of posterior cruciate ligament avulsion fixation in fresh or delayed- isolated or combined cases A.R. Kachooei1, S. Samani2, S. Razi2, R. Mahdavian Naghash Zargar2 1 Mashhad University of Medical Sciences, Emam reza Hospital, Orthopedic Department, Mashhad, Islamic, Republic of Iran, 2 Mashhad University of Medical Sciences, Mashhad, Islamic, Republic of Iran Objectives: Posterior cruciate ligament (PCL) injury is reported in 3–23% of knee injuries. Ligament rupture usually requires reconstructive surgery. Avulsion injuries in acute cases, however, are often treated by fixation of avulsed bone and ligament. Fixation of avulsions in chronic cases is somewhat controversial. Methods: In a descriptive study, from 40 male patients with a mean age of 27.87 years old who had undergone fixation of PCL avulsion

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 with maleolar screw and washer through a posteromedial approach, 20 cases with a mean of 40 ± 16.02 weeks follow-up were studied. Twelve cases had received early fixation and 8 had fixation 8 weeks or longer after the injury. The cases were evaluated with clinical evaluation of stability as well as assessment of patients’ satisfaction and return to sports or regular activities. Results: Improvement in ‘‘posterior drawer test’’ was observed in allcases. The ones who had early surgery, as well as the ones with isolated PCL injury had better results in comparison with late avulsion fixation or combined ligament injuries. Eight cases had no problem in knee function, 7 cases had some limitation in sporting, and 5 in some regular daily activities. Conclusions: Screw fixation of PCL avulsion through a posteriomedial approach, even in delayed cases, is safe, easy and effective. Other associated knee injuries are detrimental for good results.

P23-381 Radiographic landmarks for tunnel positioning in posterior cruciate ligament reconstructions A.M. Johannsen1, D.J. Westcott1, R.F. LaPrade2, C.A. Wijdicks1, L. Engebretsen3 1 Steadman Philippon Research Institute, Department of Biomechanics, Vail, United States, 2The Steadman Clinic, Department of Biomechanics, Vail, United States, 3Ullevaal University Hospital, Orthopaedic Center, Oslo Sportstrauma Research Centre, Oslo, Norway Objectives: Radiography is a standard method to assess PCL reconstruction tunnel placement both intraoperatively and postoperatively. This study aimed to establish quantitative and qualitative radiographic guidelines for identifying the femoral and tibial attachment sites of the ALB and PMB of the native PCL. Methods: Twenty non-paired fresh knees were dissected and their PCL bundles separated, excised from the bone, and attachment centers labeled using 2-mm metal spheres. Arthroscopically pertinent landmarks not readily observable on radiographs were labeled using 1-mm diameter T-pins or soldering wire. To assess bundle attachment margins, the bundle attachment area was labeled using a radioopaque barium sulfate (BaSO4) emulsion. Anteroposterior (AP) and lateral radiographs of the femur and tibia were obtained before, and lateral images repeated after the addition of BaSO4. Digital measurements were conducted using a picture archiving and communication system. Femoral and tibial measurements were made from the bundle centers and margins to radiographically pertinent landmarks and superimposed lines (Fig. 1). Clinically pertinent values are represented below and the bundle margins on the lateral femur are displayed in Fig. 1b. Results: On the AP femur view (Fig. 1A), the anterolateral (ALB) and posteromedial (PMB) bundle centers were located 14.1 (±1.3) mm and 15.8 (±2.1) mm superior to the distal joint line, respectively. The ALB center was located 34.1 (±3.0) mm lateral, and PMB center 29.4 (±3.1) mm lateral to the medial epicondyle line. Lateral femur images revealed that the ALB and PMB centers were located 4.9 (±1.3) mm and 10.8 (±1.4) mm posteroinferior and perpendicular to Blumensaat’s line. The ALB and PMB centers were located 13.6 (±1.5) mm and 14.9 (±2.2) mm superior to the distal condyle line. On the lateral tibia view (Fig. 1c), the ALB, PMB, and PCL centers were located 9.1 (±1.7) mm, 3.3 (±1.3) mm, and 6.4 (±1.6) mm superior to the champagne glass-dropoff respectively. The ALB

S231 was located 3.5 (±1.1) mm anterosuperior and perpendicular to the bundle ridge, while the PMB was located 3.6 (±0.9) mm posteroinferior and perpendicular to the bundle ridge. Lastly, the guide wire line was located 7.2 (±1.5) mm anterosuperior from the nearest concave edge of the posterior tibia. Conclusions: This study comprehensively identified radiographic guidelines of the native anatomy of the femoral and tibial bundle attachments of the PCL using clinically relevant radiographic landmarks. These findings can be directly applied to proper tunnel positioning intraoperatively during PCL reconstructions, and also to assess post-operative tunnel placement.

Fig. 1 Reference lines and measurement on the AP femur (a), lateral femur with BaSO4 (b) and the lateral tibia (c). Solid lines represent reference line and dashed lines represent measurements. PCL radiographs

123

S232 P23-396 Triple bundle PCL reconstruction with autogenous hamstring tendons: minimum 2 years follow-up T. Soejima1, T. Inoue1, K. Noguchi1, T. Kanazawa1, K. Tabuchi1, K. Nagata1 1 Kurume University, School of Medicine, Orthopaedic Surgery, Kurume, Japan Objectives: Currently, the double bundle PCL reconstruction using hamstrings tendons has been developed. However, it cannot get satisfied results yet, especially in the multiple-ligament injured knee. One of possible reasons may be that all over the femoral attachment cannot be covered by autogenous hamstrings tendons. Usually, an empty fiber-less area is appeared. To cover all over the femoral attachment as effectively as possible by even thinner grafts and to mimic only ALB and PMB but also the isometric fiber between ALB and PMB, we developed and tried the triple-bundle PCL reconstruction. Methods: Our operation technique was mentioned below. A semitendinous tendon and a gracilis tendon were harvested. A ST tendon was prepared as a V-shaped graft by folding 3 or 4 times for the ALB reconstruction. A gracilis tendon was prepared by normal manner for the PMB reconstruction. Then 3 femoral sockets (AL-high, AL-low, and PM sockets) and 2 tibial tunnels (AL and PM tunnels) were created. 2 free ends of a V-shaped graft were first introduced from the anterolateral portal to 2 femoral sockets (AL-high, AL-low) and were fixed by endbuttons, respectively. Then a loop end was pulled out to a tibial AL tunnel. Next, PMB graft was grafted into a femoral PM socket and a tibial PM tunnel as same manner. Finally, ALB graft was fixed at 90 flex. and PMB graft fixed at full flex. A total of 22 reconstructions in 21 patients were performed in our hospital. Of these, 16 reconstructions were evaluated with objective examinations including the gravity sag X-ray view, at 2 years after the operation. There were 13 males and 3 females. An average age at the operation was 27 years old. Only 3 cases were an isolated PCL injury. Results: The side to side difference of the tibial posterior displacement was 2.4 ± 1.7 mm at the time of the final evaluation, whereas it was 13.0 ± 3.8 mm before the operation. Only 1 patients were categorized in C according to the IKDC ligament evaluation form because of more than 5 mm posterior instability. Conclusions: The PCL is a very strong ligament. Its strength relate to a large cross-sectional area, and the fibres spread out to an extensive femoral attachment. When the PCL is reconstructed by the conventional double bundle procedure using thinner graft, all over the femoral attachment cannot be covered. Usually, an empty fiber-less area is appeared between ALB graft and PMB graft, and/or the most vertical arranged fiber is not reproduced. If the PCL is reconstructed by the triple-bundle procedure, it can mimic whole fiber arrangement although the volume of the grafts does not change. Because the volume of the graft increases during the remodeling process, to replicate whole fiber arrangement initially seems of great importance. There was still little number of the cases, but the triple-bundle PCL reconstruction revealed equivalent results to those of past studies in isolated cases, despite most of our cases had multiple-ligament injuries.

P23-431 Is the U sign a radiological evidence of PCL tibial avulsion fracture? S.R. Piedade1, M.M. Mischan2, D.M. Ferreira3 1 State University of Campinas, Orthopedics and Traumatology, Exercise and Sports Medicine Group, Campinas, Brazil, 2UNESP, Biostatistics, Botucatu, Brazil, 3State University of Campinas, Exercise and Sports Medicine Group, Campinas, Brazil

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: In some cases of posterior cruciate ligament (PCL) avulsion fracture at the tibial insertion, the avulsed fragment may appear on the radiographs of lateral view of the knee anterior to the original anatomic position. This particular condition may suggest fracture of the tibial plateau or even bone avulsion of the anterior cruciate ligament, which could induce the orthopedist to perform an anterior approach of the knee joint. A reassessment of our cases of PCL avulsion fracture at the tibial insertion revealed that in all the operated cases, a U-shaped radiological image was present at the anteroposterior view of the knee. This finding motivated an investigation of an anatomic relationship between the U sign and the PCL tibial insertion site. Methods: In a series of 21 patients who underwent surgical treatment for PCL avulsion fracture at the tibial insertion site, 19 preoperative radiographs were recovered and re-evaluated. A control group was established after evaluating 63 MRI scans of patients who did not present, as a final radiological diagnosis, boneligament lesions of the knee. Based on magnetic resonance scans and radiographs of anteroposterior view of the knee, the width and height of the PCL tibial insertion area and of the U sign at the avulsed tibial fracture site were measured, respectively. However, since the patients’ biotypes differed, the relationship of the heights and widths were calculated and analyzed with the largest transverse axis of the proximal tibia. In the control group (MRI scan group), the height of the PCL insertion area was measured from the superior tibial surface to the inferior distal insertion of the PCL fibers in the coronal plane, while the medial and lateral limits were defined by analyzing the PCL fibers in the axial plane. The data obtained from the two groups were tabulated and confronted. A significance level of 5% was used for statistical analysis. Results: A study of the two scans revealed that in both exams, the radiological U sign as well as the PCL tibial insertion site were topographically situated in the medial two quarters of the proximal tibial extremity. The statistical analysis did not indicate any difference between the mean height values of the tibial insertion area and that of the U sign in relation to the largest transverse axis of the proximal tibia (p = 0.72). However, an analysis of the two image exams revealed a difference between the mean values of width in relation to the largest transverse axis of the proximal tibia (p \ 0.0001). Conclusions: The data obtained in this study indicates that there is a relationship between the height of the PCL anatomic site and the height of the U sign. Therefore, when radiographs of lateral view of the knee prove inconclusive and the U sign is present on radiographs of anteroposterior view of the Knee, the diagnosis of PCL avulsion fracture at the tibial insertion should be investigated.

P23-492 Intraoperative navigation evaluation of double-bundle posterior cruciate ligament reconstruction Y. Ishibashi1, E. Tsuda1, Y. Yamamoto1, H. Tsukada1, S. Maeda1 1 Hirosaki University Graduate School of Medicine, Orthopaedic Surgery, Hirosaki, Japan Objectives: Recently, double-bundle (DB) posterior cruciate ligament (PCL) reconstruction, which reproduces both the anterolateral (AL) and posteromedial (PM) bundles, has been advocated to improve clinical outcome. The purpose of this study was to evaluate our current DB-PCL reconstruction technique with hamstring tendon, especially the function of the AL and PM bundle.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: Eight patients, who received DB-PCL reconstruction using a navigation (OrthoPilot, B/Braun Aesculap, Germany), were included in this study. Using the additional function of the navigation, posterior displacement of the tibia were measured at 15, 30, 45, 60, 75 and 90 knee flexion in four phases: before reconstruction, after PM bundle fixation, after AL bundle fixation, and after DB-PCL reconstruction (both PL and AM bundle fixation). Results: After PM bundle, or AL bundle fixation, posterior displacement significantly decreased at each knee flexion angle. However the PL bundle and the AL bundle were functionally different. Posterior displacement after DB-PCL reconstruction was significantly improved compared to those after PL or AM bundle fixation. Conclusions: To our knowledge, this is the first study that assesses the DB-PCL reconstruction in vivo using navigation system. Based on this result, DB-PCL reconstruction may be recommended for improved knee stability.

P23-595 Arthroscopic popliteus tendon reconstruction: introduction a novel technique and primary results M. Razi1, A. Safar Cherati2, H. Dadgostar3, H. Razi4, K. Ahadi5 1 Tehran University of Medical Sciences, Department of Orthopedic Surgery, Rasoul Akram Hospital, Tehran, Islamic, Republic of Iran, 2 Tehran University of Medical Sciences, Sports Medicine Department, Tehran University of Medical Sciences, Tehran, Islamic, Republic of Iran, 3Sports Medicine Rasoul Akram Hospital Tehran, Tehran, Islamic, Republic of Iran, 4Julius Wolff Institute, Center for Musculoskeletal Surgery, Charite-Universita¨tsmedizin, Berlin, Germany, 5Orthopaedic Surgeon Milad Hospital Tehran, Tehran, Islamic, Republic of Iran Objectives: The aim of the study is to introduce anovel arthroscopic method of posterolateral structure (PLS) reconstruction in Posterolateral Rotary Instability (PLRI) of the knee. In open surgical methods, apart from big surgical approach and tissue dissection, probability of injury to remaining intact parts of posterolateral structures and neurovascular damage is significantly higher; otherwise these techniques if not large enough do not allow definite assessment of anatomic landmarks. Moreover Limitation of Range of Motion and arthrofibrosis are two main pitfalls that surgeons confront in follow-up of patients in open surgery, thus the arthroscopic technique provides a better visualization of anatomic landmarks with least site morbidity. Methods: Between August 2005 to April 2010 Thirty-nine patients (8 females, 31 males) with chronic posterolateral structure injury and normal knee alignment have undergone arthroscopic Tibialis posterior allograft, posterolateral complex reconstruction, 27 of them had combined ACL and PLS injury and 9 had been complicated by PCL and PLRI, in 3 of them, injury involved ACL, PCL and PLS. Regarding to physical examination, imaging and arthroscopic evaluations, Grade I instability was treated by Modified Larson Technique with Semitendinosus auto graft when popliteus tendon had not been severely injured. Based on aforementioned assessments If injury has been evaluated as grade II involving popliteus tendon component, arthroscopic reconstruction of popliteus tendon according to open method of Muller and LaPrade was the preference method whereas the tibial tunnel has been created from medial plateau, proceeding in posterolateral direction regarding to

S233 anatomic position of popliteus tendon and in Grade III Arthroscopic popliteus tendon reconstruction and modified Larson Technique were applied concurrently . Results: All patients have been followed up for average 27 ± 1 month (6 months to 5 years) postoperatively. varus and external rotation instability restored with arthroscopic PLS reconstruction and all patients had near normal knee stability and significant improvements in the scores for pain and activities of daily living subjectively and 2000 IKDC Examination Form score for varus, external rotation, reverse pivot shift and single leg hop objectively. There were no cases of arthrofibrosis or limitations in knee motion. The advantages of arthroscopic-assisted technique of PLS reconstruction includes restoration of varus and external rotation laxity, more precious anatomical position of the graft. And reduced potential of surgical damage comparing to open PLS reconstruction. Conclusions: This arthroscopic technique of reconstruction of popliteus tendon restores near normal anatomic stability of posterolateral corner of the knee with least surgical morbidity especially for creation of tibial tunnel.

P23-619 Efficacy of posteromedial safety incision in PCL reconstruction, our experience with first 100 cases A. Goyal1, D. Chaudhary1, D. Joshi1, V. Bahl1 1 Sports Injury Centre, Delhi, India Objectives: The posterior cruciate ligament (PCL) is described as the primary stabilizer of the knee. Since PCL is located in close proximity to the neurovascular bundle, reconstructing PCL is technically demanding and associated with complications such as injury to the popliteal artery. This injury can occur during the creation of the posteromedial portal, the manipulation of the tissues in the posterior part of the capsule of the knee joint, or when drilling the tibial tunnel. To solve these problems, we used the extra capsular posteromedial safety incision (as described by Fanelli) for the trans tibial PCL reconstruction not only to minimize the iatrogenic injury to the vascular bundle, but also for creating an ideal route for the graft which includes consideration of the directions of the bone tunnel, by directly feeling the exit point of guide wire and thereby minimizing the killer turn of the graft. Methods: We evaluated 100 patients prospectively of isolated PCL injury with a follow up of 2 years. Clinical examination was the mainstay for the preoperative diagnosis. Radiographs were obtained in 2 planes in the standing position. Magnetic resonance imaging was routinely performed in all patients to rule out other ligamentous injuries. The patients were operated with trans tibial technique using the posterio-medial safety incision Each patient was evaluated on the basis of Lysholm knee score and anteroposterior translation was measured with KT-1000 arthrometer. Two year follow-up was completed in 93 patients. Results: The statistical test used in the study was paired t test with the p value being set at 0.005. Evaluation was performed using a scoring scale proposed by Lysholm and Gillquist. At 2 year follow-up, knee function was considered excellent in 22 (23.6%) patients, good-toexcellent in 61 (65.5%) patients, and fair-to-good in 8 (8.6%) patients, based on the Lysholm knee score. Results were poor in 2 (2.1%) patients. The total antero posterior translation measured by arthrometer (KT-1000, manual maximum) showed a range of 2–5.5 mm (mean

123

S234 3 mm). On the antero posterior view the exit point of the tunnel was 0.5 cm lateral to the midline (range 1–1.5 cms) in 80 patients however,13 patients had their tunnel placement 1.5 cm medial to the midline (range 0.8–2 cms). Postoperative lateral radiographs showed tunnel placement 2 cms below the joint line (range 1–2.5 cms). The average graft tibial tunnel angle was 60 (range 55–67). Conclusions: The arthoscopically assisted PCL reconstruction technique is a reproducible surgical procedure. Factors contributing to the success of surgical technique include accurate tunnel placement, minimization of graft bending and implementation of appropriate postoperative rehabilitation program however the surgeon’s finger through the extracapsular posteromedial incision in monitoring the position of the guide wire as well as the instruments used in PCL reconstruction remains one of the single most important step for success and safety of the procedure.

P23-627 High prevalence of abnormal MR findings of lateral collateral ligament and popliteus tendon in the knees without instability C.B. Chang1, J.-Y. Choi2, M.J. Chang1, J.H. Lee1, T.K. Kim1 1 Seoul National University Bundang Hospital, Orthopaedic Surgery, Seongnam-si, Republic of Korea, 2Seoul National University College of Medicine, Radiology, Seoul, Republic of Korea Objectives: Because of well known difficulty in diagnosis of chronic posterolateral rotary instability (PLRI), magnetic resonance imaging (MRI) for the posterolateral corner of the knee has received attention. However, in our practice, abnormal MR findings of the posterolateral corner structures, particularly abnormal signal intensity and abnormal thickening of the lateral collateral ligament (LCL) and the popliteus tendon were frequently found in the patients without PLRI and/or varus instability. This study aimed to determine prevalences of alterations in the signals and thicknesses of the LCL and the popliteus tendon on the MRI in the subjects without clinical evidence of knee instability and to analyze the factors associated with the alterations. Methods: MRI scans in 120 subjects (79 men, 41 women; mean age, 36.8 years; range, 14–62 years) who had not any kind of knee instability were evaluated to assess the LCL and the PT in terms of signal alteration and thickness. Demographic characteristics (age, gender, height, and weight), Kellgren-Lawrence radiographic OA grading, mechanical alignment of the knee in the coronal plane, and condition of the lateral meniscal were assessed to determine the associated factors with the MR findings of the LCL and the popliteus tendon. Results: The mean thicknesses of proximal portion of the LCL and the popliteus tendon on MR images were 5.3 ± 1.0 mm (range, 3.1–8.1 mm) and 3.8 ± 0.6 mm (range, 2.4–6.0 mm), respectively. The signal alterations of the LCL and the popliteus tendon were found in 33% and 29%, respectively, while the abnormal thickening of the LCL and the popliteus tendon were found in 33% and 8%, respectively. In the logistic regression analyses, degrees of varus alignment (mean = varus 1.5, range: varus 6.1–valgus 8.1) was found to be the only predictor for the LCL signal alteration [Odds ratio (OR) = 0.81, 95% confidence interval (95% CI) = 0.66–0.98) and abnormal thickening (OR = 0.73, 95% CI = 0.59–0.90). Other factors including gender, age, weight, height, degree of radiographic knee OA, and condition of the lateral meniscal had no associations with signal

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 alteration or thickening of the LCL on the MRI. Among the factors evaluated, any factors were not associated with the alterations of the popliteus tendon. Conclusions: Even in the subjects without varus and posterolateral rotary instability, signal alteration and abnormal thickening of the LCL and/or the popliteus tendon on the MRI could be frequently found. Thus a knee surgeon should be cautious when the results of MR features of the LCL and the popliteus tendon are being considered for the diagnosis of the chronic PLC injury of the knee, particularly in the patients with varus knee alignment.

P23-726 Femoral critical corner in posterior cruciate ligament reconstruction: a comparison of three techniques using 3-dimensional model S.-H. Kim1, Y.-M. Chun1, S.-H. Kim1, M. Jung1, S.-K. Lee1, S.-J. Kim1 1 Arthroscopy and Joint Research Institute, Yonsei University, Orthopaedic Surgery, Seoul, Republic of Korea Objectives: In arthroscopic posterior cruciate ligament reconstructions, the critical corner angle formed between the graft and the intraarticular aperture of the femoral tunnel as well as killer turn formed between the graft and the tibial tunnel in transtibial technique has been considered to be responsible for increased attritional stress. However, most studies involving the critical corner angle were done only in the context of 2-dimention instead of 3-dimention and no consideration was given to changes made with different flexion angles of the knee. Therefore, the aim of this study was to compare the critical corner angles formed by three different femoral tunneling techniques (i.e.) conventional outside-in technique, the author’s inside-out technique (Inside-out A: knee 110 flexion, far anterolateral portal used) and modified inside-out technique (Inside-out B: knee 90 flexion, anterolateral portal used), as measured by CT scan. Methods: First of all, using Visual and analysis program (Mimics 14.1, Materialise, Belgium), we created 3-dimentional images from DICOM (digital Imaging and Communications in Medicine) file obtained by CT scanning 9 fresh frozen cadaveric (age range : 45 * 77) knees with normal bone and ligaments at 0, 45, 90 and 120 flexion. Then, we passed three different guide pins from the same point on the intra-articular aperture of femoral tunnel to emerge outside of the femur, by three different techniques aforementioned. A guide pin was also inserted through anterolateral transtibial tunnel. Then, the knees were again CT-scanned. The intra-articular insertion and emerging points of the guide pins of the three different techniques were registered into the previously obtained cadaveric model. The critical corner angle in each flexion angle in each technique was measured and analysed by linear mixed model. Results: At 0, no statistically significant difference was found among the three techniques. At 45, the critical corner angles of outside-in, inside-out A and inside-out B techniques were 140.4 ± 1.4, 137.1 ± 1.4 and 117.8 ± 1.4, respectively. There was no statistically significant difference between outside–in and inside–out A techniques (p = 0.127). However, statistically significant difference was found between inside–out A and inside–out B techniques (p \ 0.0001). The same results were shown at 90(outside–in, 123.3 ± 2.3; inside–out A, 119.3 ± 2.3; inside–out B, 95.3 ± 2.3)

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S235

and 120(ouside–in, 111.4 ± 2.3; inside–out A, 107.7 ± 2.3; inside–out B, 85.1 ± 2.3). Also, the critical corner angles of outside–in and inside–out A techniques were statistically larger than the killer turn at all flexion angles (p \ 0.0001) whereas that of inside– out B technique was statistically smaller (p \ 0.0001). Conclusions: Inside–out technique for femoral tunneling through far anterolateral portal with the knee flexed 110 and the proximal tibia posteriorly translated results in similar critical corner angles throughout 0 * 120 flexion compared with outside–in technique.

P23-729 Correction of chronic irreducible posterior knee subluxation in multiple ligament injury patients using external fixation distraction: A novel technique S. Saowaprut1, T. Charoensopa2, A. Chinanuvathana2 1 Institute of Orthopaedics, Lerdsin Hospital, Ministry of Public Health, Bangkok, Thailand, 2Institute of Orthopaedics, Lerdsin Hospital, Bangkok, Thailand Objectives: Neglected traumatic knee dislocation is an uncommon but complex and difficult to treat problem. Especially, when multiple ligament injury was left untreated for months before seeking a medical service. This may cause a significant and permanent disability. The ideal goal of treatment for patient with knee dislocation is to obtain painless, stable and full range-of-motion knee. Unfortunately, to achieve all the goals can be very difficult because when the treating knee is stable, usually it will reveal limited knee range-ofmotion. Posterior sag of tibia is one of the most common problem after prolonging neglected knee dislocation. This posterior sag will eventually becomes irreducible with the times in some cases. If the knee is in irreducible position before we perform further ligament reconstruction, the reconstructed graft will be put too much stress leading to failure of reconstruction and poor outcome. Methods: We proposed novel technique to corrected irreducible posterior knee subluxation using distraction external fixator. After applying the external fixator in 90 knee flexion position, we gradually reduced the knee joint using distractor device that applied between femoral and tibial external fixators. The distraction rate is 0.5–1 mm per day and the rate was carefully adjusted according to patient0 s comfort. We reassessed the joint position weekly by radiography. After desired knee joint position was achieved, we removed the external fixator and performed further surgical treatment such as ligament reconstruction depending on patient0 s knee pathology. Results: In this present paper, we report 2 cases of chronic irreducible posterolateral knee dislocation and 1 case of neglected ipsilateral fracture of femoral shaft and proximal tibia with posterolateral knee dislocation. We had treated irreducible knee joint by external fixation distraction technique before we performed posterior cruciate ligament (PCL) and posterolateral complex reconstruction in 2 patients with chronic irreducible posterolateral knee dislocation. While the third patient with neglected floating knee and chronic irreducible posterolateral knee dislocation had used same technique to correct irreducible knee before performing high tibial corrective osteotomy to correct malunion of proximal tibia.

Preoperative picture

Distractive external fixator

Postoperative picture Conclusions: After recovery period, all the three patients0 knees were stable and achieved functional range-of-motion. Furthermore, our patients are satisfied with the results and their quality of life was improved. P23-893 Combined PCL and PLC reconstruction in chronic posterolateral instability M. Alam1, C. Zorzi2, V. Iacono3, V. Madonna2, D. Rosa4, N. Maffulli5 1 Royal National Orthopaedic Hospital, Middlesex, United Kingdom, 2 Sacro Cuore Don Calabria Hospital, Orthopaedic, Negrar, Italy, 3 Hospital ‘‘Sacro Cuore—Don Calabria’’, Orthopaedic Department, Knee Surgery Centre, Verona, Italy, 4University of Naples Federico II, Orthopaedic Department, Napoli, Italy, 5Queen Mary University of

123

S236 London, Barts and The London School of Medicine and Dentistry, Centre for Sports and Exercise Medicine, London, United Kingdom Objectives: The posterolateral corner (PLC) is more likely to be injured in combination with the posterior cruciate ligament (PCL) or the anterior cruciate ligament (ACL) than in isolation. This leads to instability of the knee and loss of function. We hypothesised that combined PCL and PLC reconstruction would restore sufficient stability to allow improvement in patient symptoms and function. Methods: 19 patients who underwent arthroscopic-assisted singlebundle PCL and PLC reconstruction by a single surgeon were analysed retrospectively. The PLC reconstruction was a modified Larson reconstruction of the lateral collateral ligament and the popliteofibular ligament. The IKDC and Tegner scores were used to assess outcome. Dial test and varus laxity were used to assess improvements in clinical laxity. Posterior laxity was tested using the KT-1000. Results: The mean follow-up was 38 months. There were no postoperative complications. All patients had \5 mm posterior step-off. 17 of 19 patients had negative dial and varus stress tests. Measured range of motion was reduced by a mean of 10, but patients did not report any daily activities restrictions. Tegner scores improved significantly (p \ 0.05) from 2.3 to 6.4 at final follow-up. The mean postoperative IKDC score was 86. Conclusions: Our results show that single-bundle PCL reconstruction with an isometric PLC reconstruction reduced abnormal knee laxity and improved functional outcome. This technique may reduce surgical morbidity. Improvements are necessary to improve the initial diagnosis, and therefore reduce the time from injury to surgery, and would need better education for primary care physicians. Additionally, further work in identifying ideal tunnel positions may improve knee function allowing return to daily activities and greater sports participation.

P23-1023 Quantification of posterior knee laxity in PCL injury: instrumented laxity measurement versus stress radiography J. Ho¨her1, R. Akoto2, S. Shafizadeh3, O. Greshake4, B. Bouillon2, M. Balke5 1 Praxis fu¨r Sporttraumatologie Ko¨ln, Ko¨ln, Germany, 2University of Witten/Herdecke, Department of Trauma and Orthopedic Surgery, Cologne Merheim Medical Center, Cologne, Germany, 3University of Witten/Herdecke, Cologne Merheim Medical Center, Cologne, Germany, 4Clinic for Sports Traumatology, Cologne, Germany, 5 Klinikum Ko¨ln Merheim, Unfallchirurgie, Orthopa¨die und Sporttraumatologie, Ko¨ln, Germany Objectives: To compare test results of stress radiographic imaging and instrumented drawer testing (Rolimeter) for the assessment of posterior knee laxity in patients with PCL injury. Methods: In 45 patients presented to our clinic with an injury to the PCL lateral radiographs at 90of knee flexion of both knees were obtained under anterior and posterior tibial load (150 N) using a standardized approach. Patients with injury of the ACL to either knee were excluded from the study. During the same clinic visit an instrumented measurement of anterior-posterior knee laxity was performed at 90 of knee flexion by an experienced examiner using the Rolimeter device. Lateral radiographs were analysed with lines being drawn on the radiograph representing the tibial plateau and perpendicular to this the posterior edges of the condyles and the posterior aspect of the tibia. The distance of these lines represent the amount of posterior drawer. For the instrumented measurements device the total AP laxity (difference between maximum anterior and maximum posterior displacement) were obtained, the difference of both represents the amount of posterior laxity. The mean side-to-side differences of both (radiographic and instrumented) measurements were compared to each other, the correlation was calculated using the Pearson coefficient.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: The mean side to side difference (injured/non-injured side) of the radiographs was 7.8 (SD 4.1) mm and of the instrumented testing 7.6 (SD 3.2) mm. The mean difference between radiographic and Rolimeter test results of 2.1 (SD 2.7) mm. The Pearson correlation coefficient revealed a strong correlation between both measurements of 0.6. Conclusions: Our results suggest that under the absence of ACL injury the instrumented laxity measurement is a suitable tool to assess the posterior laxity of the knee in the clinical setting. The Rolimeter device might replace stress radiography for follow-up visits and therefore reduce the amount of radiation to patients.

P23-1032 Posterolateral corner reconstruction using single fibular sling method J.G. Kim1, B.S. Yang1, J.K. Ha1 1 Inje University, Seoul Paik Hospital, Seoul, Republic of Korea Objectives: This study was carried out to verify the effectiveness of restoring external rotation and varus stability of novel single fibular sling. Methods: A retrospective analysis was made of 67 patients who underwent PLC reconstruction by single experienced surgeon (JGK) between October 2005 and October 2008 for PLRI. The inclusion criteria included a disruption of the PLC (diagnosed by physical examination, MRI, and arthroscopic finding), minimum follow-up of 2 years, and primary PLC reconstruction using anatomic single fibular sling method. Exclusion criteria included patients with prior surgery of the PLC, PLC reconstruction using other than single fibular sling method, an associated fracture in the same leg that knee function could be affected, and fibular head fracture. The functional and clinical outcomes were assessed by clinical examination, Lysholm scores, International Knee Documentation Committee (IKDC) subjective scores and stress radiograph. Results: The mean side to side difference in varus laxity was reduced from 2.32 ± 1.33 mm preoperatively to 0.37 ± 1.48 mm after reconstruction. The external rotation was checked at last follow up. 58 patients of 60 show grade 0 (n = 42) or grade I (n = 16). There is no grade 2 case. And there were 2 cases of failure, who showed over 10 mm of posterior drawer test with grade III external rotation instability. The mean Lysholm score was 52 ± 6.7 (range 42–64) preoperatively and 87.7 ± 12.3 (range 61–100) postoperatively. The difference between Lysholm scores before reconstruction and at last follow-up evaluation was significant (p \ .001). The mean IKDC subjective score was 44.8 ± 6.3 (range 34.5–55.1) preoperatively and 77.5 ± 14.2 (range 49.4–98.8) postoperatively. Conclusions: Anatomic PLC reconstruction using single fibular sling might be effective, reproducible and relatively simple procedure for most patients. But careful preoperative evaluation is essential and additional treatment would be considered in severe external rotational instability.

P23-1219 The use of fibroblasts for the reconstruction of anterior cruciate ligament: research on the sheep animal model P. Guillen-Garcia1, M. Guillen-Vicente1, I. Guillen-Vicente1, E. Rodriguez-Inigo1, R. Caballero-Santos1, J.M. Lopez-Alcorocho1 1 Clinica CEMTRO, Madrid, Spain Objectives: The rupture of the anterior cruciate ligament (ACL) is a relative common lesion, which currently is treated with a surgical procedure that implies the use of different tendons or ligaments to reconstruct the damaged ACL. Currently, the research on this field is

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 focused in finding a new method to reduce the time of recovery which with these techniques is now of 6–8 months. We have investigated the use of type I/III collagen membranes as a carrier for fibroblasts in the reconstruction of broken ACL in the sheep. Methods: The overall project can be divided into 2 different phases: Phase I: Set up of fibroblast cultures We have included ACL biopsies from patients with ACL rupture and from control (patients who undergone a knee surgery for reasons other than ACL rupture). Fibroblasts were isolated from ACL biopsies; the number of cells was estimated and cultured in monolayer. Phase II: Study in the sheep model of the reconstruction of the ACL ruptured with fibroblasts embedded in type I/III collagen membrane Ten female sheep with a similar age will be included in this study and were divided into 2 groups. • Group A: Implanted with 5 million fibroblasts embedded in the membrane. • Group B: Implanted with the membrane without cells. The animals undergone 2 surgeries: one surgery to take an ACL biopsy and the other one to break the ACL and implant the membrane with (Group A) or without (Group B) cells. After 3 months, the animals will be sacrificed and samples from the ACL regeneration and from healthy areas (control) will be taken. Histological and molecular studies will be performed to compare both treatments between them and with the control. Results: Phase I The fibroblast culture could be established in all cases. A negative correlation was observed between the age of the patient and the growth rate. Phase II The architecture of normal ACL was not conserved either in the ACL treated with the membrane with or without cells. However a high number of cells, similar to fibroblast was found in the celltreated ACL than in those treated only with the collagen membrane, indicating that probably these cells migrated from the membrane to the damaged ACL. RT-PCR studies performed demonstrated that these cells expressed type I collagen, tenascin-C and MMP-13; indicating the fibroblastic origin of the cells. Conclusions: We think that this novel technique could be a promising tool to treat the ACL rupture and represents a first step in the use of tissue engineering for treatment of the ACL rupture. However, further studies using another carriers and scaffolds may be performed.

P23-1268 Mid-term clinical and radiographic outcomes for surgical reconstruction of multiple ligament knee instability M. Hantes1, A. Tsarouhas1, I. Antoniou1, G. Basdekis1, K. Malizos2 1 University Hospital of Larisa, Orthopaedic, Larisa, Greece, 2 University Hospital of Larissa, Orthopaedic Surgery Department, School of Medicine, Larissa, Greece Objectives: Multiligament knee injuries are defined as rupture of at least two of the following: anterior cruciate ligament (ACL), posterior cruciate ligament (PCL), and posterolateral corner (PLC) and posteromedial corner (PMC). They are rare yet debilitating injuries. The purpose of this study was to examine the mid-term clinical and radiographic outcomes of patients who underwent arthroscopically assisted reconstruction for multiple knee ligament instability in our department.

S237 Methods: A total of 31 patients who underwent multiligament knee reconstruction between January 2003 and December 2008 were included. They were 26 males and 5 females with an average age of 30.8 ± 8.1 years. The IKDC, KOOS, Tegner, Lysholm and Euroqol5D scores were obtained. Anteroposterior knee laxity was assessed using the KT-1000 arthrometer. Progression of osteoarthritic changes on postoperative radiographs was classified using the Kellgren– Lawrence (K–L) grading scale. Results: The commonest types was ACL + PCL + PLC rupture (10 patients) and PCL + PLC rupture (9 patients), followed by ACL + PCL tear (5 patients), ACL +PLC tear (4 patients) and ACL + PCL + MCL tear (3 patients). The mean follow up was 48.25 ± 19.2 months. Two patients had peroneal nerve palsy with concurrent drop foot that was treated with tendon transfers at the time of ligament reconstruction. Two cases presented with popliteal artery thrombosis secondary to intimal tears. The mean time interval between the injury and ligament reconstruction was 14 ± 3.1 months. No complications were occured intra- or post-operatively. At the time of follow up, there was no extension deficit, while knee flexion averaged 133 (range 110–150). KT-1000 side-to-side differences were 1.7 ± 0.3 mm. The mean IKDC, KOOS, Tegner, Lysholm and Euroqol5D scores were 68.2 ± 11.2, 54 ± 11.2, 4.3 ± 0.8, 81 ± 4.2 and 74 ± 5.3 respectively. Progression by one grade in the K–L scale was evident in 11 patients (35.5%) involving osteophyte formation in all cases. Conclusions: In the mid-term, arthroscopically assisted reconstruction of multiligament knee instability provides good to excellent results in terms of stability. However, functional outcomes are variable and return to high-level activity is not common. Concomitant neurovascular injuries adversely affect functional outcomes.

Knee-Meniscus I

P24-81 Development of a novel test method for intra-articular fatigue and wear testing of meniscal reconstructions C. Halewood1, D.H. Nawabi2, A.A. Amis1 1 Imperial College London, Mechanical Engineering, London, United Kingdom, 2Hospital for Special Surgery, Adult Reconstruction and Joint Replacement, New York, United States Objectives: Artificial and tissue engineered constructs for meniscus and articular cartilage repair are being developed. Their behaviour has been studied in simple tension and compression tests, but little is known about their mechanical performance under physiological loading, particularly in the early stages of rehabilitation when tissue ingrowth has not yet occurred. An in vitro biomechanical test method has been developed so that these constructs can be loaded physiologically and be assessed for wear and fatigue and so that the articular cartilage can be examined for damage. The objective of this particular study was to examine the fatigue performance of a porous polyurethane meniscal scaffold and its effect on the articular cartilage on the femoral condyles. Methods: The ovine stifle joint was used as a model of the human knee as it is similar both anatomically and biomechanically. A test rig was designed to accommodate an ovine hind leg from below the hip to mid shank. The rig was mechanically driven and moved the femur through the range of motion of ovine gait—45–80 flexion. The joint was loaded to approximately twice body weight during the stance

123

S238 phase and the tibia was mounted on a ball joint, allowing it to rotate freely. Thus, the kinematics of the stifle joint were driven by the cruciate ligaments and other soft tissues surrounding the joint capsule. Lubrication of the joint was via a drip and catheter. The rig ran inside a refrigerator to allow for lengthy tests, with 500,000 cycles representing 6 months of walking. Partial open meniscectomies were performed on 10 stifle joints: 5 then had a biodegradable meniscal scaffold implanted; the remaining 5 were used as controls. The joints were dissected after testing and the device and articular cartilage examined for damage and graded using the ICRS scale. Particles of debris from the device were extracted from the joint and remaining portions of the device examined for pore integrity using SEM. Results: Examination of the articular cartilage after testing suggests that, without tissue ingrowth, meniscal scaffolds may abrade both the femoral and tibial articular cartilage and cause damage that is worse than with a partial meniscectomy only. The SEM examination showed some permanent deformation of the scaffold and particle analysis revealed particles of approximately 1 mm in diameter in the joint capsule. Conclusions: A novel test rig was designed, manufactured and tested to enable in vitro examination of artificial constructs for knee joints. The test represents the worst-case scenario of there being zero tissue ingrowth into the scaffold post-implantation and indicates that in this case, the scaffold may be worse than a partial meniscectomy only, in terms of damage to the articular cartilage. This test method may also be used to evaluate articular cartilage and ligament constructs.

P24-101 Inter- and intra-observer reliability of the Genovese grading to evaluate the radiological outcome after partial meniscus substitution L. Schenk1, A. Hirschmann2, L. Keller3, M.P. Arnold1, N.F. Friederich1, M.T. Hirschmann1 1 Kantonsspital Bruderholz, Klinik f. Orthop. Chirurgie & Traumatologie, Bruderholz, Switzerland, 2Kantonsspital Bruderholz, Radiology, Bruderholz, Switzerland, 3Sport Clinic, Zu¨rich, Switzerland Objectives: The Genovese grading is one of the most commonly used scores for the evaluation of radiological outcome after partial meniscus substitution using collagen meniscus. It was the purpose of our study to evaluate the intra- and inter-observer reliability of the Genovese grading on MRI in patients after collagen meniscus substitution. Methods: The MRI images of 79 consecutive patients who underwent partial meniscus substitution using collagen meniscus (CMI) were assessed. The Genovese grading was used, which consists of direct (implant morphology/size, signal intensity) and indirect (state of the corresponding cartilage of the medial/lateral femur or tibia, size of the cartilage lesions [\50%, signs of bone marrow edema). The interand intra-observer reliability was assessed using intra-class correlation coefficients (ICCs, 1 = highest, 0 = lowest). Two observers performed the grading with 2 week interval twice. The sample size was calculated according to Walter et al. Results: The Genovese grading for the morphology/size of the implant showed an ICC intra-observer-reliability of 0.456–0.775 and an ICC inter-observer-reliability of 0.256–0.614. The Genovese grading for the signal intensity of the implant showed an ICC intra-

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 observer-reliability of 0.469–0.651 and an ICC inter-observerreliability of 0.287–0.485. The Genovese grading for the bone marrow edema of the implant showed an ICC intra-observer-reliability of 0.702–0.740 and an ICC inter-observer-reliability of 0.667–0.804. Conclusions: In the clinical routine the Genovese classification shows only moderate inter- and intra-observer reliability for the evaluation of partial meniscus substitution using collagen meniscus implants. This finding is particularly true for the evaluation of the implant size. We believe that the implant size should be better evaluated on 3Dreconstructed MRI.

P24-105 Does the meniscus transplant prevent osteoarthritis? Functional and radiographic outcome 5-year follow-up H. Valencia-Garcı´a1, C. Gavı´n-Gonza´lez2, J.R. Prieto-Martinez2, J.E. Ruiz Zafra2, A. Chozas-Mun˜oz2, A. Ferrete-Barroso2 1 Hospital Universitario Fundacio´n Alcorcon, Traumatology, Alcorcon, Spain, 2Hospital Universitario Fundacio´n Alcorcon, Madrid, Spain Objectives: Partial meniscectomy is a very common procedure for treatment of meniscal injuries. The absence meniscal promoting the progression of chondral degeneration and as a result of osteoarthritis with joint space narrowing. Meniscal transplantation has been proposed for the symptomatic relief after meniscectomy, but has not yet been established whether slowing the onset of degenerative changes. Methods: We retrospectively reviewed meniscus transplants performed in our center. Inclusion criteria included at least 5 years follow-up and no other surgical maneuvers (on anterior cruciate ligament, osteotomy or chondral injuries). Finally we analyzed the results of 10 patients with mean age 34.5 years (21–45). The technique used is frozen at -80 unirradiated meniscal allograft and transplantation without bone blocks. Results: 6 were lateral and 4 medial meniscus. In all scales compared Lysholm, Tegner and VAS and joint space was measured on a preoperative radiograph charge and 5 years, and the radiological progression as Ahlba¨ck criteria. We obtained a satisfaction rate of 78% at 5 years, with an improvement in all scales (average preoperative Lysholm 55–85 to 5 years, Tegner from 3.5–6 and VAS from 6.8 to 2). We do not see a significant joint space narrowing (mean preoperative of 3.09 mm and 3.01 mm at 5 years) with 2 cases of improvement of the same (cases 2 and 7). Conclusions: The meniscus transplant is effective for pain control in symptomatic knee after meniscectomy, with a success rate of 60–88%. Also appears to decrease the short-term joint degeneration, but still do not know its importance in the long-term chondroprotective effect. The size and graft fixation are important prognostic factors transplant. The poor results are associated with graft irradiated limb malalignment or significant chondral degeneration. We did not find deterioration of affection compartment joint space at 5 years followup, but these results must be confirmed in longer term studies.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P24-122 A pilot study of the use of a polycarbonate-urethane implant for the treatment of postmeniscectomy medial knee pain P. Verdonk1, A. Dhollander2, R. Verdonk2 1 Gent University Hospital, Orthopaedics, Gent, Belgium, 2 Gent University Hospital, Orthopaedic Surgery and Traumatology, Gent, Belgium Objectives: To evaluate short-term clinical and MRI outcome of a polycarbonate-urethane meniscus replacement implant (NUsurface, Active implants) for the treatment of postmeniscectomy medial knee pain. Methods: Five patients aged 41–57 years with postmeniscectomy medial knee pain were treated with this new meniscus replacement device. Patients with evidence of grade IV articular cartilage loss were excluded. Clinical outcome was measured over 12 months by the KOOS, IKDC subjective, EQ-5D and VAS for pain. Serial MRI scans were taken at 6 weeks, 6 months and 12 months of followup. Results: The patients included in this study showed a significant clinical improvement after the procedure. The MRI findings of this pilot study were considered to be promising. No signs of deterioration of the surrounding cartilage or of the device were observed. No failures occurred among the 5 patients until now. Conclusions: This investigation provided useful information on the safety and efficacy of this treatment in a challenging patient cohort. The short-term clinical and MRI outcome are promising. Large scale trials are mandatory to confirm the results and the reliability of this device.

P24-398 A novel technique of the all-inside repair for lateral meniscus root tears using the meniscal viper: a technical note T. Soejima1, K. Noguchi1, K. Tabuchi1, T. Kanazawa1, K. Nagata1 1 Kurume University, School of Medicine, Orthopaedic Surgery, Kurume, Japan Objectives: In recent years, various repair techniques are tried to posterior lateral meniscus root tears in order to re-gain the meniscus hoop stress. However, almost of them were seem to be very difficult. Here, we report a novel simple technique of all inside repair for posterior lateral meniscus root tears using the meniscal viper. Methods: Our meniscal repair technique is mentioned below. After the rasping and abrasion of the both stump ends of the tear, one stitch of loop thread is put on one side of the stump of the tear according to normal manner of the meniscal viper. However, this thread is left there without making knots as a loop-shaped lead thread. Another loop-shaped lead thread is put on another stump and left there, as a same manner. As we pull it up through a new different thread to these two loop-shaped lead threads, the horizontal mattress suture that stepped over the tear is completed. Finally, a thread is tightened by sliding knot techniques and lets bilateral stump ends adhere. This procedure is repeated more than three times. A second thread is installed as coming to a cross-stitch suture together with the first thread in order to avoid cutting out by thread. A third thread is installed at inferior surface of the tibial side of the meniscus.

S239 In this study, 6 menisci in six cases (4 males, 2 females) who underwent an above mentioned procedure were evaluated by Lysholm score, Barrett’s parameters, and MRI, in a minimum of 1 year after the operation. The mean age at operation was 23 years old. All of them were associated with ACL injury. Results: Lysholm score was improved from 60 to 91.5 points. None of patients had any signs of Barrett’s parameters. LM extrusion in coronal views of the MRI was improved from 2.6 to 1.7 mm, but was not statistically significant. Conclusions: Posterior lateral root tears completely lose the meniscus hoop stress, but is clinically asymptomatic than other types of injury. Therefore, we must avoid any complications induced by the procedure if we challenge to repair. Our procedure can be performed through only conventional 2 anterior portals without any additional portals and incisions. Neurovascular injuries don0 t happen. Advanced skills such as mastering of 70 arthroscope are not necessary, like the suture hook method.

P24-438 A novel acellular and viroinactivated meniscus allograft N. Tan1, E. Servien2, H. Couchoux1, L. Barnouin1, P. Neyearset2 1 TBF Tissue Engineering, Mions, France, 2Hoˆpital de la CroixRousse, Centre Albert Trillat, Department of Orthopaedic Surgery, Lyon, France Objectives: Meniscal tears are among the most common knee injuries. To preserve as much as possible the joint from osteoarthritis, suturing, or for larger damages, partial and total meniscal replacements are necessary. To combine the biocompatibility and mechanical resistance of meniscus allograft to the disponibility of synthetic substitutes, we developed an acellular, viroinactivated and sterile scaffold with well-preserved structure for partial and total meniscal repair. Methods: Human menisci were collected from living donors undergoing total knee arthroplasty and were selected on macroscopic integrity criteria. They underwent chemical treatments, freeze-drying and gamma radiation. Then, efficacy of the process and safety of the product were studied. Decelluarization of menisci was analyzed by hematoxylin-eosin staining. Preservation of the matrix structure was explored by histological studies including hematoxylin-eosin staining and Safranin O staining, and by differential scanning calorimetry. Ultrastructure of the meniscal scaffold was analyzed by scanning electron microscopy. Biomechanical studies were also conducted. Finally, viroinactivation of the scaffold was investigated by viral clearance studies on 4 selected viruses. Results: Histological data evidenced complete decellularization of the menisci. The process promoted a high level of porosity within the tissue with an homogeneous pore distribution. Glycosaminoglycans, present in the center of native menisci, were eliminated by the process (Fig. 1). Differential scanning calorimetry showed a rather well-conserved collagen structure following the process. The meniscal surface as well as collagen circumferential and radial fibers were preserved as assessed by electron scanning microscopy (Fig. 2).

123

S240

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1 Histological staining of native and processed menisci. a Hematoxylin-eosin staining, B Safranin O Fast Green staining

Fig. 2 Scanning electron microscopy of processed menisci

Ultimate tensile strengths of native and processed menisci were similar. Viral clearance studies showed that each viral inactivation step led to a viral load reduction compliant with the reduction factor specified in the European guidelines for viral inactivation processes.

123

Conclusions: Safety of the scaffold is conferred by decellularization and viroinactivation of the menisci while preserving structure and mechanical resistance of the tissue. Freeze-drying and gamma radiation make it a ready-to-use product, with different sizes for partial

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 and total meniscal replacement. Efficacy of this scaffold for meniscal repair will be tested in a phase II clinical trial.

P24-440 Evaluation of two virtual reality arthroscopic simulators G.J.M. Tuijthof1, P. Visser2, I. Sierevelt2, C.N. van Dijk2, G.M.M.J. Kerkhoffs2 1 Academic Medical Center/Delft University of Technology, Department of Orthopaedic Surgery/Department of Biomechanical Engine, Amsterdam/Delft, The Netherlands, 2Academic Medical Center, Department of Orthopaedic Surgery, Amsterdam, The Netherlands Objectives: Some commercial simulators are available for training basic arthroscopic skills. However, it is unclear if these simulators allow training for their intended purposes and whether the perception of usefulness relates to level of experience. We addressed the following questions: (1) Do commercial simulators have construct (times to perform tasks) and face validity (realism), and (2) is the perception of usefulness (educational value and user friendliness) related to level of experience? Methods: Two commercially available virtual reality simulators were evaluated: TolTech Knee Arthroscopy (Simulator A) and GMV ArthroVR (Simulator B). Eleven and nine novices (no arthroscopies) were recruited, four and four intermediates (1–59 arthroscopies), and seven and nine experts ([60 arthroscopies) to test the devices. To assess construct validity, we recorded the median time per experience group for each of five repetitions of one identical navigation task. To assess face validity, we used a questionnaire to judge up to three simulator characteristic tasks; the questionnaire asked about the realism, perception of educational value, and perception of userfriendliness. Questions were answered using a 10-point numerical rating scale, where a value of 7 or greater was considered sufficient. Results: We observed partial construct validity for Simulators A and B (MannWhitneyU, p \ 0.05). For Simulator A, the first repetition was significantly slower for novices (median 447 s.; range, 181–600 s) compared to both experts (median 125 s.; range 68–245 s.) and intermediates (median 129 s.; range 60–311 s.). For Simulator B, the second and third repetitions were significantly slower for novices (median 129 s. and 90 s.) compared to experts (median 58 s. and 53. sec.). Face validity was considered satisfactory for both simulators in simulating the outer appearance and human intraarticular joint with a mean of 7.3 (standard deviation (SD), 1.4) and 6.4 (SD, 1.4) for Simulator A and 8.4 (SD, 0.6) and 6.1 (SD, 0.9) for Simulator B, respectively. Face validity of the simulated instruments was barely satisfactory with a 4.9 (SD, 1.5) for Simulator A and 5.7 (SD, 1.2) for Simulator B. Simulators A and B had equal educational value according to the participants. User-friendliness was judged better for Simulator B (8.3 [SD, 1.0]) compared to Simulator A (6.5 [SD, 1.3]), although both were graded satisfactory. The perception of usefulness did not differ with level of experience. Conclusions: Our observations suggest that training on either simulator is a reasonable preparation for real-life arthroscopy, although there is room for improvement for both simulators.

P24-545 Clinical mid-term outcome after partial meniscus substitution using the collagen meniscal implant - good or bad? M.T. Hirschmann1, L. Keller2, R. Berbig2, U. Lu¨thi2, L. Schenk1, M. Arnold1

S241 1

Kantonsspital Bruderholz, Klinik f. Orthop. Chirurgie & Traumatologie, Bruderholz, Switzerland, 2SportClinic, Zu¨rich, Switzerland Objectives: Meniscal substitution using either collagen meniscus or polyurethane meniscus is increasingly performed in patients after subtotal meniscectomy. The aim of this study was to evaluate the clinical outcome of patients who underwent collagen medial or lateral meniscus graft implantation at minimum follow-up of 1 year. Methods: 67 patients (male:female = 47:20, mean age 36 ± 10 years) underwent arthroscopic implantation of the collagen meniscus device (CMI) after subtotal medial (n = 55) or lateral meniscectomy (n = 12) due to persistent compartment pain. The lesions were in the anterior horn (n = 29) and/or corpus (n = 53) and/or posterior horn (n = 54) of the meniscus. 53 patients underwent an additional procedure such as anterior cruciate ligament reconstruction (n = 44). The patients were seen for clinical follow-up at minimum 1 year after surgery. The clinical follow-up consisted of IKDC score, Tegner score preinjury, preoperatively and at follow-up, Lysholm score and visual analogue scale for pain and satisfaction (0 best–10 worst) The follow-up rate was 90%. Implant failure was defined as infection due to the implant or mechanical failure of the device. Results: 19 patients (29%) showed a normal total IKDC score (A), 35 were nearly normal (B), 5 abnormal (C) and 1 patient was severely abnormal (D). The median Tegner score preinjury was 7 (range 2–10), it decreased preoperatively to median 3 (range 0–8). At followup the median Tegner score was 6 (range 2–10). At latest follow-up maximum 7 years after surgery it was 6 (range 3–10). The mean Lysholm score before surgery was 68 ± 20 and 93 ± 9 at follow-up. The mean VAS satisfaction preoperatively and at follow-up was 4.0 ± 0.5 and 1.9 ± 1.0. The mean VAS pain preoperatively and at follow-up was 4.4 ± 3.1 and 2.0 ± 1.0. The Tegner score and satisfaction significantly improved further up to 7 years after surgery (p \ 0.05). There was not significant difference in outcomes between the medial and lateral CMI. Conclusions: Meniscal substitution with the collagen meniscal implant showed excellent clinical 1 year results. It further provides significant pain relief and functional improvement throughout all scores at a minimum of 1 year follow-up. No difference between the medial and lateral CMI was observed.

P24-639 A 12-week exercise therapy program is feasible and improves function in middle-aged patients with degenerative meniscus tears S. Stensrud1, E.M. Roos2, M.A. Risberg3 1 University of Southern Denmark, Research Unit for Musculoskeletal Function and Physiotherapy, Norwegian Research Centre for Activ Rehabilitation, Oslo, Norway, 2University of Southern Denmark, Faculty of Health Sciences, Research Unit for Musculoskeletal Function and Physiotherapy, Odense, Denmark, 3Oslo University Hospital, Department of Orthopaedics, Norwegian Research Centre for Active Rehabilitation, Hjelp24 NIMI, Oslo, Norway Objectives: Patients with degenerative meniscus tears are at great risk of knee osteoarthritis (OA) and meniscectomy is the standard treatment. However, in patients with established knee OA exercise therapy is the primary treatment, and there is strong evidence that exercise improves physical function and reduces knee pain. Considering that function often is reduced in patient with degenerative meniscus tears and the suggestion that muscle function play a role in OA development, exercise may be beneficial. No study has yet described in details type of exercises, progression, tolerance and potential benefit from an exercise therapy program in these patients. The objective was thus to examine the feasibility of an exercise therapy program in middle-aged patients with degenerative meniscus tears.

123

S242 Methods: 20 patients with MRI-verified degenerative meniscus tears, included in an ongoing randomised controlled trial, underwent a 12-week progressive exercise therapy program. Outcome measures were the Knee injury and Osteoarthritis Outcome Score (KOOS), a 7-point global rating of change scale, isokinetic knee muscle strength test and three lower extremity performance tests. Self-reported pain was reported during and after each training session by a visual analog scale (VAS) from 0 to10. Feasibility was determined as improved patient reported outcomes, acceptable self-reported pain during and after the training sessions (B5 on the VAS), decreased or unchanged pain from the beginning to the end of the exercise therapy program, improved knee muscle strength, improved lower extremity performance and few adverse events. Results: There were significant and clinically important changes ([ 10) on all five subscales of KOOS, with mean changes between 13 and 27. 14/20 patients rated themselves as ‘‘a lot better’’ on the 7-point scale, 5/20 patients rated ‘‘better’’ and only one patient rated ‘‘unchanged’’. Acceptable pain (median 1) was reported during and after the sessions. All patients improved in peak torque knee extensor strength (5–74%) and 14/20 patients improved more than 15%, which is considered a clinically relevant change. 16/20 patients improved in lower extremity performance (0–133%), and 12/20 patients improved more than the minimal detectable change. Conclusions: A progressive exercise therapy program is feasible and should be considered in middle-aged patients with degenerative meniscus tears. All patients experienced self-perceived clinical improvements, and the majority of the patients found themselves ‘‘a lot better’’ after 12 weeks of exercise.

P24-660 Joint stressing forces when performing knee arthroscopies G. Tuijthof1, P.-B. Wulms2, I. Sierevelt3, M. Schafroth3, L. Blankevoort3, G.M.M.J. Kerkhoffs3 1 Academic Medical Center/Delft University of Technology, Department of Orthopaedic Surgery/Department of Biomechanical Engine, Amsterdam/Delft, The Netherlands, 2Delft University of Technology, Department of Biomechanical Engineering, Delft, The Netherlands, 3Academic Medical Center, Department of Orthopaedic Surgery, Amsterdam, The Netherlands Objectives: During arthroscopy of the knee joint, the available joint space is increased by stressing the lower leg. This position enables the surgeon to perform triangulation by using one hand to move the arthroscope and the other hand to perform treatment with an instrument. Triangulation is considered as one of the five important arthroscopic skills a resident should possess before training is started in the operating room. The proper amount of loading is difficult to learn by just watching the supervising surgeon. Simulators can be of help as the residents practice stressing by actual handling. So far, no feedback on stressing forces during simulation training is offered and no safe loading levels are available. The aim was to measure quantitative magnitudes and variation of joint stressing forces in vivo during arthroscopic knee surgery performed by experienced surgeons. Methods: Nineteen patients were included whose lower leg was stressed medially and laterally by four different surgeons which confirmed proper stressing using the arthroscopic view. Patients’ lower leg lengths were measured and their knee stability was documented. The stressing forces were recorded with a force sensor that was attached to a belt, which was worn by the surgeon underneath his/ her sterile clothes. A separate camera was stationed to qualitatively determine the flexion–extension angle during stressing. A data acquisition system with special software VISIONDAQ (version 1.2, MTO, AMC) recorded the force, the arthroscopic camera images and the additional camera images simultaneously. Using the arthroscopic

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 view, the periods of joint stressing were determined for which the maximum force was determined. Results: All knees were stable and the mean lower leg length was 0.42 m (standard deviation (SD) 0.02 m). The force data showed a normal distribution, and stressing was always performed with the knee slightly flexed. The mean force to open the medial knee compartment was 55 N (SD 16 N), and to open the lateral compartment was 68 N (SD 37 N). The maximum measured force was 163 N used to open the lateral compartment. Conclusions: The variation of the stressing forces is relatively high (standard deviation is 30% of the mean for medial opening and 54% for lateral opening), especially for lateral opening. Based on the results, an indicative upper level of stressing force can be suggested using the mean added by two times the standard deviation. This gives 87 N for medial opening and 142 N for lateral opening. As the joint stressing level is dependent on the patient, it is recommended to offer training modules where the complete range of stressing levels can be experienced.

P24-663 Global Rating Scales to objectify arthroscopic skills in the operating room G. Tuijthof1, B. van Ooij2, M. Schafroth2, C.N. van Dijk2, J. Dragoo3, G.M.M.J. Kerkhoffs2 1 Academic Medical Center/Delft University of Technology, Department of Orthopaedic Surgery/Department of Biomechanical Engine, Amsterdam/Delft, Netherlands, 2Academic Medical Center, Department of Orthopaedic Surgery, Amsterdam, Netherlands, 3Stanford University, Orthopedic Surgery, Redwood City, United States Objectives: As the number of operations is being reduced in academic teaching hospitals, it is worthwhile to optimize the learning effect per operation, especially for arthroscopic techniques as they require adequate psychomotor skills. Recently, new teaching aids are available for structured feedback during training in the operating room: Global Rating Scales (GRS). A GRS is a questionnaire that uses 1–5point Likert-type scale to assess overall proficiency in surgical skills. Two GRS have been proposed for application in arthroscopy: Orthopaedic Competence Assessment Project (GRS I) and Arthroscopy Global Rating Scale (GRS II). The aim is to investigate which of these two GRS is most suitable to monitor the learning curve in arthroscopic surgery. Methods: After each arthroscopy, both the resident and supervising surgeon filled out each of the two GRS independently from each other. Type of operation, year of residency and number of previously performed arthroscopies were documented. The outcomes of both GRS were compared to the experience of the residents using a linear regression analysis (p \ 0.05). The self-assessment of the residents was compared with the judgment of the supervising surgeons using a paired t test (p \ 0.05) to determine systematic differences and an intraclass correlation coefficient (ICC) to determine level of agreement. Finally, the time to fill out each GRS was measured to determine their suitability for application in clinical practice. Results: 109 knee arthroscopies were included. The GRS were filled out by 3 supervising surgeons and 17 residents. A positive and significant correlation (p \ 0.05) was found between the number of previously performed knee arthroscopies and the total GRS-scores (GRS I: R = 0.68, GRS II: R = 0.67). A stronger correlation was found between the year of residency and the total GRS-scores (GRS I: R = 0.75, GRS II: R = 0.76). No systematic difference was found between the self-assessment of the residents, which showed a mean normalized score of 73.5 (standard deviation (SD) 11.2) for GRS I and 73.4 (SD 12.6) for GRS II, and the scores of the surgeons, which were 73.7 (SD 14.0) for GRS I and 74.8 (SD 15.0) for GRS II, respectively. The ICC between the residents and supervising surgeons

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 was moderate for both scales (0.58 for GRS I and 0.66 for GRSII). The mean fill out time for the surgeons was 44 s (SD 27) for GRS I and 46 (SD 23) for GRS II. The mean fill out time for the residents was 39 s (SD 23) for GRS I and 41 (SD 20) for GRS II. Conclusions: The results show that both GRS are equally suitable and practical for clinical practice, which is due to the overlap in content of the GRS. Both demonstrate a significant correlation between prior experience of the resident and judgment by the supervising surgeons. This enables monitoring of the learning curve of arthroscopic skills. Based on these results, it is recommended to encourage supervising surgeons to use either GRS on a regular basis.

P24-670 Arthroscopic Lateral Collagen Meniscus Implant at 2 year minimum follow-up G.M. Marcheggiani Muccioli1, S. Zaffagnini1, P. Bulgheroni2, E. Bulgheroni2, A. Grassi1, M. Marcacci1 1 Istituto Ortopedico Rizzoli, University of Bologna, Biomechanics Laboratory, Bologna, Italy, 2Ospedale di Varese, Varese, Italy Objectives: The purpose of this paper is to provide results of the implantation of a bioresorbable type I collagen scaffold projected to restore the meniscal structure in patients with irreparable lateral meniscus deficiencies at 2 year minimum follow-up (FU). Methods: 24 non-consecutive patients (20 males and 4 females; mean age 35.4 ± 10.76 years, range 16–53) with irreparable lateral meniscus tears or meniscus loss requiring surgical treatment were prospectively enrolled in this Italian multicenter study (2 centers). Patients were evaluated at baseline, at 6 month and 2 at year minimum follow-up (mean 26.7 ± 7.1 month; range 24–32) after Lateral CMI implantation, with a 100-point Visual Analogue Scale (VAS) for pain, objective International Knee Documentation Committee (IKDC), Lysholm score, Tegner activity and EQ-5D survey. MRI evaluation was also performed at baseline and at 3 years mean follow-up with both Yulish and Genovese score. Results: FU evaluation showed an improving of all the clinical scores: VAS score significantly decreased from 5.5 ± 2.9 to 1.8 ± 1.8 at 6 month FU (p \ 0.0001) and remained almost stable at 2 year minimum follow-up (1.9 ± 2.5; p = 0.8259). Objective IKDC significantly improved from 6 A, 14 B, 4 C at baseline to 20 A, 3 B, 1 D (p \ 0.0001) at 6 month FU and did not changed at final FU. Lyhsolm score significantly improved from 64.0 ± 16.2 pre-op to 89.9 ± 11.4 (p \ 0.0001) at 6 month FU and showed a trend to improve at the final follow-up (92.7 ± 13.8; p = 0.1218). Tegner activity Level significantly improved from 3 (2–4) at baseline to 5 (4–7) final follow-up (p = 0.0062). The EQ-5D score improved significantly from 0.579 ± 0.28 to 0.892 ± 0.14 at the final FU (p \ 0.0001). The MRI evaluation showed a trend of the Yulish score to decrease both for lateral tibial plateau (from 1.5; 1.0–2.0 to 1.0; 1.0–0.5; p = 0.8203) and for lateral femoral condyle (from 2.0; 1.0–2.0 to 1.5; 1.0–2.0; p = 0.9697). The Genovese score at the final FU was 1.0 (1.0–1.0) for morphology and 1.0 (1.0–2.0) for intensity. No correlation was found between MRI and clinical scores. No reoperation nor complication were recorded during the follow-up. Only one patient showed unsatisfactory result. Conclusions: Based on available results, with 2 year minimum FU, 96% of the patients benefited from the Lateral CMI implantation and regained activity and mobility. Longer-term data are being collected to determine the extent and duration of the benefits as observed for Medial CMI [1]. Reference: [1] Zaffagnini S, et al. Am J Sports Med. 2011;39 (5):977–85.

S243 P24-762 A novel polycarbonate-urethane meniscal implant: a functional evaluation of sizing J. Elsner1, V. Condello2, E. Hershman3, R. Arbel4, A. Shterling1, E. Linder-Ganz1 1 Active Implants Corporation, R&D Center, Netanya, Israel, 2Sacro Cuore Don Calabria Hospital, Sports Medicine and Arthroscopic Surgery, Negrar (Verona), Italy, 3Lenox Hill Hospital, Department of Orthopaedic Surgery, New York, United States, 4Tel Aviv Souraski Medical Center, Sports Medicine and Arthroscopic Surgery, Tel Aviv, Israel Objectives: The medial meniscus plays an important role in distributing knee joint contact forces and limiting the contact pressure developed in the articular cartilage. Accurate matching of geometry and size are critical for the success of any meniscal implant aimed at restoring the contact pressure distribution and delaying joint degeneration. The goal of this study was, therefore, to study the ability of a novel freefloating elastic polycarbonate-urethane (PCU) meniscal implant to restore the pressure distribution on the articular cartilage surfaces, and to predict the necessary number of sizes needed to cover the candidate population, based on experimental and computational studies. Methods: The mean geometrical parameters of the natural medial meniscus were measured in 130 MRI scans of a mixed male/female population, which were then used to create the general shape of the synthetic PCU meniscus. Computational finite elements simulations and laboratory compression tests in cadaver knees were used to evaluate the ability of the PCU meniscus implant to distribute joint compressive loads over the tibial plateau surface. Additional evaluation of the effect of sizing on mobility and function was done on cadaver knees using fluoroscopy (Fig. 1). Results: It was found that the synthetic meniscus performs equally well in distributing joint loads in a 5% range around the ‘true’ size. Additional kinetic implant evaluation using fluoroscopy, demonstrated good functionality in terms of maintaining contact with the cartilage and smoothness of motion due to the self-adjustment ability of the implant. These findings, together with a statistical analysis of the knee sizes in the general population, imply that 9 implant sizes can accommodate most patients. Conclusions: The proposed implant was found to be able to restore the pressure distribution of the natural meniscus. Current findings suggest that for this type of implant, a relatively lenient safety-range exists for the choice of implant size by the surgeon.

Fig. 1 Fluoroscopic image of the meniscal implant

123

S244 P24-848 Implant of a polyurethane scaffold for treatment of partial meniscal lesion. Early results P. Bulgheroni1, E. Bulgheroni1, P. Cherubino1 1 Ospedale di Varese, Ortopedia e Traumatologia, Varese, Italy Objectives: In the last years scaffolds were proposed to treat meniscal defect. Since 2000 it has been marketed a collagen implant to induce regeneration of a meniscus like tissue in case of a partial meniscal defect. The results reported in the medium-long term are satisfactory, although improvable. Recently has been placed on the market a novel biodegradable polyurethane scaffold for meniscal regeneration, with different structural and mechanical characteristic and different reabsorption time. The aim of this study is to evaluate the early results and the safety of the implant. Methods: In 2009 we started implanting the polyurethane scaffold in case of knee pain after partial meniscectomy or irreparable meniscal tears. Twenty-one patients (22 implants) were operated on, 18 for medial meniscus injuries and 4 for lateral ones. Fourteen patients complained of knee pain after previous meniscectomy. In 8 cases was performed an osteotomy for associated knee malalignment and in 9 patients the ACL was reconstructed. All patients were clinically studied with Lysholm, Tegner and VAS scale and with MRI at pre-op evaluation, 6 months, 1 and 2 years after surgery. Four patients were undergone arthroscopic second look and biopsy of the implant at different timing from the surgery. Results: Eight patients have reached a 2 years follow up, 10 1 year and 3, 6 months with a significant clinical improvement. At the MRI evaluation the implant signal was evident in all the cases. No cartilage changes were observed. At the arthroscopic relook the implant was well integrated with the surrounding tissues, with unchanged size and morphology; it was stable to the probing. The histological evaluation of the biopsies showed a new tissue growth inside the scaffold. Adverse reactions to the implant were not observed. Conclusions: The ease of implantation and the early clinical results are promising, it will take a long time to evaluate the effective nontoxic degradation products of the implant and its real chondroprotective effect.

P24-858 Clinical results of arthroscopic salvage repair incorporating the base of PCL as a temporary post for medial meniscus posterior horn root tears S.E. Park1, S.K. Kim1 1 Dongguk University International Hospital, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: To introduce the clinical results and technical aspects of arthroscopic repairs including posterior cruciate ligament as posts for the treatment of medial meniscus root tears in 34 patients between the ages of 28 and 77 where treatments were indicated as total meniscectomy. Methods: Between January 2008 and April 2009, we studied arthroscopic repairs on the knees of 34 patients with symptomatic complex lateral meniscus tears that were all treated by internal repair techniques. The mean age at the time of the surgical procedure was 55.96 years and the mean follow-up period was 13.06 months (range of 11–16 months). Clinical results were evaluated using Lysholm knee scores pre-operatively and at the final follow-up. 3 patients were not followed due to other reason sixteen patients.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: 21 patients were able to return to their previous life activities, and no reoperation was required after an average follow-up period of 43 months. Mean Lysholm knee scores improved from 66.61 (range of 39–79) pre-operatively to 78.81 (range of 48–95) at the final follow-up (p \ .001). Ten patients who undergo the second look arthroscopy were well healed. Six patients were not healed, and four patients were not showed improved clinical results. Conclusions: Arthroscopic repair using posterior cruciate ligament as a post is effective for treating medial meniscus root tears.

P24-897 All-inside meniscal repair: a prospective, randomized clinical study with 4- to 7-year results S. Ja¨rvela¨1, R. Sihvonen2, T. Ja¨rvela¨3 1 Tampere University Hospital (TAYS), Orthopaedic Department, Tampere, Finland, 2Hatanpa¨a¨ Hospital, Arthroscopic Centre, Tampere, Finland, 3Hospital Mehila¨inen, Orthopaedic Department, Tampere, Finland Objectives: All-inside meniscal repairs have gained popularity in past few years, although outside-in and inside-out suture techniques have been standard techniques in repairing the meniscus in the past. However, open surgery for the meniscal repair is more time-consuming and carries some risks for the nerve and other soft tissue damage outside of the knee joint. The purpose of this study is to compare 2 all-inside meniscal repair devices in a prospective, randomized clinical trial with 4–7 years follow-up. Methods: Forty-two patients were prospectively randomized to have all-inside meniscal repair either by using bioabsorbable meniscal screws or arrows for the fixation (21 patients in each group). The mean age of the patients was 31 years (SD 9). There were 29 males and 13 females. The average time from injury to meniscal repair was 6 weeks in both groups (range, 1–26 weeks). The evaluation methods were clinical examination, Lysholm score, the International Knee Documentation Committee (IKDC) function and final score, and radiographic evaluation. The average follow-up time was 5.5 years (range, 4.0–7.0 years). All the operations were done by 2 experienced orthopedic surgeons. The follow-up was performed by one independent and blinded examiner. Results: There were no differences between the study groups preoperatively. 33 patients (79%) were available for the follow-up. 14 patients had clinical failure of the repair during the follow-up (11 patients in 6 months after the repair, and 3 patients more than 2 years after the repair) leading to partial meniscal resection (6 patients in the screw-group, 8 patients in the arrow-group, NS). Of the failures, 11 were seen on medial meniscus, and 3 on lateral meniscus. At the follow-up, the average Lysholm score was 81 (SD 15) in the screwgroup, and 89 (SD 14) in the arrow-group (NS). In the IKDC function and final scores, no significant differences were found between the groups either. However, both groups did significantly better at the follow-up than preoperatively (p \ 0.001). Also, no significant differences were found in degenerative changes seen on radiographs between the groups. However, the patients with partial meniscal resection, because of the failure of the meniscal repair, had significantly more degenerative changes in the medial femorotibial compartment of the knee than the patients, who did not have a failure of the meniscal repair (p = 0.049). Conclusions: All-inside meniscal repair with bioabsorbable meniscal screws and arrows resulted in similar clinical outcome at 4–7 years follow-up. However, the failure of the meniscal repair with a need for partial resection of the meniscus seemed to resulted in significantly more degenerative changes of the knee joint compared to healed meniscal repair.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P24-919 Effect of tourniquet use on rehabilitation and muscle damage after arthroscopic partial meniscectomy. A prospective randomized study M. Hantes1, A. Tsarouhas1, G. Basdekis1, G. Tsougias1, I. Antoniou2, K. Malizos3 1 University Hospital of Larisa, Orthopaedics, Larisa, Greece, 2 University Hospital of Larisa, Larisa, Greece, 3University Hospital of Larissa, Orthopaedic Surgery Department, School of Medicine, Larissa, Greece Objectives: Pneumatic tourniquets are used in arthroscopic knee surgery to improve surgical field visibility and operative time. However, their effect on complication rates and postoperative rehabilitation remains controversial. The purpose of the prospective randomized trial was to examine the effect of tourniquet use on the rehabilitation rate and amount of muscle damage after arthroscopic partial meniscectomy. Methods: Fifty consecutive patients who undergo arthroscopic partial meniscectomy were randomly allocated to the use of a pneumatic tourniquet (group A, n = 25) and without (group B, n = 25). Patients with concomitant ligamentous deficiency or chondral lesions requiring further intervention were excluded. Blood Creatine Kinase (CK) levels were measured pre-surgery and on days 1, 8 and 15 postoperatively to evaluaye muscle damage. Pain intensity was recorded using the Visual Analogue Scale (VAS). Postoperatively, the patients were evaluated on a weekly basis for range of motion, pain intensity number of days required to progress to full weight bearing without crutches and return to sport activities (jogging). Results: The two groups did not differ in terms of age, male/female ratio, body mass index, side affected, and preoperative IKDC, KOOS, Tegner, Lysholm and EQ-5D scores (minimum p = 0.15). Operative time was not significant between groups (mean 27.5 and 31.2 for groups A and B respectively, p = 0.83). No complications were met either intra- or post-operatively. Postoperatively, all the patients had regained full range of motion and experienced no pain by day 15. Neither of the two variables was significant between groups (minimum p = 0.22). Blood CK values were also not significant between groups during consecutive measurements (p = 0.3). The patients discontinued crutches after 13.4 and 12.9 days in groups A and B respectively (p = 0.9). Tourniquet time did not correlate with postoperative range of motion, pain intensity or CK levels (minimum p = 0.14). Similarly, time to return to jogging did no differ between groups (33.2 days for group A vs. 34.6 days for group B, p = 0.34). Conclusions: Tourniquet use does not affect postoperative pain, return to daily and sport activities or muscle damage after arthroscopic partial meniscectomy. Routine use of tourniquet in common knee arthroscopic procedures is recommended.

P24-985 A new arthroscopic technique using intraarticular portal for lateral parameniscal cyst decompression U. Haklar1, E. Ayhan2, E. Sarıoglu3 1 Acibadem Kadikoy Hospital, Istanbul, Turkey, 2Sarıyer Ismail Akgu¨n Hospital, Istanbul, Turkey, 3Medical Park Go¨ztepe Hospital, Istanbul, Turkey Objectives: We describe a new and easy arthroscopic technique for lateral parameniscal cyst decompression without using meniscal tract and report the short to midterm outcome of this technique in 19 patients. Methods: Between September 2005 and April 2011, 19 knee arthroscopies were performed for lateral parameniscal cyst treatment. All patients underwent a complete physical examination and MRI of

S245 the knee prior to surgery. The patients filled Tegner Lysholm knee scoring scale both before and after the surgery. After arthroscopic diagnosis of lateral parameniscal cyst, partial lateral menisectomy is performed. Instead of the universal method for lateral parameniscal cyst treatment, we use superomedial (SM), anterolateral (AL) and intraarticular (IA) portals to decompress the cyst. SM portal is prepared at the most proximal part of suprapatellar pouch, just medial to quadriceps tendon. An IA portal is created by shaver on the anterior synovial wall of the cyst from the AL portal. Decompression is performed through the IA portal under the view of the scope in SM portal. Patients were recalled for follow-up, and clinical statuses were classed into 4 categories as reported by Dorfmann: excellent, good, acceptable, poor. Also, follow-up MRI was assessed by both the radiologist and us in regards to recurrence of the cyst. Results: The mean age of the patients was 42.4 and mean follow up was 34 months. The clinical statuses were excellent in 15 patients and good in 4 patients. No recurrence of cyst was seen on control MRI of the patients. The average of preoperative Tegner Lysholm knee score improved from 57 to 93 postoperatively. Conclusions: In this technique, under the excellent view of the cystic lesion (SM portal) and with the nearby portal (AL portal) preference, the handling of the instrument is easier adjacent to the cyst. Also, decompression is performed through the IA portal to preserve healthy meniscal tissue. In conclusion, arthroscopic decompression by means of SM, AL, and IA portals is a new, easy, safe, and effective technique for lateral parameniscal cysts treatment.

P24-1036 Incidence of bilateral discoid lateral meniscus in Asian population: An arthroscopic assessment of contra-lateral knees J.H. Bae1, H.C. Lim2, D.H. Hwang3, J.K. Song3, J.S. Byun4, K.W. Nha3 1 Korea University Ansan Hospital, Department of Orthopedic Surgery, Ansan, Republic of Korea, 2Korea University Guro Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea, 3Inje University, Ilsanpaik Hospital, Department of Orthopaedic Surgery, Ilsan, Republic of Korea, 4Korea University Ansan Hospital, Department of Orthopaedic Surgery, Ansan, Republic of Korea Objectives: To investigate the incidence of bilateral discoid lateral meniscus (DLM) and to evaluate the arthroscopic features of lateral meniscus in asymptomatic contra-lateral knees in Asian population who presented with a symptomatic DLM. Methods: Consecutive 52 patients who underwent arthroscopic procedures for symptomatic DLM (31 complete and 21 incomplete) were prospectively enrolled in this study and consented to the examination of contralateral knees at the time of arthroscopy. Types of DLM and of meniscus tears were assessed using arthroscopic findings. Preoperative and postoperative functional outcomes were evaluated using Lysholm and Tegner activity scores. Results: Arthroscopic examinations revealed that 21 complete, 19 incomplete DLM, 11 normal and 1 ring type lateral menisci in contralateral knees. The incidence of bilateral DLM in our study population was 79% (41/52 contra-lateral knees). Furthermore, 65% of patients (34 pairs of knees) had the same DLM types. In addition, three pairs of knees with complete DLMs had menisci of different thicknesses. DLM tears were observed in two contralateral knees (1 radial, 1 longitudinal) and were treated by partial central meniscectomy. Conclusions: This study provides evidence of the high prevalence of bilateral discoid lateral meniscus.

123

S246 P24-1043 Can arthroscopic medial meniscus root repair using a suture anchor restore subluxation and hoop tension of the medial meniscus? C.-K. Lee1, D.-M. Chung1, K.-A. Jung1, S.-C. Lee1 1 Himchan Hospital, Department of Orthopaedics, Seoul, Republic of Korea Objectives: It has been controversial over the efficacy for the arthroscopic repair of medial meniscus root tear (MRT). This study was to evaluate the structural integrity and reduction status of the repaired medial meniscus by MRI after arthroscopic meniscus root repair and to identify their relevance to clinical and radiologic outcomes. Methods: From September 2009 to August 2010, 46 medial MRTs in 44 patients were repaired with one suture anchor using trans-septal technique by one senior author. 26 patients (26 knees) who had preoperative arthrosis with less than Kellgren-Lawrence grade 1 or 2, a minimum of 1 year follow-up, and postoperative MRI evaluation (mean 7.1 months, 6–8 months) were included in this study. Medial meniscal extrusion (MME) and healing state of the meniscus on MRI were evaluated. Clinical outcomes using the Lysholm and International Knee Documentation Committee (IKDC) knee score were evaluated preoperatively and at 3, 6, 12 months and final follow-up. Radiographic evaluation was performed, both preoperatively and at final follow-up. Results: The mean follow-up was 14.2 months (12–18 months). Postoperative MRI showed 3 complete healing (11.5%), 20 fibrous healing (76.9%), and 3 retears (11.5%) of the meniscus. MME on MRI significantly increased from 2.31 to 2.87 mm (p \ .05). An improvement of MME was observed in only 8 patients (30.8%). However, the mean Lysholm scores improved from 62.5 preoperatively to 84.2 at final follow-up (p \ .05), and mean IKDC scores also significantly increased from 32.39 to 61.2 (p \ .05). Of the 10 (38%) who had radiographic progression of arthrosis, one patient underwent total knee arthroplasty at the 13 month follow-up. Conclusions: Arthroscopic medial meniscus root repair using a suture anchor failed to restore subluxation and hoop tension of the medial meniscus, we supposed, which caused radiographic progression of arthrosis in this study. However, this technique may be good alternative treatment option for the medial MRT to relieve meniscal symptoms, because it showed high healing rate of repaired medial meniscus (88.5%) and good short-term results.

P24-1097 All-arthroscopic technique of biological meniscal tear therapy with collagen membrane- early results T. Piontek1, K. Ciemniewska-Gorzela1, A. Szulc2, J. Naczk3, R.P. Jakob4 1 Rehasport Clinic, Poznan, University of Medical Sciences, Clinic of Peadiatric Orthopedic Surgery, Poznan, Poland, 2University of Medical Sciences, Clinic of Peadiatric Orthopedic Surgery, Poznan, Poland, 3Rehasport Clinic Poznan, Poznan, Poland, 4Spital Tafers, Orthopaedic Department, Tafers, Swaziland Objectives: Main method for treatment combine meniscal tears in the white layer is menisectomy. An analysis of treatment results, carried out at several centres by numerous study groups, showed development of early degenerative changes in the knees of that treated patients. Aim of the study is present all arthroscopic technique to treat meniscal tears by suturing them and wrapping them in Chondro-Gide collagen membrane, followed by liquid bone-marrow collected from the tibial proximal epiphysis, into the area of lesion and 1–2 years results of treatment.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: We operated 25 patients with medial and lateral meniscal white layer tears all arthroscopic. We control patients before, and 1 and 2 years after operation by MRI scans, IKDC2000, Lysholm scale and Baret criteria of treatment. Results: We observe in MRI scans good rebuilding of the meniscal cartilage after 1 and 2 years. In Lysholm we noted improvement from 70 points before to 92 after 1 year 91 after 2 years and IKDC2000 from 50 to 83 after 1 year and 84 after 2 years. In Baret criteria we found improvement from 1 to 4 points in 90% of patients after 1 and 2 years. We observed 2 complications. One patient have been operated after 3 month because of arthofibrosis and second after 4 month because of inflammation. Conclusions: We present good alternative of treatment combine meniscal tears and promising results of using all arthroscopic meniscal membrane suturing.

P24-1117 Biological reconstruction of the failing knee: experience of meniscus allograft transplantation with osteotomy, ligament reconstruction and articular cartilage repair T. Spalding1, A. Getgood1, S. Spencer1, J. Bird1, P. Thompson1 1 University Hospitals Coventry and Warwickshire, Coventry, United Kingdom Objectives: Meniscus allograft transplantation (MAT) is an established treatment option for the painful post meniscectomy knee. The presence of osteoarthritis has been described as a relative contraindication for this procedure; however, joint replacement surgery is an unattractive option for the young active person with a failing knee. The purpose of this study was to report our experience of MAT in combination with a variety of other procedures in an attempt to reconstruct the failing knee including osteotomy, anterior cruciate ligament reconstruction, and articular cartilage repair. Methods: This is a case series of patients presenting with International Cartilage Repair Society grade IV degenerative change in one or more compartments associated with meniscus loss presenting to a single surgeon over an 8-year period. All patients were treated with meniscus allograft transplantation in combination with one or more of osteotomy, ligament reconstruction and articular cartilage surgery. A treatment algorithm was developed, aiming to address biomechanical abnormalities followed by biological treatments, i.e. osteotomy ± ACL [ MAT [ cartilage repair. MAT was performed using a novel arthroscopic technique. Osteotomy addressed the point of maximal deformity using opening or closing wedge techniques. ACL reconstruction incorporated anatomic hamstring graft insertion. Articular cartilage repair included a variety of techniques depending upon the nature of the chondral defect. Outcome data was collected prospectively at pre-op then annually, which included IKDC, Lysholm, Tegner and KOOS scores. At 1 year following surgery, further assessment of the meniscus transplant was made by way of MRI and/or second look arthroscopy. Results: Over an 8-year period, 51 meniscus allograft transplants were performed in our institution. Of those, 24 patients had grade IV ICRS degenerative change and had combination ‘salvage surgery’. The mean age at surgery was 37 years (19–49). 19 have minimum 12 month follow up with 17 (89%) of those had significantly improved knee scores at latest follow-up compared to baseline. MRI scan showed normal appearances of the graft in 17, with minimal graft extrusion. Second look arthroscopy was performed in 9 showing encouraging recovery of the chondral surface. Eight showed good peripheral integration of the graft, with only 1 failure requiring removal (15 months). One patient has been converted to arthroplasty at 4 years. Three required arthroscopic arthrolysis for adhesions.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: Salvage of the failing knee can be successful when arthroscopic meniscus allograft transplantation is combined with other procedures which aim to address biomechanical and biological deficiencies. A treatment algorithm has been developed in an attempt to better understand and treat these problem cases.

P24-1211 Arthroscopic meniscal scaffold implantation: early clinical results at 20 months of follow-up C. Zorzi1, V. Madonna1, V. Condello2, F. Cortese1, R. Giovarruscio1 1 Sacro Cuore Don Calabria Hospital, Orthopaedic, Negrar, Italy, 2 Sacro Cuore Don Calabria Hospital, Sports Medicine and Arthroscopic Surgery, Negrar (Verona), Italy Objectives: Everyday knee joint is subjected to heavy work for stability, range of motion and weight bearing. The meniscus plays a vital role in the successful performance of the knee and provides important function, including load bearing, shock absorption, stability, and joint nutrition. The treatment strategies for meniscus pathology has been evolving from resection to preservation. While many meniscus tears can be successfully repaired, not all are salvageable. Removal of all or part of the meniscus leads to degenerative changes of the articular cartilage and subsequent clinical symptoms. The goal of this study was to evaluate the ability of a tridimensional synthetic scaffold to reduce pain, to improve function and quality of life. Methods: Since 2010 we have treated 28 patients with ActifitR meniscal scaffold with average age of 40 years and 8 months (from 16 to 55 years). The mean follow-up was 11 months (from 6 to 20 months). There were 26 men and 2 woman; 4 patients received a lateral meniscus implant and 24 received a medial one. The implant of ActifitR meniscal scaffold was associated with a valgus osteotomy (HTO) in 4 cases and with supracondylar femoral varus osteotomy (HFO) in 1 case, in both with Puddu’s plate. In one patient was performed an intra-articular reconstruction of the anterior cruciate ligament (ACL) with hamstrings, fixed with TTS system on the femur and with a resorbable screw on the tibia. We have used a standard arthroscopic surgery procedure. Rehabilitation postoperative was recommended for 16–24 weeks, with patient non-weightbearing for the first 3 weeks. Partial weightbearing was permitted from week 4 onwards, with a gradual increase in loading up to 100% load at 9 weeks post-implantation. Progressive weightbearing was initiated in stages, increasing by 10 kg for week. Motion was initiated immediately after implantation, with bending up to 30 with full extension permitted in weeks 1 and 2. Flexion is progressively increased to 60 in week 3, and to 90 in weeks 4 and 5. From week 6 onwards, flexion is further increased until a full range motion is achieved. Results: Following meniscal scaffold implantation, good–excellent results are achieved in nearly 80% of cases. The IKDC showed 10 patients in group A, 13 in group B and 5 in group C. At the Lysholm score 10 had excellent results. All patients highlighted a measurable decrease of pain symptomatic, an improvement of functionality and quality of life. Implantation of meniscal scaffold combined with ACL reconstruction or correction of the limb malalignment achieved better results. Conclusions: Despite a short follow-up, implantation of ActifitR meniscal scaffold represents a reasonable treatment for symptomatic partial meniscectomized knee with early degeneration of articular surface. This device can decrease pain, increase knee function, allows pain free activies of daily living.

S247 The good results obtained represents a stimulus for future activities.

P24-1434 Hand-held ultrasound scanning diagnosis in acute knee injury B. Klos1, S. Konijnenberg1 1 ICONE, Orthopaedie, Nuenen, The Netherlands Objectives: Muscoloskeletal ultrasonography (US) is a sufficient tool to decide on surgical indication. In acute knee injury diagnosis is usually performed by X-ray and or MRI. We conducted a study using US for additional imaging after plain X-ray. Because of the low-cost and availability of US we used US for pre operative decision making. Methods: In a 2 year period (2010–2011) two orthopedic surgeons identified 55 patients with a acute knee injury with a mean 2.4 weeks after trauma. All patients had X-ray and US with a mean 2.9 weeks after injury. Patients with isolated MCL injuries, muscle injuries or tendinopathy were treated conservatively. Patients with a mechanical extension loss and or with US suggesting meniscal tears and or loose bodies were treated with arthroscopy. Results: 55 patients were included. 25 patients were treated conservatively. 24 patients had no recurrence complains after a mean 12.1 months follow-up. 1 patient underwent arthroscopy because of persistent pain. In 30 patients primary arthroscopy was performed. In 93% the US diagnosis was confirmed and in 7% a solitary anterior cruciate ligament entrapment injury was found without meniscal lesion (Table 1). Conclusions: This validation study justified further investigation which could be a level 1 study in with the surgeon is not aware of the preoperative findings. We did not include MRI findings. Previous studies compared preoperative findings of MRI and US and found high sensitivityfor meniscal lesions. New techniques using power Doppler and improved image quality are expected to produce even better results for sensitivity and specificity. The current practice using MRI preoperative imaging can be time consuming and expensive. Large cost reduction is achieved by using US imaging as an alternative tool. US is a valuable alternative tool in surgical decision making in acute knee injury. In our study the method was validated and deserves further research.

Table 1 Pre operative US findings versus arthroscopy findings in acute knee injury US findings versus arthroscopy findings

Arthroscopy positive

US positive

28

2

30

US negative

1

24*

25

29

26

55

Total

Arthroscopy negative

Total

Sensitivity 0.93 Specificity 0.96 Pos. predicted value 0.97 Neg. predicted value 0.92 * Conservatively treated Group, mean follow-up 11 months

123

S248 P24-1437 Comparison between arthroscopic meniscal allograft transplantation and polyurethane scaffold implantation: clinical outcome, reinterventions and implant intergrity after a minimum follow-up period of 2 years. A prospective clinical trial P. Verdonk1, K. Moens2, L. Willemot2, T. Tampere2, R. Verdonk3 1 Gent University Hospital, Orthopaedic Surgery, Gent, Belgium, 2 Gent University Hospital, Gent, Belgium, 3Gent University Hospital, Orthopaedic Surgery and Traumatology, Gent, Belgium Objectives: There is growing evidence in literature that meniscal substitution results in significant pain relief and functional improvement after meniscectomy. In this study we compared the patient clinical profile and outcome after implantation of meniscal allograft transplantation (MAT) with polyurethane scaffold (Actifit). Although both are arthroscopic procedures, the indications are strictly different: meniscal allograft transplantation is used in patients who underwent total meniscectomy whereas the polyurethane scaffold is implanted in patients with partial meniscectomy. Goal of this study was to define clinical outcome, reintervention rate and implant structural failure rate after a minimum follow-up period of 2 years to establish its safety and reproducibility. Methods: We evaluated 73 patients after a mean follow-up period of 3.3 years (range 2–5 years). 36 patients underwent an arthroscopic MAT at a mean age of 31.9 years (range 17–47 years). The second group consisted of 37 patients who received a polyurethane scaffold because of partial meniscus loss at a mean age of 31.43 years (range 15–50 years). Patients were evaluated clinically preoperatively and at final follow-up using KOOS, HSS, VAS, Tegner, Lysholm and IKDC questionnaires. Complications were defined as reintervention surgeries. Implant structural failure was defined as ruptures or tears. Results: Clinical outcome improved significantly for both MAT and scaffold in a similar fashion. In 50% of the MAT cases resurgery was necessary after a mean of 14.8 months (range 1–42 m). Structural failure was observed in 4 of 36 (11%) cases (1 partial meniscectomy, 1 graft removal, 2 ruptures). In two cases septic arthritis was seen. In the scaffold group reintervention was necessary in 45.9% of cases. 8 of 37 (21%) polyurethane scaffolds showed tissue/construct damage necessitating debridement. Conclusions: Although both meniscus substitution procedures have strictly different indications, the clinical outcome and reintervention rate are very similar for meniscal allograft transplantation and polyurethane scaffold implantation: Clinical follow-up showed significant pain relief and improvement of function of the knee joint, while a reintervention rate of approximately 50% is observed. The structural integrity of MAT is 89% while 79% for the scaffold group at a minimum follow-up of 2 years.

P24-1508 Early results after arthroscopic meniscus suture in patients with a ,,relative’’ indication P.R. Melinte1, C. Rodica1, D. Niculescu1, S.C. Didu1, C. Mercut1 1 University of Medicine and Pharmacy, Orthopaedic and Traumatology, Craiova, Romania Objectives: The purpose of this work is to evaluate the precocious and late results of arthroscopic meniscus suture with a ‘‘relative’’ indication. Methods: 54 patients (from 923 arthroscopies—5.85%) benefited by this arthroscopic meniscus suture, between 2007 and 2011. Results: The precocious and late results were appreciated on clinical, imagistic and arthroscopic criteria (‘‘second-look’’). The groups of patients identified with a relative indication of meniscus suture were characterized by the one or more of the following: the age over 30 years (the superior limit—39 years old—20 cases/37.3%); complex meniscus’ lesion (longitudinal lesion in the vascular area,

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 associated with minimal radiar lesion of the free edge)—15 cases/ 27.77% (14 cases/93.33% completely healed and 1 case/6.66% demanded completion of free edge resection); ancient lesion of the meniscus (between 1 and 6 months; 39 cases/72.22%); unstable knee (ACL lesion -32 cases/59.25%)—the lesion was later found completely healed during the ACL reconstruction, that was performed only in 27 cases (84.37%), 4 patients refusing this procedure because the instability disappeared after the meniscus reinsertion; lesion in the ‘‘red -white’’ area (8 cases/14.81%)—3 failures (37.5%). The sutures healed nicely. Conclusions: The extension of indication of meniscus suture to those cases considered ‘‘at the limit’’, is generally successfully, while respecting the proper surgical technique (a good abrasion of both edges of the lesion, with the shaver) and a sufficient immobilization after the surgery (30 days). Poor results were obtained only when the rupture was in a hypo vascular area, the placement of the lesion remaining an essential criterion for meniscus suture indication. The association between ACL rupture and meniscus tear that is fit for suturing is frequent and the results are promising.

Knee-Mensicus II

P25-10 Width is more reliable than length in graft sizing using plain radiography prior to lateral meniscal transplantation S.I. Bin1, B.-S. Lee2, J.-W. Chung3, J.-M. Kim1, K.-A. Kim4 1 Asan Medical Center, Department of Orthopedic Surgery, Seoul, Republic of Korea, 2Incheon St. Mary’s Hospital, Catholic University, Orthopaedic Surgery, Incheon, Republic of Korea, 3CHA Bundang Medical Center, Seongnam-Si, Republic of Korea, 4 Chungbuk National University, Cheongju, Republic of Korea Objectives: Accurate graft sizing is important for anatomical positioning of meniscal transplants. Plain radiographs are frequently used to select appropriately sized meniscal allografts, of which the width and length are not always perfect simutaneously. The purpose of this study was to determine which of these dimensions should assume priority when matching is considered for adequate positioning of lateral meniscal transplants. Methods: The positions of 34 lateral, fresh-frozen meniscal allografts, transplanted using the central bone bridge method, were evaluated by magnetic resonance imaging (MRI) 2 days after surgery. Allograft size was estimated on magnification controlled plain radiographs and meniscal transplant dimensions were intraoperatively measured. A size mismatch was defined as a difference between preoperative size and a real dimension of the transplants. The anterior and posterior horn positions on the sagittal plane and the lateral subluxation of the midbody on the coronal plane were evaluated by MRI. The reliability of preoperative radiographic measurements was determined by assessing the association between size mismatch and the meniscal position on each plane. Results: The mean differences between real allograft dimensions and preoperative radiographic measurements were averaged -0.2 ± 0.9 mm (-0.6 ± 2.9%) for width and 0.1 ± 2.3 mm (0.5 ± 6.2%) for length. The mean lateral subluxation of the midbody on the center of coronal sections was 1.7 ± 1.8 mm. The anterior and posterior horns were located 2.0 ± 2.1 mm and -3.8 ± 2.7 mm from the articular edge, respectively, in the center of sagittal images. Lateral subluxation was significantly associated with width mismatch (r = 0.415 * 0.486, p \ 0.05), but length mismatch was not significantly correlated with the anterior or posterior horn positions on sagittal images (p [ 0.1).

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: Width matching using plain radiographs is more reliable than is length matching when it is sought to assure adequate positioning of meniscal transplants, if both dimensions cannot be simultaneously matched.

P25-24 Meniscectomy after failure of meniscal repair: a review of 37 cases N. Pujol1, O. Barbier1, P. Boisrenoult2, P. Beaufils3 1 Hopital Andre Mignot, Orthopaedic Department, Le Chesnay, France, 2Hopital Andre Mignot, Chirurgie Orthope´dique, Versailles, France, 3Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France Objectives: Little is known about the amount of meniscal tissue removed after failed meniscal repair. The objective of this study was to evaluate wether volume of subsequent meniscectomy after failed meniscal repair is not increased when compared to the meniscectomy if performed initially, and to look for early chondral degeneration. Methods: From January 2000 to December 2009, 295 knees underwent arthroscopic meniscal repair for unstable peripheral vertical tears. When present (219 cases), all anterior cruciate ligament (ACL) tears underwent reconstruction. Patients with multiple ligament injuries and posterior cruciate ligament injuries were excluded from the analysis. Thirty-two medial and 5 lateral menisci underwent subsequent meniscectomy after failed repair at a mean of 26 months postoperatively (range, 3–114). Five parameters were specifically evaluated: the amount of meniscectomy related to the initial tear, the ACL status, the appearance of chondral lesions, the time from the initial injury to meniscal repair, and the time from repair to meniscectomy. Results: The posterior segment of the meniscus was involved in all tears and re-tears. Among failures, resection of the meniscal segments primarily repaired occurred for 17 (52%) medial and 2 lateral meniscal tears; the tear extended in 5 cases (All medial menisci), and a healing of some repaired segments led to a partial resection of the initial lesion in 35% of cases (10 medial menisci, 3 lateral menisci). The time from injury to meniscal repair was correlated with an increasing volume of meniscus removed (p \ 0.05), and with the presence of stage 2–3 chondral lesions at revision (p \ 0.03). All knees with extended tears (5 cases) and/or with significant chondral degeneration (8 cases) occured in ACL reconstructed knees. Among them, 50% of ACL reconstructed knees were ACL deficient. Conclusions: There are few detrimental effects when repairing a repairable meniscal lesion, even if it fails. The amount of meniscectomy is rarely increased when compared to the initial lesion. This study supports the hypothesis that the meniscus can be partially saved and that a risk of a partial failure should be taken when possible.

P25-30 Midterm outcomes after meniscal allograft transplantation: comparison of cases with extrusion versus without extrusion S.I. Bin1, D.H. Lee2 1 Asan Medical Center, Department of Orthopedic Surgery, Seoul, Republic of Korea, 2Korea University, Anam Hospital, Seoul, Republic of Korea Objectives: Although graft extrusion is of concern after meniscal allograft transplantation (MAT), the correlation between extrusion and clinicoradiological outcomes remains unclear. Hypothesis Patients with graft extrusion after MAT have worse clinical outcomes and greater arthritic change than those without graft extrusion. Methods: Forty-three patients underwent MAT between 1999 and 2004. Grafts were classified as either nonextruded (\3 mm) or extruded (C3 mm) according to 1-year postoperative magnetic

S249 resonance imaging (MRI) findings. The mean patient age at the time of surgery was 33.5 years (range, 17–43 years), and the mean followup period was 5.1 years (range, 3.5–8.3 years). The Lysholm score was used to evaluate knee function. In addition, preoperative and postoperative (final follow-up) measurements were taken to determine the absolute and relative (affected side/normal side) joint space width (JSW) on a standing 45 flexion posteroanterior view. Results: Magnetic resonance imaging at 1 year showed the mean graft subluxation was 3.03 ± 0.872 mm across all patients. Twenty-six knees (60%) were classified as nonextruded and 17 (40%) as extruded. No statistical difference was found between these 2 groups regarding Lysholm score improvement after MAT. Overall, absolute and relative JSWs were slightly narrower postoperatively compared with preoperatively across all patients (mean absolute difference, 0.283 mm; mean relative difference; 4.79%; p \ .001). However, extruded and nonextruded knees were similar in terms of absolute (p = .764) and relative (p = .482) JSW after MAT. The amount of extrusion did not correlate with Lysholm score or the relative difference between preoperative and postoperative JSW across all patients or in either group. Conclusions: Joint space width was slightly narrower after MAT. Extrusion had no effect on 5-year clinical or radiological outcomes.

P25-48 Meniscal healing after repair of chronic meniscal tears: a mri arthrography assessment D. Popescu1, R. Diego2, S. Sastre3, L. Lozano4, J.C. Martinez-Pastor5, F. Macule5 1 Hospital Clinic i Provincial, Barcelona, Spain, 2Hospital Clinic Barcelona, Barcelona, Spain, 3Hospital Clinic i Universitari de Barcelona, Knee Unit. Orthopaedic Surgery Department (ICEMEQ), Barcelona, Spain, 4Hospital Clinic de Barcelona, Knee Unit, ICEMEQ, Barcelona, Spain, 5Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain Objectives: The aim of this study was to assess the healing rate for arthroscopic meniscal repair in chronic lesions and to detect if there was any correlation between the clinical outcomes and the anatomic results on an arthrography combined with a Magnetic Resonance (arthro-MRI). Methods: A meniscal suture was carried out on 38 patients between January 2006 and December 2008. On 29 patients the meniscal healing was evaluated with an arthro-MRI, at least 6 months after the surgery. The mean waiting time until surgery was 24.7 months (3–80). Average follow-up was 23.6 months (14–48). 15 patients (52%) required reconstruction of an Anterior Cruciate Ligament (ACL) associated lesion. 20 patients (69%) showed medial meniscus lesions. All lesions were located in the red zone or red-white zone. Clinical evaluation included Barett’s healing criteria and the Lysholm and Tegner scores before and after the operation. The arthro-MRI parameters were— healing in thickness (Henning criteria) and the overall healing rate. Results: According to the Barett criteria, meniscal lesion healing was achieved in 24 patients (82%). Scores on the Lysholm and Tegner scales improved from 59.5 preoperatively to 92.3 postoperatively and from 2.9 preoperatively to 6 postoperatively, respectively. Three arthroscopic revisions were carried out and a partial meniscectomy was performed. According to Henning’s criteria, 13 of the menisci healed completely, 9 partially, and 7 failed. The overall healing rate was 76%. Tears located in the posterior horn had a healing rate of 58%. Tears extending from the posterior to the middle part had a healing rate of 83.1%. Better healing results were observed in tears associating an ACL reconstruction. All the tears with negative Barett criteria had a failed healing according to Henning anatomic criteria. Conclusions: The results obtained prove that, despite the chronic nature of the meniscal lesion, good clinical and anatomic outcomes

123

S250 can be achieved. ACL reconstruction had a favourable effect on meniscal healing. Partial healing occurred often, but the meniscus was painless and stable. A correlation was found between the Barett clinical criteria and the Henning anatomic ones.

P25-100 Radiological mid-term outcome after partial meniscus substitution using the collagen meniscal implant using MRI M.T. Hirschmann1, L. Keller2, L. Schenk1, R. Berbig2, U. Lu¨thi2, M.P. Arnold1 1 Kantonsspital Bruderholz, Klinik f. Orthop. Chirurgie & Traumatologie, Bruderholz, Switzerland, 2Sportclinic Zu¨rich, Zu¨rich, Switzerland Objectives: Meniscal substitution using collagen or polyurethane meniscal implants is a promising procedure in patients after subtotal meniscectomy. The aim of the study was to evaluate the radiological outcome of a collagen meniscus graft implanted in an injured medial or lateral meniscus after a minimum follow-up of 1 year. Methods: 81 patients (male:female = 55:26) underwent arthroscopic implantation of the collagen meniscus device (CMI) after subtotal medial (n = 62) or lateral meniscectomy due to persistent compartment pain. The lesions were in the anterior horn (n = 33), corpus (n = 65) and/or posterior horn (n = 65) of the meniscus. 63 underwent an additional procedure such as anterior cruciate ligament reconstruction (n = 52). The patients were seen for radiological follow-up using MRI. MRI scans were performed in all patients at minimum 1 year follow-up and analyzed according to the Genovese criteria. The Genovese criteria consisted of direct (implant morphology/size, signal intensity) and indirect (state of the corresponding cartilage of the medial/lateral femur or tibia, size of the cartilage lesions [\ 50%, signs of bone marrow edema). The extrusion of the meniscus was noted as \ and [3 mm. Results: In 10 patients (12%) the meniscus implant was entirely resorbed, in 47 (57%) partially resorbed and in 25 (31%) entirely preserved. It was in particular obvious that sagittal, coronal and axial slices have to be taken into consideration to evaluate size of the meniscus. In 15 patients (18%) the meniscus graft was isointense, 63 (78%) were slightly hyperintense and 3 (4%) highly hyperintense. The size of the cartilage lesion was \50% in 47 patients (58%) and [50% in 34 patients (42%). 45 patients (56%) showed no signs of bone marrow edema. 55 (68%) patients showed an extrusion of the meniscus [3 mm at last follow-up. Conclusions: The MRI results clearly indicate that the collagen meniscus implant undergoes significant remodeling, degradation and extrusion in a significant number of patients. However, to draw conclusions on the success of this procedure, the clinical outcomes have to be taken into consideration.

P25-230 Morphologic changes in fresh-frozen meniscus over 1 year: a prospective magnetic resonance imaging study on the width and thickness of transplants S.I. Bin1, B.-S. Lee2, J.-M. Kim1, W.-J. Cho1 1 Asan Medical Center, Department of Orthopedic Surgery, Seoul, Republic of Korea, 2Incheon St. Mary’s Hospital, Catholic University, Incheon, Republic of Korea Objectives: Little is known about morphologic changes in remodeling period following human MAT. It was hypothesized that the gross morphology of meniscal transplants may be altered significantly in width and thickness during early remodeling periods. Using serial magnetic resonance imaging (MRI), we assessed the rate of meniscal transplant deformation, its time of occurrence and the characteristics

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 associated with deformation over 1 year, as well as perioperative risk factors for deformation. Methods: Thirty patients who underwent meniscal allograft transplantation (MAT) (9 medial and 21 lateral menisci) between 2008 and 2009 were prospectively evaluated by serial MRI scans 2 days, 6 weeks, and 3, 6, and 12 months after surgery. At each time point, the width and thickness of the menisci were measured on midcoronal images. The relative degree of shrinkage after 1 year was categorized as minimal (\10%), mild (10 * 25%), moderate (25 * 50%), or severe ([50%). Logistic regression analysis was used to determine whether perioperative factors, including preoperative alignment, cartilage status, age, gender, and time from previous meniscectomy, were associated with more than moderate shrinkage. Results: At baseline, mean meniscal width and thickness were 9.45 ± 1.35 mm and 6.77 ± 1.09 mm, respectively. At 1 year, width decreased to 8.44 ± 1.85 mm (p \ 0.01), while thickness increased to 7.63 ± 1.51 mm (p \ 0.01). Shrinkage was observed for 3 months after MAT, but there were no significant changes thereafter. The thickness of the meniscal peripheral rim did not change before 3 months, but increased afterward, for up to 1 year. Of the 30 patients, 19 (63.3%) had minimal shrinkage, 6 (20.0%) had mild shrinkage, 5 (16.7%) had moderate shrinkage and none had severe shrinkage over 1 year. Preoperative alignment, cartilage status, age, gender, and time from previous meniscectomy did not affect the degree of shrinkage. Conclusions: Gross morphologic alterations, as determined by width and thickness, were significant during the first postoperative year. The perioperative conditions of the recipient knee did not affect these changes.

P25-293 Long term follow-up after meniscal repair A. Liantsis1, L. Willberg1, B. Engstrom1 1 Karolinska Institutet, Capio Artro Clinic & Stockholm Sports Trauma Research Center, Stockholm, Sweden Objectives: The aim of this study was to analyse the clinical results of meniscal repair and also to identify factors that correlate with the success or failure of this procedure after a long term follow-up. Methods: A retrospective study of 193 out of 637 patients who underwent meniscal suture in the period 1997–2010 at the Capio Artro Clinic. The average follow-up was 84 months. Patients age was 13–61 years (mean 27.5 years), 94 male and 99 female. All 193 patients were evaluated with the Knee injury and Osteoarthritis Outcome Score (KOOS) and with nine questions ‘‘Patient Questionnaire’’ concerning the state of their knee. Results: The overall failure rate of the meniscal repairs was 20%. From the total 193 patients, 126 had as well anterior cruciate ligament rupture (ACL). 72 of them underwent ACL reconstruction at the same time as meniscal repair, 29 later and 25 had no ACL reconstruction. The failures in each category was 14 (19%), 7 (24%) and 5 (20%) respectively (N.S). From those patients who had no ACL rupture (67 patients) 13 (19%) failed (N.S). There was no correlation between meniscal failures and time elapsed from injury to surgery (p = 0.342). Neither any significant difference was found between the age groups ([25 years and B25 years) regarding success or failure of the meniscal suture. The decreased level of activity was statistically correlated with: the patients who had simultaneously cartilage injury (p = 0.01) and also with ages over 25 years (p = 0.02). Conclusions: We found an overall failure rate of 20% in this group of meniscal repairs. Factors as ACL rupture or time from meniscal repair to ACL surgery showed no statistical difference in the outcome of failures. Decreased level of activity was correlated with cartilage injury and with age over 25 years.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P25-401 Pullout failure strength of the posterior horn of the medial meniscus with root ligament tear: a biomechanical study Y.-M. Kim1, Y.-B. Joo1, T.-H. Kang1, Y.-S. Jeon1, S.-C. Han1 1 School of Medicine, Chungnam National University, Department of Orthopaedic Surgery, Daejeon, Republic of Korea Objectives: To evaluate the validity of reparability of the posterior horn of the medial meniscus (PHMM) with the root ligament tear by measuring the real pullout failure strength of simple vertical suture of arthroscopic subtotal meniscectomized PHMM. Methods: From November 2009 to May 2010, 9 specimens of the PHMM were obtained from arthroscopic subtotal meniscectomy as a treatment of the root ligament rupture of the PHMM. Simple vertical suture using a 0 Ethibond was performed at 7 mm medial point from the torn root ligament of the PHMM and baseball suture using two 0 Ethibonds were done on the resected medial margin. Afterwards, while covered in normal saline wet gauze, the specimen was kept in frozen state at -70C. The degree of degeneration, extrusion, and medial displacement of the medial meniscus (MM) were evaluated through MRI. Kellgren– Lawrence (K–L) classification was measured in standing plain radiography, and mechanical alignment was measured using orthoroentgenography. In arthroscopy, by the torn site of the root ligament of the PHMM, torn morphologies were classified into ligament proper type or meniscoligamentous junctional type, and associated injuries were checked if present. Results: The mean pullout failure strength of the PHMM was 71.613 ± 23.181 N (range 41.434–107.731 N). Degeneration of the PHMM in MRI showed statistically significant correlation with pullout failure strength and K–L classification. Pullout failure strength showed correlation with mechanical alignment and K–L classification (p \ .05). Conclusions: The measurement of pullout failure strength of PHMM with root ligament tear showed a result of repairable degree. The degenerative degree of the PHMM on MRI showed significant correlation with the pullout failure strength, and pullout failure strength was also confirmed to be correlated with degeneration degree of the PHMM, mechanical alignment, and K–L classification which represent bony degenerative change.

P25-404 Role of the mechanical axis of lower limb and body weight in the horizontal tear and root ligament tear of the posterior horn of the medial meniscus Y.-M. Kim1, Y.-B. Joo1, T.-H. Kang1, Y.-S. Jeon1, S.-C. Han1 1 School of Medicine, Chungnam National University, Department of Orthopaedic Surgery, Daejeon, Republic of Korea Objectives: To compare and analyze about the relationship between horizontal tear and root ligament tear of the posterior horn of the medial meniscus (PHMM) and the degree of varus of the axis of lower limb and body weight. Methods: Of the 129 patients who overwent surgical treatment as they were diagnosed medial meniscus tear in our hospital from May 2006 to December 2009, 19 cases (group 1), who overwent partial meniscectomy as they were confirmed to solely have horizontal tear of the PHMM on arthroscopic examination, and 27 cases (group 2), who overwent subtotal meniscectomy as they were confirmed to solely have root ligament tear of the PHMM on arthroscopic examination, were chosen for retrospective study. We took a picture of orthoroentography for every case of prior to arthroscopic surgery and measured varus angle. Also, we checked body mass index (BMI) of two groups. The difference of varus angle and BMI between two groups were statistically verified using the Levene’s test, paired t test.

S251 Results: Group 1 showed mean value of varus angle of 2.30 ± 0.54, and BMI of 25.32 ± 3.23. Group 2 showed mean value of varus angle of 5.64 ± 0.54, and BMI of 25.67 ± 3.12. The degree of varus of group 2 was statistically significantly higher than group 1 (p = 0.002). Comparison between the BMI of two groups showed no statistical significance (p = 0.053). Conclusions: Authors have found out through a comparative study of sole horizontal tear and root ligament tear of the PHMM and, that sole root ligament tear of the PHMM is more relative to the genu varum than sole horizontal tear of the PHMM. However, body weight was statistically irrelevant to the occurrence of the two lesions.

P25-567 Learning and retaining complex arthroscopic knee skills T. Khan1, W. Jackson2, A. Alvand1, H. Gill3, A. Price4, J. Rees2 1 University of Oxford, Nuffield Department of Orthopaedics, Nuffield Orthopaedic Centre, Oxford, United Kingdom, 2University of Oxford, Nuffield Department of Orthopaedic Surgery, Nuffield Orthopaedic Centre, Oxford, United Kingdom, 3Nuffield Department of Orthopaedics, Botnar Research Centre, University of Oxford, Oxford, United Kingdom, 4Oxford University, Nuffield Department of Orthopaedic Surgery, Oxford, United Kingdom Objectives: Previous studies investigating the retention of surgical skills have suggested that objective loss of technical performance occurs after a 6-month period without practice. The aims of this study were to objectively demonstrate the learning curve for arthroscopic meniscal repair using a motion analysis tracking system and to determine the impact of task repetition on the retention of this skill. Methods: 19 orthopaedic residents with experience of routine knee arthroscopy, but not arthroscopic meniscal repair were recruited to a randomized study. During the ‘Initial Learning Phase’, all subjects performed 12 meniscal repairs on a knee simulator over a 3-week period. A validated motion analysis tracking system was used to objectively record their learning and performance using the outcomes of ‘time taken to complete task’, ‘total distance travelled’ and ‘total number of hand movements’. The subjects were then randomized into 3 groups: Group A continued to perform one meniscal repair episode each month. Group B performed one meniscal repair episode at 3 months and group C performed no repairs during this ‘Interim Phase’. All 3 groups then returned at the 6-month point and carried out a further 12 meniscal repairs over 3 weeks as in the initial learning period (‘Final Assessment Phase’). Results: All subjects demonstrated significant objective improvement over the initial 12 episodes for all three motion analysis parameters (p \ 0.0001, Wilcoxon Signed Rank test) demonstrating a clear learning curve. Although some residents had reached a learning plateau by 12 episodes, others continued to make further small improvements for up to another 6 episodes. Importantly, Group C did not display any loss of skill despite a 6-month break in task repetition (p [ 0.05, Wilcoxon Signed Rank test). Conclusions: This study further highlights the useful role of surgical simulation for resident training programs, especially for rarely performed procedures. In contrast to some previous studies, this study does not show any loss of surgical skill after a 6-month period of absence. These findings suggest the presence of taskspecific or surgeon-dependent factors that affect the retention of arthroscopic skills. We suggest that the use of generic guidelines on minimum task frequency for surgeons to maintain optimal performance at arthroscopic tasks may not always be appropriate.

123

S252 P25-662 Probing forces of menisci: what levels are safe for arthroscopic surgery? G. Tuijthof1, T. Horeman2, M. Schafroth3, L. Blankevoort3, G.M.M.J. Kerkhoffs3 1 Academic Medical Center/Delft University of Technology, Department of Orthopaedic Surgery/Department of Biomechanical Engine, Amsterdam/Delft, The Netherlands, 2Delft University of Technology, Department of Biomechanical Engineering, Delft, The Netherlands, 3Academic Medical Center, Department of Orthopaedic Surgery, Amsterdam, The Netherlands Objectives: To facilitate effective learning, feedback on performance during arthroscopic training is essential. Less attention has been paid to monitoring and providing feedback on safe handling of delicate tissues such as meniscus. Meniscal and cartilage tissue are especially at risk of unintentional damage as they have little to no healing potential and are frequently probed even outside the arthroscopic view. The goal is to measure in vitro probing forces of menisci and compare them with a theoretical maximum probing force (TMPF). Methods: Menisci samples of ten cadavers were mounted on custom made Force Platforms to measure probing forces up to 20 N in three directions. Each meniscus was firmly secured with two clamps to prevent slip and to imitate anatomic attachment. A Perspex cylinder was placed over the Force Platform in which access portals were made through which a routinely used arthroscopic probe could be inserted. Nineteen subjects participated: six novices (experience \60 arthroscopies in total), ten intermediates ([60 arthroscopies in total), and three experts ([250 a year). All had to perform three tasks on each meniscus sample with a probe: push three times on the superior meniscal surface, perform one continuous run on the superior meniscal surface and back, and pull three times on the inferior meniscal surface. The absolute maximum probing force (AMPF) was determined for each condition. A multivariable linear regression analysis was performed to assess the influence of experience on the force magnitude (p \ 0.05). AMPFs were compared to the TMPF, which was estimated to be 8.5 N. Results: The AMPF of the push task was on average 2.8 N (standard deviation (SD) 0.8 N), of the continuous run task 2.5 N (SD 0.9 N), and of the pull task 3.9 N (SD 2.0 N) for all participants. A significant difference was present between experts and novices (p \ 0.05), where the experts performed all tasks with higher forces. Additionally, a significant difference was found between the forces applied by the intermediates versus the experts for the continuous run task. The maximum probing force levels of AMPF per task and experience ranged from 3.4 to 9.1 N, which are in the same order of magnitude as the TMPF. Furthermore, the experts indicated that a safety margin of 20% would be recommended. This equals a maximum probing force of 6.8 N. Conclusions: The results indicate that the theoretical maximum probing force of 8.5 N is exceeded by some of the measured participants. Therefore, it is necessary to use a force threshold value for tissue manipulation when training arthroscopy. This threshold should inform the trainee that the force exceeds dangerous force levels when probing meniscus tissue. Both virtual and physical medical simulators can benefit by implementing the results into their performance monitoring systems.

P25-783 MRI and clinical assessment of the contralateral knee of symptomatic discoid lateral meniscus J. Takigami1, Y. Hashimoto2, S. Yamasaki2, H. Nakaura1 1 Osaka City University Graduate School of Medicine, Orthopaedics, Osaka, Japan, 2Osaka City University Graduate School of Medicine, Department of Orthopaedic Surgery, Osaka, Japan Objectives: Discoid lateral meniscus has reported prevalence that ranges from 0.4 to 17%, occurring at a higher rate in Asian

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 populations. But few reports have described the prevalence of bilateral discoid lateral meniscus. The purpose of this study was to evaluate the contralateral knee of symptomatic discoid lateral meniscus with magnetic resonance imaging (MRI) and clinical examination. Methods: 22 patients underwent arthroscopic surgery for symptomatic discoid lateral meniscus between March 2007 and October 2011. 11 of 22 patients were examined MRI of contralateral knee during the follow-up. The patients were all young Japanese descent, six were males and five were females. The mean age of patients was 11.6 years (range, 8–14 years). Their contralateral knees were evaluated with MRI and clinical examinations. We categorized the lateral meniscus into normal meniscus (NM), incomplete discoid meniscus (ICDM), and complete discoid meniscus (CDM), and evaluated meniscal tear, meniscal shift, chondral lesion and the size of the popliteal space with MRI. For clinical assessment, McMurray test with pain and/or click were recorded. Results: We found that all of patients had a complete or incomplete discoid lateral meniscus in the contralateral knee. 10 (90.9%) were complete and one (9.1%) was incomplete. Operative sides of 11 patients were all complete discoid lateral meniscus knees. Meniscal tears of operative knees and contralateral knees were 9 knees (81.8%) and 0 knee (0%), meniscal shift were 4 knees (36.4%) and 0 knee (0%), chondral lesion were 2 knees (18.2%) and 2 knees (18.2%), widening of the popliteal space being larger than 5 mm on both the coronal and sagittal images were 1 knee (9.1%) and 0 knee (0%), respectively. Clinical assessment were identified that pain was observed in 2 knees (18.2%), and click was observed in 6 knees (54.5%) with McMurray test and no cases with limitation of knee extension or locking history were observed in contralateral knees. Conclusions: Discoid meniscus was previously reported an atavistic anomaly and usually occurred laterally. Previous Japanese cadaveric study with macroscopic anatomy reported that 192 of 577 knees (33.2%) were the incomplete and complete discoid lateral meniscus and the prevalence of bilateral discoid lateral meniscus was 81.5%. In this study evaluated the occurrence ratio of bilateral discoid lateral meniscus with MRI, all of patients (100%) with a symptomatic discoid lateral meniscus had ICDM or CDM in the contralateral knee in Japanese case. The limitation of this study is the number of patients evaluated was small and not prospective study.

P25-808 The impact of free or restricted rehabilitation after meniscus repair. A prospective randomized clinical trial M. Lind1, B. Lund1, P. Faunoe1, S.E. Christiansen2 1 University Hospital of Aarhus, Department of Sportstraumatology, Aarhus C, Denmark, 2University of Aarhus, Sports Trauma Division, Orthopedic Department, Skanderborg, Denmark Objectives: The optimal rehabilitation after meniscus repair has not been established. No controlled trials exist in the literature and numerous regimes have been suggested as beneficial for meniscus healing. The purpose of this study was to investigate the outcome of meniscus repair with an either free or restricted rehabilitation regimen in a prospective randomized controlled clinical trial. Methods: 60 patients were included in the trial. Patients were included into the trial within 4 days postoperatively if meniscus repair was performed during an arthroscopy procedure. Only vertical meniscus lesions close to capsule was repaired. Meniscus repair was performed with all-inside technique using Fast-Fix or Meniscus Arrow implants. No patients with concomitant ligament reconstruction or cartilage repair surgery were included in the study. 33 patients and 27 patients were randomized to free and restricted rehabilitation respectively. Free rehabilitation consisted of 2 weeks in brace 0–90 and touch weight bearing, where after free rehabilitation was allowed. Restricted rehabilitation consisted of 6 weeks in brace with gradual range of

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1 motion increase to 90 and only touch weight bearing during the 6 weeks. Patients were seen for follow-up at 3, 12, 24 months. Any patients with continuous joint line pain at 3 months had MRI and subsequent arthroscopy to evaluate healing of the repaired meniscus. At follow-up visits, patients were evaluated by Knee Osteoarthritis Outcome Score (KOOS), Tegner Score, pain assessment and patient satisfaction. Data from the present study are results after 1-year follow-up. Results: Six patients were lost to follow-up. Re-arthroscopy in patients with continuous symptoms demonstrated partial healing or lack of healing in 33% and 24% of patients in the restricted and free rehabilitation groups respectively. KOOS score was similar between groups (Fig. 1). Also Tegner score and patient satisfaction was similar between groups. Conclusions: Free rehabilitation after meniscus repair is safe with a tendency to better healing rates than after restrited rehabilitation. Functional outcome at 1 year was not affected by rehabilitation regimen. An overall lack of healing of 28% for patients with isolated meniscus lesions is a concern.

P25-820 Second-look arthroscopic findings after open wedging high tibia osteotomy of posterior root tears of the medial meniscus K. Nha1, J.I. Kim2, J.H. Kwon2, B.H. Jang2, H.D. Lee1 1 Inje University, Ilsanpaik Hospital, Koyangsi, Republic of Korea, 2 Inje University, Ilsanpaik Hospital, Koyang, Republic of Korea Objectives: The purpose of this study was to evaluate the results of open wedge high tibial osteotomy (HTO) focused on RTMMP by second-look arthroscopy. The hypotheses of this study were (1) the degree of alignment correction is related to the healing of RTMMP and (2) healing of RTMMP would be beneficial in the cartilage lesion and clinical results. Methods: Among 31 consecutive patients who underwent HTO without meniscectomy or pullout repair about RTMMP, 20 patients (mean, 28 months; range, 24–34 months) were followed more than 2 years and available for second-look arthroscopic evaluation. The healing status of the RTMMP was classified as complete, incomplete, and no healing. The difference of weight bearing line between preoperation and last follow up was evaluated. Osteoarthritis and chondral lesion were evaluated by Kellgren–Lawrence (K–L) grade and Outerbridge classification, respectively. Clinical results were also evaluated by the Lysholm and the Hospital for Special Surgery (HSS) scores. We evaluated the correlation between healing status and other variables (weight bearing line, K–L grade, Outerbridge classification, and HSS scores). Results: There were 10 (50%) cases with complete healing, 6 (30%) with incomplete healing, 4 (20%) with no healing. The K–L grade was

S253 not improved with the standing plain radiograph (p = .09), but no progression of the chondral lesion was observed in second-look arthroscopy; some improvement was even observed (p = .002). The mean Lysholm score improved from 56.1 preoperatively (range, 41–71) to 89.0 at the last follow-up (range, 69–94; p = .003). The mean HSS score also increased significantly from 64.1 (range, 50–76) to 86.4 (range, 80–92; p = .003). The difference in the weight bearing line between before surgery and the last follow up correlated with the healing status (p = .04, correlation coefficient: .463) and the degree of the weight bearing line before surgery was negatively associated with the healing status (p = .049, correlation coefficient: -.427). Conclusions: Alignment correction and load re-distribution by HTO could cause healing of the RTMMP, and HTO offers benefit to cartilage lesions and the clinical results. Healing was positively correlated with the magnitude of correction and negatively correlated with the weight bearing line before surgery.

P25-906 Treatment of meniscus horizontal tear M.C. Lee1, J.K. Lee1, S. Lee1, J. Jang1, S.H. Chun1, S.C. Seong1 1 Seoul National University College of Medicine, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: Meniscus horizontal tears are usually associated with degeneration. Because there is no disruption in the continuity of circumferential fibers, load bearing, shock absorbing functions are largely preserved. In this study, we tried to find out the factors to successful conservative and arthroscopic treatments of meniscus horizontal tear, respectively. Methods: One hundred and sixty-six meniscus horizontal tear patients without combined ligament injury, other types of meniscus tear such as longitudinal, radial, flap, complex were included in this study. Patients who had non-arthroscopic surgeries or operations at other hospital were excluded. All patients were followed up for minimum of 2 years. We generally recommended arthroscopic surgery to patients with definite mechanical symptoms (positive McMurray test, locking, giving way, clicking). Factors considered were age, gender, joint line tenderness, mechanical symptom, widest tear gap width on MRI sagittal view, grade of joint arthritis (International Cartilage Repair Society (ICRS) grade), combined root tear, discoid meniscus, tear site, and joint alignment on X-ray. Chi square test and multiple logistic regression analysis were done to determine the factors that lead to successful conservative and arthroscopic treatment, respectively. Results: The patients whom we recommended arthroscopic operation were younger (patients over 60 compared to patients under 50, p value = 0.022, odds ratio : 0.132, 95% confidence interval : 0.031–0.568) and had wider tear gap (widest gap width more than 1.5 mm compared to widest gap width \1.0 mm, p value \ 0.001, odds ratio : 49.130, 95% confidence interval : 9.071–266.082). Combined root tear was the only meaningful factor that led to failure of the conservative treatment (p value = 0.025, odds ratio : 2.860, 95% confidence interval : 1.140–7.175). Other factor that could be considered was joint line tenderness (p value = 0.136, odds ratio : 2.179, 95% confidence interval : 0.782–6.072). ICRS grade was the only significant factor that led to the failure of arthroscopic meniscectomy (ICRS grade 3 or more compared to ICRS grade 0, p value = 0.035, odds ratio : 33.383, 95% confidence interval : 1.288–864.902). Conclusions: Meniscus horizontal tear patients with combined root tear and joint line tenderness did not respond well to conservative treatment. Patients with advanced arthritis had poor outcomes after arthroscopic meniscectomy. When deciding the treatment option for meniscus horizontal tear, other factors should be considered thoroughly in addition to mechanical symptom.

123

S254 P25-945 Meniscal repair in soccer players using all-inside meniscal suture device. Long term results P. Alvarez1, F. Llobet1, G. Nelson1, G. Steinbacher1, C. Ramon1 1 Catalonian Soccer Federation, Orthopedic Surgery and Traumatology, Barcelona, Spain Objectives: The aim of this study was to analyse the long-term results of meniscal tears repaired with Fast-Fix All-inside suture of the federated Catalonian soccer players. Methods: 81 Patients operated by suture meniscal Fast-Fix device and with a minimum of 5 years of follow-up, had been reviewed. The clinical histories were reviewed and patients were called telephonically to value their current condition. We contact 29 patients, 15 of them suffered an associated surgery of reconstruction of the LCA during the same surgery. The average of follow-up for these patients was 6.2 years. Results: All the patients (100%) returned to play soccer to a competitive level after the operation. Five patients (17.2%) presented persistent pain after the surgery and 2 of them (6.7%) needed arthroscopic revision and meniscectomy. A patient was re-operated due to a new injury playing soccer. No extrarticular complications were founded. Nowadays 13 patients (45%) continue playing soccer, 8 of them (28%) do it to the same level as before the surgery. Conclusions: The suture meniscal with FasT-Fix is a good option for the treatment of the meniscal ruptures in soccer players who play to competitive level.

P25-966 Relationship between meniscal extrusion and two different fixation techniques in meniscal allograft transplantation F. Abat1, G. Gonzalez-Lucena1, P.E. Gelber2, J. Erquicia3, X. Pelfort4, J.C. Monllau5 1 University of Barcelona. Hospital de la Santa Creu i Sant Pa, Department of Orthopaedic and Traumatology, Barcelona, Spain, 2 Hospital de la Santa Creu i Sant Pau, Department of Orthopaedic and Traumatology, Barcelona, Spain, 3ICATME- Institut Universitari Dexeus. UAB., Traumatology and Orthopaedia, Barcelona, Spain, 4 ICATME- Institut Universitari Dexeus. UAB., Orthopaedia, Barcelona, Spain, 5University of Barcelona. Hospital de la Santa Creu i Sant Pa, Barcelona, Spain Objectives: To study the differences between the degree of meniscal extrusion (ME) of the transplanted meniscal allograft in relation to two types of graft fixation used. Methods: Prospective series of 88 meniscal allograft transplantations (Fig. 1). Thirty-three of the cases were performed with an only-suture technique (Group A). The remaining 55 cases were operated on with bony fixation method (Group B). All patients were studied with magnetic resonance imaging (MRI) in order to determine the degree of meniscal extrusion (Fig. 2). Patients with malalignment of the lower extremity were excluded. Time between surgery and MRI evaluation was 4 years in both groups. ME was assessed in MRI coronal views. ME was evaluated measuring the distance between the peripheral edge of the transplanted meniscus to the edge of the tibia, divided by the total width of the meniscus. In order to standardize this measure for the different knee sizes, the value is presented as percentage. Variability between compartments was kept in mind and the degree of extrusion was correlated to the fixation method used in the surgery. Results: The percentage of ME in Group A was 36.3% ± 13.7% with a no difference (p = 0.84) between medial (35.9% ± 18.1%) and lateral (38.3% ± 14.4%) compartment. Group B had an average percentage of ME of 28.13% ± 12.2% without differences between the medial compartment at (25.8% ± 16.2%) and lateral (30.14 ± 13.5%). Upon

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1 Meniscal allograft transplant

Fig. 2 MR study for ME

comparing meniscal extrusion with the type of fixation employed, a higher percentage of extrusion was found in Group A. Conclusions: Meniscal allograft fixed with an only-suture technique had a greater tendency to extrusion tan those fixed with bony fixation method.

P25-1064 Repair for radial tears of the midbody of the lateral meniscus using the new suture technique T. Suzuki1, K. Nakata2, H. Otsubo1, K. Watanabe3, T. Yamashita3 1 Sapporo Medical University, Orthopaedic Surgery, Sapporo, Japan, 2 Osaka University, Suita City, Japan, 3Sapporo Medical University School of Medicine, Orthopaedic Surgery, Sapporo, Japan Objectives: It has been known that radial tears of the meniscus dramatically reduced hoop strain, and leads to dysfunction of load distribution. Repair for radial tears of the midbody of the lateral meniscus have been challenging and rarely reported. The purpose of this study was to describe the surgical technique and clinical outcomes with the second-look arthroscopy and functional 3D-MR images after inside-out suture repair for radial tears of the midbody of the lateral meniscus. Methods: Three patients with radial tear of the lateral meniscus underwent Arthroscopic meniscus repair. Two male and one female

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 whose age were 24, 17, and 45 years. The causes of injury were sports (badminton, rugby, tennis). The time from injury to operation were 3, 6, and 4 weeks. Discoid meniscus was not included. All patients underwent new repair procedure ‘‘tie grip’’ suture using inside-out technique. First, two vertical divergent sutures were placed in both ends of radial tear for reduction and horizontal mattress sutures were secured over the tear site and the grip to make tie grip suture. After surgery, non-weight bearing were recommended for 6 weeks, followed by a gradual increase of weight bearing over the following 2 weeks. Range of motion exercises were allowed from 0 to 90 for 6 weeks and full squatting and jogging were permitted 12 weeks. All cases were taken 3 patterns (1, non weight-bearing 2, loading + full extension 3, loading + 45 flexion) images of functional 3D MRI using DynaWell L-spine compression device at 16 weeks after surgery. Results: The average follow-up was 19.6 months. No patients complained of lateral joint pain, click and loss of range of motion at last follow-up. The postoperative mean Tegner activity level was 8 and mean Lysholm score was 95. Second look arthroscopy of one patient showed complete healing. Follow-up functional 3D MRI showed normal meniscus function against simulated loading. Conclusions: Clinical outcomes after repairs of radial tears of the lateral meniscus showed satisfactory results by subjective and objective evaluations. Therefore, meniscal repair rather than meniscectomy should be considered for high activity patients.

P25-1067 The relationship of neural structures to arthroscopic posterior portals according to knee positioning S.H. Lee1, J.H. Ahn2 1 Kyung Hee University, Orthopaedic Surgery, Seoul, Republic of Korea, 2Kangbuk Samsung Hospital, Sungkyunkwan University, Orthopaedic Surgery, Seoul, Republic of Korea Objectives: The purpose of this study was to investigate the relationship between the proximity of neural structures to standard posterior portals in different knee positions. Methods: Ten fresh cadaveric knees were used to establish the standard posteromedial and posterolateral portals using an outside-in technique with arthroscopic transillumination. The distance from each portal site to the adjacent neurovascular structures (infrapatellar branches of the sapenous nerve) and the sartorial branch of saphenous nerve from the posteromedial portal, and the common peroneal nerve from posterolateral portal was measured using a precision caliper. Distances were recorded with the knees in extension, 45 of flexion and 90 of flexion in order to examine the effect of dynamic knee motion on nerve position. Results: The mean distance between the posteromedial portal and the superior and inferior branches of infrapatellar branches of the sapenous nerve at full extension was significantly greater than at 90 of flexion. However, there was no difference observed between at 45 and at 90 of flexion. The mean distance from the posteromedial portal to the sartorial branch of saphenous nerve at 90 of flexion was significantly greater (26.1 mm, SD: 4.7) than that at 45 of flexion (18.4 mm, SD: 4.6) and at full extension (14 mm, SD: 4.3) (p \ 0.0001). The mean distance between the posterolateral portal to the common peroneal nerve at 90 of flexion was also significantly greater (25.4 mm, SD: 9.2) that that at 45 of flexion (22.5 mm, SD: 8.1), and at full extension (20.1 mm, SD: 9.1) (p \ 0.0001). Conclusions: In this cadaveric study, it shows that position of 90 knee flexion is reasonable safe in order to establish a posterior portal

S255 in knee arthroscopy using the technique described in the current study, especially to avoid the sartorial branch of saphenous nerve and the common peroneal nerve injuries. Results of this study suggest that the establishment of posterior portals while the knee is flexed at an angle of 90 is recommended to ensure the safety of these structures.

P25-1247 The effectiveness of comprehensive physiotherapeutic programme in patients after AMMS (Artroscopic Meniscus Membrane Suturing) A. Prusin´ska1, T. Piontek2, K. Ciemniewska-Gorzela2, J. Naczk1, M. Grygorowicz1, W. Dudzin´ski1 1 Rehasport Clinic, Poznan´, Poland, 2Medical University of Poznan´, 2nd Department of Pediatric Orthopedics and Traumatology, Poznan´, Poland Objectives: Meniscus is crucial for appropriate functioning of the knee joint. In our clinic most often damaged meniscus is sutured during surgery. Since 2010 we have used a technique of covering the damaged meniscus with the Chondro-Gide collagen membrane and sealing it with staples. The aim of the study is to compare the results of biomechanical evaluation of a group of patients after a classical procedure of meniscal suturing with the results of a group of patients after the procedure of covering the damaged meniscus with the Chondro-Gide collagen membrane. In all patients, apart from the relevant procedure on the damaged meniscus, we have simultaneously performed anterior cruciate ligament (ACL) reconstruction. Methods: The procedure of meniscal repair with ACL reconstruction was performed in 21 patients. In 11 patients (age 35.8 ± 3.09 years; weight 79.81 ± 4.33 kg; height 178.09 ± 2.53 cm) the damaged meniscus was covered with the Chondro-Gide collagen membrane and sealed. In the remaining 10 patients (age 35.1 ± 3.24; weight 81.8 ± 4.54 kg; height 178.1 ± 2.66 cm) the meniscus was sutured using the classical method. To compare the results of patients who underwent two different procedures of meniscal repair, in the 12th week after the surgery we have conducted postural strategy and neuro-muscular control evalution of lower limbs and spine using the DELOS system, as well as muscle force test in isokinetic conditions at 60/s and 240/s angular velocity (Biodex System 3 Pro). Non parametrical tests were used for the statistical analysis. The level of significance was set at p \ 0.05. Results: The Mann–Whitney U test did not reveal any statistically significant differences between the Peak Torque (PT) and PT/Body Weight in the treated limb for flexion and extension at 60/s isokinetic angular velocity. The same test for flexion at 240/s isokinetic angular velocity did not show any differences between Total Work (TW) and TW/BW. Significant differences were observed at 240/s isokinetic angular velocity for extension between TW and TW/BW. No statistically significant differences between postural priority values were found between the groups in a static tests on a Delos platform. The group treated with the classical method recorded a statistically longer time of support in the dynamic test. Conclusions: (1) The surgical procedure used in the study may serve as an alternative, because the results of pilot studies prove that there are no differences in muscle force of the treated limbs, in both, flexion and extension. (2) The group of patients whose meniscus was covered with the Chondro-Gide collagen membrane and then sealed demonstrated better neuromuscular control, as proved by shorter support time during dynamic tests on the Delos platform. (3) The patients treated with the classical method displayed better muscular endurance during extension.

123

S256 P25-1315 Medial meniscal root repair: the biomechanical effectiveness of pull-out sutures E.F. Ro¨pke1, R. Becker2, C.H. Lohmann1, C. Sta¨rke1 1 Otto-von-Guericke Universita¨t Magdeburg, Orthopedics Department, Magdeburg, Germany, 2Sta¨dtisches Klinikum Brandenburg, Zentrum fu¨r Orthopa¨die und Unfallchirurgie, Brandenburg, Germany Objectives: Recently it was shown that the effect of meniscal root tears equals that of a total meniscectomy in certain aspects. Surgical repair has thus been advocated. Biomechanical research indicates that relatively high tensile forces are generated at meniscal roots under tibio-femoral loads. It is therefore unclear if a repaired meniscal root remains functional under low-level, repetetive tibio-femoral loads, as they can be encountered post-operatively. The aim of the study was to investigate if a meniscal root repair remains stable and functional in terms of cartilage protection under cyclic tibio-femoral loads of moderate magnitude. Methods: In porcine knees (n = 8) a subminiature displacement transducer was installed such that it measured compressive deformations of the cartilage as they occur with tibio-femoral loads. The knees were mounted in a materials testing machine at full physiologic extension. The specimens were then subjected to a cyclic load (n = 100) with an amplitude of 100 N. The deformation of the cartilage was recorded simultanously at a sampling frequency of 100 Hz. Next, the posterior root of the medial meniscus was completely cut. The loading cycles were repeated with the same parameters and the cartilage deformation recorded. Finally, an anatomic repair of the meniscal root with a transtibial pull-out suture was performed and the loading cycles and measurements repeated again. Results: The deformation of the cartilage increased during the course of cyclic loading independently of the state of the root. However, detachment of the meniscal root increased the magnitude of deformation after 100 cycles significantly (p \ 0.001). Repair of the meniscal root did not restore the magnitude of deformation to normal and was not significantly different from the state of a detached root (p \ 0.24). Conclusions: Although earlier studies suggested that repairing meniscal roots restores the pressure distribution in the respective compartment to normal, our results indicate that this effect is not maintained under low level cyclic loads, at least in the case of an anatomic trans-osseous repair. This implies that either strict unloading is necessary, more stable repair techniques should be sought, or that a certain level of pretension on the repaired root is necessary to account for the loss in function.

P25-1359 Meniscus transplantation and open wedge proximal tibia osteotomy in biologically young patients with varus axis of the knee and chondromalacy L. Pasa1, S. Kalandra1, V. Radek1, M. Kelbl1, J. Kuzma1 1 Trauma Hospital Brno, Dpt. of Traumatol. of Medical Fac Masaryk University Brno, Brno, Czech Republic Objectives: Authors present their experience with deep frozen medial meniscus transplantation in young patients with weight bearing pain and chondromalacy in injured compartment after menisectomy with contemporary varus angulation of the injured knee. Methods: Chondromalacy and pain begins usually within a few years after subtotal menisectomy in injured compartment, especially in varus axis of the injured knee. Meniscus transplantation could improve forces transmission and lubrication in the joint and could help to cartilage healing if it is changed. There is a problem with contemporary varus angulation of the knee and overloading implanted

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 meniscus and cartilage too. Open wedge valgus osteotomy improves condition for meniscus transplant and cartilage healing. From September 2007 to June 2010 authors operated 12 medial meniscus transplantations with contemporary open wedge osteotomy in 12 patients. 7 women and 5 men, at the age 32–42, 5–9 years post menisectomy. 8 patients were with local chondromalacy gr. III sec Outerbridge, treated by microfracture, 4 patients were with chondromalacy gr II. All meniscus transplantations were done arthroscopically, valgus osteotomies were fixed by Tomofix plate with allogenous bone substitution to the osteotomy. Patient0 s outcome evaluations were performed using Lysholm and IKDC scores. Results: All patients were healed without complications. Three months after an operation 10 patients had knee movement S-0-0-120. All patients had no walking pain 4 months after their operation and 8 patients had no pain while jogging 6 months after their operation. Control arthroscopy was made in 6 patients. Meniscus and chondral defect with fibrocartilage were healed in all patients. The mean follow-up was 18.2 months. The mean Lysholm score improved from 64.2 to 88.7. The IKDC subjective knee score improved from 58.2 to 84.4. Joint medial space was enlarge on control X-ray in 9 patients. Conclusions: Meniscus transplantation and valgus proximal tibia osteotomy could be a promising method for improving of the injured knee after medial subtotal menisectomy in young patients with contemporary varus axis.

P25-1367 Postoperative care and rehabilitation in patients after arthroscopic medial meniscus repair with polyurethane implants E. Kurowska1, K. Slynarski2, M. Pia˛tkowski1 1 Sports Medicine Center CMS, Rehabilitation, Warszawa, Poland, 2 Sports Medicine Center CMS, Warszawa, Poland Objectives: The meniscus performs crucial functions within the knee, and its loss frequently leads to osteoarthritis and irreversible joint damage. There are much more study about meniscal transplantation with use scafolds but the rehabilitation of meniscus implantation is not documented in literature. The aim of this study is examine the main rehabilitation components and discuss their practical applications within the overall treatment program. Methods: This study showed outcome of 30 patients after arthroscopic medial meniscus repair with polyurethane implants. They were evaluated with KOOS score, analogue pain score and patient’s subjective assesment. Meniscus implant significantly limited ROM and weight bearing status but rehabilitation program begins immediately after surgery, to achieve full extension, strengthenig of quadriceps muscle and proprioception with care of pain control. Range of motion, weight bearing status and exercises, are adapted individually to the patients condition. Patients had 3–5 rehabilitation session per week. A single session lasted for 2 h. The total duration of rehab was 6 months. Results: All patients significantly improved in pain during the observation period of 12 months and were satisfied with the results. The patients filled in the evaluation forms preoperativly, 3 months after the procedure and then 6 months and finally 1 year postoperatively. Results of KOOS score increased significantly for pain from 64 preoperatively to 90 postoperatively. Their daily life activity has improved from 52 to 96. Similarly the remaining three components of KOOS score also rose:symptoms from 57 to 91; sport 32 from to 89 and quality of life from 45 to 90. Conclusions: From the point of view of patient we need to fasten as soon as possible rehabilitation programme but meniscus implant need gradual progression both ROM and weight bearing status. In our opinion good clinical results allowing patients to return to normal daily life activity without pain and to sports validate proposed surgery followed by personalized rehabilitation.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P25-1423 Simulation training in arthroscopic surgery. A proficiency based training curriculum for arthroscopic skill aquisition R. Vega1, M. In˜iguez2, J. Varas3, O. Padilla4, R. Vega Jr5 1 Universidad Cato´lica de Chile, Orthopaedic Surgery, Santiago, Chile, 2Karolinska Institute & Capio Artro Clinic-Sweden/ Universidad, Orthopaedic surgery/Stockholm Sport Trauma Research Center, Santiago, Chile, 3Universidad Cato´lica de Chile, Simulation Unit, Santiago, Chile, 4Universidad Cato´lica de Chile, Statistics Department, Santiago, Chile, 5Universidad Cato´lica de Chile, Santiago, Chile Objectives: There are many programs to improve arthroscopic skills. Simulator training for arthroscopic surgery may have benefits by enabling surgical trainees to perform basic arthroscopic procedures and gain the necessary skills in a safe environment. Purpose: The aim of this study was to asses the surgical skills before and after an arthroscopic simulation training. Methods: 23 orthopaedic residents, who approved the basic arthroscopic instuctional course, were evaluated performing a single specific procedure (meniscal suture) in an arificial model. Each procedure was recorded on video and then assessed by an expert using a modified and validated global and specific rating scales for objetive stuctured assessment of technical skills (OSATS). In addition, outcomes such operative time, number of attempts and number of divices were evaluated. The evaluation was carried out using the Imperial College Surgical Assessment Device (ICSAD) allowing objetive quantification of movements and traveled path length of each hand (Aggarwal, Dosis et al. 2006). After that, residents were ready for a 10 sessions training program. The curriculum is designed as a progressive cumulative experience, where each individual learns a specific task, repeats it, recives feedback to achieve proficiency, and then continue repeating the task through the 10 sessions. After the training period, they were ready to be re-evaluated. Once again using the ICSAD and recording each procedure on video which are assessed by an expert using OSATS. Statistics assesment was done using the Kruskal-Walllis method and Mann–Whitney. Results: After training the residents improve the OSATS score significantly, from a mean score from 8 to 22 points. The objetive cuantification of hands movement and travel path length with ICSAD, dropped from a mean of 340–120 m after the training period (p \ 0.001). Conclusions: A proficiency based training curriculum for arthroscopic skill aquisition, allows resident surgical skills improvement . This estudy showed improvement of skill levels for basic arthroscopy trained residents.

Knee-Cartilage I

P26-44 A pilot study of the use of an osteochondral scaffold plug for cartilage repair in the knee and how to deal with early clinical failures? A. Dhollander1, K. Liekens1, K.F. Almqvist2, R. Verdonk1, P. Verdonk2 1 Gent University Hospital, Department of Orthopaedic Surgery and Traumatology, Gent, Belgium, 2Gent University Hospital, Orthopaedics, Gent, Belgium Objectives: The purpose of this pilot study is to present our short-term experience with an osteochondral scaffold plug (TruFit plug, Smith & Nephew, Andover, MA) for cartilage repair in the knee, but more importantly to discuss our approach to treat early clinical failures.

S257 Methods: Twenty patients were consecutively treated for their cartilage lesion with this plug technique. These patients were prospectively clinically evaluated at 6 and 12 months of follow-up. Magnetic resonance imaging (MRI) was used for morphological analysis of the cartilage repair. Biopsy samples were taken from 3 cases during revision surgery, allowing histological assesment of the repair tissue. Results: The short-term clinical and MRI outcome of this pilot study are modest. No signs of deterioration of the repair tissue were observed. Three of the 15 patients followed during 1 year (20.0%), displayed persistent or even more clinical symptoms after insertion of the plug. These patients were considered as failures and therefore eligible for revision surgery. During revision surgery the repair tissue was carefully removed. The remaining osteochondral defect was filled with autologous bone grafts. Immediate and persistent relieve of symptoms was observed in all 3 patients. Histological assessment of biopsy specimens taken during revision surgery revealed fibrous vascularized repair tissue with the presence of foreign-body giant cells. Conclusions: The overall short term clinical and MRI outcome of the osteochondral scaffold plug for cartilage repair in the knee is modest. In this pilot study a modest clinical improvement became apparent at 12 months of follow-up. MRI data showed no deterioration of the repair tissue. Three of the 15 patients (20%) had persistent clinical symptoms after surgery. These patients were succesfully treated with removal of the osteochondral plug remnants and the application of autologous bone grafts.

P26-45 Autologous matrix-induced chondrogenesis combined with platelet-rich plasma gel A. Dhollander1, F. De Neve1, K.F. Almqvist2, R. Verdonk1, P. Verdonk2 1 Gent University Hospital, Department of Orthopaedic Surgery and Traumatology, Gent, Belgium, 2Gent University Hospital, Orthopaedics, Gent, Belgium Objectives: This pilot study was designed to describe the technical details and to present the preliminary outcome of autologous matrixinduced chondrogenesis (AMIC) combined with platelet-rich plasma gel, the so called AMIC plus technique, for the treatment of patellar cartilage defects in the knee. Methods: The AMIC plus technique was used for the treatment of (osteo) chondral patellar lesions in the knee. The surgical technique is extensively described. Five patients were clinically prospectively evaluated during 2 years. MRI data were analysed based on the original MOCART (Magnetic Resonance Observation of Cartilage Repair Tissue) and modified MOCART scoring system. Results: A clinical improvement became apparent after 24 months of follow-up. Both MOCART scoring systemsrevealed no significant deterioration or improvement ofthe repair tissue between one and 2 years of follow-up. However, all cases showed subchondral lamina and bone changes. The formation of intralesional osteophytes was observed in 3 of the 5 patients during the 2 years of follow-up. Conclusions: AMIC plus is feasible for the treatment of symptomatic patellar cartilage defects and resulted in a clinical improvement in all patients. The favourable clinical outcome of the AMIC plus technique was not confirmed by the MRI findings.

P26-50 Mid-term results of the treatment of cartilage defects in the knee using alginate beads containing human mature allogenic chondrocytes A. Dhollander1, P. Verdonk2, R. Verdonk1, K.F. Almqvist2

123

S258 1

Gent University Hospital, Department of Orthopaedic Surgery and Traumatology, Gent, Belgium, 2Gent University Hospital, Orthopaedics, Gent, Belgium Objectives: The treatment of chondral lesions is still an important challenge for the orthopaedic surgeon. Attempts have been made to restore cartilage lesions by filling the defects with a temporary biocompatible matrix. The purpose of this paper was to present our midterm experience with the implantation of alginate beads containing human mature allogenic chondrocytes for the treatment of cartilage lesions in the knee. Methods: A biodegradable, alginate-based biocompatible scaffold containing human mature allogenic chondrocytes was used for the treatment of cartilage lesions in the knee. Twenty-one patients were clinically prospectively evaluated with use of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and a Visual Analogue Scale (VAS). The mean follow-up time was 6.3 years (5–8 years). MRI (Magnetic Resonance Imaging) data were analyzed based on the MOCART (Magnetic Resonance Observation of Cartilage Repair Tissue) system, allowing morphological assessment of the repair tissue. MRI images were taken at 1 year of followup and at a mean follow-up of 6.1 years (5–7 years). Results: During the follow-up period the WOMAC and VAS scores improved significantly. No signs of clinical deterioration or adverse reactions to the alginate beads/allogenic chondrocyte implantation were observed. Four failures occured during the follow-up period in this study (19.05%). The MOCART scores were moderate and remained stable in time. Conclusions: This investigation provided useful information on the efficacy of the implantation of alginate beads containing human mature allogenic chondrocytes for the treatment of cartilage lesions in the knee. The mid-term clinical outcome of the presented technique was satisfactory. However, these results were not confirmed by the MRI findings.

P26-172 Long-term benefits comparing intra-articular betamethasone with sodium hyaluronate in patients with knee osteoarthritis K.F. Almqvist1, J. Van Lauwe2, H. Van der Bracht3, R. Verdonk4, F. Luyten5 1 University Hospital Gent, Orthopaedics, Gent, Belgium, 2University Hospitals Leuven, Orthopaedics, Leuven, Belgium, 3University Hospital Gent, Gent, Belgium, 4University Hospital Gent, Department of Orthopaedic Surgery and Traumatology, Gent, Belgium, 5 University Hospitals Leuven, Reumatology, Leuven, Belgium Objectives: The primary objective was to measure in patients with symptomatic knee osteoarthritis the time to reappearance of clinically relevant symptoms with the need of additional treatment comparing two treatment approaches. One treatment arm consisted of three intraarticular injections of sodium hyaluronate (Ostenil) and was compared to the effect of one injection betamethasone (Diprophos) and two sham injections. Methods: Patients with symptomatic primary osteoarthritis of the knee were randomly assigned to either the sodium hyaluronate (Ostenil) or betamethasone (Diprophos) arm in a two-centre, masked-observer, cross-over study. The patients were injected either one of the two medications during a first treatment cycle. Results were evaluated at 1, 12, 24, 36 and 48 weeks post the first cycle of 3 consecutive weekly injections. The cross-over was triggered through request of the patient to initiate a second treatment cycle of injections followed by evaluations at 1, 12, 24, 36 and 48 weeks post the last injection. The second cycle terminated when the patient demanded further treatment. The primary evaluation was the time to request for a second treatment cycle. The secondary evaluation criteria included the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index, visual analogue score (VAS) pain evaluation, clinical

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 assessment and consumption of escape medication and were analyzed in an exploratory manner. All adverse events were recorded. Results: A total of 80 patients (Kellgren grade II or III) were randomly assigned. The time to event was a median of 350 days for the Ostenil group and 339 days for the Diprophos treatment, not statistically different when the two treatments were compared (p = 0.5065). Several of the secondary outcome measures (WOMAC total score, VAS pain and clinical global impression (CGI)), exhibited proven or observed superiority for the Ostenil group. No striking safety problems were reported with either treatment. Conclusions: Intra-articular treatments with sodium hylaruonate or betamethasone are beneficial for the patient with symptomatic osteoarthritis of the knee. In a real life clinical practice setting, time to recurrence of symptoms and need for a new treatment cycle of intraarticular injections appears to be similar for both treatments.

P26-175 Long-term success in the treatment of diffuse pigmented villonodular synovitis of the knee with subtotal synovectomy and radiotherapy L. Carvalho Jr.1, O. Melo-Silva Jr.1, L.F.M. Soares1, E.F. Temponi1, M.B. Jacques Gonc¸alves1, L.P. Costa1 1 Hospital Madre Teresa, Grupo de Cirurgia do Joelho, Belo Horizonte, Brazil Objectives: The objective of the present study was to evaluate the effectiveness and long-term outcome of the treatment of patients suffering from diffuse pigmented villonodular synovitis (PVNS) of the knee with subtotal arthroscopic synovectomy and subsequent external-beam radiotherapy. Methods: Eight patients diagnosed with diffuse PVNS by clinical and magnetic resonance imaging, were treated surgically and followed up for an average of 8.6 years in order to monitor remaining lesions and to detect new occurrences of the condition. Results: None of the subjects presented major postoperative complications, although three patients exhibited late minor complications (peripatellar pain, articular effusion and persistent quadricipital muscle atrophy). Only one patient (12.5%) presented with recurrence of the disease during the follow-up period. Clinical data of patients presenting diffuse PVNS Patient Sex

Age

Knee

(years) joint

Number of

Follow-

recurrences up

Postoperative complications

(years) 1

Female 50

Right 0

14

Peripatellar pain and articular

2

Female 50

Left

0

14

Atrophy of quadricipital muscle

3

Female 32

Right 0

8

4

Male

Right 0

8

5

Female 35

Left

1

8

6

Male

30

Left

0

8

7

Male

27

Right 0

6

8

Female 50

Left

3

effusion

32

0

Atrophy of quadricipital muscle

Conclusions: Subtotal arthroscopic synovectomy with subsequent local external-beam radiotherapy is very effective for the treatment of PVNS, and presented few complications, a low rate of recurrence and good long-term prognosis.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P26-209 Long term management of knee osteoarthritis with hylan GF-20: Efficacy and safety of repeat treatments over 4 years R. Raman1, C. Shaw2, G. Johnson3, A. Dutta2, H. Sharma2 1 Hull Royal Infirmary, Trauma and Orthopaedics, Swanland, United Kingdom, 2Hull Royal Infirmary, Hull, United Kingdom, 3Academic Departmentof Orthopaedics, Hull Royal Infirmary, Normanton, United Kingdom Objectives: OA is a chronic disease and its management is a constant process. Pain and symptom relief from viscosupplementation is variable ranging from 3 to 12 months necessitating repeat courses of treatment. The aim of this study was to assess the safety and efficacy of multiple repeat intra-articular injections of hylan GF-20 (Synvisc) in the treatment of osteoarthritis of the knee. Methods: Patients with OA of the knee were offered initial treatment with hylan GF-20 and repeat courses after a minimum of 6 months from the previous injection (s). The inclusion criteria was pain score of [6 on a VAS (0–10) in the target knee. Patients with at least 2 treatment courses were included in the study. This is a prospective, longitudinal independent study over 6 years in the same institution, where all patients were reviewed by blinded assessors at pre injection, 1 week, 6 weeks, 3, 6 months after every treatment course. The primary outcome variable was knee pain on VAS at 6 months. Secondary outcome measures were WOMAC, Oxford knee score and SF-12. All adverse events (AE) were recorded. Results: From our arthritis database, we identified 1,103 patients who had repeat treatments (2–8 courses, median-3) with hylan GF-20. The mean time to repeat treatment was 45.6 weeks (27–104, weeks). Knee pain on VAS improved from 6.6 to 3.7 at 6 months (p = 0.02) over all courses. Mean improvement in knee pain was 51% after the first repeat course and 49% at the last repeat course. Significant improvements from the baseline in the WOMAC pain and function subscales and Oxford knee scores were observed during all repeat courses. Overall incidence of AE was 13.4% (11.9% in initial course) from the pooled data. The incidence of AE had no correlation to the number of repeat treatments. Conclusions: Pain relief and improvement of function are consistently observed following repeat treatment of symptomatic OA of the knee with hylan G-F 20. Longevity and magnitude of symptom control are similar to the first course of treatment. Repeat courses are well tolerated with low adverse events. This study demonstrates that viscosupplementation with hylan GF-20 is an invaluable tool in the multimodal treatment and forms part of the armamentarium of OA management.

P26-220 How to treat osteocondritis dissecans of the knee: surgical techniques and new trends E. Kon1, F. Vannini2, R. Buda2, G. Filardo3, M. Marcacci4, S. Giannini2 1 IOR, IX Divisione-Biomechanics Lab, Bologna, Italy, 2Istituto Ortopedico Rizzoli, Bologna University, VI Department of Orthopaedics and Traumatology, Bologna, Italy, 3Rizzoli Orthopaedic Institute, Biomechanics Lab., Bologna, Italy, 4Instituto Ortopedico Rizzoli Bologna, Bologna, Italy Objectives: Osteocondritis dissecans (OCD) is a relatively common cause of knee pain. Ideal treatment is still controversial. Aim of this exhibit is to describe the outcomes of 5 different surgical techniques in a series of 63 patients. Methods: 63 patients (age 22.5 ± 7.4 years) affected by OCD of the femoral condyle (45 medial and 17 lateral) were treated by either osteochondral autologous transplantation, autologous chondrocyte implantation with bone graft, biomimetic nanostructured osteochondral scaffold (Maioregen) implantation, bone-cartilage paste graft or

S259 bone marrow derived cells transplantation ‘‘one-step’’ technique. Patient evaluation included IKDC score, eq-vas score, X-Rays and MRI preoperatively and at follow-up. Results: Global mean IKDC improved from pre-operative 40.1 ± 14.6 to 77.2 ± 21.3 (p \ 0.0005) at mean 5.3 ± 4.7 years follow-up, while eq-vas improved from 51.7 ± 17.0 to 83.5 ± 18.3 (p \ 0.0005). No influence of age, size of the lesion, length of followup and associated surgeries on the result was found. No differences were found between the results obtained with different surgeries except a slight tendency of better improvement in the result following autologous chondrocyte implantation (p \ 0.01). Control MRI evidenced a satisfactory repair of cartilaginous layer and subchondral bone. Conclusions: The techniques described were effective in providing good clinical and radiographic results in the treatment of OCD and confirmed the validity of autologous chondrocyte implantation over time. Newer techniques such as Maioregen implantation and ‘‘onestep’’ base on different rationales, the first relying on the characteristics of the scaffold and the second on the regenerative potential of mesenchymal cells. Both of them have the advantages to be minimally invasive surgeries and to require a single operation.

P26-240 Does the mechanical alignment correlate with the tracer uptake pattern and intensity in SPECT/CT? A retrospective series on 104 knees M.T. Hirschmann1, S.N. Schoen1, F.K. Afifi1, H. Rasch2, M.P. Arnold1, N.F. Friederich1 1 Kantonsspital Bruderholz, Klinik f. Orthop. Chirurgie & Traumatologie, Bruderholz, Switzerland, 2Kantonsspital Bruderholz, Institut fu¨r Radiologie und Nuklearmedizin, Bruderholz, Switzerland Objectives: SPECT/CT is a promising imaging technology, which promises the combined evaluation of mechanical and biological information. If the loading pattern and intensity values of the SPECT/ CT correlate with the anatomical and mechanical alignment it could be used for evaluation of postoperative patients after realignment procedures such as high tibial osteotomies or patellofemoral surgery. Methods: 99mTc-HDP-SPECT/CT of 76 patients (mean age 49 ± 17) and 104 knees were analysed using a previously validated localisation and grading algorithm. The maximum intensity in each femoral, tibial and patellar joint compartment (medial, lateral, central, superior, inferior) was noted (0–10). In addition, anterior-posterior and lateral weight bearing radiographs as well as Rosenberg and skyline view were analysed with regards to the Kellgren-Lawrence osteoarthritis score. Long leg radiographs were used to assess the mechanical and anatomical leg alignment, which was then classified as varus, valgus or neutral. We correlated the mechanical and antomical alignment with the intensity of tracer uptake in each area of interest. In addition, the Kellgren Lawrence socre was also correlated. A sample size calculation was performed. The level of statistical significance was p \ 0.05. Results: The intensity of 99mTc-HDP tracer uptake on the medial compartment significantly correlated with anatomical and mechanical varus alignment of the knee (p \ 0.05). The intensity of 99mTc-HDP tracer uptake on the lateral compartment significantly correlated with anatomical and mechanical valgus alignment of the knee (p \ 0.05). In patients having higher Kellgren Lawrence scores, which reflects a higher degree of osteoarthritis, higher tracer uptake values were found in the corresponding joint compartments. Conclusions: SPECT/CT reflects the loading pattern of the knee joint with regards to the mechanical and anatomical alignment. SPECT/CT is then a promising imaging technology in particular for follow-up of hight tibial osteotomies, deloader braces treatment and patellofemoral realignment procedures.

123

S260 P26-289 Isokinetic knee extensor strength deficit following matrix-induced autologous chondrocyte implantation J. Ebert1, D. Lloyd1, T. Ackland1, D. Wood2 1 The University of Western Australia, School of Sport Science, Exercise and Health, Perth, Australia, 2The University of Western Australia, School of Surgery, Perth, Australia Objectives: Matrix-induced autologous chondrocyte implantation (MACI) has become an established technique for the repair of full thickness chondral defects in the knee. However, despite the reported improvement in knee pain and symptoms, as well as the successful regeneration of hyaline-like repair tissue, little has been reported on the recovery of knee strength and function. The aim of this study was to investigate isokinetic knee strength in patients at 5 years following MACI. Methods: A total of 65 patients who underwent MACI to the medial or lateral femoral condyle underwent strength testing at 5 years postsurgery. Using an isokinetic dynamometer, and during isokinetic knee extension and flexion angular velocities of 60, 90 and 120/s, the peak torque (PT), knee range of motion at PT, PT at 45 of knee flexion and hamstrings/quadriceps ratio was obtained, in both the operated and non-operated sides. Pain at the time of assessment using the Knee Injury and Osteoarthritis Outcome Score (KOOS) and visual analogue pain scale (VAS) was also taken. Results: While the peak knee extensor torque was greater in the nonoperated leg, when compared to the operated leg at all angular velocities (60, 90 and 120/s), these differences were not significant (p [ 0.05). There were no significant differences (p [ 0.05) between the operated and non-operated legs in the peak knee flexor torque or knee flexor torque at a knee flexion angle of 45, at all angular velocities. However, a significantly greater (p \ 0.05) knee extensor torque at a knee flexion angle of 45, was observed at all speeds. Across all three angular velocities, patients produced a peak knee flexor torque on their operated legs within 96–101% of their nonoperated side. This difference was 86–87% for their peak knee extensor torque. The peak knee extensor torque parameters exhibited a significant positive correlation with body weight and height, and a significant negative correlation with age. There were no significant correlations between knee pain and any of the torque values. Conclusions: While patients had recovered their knee flexor strength, they still demonstrated a reduced knee strength extensor profile at 5 years following MACI, unrelated to the speed of isokinetic testing. There was no correlation between knee pain and any of the torque values, indicating pain did not contribute to the observed knee extensor torque differences. This demonstrates that the early supervised rehabilitation phase following MACI is not sufficient to restore long-term knee strength, and ongoing patient advice and rehabilitation is required extending beyond this early post-operative supervised care. It is yet unknown how this prolonged reduction in knee strength and subsequent joint support may affect long-term graft outcome.

P26-290 Clinical and radiological outcome following matrix-induced autologous chondrocyte implantation in the patellofemoral joint: 5 year results J. Ebert1, D. Meyerkort2, M. Fallon3, T. Ackland1, M.H. Zheng2, D. Wood2 1 The University of Western Australia, School of Sport Science, Exercise and Health, Perth, Australia, 2The University of Western Australia, School of Surgery, Perth, Australia, 3Perth Radiological Clinic, Perth, Australia Objectives: This prospective study presents radiological and clinical outcome to 5 years post surgery, for a consecutive series of patients following matrix-induced autologous chondrocyte implantation

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 (MACI), to evaluate whether MACI provides a suitable mid-term treatment option for articular cartilage defects in the patellofemoral joint of the knee. Methods: A prospective study design was used to assess clinical and radiological outcome in 23 patients (24 grafts) to 5 years following MACI for the treatment of symptomatic full thickness lesions of the patella or trochlea groove. Nine of these patients underwent adjunct tibial tubercle transfer (TTT) to address patellofemoral malalignment, defined as [0.9 cm of lateralization of the tibial tubercle. Following surgery, patients underwent a structured, supervised rehabilitation program of progressive exercise and knee joint range of motion to protect, and then stimulate the healing process. Clinical outcomes were measured using the Knee Injury and Osteoarthritis Outcome Score (KOOS), the Short-Form Health Survey (SF-36), the 6-min walk test and knee range of motion. High resolution magnetic resonance imaging (MRI) was undertaken at 3 months, 1, 2 and 5 years post-surgery, to describe the quality and quantity of repair tissue. Results: Patients demonstrated an increased MRI composite score over time that improved significantly (p \ 0.0001) from 3 months to 5 years post-surgery. Patients demonstrated a significant improvement (p \ 0.05) throughout the post-operative timeline for all five subscales of the KOOS, the physical component scale of the SF-36 and the 6-min walk test. There were no significant differences (p0.05) in any of the MRI scoring parameters or clinical measures at 5 years, between those with MACI alone and MACI in combination with TTT. Conclusions: The outcomes from these 5 year MRI results indicate that MACI grafting to the patellofemoral joint can result in a durable graft with associated marked improvement in pain and symptoms. The MRI data provides an accurate assessment of the graft status which is not confounded from the results of TTT, which alone can affect clinical results.

P26-424 Membrane-guided regeneration augmented with bone marrow concentrate (BMC) for cartilage repair in the knee. Histological results D. Enea1, S. Calcagno2, B. Alberto1, S. Cecconi1, S. Manzotti1, A. Gigante1 1 Polytechnic University of Marche, Orthopedics, Ancona, Italy, 2 Rapallo Hospital, Orthopedics, Rapallo, Italy Objectives: The treatment of knee chondral lesions is still an important challenge for the orthopaedic surgeon. One-step cartilage repair procedures adopting microfractures coupled with a collagen membrane are increasingly performed in an attempt to recreate native articular cartilage. However, to date, only a little is known about histological outcomes of such techniques. This study aimed to evaluate the quality of the repair tissue obtained from biopsies harvested during second-look arthroscopy following arthroscopic membraneguided regeneration (MeRG) augmented with bone marrow concentrate (BMC). Methods: Ten consecutive patients affected by a focal isolated cartilage lesion in the knee underwent arthroscopic microfractures and the implant of a collagen membrane (MeRG) augmented with autologous BMC obtained from the iliac crest. After full informed consent had been given, second-look arthroscopy was performed and core biopsy was harvested at an average of 12 months follow-up. At the time of biopsy the surgeon reported on patient’s functional status and on the quality of the repair tissue using the standard ICRS Cartilage Repair Assessment (CRA). Every biopsy underwent blind histological evaluation according to the ICRS II Visual Histological Assessment Scale. Results: Patients (mean age 40.7 years) had isolated lesions (mean size was 3.5 cm2) at the medial femoral condyle. Nine patients were asymptomatic and one patient was in pain at the time of biopsy. At

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 arthroscopic evaluation 8 implants appeared nearly normal and 2 implants appeared abnormal according the ICRS CRA. The mean overall ICRS II histological score was 55.6 ± 15.3. Predominantly hyaline matrix was found in 2 cases, a mixture of hyaline/fibrocartilage was found in 2 cases and fibrocartilage was found in 6 cases. When the hyaline/fibrocartilage mixture was observed, hyaline-like cartilage was found next to the osteochondral junction and fibrocartilage was found on the articular side of the bioptic cylinder. Conclusions: Our data suggest that at an average of 12 months follow up, BMC-augmented MeRG yielded a nearly normal arthroscopic appearance of the implants. Even though, on average, a satisfactory quality of repair tissue was obtained (ICRS II overall), the potential of this technique to recreate hyaline-like matrix was still limited at 12 months follow up. We hypothesize that the biopsies with a mixture of hyaline-like cartilage and fibrocartilage may represent an example of cartilage bottom-to-top maturation. Further studies with longer follow-up are needed to understand the true potential of one-step procedures in the repair of focal chondral lesions in the knee.

P26-485 Autologous chondrocyte implantation on the femoral condyles: long term clinical and radiographic results M. Drobnic1, D. Martincic1, D. Radosavljevic1 1 University Medical Centre Ljubljana, Department of Orthopaedic Surgery, Ljubljana, Slovenia Objectives: A prospective non-randomized study analyzed clinical and radiographic results in a cohort of patients operated on the femoral condyles with the periosteum autologous chondrocyte implantation (ACI) over 10 years ago. Methods: 31 out of the 45 patients in the initial cohort (3 failures, 9 nonresponders, 2 other diseases or injuries) were available for a continuous clinical (Lyshom/Tegner, IKDC, KOOS) and radiographic (KellgrenLawrence) follow-up at 0, 2, 5, and 10 years after the ACI procedure on the femoral condyles. The patients were further classified into focal lesions (FL)—10, osteochondritis dissecans (OCD)—12, and lesions with simultaneous ACL reconstruction (ACL)—9 subgroups. Results: Lysholm and IKCD subjective scores revealed stable results over the period from 2 to 10 years with a significant improvement toward the pre-operative levels. Tegner activity levels followed the same pattern, but the patients had not reached their pre-injury levels. KOOS profile at 10 years was: pain 78.6, symptoms 78.1, activities of daily living 82.5, sports 56.9, quality of life 55.1. A 10-year IKDC knee examination classified knees as: 14 normal, 10 nearly normal, 5 abnormal, and 2 severely abnormal. Kellgren-Lawrence scores of 2 and above were found in 10 patients (5 FL, 0 OCD, 5 ACL). Conclusions: Clinical results demonstrated stable performance of operated knees over 10 years. High incidence of knee osteoarthritis in ACL subgroup and low incidence in OCD indicate that ACI works best for localized low-impact cartilage lesions in young patients.

P26-608 Comparison of ‘‘autologous chondrocytes in an agarose and alginate gel’’ implantation versus Mosaicplasty: 1 year follow-up of a phase III clinical trial J.-F. Potel1, H. Robert2, C. Bussie`re3, E. Servien4, P. Neyearset4, F. Dubrana5 1 Medipole Garonne, Toulouse, France, 2Centre Hospitalier North Mayenne, Department of Orthopaedic Surgery, Mayenne, France,

S261 3

Centre Orthope´dique Me´dico-Chirurgical de Dracy-le-Fort, Dracyle-Fort, France, 4Hoˆpital de la Croix-Rousse, Centre Albert Trillat, Department of Orthopaedic Surgery, Lyon, France, 5Hoˆpital Cavale Blanche, Orthope´die, Brest, France Objectives: In 2007, a Phase III clinical trial was initiated to assess the efficacy of CARTIPATCH, autologous chondrocytes implants containing an agarose/alginate gel- compared to mosaicplasty as a firstline treatment for symptomatic osteocartilaginous femoral condyle defects. The present study provides the 12-months follow-up preliminary results of this 24-month long trial. Methods: Inclusions in this multicenter, prospective, randomized, controlled trial versus Mosaicplasty started in April 2007 and ended in April 2010. Fifty-seven patients were included and underwent knee surgery, 30 were implanted with CARTIPATCH and 27 underwent a Mosaicplasty procedure. Efficacy of the treatment was investigated through analysis of the mean subjective IKDC scores (primary endpoint of the trial). Results: Both CARTIPATCH- and Mosaicplasty-treated groups had comparable inclusion characteristics (mean subjective IKDC score, defect size and age). Knee function improvement was assessed by sorting subjective IKDC scores in 2 subgroups, depending on the size of the osteochondral defect: patients presenting a defect B3.5 cm2 or [3.5 cm2. Figure 1 summarizes the mean IKDC scores of both groups depending on the defect size. Patients treated with CARTIPATCH (n = 7) and presenting lesions B3.5 cm2 showed better improvement of their knee function (+39 points, p \ 0.005) as compared to Mosaicplasty-treated patients (n = 7; +16 points). For lesions exceeding 3.5 cm2, both arms gave similar results with a mean IKDC improvement of +35 points (n = 7 for both arms). In terms of treatment success, one failure was observed for a Mosaicplasty-treated patient, while all CARTIPATCH-treated patients showed a significant improvement of their knee function (i.e. improvement of their IKDC [10 points). Conclusions: The 1-year follow-up preliminary results from this phase III multicenter controlled study show a therapeutic effect of CARTIPATCH in the treatment of cartilage defects with a traumatic or pathological etiology. While both techniques resulted in a similar improvement of knee function for patients with rather large lesions, a clear clinical benefit of autologous chondrocytes in a gel implantation over Mosaicplasty for patients with smaller lesions was shown. We are looking forward to completing these preliminary results with the 24-months follow-up data and arthroscopic- histological analyses to fully assess the clinical benefits of CARTIPATCH.

Fig. 1 Subjective IKDC scores versus defect size

123

S262 P26-671 A pilot study of the use of a contoured articular prosthetic device as a salvage procedure for the treatment of focal cartilage defects in the knee A. Dhollander1, P.-J. Vandekerckhove2, K. Moens2, P. Verdonk3, K.F. Almqvist4 1 Gent University Hospital, Orthopaedic Surgery and Traumatology, Gent, Belgium, 2Gent University Hospital, Gent, Belgium, 3Gent University Hospital, Orthopaedic Surgery, Gent, Belgium, 4Gent University Hospital, Orthopaedic, Gent, Belgium Objectives: To evaluate short-term clinical and radiological outcome of a contoured articular prosthetic device (HemiCAP) for the treatment of cartilage defects in the knee in challenging patient cohort. Methods: Fourteen patients aged 29–51 years with a focal cartilage defect (grade III/IV) were treated with this device (HemiCAP, Arthrosurface Inc., Franklin, MA, USA) and prospectively followed during 2 years. All patients included in this pilot study underwent previous surgery (failed cartilage repair procedures, partial meniscectomies,…). Clinical outcome was measured by the KOOS, the Tegner activity scale and the VAS for pain. Serial radiographs were taken at 1 week, 6 months, 12 months and 24 months of follow-up. Results: The patients included in this study showed a significant clinical improvement after the procedure. The analysis of the serial radiographs of this pilot study showed no osteoarthritic changes in the affected knee. No signs of loosening of the device were observed. No failures occurred among the 14 patients until now. Conclusions: This pilot investigation provided useful information on the safety and efficacy of this device in a challenging patient cohort. The short-term clinical and radiographical outcome are promising. Large scale trials are mandatory to confirm the results and the reliability of this device.

P26-680 Comparison of chondrotoxic effects of anaesthetics for arthroscopic application as single shot or repetitive use M. Herten1, J. Kircher1, I. Ickert1, M. Vogl1, T. Patzer1, R. Krauspe1 1 Heinrich-Heine University, Orthopaedic Department, Du¨sseldorf, Germany Objectives: Local anaesthetics (LA) are frequently used in arthroscopic practice for diagnostic and therapeutic purposes. The intraarticular application ranges from single shot to repetitive use. The fulminant effect of chondrolysis leading to severe consequences has been shown in several case studies after the use of LA of amidtype in pain pumps [1–3]. This leads to the necessity of the re-evaluation of LA in routine use in regard to chondrocytes viability [4]. As an alternative opioids have been proven effective and suitable for clinical application [5]. Possible side effects of Morphine in clinical application (e.g. nausea) seem to be reduced using its metabolic form, Morphine 6-Glucuronide (M6G). In the present study amid-type LA, Morphine and M6G are compared with NaCl control regarding potential harmful effects on human chondrocytes concerning cell viability. The test arrangement includes the effect of time and concentration in different clinical settings. Methods: Nine explants of hyaline cartilage were cultivated and cells were harvested and transferred into alginate bead matrixes. Cells were characterized via Safranine-O and MMB staining. After 16 days of culture the alginate beads were incubated with LA (Bupivacain 0.5 mg/ ml, Ropivacaine 0.75 mg/ml), Morphine 0.5 mg/ml and M6G 0.5 mg/ml or NaCl 0.9% with different exposition times (15, 60, 240 min) and in full, half (‘) and one quarter ()) concentration-dependencies. At day 3 the ATP concentration for cell viability was measured. Results: In single shot (15 min) applications, only Bupivacaine showed a concentration dependent, highly significant reduction

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 (-83%) of cell viability while Ropivacaine and the opioid derivates did not decrease cell viability in any concentration. To explore the long lasting effect the cells were incubated with the substances in full concentration for up to 240 min. Using Bupivacaine no vital cells were detectable at all whereas Ropivacaine showed a significant reduction (-70%) in cell viability after 240 min. In contrast to this both opioids tested did not alter the viability of the chondrocytes at all even after long-time exposition. Conclusions: The present in vitro simulation of a clinically relevant single shot and long lasting application demonstrated the cytotoxic effect of local anaesthetics of amid type on human chondrocytes. These data suggest a very limited intraarticular usage of amide-type local anaesthetics for special indications only. Ropivacaine therefore should be preferred to Bupivacaine and only in low dosage ‘,  and in single shot. For continuous use in pain pumps or repetitive use Morphine or M6G seems to be the better alternative. Alternative opioid drugs are under investigation for intraarticular application in routine practice. References 1. Dragoo J L et al. Am J Sports Med, 36(8):1484–8, 2008. 2. Piper, SL et al. J Bone Joint Surg Am, 90(5):986–91, 2008. 3. Lo IK et al. Arthroscopy, 25(7):707–15, 2009. 4. Kircher J et al. P67, AGA-Congress 2011, p.143 5. Kalso E et al. Pain, 98(3):269–75, 2002

P26-697 Long- term follow- up study of autologous chondrocyte implantation and survivorship analysis A. von Keudell1, T. Bryant2, T. Minas1 1 Brigham and Women’s Hospital/Harvard Medical School, Boston, United States, 2Brigham and Women’s Hospital, Boston, United States Objectives: Autologous chondrocyte implantation has proven to have good to excellent results in about 80% of patients under 55 years up to 10 years. There is only scarce literature in cartilage restoration after 10 years, especially for autologous chondrocyte implantation. Methods: 210 patients (238 knees, RK = 128, LK = 110, ØBMI = 27 kg/m2) were treated with autologous chondrocyte implantation between 1994-2001 for cartilage lesions (Outerbridge 4) in the knee. The mean age at surgery was 36 ± 9 and total defect size measured 8.2 ± 5.3cm2. High tibial osteotomy (HTO) was performed in 51, Tibial Tubercle Osteotomy (TTO) in 69 (44 Fulkerson, 25 McKay), combined HTO/TTO in 16 and distal femoral varus osteotomy in 2 patients. ACL was repaired in 10, LCL in 1, and combined ACL/LCL in another patient. Modified Cincinnati was collected pre- and postoperatively at the last follow-up. Survival analysis was completed using Kaplan-Meier curve. Results: Out of the 210 patients who were treated for cartilage defects, 205 patients were available for postoperative evaluation. The defect type classified as simple (type 1, single unipolar grade 3/4 lesion on femur or grade 2 or less on the tibia or patella) in 18, complex (type 2, multifocal unipolar grade 3/4 chondral lesions on femur, concurrent HTO/TTO, OCD, unipolar lesions on tibia or patella) in 113 and salvage (type 3, Bipolar focal chondral lesions, generalized chondromalacia grade 2 or greater) in 107 patients. An average of 1.7 defects/patient were treated. After a mean of 148 ± 19 months, the self reported modified Cincinnati increased from an average 3.7 ± 1.4–5.6 ± 1.8 (p \ 0.0001) and clinical documented modified Cincinnati from 3.7 ± 1.4–5.6 ± 1.9 (p \ 0.0001). At 16 years follow-up 62 (26%) patients had at least one failed autologous chondrocyte graft. Failure was defined as revision cartilage repair procedure for a minimum of one lesion in 31, partial or total knee replacement in 23 patients. 4 patients were lost to follow-up and one

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 declined further treatment. Kaplan–Meier survival demonstrated significantly different chances for survival between simple, complex, and salvage cases, respectively. Conclusions: Up to 16 years, autologous chondrocyte implantation provides a durable repair of cartilage in patients for the treatment cartilage defects in the knee. Our results advocate good to excellent results in clinical assays after a minimum follow-up of 10 years and can substantially delay the need for prosthesis in the young.

P26-751 Knee arthroscopy on the internet: How well informed are your patients? A. Prinja1, J. Neffendorf2 1 Bedford Hospital NHS Trust, Bedford, United Kingdom, 2 Stoke Mandeville Hospital, Aylesbury, United Kingdom Objectives: By 2001, 52 million adults had used the internet to obtain medical information. A systematic review found that quality was a problem in 70% of health websites. This study assesses the quality of medical websites with information on knee arthroscopy. Methods: We searched the keywords ‘‘knee arthroscopy’’ (English Language, exact phrase setting) in the three most popular search engines: Google, Yahoo and MSN/Bing. The top 50 websites were evaluated from each search engine. Exclusion criteria were irrelevant information, repetition or inaccessibility of the website. Readability was assessed using the Gunning-Fog Index (GFI, measure of years of schooling needed to understand content) and the Flesch Reading Ease Score (FRES, index rating - score/100). We then used the LIDA tool (an online validation instrument of medical websites) to assess the accessibility, usability and reliability of the websites. Results: 49 appropriate websites were analysed out of 150. Websites were excluded due to irrelevant information (30), repetition (60) and inaccessibility (11). The mean GFI result showed the average website was similar to reading the Wall Street Journal (mean GFI = 12.18, SD 1.90). The mean FRES was 52.18 (SD 10.5), which is below the universally recommended target of 60–70. The mean results of the LIDA medical website validation tool were accessibility 81.20%, usability 54.86% and reliability 39.67%. Conclusions: We found readability and reliability of the websites was variable and generally very poor. The best resources are those belonging to recognised medical and academic institutions, as well as those without financial interests in the field. In conclusion, since patients are likely to be heavily influenced by what they read on the internet, it is essential that we guide patients by identifying reliable sources of information.

P26-857 Entrapment of TGF-b1 in photo-responsive hydrogels stimulated chondrocytes differentiation in bioreactor cultivation L.-Y. Chen1, C.-H. Chen2 1 Keelung Chang Gung Memorial Hospital, Orthopaedic Department, Keelung City, Taiwan, Republic of China, 2Keelung Chang Gung Memorial Hospital, Keelung, Taiwan, Republic of China Objectives: The use of autologous chondrocytes implantation and its further development combining autologous chondrocytes with bioresorbable matrices may represent a promising new technology for cartilage regeneration in orthopaedic research. Photopolymerized hydrogels are being investigated for a number of medical applications because of fast curing rates, spatial and temporal control and minimal

S263 heat production. A photopolymerizing hydrogel system provides an efficient method to encapsulate cells. Chondrogenesis of rabbit chondrocytes encapsulated in a photo-responsive hydrogels, poly (ethylene glycol) diacrylate (PEGDA) was studied in the presence and absence of transforming growth factor beta-1(TGF-b1) to better understand the role of the cell-polymer gel constructs in rotating bioreactor cultivation. Methods: Chondrocytes were isolated from the articular cartilage of the knee of a rabbit using the collagenase digestion method. Chondrocytes were expanded in vitro and encapsulated three-dimensionally in semi-interpenetrating networks with or without TGF-b1 using a photopolymerization process. The capacity of dedifferentiated chondrocytes to re-differentiate was evaluated over an 8-week period of dynamic culture in rotating bioreactor. Proliferation of, and matrix production by, chondrocytes were assessed by determination of the DNA and glycosaminoglycan (GAG) content, respectively. Chondrogenic differentiation was evaluated by histological, immunostaining and RNA analyses for the expression of cartilage extracellular matrix (ECM) components. Results: To increase cell survival, TGF-b1 was incorporated into gels. A concentration of 500-ng/ml TGF-b1 incorporated into the network maintained higher cell viability in cultures. DNA assay revealed that higher relative cellular number was acquired in the combinations of TGF-b1 indicating proliferation of chondrocytes. Under this condition, cell-polymer gel constructs also show a comparable increase in the total amount of retained GAGs, indicating synthesis of proteoglycans. Further studies demonstrated that 500-ng/ml TGF-b1 chondrogenic cultures had greater gene expression for aggrecan and collagen II. Conclusions: Incorporation of this growth factor not only allows for sustained viability, but also contributes to initiating chondrogenesis. Synthetic-biological (PEGDA-TGF-b1) composites therefore demonstrated the ability to support chondrocytes function and affect cell response to cartilage extracellular matrix in bioreactor cultivation.

P26-891 Clinical results of the arthroscopic treatment in knee chondral defects- a 7 years clinicasl follow-up B. Tiberiu1, R. Opris2, S. Bataga2, B. Voicu2, R. Melinte2, A. Ivanescu2 1 University of Medicine Tg-Mures, Orthopedic and Sport Traumatology, Tg-Mures, Romania, 2University Emmergency Hospital Tg-Mures, Tg-Mures, Romania Objectives: We describe the clinical results of the mosaicplasty technique in the knee chondral defects. Methods: Between April 2004 and April 2011, 267 patients underwent mosaicplasty for knee chondral defects. The mean age of the patients was 3 years and 187 patients were male and 80 patients were female. All the parients related pain sweelling and functional disconfort of the affected knee and had a medium level of sports activity. All the patie4nts underwent arthroscopic surgery and mosdaicplasty. 167 patients of the patients had OCD and 100 patients had chondral lesions of traumatic origin relates to sports activity. All had Outerbridge lesions typ 3 or 4, with mean size 3 cm2.73% of the lesions were located to MFC,21% of the LCF and 6% to the patella. Results: In the first 2 months all the patients related swelling, after 4 months one of five cases eas notice and at 12 months none had swelling or pain. The second look were performed after 1 years, and showe a good chondral integration of the bone grafts. The ROM was

123

S264

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

normal for all the patients, and 12 months later 87% were pain free and returne to the anterior sports activity. In every years the clinical results were evaluated using the Tegner Lysholm scale and showe in 85% excellent results, in 12 cases good results and in 3% fair results. Conclusions: The mosaicplasty hade excellent results in chondral defects smaller than 8 cm2. The disadvantage of donor site morbidity seems to be well compensated to all patients. The application of this method is generally accepted as a joint salvage procedure, both patient and the surgeon, before make other surgical procedures.

P26-910 G-CSF and cartilage fragments. A potential new strategy for increasing chondrocyte outgrowth for one-stage repair. In vitro study A. Marmotti1, G. Peretti2, M. Bruzzone3, D.E. Bonasia4, R. Rossi5, F. Castoldi6 1 University of Torino, Mauriziano Hospital, Department of Orthopaedics, Torino, Italy, 2San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy, 3 University of Torino, Mauriziano ‘Umberto I’ Hospital, Department of Orthopaedics and Traumatology, Torino, Italy, 4University of Torino, Torino, Italy, 5University of Torino, Mauriziano Umberto I, Torino, Italy, 6Mauriziano Hospital, University of Torino, Department of Orthopaedics and Traumatology, Mauriziano Hospital Turin, Torino, Italy Objectives: Aim of the study was to verify the capability of G-CSF exposure on enhancing chondrocyte outgrowth into a hyaluronic acid derivative (HA- HYAFF-11/fibrin glue-Tisseel/platelet rich plasma (PRP) scaffold and to evaluate its effects on chondrocyte behaviour, compared to TGF-beta exposure. Methods: Cartilage fragments construct preparation: articular cartilage from 23 human knees (\ 35y) was harvested and minced into small fragments and loiaded onto the scaffolds. Constructs were cultured for 1 month both in standard culture medium and under exposure to G-CSF (10 ng/ml) and/or to TGF-beta (10 ng/ml). Constructs were evaluated histologically and with immunofluorescence. Immunofluorescence was also performed on cells which were migrated from the constructs and found adherent to the plastic surface at 1 month. Results: Compared to unstimulated cultures, chondrocyte outgrowth into the scaffold at 1 month was greater when exposed to G-CSF and/ or TGF-beta (p \ 0.05) (Fig. 1). No statistical differences were observed between exposure to G-CSF or TGF-beta or to both factors (p [ 0.05). Immunofluorescence of migrating cells inside the constructs was positive for sox9, CD151, CD49c and G-CSF Receptor, and was negative for CD105, consistent with a predominant chondrogenic phenotype (Fig. 2). In cells adherent to the plastic surface after migration from constructs, the exposure to G-CSF slightly decreased SOX-9 expression and increased PCNA and beta-catenin expression (Fig. 3). Conclusions: G-CSF exposure improves chondrocytes outgrowth from human cartilage fragments loaded into a HA/fibrin glue/PRP scaffold. The changes in expression of Sox-9 (slightly decreased) and PCNA and beta-catenin (increased) under G-CSF exposure suggest a proliferative attitude of cells. This supports a potential role of G-CSF in increasing chondrocyte outgrowth during in vivo one stage cartilage repair with minced human cartilage fragments. We gratefully acknowledge prof. C. Tarella for the precious suggestions in the planning and completion of this work.

123

Fig. 1 Mean number of migrating cells (and SD) at 1 m

Fig. 2 Immunephenotype of cells into the scaffold

Fig. 3 Immunophenotype of cells on plastic

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P26-915 Chondrogenic potential of haemarthrosis-derived MSC for cartilage repair after abrasion arthroplasty C. Hartz1, F. Pries2, S. Lippross1, T. Pufe3, A. Seekamp1, D. Varoga1 1 Universita¨tsklinikum Schleswig–Holstein, Campus Kiel, Klinik fu¨r Unfallchirurgie, Kiel, Germany, 2MARE Klinikum, Department fu¨r Arthroskopische Chirurgie und Sporttraumatolo, Kiel, Germany, 3 RWTH Aachen, Department of Anatomy and Cell Biology, Aachen, Germany Objectives: Mesenchymal stem cells (MSC) from bone marrow are a very attractive tool in the context of repair and regeneration of cartilage lesion. Arthroscopic treatment of OA includes bone marrow stimulation technique such as abrasion arthroplasty (AAP) and microfracturing (MF). These procedures are proved to stimulate cartilage regeneration. Recently we have shown the release of growth factors such as insulin-like growth factor-1 (IGF-1) and transforming growth factor beta-1 (TGF-b1) in the postoperative haemarthrosis depending on the choice of arthroscopic procedure. They play a pivotal role in the regeneration process of chondral defects and chondrogenic differentiation of MSC. The aim of the current study was to characterize the mononuclear cells after bone marrow stimulation techniques and to determine their regeneratory potential. Methods: Haemarthrosis was collected from the drainage bottle 22 h (h) (n = 164) and 1 week (n = 10) after different arthroscopic knee procedures. Mononuclear cells were isolated by ficoll density gradient centrifugation. Adherent cells were were characterized using fluorescence-activated cell-sorting (FACS) analysis, immunohistochemistry (IHC) and immunofluorescence for characteristic stem cell markers. Thereafter, MSC were seeded in a high-density culture. To determine the chondrogenic potential of different serum media, MSC were either cultured with basal stimulation medium, chondrogenic differentiation medium or with haemarthrosis-serum. The 3D cell pellet was characterized (e.g. collagen type II, chondroitin-4-sulfate, SOX-9) by IHC. Results: After 22 h AAP release more cells comparing to chondral procedures (CP) in the haemarthrosis while 10 days after AAP most cells were countered. Their morphology changed from spindle-shaped fibroblast like cells to rounded chondrocyte-like-cells. Using FACS analysis haemarthrosis-derived cells 22 h after AAP and CP are positive for CD 44, 73, 90, 105 and negative for 34. In contrast to MSC after AAP, MSC after CP do not proliferate wich is negative for the proliferation marker Ki-67. In a high density culture, MSC differentiate to a chondrocyte-like cell type and produce an extracellular matrix which is positive for SOX-9, chondroitin-4-sulfate and collagene type II. Comparing the chondrogenic potential of different cell media after co-culturing with MSC no differences regarding their number and quality were observed. Conclusions: Interestingly, mononuclear cells after solely chondral procedures display characteristic MSC markers but do not proliferate in a high density culture and thereby explain the poor clinical results after chondral procedures. This study provided evidence that haemarthrosis-derived cells after AAP and MF can differentiate to chondrogenic lineage cells in vitro. The morphological convergence of the cell culture to hyaline cartilage dependent on the kind of arthroscopic procedure underlines the benefit of bone marrow stimulating techniques in the arthroscopic treatment of cartilage defects.

P26-925 High tibial ,,open-wedge’’ oteotomy versus unicompartmental knee arthroplasty in unicompartimental knee joint osteoarthritis: 5-year follow up retrospective study S. Metzlaff1, A. Achtnich1, P. Forkel1, W. Petersen2

S265 1

Martin-Luther-Krankenhaus, Orthopaedic and Trauma Surgery, Berlin, Germany, 2Martin-Luther-Krankenhaus, Klinik fu¨r Orthopa¨die und Unfallchirurgie, Berlin, Germany Objectives: The surgical treatment for unicompartmental gonarthrosis is still a controversial issue today. High tibial osteotomy and unicompartmental joint replacement represent two common operative options to treat this disease. The purpose of this study was to evaluate the clinical outcomes and Patient satisfaction of hight tibial ,,open- wedge’’ osteotomy and unicompartimental knee arthroplasty. Methods: A total of 25 consecutive patients recieved a unicompartmental arthroplasty and 23 patients a high tibial ,,open-wedge’’ osteotomy. The unicompartmental prosthesis used in this study was the Oxford III prosthesis, (Biomet Inc., Warsaw, USA). It was implanted with cement over a minimalinvasive medial approach according to the operating manual. Postoperatively, full weight bearing was allowed. All the osteotomies were carried out in ,,open wedge’’ technique over a medial approach using the Tomofix TM angular locking plate (Synthes, Umkirch). Postoperatively, partial weight-bearing with 10 kg was performed for 6 weeks. An objective assessment of the succes of the operation was obtained using the Knee injury and Osteoarthritis Outcome Score (KOOS) and the Hospital for Special Surgery Score (HSS). The Patient satisfaction was measured using the visual analogue scale (VAS). Results: 80% of patients after replacement and 65% after osteotomy had a Hospital for Special Surgery Score (HSS) of excellent or good 5 years postoperatively. The Kaplan- Meier survival analysis 5 years postoperatively showed a survivorship of 96% for UKA and 95% for HTO. Using the Knee injury and Osteoarthritis Score (KOOS) the replacement group improved from 42% pra¨-operative to 77% 5 years postoperative. The osteotomy collective improved from 42% pra¨operative to 58% 5 years post-operative. Conclusions: We conclude that both operative techniques are sufficient methods to treat medial gonarthritis though the UKA collective showed better results 5 years post-operative.

P26-1112 Autologous Membrane Induced Chondrogenesis for the treatment of chondral lesions in the knee. A retrospective clinical evaluation with mid-term results of 40 patients L. de Girolamo1, P. Adravanti2, H. Schoenhuber1, P. Banchini2, P. Volpi3 1 IRCCS Galeazzi Orthopaedic Institute, Milan, Italy, 2Casa di Cura ‘Citta` di Parma’, Orthopaedic Department, Parma, Italy, 3IRCCS Istituto Clinico Humanitas, Rozzano, Italy Objectives: The aim of our multicentric study is to retrospectively evaluate the efficacy of the AMIC (Autologous Matrix Induced Chondrogenesis), combining microfracture with the application of a collagen I/III bilayer matrix (Chondro-Gide) for the treatment of chondrla lesions. Methods: 40 patients (28 male, 12 female) were treated with the AMIC technique in two centers. Patients were evaluated pre- and post-operatively by VAS Pain Scale, Lysholm Score and IKDC form; MRI was performed at 12 and 24 months. Mean patient age was 29 ± 11y/old (range 16–47). All patients presented grade IV chondral defects (ICRS classification), with a mean lesion size of 3.1 ± 1.2 cm2 (range 1–6 cm2). Defect location was medial femoral condyle (n = 22), lateral condyle (n- = 9) and patella (n = 9). Average follow up was 32 months (range 18–55 months). Statistical analysis was performed using Student t test and v2 test. Data ar presented as mean ± SD; p \ 0.05 was considered statistically significant.

123

S266

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Table 1 Comparison between patella and total population Time point (month)

Lysholm score

VASSCORE

Total (%)

Pattllar lesions (%)

Total (%)

Pattllar lesions (%)

6

+22

+42

-50

-63

12

+19

+41

-65

-71

24 36

+23 +25

+47 +45

-80 -81

-91 -89

48

+24

+43

-79

-90

Males obtained better results compared to females (p \ 0.05 at all time points). MR revealed a significant reduction of the defect, with a complete filling of the lesion in 75% of the patients after 12 months and in 84% after 24 months. Conclusions: AMIC is an effective treatment for chondral lesions. Significant improvements are already obtained after 6 months, with a further significant increase of the scores starting from 24 months from intervention. These improvements are stable at least until 48 months of follow up.

Fig. 1 Lysholm Score and VAS Scale

Fig. 2 IKDC Results: Significant improvements in term of objective and subjective functional improvement and pain relief was observed already starting from 6 months post-operatively (p \ 0.05). All scores were further improved significantly until 24 months. These results were then maintained at least until 48 months (Figs. 1, 2 where *p \ 0.05, each time point Vs. pre-op; #p \ 0.05 each time point Vs. 6 months). Patients treated for patellar chondral lesion showed a relevant higher improvement in term of Lysholm Score and VAS Scale with respect to the whole population, but not statistically significant (Table 1).

123

P26-1120 Critical bone and osteochondral defects: An innovative approach using magnetic scaffold A. Russo1, S. Panseri1, M. Sandri2, A. Tampieri2, V. Dediu3, M. Marcacci4 1 Rizzoli Orthopaedic Institute, University of Bologna, Bologna, Italy, 2 Institute of Science and Technology for Ceramics -ISTEC-CNR, Faenza, Italy, 3Institute of Nanostructured Materials, ISMN-CNR, Bologna, Italy, 4Instituto Ortopedico Rizzoli Bologna, Bologna, Italy Objectives: Tissue engineering approaches aim to reconstruct critical tissue defect using biodegradable scaffolds as a template for cells attachment, proliferation, differentiation and extracellular matrix (ECM) deposition. Regenerating a tissue, with controlled threedimensional architecture able to reproduce native biological and mechanical characteristics, represent a major challenge in bone tissue engineering. As showed by few Authors, the application of magnetic forces can orient, in vitro, the osteoblasts and the collagen fibrils according to magnetic field lines and moreover cells proliferation and differentiation result accelerated. The objective of the present study is to demonstrate, in vivo, that magnetic forces are able to induce bone tissue regeneration in a defined three dimensional pattern according to the applied magnetic field, leading, for the first time, to the possibility of recreating bone functional architecture in vivo. Methods: Magnetic scaffolds for bone regeneration (whose safety and biocompatibility were previously assessed) were implanted into the lateral aspect of femoral condyles of twelve male rabbits; they were in direct contact with a press fit implanted 1.2 T Cylindicar NdFeB magnets (diamater2.00 mm; length 8.00 mm) (experimental times: 4–12 weeks). Scaffolds (diameter 4 mm; thickness 4 mm) were made by Hydroxyapatite/collagen (70:30) and magnetized with magnetic nanoparticles. Non magnetic scaffolds were implanted in the controlateral side, as a control group, in direct contact with the same permanent magnet. Magnetic forces through the scaffold were calculated by finite element software (COMSOL Multiphysics, AC/DC Model). Results: Bone tissue regeneration was revealed in magnetized scaffolds at both follow-up times (4–12 weeks). After 4 weeks the collagen fibrils of the magnetic scaffold oriented parallel to the magnetic field lines, whereas a 12 weeks new bone trabecolae were oriented perpendicular to magnetic field lines. On the contrary bone regeneration in the control group revealed random tissue organization.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: These findings may open the way to new strategies to reconstruct tissue functional structure in bone tissue engineering. This evidence leads to the fascinating possibility of recreating in vivo the three dimensional functional architecture of bone tissue guided by magnetic forces.

P26-1226 All arthroscopic Autologous Matrix Induced Chondrogenesis cartilage reconstruction in the knee - early results K. Ciemniewska-Gorzela1, T. Piontek1, A. Szulc2, J. Naczk3 1 Rehasport Clinic, University of Medical Sciences, Clinic of Peadiatric Orthopeadic Surgery, Poznan, Poland, 2University of Medical Sciences, Clinic of Peadiatric Orthopedic Surgery, Poznan, Poland, 3Rehasport Clinic, Poznan, Poland Objectives: One of the methods of cartilage knee reconstruction is Autologous Matrix-Induced Chondrogenesis (AMIC). AMIC is a variant of the microfracture method. It makes use of a collagen membrane that serves as a scaffold for new cells and allows effective reconstruction of even large fragments of damaged cartilage surface. Currently, such procedures are performed by means of surgical opening of the knee joint. The goal of study is to present an arthroscopic technique for reconstructing damaged fragments of knee cartilage using the AMIC technique with collagen matrix and present 1 year results after operation. Methods: Twenty patients with cartilage defects of the knee were operated with AMIC all arthroscopic technique. Defects sizes were 2–6 square cm on all femoral, tibial and patellar localization. We control patients before and 3, 6, 12 month after operation using MOCART MR scans scale, Lysholm and IKDC2000 scale and Baret criteria of clinical assessment. Results: We observe in MR scans good covering of the cartilage defects by the new tissue and in MOCART scale 75–100% good response for the treatment after 6 and 12 month. We observed improvement in Lysholm scal from 60 to 90 after 1 year and IKDC2000 from 40 to 70 after 1 year. In Baret criteria we found improvement from 1 to 4 points in 85% of patients. Conclusions: We have presented a simple, entirely arthroscopic technique for reconstructing extensive cartilage defects without bone defects and results of treatment are good promising for future investigations.

P26-1235 Long-term results of abrasion arthroplasty of the medial femoral condyle of the knee for full-thickness chondral defects V. Sansone1, V. Pascale1, L. de Girolamo2 1 Universita’ degli Studi di Milano, Orthopaedics and Traumatology, Istituto Ortopedico Galeazzi IRCCS, Milano, Italy, 2Galeazzi Orthopaedic Institute, Orthopaedic Biotechnologies Lab, Milano, Italy Objectives: Abrasion arthroplasty is a technically simple and low cost procedure compared to other treatments for full thickness chondral injury. Nevertheless, there is little data on the long-term outcomes of this procedure. We present our results of medial femoral condyle abrasion arthroplasty, with a minimum 10-year follow-up. To our knowledge, this is the longest follow-up to date for this procedure. Methods: From 3/1990 to 12/2001, 76 consecutive patients (37 male and 39 female) suffering from a full thickness chondral defect of the medial femoral condyle underwent abrasion arthroplasty performed by a single surgeon. The average age of the cohort was 65 years (range:24–79, SD 15.44 years). Inclusion criteria were having undergone abrasion arthroplasty, with a minimum 10-years follow-up (mean 16.1 years, range 10.2–20.1 years, SD 3.26). Patients with

S267 ligament injuries, multiple compartment involvement, varus angulation [10 and inflammatory arthritis were excluded from the study. Preoperative weight bearing radiographs were obtained in all cases. At final f-u patients completed a questionnaire based on their current knee symptoms and need for further surgery. The questionnaire included the IKDC subjective score and a question regarding medication used for the operated knee. Patients were considered to have a successful result if they had no pain or only occasional pain with activity, no limitation of their activities due to the knee, and had not undergone repeat surgery. Results: 65 patients (85.5%) were available for follow-up (3 patients died and 8 were lost at f-u). Patients with lesions less that 4 cm2 (n = 30) had a 73% success rate at final f-u. Patients with lesions greater than 4 cm2 (n = 32) had a 21% success rate which was statistically significant (p \ 0.005). The differences in outcome with regard to meniscectomy, as well as the difference in distribution of large lesions between the two patient groups were not statistically significant. Patient’s age did not affect outcomes, although there was a trend of poor outcomes associated with age of 50 years or older. The overall re-operation rate was a 31%. Re-operation was on average 6.2 years after the arthroscopic procedure. Conclusions: Abrasion arthroplasty aims to seal full thickness chondral defects, avoiding the continued degeneration that is the natural history of such lesions. It is a temporary measure to delay the progression of arthritis. In our series it seems that dimensions of the lesion was crucial for the quality of the result. The procedure is effective in minimizing the progression of symptoms at long-term follow-up for small lesions, regardless of patient age. The trend toward poor outcomes in patients over fifty is likely due to the higher percentage of large or lesions rather than patient age itself. Abrasion arthroplasty should not be performed in patients with chondral lesions larger than 4 cm2.

P26-1278 Comparison of factors associated with early osteoarthritis versus moderate to severe osteoarthritis of the knee K. Briggs1, L.M. Matheny1, J.R. Steadman2, W.G. Rodkey3 1 Steadman Philippon Research Institute, Clinical Research, Vail, United States, 2The Steadman Clinic and, Steadman Philippon Research Institute, Vail, United States, 3Steadman Philippon Research Institute, Vail, United States Objectives: Knee osteoarthritis (OA) has been shown to cause significant pain and disability, which can lead to limitations of daily activities, a sedentary lifestyle and ultimately a lower quality of life. No risk factors have been identified for early OA, making it difficult to intervene and prevent the progression of knee OA. The purpose of this study was to identify factors associated with early knee OA (defined as Kellgren-Lawrence (KL) grade 2) and compare to factors associated with moderate to severe OA (defined as KL grade 3 or 4). Methods: This study was institutional review board approved. Patients over the age of 18 years and patients with a KL grade 2 or greater were included. Data were collected prospectively and reviewed retrospectively. In addition to Kellgren-Lawrence grade, knee alignment, time from injury to presentation, surgical history, body mass index (BMI), age and gender were documented. In addition, all patients completed a subjective questionnaire at time of evaluation that included WOMAC score. There were 1,208 patients who met the inclusion criteria. Patients were then dichotomized according to KL grade, early OA was operationally defined as KL grade 2 and moderate to severe OA defined as KL grades 3 or 4. Results: There were 510 (42.2%) patients in the early OA group (KL grade 2) and 698 (57.8%) patients in the moderate to severe OA group (KL grade 3/4). Patients’ average age in the early OA group was

123

S268 54.9 years [95%CI: 53.8–56.0] and 59.6 years [95%CI:58.8–60.5] in the moderate to severe OA group. Mean body mass index (BMI) for all patients was 26.0 kg/m2 (range:16.5–50.2 kg/m2). Average WOMAC score for early OA group was 29.0 [95%CI:27.5–30.6] and 34.0 [95%CI:32.8–35.3] for the moderate to severe OA group. In the early OA group, alignment outside of neutral was seen in 45.5% of patients and 68.3% of patients in the moderate to severe OA group. Patients with no previous surgery were 1.5 times more likely to have early OA [95%CI:1.1–1.9] versus moderate to severe OA. Patients who presented within 2 years or less from time of injury were 2.0 times more likely to have early OA [95%CI:1.6–2.8] versus moderate to severe OA. Conclusions: This study identified factors associated with early OA as compared to moderate to severe OA. Patients with early OA were younger, sought treatment for their injury sooner, had less knee malalignment and reported less disability when compared to patients with moderate/severe OA. These factors may be important in defining programs and early interventions to slow the progression of OA.

P26-1389 Miniprosthesis addressing cartilage lesions in the knee - are they usable? J.O. Laursen1 1 Orthopedic Department, County Hospital of South Jutland, Sønderborg, Denmark Objectives: In 2003 a new type of knee mini-prosthesis was introduced in USA and Europe. HemiCAP was primary developed for articular cartilage lesions in knee, toe, shoulder and hips. In 2006, it was introduced to the Danish market. By then there was clinical follow up in USA and Europe for 3 years in around 5.000 patients, and the results were promising. Four years later, the UNICAP was introduced for more advanced cartilage lesions on both femoral and tibial side. Over the years if you deal with surgery of the knee -arthroscopies or total knee replacements- you have seen many patients with minor cartilage lesions and perhaps have tried microfracturing (am. Steadmann), transplant cartilage or utilize chondrocyte transplantation with more or less success. I introduced these implants in 2007 and 2009 respectively, and have so far implanted 69 Hemicaps and 80 Unicaps. Methods: There was a protocol for the operations: all patients had preoperative X-rays and MRI taken and all operations started with arthroscopy to confirm if the cartilage damage was suitable for further CAP-surgery and to deal with other damages to meniscus or cartilage. All operations were done in spinal or GA. PC preoperatively and with use of a tourniquet. Midincision -miniarthrotomy. Postoperative full weight-bearing and range of motion. Discharge 1–3 day postoperatively and controlled by X-rays postop. at 6, 12 weeks and 1, 2, 3, … years including clinically controls with assessing of KSS. All patients were registered in the Danish Knee Society register. VAS-score was registered preoperatively and at follow-ups including KSS, complications and revisions. Results: Unicaps: 35 ptt’s in 2009, 27 in 2010 and 18 in 2011 (only on the femoral condyle). 5 ptt’s revised to TKA because of too extensive osteoarthritis in more than one chamber! 2 ptt’s revised to TKA after postop. infection (second to re-arthroscopy), and 3 ptt’s revised to TKA because of aseptic loosening. 69 patients operated with the Hemicap and 4 revised, 3 due to more extencive lesions and arthrosis and one aseptic loosening. Total caps: 149 in 5 years and 14 revised (9%), and cummulative Survival rate at 92.7% at 5 years. Conclusions: So far I have found the HEMICAP and UNICAP promishing for restoration of focal cartilage lesions in the knee in the ‘‘millennium patient’’ with isolated cartilage lesions or minor arthrosis. Further study and follow-ups are nessessary to confirm this,

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 and RCT-studies against Steadman’s micro-fracturing or cartilage transplantation would be of great clinical interest.

P26-1393 A tissue engineered osteochondral composite for cartilage repair: an in vivo study G. Peretti1, C. Sosio1, D. Deponti2, M. Melato3, A. Di Giancamillo3, F. Gervaso4 1 San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy, 2University of Milan, Faculty of Exercise Sciences, Milan, Italy, 3University of Milan, Milan, Italy, 4University of Salento, Lecce, Italy Objectives: The aim of this work is to validate the efficacy of a tissue engineered osteochondral composite for the treatment of cartilage lesion produced in adult pigs. The osteochondral composite was manufactured by combining an osteo-compatible cylinder and a neocartilagineous tissue obtained by seeding swine articular chondrocytes into a collagen scaffold. Methods: Articular cartilage were harvested from the trochlea of six adult pigs; the cartilage was enzymatically digested and the chondrocytes isolated were expanded in monolayer culture in presence of growth factors. Expanded chondrocytes were seeded onto a collagen scaffold. The collagen scaffold was preintegrated in vitro, macroscopically and microscopically, to a an osteo-compatible cylinder. The seeded osteochondral scaffold were left in standard culture condition for 3 weeks with the addition of growth factors. At the end of culture time the osteochondral scaffolds were surgically implanted in osteochondral lesion performed in the trochlea of the same pigs from which the cartilage was initially harvested. As control, some osteochondral lesions were treated with acellular scaffolds and others were left untreated. After 3 months, the repair tissue of the three experimental groups was macroscopically analyzed and processed for histological and biochemical analysis. Results: The hystologic ICRS II scale showed a statistically significant difference between the three experimental groups only in the parameters regarding the cell morphology and the surface/superficial assessment: the lesion treated with the unseeded osteochondral scaffolds showed higher values in chondrocytes morphology and in the superficial layer recovery, with respect to the lesions treated with the seeded scaffolds or left untreated. The biochemical analysis showed a higher DNA content in the lesion repaired with cellular scaffold and a higher GAGs/DNA ratio in the lesions with a spontaneous repair. Conclusions: The result of this study demonstrate that an osteochondral scaffold was able to repair an osteochondral lesion in an in vivo model of adult pigs, showing a good integration with the surrounding tissue. The quality of the repair was higher when the scaffold was not seeded with chondrocytes, but filled with cells migrated from subchondral bone. This tissue engineered osteochondral composite could represent a valuable model for further in vivo studies on the repair of chondral/osteochondral lesion. The Authors acknowledge Professor Cinzia Domeneghini, Mr. Paolo Stortini, Mr Alessandro Pozzi, Mrs Rosa Ballis, Prof. Alessandro Sannino for the completion of this work.

P26-1430 Microfracture treatment In athletes with knee grade IV chondral lesion: a 10 year follow up A. Gobbi1, A. Kumar1, G. Karnatzikos1 1 Orthopaedic Arthroscopic Surgery International, Department Sport and Medicine, Milano, Italy

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Microfracture is a well-established treatment procedure for chondral defects in high demand population with good short-term results. However, the long-term durability is under debate and few studies have been published with long-term results in athletic population. The purpose of this study was to evaluate long-term clinical outcome of microfracture treatment in athletic population with knee grade IV chondral lesion. Methods: We studied a cohort series of athletic patients with a minimum follow-up of 10 years. Between 1991 and 2001, a total of 170 patients underwent microfracture treatment of their knee, out of these 65 patients met the following inclusion criteria: (1) traumatic fullthickness chondral defect and (2) athletic individuals, whom we considered as those playing in competitive sports or participating in active sports at least 2 times per week. Out of 65 patients, 2 died and 6 changed address. Finally, 57 patients completed self-administered questionnaires preoperatively and postoperatively, at 5 years and were recalled after completing 10 years of follow up. Lysholm, KOOS, Tegner and IKDC scores were utilized. Results: The following results were significant at the p \ 05 levels. IKDC Subjective evaluation was 40/100 preoperatively and 60/100 at final follow-up. Lysholm was 56.8 pre-op and 60.2 final. IKDC objective analysis revealed: 0 A, 3 B, 40 C and 14 D preoperatively while at final follow up 60% scored B or C. Tegner score at final follow-up (80%) showed a decline in sport activity level (Tegner 5). Pre-operatively KOOS was p = 62.2/ S = 61.7/ADL = 50/SP = 40.3/QOL = 40.8 and at the end of follow up p = 68.4/S = 64.4/ADL = 60/SP = 48.4/QOL = 50.8. Only 4 patients could continue at the previous level of sports. Conclusions: Over 10–15 years follow-up period (average 13.2 years), the athletic patients with big lesions ([3 cm2) who underwent microfracture procedure for full-thickness chondral defects of the knee failed within 1 year and they required a second surgery. Patients with smaller lesions (\2.5 cm2), showed statistically significant improvement in function in the first 5 years. After 5 years most of them deteriorated gradually due to lower activity levels. Although the activity level of athletes with respect to sports had declined, they showed significant improvement in function and decrease in pain. We believe microfracture can be a good option to treat small chondral defects in active individuals but competitive athletes should be advised that the improvement seen would decline with time.

Knee-Cartilage II

P27-39 Osteochondritis dissecans of the knee. Autologous chondrocyte transplantation, in combination with corrective osteotomy of the lower limb G. Goudelis1, K. Samaras1 1 Interbalkan European Medical Center, Orthopaedic and Sports Medicine Department, Private Hospital, Thessaloniki, Greece Objectives: Osteochondritis dissecans is a necrotic focus with surrounding sclerosis of the bone, which can then become detached and remain in the joint as a loose body (joint mouse). The chondrocyte transplantation is now an established method for treating large damage of articular cartilage. Corrective osteotomy of the lower limb helps to redistribute the forces exerted on the joints, reducing the loads on the suffering part of it. Methods: 15 knees—14 men and 1 woman—underwent chondrocyte transplantation in combination with corrective osteotomy of the affected leg. The lesion was located on the medial femoral condyle in 13 patients and on the lateral femoral condyle in 2 patients. Valgus open wedge tibial osteotomy was done in 13 individuals and varus open wedge femoral osteotomy in 2 patients. The average age was 24 years (16–40). 6-year post-op follow up.

S269 Results: The control patients was done with MRI examination (after removal the osteotomy plates), arthroscopicaly and with the SF-36 questionnaire (quality of life measurement) before surgery and 3 months, 6 months, 12 months, 2 years and 6 years after surgery. The results showed clinical improvement of patients 90%, patients were satisfied and all returned to their previous activities (daily activity or sports activity). Conclusions: The chondrocyte transplantation is now an established method for the treatment of osteochondritis dissecans. Combining the method with corrective osteotomy of the leg due to the new distribution of forces exerted at the knee joint, is that it improves the final result for the painless restoration of the individual and return to previous activities.

P27-65 Platelet-rich plasma in treatment of tibiofemoral chondromalacia R. Hart1, A. Safi1, M. Komza´k1, P. Jajtner2, M. Puskeiler3, P. Hartova´4 1 General Hospital, Department of Orthopaedics and Traumatology, Znojmo, Czech Republic, 2General Hospital, Department of Haematology, Znojmo, Czech Republic, 3General Hospital, Department of Radiology, Znojmo, Czech Republic, 4General Hospital, Department of Psychiatry, Znojmo, Czech Republic Objectives: Recently an articular cartilage repair has been given much attention in the orthopaedic field. Cartilage regeneration capacity is very limited. Optimal approach could be a delivery of natural growth factors. Autologous platelet-rich plasma (PRP) contains many proliferative and chemoattractant growth factors. The objective of the present study was to confirm or refuse two hypotheses: 1) PRP enhances tibiofemoral cartilage regeneration and 2) PRP improves knee function. Methods: Fifty consecutive and strictly selected patients (29 men and 21 women) with a mean age of 58 years were enrolled in the prospective study approved by the Ethical Committee. All underwent arthroscopy between January and June 2009 with the diagnosis of isolated Grade II (21 knees) or III (29 knees) tibiofemoral chondromacia (Outerbridge classification). The most degenerated (and investigated) was cartilage on the medial femoral condyle in 22, on the lateral femoral condyle and medial tibia identically in 11, and on the lateral tibia in 6 cases. Main exclusion criteria: patellofemoral damage, associated intra- or extraarticular lesions, axial deviations, BMI C 35, any intraarticular aplication 6 months prior the arthroscopy, administration of NSAID, SYSADOA or steroids, rheumatic diseases, gout. Patients underwent 1 year treatment—9 injections of 6 ml of autologous PRP in a liquid form with 2.0–2.5-fold platelets concentration (407,000–513,000/ll). First 6 injections were applied at 1-week intervals starting 6 weeks after the arthroscopy. Last 3 injections were applied at 3-month intervals. Outcome measures included the Lysholm, IKDC, and Cincinnati scores. Magnetic resonance (MR) imaging was used to evaluate cartilage thickness and degree of degeneration at the beginning and after 1 year in the end of the study. Initial MR evaluation of the cartilage structure (Grade II—fibrillation, Grade III—deep fissuring) didn0 t differ from the arthroscopy finding in any case. Nonparametric Wilcoxon and Mann–Whitney tests were used to evaluate outcomes. Results: The study demonstrated significant improvement in Lysholm (p \ 0.0001) and Cincinnati (p \ 0.00001) scores. The IKDC subjective score improved from 51.2 to 63.8 (p \ 0.02). Most improvements were obtained in pain, swelling, stairs and walking. The IKDC objective score remained without significant difference. MR cartilage assessment revealed no significant cartilage regeneration in terms of thickness and structure (p \ 0.05).

123

S270 There were no adverse events observed. Mild pain was reported after 17 of all 450 applications. In no case it persisted more than 5 days. Conclusions: MR imaging didn0 t confirmed any significant cartilage condition improvement. No worsening was observed. We had to refuse the 1st hypothesis. PRP treatment significantly reduces pain and improves quality of life in patients with low degree of tibiofemoral cartilage degeneration. We have confirmed the 2nd hypothesis. Supported by the grant No. NT 12057-5/2011 of the Ministry of Health.

P27-69 Short-term outcome of the second generation characterized chondrocyte implantation for the treatment of cartilage lesions in the knee A. Dhollander1, P. Verdonk2, R. Verdonk1, K.F. Almqvist2 1 Gent University Hospital, Department of Orthopaedic Surgery and Traumatology, Gent, Belgium, 2Gent University Hospital, Orthopaedics, Gent, Belgium Objectives: To evaluate short-term clinical and MRI outcome of the second generation characterized chondrocyte implantation (CCI) for the treatment of cartilage defects in the knee. Methods: Patients aged 15–51 years with single International Cartilage Repair Society (ICRS) grade III/IV symptomatic cartilage defects of different locations in the knee were treated with CCI using a synthetic collagen I/III membrane to cover the defect. Clinical outcome was measured over 36 months by the Knee injury and Osteoarthritis Outcome Score (KOOS) and Visual Analogue Scale (VAS) for pain. Serial magnetic resonance imaging (MRI) scans were scored using the original and modified Magnetic resonance Observation of Cartilage Repair Tissue (MOCART) system. Results: The patients included in this study showed a significant gradual clinical improvement after CCI. The MRI findings of this pilot study were considered to be promising. No signs of deterioration were observed. A complete or hypertrophic filling was observed in 76.50% of the cases at 24 months of follow-up. No preventive effect of an avital membrane on the occurrence of hypertrophic repair tissue was observed on MRI. Three failures were observed among the 32 patients until now (9.4%). Conclusions: This investigation provided useful information on the efficacy of this treatment in a challenging patient cohort. The shortterm clinical and MRI outcome are promising. Large scale and longterm trials are mandatory to confirm the results and the reliability of this procedure.

P27-97 Results prediction of knee cartilage injuries treatment M. Golovakha1, R. Schabus2, W. Orljanski2 1 Zaporozhje State Medical University, Orthopedics & Sport Trauma Faculty, Zaporozhye, Ukraine, 2Wiener Privat Klinik, Orthopedics & Sport Trauma, Wien, Austria Objectives: An analysis of the current literature about chondroplasty for knee osteoarthritis has shown that despite the numerously developed treatment methods prediction of treatment results remains difficult. First of all this testifies that indications for the developed treatment methods are insufficiently precise. Aim of the study: to reveal factors which essentially influence treatment result after repair of cartilage defects in the knee joints based on a clinical study.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: In this work the treatment results of 176 patients with knee osteoarthritis with a minimum follow-up time of 8 years were analyzed. The cartilage defects were surgically treated in all patients. The used techniques were microfracture in 142 patients and mosaic plastic in 34 patients. Average age of the patients was 47 years (range 37–61 years). 79 patients had cartilage lesions of the internal femoral condyle; other patients had defects of both articular surface of the internal condyles of the femur and of the articular cartilage of the tibia. Only cartilage lesions of III–IV degree by Outerbridge were considered for the analysis. 149 patients out of 176 treated were followed up (84.7%). Follow-up time was between 8 and 14 years. Outcome measures were Lequesne index and radiological progression of the osteoarthritis based on the height of the joint space. The statistical analysis was performed using nonparametric methods, comparative tables and assessing correlations. Results: Patient age was a statistically significant predictor for progression of knee osteoarthritis. Best results were seen in patients younger than 40 years (p \ 0.001). In the age group of 40–55 years the results were satisfactory (p \ 0.01). Influence of the lesion area on the treatment results was not significant (p [ 0.05), as also the correlation between the lesion area and the follow-up results was low (Rxy = 0.21). An important significant predictor for treatment results was the initial condition of the knee joint. High correlation was seen between results and the preoperative and height of the joint space (Rxy = 0.21). The most significant predictor for good treatment results was medial tibial angle. The treatment results for the most of patients with an angle 86 and below were fair or poor. The geometry of a distal femur has not appeared very variable, and therefore did not essentially correlate with the treatment results. Conclusions: In conclusion, results of restoration of a condyles cartilage lesions in knee joints depends on initial knee joint status and biomechanical relations in the joint and not on the lesion area.

P27-120 The chondrotoxic effect of single-dose local anesthetics J.L. Dragoo1, H.J. Braun1, H.J. Kim1, S.R. Golish1 1 Stanford University, Orthopaedic Surgery, Redwood City, United States Objectives: Single-dose intra-articular injections of local anesthetics such as lidocaine, bupivacaine, and ropivacaine are used to relieve pain after injury or surgery, to provide anesthesia for diagnostic tests, or to decrease chronic joint pain due to inflammation in combination with corticosteroids. The chondrotoxicity of local anesthetics has been documented in vitro and is a cause for clinical concern. Continuous administration of local anesthetics into human and animal joints has also been linked to subsequent chondrolysis. The effect of single-doses of local anesthetics has not been widely investigated, and the evaluation of chondrocyte viability has not been performed with respect to the average clinical duration of action of the medications. This study was performed to evaluate the in vitro chondrotoxicity of single-doses of 1% lidocaine, 0.25% bupivacaine and 0.5% ropivacaine on human chondrocytes over the clinical duration of action of each drug. Methods: Human chondrocytes were seeded at a density of 0.5 9 10^6 cells/well. A bioreactor was used to simulate normal joint fluid metabolism. The clinically acceptable dose of 10 cc was adjusted to account for decreased cartilage surface area of experimental conditions versus human knee, and three anesthetics were tested: 1% lidocaine, 0.25% bupivacaine and 0.5% ropivacaine. Each medication was delivered to the chondrocytes over the average duration of chemical action. Cell viability was assessed with a two-

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 color fluorescence assay. Data were analyzed by Chi-square test and z-test. Results: Chondrocytes treated for 3 h with 1% lidocaine exhibited significantly higher cell death than those in control media by Chisquare test (7.9 vs. 2.9% cell death, p \ 0.001). Chondrocytes treated for 6 h with 0.25% bupivacaine exhibited no difference in cell death compared to those in control media (2.7 vs. 2.8% cell death, p = 0.856). Similarly, cells treated for 12 h in 0.50% ropivacaine showed no difference in cell death compared with the control group (2.9 versus 2.4% cell death, p = 0.084). Conclusions: Our data suggest that when normal human chondrocytes are exposed to a single injection dose of 1% lidocaine for the average duration of action of 3 h, there is a significant decrease in chondrocyte viability compared with control cultures. Conversely, when chondrocytes are exposed to single injection doses of 0.25% bupivacaine and 0.5% ropivacaine over the medication’s average duration of action (6 and 12 h, respectively), there is no significant decrease in cell viability. Although there have been many studies examining the chondrotoxic effect of this class of medications, this investigation has been the only one to focus on identifying whether these medications are chondrotoxic in a model of single-dose administration based on the clinical duration of action. The chondrotoxicity observed in this study suggests that intra-articular administration of lidocaine should be used with caution, even in a single dose.

P27-141 Specific compartmental analysis of cartilage status in patients with double bundle ACL reconstruction: a comparative study using a pre- and post-operative MRI Y.S. Lee1, J.A. Sim1, J.H. Kwak1, S.W. Nam1, B.K. Lee1 1 Gachon University School of Medicine, Department of orthopedic surgery, Incheon, Republic of Korea Objectives: The purpose of this study was to evaluate the change of site specific cartilage status after double bundle ACL reconstruction by comparison between preoperative and follow-up MRI analyses. Methods: From February 2001 to August 2009, 36 double bundle ACL reconstructed patients had preoperative and follow up magnetic resonance imaging (MRI) and satisfied our inclusion criteria. Meniscal injury was included in this study and we divided patients 2 group by combined meniscal injuries. We classified cartilage morphology into 6 scales (from 0 = normal thickness and signal; to 6 = diffuse full-thickness loss ([ 75% of the region)). We evaluated 14 sites and evaluated the change of cartilage status after double bundle ACL reconstruction. Results: Cartilage changes were noticed in all sites and they were classified with the degrees of grade change and sites. Majority of changes were grade 0 and 1 changes and they were 68 and 16.8%, respectively. Patella medial facet, lateral femur anterior region, and medial femur central region showed statistically significant cartilage change (loss) than medial tibia posterior, lateral tibia central, lateral tibia anterior, and medial tibia anterior region. There was no statistical significance in the comparison according to the combined injury (p = 0.68). However, they showed different pattern of cartilage change by showing different rank of grade change. The time of MRI acquisition was related with cartilage changes of lateral femur posterior and medial tibia anterior regions (correlation coefficient 0.41 and 0.34). Conclusions: Double bundle ACL reconstruction could not avoid cartilage change, but majority of changes were minimal and showed some site specific variations. Combined meniscal injury did not affect on the change of cartilage status and the time of MRI acquisition was related with small correlation only in specific sites.

S271 P27-195 Prosthetic inlay resurfacing for the treatment of focal, full thickness cartilage defects of the femoral condyle: a bridge between biologics and conventional arthroplasty P. Bollars1, M. Bosquet2, B. Vandekerckhove3, J. Bellemans4 1 KU Leuven, Orthopaedic Surgery, Leuven Pellenberg, Belgium, 2 Europa Hospital Brussels, Orthopaedic Surgery, Brussels, Belgium, 3 AZ St. Jan Brugge, Orthopaedic surgery and Traumatology, Brugge, Belgium, 4University Hospitals Leuven, Catholic University Leuven, Orthopaedic Department, Pellenberg, Belgium Objectives: Localized full thickness defects of the femoral condyle can be highly symptomatic. Treatment options for these lesions are numerous in young patients, however they become increasingly challenging in middle aged and older patients. In order to delay traditional joint replacement procedures and to provide a soft tissue and bone sparing alternative, a focal inlay resurfacing procedure was introduced in 2004. Methods: Between 2004 and 2008, a consecutive series of 27 patients were treated with the Arthrosurface HemiCAP Focal Femoral Condyle Resurfacing Prosthesis and were assessed to study the clinical benefit of this procedure. Outcome measures included the KOOS, IKDC and WOMAC as well as physical and radiographic evaluation. Results: Nineteen patients met the inclusion and exclusion criteria, 18 were available for review at an average follow-up of 35.3 months (range: 20–57 months). The average age was 50.7 years (range: 43–78). 63% were diagnosed with early arthritis, 5.2% had localized osteonecrosis, and 31.6% had a focal traumatic full thickness defect. The total WOMAC score averaged 90.1, The KOOS showed very good to excellent scores in all domains at final follow-up and when compared to age matched normative data. Significant improvement was seen at final follow-up with the HSS Knee and Function Score. On IKDC examination, 83.4% had normal or nearly normal results. Conclusions: Focal femoral condyle resurfacing has demonstrated excellent results for pain and function in middle aged, well selected patients with full thickness cartilage and osteochondral defects. Patient profiling and assessment of confounding factors, in particular mechanical joint alignment; meniscal function; and healthy opposing cartilage surfaces, are important for an individual treatment approach and successful outcomes.

P27-208 Safety, efficacy and outcome of 4,400 patients treated with viscosupplementation for osteoarthritis of knee R. Raman1, C. Shaw2, H. Sharma2, G. Johnson3, A. Dutta2 1 Hull Royal Infirmary, Hull, Orthopaedics, Swanland, United Kingdom, 2Hull Royal Infirmary, Hull, United Kingdom, 3Academic Department of Orthopaedics, Hull Royal Infirmary, Normanton, United Kingdom Objectives: Viscosupplementation is used widely to provide symptomatic relief to patients with knee OA. This study is aimed to assess the safety, analgesic efficacy and functional outcome of hylan GF-20 in a large cohort of patients. Methods: This is an independent, prospective, blinded (reviewers), longitudinal cohort study over 6 years in a single institution. Inclusion criteria was OA knee pain of at least 60 mm on a 100 mm VAS; no prior intra articular (IA) injection. Patients received 3 9 2 mL hylan G-F 20 or 1 9 6 mL (from 2009). Follow-up at 1, 6, 12 and 26 weeks by blinded reviewers. Analgesics prohibited for 24 h prior to followup assessments and NSAIDs for 26 wks. All adverse events (AE) were recorded. Primary outcome measure: Target knee pain (VAS) at 26 weeks. Secondary outcome measures included WOMAC, Oxford knee score, SF12, Euroqol Eq 5D. Results: A total of 4,400 patients were recruited from 2004 to 2010. All patients received hylan GF-20 (3 9 2 mL N = 3,352,

123

S272 1 9 6 mL N = 1,048). Mean age was 61.2 years and 71% had Grade III (K–L) disease. Injection and/or treatment-related AE in the target knee were reported in 11.2% (3 9 2 mL) and 9.8% (1 9 6 mL) of patients. Significant pain reduction ([ 40%) was observed in both groups at 26 weeks (3 9 2 mL: 57% (mean) decrease from baseline, 1 9 6 mL: 51%). Overall knee pain on VAS improved from 69 to 37 at 6 months (p = 0.02). Significant improvements from the baseline in the WOMAC pain and function subscales (65% improvement at 26 wks) and Oxford knee scores at 3 and 6 months were observed. There were no significant differences between groups in demographics or for any of the primary or secondary outcome measures at 26 weeks. Conclusions: Viscosupplementation with hylan GF-20 is an efficacious and safe treatment option for patients with symptomatic OA of knees. Pain relief lasts for at least 6 months after a first course of treatment. Single dose of 6 mL hylan G-F 20 offer comparable safety and efficacy to 3 9 2 mL at 26 weeks. It provides both the patients and physicians a choice with the potential additional benefits of reduced operational costs.

P27-448 The evaluation using T2 mapping and the pathological specimen for the degenerative change of articular cartilage of the knee joint T. Tatsuhiro1, J. Nakase1, K. Masahiro1, O. Yoshinori2, T. Hiroyuki2 1 Kanazawa University, Orthopaedic surgery, Kanazawa, Japan, 2 Kanazawa University, Kanazawa, Japan Objectives: There have been reports on MRI T2 mapping that can evaluate arrangement and moisture content of the collagen in the articular cartilage. It can be useful at detect degeneration of cartilage at early stage or the quantitative evaluation of the level of cartilage denaturation. The purpose of this study was to compare the articular cartilage of the osteoarthritis of the knee joint between histopathological findings and imaging findings with MRI T2 mapping. Methods: Subjects were 11 patients (12 knees) with medial compartment osteoarthritis of the varus knee who were underwent total knee arthroplasty. They were 1 male and 10 female aged 74.5 in average (range, 67–83 years). For the histopathological findings, the decalcified pathological specimens of lateral condyle center of the femur excised by the operation were produced and safranin O staining was performed. For the imaging findings, MRI was performed before operation using a1.5T or 3T MRI (GE, Signe EXCITE HDxt), and T2 mappings were calculated from a sagittal multi-echo acquisition (Section thickness 3.0 mm, TR 1,200 ms, TE 7.9–62.8 ms, FOV 160 9 160 mm,Matrix 256 9 192). Four regions of interest (ROIs) were positioned within medial femoral condyle cartilage both the pathological specimens and T2 mapping images as the same part. They were evaluated the correlation between the average T2 value of T2 mapping images of each ROI and average Mankin score of the pathological specimens. Moreover, the results of Mankin score were classified into 2 groups: 0–7 as a low degenerative group, 8–14 as a high degenerative group. Statistical analysis was performed between two groups using spearman0 s rank-correlation coefficient and one-way analysis of variance, and a p value of \0.05 was considered significant. Results: The average T2 value was 39.3 ± 5.9, and the average Mankin score was 4.0 ± 2.7. The correlation coefficients are 0.48 and there was positive correlation between T2 value and Mankin score. Also, significant difference was observed in the average of each T2 value of low degenerative group and high degenerative group (p = 0.01). Conclusions: This study showed that it was possible to evaluate the degree of degeneration of the articular cartilage of osteoarthritis of the

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 knee visually from T2 value and to have quantitated degree of cartilage degeneration. It was suggested that T2 mapping was noninvasive and suitable as a method of evaluating the cartilage degeneration visually, and it seemed useful to diagnosis of the degree of cartilage degeneration and evaluation of medical treatment.

P27-518 PRP injections versus viscosupplementation for early knee osteoarthritis: a randomized double-blind study G. Filardo1, E. Kon1, A. Di Martino1, S. Patella1, B. Di Matteo1, M. Marcacci1 1 Rizzoli Orthopaedic Institute, Bologna, Italy Objectives: The influence of growth factors (GFs) on cartilage repair is not yet widely studied and its application in clinics is still experimental. Platelet Rich Plasma (PRP), a blood product rich in GF, is a promising support for treating cartilage defects. Aim of this study is to evaluate and compare the efficacy of PRP and Viscosupplementation (Hyaluronic Acid—HA) injections for the treatment of chondropathy or early osteoarthritis (OA) of the knee. Methods: 150 patients were enrolled: 72 were evaluated at 6 months follow-up. The study involved patients affected by chondropathy or early OA. All patients underwent an autologous blood harvesting, then 36 patients were randomized in the PRP group and 36 in the HA group. A cycle of 3 weekly injections was administered blindly. All patients were clinically evaluated at the enrolment, 2 months after the treatment and at 6 months follow up. IKDC, EQVAS, TEGNER and KOOS scores were used to clinically evaluate the patients. Adverse events and patient satisfaction were also recorded. Results: No complications like infection, marked muscle atrophy, deep vein thrombosis, fever, haematoma, tissue hypertrophy, adhesion formation or other major adverse events occurred among study subjects. Only minor adverse events were detected in some patients, as mild pain reaction and effusion after the injections, in particular in the PRP group, but they lasted for no more than a few days. At the follow-up evaluations, both groups showed a significant improvement in terms of function and quality of life. The preliminary comparison between the outcomes of the two groups showed a not statistically significant difference, but with a trend slightly favourable for the PRP group at 6 months. Conclusions: PRP is a simple, low cost and minimally invasive approach to treat chondropathy and early OA. The clinical results of our study are encouraging and suggest this method may be used to treat the degenerative articular pathology of the knee, leading to results at least comparable with those of HA. Long-term results will confirm the reliability and evaluate the durability of this promising procedure.

P27-577 Osteochondral regeneration in goat model using a new aragonite-hyaluronic acid bi-phasic implant E. Kon1, K.R. Zaslav2, D. Robinson3, Y. Chorev4, J.A. Eisman5, A.S. Levy6 1 Rizzoli Orthopaedic Institute, Bologna, Italy, 2Advanced Orthopedic Centers, Richmond, Virginia, United States, 3Hasharon Hospital, Rabin Medical Center, Petah Tikwa, Israel, 4Laniado Hospital, Netanya, Israel, 5Garvan Institute of Medical Research, Darlinghurst Sydney, Australia, 6Center for Advanced Sports Medicine Knee and Shoulder, Morristown, United States Objectives: Although coral is an excellent material for bone repair, as a native stand-alone material it cannot regenerate native hyaline cartilage. Mechanical modification with drilled channels and

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 impregnation of HA could enhance the scaffold regenerative potential. The objectives of this study were to examine whether different mechanical modifications and/or impregnation of Hyaluronic Acid (HA) can enhance Aragonite-based scaffold properties in the regeneration of cartilage and bone. Methods: A new bi-phasic osteochondral implants (Agili-C, Cartiheal, Israel) were prepared using coralline aragonite. 28 osteochondral defects (6 mm diameter and 8 mm depth) were created in the medial and the lateral femoral condyles of 14 mature goats. Animals were randomized into different groups: bi-phasic scaffold with drilled channels at the cartilage phase, bi-phasic scaffold with drilled channels at the bone phase, native unmodified coral (mono phasic) and control (empty defect). Within each of the treated groups half of the implants were impregnated with HA at the cartilage phase. Six months after surgery the animals were sacrificed and regeneration properties were evaluated. Results: The group with the mechanical modification and impregnation of HA at the cartilage phase out-performed all other types of biphasic implants, as well as the native coral and the empty defect. In this group the implants were completely biodegraded, and were replaced by a newly formed hyaline articular cartilage and subchondral bone. Moreover, in this group the regenerated cartilage showed a smooth contour and was well integrated into the adjacent native cartilage. Evidence of hyaline cartilage regeneration was confirmed by the marked presence of proteoglycans, a marked grade collagen type II and absence of collagen type I. The native unmodified corals, and the empty defects, showed the lowest healing performance. Conclusions: The effect of the mechanical modifications together with the composite aragonite-HA in the cartilage phase turned out to be an optimal scaffold, showing a high level of healing performance for the treatment of osteochondral defects with both articular hyaline cartilage and subchondral bone regeneration.

P27-615 An extra-capsular load absorber as an early surgical option for patients with medial knee OA N.J. London1, Z. Dahabreh1, C. Waller2, D. Hayes3, COAST Study Group 1 Harrogate and District NHS Foundation Trust, Harrogate, United Kingdom, 2St Vincent’s Private Hospital, Darlinghurst, Australia, 3 Brisbane Private Hospital, Brisbane, Australia Objectives: Treatment of osteoarthritis (OA) is a dilemma for young, active patients who have failed conservative therapies but are not ideal candidates for anatomy-altering surgery (arthroplasty or osteotomy). Increasing obesity rates and high activity demands of these patients contribute to the problem. Often, these patients face a debilitating spiral of disease progression, increasing pain, and decreasing activity until they finally become ‘suitable’ arthroplasty patients. The KineSpring System (Moximed, Inc., USA), an implantable load absorbing device, was evaluated for the treatment of medial knee OA in this patient population. The extra-capsular implant is interesting as a first surgical implant for young and active patients. By preserving bone and the native joint capsule, the procedure is a natural extension beyond early cartilage repair and does not preclude future intraarticular treatments if needed. The purpose of the study was to evaluate the KineSpring System as a treatment option for patients with medial knee OA. Methods: The KineSpring System was implanted in 97 patients, with the longest duration exceeding 3 years. The treated group had medial knee OA, included younger or more active OA sufferers (mean 52.6 years, range 31.7–75.1), with a mean BMI = 30.3 kg/m2 (range

S273 20.9–44.6). Adverse events and clinical outcomes using the WOMAC and Visual Analog Scale (VAS) scores were recorded at pre-op, postop, 2 and 6 weeks, and 3, 6, 12 and 24 months post-op. Results: Mean procedure time was 69.1 min (SD 16.6) and mean hospital length of stay was 1.5 days (SD 1.4). Patients recovered rapidly, achieving full weight bearing within 1–2 wks and normal range of motion after 6 weeks. Patients experienced clinically meaningful and statistically significant early pain reduction by 6 weeks, and the results remained durable at 24 months. Mean WOMAC pain scores (0–100 scale) improved from 45.1 at pre-op to 10.8 at 24 months (p \ 0.001), with 90% of patients reporting improvement [30% at 24 months. Mean WOMAC function scores (0–100 scale) showed similar improvement from 44.3 at pre-op to 19.5 at 24 months (p \ 0.001), with 85% of patients reporting improvement [30% at 24 months. Patients achieved similar pain relief as measured by the VAS scores, which improved from 59.0 to 25.0 (p \ 0.001). Complications in the initial phase have been effectively resolved through revised surgical technique and minor design modifications and confirmed with the current design in subsequent clinical studies. Conclusions: The KineSpring System should be considered an early surgical option for patients with medial knee OA. The substantial pain relief, short recovery time, and preservation of native anatomy position this procedure as a first surgical treatment beyond early cartilage repair.

P27-668 Predicting symptomatic progression of osteoarthritis A. Ghodadra1, F. Sakamoto2, P. Rajiah2, C. Winalski2, A. Miniaci3, M. Jones4 1 Cleveland Clinic, Lerner College of Medicine, Cleveland, United States, 2Cleveland Clinic, Imaging Institute, Cleveland, United States, 3 Cleveland Clinic, Department of Orthopaedic Surgery, Cleveland, United States, 4Cleveland Clinic, Cleveland, United States Objectives: The goal of the proposed study is to analyze MRI’s from the Osteoarthritis Initiative (OAI) and determine which baseline MRI findings best correlate with symptomatic osteoarthritis (OA) progression. Methods: Data were taken from the progression subcohort of the OAI. The OAI is a multi-center, four-year observational study focusing on OA incidence and progression. Patient-reported outcomes included four KOOS subscores. Performance measures included the 20-m walk pace and chair stand pace. Outcome measures at baseline and 3-year follow up were included in our analyses. Baseline MRI images of 140 patients randomly selected from the OAI were scored using portions of the Boston-Leeds Osteoarthritis Knee Score (BLOKS). Cartilage, bone marrow lesions (BML), meniscus morphology and meniscus extrusion were scored. Scores were weighted and combined within physiologic compartments (medial, lateral and patellofemoral). BML size was given the most weight within BML score and cartilage loss size was given the most weight within Cartilage score. Linear regression modeling was implemented via the statistical software package JMP 9 (SAS Institute, Cary, NC). Each model consisted of one Outcome Measure and 14 predictor variables (3 demographic variables, 10 MRI scores, 1 baseline symptom score). Results: Patients with higher scores at baseline tended to have higher scores at follow-up for all outcome measures. For 3-year follow-up, patellofemoral (PF) cartilage loss and lateral meniscus extrusion correlated with clinically significant worse KOOS Quality of Life (QOL) scores (Fig. 1) and a trend towards significance for the remaining KOOS domains. Presence of patellofemoral bone marrow lesions correlated with better scores.

123

S274

Fig. 2 Modelling comparison of 3-years KOOS QoL Score. For each predictor and comparison (in parentheses) the mean change (black bar) is shown. Improvement is positive change and worsening is negative change. A result is statistically significant if the 95% confidence interval (gray bar) does not cross zero. A result is clinically significant if the mean change is [8 or \-8. Modelling comparison of 3-years KOOS QoL Score

Conclusions: Our work indicates that lateral meniscus extrusion and patellofemoral cartilage loss are clinically important correlates of future worsening of KOOS QOL. Interestingly, patients with higher PF BML scores had higher KOOS QOL scores at follow-up, which could be explained if PF BML’s are a sign of end-stage disease that cannot progress further. However, this study is limited by the sample size and larger future studies are needed.

P27-940 The evaluation of the pull out force between the osteochondral plug and the receiving tunnel in autologous osteochondral transplantation R. Fleaca1, M. Roman1, V. Oleksik2, A. Pascu2 1 University of Sibiu, Faculty of Medicine, Orthopaedics and Trauma Surgery, Sibiu, Romania, 2University of Sibiu, Faculty of Engineering, Sibiu, Romania Objectives: The stability of the plugs in Osteochondral Autologous Transplantation influence the outcome and the rehabilitation. The purpose of this study is to determine the pull out force between the plug and the receiving tunnel, with role in the primary stability of the transplant. Methods: To achieve this purpose we have performed 4 tunnels 20 mm deep in the femoral condyles of a fresh bovine femur. The tunnels were positioned to allow after the fixation of the femur at the inferior device of the stand on the Instron 5,587 Testing Machine the alignement of the tunnels in the vertical axis. Then, we have grafted 4 plugs 8 mm diameter and 15 mm long from the femoral

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 trochlea of the same femur. Each cylinder was then drilled in it’s axis and thru this hole was introduced a metal wire with a ring fixed at the base of the osteochondral plug and the free end in the superior mobile sleeper of the machine. After performing each osteochondral transplant the plug was submitted to a traction force from 0 N up to the pullout of the plug in order to determine the pullout force of the cylinder from the receiving tunnel. The exact lenght of this plugs was measured for the calculation of the pullout forces. This test was repeated 4 times, each with a new plug transplanted in a different receiving tunnel. The parameters of the loading and displacement forces of the plugs were obtained from data received from the testing machine soft. Results: The results shows that the maximum pull out force is around 500 N (average 474, 15 N), showing that the transplanted plug has a good pull out resistance. There is a good statistical distribution of the results. The pullout force during the experiment is increasing quickly until it’s maximum then it’s decreasing slowly at 100 N, when the test was stopped, considering the plug pulled out. The plug is moving inside the receiving tunnel 5–6 mm (35–40%) of it’s lenght until it reach the maximum pull out force, at the final of the test being at approximately 14 mm (80–90%) outside the femur. One can see also that on the descending slope of the pullout force versus cylinder displacement curve that there are some small variations of the pullout force, due probably to the fact that the holding is not constant on all the exterior surface of the plug because both the plug and the receiving tunnel are not geometrically perfect. This fact is not influencing significant the pullout force, because all these inperfections are producing effects after the maximum pullout force was reached. Conclusions: The osteochondral plug has a good pull out resistance. The pullout force during the experiment is increasing quickly until it’s maximum. The plug is moving inside the tunnel 35–40% of it’s lenght until it reach the maximum pull out force. The holding is not constant on all the exterior surface of the plug because both the plug and the receiving tunnel are not geometrically perfect.

P27-970 Newly developed nano-structural bio-mimetic scaffold for the treatment of osteochondral defects at the knee. Preliminary results S. Alevrogiannis1, G. Skarpas1 1 Athens Sports Clinic, Athens, Greece Objectives: To report our preliminary results in using a novel nano composite multilayer biomaterial for the treatment of multifocal cartilage lesions at the knee. Methods: We assessed 11 non elite athletes, with a mean age of 42, suffering from chondral defects. The mean size of the lesions was 5.6 cm2. 5 of them had an osteochondral defect of the MFC due to OCD and the remaining 6 of them had a cartilage lesion (all type IV in Outerbridge scale) of the weight bearing surface of MFC due to previous trauma (4) and trochlea area (2), due to patellar mal-alignment. In 2 of them the chondroplasty was accompanied by ACL reconstruction and in another 2 cases by a MOWHTO simoultaneously due to axial deformity. All of them were treated successfully in one stage procedure via mini arthrotomy, using a newly developed biomimetic nanostructured osteochondral bioactive scaffold. All the patients underwent a specialized rehab protocol. Preop. and postoperative evaluation of patients was performed using the Modified Cincinatti (MC) Rating System, the Visual Analogue Pain score (VAS) and the IKDC Knee examination score and Patient Outcome Function score. MRI was performed to all of them at 1 year post op. and the MOCART score was used for evaluation.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: All cases progressed uneventfully. We evaluated the patients at 6 and 12 months post-op. The IKDC score improved significantly while there was a great progress in Patient Outcome Function. Pain significantly reduced in VAS. MRI evaluation showed adequate filling of the defect without significant subchondral bone oedema and a hyaline-like signal of the implant. All patients were pain free, had full ROM and returned to their previous athletic activities with no reduction in performance. Conclusions: Promising results have been obtained using this novel nano- composite bioactive scaffold. There is no need for a long learning curve for the surgeon and specific instrumentation. Longer follow-up and greater number of cases, as well as multicenter trials, are needed to prove the efficacy of the method.

P27-993 Platelet Rich Plasma (PRP)–a new hope for degenerative lesions of the knee A. Gobbi1, A. Kumar1, G. Karnatzikos1 1 Orthopaedic Arthroscopic Surgery International, Department of Sport and Medicine, Milano, Italy Objectives: Articular cartilage has limited intrinsic healing potential; therefore trauma and/or chronic irritation may lead to progressive damage, joint degeneration and early OA. The purpose of our study was to determine the efficacy of the use of platelet rich plasma (PRP) obtained from a simple autologous blood extraction in the degenerative lesions of the knee. Methods: We prospectively followed up 80 patients with degenerative lesions of the knee (mean age 46.7 years) with a minimum follow up of 12 months. All patients were treated with 2 intraarticular injections at monthly interval with autologous plasma rich in growth factors (PRGF) utilizing the kit provided by RegenLabPRP. Forty of the patients (50%) had undergone a previous surgery on the ipsilateral knee. VAS and KOOS scores were collected at 6 and 12 months post treatment. Non-parametric analysis was performed with the Wilcoxon rank test to compare the variations of the KOOS, VAS, Tegner, IKDC and MARX scores from preinjection to 6 and12 months. The non-parametric Mann–Whitney U test was performed to analyze difference in improvement between the subgroups of patients with or without previous surgery and between male and female. All reported p values were two tailed, with an alpha level of 0.05 indicating significance. Results: Patients showed significant improvement in all scores at final follow-up (p \ .005). Mean pre-treatment values were: KOOS Scores: p = 73.6/S = 72/ADL = 77.8/SP = 42.3/QOL = 41.3, VAS 4.1, Tegner 3.6, IKDC 53.4 and Marx 3.8. At final follow up mean scores were: KOOS Scores: p = 88.1/S = 86.0/ADL = 94.8/SP = 64.2/ QOL = 67.8, VAS 1.3, Tegner 5.2, IKDC 68.5 and MARX 9.4. There was significant difference between ‘‘patients with’’ and ‘‘patients without previous surgery’’ only in VAS score, which showed higher reduction in the 1st group, in KOOS (Sport) subgroup from preinjection to 6 month follow up and in KOOS (QOL) subgroup from 6 to 12 month follow up scores, which showed higher improvement in the 2nd group. There was no significant difference in improvement between males and females. Conclusions: This study showed that the use of PRP could act as a new hope as a preventive agent in patients with chronic and degenerative disease of the knee by diminishing pain and improving symptoms and quality of life.

S275 P27-1022 The use of Agili-C, a novel aragonite-hyluronate biphasic implant, in the regeneration of chondral and osteochondral defects: initial clinical results M. Drobnic1, E. Kon2, W. Widuchowski3, K.R. Zaslav4, D. Robinson5, A.S. Levy6 1 University Medical Centre Ljubljana, Department of Orthopaedic Surgery, Ljubljana, Slovenia, 2IOR, IX Divisione-Biomechanics Lab, Bologna, Italy, 3District Hospital of Orthopedics and Trauma Surgery, Piekary Slaskie, Poland, 4Advanced Orthopedic Centers Richmond Virginia, Richmond VA, United States, 5Hasharon Hospital, Rabin Medical Center, Department of Orthopedic Surgery, Petah Tikwa, Israel, 6Center for Advanced Sports Medicine, Knee and Shoulder, Millburn NJ, United States Objectives: After successful healing performance of a novel biphasic aragonite-hyluronate implant (Agili-C, CartiHeal, Israel) in the goat model, resulting in regeneration of hyaline articular cartilage and subchondral bone, a prospective non-randomized multi-center study was initiated. We herein present the initial clinical and MR results of the first patients. Updated data will be given during the presentation. Methods: A 2-year long multi-center study aimed to recruit 20 patients with symptomatic focal, isolated, osteochondral or chondral lesions in the knee joint. The lesion size limitation was set at 2 cm2. Concomitant pathology (e.g. ACL reconstruction or partial menisectomy) can be treated prior or during Agili-C implantation. Four patients in two centers were already operated on by the time of at abstract submission. So far, in each one of the cases two devices of different dimensions (e.g. 8–10 mm in diameter, and 10–15 mm in length) were used to cover the lesions. Post-operative patients practiced touch-toe weight-bearing for the first 3 weeks, advancing to full weight-bearing. Results: No adverse clinical findings were noted in any of the patients. The patients regained their full knee ROM in 4 weeks. A 3-month radiographic and MR analysis (at present for one patient only) demonstrated good incorporation of the implant. Ongoing implant absorption was noted, which was replaced with regenerated tissue. Clinical examination according to IKDC was ‘‘normal’’. Conclusions: In the very limited case series, with short follow-up so far, our observations are that the implant has not caused any adverse effects, that the surgical implantation is technically undemanding and fast, initial implant absorption was noted already during the 3 month follow-up visit, and that the patients are in good clinical condition.

P27-1153 Positive reaction of bone and surrounding cartilage to a metal implant after 12 months in the sheep’s knee N. Martinez-Carranza1, H. Berg1, H. Nurmi-Sandh2, L.E. Ryd1, A.-S. Lagerstedt2 1 CLINTEC, Karolinska Institute, Department of Orthopedics, Karolinska University Hospital, Stockholm, Sweden, 2Inst. of Clinical Sciences, Swedish Agricultural University, Uppsala, Sweden Objectives: Treatment of cartilage injuries is still a challange since hyaline cartilage does not heal by itself. Biological methods such as ACI, MACI and microfracturing are used as well as methods to ‘‘move cartilage’’ like mosiacplasty and allografts. An alternative method is the implantation of biomaterial into a cartilage defect. For such a method to work, the implant must adhere to the underlying bone and be accepted by the surrounding cartilage. We have tested a concept where a metal implant (Cr-Co), coated with a double layer of

123

S276 first titanium and, most superficially, hydroxyapatite of the surfaces facing bone and surrounding cartilage was tested in sheep. Methods: 6 female sheep (mean age 4 years, mean weight 82 kg) were operated on the medial femoral condyle of one knee. A defect, 7 mm in diameter, was created and an implant was immediately inserted. The implant had a double-curved outer surface coinciding with the curvature of a ‘‘standard sheep knee’’. The under surface was linear with a 7 mm long, 2 mm thick pin protruding perpendicularly into the bone. This pin was inserted into a 1.8 mm predrilled hole for immediate interference fit. The implantation was such that the outer surface was counter-sunk approx. 0.5 mm into the surrounding cartilage and the implant protruded approx. 1 mm into the subchondral bone. The implant articulated against the natural tibial condyle cartilage. After 12 months the animals were sacrificed and the knees with implants were sectioned for histological studies with the implants in situ. Results: All animals moved freely and symmetrically 2–3 weeks postoperatively. Histologic examination showed excellent bony ongrowth close to 100% in all 6 implants. Where cartilage met the hydroxiapatite-coated periphery of the implant, the cartilage adhered closely such that a virtual ‘‘seal’’ appeared to be in place. Conclusions: A small metal implant that rests on subchondral bone with a 2 mm pin inserted in a undersized hole for immediate fixation and with hydroxyapatite for permanent fixation showed excellent bonding to bone after 12 months. An apparent seal between the surrounding cartilage and the implant suggests a protective effect against joint fluid intrusion. Such an implant may show excellent long-term bonding in the joint.

P27-1289 Radiosynoviorthesis improves functional outcome after arthroscopic synovectomy B. Pompe1, L. Suhodolcan2, M. Grmek3, O. Zupanc2 1 Clinical Center Ljubljana, Department of Orthopaedic Surgery, Ljubljana, Slovenia, 2Medical Centre Ljubljana, Department of Orthopaedic Surgery, Ljubljana, Slovenia, 3Medical Centre Ljubljana, Department of Nuclear medicine, Ljublajna, Slovenia Objectives: The report evaluated effects of radiosynoviorthesis (RS) subsequent to arthroscopic synovectomy and to disclose indications for its use. Methods: Single joint in 10 patients was treated with combination of arthroscopic synovectomy followed by RS, between January 2007 and June 2010. Four patients had rheumatoid arthritis, four patients had psoriatic arthritis and two had nonspecific chronic synovitis. RS, following EANM procedure guidelines for radiosynovectomy, were performed 2–6 months after arthroscopic resection of actively inflamed synovium, using shaver. For RS of knee joint (9 patients) 185 MBq of 90Y colloid was used. In one patient with elbow involvement, 186Re colloid activity 74 MBq was used for RS. Intraarticular application of the radiopharmaceutical was performed by an experienced orthopaedic surgeon. After application of the radiotherapeutical, the joint was immobilized in a plaster splint for 4 days. The effectiveness of treatment was evaluated at regular follow-up visits (up to 12 months after RS) and special attention was paid to range of motion, presence of joint effusion and patient satisfaction. Results: The patient satisfaction and functional outcome following RS improved in 8 patients (80% of treatments), which consisted of improved range of motion in 2, no recurrence of joint effusion in 2, or both in 4 patients. RS was not effective in 2 cases (20% of treatments), in one case, joint effusion persisted, where the other suffered from septic arthritis 5 weeks after RS. Our recommendations for use of RS after arthroscopic synovectomy include: • Insufficient removal of the inflamed synovial membrane.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 • •

Persistent joint effusions. Limited range of motion.

Conclusions: This report demonstrated that the chronic inflammed synovitis can been successfully treated by combination of arthroscopic synovectomy and RS, where residua of hypertrophic synovial membrane can only be effectively eliminated with radiosynoviorthesis. For favourable results close collaboration between orthopaedic surgeon and nuclear medicine physician is essential.

P27-1307 Synthetic polymer scaffolds vs. microfracture: two year follow-up prospective randomised trial G. Camillieri1, V. Calvisi2, M. Ferretti3 1 II Faculty, Roma, Italy, 2University of L’Aquila, Rome, Italy, 3Villa Aurora Hopsital, Rome, Italy Objectives: To evaluate the clinical outcome and MRI findings of synthetic resorbable biphasic implant (TruFit Plug; Smith & Nephew, San Antonio, TX) that immediately fill a chondral or osteochondral defect vs. microfracture at 6 months, 1–2 years of follow-up. Methods: 36 consecutive patients were included in the study and randomized in two groups and evaluated clinically using the Cartilage Standard Evaluation Form as proposed by ICRS and the Tegner score. For the description and evaluation of MRI findings, we employed the MOCART-scoring system. SPSS software was used for statistical analysis. Limit of confidence was set at p \ 0.05. Results: A statistically significant improvement in all clinical scores was observed at 1 year in both groups. At 2 years follow-up the TruFit group showed better clinical results than microfracture group. Actually, at 2 years follow-up, the microfracture group showed decreased clinical and MRI scores than 1 year follow-up. The total MOCART score and the signal intensity of the repair tissue were statistically correlated to the IKDC subjective evaluation. Any correlation between size and position of the treated cartilage lesions, with clinical outcomes and MRI findings were found regarding to TruFit Plug. Seven implants showed a delayed incorporation into the bone (MRI evaluation). Three of them had worse clinical outcomes. Second look arthroscopy was performed in six patients (4 TruFit, 2 Microfracture). Conclusions: At 2 year FU, TruFit implants showed better and more stable clinical outcomes than microfracture. MRI (MOCART score) is a reliable method to evaluate the integration and evolution of cartilage procedures showing better findings for the TruFit group at 2 years FU. Longer FU and further studies are necessary to confirm the impact of synthetic scaffolds on cartilage and osteochondral repair.

P27-1313 Outcomes and survivorship at 10 years following arthroscopic treatment package for osteoarthritis of the knee J.R. Steadman1, H.B. Ellis2, K. Briggs3, L.M. Matheny3 1 The Steadman Clinic and, Steadman Philippon Research Institute, Vail, United States, 2Steadman Philippon Research Institute, Vail, United States, 3Steadman Philippon Research Institute, Clinical Research, Vail, United States Objectives: The purpose of this study was to document long-term outcomes and total knee arthroplasty (TKA) rate following arthroscopic treatment for osteoarthritis, and to identify risk factors associated with total knee arthroplasty. Methods: Eighty-one knees in 73 patients (49 males, 32 females), with a mean age of 58 years (range:37–79 years) who were treated with an arthroscopic regimen for knee OA between August 2000 and

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 November 2001 were included in this Institutional Review Boardapproved study. Inclusion criteria were patients with severe OA determined by radiographic and intraoperative criteria who had failed conservative therapy. All patients had Kellgren-Lawrence (KL) grade 3 or 4 and were candidates for knee arthroplasty; however, did not wish to undergo arthroplasty for various reasons. Arthroscopic data were prospectively collected. Arthroscopic regimen included joint insufflation, lysis of adhesions, anterior interval release, contouring of cartilage defects to stable rim, shaping of meniscus tears to stable rim, synovectomy, removal of loose bodies and osteophytes that affect terminal extension. Survivorship analysis was performed. Endpoint was defined as total knee arthroplasty (TKA). In patients who did not undergo TKA, outcome measures were collected at a minimum 10 years following treatment which included, Lysholm, Tegner Activity Scale, patient satisfaction with outcome and WOMAC. Results: Of 81 knees, seven were decreased and two refused to participate, leaving 72 available for follow-up. Follow-up was obtained on 96% of patients (n = 69). Forty-three knees (60%) were converted to TKA at a mean of 4.4 years (range:1.0–9.6 years) following index knee arthroscopy. Repeat arthroscopy was performed in 15 knees (21%) due to recurrent mechanical symptoms. Survivorship analysis showed mean survival time of 6.8 years[95%CI: 5.9–7.6 years]. Survivorship was 60% at 5 years, 47% at 7 years and 43% at 8 years, and 40% at 9 and 10 years following index arthroscopy. There was no association between failure and gender or malalignment. Failures were significantly older compared to survivors (mean difference = 8.8 years) [95%CI: 2.7–14.9](p = 0.005). Patients who had a KL grade 4 were 5.3 times more likely to fail [95%CI: 1.3–23.4] compared to KL 3 (p = 0.012). Twenty-six knees (36%) did not undergo arthroplasty at 10 years post-arthroscopy. For patients who did not undergo arthroplasty, mean Lysholm score was 74.4 [95%CI: 67–80], median Tegner Activity Scale was 3 (range: 0–8), median patient satisfaction with outcome was 9 (range: 1–10) and mean WOMAC score was 18.5[95%CI: 13–24]. Conclusions: This study identified risk factors for total knee arthroplasty in patients with moderate to severe knee OA. Patient age and degree of osteoarthritis were shown to play a role in progression to knee arthroplasty. Although there was a wide range of end points for TKA, this study also demonstrated that older patients who were originally candidates for TKA were able to delay arthroplasty, even at 10 years for some.

P27-1331 Tissue engineering for cartilage repair: In vitro development of an osteochondral scaffold D. Deponti1, A. Di Giancamillo2, F. Gervaso3, C. Domeneghini4, A. Sannino3, G.M. Peretti1 1 University of Milan, Faculty of Exercise Sciences, Milan, Italy, 2 University of Milan, Milan, Italy, 3University of Salento, Lecce, Italy, 4University of Milan, Department of Veterinary Sciences and Technologies for Food, Faculty of Veterinary Medicine, Milan, Italy Objectives: Articular cartilage lesions have poor healing potential. Basic science and clinical investigation have led to different approaches to address this problem, in particular with the use of biocompatible scaffolds seeded with autologous chondrocytes. We developed an osteochondral scaffold with a biphasic structure: a chondral phase, designed for hosting the autologous chondrocytes,

S277 and a bony phase structured for the scaffold integration into the bone beneath the chondral layer. The aim of the present work was the in vitro optimization of the chondral phase, and, in particular, the analysis of the effect of the addition of the fibrin glue as embedding scaffold for the seeded chondrocytes. Methods: First, fresh chondrocytes were isolated from swine articular cartilage and seeded onto the chondral scaffold: some samples were seeded with cells resuspended in medium, other samples were seeded with cells resuspended in fibrinogen that was then polymerized into fibrin glue by the addiction of thrombin. Then, in the second part of the study, different in vitro times were assessed for the optimal in vitro maturation of the chondral composite. First, chondrocytes were isolated from swine articular cartilage and expanded in the presence of specific growth factors (FGF-2 and TGFb1) to stimulate proliferation and the maintenance of a chondral phenotype, then they were resuspended in fibrinogen and seeded onto the chondral composite that was cultured in vitro for 1, 3 and 5 weeks in the presence of TGFb3: the optimal time for a preliminary maturation of the composite was identified by morphological, phenotypical and biomechanical analyses. Results: The data obtained by histological and immunohistochemical analyses of the samples demonstrated that the presence of fibrin glue ameliorated cell distribution and survival into the chondral composite. Data from the second part of the study showed that chondrocytes’ phenotype was rescued after 3 weeks of in vitro culture and maintained for the following weeks; the biomechanical properties improved during time, but cell survival decreased after 3 weeks, probably because of the lack of physiological signals, suggesting that the in vitro culture of the chondral composite should be maintained for 3 weeks before performing the implantation in the animal model in future studies. Conclusions: These in vitro data demonstrated that chondrocytes can grow and promote the formation of a mature cartilaginous tissue when seeded on the chondral scaffold proposed in this study; their survival and activity are ameliorated by the presence of fibrin gel as embedding scaffold and by maintaining the vitro culture for 3 weeks in the presence of specific growth factors.

Knee-Patellofemoral

P28-12 Localized pigmented villonodular synovitis of the knee: clinical and MRI outcome of arthroscopic treatment with mid term follow up P. Loriaut1, P. Djian1, T. Boyer2, J.-P. Bonvarlet2, C. Delin3 1 Cabinet Goethe, Paris, France, 2IAL Nollet, Paris, France, 3 RIM Maussins-Nollet, Paris, France Objectives: The purpose of this study was to review the outcome of arthroscopic treatment of localized pigmented villonodular synovitis (LPVNS) of the knee and to determinate recurrence rate with clinical and magnetic resonance imaging (MRI) evaluation at mid term follow-up. Methods: A retrospective evaluation of 30 consecutive patients diagnosed with LPVNS of the knee and treated arthroscopically between 1990 and 2008 was performed. Assessments were based on clinical records, Lysholm Knee Scoring Scale (LKSS), operative reports, X-rays and MRI of the knee. Adequate pre operative and follow-up data were available for 20 patients (average follow-up 75 months). Clinical and MRI recurrence were analysed and recurrence-free survival was measured by the actuarial method of Kaplan and Meier.

123

S278 Results: The mean age at presentation was 46 years.

Incidence of localized pigmented villonodular syno Symptoms were discomfort of the knee (100%), swelling (90%), locking (50%), joint line pain (10%), palpable mass (15%) and an history of knee trauma (10%). At surgery, nodules were localized in the gutters (45%), the suprapatellar pouch (26%), the patellar fat pad (13%), the posterior compartment (13%) and the femoral notch (9%). No postoperative complications occured. The mean preoperative and last follow-up LKSS were respectively 56.7% and 85.3%. Clinical and MRI recurrence free survival were respectively of 90% and 80% at 6 years.

Clinical recurrence survival free curve

MRI recurrence survival free curve

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: MRI is essential for diagnosis and treatment strategy of LPVNS. Arthroscopy is a safe procedure for effective resection of LPVNS. Long term follow-up is recommended as clinical and imaging recurrences occurs.

P28-23 Does pre-op chondral damage predispose to subsequent patellofemoral arthritis? 10–15 year follow up results of athroscopic Elmslie-trillat osteotomy from Bristol M. Naveed1, C. Ackroyd2, A. Porteous3 1 North Bristol Trust, Southmead Hospital, Bristol, United Kingdom, 2 Avon Orthopaedic Centre, Orthopaedics, Bristol, United Kingdom, 3 Avon Orthopaedic Centre, Southmead Hospital, Bristol, United Kingdom Objectives: Many investigators have documented patello femoral arthritis following Elmslie-trillat tibial tubercle osteotomy in long term. We wanted to answer question if pre existing chondral damage is the cause of subsequent arthritis therefore we documented arthroscopic cartilage condition at the time of operation. Methods: 31 patients (34 knees) underwent arthroscopy at the time of Elmslie–Trillat tibial tubercle transfer and cartilage condition was documented. At latest follow up 25 patients (28 knees) were available for review. All patients had pre-op, post op and 10–15 year follow up xrays and Cox Insall scores. Statistical analysis was done by using student’s t test and means of pre-op and follow up results calculated. Results: At 10–15 year follow up, 5 (20%) knees with no pre-op chondral damage had no signs of arthritis, 13 knees (45%) with grade 1–2 chondral damage had early to moderate signs of arthritis and 10 knees (35%) with grade 3–4 damage 4 showed moderate to high grade arthritis (4 knees had undergone arthroplasty). All patients had poor Cox/insall scores pre op but improved to 70% at early follow up and it was 50% at 10–15 year follow up. 9 patients had screw removal. Conclusions: Patients with pre-existing chondral damage developed long term arthritis, higher the grade of arthritis, worse the outcome. Fixation screws were a problem and many required removal.

P28-201 ‘Basket weave technique’ for MPFL reconstruction: a prospective study P. Kodkani1 1 Arthroscopy & Sports Medicine Institute, Mumbai, India Objectives: Medial patellofemoral ligament (MPFL) of the knee is known to be the primary medial stabilizer for the patella. The MPFL also has a poor healing capacity with poor results following repairs and medial plication. An MPFL reconstruction therefore is a necessity following a patellar dislocation to prevent recurrence. Described techniques for MPFL reconstruction have their own pitfalls and disadvantages. To overcome these disadvantages and complications, a new technique of reconstruction was implemented. Its efficacy and results assessed prospectively. Methods: 23 knees of lateral patellar dislocation were treated in the past 26 months by MPFL reconstruction. 19 were females and 4 males. Age ranged from 13 to 45 years, an average of 25 years. Average follow up was for 11 months. Cases excluded were those with single traumatic episode without any osteochondral injuries, cases presenting with only instability without any dislocations, cases requiring any supplementary procedure for trochlear dysplasia or extensor realignment. The cases included had more than one episode of dislocation or had associated osteochondral injury in first time dislocations. 5 knees presented with loose bodies which were

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 removed arthroscopically. 4 cases having a tight lateral retinaculum underwent an arthroscopic lateral release. Reconstruction was performed with hamstring autograft from the ipsilateral knee. To avoid bone tunnels, implants and its pitfalls, a soft tissue fixation was used. An anatomical MPFL reconstruction was performed. The graft was looped around a ligamento-periosteal sleeve from the saddle groove in between the adductor tubercle and the medial epicondyle which is the anatomical attachment of MPFL. It was sutured for fixation. The two limbs of the graft were routed between the first and second layer of the medial retinaculum along the path of the MPFL. They were sutured to the extensor retinaculum over the patella in a ‘basket weave technique’ at the anatomical insertion site. The patients were followed up prospectively at 1 month intervals. Results: All achieved full range of motion and normal mediolateral stability. There was no recurrence of dislocation. No major complications were encountered. 3 cases actively involved in sports could return back to their sporting activities. 5 cases with osteochondral fractures had occasional pain following overactivity. Osteochondral fracture therefore was a relatively poor prognostic factor. Average Kujala score was 94.3. KOOS score was near normal in all, except for patients with osteochondral injury where it was compromised due to occasional pain and swelling. Conclusions: This new method of MPFL reconstruction with soft tissue fixation gives excellent results comparable to other series. It avoids complications related to bone tunnelling and implants. It can therefore also be implemented in skeletally immature knees. It is a safe, effective, reliable and reproducible technique. Therefore ideally suited for MPFL reconstructions.

P28-225 Atraumatic locked knee from superior dislocation of the patella-diagnosis and management of a rare injury M.A. Siddiqui1, M.H. Tan1 1 Singapore General Hospital, Department of Orthopaedic Surgery, Singapore, Singapore Objectives: Evaluation of a patient with a locked knee without any history of trauma. Subtle points that may alert the clinician to the diagnosis of locked knee secondary to interlocking osteophytes between the patella and femoral condyle is discussed. Subsequent management of this very rare injury is also presented. Methods: The patient0 s dislocation was reduced under sedation. Results: X-rays and 3-dimensional computed tomography scans of the knee reveals the corresponding locking osteophytes which caused patella locking. Patella tendon ruptures need and can be differentiated from superior dislocation of the patella. Conclusions: There are \20 reported cases of superior dislocation of the patella producing a locked knee. This rare case discusses tips in diagnosis and subsequent management of the patient that comes in through the emergency department.

P28-234 Correlation between different patella alta measurements in patients with and without patella-femoral instability S. Said1, P. Faunoe2, B. Lund3, S.E. Christiansen4, M. Lind5 1˚ ˚ rhus, Denmark, 2Aarhus University Arhus University Hospital, A Hospital, Department of Sports Traumatology, Aarhus, Denmark, 3 University Hospital of Aarhus, Department of Sportstraumatology, Aarhus C, Denmark, 4University of Aarhus, Sports Trauma Division, Orthopedic Department, Skanderborg, Denmark, 5Aarhus University Hospital, Division of Sportstrauma, Aarhus C, Denmark Objectives: There is an increasing focus on the influence of patella alta in the pathogenesis and surgical treatment of patello-femoral instability. However there is no consensus of which radiological methods is reliable for patella alta characterization and surgical

S279 decision-making. The present study aims to determine the correlation between three different MRI based patella alta measurements in patients with and without patella instability. Methods: MRI scanning of 100 patients was analyzed. There were 50 patients with patella instability and 50 patients with meniscus lesions without patella instability representing a control group. The CatonDeschamp (CD), Insall-Salvati (IS) and Patella-trochlear index (PT) was measured and calculated. Correlations between the different indexes were determined with non-parametric correlation test. Patella-femoral dysplasia, sulcus angle and Tibia Tuberosity trochlea Groove (TTTG) distance were determined. Results: Patients with patella instability had significantly higher TTTG distance, sulcus angel and patella alta index measured with CD and IS methods. The PT patella alta index was not correlated to patella instability. Patella alta index measured with CD and IS were significantly correlated in patella instability group (p [ 0.00) and in the control group (p [ 0.003), whereas the PT index was not correlated to CD and IS indexes. Conclusions: Increased Patella Alta index measured with SI and CD methods, TTTG distance and sulcus angel are characteristic for patella instability patients, while patella-trochlea index is not related to patella instability patients.. The Patella-trochlea index did not correlate to the CD and IS indexes or patella instability and is thus not suitable for evaluation of patella instability.

P28-295 Treatment of persistent patellofemoral instability after failed previous surgery L. Kohn1, G. Meidinger2, K. Beitzel3, I. Banke2,4, A.B. Imhoff5, P. Scho¨ttle6 1 Technische Universita¨tsklinik, Klinikum rechts der Isar, Abt. fu¨r Orthopa¨die und Sportmedizin, Mu¨nchen, Germany, 2Klinikum rechts der Isar, TU Mu¨nchen, Abteilung und Poliklinik fu¨r Sportorthopa¨die, Mu¨nchen, Germany, 3University of Connecticut Health Center, Department of Orthopaedics, West Hartford, United States, 4 Technische Universita¨t, Abt. fu¨r Orthopa¨dische Sportmedizin, Mu¨nchen, Germany, 5TU Mu¨nchen, Klinikum rechts der Isar, Abteilung und Poliklinik fu¨r Sportorthopa¨die, Mu¨nchen, Germany, 6 Orthopa¨die am Zu¨richberg, Praxis Munzinger, Zu¨rich, Switzerland Objectives: Overviewing the literature, pain and redislocation after surgical treatment of patellofemoral instability (PFI) is described with up to 30%, especially with techniques involving the extensor mechanism, the lateral retinaculum or the medial vastus. However, outcome data about revision surgery is missing. Therefore, it was the purpose of this prospective study to evaluate the clinical outcome after revision surgery with an isolated or a combined reco. of the medial patellofemoral ligament (MPFL) with a follow up of 24 months. The indication for additional procedures such as distal femoral osteotomies (DFO), trochleoplasty, reclosing of the lateral retinaculum (RLR) or lateralisation of the tuberosity (LT) were performed depending on the clinical and radiological pathomorphologies. Methods: In between 3/07 and 4/09, 28 pat. with a mean age of 24 years (13–46 years) were revised due to persistent PFI after mean 1.5 failed previous operations (lateral release, medial gather/VMOdistal., medial. of the tuberosity) in our department. An isolated reco. of the MPFL was performed in 12 cases, while a combination with a DFO due to massive femoral axis deformity (n = 4), trochleoplasty due to a convex trochlear morphology (n = 1) and/or LT (n = 1) and/ or RLR due to medial instability or lateral pain (n = 12) was done in 16 cases. The clinical result was evaluated by the pre- and postop. IKDC/Kujala/Tegner—score and by a subjective questionnaire. Radiologically, the patellashift/-tilt/-height and level of the eventual degeneration were defined preop. and at the point of follow up with the help of straight lateral radiographs and by MRI. Significance level

123

S280 was set at p = .05, statistical calculation was done by the use of the t test. Results: 89% were very satisfied/satisfied with the treatment. None redislocation of the patella occurred during the follow up. We could evaluate a significant decrease of the pain during daily activities. The IKDC/Kujala/Tegner score have shown a significant increase overall. However, a significant difference concerning the scores could not be noticed in between the groups without and with additional procedures. Patellar shift, tilt, height decreased significantly to anatomical values. The level of preexisting degeneration showed no aggravation. Conclusions: Since it is known that PFI is a multifactorial problem, revision surgery should set about only after comprehensive examinations. Regarding our results, isolated or combined MPFL reco. seems to be a very effective treatment of recurrent patella dislocation and results in a significant increase of stability, functionality as well as in a reduction of pain. Additional pathomorphologies of the bony structures or a lateral release, which causes pain or an aggravation of instability, have to be addressed additionally to reach same results like with an isolated therapy. Level of Evidence: II

P28-306 Initial validity and reliability of the Banff Patellar Instability Instrument (BPII) L. Hiemstra1, M. Lafave2, S. Kerslake3, S.M. Heard1, G. Buchko1 1 Banff Sport Medicine, Orthopaedic Surgery, Banff, Canada, 2Mount Royal University, Physical Education, Calgary, Canada, 3Banff Sport Medicine, Research, Banff, Canada Objectives: Patellar instability is a common problem yet there are currently no published, patient-reported, outcome measures that are disease-specific for this population. The purpose of this study was to determine if the Banff Patellar Instability Instrument (BPII) is a valid and reliable outcome for measuring quality of life in subjects with recurrent patellar instability. Methods: A modified Ebel procedure was employed to validate the content of the BPII. This procedure is a three stage process consisting of: 1. simple validation by a local group of experts; 2. formal adoption from an international group of experts; 3. face to face discussion of international experts followed by modification of the instrument if 80% agreement did not occur in the second stage. As a measure of internal consistency, Cronbach’s Alpha was utilized to assess how reliably the 32 items of the BPII measured a similar construct (patellar instability). One hundred and twenty patients with a confirmed diagnosis of patella instability from one of the three orthopaedic surgeons completed the BPII at the initial orthopaedic consult (pre-surgery), 6 month follow-up appointment (post-surgery) and 12 month follow-up appointment (post-surgery). Surgical procedures included medial patellofemoral ligament (MPFL) reconstruction, MPFL imbrication, tibial tubercle transfer, knee arthroscopy or any combination thereof. Data from each item of the BPII, as well as the cumulative score, was used in the Cronbach’s Alpha Reliability Coefficient analysis. Results: Content validity was clearly established as each item in the BPII achieved a minimum of 83.3% (range 83.3–100%) agreement for relevance among the expert panellists at the second stage of the modified Ebel procedure. The average agreement was 96.9% and 24/32 questions achieved 100% agreement. Reliability of the BPII was established at the initial orthopaedic consult (reliability co-efficient = 0.91), 6 months post-operatively (reliability co-efficient = 0.96) and 12 months post-operatively (reliability co-efficient = 0.96).

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: This study has established the BPII is valid and reliable in patients with recurrent patellar instability as well as patients who have had a patellar stabilization procedure.

P28-315 Displacing lateral meniscus masquerading as patella dislocation E. Arendt1, C.A. Fontbote2 1 University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States, 2Pontificia Universidad Catolica de Chile, Orthopaedic Surgery, Santiago, Chile Objectives: A series of 11 knees in 10 patients were referred to an orthopedic surgeon with a diagnosis of recurrent lateral patella dislocation (LPD). 3 knees had undergone patella realignment surgery with continuance of symptoms. 9 of 11 knees had MR’s prior to referring MD presentation with the MR read as no meniscal pathology & no acute patella injury. All patients had ‘‘failed’’ previous management including physical therapy & the use of a patella stabilizing brace. All patients presented for a consult with a similar history; their knee locked in low flexion after a flexed knee movement. Reduction, or ability to straighten the knee, occurred either spontaneous or by forced knee extension. Reduction was associated with a sense of patella movement. Methods: Retrospective chart review. Results: All patients were treated by the same surgeon. At time of consult all patients had a physical exam that was inconsistent with patella dislocation; \3 quadrants lateral patella mobility with a solid medial check rein; no apprehension sign present. Age range at time of first symptoms was 8–63 year (mean age 26). The average time from initial knee locking event to time of definitive diagnosis by the consulting MD was 2.8 years (range 3 months to 10 years). 4 knees had ‘dislocation events’ witnessed by a MD with continued diagnosis of patella dislocation. Of 9 MR’s obtained prior to definitive treatment, 8 were read as having no lateral meniscus pathology. 5 were re-read by the consulting MD & a radiologist with concurrence of a lateral meniscal capsular fascicle tear or attenuation; 1 MR showed a displaced lateral meniscus tear. 1 patient had an ultrasound when the knee was in the locked position documenting a displaced meniscus. All knees underwent an exam under anesthesia & arthroscopy. All patients had stable patella as judged by lateral patella translation under anesthesia. At time of surgery, 6 patients had a frank tear in the lateral meniscus, all were readily displaceable. 5 knees showed a displaceable lateral meniscus that was able to traverse[50% width of the lateral femoral condyle with attenuation but not a frank tear seen in the meniscal synovial junction; 2 of 5 menisci could be displaced and ‘‘locked’’ in the forward position under anesthesia. 8 menisci were treated with repair, 3 knees underwent partial lateral meniscectomies. All patients were able to return to previous work/sport activities at a minimum patient f/u of 6 mos. All patients reported no further episodes of ‘‘knee cap dislocation’’ or symptoms of catching and locking. Conclusions: The MD treating a patient with a history of knee locking in flexion should have a high index of suspicion for a lateral meniscus tear/unstable hypermobile lateral meniscus, despite patient report of perceived patella movement. History of symptoms occurring in flexion and attention to patella physical exam are key factors in this diagnosis. Ultrasound is helpful as a dynamic tool for this diagnosis.

P28-316 Location of abnormally elevated T2-relaxtion values in patellar cartilage correlates with the degree of patella alta E. Arendt1, N. Lemeister2, J.M. Ellermann3 1 University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States, 2University of Minnesota, Medical

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 School, Minneapolis, United States, 3University of Minnesota, Department of Radiology, Center for Magnetic Resonance Research, Minneapolis, United States Objectives: Patella alta is a known risk factor for patellofemoral (PF) joint dysfunction. The increased vertical position of the patella is an important structural variable that results in reduced contact area in patients with patella alta. This reduced contact area can result in increased PF stress (PF joint reaction force/PF contact area) which can lead to altered cartilage homeostasis. T2 relaxation time mapping provides information about the structural integrity of cartilage extracellular matrix, specifically the collagen network. T2 values have been reported to increase in spontaneous degeneration of bovine cartilage, osteoarthritis, and with age. We hypothesized that increased cartilage stress would result in altered cartilage health as evidenced by elevated T2 values, and that the geographic location of this elevated area would correlate with patella height as determined by the InsallSalvati (I/S) ratio. Methods: Clinical 3T Magnetic resonance imaging (MRI) was performed on 7 patients with a history of PF pain &/or instability. Patella height was measured utilizing the I/S ratio (range 1.0–1.7). Multiecho Spinecho images in the axial plane at the level of the patella were obtained; subsequently T2 maps were calculated from these images. The patellar cartilage was delineated and for each patient areas of elevated T2 relaxation times were identified on the maps. Using a fat suppressed sagittal fluid sensitive proton density image, the distance (mm) from the most inferior border of the patellar cartilage to the site of elevated T2 values was measured. This measurement was normalized by dividing it by the total length of patellar cartilage from inferior to superior. The two sets of data were plotted against each other with I/S index as the independent variable and the normalized distance to elevated T2 relaxation values in ms as the dependent variable at which time a regression analysis was performed (Fig. 1). Results: Negative correlation between I/S and the first (inferior) slice showing a region with elevated T2 (Fig. 2). Conclusions: Cartilage degradation as defined by increased T2 relaxation values are associated with patella alta in their geographic location on the patella in this small patient population. Further longitudinal studies correlating quantitative cartilage mapping with

Fig. 1

Fig. 2

S281 clinical symptoms and PF morphology may aide in understanding the pathogenesis of PF pain and chondrosis.

P28-340 Reconstruction of the knee extensor mechanism and free vascularized latissimus dorsi musculocutaneous flap for treating traumatic complete extensor loss with extensive soft tissue defect of the anterior aspect of the knee: a case report K. Otoshi1, S. Kikuchi1, H. Numazaki1, H. Kobayashi1, S. Ejiri1, S. Konno1 1 Fukushima Medical University School of Medicine, Orthopaedic Surgery, Fukushima, Japan Objectives: High-velocity open knee injuries sometimes result in several degrees of knee extensor mechanism disruption with severe soft tissue damage around the knee joint. We demonstrated our experience for treating traumatic complete extensor mechanism loss with extensive soft tissue defect of the anterior aspect of the knee. Methods: A 58-year-old male injured his right leg while operating a snowplow. There was extensive soft tissue defect at the anterior aspect of the knee with a defect of patella, patellar tendon, and distal end of quadriceps and its tendon. To regain the function of knee extensor and cover the large soft tissue defect, reconstruction of the knee extensor mechanism using bilateral autologous semitendinosus tendons and the free vascularized latissimus dorsi musculocutaneous flap were performed simultaneously. Results: He could extend his knee without extension lag and return to his previous work 6 months after surgery without any difficulty except for slight degree of tiredness in his right leg when going up and down the stairs. The recovery of knee flexion strength was sufficient whereas knee extension strength was less than 30% compared with the contralateral side even 2 years after surgery. Conclusions: Combined reconstruction of the knee extensor mechanism using an autologous hamstrings tendon with covering of the soft tissue defect using a free vascularized musculocutaneous flap appears to be a treatment option for extensive loss of the knee extensor mechanism with a soft tissue defect around the anterior aspect of the knee. Hamstrings tendons have sufficient strength, and their length is easy to adjust according to the gap, and a free vascularized latissimus dorsi musculocutaneous flap has sufficient size and thickness regardless of the defect size.

P28-383 Tunnel positioning analysis post medial patellofemoral ligament reconstruction. Correlation with clinical outcomes N. Howells1, A. Barnett1, J. Eldridge1 1 University Hospitals Bristol, Trauma and Orthopaedics, Bristol, United Kingdom Objectives: To determine whether tunnel position during MPFL reconstruction surgery has an influence on post-operative patient reported outcomes. Methods: 94 patients who underwent MPFL ligament reconstruction between October 2005 and January 2011 were included in the study. All procedures were performed by or under supervision of the senior author using a standardised semitendinosus autograft technique. Patient follow-up was minimum 6 months (mean 15, range 6–36 months). True lateral radiographs with no rotation were available for all included patients and assessed for femoral and patella tunnel position. All radiographs were standardised for magnification. Position was compared to the anatomical insertion of the MPFL footprint and acceptable limits around this were determined. Functional outcomes were analysed to identify correlations with tunnel position. Up to 5 mm radius outside the footprint of the femoral MPFL insertion as determined

123

S282 by previous studies was taken as acceptable. The centre of the patella insertion has been described as ratio of the distance from the proximal pole over the patella length. Acceptable patella tunnel position was within 0.1 of the anatomically defined ratio. Results: 9 (9.4%) patients were identified as having mal-positioned femoral tunnels. No significant difference in outcomes of patients with acceptable and outlier femoral tunnel positions was seen. However of these 9 patients, 2 (2.1%) outliers were mal-positioned in both sagittal and axial planes. Comparison of outcomes of these biplanar outlier patients did identify significantly worse outcomes as assessed by Oxford Scores (p = 0.048) and Tegner Scores (p = 0.040) with a marked trend towards worse outcomes seen for Kujala, IKDC, WOMAC, Fulkerson and SF12 scores approaching but not reaching significance. Patella tunnel position analysis revealed no outliers in the sagittal plane. In the axial plane 15 (15.9%) patients were deemed malpositioned. Comparing outcomes of patients with overall acceptable and outlier patella tunnel positions, no significant difference was seen. If outliers were broken down into proximal and distal outliers however, a marked trend was identified towards worse outcomes in proximal outliers (6 patients) when compared to acceptable position patients. This trend approached but did not reach statistical significance. (Proximal outlier mean score, acceptable position mean score): IKDC (59.7, 74.1) Kujala (69.7, 80.9) Oxford (32.8, 40.4) WOMAC (84, 92) Fulkerson (63.5, 82.6,) Tegner (4.2, 5.3) SF12 pcs (47.7, 50.2) SF12 mcs (49.5, 55.6). Conclusions: Significantly worse outcomes for patients with bi-planar femoral tunnel mal-position and trends towards worse outcomes for proximal patella tunnel mal-position have been identified following MPFL reconstruction. Attention to accurate anatomical tunnel placement is important in order to optimise outcomes following MPFL reconstruction surgery.

P28-392 Medial patellofemoral ligament reconstruction: prospective outcome assessment of a large single centre series N.R. Howells1, A. Barnett1, A. Ansari1, N. Ahearn1, J. Eldridge1 1 University Hospitals Bristol, Bristol, United Kingdom Objectives: Published literature reporting outcomes following medial patellofemoral ligament reconstruction to date has been confined to small number case series of less than 50 patients and systematic reviews combining these studies. This study aim was to report the prospective analysis of clinical outcome in a large single centre cohort. Methods: 201 consecutive patients underwent 219 procedures between October 2005 and January 2011. Follow up is presented for 193 patients and 211 MPFL (96.3%) procedures. All procedures were performed by the senior author using a standardised technique. A semitendinosus autograft is passed extrasynovially between the isometric attachments to the patella and femur and secured with an endobutton and interference screw. The technique includes arthroscopic assessment through a superolateral portal before and after graft placement ensuring correct graft tension and patella tracking before graft fixation. Results: 193 patients underwent 211 procedures with mean age 26 (range 16–49) and mean follow-up of 16 months (range 6–42 months). 92 were male and 119 female. Indications were atraumatic recurrent patella dislocation in 149 (68%) patients, traumatic recurrent dislocation in 50 (22.8%) patients, instability in 11 (5%), single dislocation in 6 (2.7%) patients and anterior knee pain in 3 (1.4%) patients. 57% had moderate trochlea dysplasia, 35% had mild dysplasia and 7% had normal trochlear morphology. There have been no recurrent dislocations/subluxations. 10 (4.7%)

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 patients have required further surgery. 4 have had removal of a prominent patella endobutton, 2 have had removal of a prominent femoral screw, 1 had arthroscopy to investigate a medial swelling, 1 had arthroscopic partial graft release, 1 had acute tunnel repositioning for mal-position and 1 underwent patellofemoral joint replacement for progressive OA. Available pre-op Kujala Scores were mean 55 (SE 5.21). Post-op Kujala scores for 193 patients improved to mean 82 (31–100) (SE 1.18). This was statistically significant (p \ 0.001). This improvement and significance is mirrored with Oxford (27–41), WOMAC (76–93), Fulkerson (53–83), IKDC (46–75), Tegner (4.1–5.3) and SF12 (38–51) scores (p \ 0.005). 93% of patients were satisfied with their operation. Indication for surgery, degree of dysplasia and evidence of associated cartilage damage at surgery did not result in any significant difference in outcome achieved. History of prior realignment surgery was associated with a significantly worse outcome in comparison to patients for whom MPFL reconstruction was their first realignment procedure (p \ 0.05). Conclusions: This series is to our knowledge larger than any previously reported in the literature for any technique of medial patellofemoral ligament surgery by a substantial margin. This particular technique allows for objective intra-operative evaluation of the required graft tension to optimise patella tracking.

P28-393 Medial patellofemoral ligament reconstruction for patellofemoral instability in patients with hypermobility N. Howells1, J. Eldridge1 1 University Hospitals Bristol, Trauma and Orthopaedics, Bristol, United Kingdom Objectives: Hypermobility is an acknowledged risk factor for patellofemoral instability. This study is a prospective analysis of the influence of hypermobility on clinical outcome following medial patellofemoral ligament reconstruction for patellofemoral instability. Methods: 25 patients who met clinical criteria for hypermobility as determined by Beighton Criteria were assessed and compared to a control group of 50 patients who were 2:1 case matched for age, sex, indication for surgery and degree of trochlea dysplasia. All patients underwent MPFL reconstruction performed by or under supervision of the senior author using a standardised technique. A semitendinosus autograft was passed extrasynovially between the isometric attachments to the patella and femur and secured with an endobutton and interference screw. Patients received pre and post operative clinical evaluation, radiological assessment and outcome scoring systems. Results: The 25 hypermobility patients had a Beighton Score of 6 or more compared to the 50 control patients whose score was \4. Mean Age was 25 in both groups (SE 1.0 and 1.3). Mean follow-up was 15 months in both groups (range 6–34). A significant difference in post-operative patient reported outcomes was determined between the two groups in IKDC (78 vs.55 p \ 0.0001), Kujala (85 vs. 64 p \ 0.0001), Oxford (42.7 vs. 33.3 p \ 0.0001), WOMAC (93 vs. 78 p \ 0.01), Fulkerson (86 vs. 65 p \ 0.001), Tegner (5.5 vs. 4.1 p \ 0.01), SF12 (52 vs. 44 p \ 0.01). Interestingly there was no significant difference between the 2 groups when they were asked subjective questions regarding satisfaction following surgery, rates of perceived improvement, willingness to repeat if the situation arose and likelihood of recommendation of the procedure to friends. Conclusions: Patients with hypermobility who undergo MPFL reconstruction have been shown to achieve significantly worse functional outcomes in comparison to case-matched controls however this does not appear to affect their rates of satisfaction or perceived improvement following surgery.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P28-515 Medial patellofemoral ligament reconstruction: mid term results A. Schiavone Panni1, S. Cerciello1, M. Vasso1 1 Universita del Molise, Science for Health, Campobasso, Italy Objectives: Patellar dislocation is a relatively common occurrence in young and active patients. Its onset is generally post traumatic. Several anatomic factors can predispose to this problem. Aim of our study was to prospectively analyze the functional results of a modified medial patello-femoral reconstruction technique in a population affected by recurrent patellar dislocation. Methods: Forty eight patients (51 knees) with at least 3 episodes of dislocation were included in these study. All these patients have had traumatic onset and have been unsuccessfully treated with a 6 months rehabilitation protocol. All patients practised sports activities. They were evaluated with complete imaging study as well as with Kujala, Larsen, Tegner and Fulkerson scoring questionnaires. Results: Three patients were lost at the last FU. The mean Kujala score improved from 57 preoperatively to 87, the mean Larsen score improved from 12 preoperatively to 17, the mean Fulkerson score improved from 59 to 90. The mean Tegener score improved from 58 preoperatively to 88. 64% of patients returned to the same type of sport at the same level, 16% reduced the level or type of sport for reasons unrelated to the surgery, while 20% reduced the level of sport or changed it for reasons related to surgery. Conclusions: Our results of MPFL reconstructions are encouraging, with minimal risks of redislocations and an overall patient satisfaction rate of over 80%. These early and medium term results are comparable to other MPFL reconstruction techniques reported in the literature. They confirm the efficacy of this procedure, alone or in combination with additional procedures, in treating patellar dislocation.

P28-544 Recalcitrant symptomatic synovial plica: validation of unique clinical exam findings and causal relationship to anterior knee pain J.P. Albright1, P. McCunniff2 1 The University of Iowa, Department of Orthopaedics and Rehab, Sports Medicine Institute, Iowa City, United States, 2University of Iowa, Department of Orthopaedic Surgery, University of Iowa, Iowa City, United States Objectives: Symptomatic synovial plica is identified at arthroscopy for meniscus tear when a thick medial band is discovered. By exclusion, it is felt to be responsible for complaints of knee pain. The Hypotheses: 1. Rather than limited to medial parapatellar location, plica-based pain is often located in multiple areas about the anterior knee. 2. The presence/locations of this sensitive tissue can be detected preoperativly by specialized clinical exam. 3. In patients without chondromalacia where pain cannot managed with conservative measures, surgical excision can be effective. Our objectives are to describe the clinical history/exam findings as well as the operative findings and short-term results of surgical excision of synovial plicas in patients with symptoms that were not responsive to non-operative treatment. Methods: This is a retrospective study of 80 consecutive patients with chronic recalcitrant painful synovial plicas with complete medical record documentation. The diagnosis and anatomic locations of the painful plicas were established from unique clinical exam features

S283 prior to surgery. The pre-op diagnosis was to be confirmed or dismissed at surgery. The short-term success in producing significant relief of pain in patients without concomitant chondromalacia patella was to be established. Finally, the causal relationship of the plica to the pain syndrome was to be confirmed by post excision relief of pain. Results: True positives: In 70 knees (87.5%) were correctly diagnosed by preop clinical exam, as confirmed at surgery. False negatives: In 9 knees (11.25%) a plica was not suspected until surgery but removal led to symptomatic relief. False positives: In 1 knee (1.25%) the clinical exam indicated plica but none was found. The Sensitivity of the clinical exam features that led to the diagnosis of a synovial plica was 70/79 (88.6%). The positive predictive value that the pre-operative diagnosis of symptomatic synovial plica was actually verified by the findings at surgery was 70/71 (98.6%). The chances that there had been a causal relationship between the painful synovial plica and the pain syndrome that had led the patient to seek treatment was realized in all 79 patients (100%) whether they had been correctly diagnosed prior to surgery. The time it took to reach these results was 7 weeks (2–36) after plicectomy in 75 patients who recovered without complication. In the four patients (5%) who required manipulation, it took 9 weeks to realize significant benefit from the procedure. Conclusions: Painful synovial plicas should be considered in the differential diagnosis of acute and acute-on-chronic anterior knee pain. Contrary to popular belief, symptomatic synovial plicas are often found in multiple areas around the patella. The existence and location of symptomatic plicas can be reliably detected pre-operatively by recognizing unique physical exam features. Surgical removal of otherwise recalcitrant plicas predictably yields excellent pain relief.

P28-569 MPFL reconstruction with less implants in recurrent patella dislocations-first results in 15 consecutive patients C. Gatzka1, G. Pap1 1 Parkkrankenhaus Leipzig, Orthopaedic Surgery, Leipzig, Germany Objectives: MPFL reconstruction with gracilis tendon is widely accepted as one successful treatment option in recurrent patella dislocation. To date most surgeons use screws, anchors or pins to fix the tendon at the patella side. These fixation methods bear the risk of implant loosening and foreign body reaction. Additionally they are cost intensive. We describe our technique of implant free gracilis tendon fixation on the patella side and present first results in 15 consecutive patients. Methods: All patients operated in our institution between 03/09 and 03/11 were included in this study. In all patients we performed gracilis tendon fixation on the patella side without implants. To fix the tendon we drilled two slightly converging parallel holes from the medial proximal edge to the central part of the patella. Both holes had ‘‘blind’’ ends. We then drilled another two anterior holes in a 90 angle to the first holes in direction to the ‘‘blind’’ ends of the first two holes. The two anterior holes were then connected via a grove made with a burr. The gracilis tendon was then pulled through the holes with the help of the passing wire. The two ends of the gracilis tendon where then passed through the second and third layer of the knee capsule and pulled into a previously drilled blind hole at the origin of the femoral MPFL insertion. After arthroscopically controlled tensioning of the gracilis tendon both ends were fixed with a biodegradable screw on the femoral side. Additional operations were: Distalization of the tuberositas tibia and trochleaplasty in one patient,

123

S284 femoral closed wedge osteotomy with rotational correction in one patient and ACT on the patellar side in one patient. Operation time varied between 60 and 145 min. All patients had clinical follow-up examinations at 6, 12 weeks and 6, 12 months. Outcome was evaluated with Kujala and modified Lysholm score. Dislocation rate, knee function and complications were further documented. Results: In all cases implant free gracilis fixation on the patella side was possible with the above described technique. No patient was lost to follow up. Average follow up time was 11 months (6–18 months). There was no patella dislocation postoperatively. In one female patient we documented an in part ‘‘cut through’’ of the gracilis tendon in the patellar bone after traumatic fall on the knee. This was fixed via intraosseos ligament fixation. No dislocation after this reoperation up to now. We saw no infection or impairment of knee function after surgery. Lysholm Score improved from 56 to 86 Kujala score improved from 52 to 82. Conclusions: MPFL reconstruction is a safe and successful technique for the treatment of recurrent patella dislocation. Implant free gracilis tendon fixation on the patella side was found to be efficient. We were able to demonstrate the potential advantages like missing implant associated risks and lower costs of implant free fixation in comparison to implant based fixation techniques.

P28-652 Anatomic femoral tunnel position for medial patellofemoral ligament reconstruction J. Stephen1, D. Kader2, P. Lumpaopong1, D. Deehan3, A.A. Amis1 1 Imperial College London, Mechanical Engineering Department, London, United Kingdom, 2Queen Elizabeth Hospital, Orthopaedics, Tyne and Wear, United Kingdom, 3Freeman Hospital Newcastle, Orthopaedics, Newcastle, United Kingdom Objectives: To recommend a reproducible femoral attachment site for undertaking anatomic MPFL reconstruction. Methods: 8 fresh frozen cadaveric knees were defrosted and dissected of skin and subcutaneous fat. A 1 mm diameter pin was inserted midway between the medial femoral epicondyle and the adductor tubercle, at the centre of the origin of the MPFL. Radiographic assessment of the femoral attachment points was achieved with standardised lateral view X-rays superimposing the posterior femoral condyles. The knees were flexed to 30 over a wooden block and securely packaged in polyethylene boxes, with a cross drawn on the outside of the box in line with the medial epicondyle of the femur to ensure standardisation. Measurements of the medial condylar diameter, femoral cortex diameter and the distance of the MPFL insertion from the anterior, posterior and distal femoral edges (Pic 1) were taken using ImageJ (Maryland, USA). Pearsons correlations were undertaken to investigate any relationships between the measurements. Data was also analysed to calculate the anterior/posterior and proximal/distal position of the MPFL insertion point in relation to the medial femoral condyle border. This was done by calculating the MPFL distance in each direction as a percentage of the medial femoral condyle AP diameter: (distance/diameter) 9 100. This enabled individual knee size variation to be accounted for. Results: The femoral condyle diameter was significantly related to the MPFL insertion anteriorly (r = 0.794: P = 0.019), posteriorly (r = 0.935: P = 0.001) and distally (r = 0.987: P \ 0.01). Additional relationships were identified between the MPFL insertion distally and anteriorly (r = 0.867: P = 0.005) and distally and posteriorly (r = 0.874: P = 0.004) and between the AP diameters of the medial condyle and the femoral cortex (r = 0.769: P = 0.026).

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1

The MPFL insertion point was 41% ± 2 (37–43) (mean ± SD; range) from the posterior border, 59% ± 2 (57–63) from the anterior border and 51% ± 1 (49–52) from the distal border of the medial condyle. This radiographic point was 0.0 mm ± 0.7 mm in the proximal/distal and 0.05 mm ± 1.7 mm A/P to the anatomic MPFL insertion determined by the current study Conclusions: The key findings of this work allow the surgeon to locate through the use of reliable anatomical landmarks and radiographically the most anatomic tunnel placement for MPFL reconstruction. It is hoped that this ‘40–50–60% rule’ will be incorporated within the surgeons armamentarium. Clinical studies are now required to confirm the clinical efficacy of such a rationale.

P28-698 Autologous chondrocyte implantation to isolated patella defects A. von Keudell1, T. Bryant2, T. Minas1 1 Brigham and Women’s Hospital/Harvard Medical School, Boston, United States, 2Brigham and Women’s Hospital, Boston, United States Objectives: Autologous Chondrocyte Implantation has been proven to be a lasting treatment option for patients with chondral defects. Up to now, there has been limited evidence of the efficacy of ACI to single defects in the patella. Methods: Between May 1995 and May 2009, 510 patients were treated with ACI at our institution. Among those were 30 symptomatic focal chondral patella (4 Type II, 7 Type III, 7 Type IV, 12 Type IVa) defects that we treated with autologous chondrocyte implantation and 22/30 underwent additional Tibial Tubercle Osteotomy (20/ 22 Fulkerson and 2/22 McKay TTO). 24/30 patients experienced either sports-, motor vehicle- or fall- related trauma. The defect size averaged at 4.7 ± 2.1cm2. M/F ratio was 12/18, Ø BMI = 27 and Ø age at time of surgery was 32 ± 10 years. Prospective evaluation of patients with a series of validated clinical assays and subjective clinical rating (Patient Satisfaction, modified Cinncinati, Knee Society Score, WOMAC, SF36) was performed at a minimum of 24 months. Magnetic resonance imaging using the MOCART score was performed at a mean of 31 months. Results: At the time of last follow-up, knee function was rated good to excellent in twenty-five patients (83%), fair in four patients (13%) and poor in one patient (3%). Three patients failed after a mean of 75 months. Failures were defined as [25% delamination, revision cartilage repair surgery or prothesis implantation. Significant increases in all clinical and health utility outcome scores were seen. A body

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 mass index above[26 was associated with lower postoperative scores (r = 0.6). Magnetic resonance imaging at 31 ± 25 months postoperative in twenty-four knees demonstrated good repair- tissue fill in the defect in seventeen patients (71%), moderate fill in six (25%), and poor fill in one patient (4%). The fill grade correlated with the clinical scores (r = 0.41). All knees with good fill demonstrated improved knee function and pain, whereas poor fill grade was associated with limited improvement and decreasing functional scores after 24 months. Conclusions: Autologous chondrocyte implatantion to isolated patella defects results in significant functional improvement at a minimum of 24 months. The best outcome was seen with good fill grade. High body-mass index is associated with lower scores. More research is needed to confirm the data.

P28-699 Bicompartmental knee arthritis: a prospective matched short term study using UKR + PFA N. Confalonieri1, A. Manzotti1 1 CTO Hospital, 1st Orthop Dept, Milan, Italy Objectives: The authors performed a short term prospective study of unicompartmental knee replacement (UKR) associated to patellafemoral arthroplasty (PFA) for the treatment of bicompartmental symptomatic knee arthritis. At the latest follow-up all the implants were matched to a similar computer assisted Total Knee Replacements (CAS-TKR) group implanted for the same indications. Hypothesis of the study was that this bicompartimental implants could achieve comparable outcomes to TKR with a more conservative surgery and a higher joint function. Methods: 19 cases of anteriomedial (12) or anterolateral (7) arthritis in 19 stable knees were prospectively involved in the study. All the knees underwent to a selective reconstruction using simultaneously both UKR and PFR using the same surgical technique. All bicompartmental implants were performed by the same surgeon. Surgical time, hospital staying and all intra and post operative complications were registered. At a minimum follow-up of 20 months, every single case was marched to a similar case where had been implanted a computer assisted cruciate retaining TKR. Criteria of matching were: sex, age, pre-operative range of motion and arthritis grade. In both the groups all the cases were assessed clinically using WOMAC, KKS and GIUM scores. All the knees were radiologically investigated using the same radiological protocol. Results: Intra operatively we did not registered any complication. No revision in both groups. The mean surgical time was 86 min (range: 78–121) in UKR + PFR group and 81 min (range: 71–112) in CASTKR group. There were no statistical significant differences in the hospital stay. No statistically significant difference was seen for the Knee Society, Functional and GIUM scores between the 2 groups. Statistically significant better WOMAC Function/Stiffness indexes were registered for the UKR + PFR group. CAS-TKR implants achieved a statistically better aligned mechanical axes. Conclusions: The results of this prospective short term prospective study suggest that UKR + PFR implant is a viable option for bicompartmental anteromedial arthritis at least as well as TKR but maintaining an higher level of function.

P28-754 Patellofemoral realignment with medial patellofemoral ligament reconstruction as a novel procedure for correction of patellar maltracking after TKA in selected cases S. van Gennip1, J. Schimmel2, G. van Hellemondt1, A. Wymenga3 1 Sint Maartenskliniek, Department of Orthopaedic Surgery, Nijmegen, The Netherlands, 2Sint Maartenskliniek, RD&E,

S285 Nijmegen, The Netherlands, 3Sint Maartenskliniek, Orthopedic Surgery, Knee Reconstruction Unit, Nijmegen, The Netherlands Objectives: Maltracking of the patella after total knee arthroplasty (TKA) remains a common problem. The medial patellofemoral ligament (MPFL) has shown to be important for patellar stabilization and reconstructions of the MPFL have already shown excellent functional outcomes in case of habitual patellar instability. Nevertheless, there is only limited literature on using a MPFL reconstruction for correction of patellar maltracking after TKA. In this study we evaluated a consecutive case series, retrospectively. Methods: Between January 2007 and January 2011 nine patients (nine knees) with anterior knee pain and symptomatic (sub)luxations of the patella after primary or revision TKA were treated by reconstruction of the MPFL in combination with a lateral release. In two cases, an additional tuberosity transfer was performed, due to insufficient correction peroperatively. Pre-operative work-up comprehended a CT-scan to exclude component malrotation and disorders in limb alignment. Patellar displacement and lateral patellar tilt were measured on axial radiographs as described by Heesterbeek et al. Clinical outcome was evaluated using the visual analogue scale (VAS) satisfaction, VAS pain, luxation-rate, Bartlett knee score, Kujala knee score and KOOS. Results: Median lateral patellar tilt was 40˚ pre-operative and changed to a median 12˚ post-operative. Median patellar displacement improved from 35 to 7 mm post-operative. Median VAS satisfaction was 7.8 and only one patient reported patellar (sub)luxations afterwards. Functional outcomes displayed a diverse picture. Conclusions: Patellar maltracking after TKA can be effectively treated by a MPFL reconstruction in combination with a lateral release. Only in limited cases an additional tuberosity transfer is needed. Exclusion of component malrotation and disorders in limb alignment are essential.

P28-798 Knee function in patients with two or more episodes of patella dislocations J.R. Mikaelsen1, I. Skra˚mm1, B.V. Evensen1, S. Pedersen1, J. Brand2, A. A˚røen1 1 Akershus University Hospital, Lørenskog, Norway, 2Heartland Orthopedic Specialists, Alexandria, United States Objectives: Recurrent patella dislocation, causes rupture of the medial patellofemoral ligament (MPFL), and often results in avoidance of sport activities among young active people. Patients with recurrent instability are candidates for MPFL reconstruction due to failure of nonoperative approaches. Little is known of the impact of recurrent instability on knee function and on activity level of the patients who are candidates for MPFL reconstruction. Methods: Patients aged 12–30 years, with two or more patella dislocations, were consecutively included into a randomized prospective study. Exclusion criteria include bilateral dislocations, non compliance to rehabilitation, established patellofemoralarthrosis and tibial tuberosity trochlear groove (TT-TG)—distance of more than 20 mm. All patients underwent clinical examinations, functional tests, arthroscopy, Computerized Tomography (CT) and were assessed with the Kujala score, Lysholm score, Tegner activity score, visual analogue scale (VAS) and Activity Scale Knee injury and Osteoarthritis Outcome Score (KOOS), 25 patients were evaluated in the study and 15 of those fulfilled the inclusion criteria that are reported in this investigation. Results: There were 4 males and 11 females, mean age 16.4 years. Injury mechanism is sport 53%, dance 27 and 20% were of varied injury. Clinical examination showed that the patella was hyper mobile

123

S286

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

and the apprehension test was positive in all cases. Functional tests showed no significant side to side difference between injured and healthy knees. At arthroscopy 10 of 15 patients had articular cartilage changes. CT examination revealed a TT-TG value of 15.8 (11–19). Kujala score 66 (36–98), Lysholm score 59.4 (39–95), Tegner score 5.3 (1–10), VAS 4.3 (0–9), Activity Scale Level 2.5–69 (Level 1–4, 0–100), KOOS Pain 70 (39–100), KOOS Symptoms 71 (54–93), KOOS ADL 84 (59–100), KOOS Sport/Rec 51 (5–80), KOOS QOL 43 (12–75). Conclusions: The results show a low level of functioning in this patient group in relation to sport activity and knee function. These patients are significantly impacted and need an effective treatment. This is the initial report of a prospective randomized study that will compare reconstruction of the MPFL to 6 months of a standardized training program.

P28-1106 The isometric point for medial patellofemoral ligament reconstructions: an in vivo analysis using 3-dimentional computed tomography S.Y. Song1, Y. Yoo2, Y.-J. Seo3, Hallym sports medicine research group 1 Hallym University, Seoul, Republic of Korea, 2Hallym University, Chuncheon, Republic of Korea, 3Hallym University, Orthopaedic surgery, Hallym sports medicine research group, Seoul, Republic of Korea Objectives: The ideal tunnel position for a successful MPFL reconstruction remains controversial. With this reason, numerous MPFL reconstruction procedures with various graft attachment sites are being used today. The maintenance of constant length during the range of knee motion has been advocated as one of the important goal in MPFL reconstruction. The purpose of this study was to determine the most isometric points for femoral and patellar attachment sites at which the length changes of the MPFL would be minimized during knee flexion. Methods: The subject compromised ten male volunteers with no history of knee pathology. The mean age was 32.1 ± 3.8 years. The right knee of each subject was scanned with a CT scanner at five different knee flexion angles (0, 30, 60, 90 and 120). Software was used to created, manipulate, and analyze the 3D model. Based on recent anatomic studies and current surgical techniques, the four possible origin points were marked on the femoral side: the adductor tubercle (A), the center between the medial epicondyle and the adductor tubercle (B), the medial epicondyle (C), and 10 mm distal to the adductor tubercle (D). On the patellar side, the three possible insertion points were marked: 30% (1), and 45% (2) from the proximal pole of the patella (Fig. 1). We thus created 8 virtual bundles on the 3D knee models, and digitally measured the length of the different bundles was at 5 different knee flexion angles. This provided 40 measurements per knee. We determined the most isometric point of the reconstructed MPFL at which the length change was the smallest. Results: Two virtual bundles included in point C (C1 and C2) were less isometric than the other 6 bundles. No significant differences were found among the bundles included in point B, in point D, and in point A (Fig. 2). With regard to length change pattern (Fig. 3), 2 bundles inserting into the adductor tubercle (A1, A2) showed a increasing tendency while the knee was flexed over 60, which would increase force and the pressure applied to cartilage of the patellofemoral joint. In contrast, 2 bundles inserting into the medial femoral epicondyle (C1, C2) lead to excessively decrease the distance during the knee flexion, which could reduce the resistance to lateral patellar subluxation. Therefore, the adductor tubercle and the medial femoral epicondyle are not ideal femoral site. In remaining 4 bundles (B1, B2,

123

Fig. 1

Fig. 2

Fig. 3 D1, D2), similar to previous studies, the distances (lengths) increased slightly as the knee flexed from 0 to 30, and decreased slightly as the flexion angle increased over 30.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: To obtain more stability throughout the range of knee flexion, we recommend that a femoral drill hole be located at the center between the medial epicondyle and the adductor tubercle, or just distal to the adductor tubercle. The best angle for graft fixation would be near 30 of knee flexion.

P28-927 MPFL: reconstruction with gracilis tendon for patellar instability—3 years results A. Fechner1, G. Godolias1, O. Meyer1 1 St. Anna-Hospital, Center for Orthopaedics and Traumatology, Herne, Germany Objectives: Over the last years it was proven that the conservative therapy of the patellofemoral instability is limited by a narrow therapeutic window. The reconstruction of the medial patellofemoral ligament arose to a steady therapeutic concept. Various operation techniques are based on anatomic studies. Goal of this study was an evaluation of the middle term clinical outcome and radiological results while considering the range of indications of this procedure. Methods: Between 01/2006 and 03/2008 an MPFL using the gracilis tendon was conducted on 42 patients (45 knee joints). Study exclusion criteria account a TTTG [ 20 mm, a pronounced dysplastic trochlea (Typ C and D of the Dejour classification), or a femoral Axis deviation, which needs to be treated. The age of the patients averaged to 25.6 years (16–39 years old). Patients that were included in this prospective study where examined preoperatively and 3 months, 1 and 3 years postoperatively via clinical examination and vian MRI. The clinical evaluations where conducted with help of the Kujala Score and the Tegner activity score. Results: When the postoperative examinations where conducted none of the examined patients experienced another patellar luxation. 37 patients achieved their planed level of activity. The average Kujala Score improved from 53.2 to 88.8 in the 3 year follow up examination. The radiological follow up showed that the patellar tilt sunk significantly to 7.0. The congruence angle improved significantly to 6.1 postoperatively. 3 patients reported pain peripatellar, no operative revisions where necessary. Conclusions: The reconstruction of the MPFL is a safe, reproducible and minimal invasive method when treating patellar instability. When considering the pathology of the instability it widens the spectrum of operative patellar stabilizing treatment, though it isn’t the solution of all types of patellar instability.

P28-941 Incidence of cartilage lesions and early osteoarthritis in patients with patellar dislocation S. Scheffler1, B. Vollnberg2, T. Koehlitz2, A. Hoburg3, T.M. Jung4, G. Diederichs2 1 COPV, Chirurgisch Orthopa¨discher PraxisVerbund, Berlin, Germany, 2Charite´, University Medicine Berlin, Department of Radiology, Berlin, Germany, 3Charite´, Universita¨tsmedizin Berlin, Klinik fu¨r Orthopa¨die und Unfallchirurgie, Berlin, Germany, 4 Virchow-Klinikum/Charite´, Unfallchirurgie & Orthopa¨die, Sektion Sporttraumatologie & Arthroskopie, Berlin, Germany Objectives: To assess cartilage lesions and osteoarthritis (OA) of the patellofemoral joint in patients following first and recurrent lateral dislocation of the patella using magnetic resonance imaging (MRI).

S287 Methods: MR images of 129 knees (mean age 26 years, range 11–56) grouped in acute (A), recurrent (B), and chronic (C) dislocators were analyzed. Incidence, severity, location, and morphology of cartilage lesions of the patella and trochlea were quantified. Grades and location of patellofemoral OA were assessed using a modified WORMS system. Results: In groups A, B, and C the incidence of cartilage lesions was 71, 82, and 97%, respectively. Most lesions were located in the central dome of the patella in groups A and B (central 69 and 78%; medial 56 and 47%; lateral 31 and 42%), whereas group C showed cartilage damage in all regions (73, 61, and 67%). Mild OA was present in 14% of group A and 64% of group B. Group C showed mild OA in 62% and moderate OA in 18%. The size of cartilage defects and presence of OA were significantly correlated with the number of previous dislocations (R = 0.41 and R = 0.59; both p \ 0.001). Conclusions: Cartilage lesions are very common in patients after acute and recurrent patellar dislocation, and their severity varies with the number of previous dislocations, suggesting a causal relationship. Signs of early OA were also present, especially in recurrent and chronic dislocators. Considering the young age of the patients this finding might have implications for future treatment approaches.

P28-967 The width: thickness ratio of the patella and its relationship to pain and functional outcome post patellofemoral joint arthroplasty L. Eddie1, M. Hassaballa2, J. Robinson2, A. Porteous3, J. Murray4, The Bristol Knee Group 1 Bristol, Bristol, United Kingdom, 2The Bristol Knee Group, Orthopaedic Surgery, Avon Orthopaedic centre, Bristol, United Kingdom, 3Avon Orthropaedic Centre, Southmead Hospital, Bristol, United Kingdom, 4The Bristol Knee Group, Ortopaedics, Bristol, United Kingdom Objectives: The aim of this study is to investigate if the anatomical width: thickness ratio of the patella is reproducible, and whether reproducing this ratio is a correlated to clinical outcome. Methods: To establish patella width: thickness ratio, 100 digital skyline radiographs in 89 patients who underwent uni-compartmental knee replacements were selected. The Centricity Picture Archiving and Communication Systems (PACS) on screen tool was used to measure patella width and thickness. To establish the effect of the ratio upon clinical outcome following patellofemoral joint arthroplasty, 81 post-operative digital skyline patellofemoral radiographs were used to measure patella-bone prosthesis construct width and thickness. Oxford knee and WOMAC post-operative scores analyzed. Results: Patellofemoral joint measurements in the uni-compartmental knee replacement (UKR) cohort showed that patella thickness correlated (r = 0.84, p \ 0.001) with its width. The mean width to thickness ratio was 2.07. According to the confidence interval of the non-pathological patella ratios from the uni-comparmental knee replacement cohort, ratios from the resurfaced patellae cohort were categorised into ‘under-stuffed’ (ratio [ 1:2.1), normal (ratio 1:2.04–1:2.10) and ‘over-stuffed’ (ratio \ 1:2.04). Statistical analyses assessed correlation between the ratios with WOMAC and OKS. Results showed that the ‘normal’ subset had better outcomes in all scores however this was not statistically significant on ANOVA testing (p [ 0.05). Conclusions: The results of this study imply that if patellofemoral joints are resurfaced using the thickness to width ratio range 2.04–2.10 better clinical outcomes may result.

123

S288 P28-981 Medial patellofemoral ligament avulsion injury from the patella: classification, surgical technique and clinical outcome P. Sillanpa¨a¨1, H. Ma¨enpa¨a¨2, H. Pihlajama¨ki3 1 Tampere University Hospital, Orthopeadic Surgery and Trauma, Tampere, Finland, 2Tampere University Hospital, Orthopaedics and Traumatology, Tampere, Finland, 3Helsinkin University Hospital, Orthopaedics and Traumatology, Helsinki, Finland Objectives: Primary traumatic patellar dislocation results in medial patellofemoral ligament (MPFL) injury and may lead to chronic patellar instability. However, characteristics and clinical outcome of patellar attachment MPFL avulsion is unknown. Methods: Magnetic resonance imaging (MRI) was used to retrospectively assess MPFL avulsion injuries, from the medial margin of the patella, in patients with primary traumatic patellar dislocation. Fifty-six patients with patellar MPFL avulsion were found. Patellar MPFL avulsions included 13 patients, who underwent surgical fixation of the avulsed patellar medial margin osteochondral fracture and the remaining patellar MPFL avulsions were treated without surgical stabilization. Median age was 23 years (range, 15–31). Fourty-four patients were available for follow-up of median 4 years (range 1–10). The follow-up evaluation included recurrent patellar instability, subjective symptoms and functional limitations were evaluated. Results: Three types of patellar MPFL injuries were found; type P0 with ligamentous disruption at patellar attachment, type P1 with bony avulsion fracture from the medial margin of the patella and type P2 with bony avulsion involving articular cartilage from medial facet of the patella. Of the patellar MPFL avulsions, which underwent fragment fixation, two patients [2/13 (15%)] reported an unstable patella at follow-up. Correspondingly, patellar MPFL avulsions, which were treated without fragment fixation, had recurrent patellar instability in 55% of cases (17/31) (P = .12). The median Kujala score was 90 for patellar avulsion with fragment fixation and 86 for patellar avulsion with fragment fixation (P = .68). Conclusions: Patellar MPFL avulsion with an osteochondral fracture of the medial patellar margin may benefit from initial surgical fixation. Surgery should most likely to be aimed at cases with articular surface involvement. MRI is recommended to assess the injury pattern. Prospective studies are required to confirm the role of surgical treatment.

P28-1010 Biologic arthroplasty in patello-femoral cartilage lesions A. Gobbi1, A. Kumar1, G. Karnatzikos1 1 Orthopaedic Arthroscopic Surgery International, Department Sport and Medicine, Milano, Italy Objectives: The purpose of our study was to determine the effectiveness of cartilage repair in patello- femoral (PF) lesions utilizing a one step surgery with Bone Marrow Aspirate Concentrated (B.M.A.C.) and collagen I/III matrix. Good results have been published with the two-step Autologous Chondrocyte Implantation but very few studies have analyzed single step procedure with bone marrow derived mesenchymal cells for PF cartilage lesions. Methods: We prospectively followed up 25 patients (16 males and 9 females) operated for large cartilage lesions of the knee (grade 4 of ICRS classification), for a minimum follow-up of 3 years. All patients underwent a mini arthrotomy and concomitant transplantation with BMAC covered with the collagen matrix (Chondro-GideGeistlich Wolhusen, CH). Bone marrow was harvested from ipsilateral iliac crest and subjected to concentration and activation with Batroxobin solution (Plateltexact-Plateltex S.R.O. Bratislava, SK). The mean age of the patients was 46 years. Ten patients had multiple chondral lesions; the location of the lesions was 12 patella, 11 trochlea, 7 medial tibial plateau, 10 medial, and 3 lateral femoral condyle.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Lesions of the PF joint were seen in 19 of the patients (9 in the patella, 8 in the trochlea and 3 PF kissing lesions; 1 patient had 2 patellar lesions). Average lesions size was 8.3 cm2. Co-existing pathologies were treated before or during the same surgery in 19 of our patients. All patients followed the same specific rehabilitation program for a minimum of 6 months. X-rays, MRI and VAS, IKDC, KOOS, Lysholm, Marx and Tegner scores were collected at pre-op and at 6, 12, 24 months and at final follow-up; nonparametric analysis was performed with the Wilcoxon rank test to compare these variations. Six patients gave their consent for second look arthroscopy and five of them for a concomitant biopsy. Results: Patients showed significant improvement in all scores at final follow-up (p \ .005). Mean preoperative values were: VAS 5.2, IKDC subjective 43.6, KOOS Scores P = 66.2/S = 68.2/ ADL = 70.0/SP = 41.6/QOL = 37.2, Lysholm 60.4, Marx 4.2 and Tegner 2.0. At final follow-up mean scores were: VAS 0.6, IKDC subjective 81.2, KOOS P = 93.8/S = 91.4/ADL = 94.8/SP = 72.1/ QOL = 78.6, Lysholm 91.2, Marx 10.2, and Tegner 5.2. MRI showed good coverage of the lesion and tissue quality in all patients in accordance with clinical results. Good histological findings were reported for all the specimens analysed who presented hyaline-like features. No adverse reactions or post-operative complications were noted. Conclusions: This study showed that the use of autologous BMAC and collagen I/III matrix in a one-step procedure could represent an improvement on the currently available techniques for cartilage repair and could be a viable technique in the treatment of grade IV PF chondral lesions justifying it as a Biologic Arthroplasty.

P28-1016 Long-term results of patello-femoral realignment procedures M. Drobnic1, G. Vivod1, K. Strazar1, O. Zupanc1, P. Verdonk2 1 University Medical Centre Ljubljana, Department of Orthopaedic Surgery, Ljubljana, Slovenia, 2Gent University Hospital, Orthopaedic Surgery, Gent, Belgium Objectives: Several realignment techniques are available for the treatment objective patello-femoral (PF) instability. In general, they can be divided into proximal procedures, distal procedures and trochleoplasties, or a combination of the above. Although these procedures result in an instant increased stability of the PF joint, the long term clinical and radiological outcome of them remain to be determined. Methods: The operation registries at our university orthopedic department were reviewed from the years 1963–1994 (total 547 PF procedures). 39 patients (78 knees) were available for the long term re-evaluation. They were operated on by various orthopedic surgeons using different PF stabilization methods. We sub-grouped their knees into: proximal procedures—22 (Insall, Madigan, Green, or simple medial reefing), distal procedures—32 (Elmslie-Trillat) with or without Albee trocheoplasty, and non-operated knees—24. Subjective (KOOS, Kujala), clinical (PF tracking, apprehension, and lateral motion), and radiographic (patella height, Kellgren-Lawrence OA scale) evaluation was performed. Results: Patients with proximal procedures were younger (38.5 vs. 44.0 years), and operated on at younger age (16.0 vs. 25.0 years). PF redislocation rate was similar in both operated subgroups (proximal 36% vs. distal 25%). Proximal subgroup revealed less centrally positioned patellas (64 vs. 94%), more frequent PF apprehension test (82 vs. 47%), and increased PF lateral motion (82 vs. 47%). KOOS and Kujala scores were similar between proximal and distal subgroup. Both subgroups reached lower values that the non-operated knees in all categories. Patellas were positioned lower after distal procedures (Caton-Descamp index: distal 0.8 vs. proximal 1.0). Distal procedures induced higher rate of tibio-femoral OA (63 vs. 37%) and PF OA (76

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 vs. 64%) than the proximal ones. There were no significant differences between the distal procedures with or without a trochleoplasty in all categories. Conclusions: The presented retrospective study gives one of rare longterm insights into the PF instability surgery. High PF re-dislocation rates together with a very high incidence of PF and tibio-femoral OA indicate that PF realignment strategies used at the time had failed to reach their expectations. As the proximal procedures allowed for more residual PF mobility, they seem to have induced less stress on cartilaginous surfaces and therefore protected from early OA. On the other hand, more constrained PF joints with low-riding patellas after distal procedures made OA outcome worse. Hopefully, better pre-operative planning and modern anatomical PF reconstruction techniques will improve the currently inferior long-term outcome.

P28-1056 J-sign, TT-TG, and dynamic MRI in establishing proper tibial tubercle transfer In patients with patellar instability J.P. Albright1, S. McDermott2 1 The University of Iowa, Department of Orthopaedics and Rehab, Sports Medicine Institute, Iowa City, United States, 2University of Iowa Carver College of Medicine, Iowa City, United States Objectives: The most popular surgery for patients who suffer from patellar dislocations involves medialization of the tibial tubercle. A major problem is how far one needs to transfer the tubercle in order to get maximum congruency of the patella sitting within the trochlear edges at 0 and 30 flexion. Currently, most surgeons take into consideration the preoperatively measured TT-TG distance. The distance of medialization recommended by Dejour et al. as well as many others for a TT-TG of 2 cm is 1 cm maximum (2:1 ratio). At our institution, the target is achieving complete patellofemoral congruency during active leg extension both with the patient awake and under anesthesia (femoral nerve stimulation) by direct observation during surgery. Our hypotheses were that (1) the 2:1 ratio of TT-TG to medialization was inadequate at eliminating the J-sign by creating congruency (2) assessment of the dynamic function of the extensor mechanism preoperatively and intraoperatively is superior to TT-TG calculation alone. Methods: We examined 20 knees in 18 patients who had a history of recurrent lateral dislocations. Preoperatively they were examined clinically for congruency of the patella in relation to the trochlea of the femur at 30 flexion and complete hyperextension. This was quantified by obtaining an MRI with the quadriceps contracted at 30 flexion and at complete extension. TT-TG measurements were obtained in a manner as described by Dejour. At surgery, the tibial tubercle was medialized according to (1) the 2:1 TT-TG ratio explained by Dejour (2) a 1:1 ratio of the TT-TG calculation (3) active quadriceps contraction through stimulation of the femoral nerve and direct observation of the relationship of the patella to the trochlear edges at 30 flexion and complete extension. Results: Thirteen knees preoperatively demonstrated a true J-sign with the patella subluxed laterally at full extension but within the confines of the groove at 30 flexion with the quadriceps contracted before and during surgery. Of this group the TT-TG was accurate if the distance medialized was 1:1 with the measured TT-TG rather than the 2:1 ratio described by Dejour. In using 2:1 ratio as the transfer distance, 100% of the time it did not correct or eliminate the J-sign. Using a 1:1 ratio, it was successful 92% of the time in accepting a 5 mm error. In 7 out of the 20 knees, the patella was subluxed at both 30 flexion and in complete extension. In this group the TT-TG underestimated the transfer required for congruency at 0 and 30 flexion on average 5 mm with a range from 0 mm to 12 mm even when using the 1:1 ratio. Conclusions: (1) The 2:1 ratio of TT-TG to medialization is always inadequate at eliminating the J-sign to create patellofemoral

S289 congruency at both 30 flexion and extension with the quadriceps contracted. (2) In the population of patients where the patella remains subluxed at 30 flexion and full extension, even the 1:1 ratio remains inadequate to reliably produce congruency when the muscle is stimulated.

P28-1069 Clinical outcome after MPFL reconstruction in primary patellar dislocation versus salvage surgery P. Verdonk1, A. Mulliez2, E. Thibaut2, D. Verbruggen1, D. Lambrecht1, J. Victor3 1 University Hospital Gent, Orthopaedic Surgery, Gent, Belgium, 2 University Hospital Gent, Department Orthopaedic Surgery and Traumatology, Gent, Belgium, 3AZ St Lucas Brugge, Department of Orthopedics, Brugge, Belgium Objectives: There are several surgical options for recurrent lateral dislocations of the patella. As the reconstruction of the medial patellofemoral ligament reconstruction (MPFL) has been proven to restore stability, it has become more accepted by surgeons. No study to date has examined the difference in clinical outcome between patients with a primary MPFL reconstruction (group 1) versus MPFL reconstruction in previously failed patellar stabilizing operations (transposition of tibial tubercle, trochleoplasty, Insall plasty) (group 2). Methods: Study type: Prospective cohort study with minimal followup of 2 years (range 2–4 years, mean 3.2 years). 65 knees of 60 patients have been treated between April 2007 and December 2009. Forty-five knees of 41 patients received primary MPFL reconstruction of which 12 were in combination with a transposition of the tibial tubercle (group 1). Twenty knees of 19 patients already had a patellar stabilizing operation in the history of which 2 Insall plasty, 6 trochleoplasty and 12 TTT (group 2). The clinical follow up was evaluated using KOOS and KUJULA scores preoperatively and at 1, 2 year yearly thereafter. Results: For group 1, all KOOS and KUJALA score increased significantly after 2 year, while only significant improvement occurs on KOOS subscores sports and quality of life in group 2. The net gain for both group 1 and 2 is similar for both groups. However, the preoperative KOOS and KUJALA score was significantly worse in group 2 compared to group 1, with a difference of about 20 points Thus, overall the outcome after 2 years is significantly better for group 1 versus group 2. Conclusions: MPFL reconstruction is a viable treatment option for episodic patellar dislocation both in a primary setting as well as in a secondary setting for failed surgery cases. Both group improve significantly with a similar amount. The clinical outcome is nevertheless significantly higher in primary cases.

P28-1131 Improved TTTG accuracy using 3D landmarks to compensate for misalignment S.J. Harris1, J. Cobb2 1 Imperial College, Surgery and Cancer, London, United Kingdom, 2 Imperial College London, Charing Cross Hospital, Orthopaedic Surgery, London, United Kingdom Objectives: The tibial tubercle to trochlear groove distance is used as a metric to indicate whether medialisation of the tubercle is an appropriate surgical technique to improve patellar-femoral stability in individual patients. The standard method of measurement is to obtain axial CT slices through the femur and tibia and determine the horizontal offset of the trochlear groove on one slice from the tubercle on another. The horizontal reference is taken from the posterior femoral condyles. In

123

S290

Fig. 1 Range of absolute deviations measured the literature only the slices around the knee are used with no reference to the overall leg. When measuring anatomical features of the knee to plan joint replacement surgery it had been observed that patients were generally not exactly aligned with the axis of the CT scanner. The objectives of this study were to measure the amount of misalignment typically encountered, determine its effect on TTTG measurements and provide a method of compensating for misalignment. Methods: Thirty-two anonymised patient datasets were analysed. Each dataset contained hip, knee and ankle data. Hips and ankles were scanned with an increased slice spacing to reduce X-ray exposure. The legs were axially aligned by rotating the scans until the posterior condyles were level. Landmarks were placed on the femoral head surface and a spherical least-squares fit obtained to determine head centre. Landmarks were placed on the talus centre (a slice was chosen where it appeared trapezoid and the crossing point of lines between the four corners was computed). The angle of the line from the femoral head centre to the talus centre was then measured in the AP view, relative to the longitudinal axis of the CT scan. Results: Figure 1 shows the deviations measured. The mean absolute deviation was 1.9. Both varus and valgus deviations occurred. A geometrical model of the geometry of the TTTG measurement was made using typical knee dimensions—a TTTG of 18 mm and a distance of 70 mm between the femoral and tibial measurement slice. It was calculated that each 1 of alignment error caused a TTTG measurement under- or over-estimate of approximately 1.1 mm. In the study, this indicates a TTTG measurement error on average of approximately 2 mm, and up to 5.5 mm in the worst case observed. Conclusions: It has been shown that patients are not always scanned with their legs straight in the CT scanner. In many cases the error will be small, however outliers may cause mis-classification of the extent of the TTTG distance. By determining the deviation of the limb from the scanner axis using proximal and distal landmarks, it is possible to estimate the TTTG error and compensate, allowing measurements to be made irrespective of leg alignment.

P28-1158 Patello-femoral kinematics in relation to medial patello-femoral ‘ligament’ F. Colle1, B. Sharma2, S. Bignozzi1, D. Dejour3, S. Zaffagnini2 1 Istituto Ortopedico Rizzoli, University of Bologna, Laboratorio di Biomeccanica e Innovazione Tecnologica, Bologna, Italy, 2Istituto

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Ortopedico Rizzoli, University of Bologna, Bologna, Italy, 3LyonOrtho-Clinic, Lyon, France Objectives: Understanding of the patello-femoral joint pathology, its natural history, rationale for various treatments and evaluation of their efficacy is currently challenged by the lack of standardized information on the kinematics of this joint. In this setting, studies are needed to analyze association between the morphology of the joint in relation to the distal femur, its restraints and kinematics. Starting from the hypothesis that the medial patellofemoral ‘ligament’ (MPFL) is only a restraint during motion, against a load inducing lateral shift, functioning as an aponeurosis, to guide the patella through a trochlear groove that is related to the trans-epicondylar and posterior condylar axis, this study wants to analyze the morphology of the patellotrochlear joint and its influence on patello-femoral kinematics with reference to the presence or absence of the medial patellofemoral restraint. Methods: This is a time zero, in vitro study. A kinematic analysis of six cadaveric knees, three of each side, was performed recording with a navigation system the passive flexion–extension range of motion (ROM) between 10 and 120, using the center point of patella as reference over posterior condylar and mechanical axis of the femur. Patello-femoral kinematics was recorded under an axial quadriceps load of 60 N, with free tibial rotation and eliminated femoral anteversion. Patella kinematic tests were conducted with and without a lateral load of 25 N, to evaluate tilt and lateral shift in two different anatomical condition, with natural and without the medial patellofemoral ligament. Results: A wide variation in MPFL femoral insertion was noted. In comparison to the MPFL-intact state, the patella lateralized the path in MPFL-deficient state, even without lateral load. The variability in kinematics could not be explained on the basis of variation in trochlear morphology. MPFL was anisometric, the insertion points of the inferior bundles coming closer in flexion. Conclusions: While, MPFL may guide the patella shift and tilt during knee motion, in normal knee trochlear morphology does not influence kinematics. The ligament act only as a passive restraint and its complex anatomical structure allows it to be anisometric during range of motion.

P28-1184 In vitro measurement of native patella kinematics in different loading conditions and motor tasks H. Vandenneucker1, L. Labey2, Y. Chevalier2, B. Innocenti2, J. Victor3, J. Bellemans4 1 UZ Pellenberg KU Leuven, Department of Orthopaedics, Knee and Sports Medicine, Pellenberg-Lubbeek, Belgium, 2Smith & Nephew, Inc, European Centre for Knee Research, Leuven, Belgium, 3AZ St Lucas Brugge, Department of Orthopedics, Brugge, Belgium, 4 University Hospitals Leuven, Catholic University Leuven, Orthopaedic Department, Pellenberg, Belgium Objectives: Clinical problems of the knee joint are often related to the patella. Despite this, patella biomechanics has not been investigated to a great extent dynamically but mainly in passive or quasi-static conditions. The objective of this study was to document patella kinematics in different loading conditions and motor tasks, using a well defined methodology. Methods: Twelve fresh frozen full leg cadaver specimens were tested. Frames with reflective markers were rigidly fixed to tibia, femur and patella and a computed tomography (CT) scan was made. Femur and tibia were embedded, properly aligned in frontal and sagittal planes. Medial and lateral hamstrings tendons were prepared for attachment to constant load springs (50 N each). The quadriceps tendon was prepared to be clamped to the motor of a knee rig.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1 3D models of the bones were used to identify landmarks and define coordinate frames. Patella marker trajectories were then transformed to following rotations and translations: patella flexion (around Ft), rotation around Pfl) and tilt (around Py); anteroposterior (AP), mediolateral (ML) and inferior-superior (IS) translations, all along the coordinate axes of the femur

The knee rig provides 6 of freedom to the knee joint. It simulated loaded open chain and squatting motions while infrared cameras recorded the marker trajectories. Four motor tasks were tested: passive motion, open chain extensions with and without a weight of 30 N fixed to the distal tibia, and squats with constant vertical ankle force of 130 N. During each motor task, four muscle load combinations were tested: quadriceps only, quadriceps + medial hamstrings, quadriceps + lateral hamstrings and quadriceps + both hamstrings. Patella kinematics was calculated using the coordinate system in Fig. 1 and is reported as a function of knee flexion angle. Results: Variability among specimens was high, but some consistent motion patterns were clearly visible. On average, patella flexion was linearly related to knee flexion. We found almost constant patella rotation, tilt and ML translation. The patella shifted almost linearly posteriorly and inferiorly with increasing knee flexion. There were some differences in kinematics between motor tasks. Compared to squatting, passive and open chain motions showed more patellar flexion and a more inferior and posterior position of the patella. During passive motion, the patella gradually rotates externally compared to squatting. Patella tilting and ML translation were the same for all motor tasks. Hamstrings forces only affected patella flexion in motor tasks with small quadriceps force. Conclusions: The objective of this study was to investigate patella kinematics and how it is influenced by motor task and muscle load. This was done in a reproducible and reliable way. The data show that patella kinematics is indeed different between motor tasks, mainly due to differences in quadriceps forces. Hamstrings loads, though important for tibio-femoral kinematics, have less impact on patella kinematics.

P28-1217 The biomechanical relationship between Anterior Knee Pain Syndrome (AKPS) and Functional Hallux Limitus (FHL). A case series review S. Diehl1, J. Vallotton1, Functional Hallux Limitus Study Group 1 Medicol Sa`rl/Clinique Bois Cerf, Lausanne, Switzerland

S291 Objectives: To determine the close relationship between AKPS and the presence of FHL. Demonstrate that the sagittal plane blockade caused by the Flexor Hallucis Longus tendon entrapment at the retrotalar pulley is a major predisposing factor for AKPS of unexplained origin. Describe a conservative and surgical treatment protocol for patients with AKPS and FHL. Methods: A prospective cohort of patients with AKPS, whose physical therapy program failed and were treated with endoscopic retrotalar release of the Flexor hallucis longus (Fhl) tendon, was compared with a retrospective cohort of patients treated conservatively with success. A complete orthopaedic examination was performed to exclude any other possible etiology for AKPS. FHL was diagnosed with a specific stretch test. Conservative treatment consisted of subtalar manipulation, unlock of the subtalar joint with the Hoover cord maneuver, muscular strengthening and propioceptives exercises. Surgical treatment consists of section of the retrotalar pulley endoscopically. For both cohort of patients a questionnaire for data collection was done. Results: Patients with AKPS who were treated for FHL in the same time, conservatively or surgically, show a high rate of treatment success with disappearance of symptoms. When conservative treatment fails the surgical release is a safe and effective procedure to restore normal biomechanics in the PF joint. High rate of patient’s satisfaction is found in this series without complications. Conclusions: There exists a strong relation between AKPS and FHL. Location of the pain on the medial aspect of the patello-femoral joint can be explained by the asynchronism in gait and a modification in impact forces at heel strike followed by an hyperpronation in final stance phase. Treatment of the FHL, conservatively or surgically, is a good safe and reliable strategy in patients with AKPS of unexplained origin.

P28-1232 The value of the MPFL suture for acute patella luxation in consideration of risk factors D. Wagner1, C. Loebig1, F. Pfalzer1, J. Huth1, F. Mauch1, G. Bauer1 1 Sportklinik Stuttgart, Stuttgart, Germany Objectives: Patella dislocation is an injury often found in young, physically active patients. In cases of lateral dislocation of the patella, it is generally accompanied by an injury of the MPFL (medial-patella-femoral-ligament), which has been identified over the last few years as the main soft tissue stabilizer of the patella. Conservative treatment of acute patella dislocation leads to high rates of recurrent dislocation (30–60%) and thus to dissatisfactory results for young and active patients. An established procedure to treat patella dislocation is to gather the medial ligament complex (Yamamoto sutures). This procedure implements unspecific suturing of the medial patellar ligament and does not take risk factors into consideration. MRT diagnostics permit the exact localization of the MPFL injury (near the patella, femoral or mid-substance). This allows the exact treatment of the injured MPFL with local sutures. Methods: In this study, 50 patients (27 men, 23 women) with acute patella dislocations were examined prospectively, underwent arthroscopic surgery, and were treated with a local suture. Following the clinical examination, MRT and X-ray diagnostics were done to determine the precise location of the rupture, to assess accompanying injuries (particularly cartilage lesions) and to record other anatomical variables (degree of trochlear dysplasia, Insall Salvati index, TTTG distance, trochlear SLOPE, Patella-TILT). After 1 year p.o., subjects underwent a clinical re-examination with MRT diagnostics to assess the medial patellar ligament complex, changes in Patella-TILT, and to monitor accompanying injuries of the cartilage. In addition, subjects were questioned about their ability to

123

S292 participate in sports (Valdarrabano), recurrent dislocations, and their satisfaction with p.o. result. The Lysholm and Kujala score was also determined at this time. The recorded results were then related to the known risk factors. Results: We found a 6% rate of recurrent dislocation. 1 patient reported a subluxation. 80% of the subjects were able to achieve their pre-operative level of sports activity. The average Kujala score was 88 points. The average patellar tilt was i15.4. 95% of the patients were satisfied with the results and would undergo the intervention again. All patients with recurrent dislocation and subluxation showed higher or combined risk factors. In addition, we noted a trend towards lower scores and decreased sports activity which went along with higher risk factors. Conclusions: In the authors’ opinions, the described procedure is indicated in patients with acute injuries and those with slight to moderate risk factors. MPFL reconstruction and/or the treatment of the additional factors should be discussed in the face of greater risk factors.

P28-1320 Long term results after arthroscopic suturing of the medial patello-femoral ligament in children after acute, first episode of patella dislocation M. Drwiega1, U. Zdanowicz2, R. Smigielski3 1 Praktyka Lekarska Michal Drwiega, Orthopaedics, Warsaw, Poland, 2 Carolina Medical Center, Orthopaedics, Warsaw, Poland, 3Prywatna Praktyka Lekarska Robert Smigielski, Nieporet, Poland Objectives: Dislocation of the patella are very often cause of knee dysfunction in children. This is a result of abnormal knee anatomy as well as an inborn joint laxity or direct injury to the kneecap. The most common way of treatment of this kind of injury is joint punction and immobilization for a few weeks in a plaster of Paris. After that, children are usually preserved from physical activity. According to literature the average rate of recurrent dislocation after conservative treatment reaches from 32 to 48%. We present long term results of an arthroscopic procedure that allowes to stabilize patella in children after acute, first episode of patella dislocation. Methods: 48 patients were included into retrospective evaluation. Inclusive criteria were first episode of patella dislocation, only arthroscopic procedure performed and open growth plates around the knee at the time of surgery. The mean age of children at the time of surgery was 14.8 years (range 11–17). During each procedure the medial patello-femoral ligament was sutured arthroscopically with use of two or three bio-absorbable sutures. 5 patients did not respond the request for evaluation. 43 patients (18 girds and 25 boys) underwent both clinical and radiological evaluation. Lysholm score questionnaire and IKDC form were fulfilled. Results: Results of this kind of surgery are very promising. A good stability of patella was achieved in most of patient. In 3 cases redislocation of the operated knee cup was recorded. According to Lysholm score questionnaire and IKDC 78% reported good and excellent results, 15% fair and 7% poor results (3 cases of redislocation). Conclusions: Presented method of treatment seems to be efficient in stabilization of the patella in children after acute episode of patella dislocation. The whole procedure is short and restore initial length of medial patellar retinaculum. It do not guarantee full stability, but increases chance of good and excellent result of treatment in children,

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 were most of realignment procedures are contraindicated because of open growth plates.

P28-1324 The relation of the distal femoral physis and the medial patellofemoral ligament M. Nelitz1, S. Lippacher2, D. Dornacher2, H. Reichel1 1 Ulm University, Department of Orthopaedic Surgery, Centre of Musculoskeletal Research Ulm, Ulm, Germany, 2Ulm University, Orthopaedic Surgery, Ulm, Germany Objectives: The purpose of this study was to analyse true lateral radiographs of children and adolescents to determine the relation of the origin of the MPFL and the distal femoral physis considering the complex anatomy of the physis. The hypothesis was that the femoral insertion of the MPFL is distal to the growth plate. Methods: Antero-posterior and true lateral radiographs from PACS computer records of 27 patients with a history of patellofemoral instability were assessed. To determine the femoral origin of the MPFL, the method by Scho¨ttle et al. and the method by Redfern et al. were applied independently. On the anteroposterior radiograph, the distance between the medial most part of the physis and the central part of the physis was measured to quantify the concave curvature of the physis. To cross-reference the femoral insertion of the MPFL onto an ap view, the projected MPFL origin-physis distance was subtracted from the distance between the most medial part of the physis and the central part of the physis. Results: The projected median origin of the MPFL as measured on a lateral radiograph was located 3.2 mm (1.2–5.8 mm) proximal to the physis. The median distance between the most medial part of the physis and the physeal line on the anteroposterior radiograph was 9.9 mm (4.1–12.0 mm). Subtracting the two measured values, the median origin of the MPFL as seen on the ap view was 6.4 mm (2.9–8.5 mm) distal to the femoral physis. Conclusions: Considering the concave curvature of the distal femoral physis, it can be assumed that the femoral insertion of the MPFL is distal to the femoral physis. As a too proximal insertion of the graft can cause unintentional tightening of the MPFL in knee flexion, these results have to be considered when performing reconstruction of the MPFL in children and adolescents with open growth plates.

P28-1340 Analysis of failed surgery for patellar instability in children with open growth plates M. Nelitz1, J. Woelfle2, S. Lippacher2, H. Reichel1 1 Ulm University, Department of Orthopaedic Surgery, Centre of Musculoskeletal Research Ulm, Ulm, Germany, 2University of Ulm, Orthopaedic Surgery, Ulm, Germany Objectives: Many surgical procedures have been proposed to treat recurrent patellar dislocation in children. In recent years a more tailored approach considering the underlying pathology has been advocated. The aim of the study was to analyze a group of patients with recurrent patellofemoral instability after unsuccessful operative stabilization (Roux-Goldthwait procedure, lateral release, medial reefing or in combination) in childhood and adolescence. Methods: 37 children and adolescents with recurrent patellofemoral instability despite previous surgery were analyzed retrospectively. Radiographic examinaton included AP and lateral views to assess

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S293

patella alta and limb alignment. MRI was performed to evaluate trochlear dysplasia and tibial tubercle—trochlear groove (TTTG) distance. As a control group 23 age- and sex-matched adolescents that were treated with a favourable outcome after medial reefing alone or combined with a Roux-Goldthwait procedure were analyzed. Results: Severe trochlear dysplasia (type B–D according to Dejour) as detected on MRI scans was found significantly more often in the study group (89%) than in the control group (21%). No statistical difference of patellar height ratio (Insall-Salvati index) and TTTG distance between the two groups could be found. Conclusions: Of the measured parameters only the incidence of trochlear dysplasia was increased. Trochlear dysplasia therefore seems to be a major risk factor for failure of operative stabilization of recurrent patellofemoral instability in children and adolescents. The results in children are in consensus with the literature in adults that a more tailored operative therapy including reconstruction of the MPFL and trochleaplasty has to be considered.

about an improvement of their pain situation. This fact is reflected in a highly significant improved IKDC- (from 47.5 ± 12.5 to 74.9 ± 20.7, p \ 0.001) and Kujala-score (from 52.3 ± 14.2 to 80.8 ± 22.2, p = 0.001) at a mean follow-up of 26 months. As far as sports activity is concerned, a statistically significant improvement of the Tegnerscore up to 5.1 ± 1.8 (preoperative 2.1 ± 1.0, p \ 0.001) could be achieved. The radiological analysis showed a significant decrease of both, patellar tilt (from 16.7 ± 7.0 to 10.4 ± 6.3, p = 0.003) and patellar shift (from 3.9 ± 3.8 to 2.4 ± 1.5, p = 0.029). Conclusions: On the basis of these results it can be shown that by the reconstruction of the MPFL and the lateral retinaculum both, medial and lateral stability of the patella are regained. After one or more unsuccessful previous efforts of stabilizing the patella by LR, the technique performed in this essay leads to a considerable improvement of subjective scores, objective clinical and radiological parameters as well as sports activity.

P28-1343 Reconstruction of the lateral retinaculum after failed lateral release in case of patellofemoral instability: a prospective study G. Meidinger1, M. Mu¨nch1, I. Beermann1, I.J. Banke1, A.B. Imhoff1, P. Scho¨ttle2 1 TU Munich, Klinikum rechts der Isar, Department of Orthopaedic Sports Medicine, Munich, Germany, 2Orthopa¨die am Zu¨richberg, Praxis Munzinger, Zu¨rich, Switzerland Objectives: For a long time lateral release (LR) was performed as a standard procedure both, for patellofemoral pain syndrome (PFPS) and patellofemoral instability (PFI). However, recent biomechanical studies have shown that LR is not useful for decreasing the lateral force onto the patella, but is increasing not only medial but also lateral PFI. Furthermore, pain on palpation over the lateral patellofemoral joint space can result in patients treated with extensive LR. We postulate that in case of persistent PFI or PFPS after failed LR the reconstruction of the lateral retinaculum as an addition of the reconstruction of the medial patellofemoral ligament (MPFL) is necessary in terms of decreasing medial PFI as well as lateral pain. Methods: In between 03/07 and 04/09 we have seen a total of 15 patients (13 f, 2 m) with persistent PFI and palpatory pain over the lateral retinaculum due to unsuccessful treatment of PFI with a LR. These patients have undergone revision surgery with an anatomical reconstruction of the released lateral retinaculum in combination with a reconstruction of the medial patellofemoral ligament (MPFL) in an aperture-technique using the gracilis tendon. The average age at time of operation was 25.3 ± 8.3 years. Preoperatively, as well as 6 weeks, 3 months, 6 months, 12 months and 24 months postoperatively, clinical examinations were performed and subjective as well as objective scores (Kujala-, Tegner-, IKDC-score) were evaluated. Regarding radiological parameters measurement of patellar tilt and shift was carried out on axial radiographs before and after the operation. Results: During the first 2 years after the operation only one redislocation could be recorded. Pain on palpation over the reconstructed lateral retinaculum was remaining in two patients (13%). Although some patients complained about persistent patellofemoral pain on exertion or after enduring flexion of the knee, all of them reported

P28-1346 In vitro navigated kinematic validation of the tensioning of MPFL reconstruction D. Dejour1, B. Sharma2, N. Lopomo3, S. Bignozzi4, S. Zaffagnini5 1 Lyon-Ortho-Clinic, Lyon, France, 2Istituto Ortopedico Rizzoli, University of Bologna, Bologna, Italy, 3Istituto Ortopedico Rizzoli, Laboratorio di Biomeccanica, Bologna, Italy, 4Rizzoli Orthopaedic Institute, Biomechanics Lab, Bologna, Italy, 5Istituto ORTOPedico Rizzoli Bologna, Bologna, Italy Objectives: The goal was the validation of a MPFL bony reconstruction, in terms of affect of tunnel positioning and tension method, on the patellar shift and tilt at 0, 30, 60, 90 of knee flexion. Methods: Six fresh-frozen, cadaveric lower limbs, were tested. Three acquisitions was done for each configuration; Normal MPFL, MPFL cut, MPFL reconstructed. Two bi-cortical pins were fixed to the femur and tibial shaft to load the navigation reference frames and a navigation tracker was fixed on the patella with a single screw. The arrangement allowed positional measurements of patellar motion to be tracked in 6 of freedom, while a 60 N axial force was applied to the quadriceps tendon. A 25 N lateral load was applied at 0, 30, 60, 90. The medio-lateral patellar shift and tilt were recorded. The MPFL reconstruction was done with a Gracilis tendon passed in a loop in two tunnels in the proximal anterior patellar cortex. Then the two strands were tensioned in a femoral tunnel in the anatomic position. A temporary fixation with tension adequate to engage the patellar lateral facet in extension was applied to the MPFL, and after ten cycles, the final fixation was done at 70 knee flexion. Results: Anatomical and kinematic data were compared between the normal MPFL, cut MPFL and reconstructed. There was a significant difference in M-L shifting and tilting of the MPFL-intact and reconstructed, compared to the MPFL-cut. The static patellar shift in MPFL-reconstructed, with or without load, was comparable to patellar shift in MPFL-intact in the current study (Fig. 1). On average the MPFL femoral tunnel was placed more proximal to the natural MFPL barycentre (Fig. 2). The dynamic PF shift kinematics recorded an under-constrained patella in early flexion and over-constraint in late flexion, while an opposite effect was recorded in patellar tilt (Fig. 3).

123

S294

Fig. 1

Fig. 2

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 the appropriate form of treatment. Many imaging modalities and measurements have been described to investigate the patellofemoral joint. MRI has evolved as an important tool for investigation since it provides invaluable information on the multifactorial aspects of patellofemoral instability. However, patients are generally examined in prone position with relaxed muscles with certain limitations in displaying the complex dynamic pathology. We hypothesized that results for common measurements for evaluation of instability factors are different with weight bearing compared to non-weight bearing conditions at MRI examination. Methods: In a diagnostic clinical study, only patients with a history of C2 symptomatic lateral patellar dislocations were included. MR scans were taken in an open 0.6 Teslan MRI (Upright MRI, FONAR Inc., Melville, NY, U.S.A) with T1 weighted sagittal and axial Spin echo images (TR 140 ms; TE, 15 ms; slice thickness, 2.5 mm). Examinations were done with the knees in 0 of flexion and the foot in 15 external rotation. First, the patients were evaluated in an upright standing position with full weight bearing and second in prone position. Measurements of MR images were analyzed with a common software (JiveX, VISUS Technology Transfer GmbH, Bochum) and included the relative lateral patellar displacement (Bisect Offset), the patellar tilt angle, different patellar height ratios (IS: Insall Salvati, CD: Caton-Deschamps, BD: Biedert-Albrecht) and the tibial tubercle trochlear groove distance (TT-TG). A paired samples t test and the Pearson correlation coefficient were used for statistical analysis with a significance level of p \ 0.05. Results: To date, 12 patients (8 female, 4 male) with an average age of 21.8 ± 3.2 years participated in the study. With weight bearing, the relative lateral displacement was significantly higher (10.4 ± 0.2 mm) compared to non-weight bearing conditions. The patellar tilt was increased by 2.2 ± 0.2. The patellar height ratios were also increased with weight bearing (IS: +0.04; CD: +0.03, BD: 7 ± 3.6%). In contrast, the average TT-TG was significantly lower (2.2 ± 0.71 mm). Significant correlations were found for the lateral displacement and patellar tilt as well as for the patellar height ratios. Conclusions: The analysis of common measurements for evaluation of instability factors in patients with patellofemoral instability revealed significant differences with weight bearing compared to nonweight bearing conditions at MRI examination. More information is needed on a larger amount of patients, the differences at higher flexion angles and a matched control group without patellofemoral instability.

Fig. 3

Conclusions: This study confirm the major role of the MPFL in case of medial loading between 0 and 60, validated the MPFL graft positioning and the original method of tensioning which will prevent the over tensioning leading to pain and stiffness.

P28-1352 MRI-assessment of various contributing factors in patellofemoral instability with and without weight bearing C. Becher1, M. Rase1, T. Calliess1, M. Ettinger1, C. Siebert1, S. Ostermeier1 1 Hannover Medical School, Orthopaedic Department, Hannover, Germany Objectives: The analysis of contributing factors in patients with patellofemoral instability is of high importance in decision making for

123

P28-1381 Overmedialized tibial tubercle as a failed surgery for patellar instability: principals of surgical correction and clinical outcome of revision surgery P. Sillanpa¨a¨1 1 Tampere University Hospital, Orthopeadic Surgery and Trauma, Tampere, Finland Objectives: Medialization of the tibial tubercle (TT) has been used as a surgical solution for patellar instability. Overmedialization results in iatrogenic abnormality and may not be able to stabilize patella. However, clinical characteristics overmedialized TT and results of surgical correction are unknown. Methods: Seven patients with failed previous TT medialization requiring revision surgery due to recurrent patellar instability were included in the study. Standard radiographs, magnetic resonance imaging (MRI), lower limb axial radiographs and rotational CT scans were analyzed to assess any abnormalities in affected limb. Postsurgical TT-trochlear groove (TT-TG) distance was analyzed by MRI as

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 well as patellar height (patello-trochlear index) and trochlear dysplasia. Varus-valgus alignment was assessed by axial radiographs and rotational deformities of the lower limb by rotational CT study. Previous MRI scans were also reviewed to assess the preoperative TT-TG value prior failed TT medialization. Median age was 23 years (range, 17–31). Follow-up was 6–36 months and the follow-up evaluation included recurrent patellar instability, assessment of subjective symptoms and functional limitations. Results: In patients with failed TT medialization, the mean postsurgical TT-TG value was -2–8 mm, preoperatively being 9–20 mm. In all patients, TT medialization had been performed without existing anatomical abnormality. Following surgical procedures were considered necessary to address major osseus abnormalities found in imaging studies: correction of the malpositioned TT at its anatomical location (to preoperative location) in all patients, distal femoral derotational osteotomy in three patients, distal femoral varising osteotomy in one patient, combined derotational and varising osteotomy in one patient, TT distalization in two patients. Medial patellofemoral ligament (MPFL) reconstruction was performed in all patients. Two most severe cases with less than zero TT-TG value, presented limping with internal rotation of the tibia in gait analysis. All patient had developed core stability dysfunction and extensor muscle weakness. All revision TT osteotomies healed without complications. At follow-up, all patients were satisfied with the result and would have chosen to undergo the same revision surgery again. None had recurrent patellar instability. The median Kujala score at followup was 90 (range 79–100) and 71 (range 54–80) before revision surgery. Conclusions: Overmedialization of the TT may fail to stabilize patella and does not correct the other abnormalities related to patellar stability. Overmedialized TT can be revised by correcting the malpositioned TT to its anatomical location. In this study, other osseus abnormalities were frequently found and required surgical correction. MPFL reconstruction is recommended to assess the soft tissue restraint insufficiency. Long-term studies are required to confirm the role of revision surgery after failed TT overmedialization.

P28-1481 Presence of trochlear dysplasia associated with less progression of tibiofemoral osteoarthritis following patellofemoral arthroplasty M.M. Kalisvaart1, S.W. Slettedahl2, D.L. Dahm1 1 Mayo Clinic, Orthopedic Surgery, Rochester, United States, 2 Mayo Clinic, Biomedical Statistics and Informatics, Rochester, United States Objectives: Patellofemoral arthroplasty designs are intellectually appealing because of the contention that they can be an effective treatment for patients with isolated anterior compartment osteoarthritis of the knee. However, long-term failure as a result of tibiofemoral degeneration has been reported to occur in up to 25% of patients. We carried out this retrospective review of the results of patellofemoral arthroplasty performed by a single surgeon at a single institution to determine factors associated with better clinical patient outcomes and progression of tibiofemoral degenerative joint disease. Methods: Sixty-one patients with isolated patellofemoral osteoarthritis were treated with a patellofemoral arthroplasty by a single surgeon between 2003 and 2009. Fifty-nine patients were available for analysis with a mean follow-up of 3.5 years. Patients were evaluated by measuring range of motion and with the use of the Knee

S295 Society clinical rating system, the Tegner Activity Level scale, and the UCLA Activity Score. In addition, preoperative radiographs were reviewed for evaluation of patellofemoral and tibiofemoral compartmental osteoarthritis and presence of trochlear dysplasia, and postoperative radiographs were reviewed for evaluation of progression of tibiofemoral degenerative arthritis. Furthermore, multivariate statistical methods were applied to study factors that may influence the final outcome. Results: There was no statistically significant association between age, gender, history of prior knee surgery, patellar height, patellofemoral osteoarthritis severity, patellar and femoral component size, or performance of lateral release with patient pain and function (as measured by the Knee Society scores) or progression of tibiofemoral joint osteoarthritis at final follow-up. Patients with preoperative trochlear dysplasia, however, did have significantly less radiographic evidence of tibiofemoral joint osteoarthritis progression compared with patients without trochlear dysplasia at final follow-up (p \ 0.0001). Conclusions: In this retrospective study, patients with preoperative radiographic evidence of trochlear dysplasia experienced less progression of tibiofemoral degenerative joint disease than patients without trochlear dysplasia at a mean follow-up of 3.5 years.

P28-1483 Validation of a classification for patients with patellar instability L.A. Hiemstra1, M.R. Lafave2, S. Kerslake3, S.M.A. Heard1, G.L. Buchko1 1 Banff Sport Medicine, Orthopaedic Surgery, Banff, Canada, 2Mount Royal University, Physical Education, Calgary, Canada, 3Banff Sport Medicine, Research, Banff, Canada Objectives: Patients with patellar instability present with a wide range of etiologies and symptoms as well as varied anatomic and neuromuscular characteristics. As with shoulder instability, these patient types present as a continuum. For patella instability, the authors propose that two main subsets of patients present. This study introduces a novel classification system, WARPS (Weak, Atraumatic, anatomy Risky, Pain, and Subluxation) and STAID (Strong, Traumatic, Anatomy normal, Instability and Dislocation) The purpose of this study was to demonstrate inter-observer reliability of the classification continuum of WARPS and STAID for patella instability. Methods: Twenty-five consecutive patients with a confirmed diagnosis of patellar instability were included. Each patient underwent a thorough history and physical examination concurrently with 3 assessors (orthopaedic surgeon, physiotherapist and athletic therapist). Each characteristic of the WARPS-STAID classification continuum was graded independently by each examiner on a 10 cm VAS. Each assessor also assigned a cumulative WARPS-STAID score to each patient. Intraclass Correlation Coefficient (ICC 3.3) was calculated for each pair of characteristics on the continuum and for the cumulative scores. Results: Intraclass Correlation Coefficient (3.3) of the WARPSSTAID classification continuum were W–S = 0.92; A-T = 0.95; R-A = 0.81; P–I = 0.84; and S-D = 0.91. The ICC (3.3) of the composite score was 0.84. Conclusions: These results demonstrate strong reliability of the classification system, WARPS-STAID for patients with patella instability. The classification system reliably differentiates between two distinct subsets of patella instability patients.

123

S296

Knee-Osteotomy

P29-4 A ‘‘safe zone’’ in medial open wedge high tibia osteotomyto prevent lateral cortex fracture K. Nha1, S. Han2, D.H. Lee3, K.H. Wang1, H.D. Lee1 1 Inje University, Ilsanpaik Hospital, Koyangsi, Republic of Korea, 2 Korea University, Anam Hospital, Orthopaedic Surgery, Seoul, Republic of Korea, 3Korea University, Anam Hospital, Seoul, Republic of Korea Objectives: Lateral cortex fracture in open high tibia osteotomy (OHTO) may result in displacement of the osteotomy, and hence lead to the loss of correction, and this way it may contribute to recurrent varus deformity. It has not been previously described how the level of osteotomy affect the lateral cortex fracture. The purpose of this study is to find the safe zone in lateral cortex according to the level of osteotomy and dissect the lateral structures that act the lateral cortex hinge. Methods: Medial OHTO was performed in nine fresh frozen human cadavers (eighteen paired knees). The left and right legs of each specimen were randomly assigned to a ‘‘safe zone’’ osteotomy (group A: between the fibular tip and circumference line of the fibular head) or a lower level osteotomy (group B: distal to the circumference line of fibular head) under the fluoroscopy. Lateral cortex fracture and fracture gap after the osteotomy were measured and compared group A and B using AO tomofix spreader. We dissected the lateral fracture site and analysed the soft tissue structures which correlate to the fracture site and measured the length of lateral capsule. Results: In group A, there was no lateral cortex fracture and fracture gap. In group B, six out of nine knees (67%) developed the lateral cortex fracture, and fracture gap was found five of six cases when the osteotomy site was distracted to a maxiumu of 20 mm. All fracture site or gap were found in the extensor muscle area, not in the lateral capsule. Following group B, lateral cortex fractures were significantly developed than group A (P \ 0.05). Length of lateral capsule was from 3 mm to 18 mm from joint line. Conclusions: The level of osteotomy site in OHTO affects the lateral cortex fracture and fracture gap. Group B increased the lateral cortex fracture and fracture gap after OHTO. Safe zone is the lateral capsular area, which is from tip to circumference line in fibular head.

P29-77 The effect of medial opening and lateral closing high tibial osteotomy on leg length E. Servien1, S. Lustig2, G. Demey3, P. Neyret4 1 Hopital de la Croix-Rousse, Centre Albert Trillat, Hospices Civils de Lyon-Lyon University, Lyon, France, 2Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 3Centre Albert Trillat, Lyon-Caluire, France, 4Hopital Croix-Rousse, Centre Livet, Chirurgie Orthopedique, Caluire-et-Cuire, France Objectives: High tibial osteotomy (HTO) is a common treatment for medial compartment arthritis of the knee in younger, more active patients. An HTO also potentially affects leg length. Mathematical models predict that the osteotomy type (medial opening-wedge vs. lateral closing-wedge) and the magnitude of the correction determine the change in leg length, but no in vivo studies have been published. The purpose of this study was undertaken to quantify and compare leg-length change after opening-wedge and closing-wedge HTO.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: Thirty-two medial opening-wedge and 32 lateral closingwedge HTOs were selected from patients treated at the authors’ institution. Preoperative and postoperative coronal plane alignment and leg length were measured and surgical details were collected. Results: The 64 osteotomies were performed at an average age of 57 years. The mean opening wedge was 9.3 mm and the mean closing wedge was 8.0 mm. Mean knee alignment changed from 174 preoperatively to 183 postoperatively in both groups. In the medial opening-wedge group, entire leg length changed from 836.3 ± 63.5 mm to 841.8 ± 64.1 mm, an increase of 5.5 ± 4.4 mm (P \ .0001). Tibia length changed from 368 ± 30.9 mm to 372.3 ± 31.2 mm, an increase of 4.3 ± 2.3 mm (P \ .0001). In the lateral closing-wedge group, entire leg length changed from 840.6 ± 51.5 mm preoperatively to 837.9 ± 52.0 mm postoperatively, a decrease of 2.7 ± 4.0 mm (P = .0008). Tibia length changed from 365.1 ± 23.2 mm to 361 ± 22.9 mm, a decrease of 4.1 ± 2.9 mm (P \ .0001). The difference in mean leg-length change between opening-wedge and closing-wedge osteotomies was 8.2 ± 5.9 mm (P \ .0001). Conclusions: Both medial opening-wedge and lateral closing-wedge HTO can result in significant leg-length change, but changes are generally less than mathematical models predict.

P29-125 The role of high tibial osteotomy with ACL reconstruction in ACL-deficient varus knees with medial compartment OA: systematic review A. Anand1, A. Patel2, B. Anand3 1 West Hertfordshire Hospitals NHS Trust, Trauma & Orthopaedics, London, United Kingdom, 2Barnet Hospital, Trauma & Orthopaedics, London, United Kingdom, 3Chelsea & Westminster Hospital, Trauma & Orthopaedics, London, United Kingdom Objectives: Young, active patients with symptomatic ACL-deficient knees, varus malalignment and symptomatic medial compartment osteoarthritis are often a challenge to manage. There is considerable controversy regarding the management of these patients with various surgical options described, including HTO alone, ACL reconstruction alone, staged HTO with ACL reconstruction, simultaneous HTO with ACL reconstruction, and more recently UKR with ACL reconstruction. (1) The primary aim of this systematic review was to determine whether there is improvement in knee pain and functional stability in patients undergoing simultaneous HTO with ACL reconstruction for instability and medial compartment OA symptoms. (2) Secondary aims were to assess whether these patients where able to return to sporting activity and delay the progression of radiological osteoarthritis. Methods: A literature search on MEDLINE (Ovid), PubMed, EMBASE and the Cochrane library databases was performed on 21st January 2010. Eligibility criteria included studies in English, patients aged 18–50 years, and undergoing a simultaneous HTO with ACL reconstruction using either autograft or allograft hamstring or bonepatellar-bone grafts. Studies with patients with concomitant PCL, MCL, LCL or PLC injuries were all excluded, in an attempt to reduce the influence of confounding factors. We also excluded small studies (n \ 6) and studies looking at patients with ACL ruptures with varus knees, but with no medial compartment symptoms. Results: Six studies fully met the eligibility criteria and were critically appraised. No randomised controlled trials or meta-analyses were found. All of the studies demonstrated that pain, subjective and objective knee stability had improved post-operatively. A high degree of variability was reported for major complications (5–37%), with most caused by technical errors in performing the osteotomy. 1 study demonstrated that the valgus tibial osteotomy may reduce the

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 progression of medial compartment OA, at the expense of increasing radiographic lateral and patello-femoral compartment OA. Conclusions: Simultaneous HTO and ACL reconstruction offers young, active patients a procedure that can improve knee pain and functional stability. Patients are often able to return back to sporting activity, but rarely to their pre-injury or competitive level. The combined procedure has the advantage of only one rehabilitation period and should be advocated early before the onset of severe medial compartment OA. Limitations and Future research. All the studies published and critically appraised on this area are retrospective case-series. Very few studies were able to provide functional outcome scores pre- and post-surgery. Further research in the form of large well-designed prospective randomised controlled trials with rigorous methodology are needed comparing the various treatment options for these patients, in order to ensure that these young, athletic patients receive the best treatment available.

P29-370 How much is the tibial width in a high tibial osteotomy important? V.J. Leo´n Mun˜oz1, A.J. Liso´n Almagro1, I. Martı´nez Bravo1, D. Ruiz Macia´1, F. Caiguelas Villa1 1 Hospital de la Vega Lorenzo Guirao, Department of Orthopedics, Cieza (Murcia), Spain Objectives: High tibial osteotomy is frequently used for the surgical treatment of patients with isolated medial knee osteoarthritis associated with varus deformity. The classic rule (Coventry, Bauer) of which every millimeter of the basis of the lateral closed-wedge is equivalent to 1 of correction is only true when the distance between the most cephalic points of the wedge is 56.712819 mm. The aim of this study was confirming or rejecting the hypothesis of which the distance of interest is significantly equal to 56.7 mm. Methods: We have done a prospective blind study. Four specialists in Orthopaedic Surgery have measured 30 full leg standing radiographs (60 knees) of random assignment. We have measured with conventional transparent metric ruler with millimeter markings, the distance between the most cephalic points of the wedge that is planned for a lateral closed-wedge osteotomy. In addition the measurement of the distance has been realized to 10, 15, 20 and 25 mm from the joint line. Statistical analysis was performed using SPSS for Windows statistical package (version 15; SPSS, Chicago, IL, USA). Kolmogorov–Smirnov tests revealed that ratio scale data consistently displayed a normal distribution, and thus, the one-sample t test (with test value 56.712819) and the Pearson correlation coefficient, as measure of linear association between two variables, were used for comparative analysis. Statistical significance was set at p B 0.05. Results: Age (year): 55.53 ± 18.4. Gender (M/F): 9/21. Weight (kg): 74.7 ± 14.89. Height (cm): 159.9 ± 8.31. BMI: 29.12 ± 5.27. Distance between the most cephalic points of the wedge (in millimeters; the values are given as the mean and the standard deviation): Observer 1: 71.25 ± 6.32. Observer 2: 70.71 ± 6.45. Observer 3: 64.24 ± 7.61. Observer 4: 68.96 ± 6.26. 10 mm from the joint line: 73.99 ± 5.77. 15 mm from the joint line: 71.46 ± 6.4. 20 mm from the joint line: 67.57 ± 7.32. 25 mm from the joint line: 61.12 ± 8.25. The average of the distance has differed significantly with the test value in all the variables, except in the measurement realized to 25 mm of the joint line in females. Significant correlation has existed between the different observers. Conclusions: The measure of the distance between the most cephalic points of the wedge in a sample of standard population is significantly greater than 56.7 mm. Careful preoperative planning is mandatory to

S297 avoid under correction or overcorrection, two factors responsible for early failure of the procedure.

P29-445 Femoral opening wedge osteotomy for valgus knees. A prospective study with 5 years follow-up A. Ekeland1, K. Nerhus2, S. Dimmen3, S. Heir1 1 Martina Hansens Hospital, Bærum Postterminal, Norway, 2Martina Hansens Hospital, Bærum, Norway, 3Lovisenberg Deaconal Hospital, Orthopedic Department, Oslo, Norway Objectives: The surgical technique for opening wedge osteotomies has been simplified by the use of the Puddu-plate. This study presents results after use of Puddu-plates on distal femoral osteotomies. Methods: Twenty-four distal femoral osteotomies have been performed with opening wedge technique due to unilateral knee osteoarthritis with valgus malalignment in the period 2000–2008. The osteo-arthritis was mainly due to previous meniscal extirpation. The mean age of the patients was 48 years (31–62 years), and 11 females and 13 males were operated. The patients had a mean preoperative tibiofemoral valgus angle of 12 (8–20). The osteotomy was fixed with a Puddu-plate where the tooth of the implant secured the planned angular correction. The osteotomy cleft was filled by autogenous pelvic bone. The mean follow-up time was 5 years (2–10 years). Results: The mean width of the osteotomy cleft was 8.8 mm (7.5–12.5 mm), and the mean angular correction measured on preand post-operative radiographs was 8.8 (6–16). The osteotomy cleft healed after a mean of 12 weeks (11–26 weeks). The knee injury and osteoarthritis outcome score (KOOS) increased significantly during the observation period (P \ 0.001). A score of 100 means no complains. For pain the mean preoperative score was 52 and the score at follow-up was 70. The corresponding scores for symptoms were 51 and 63, for activity of daily life (ADL) 67 and 81, for sport and recreation 19 and 40 and for quality of life 29 and 56 (P \ 0.001). About 85% of the improvement occurred during the first year with a small improvement the second year which was almost kept for those followed for 5 and 10 years. One patient had reduced flexion postoperatively due to intraarticular scar tissue, and one osteotomy healed in 15 antecurvation. No infections or osteosynthetic failures were observed. Four knees (16%) have been converted to a TKA mean 5 (4–8) years postoperatively. Conclusions: The results after opening wedge osteotomy using the Puddu-plate show 25–120% improvement according to KOOS during the first 2 years. The mean angular correction in degrees agreed well with the mean width of the tooth of the implant in mm.

P29-447 Tibial opening wedge osteotomy for varus knees. A prospective study with 5 years follow-up A. Ekeland1, K. Nerhus2, S. Dimmen3, S. Heir1 1 Martina Hansens Hospital, Bærum Postterminal, Norway, 2Martina Hansens Hospital, Bærum, Norway, 3Lovisenberg Deaconal Hospital, Orthopedic Department, Oslo, Norway Objectives: The surgical technique for opening wedge osteotomies has been simplified by the use of the Puddu-plate. This study presents results after use of Puddu-plates on proximal tibial osteotomies. Methods: Fifty-two proximal tibial osteotomies have been performed with opening wedge technique due to unilateral knee osteoarthritis with varus malalignment in the period 2000–2008. The knee osteoarthritis was mainly due to previous meniscal extirpation. Seventeen patients had an additional rupture of the anterior cruciate ligament. The mean age of the patients was 47 years (31–64 years), and 20

123

S298 females and 32 males were operated. The patients had a mean tibiofemoral varus angle of 1 (7 varus–3 valgus). The osteotomy was fixed with a Puddu-plate where the tooth of the implant secured the planned angular correction. The osteotomy cleft was filled by autogenous pelvic bone. The mean follow-up time was 5 years (2–10 years). Results: The mean width of the osteotomy cleft was 8.5 mm (5–12.5 mm), and the mean angular correction measured on pre- and postoperative radiographs was 8.0 (5–12). The osteotomy cleft healed after a mean of 12 weeks (7–18 weeks). The knee injury and osteoarthritis outcome score (KOOS) increased significantly during the observation period (P \ 0.001). A score of 100 means no complains. For pain the mean preoperative score was 51 and the score at follow-up was 85. The corresponding scores for symptoms were 54 and 82, for activity of daily life (ADL) 62 and 90, for sport and recreation 26 and 65 and for quality of life 30 and 71 (P \ 0.001). About 80% of the improvement occurred during the first year with a minor improvement the second year which was almost kept for those followed for 5 and 10 years. One lower leg thrombosis, two wound infections and one osteosynthetic failure were observed in the postoperative period. Four knees (8%) have been converted to a TKA mean 4 (2–7) years postoperatively. Conclusions: The results after opening wedge osteotomy using the Puddu-plate showed 40–140% improvement according to KOOS during the first 2 years. The mean angular correction in degrees agreed well with the mean width of the tooth of the implant in mm.

P29-555 Dome osteotomy with dome stabilizer: a patient friendly procedure P. Kodkani1 1 Arthroscopy & Sports Medicine Institute, Mumbai, India Objectives: Hight tibial osteotomies are very often avoided due to procedural drawbacks or patient compliance. In order to overcome these drawbacks, dome osteotomy was performed using a newly designed uniplanar bilateral fixator. Dome osteotomy has a number of advantages over the other osteotomies. The indigenous fixator is compact, unplanar bilateral fixator, permits postoperative alterations in correction, immediate postoperative mobilization with return to activities of daily living without pain by 2 weeks and is economical. Methods: 64 osteoarthritic knees with medial compartment osteoarthrosois associated with genu varum were treated by dome osteotomy using this fixator in past 8 years. The age group varied from 30 to 72 years. Average followup is for 5 years. The dome osteotomy was performed under fluoroscopic guidance. Range of motion exercises could be started by second postoperative day. Partial weight bearing could be started by the 3rd day followed by full weight bearing depending on the patients comfort levels. Weight bearing scannograms were assessed and if any alteration in correction was required it was done using the fixator mechanism to achieve precision in first 10 days. X-rays were assessed at 1 monthly intervals. Once union was confirmed, the fixator was loosened and was removed after 3 days if there were no fresh complaints. Results: 2 cases had infection in immediate postoperative period. The fixator was removed, infection treated and the osteotomy bone grafted to achieve union. 10 cases had superficial pin tract infections. 2 cases with diabetes and osteoporosis showed pin migration and delayed union. 8 cases required alteration of correction in postoperative period. Cases which were not complicated could resume their activities of daily living by 2 weeks postoperatively.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 On an average the preoperative varus was 191.1 (HKA), with a maximum of 204 HKA (Hip-Knee-Ankle) angle. An average of 173.1 (HKA) valgus was achieved, with a valgus alignment achieved in all the knees. All the osteotomies united within 6–8 weeks. Due to the over correction, lower ‘Knee Score’ were achieved in resulting in ‘good’ result in 4 patients and ‘excellent’ in the rest. The Knee Function Score however was ‘excellent’ in all except for those complicated by infection. The Knee Society Score of all patients was ‘good’ to ‘excellent’ by the end of 2 months, which was maintained in their further follow ups. Conclusions: This new fixator gives all the advantages of an external fixator permitting early joint mobilization, early full weight bearing and early return to activities of daily living. It has better social acceptability and patient compliance. The fixators postoperative management is much simpler for the surgeon & the patients to maintain. Being economical also adds to its significance. Combining the advantages provided by a dome osteotomy with the added advantage of this new fixator makes the treatment very practical with better patient compliance.

P29-589 Medial opening wedge high tibial osteotomy (OWHTO) for medial compartment overload/arthritis in the varus knee R. Rossi1, F. Dettoni2, F. Castoldi3, D. Blonna3, G. Sito1, D.E. Bonasia4 1 University of Torino, Mauriziano ‘Umberto I’ Hospital, Torino, Italy, 2University of Torino, Mauriziano ‘Umberto I’ Hospital, Department of Orthopaedics and Traumatology, Torino, Italy, 3 Mauriziano Hospital, University of Torino, Department of Orthopaedics and Traumatology, Torino, Italy, 4University of Torino, Torino, Italy Objectives: The goal of this prospective study regarding OWHTO was to evaluate the midterm results, survivorship analysis, and the variables associated with the outcome. Methods: From Jan 2001 to Dec 2009, 141 consecutive OWHTOs with the Puddu plate were performed, planning a 5–7 of valgus overcorrection. Only OWHTOs for symptomatic medial knee overload/arthritis were included in the study. The patients were evaluated with: 1. Knee Society Score, 2. Womac scale, 3. another self evaluation scale (including VAS for pain in the last week, a scoring of the knee and surgery from 0 to 10, and a last question where the patient was asked if he/she would undergo the surgery again), 4. AP, lateral, Merchant and long leg X-rays. Numerous variables were investigated to find an association with the outcome, including: pre-operative (age, sex, BMI, amount of varus, ROM, Knee/Function Score, Womac Score, and arthrosis in the medial, lateral and patellofemoral compartments), intra-operative (combined arthroscopic procedures, size of the plate, type of osteotomy gap augmentation), and post-operative (ROM, alignment, residual medial laxity) variables. A paired t test was used to compare Knee Society and Womac scores before and after surgery. Pearson’s Chi-squared test was used for categorical variables, while simple logistic regression was performed for nominal variables. All statistically significant variables were then inserted in a multiple regression model. P \ 0.05. A Kaplan–Meier survival analysis was performed. Results: 39 patients were lost to follow-up or excluded from the study, leaving 99 knees (84 patients) for the present study. The mean age of patients at the time of surgery was 54.5 years (SD 9.2 years). The

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 mean follow up was 51.5 months (SD 23.8 months). Pre-operatively on average the Knee Society Score (Knee Score + Function Score) was 135.6 (SD 33.9), the Womac score was 50.7 (SD 20.8), and the limb alignment was 7.62 of varus (SD 3.33). Post-operatively on average the Knee Society Score was 160.5 (SD 26.3), the Womac score was 76.1 (SD 18.5), and the limb alignment was 0.1 of valgus (SD 2.8). The Knee Society and Womac Scores significantly improved after surgery (p \ 0.001). The variables significantly related to a poor outcome were: (1) preoperative BMI [ 30 (p = 0.038), (2) age [ 56 years (p = 0.008), (3) pre-operative knee flexion \ 120 (p = 0.033), (4) post-operative knee flexion \ 120 (p \ 0.001). The variables significantly related to a good outcome were: (1) Ahlba¨ck grade ‘‘0’’ arthritis of the medial compartment (p \ 0.001), (2) pre-operative excellent Knee Score (p \ 0.001). 6 patients underwent a total knee replacement (TKR) or were indicated for TKR during the follow-up period. The Kaplan–Meier analysis showed a 98.7% survival rate at 5 years of follow-up, and 75.88% at 7.5 years. Conclusions: The outcomes reported in the present study are similar to other studies, although the variables significantly associated with the outcome slightly differ from the existing literature.

P29-649 The use of toothed plate in supracondylar femoral osteotomy to correct genu valgum M.M. Abdel-Hamid1 1 Assiut University, Orthopedics and Traumatology, Faculty of medicine, Assiut, Egypt Objectives: Traditionally distal femoral varus osteotomy has been used for redistribution of articular surface load in young patients who have genu valgum deformity. Many methods of fixation of the osteotomy have been described. The toothed plate has been used in medial wedge opening high tibial osteotomy for genu varum for treatment of osteoarthritis. This plate was first presented in the Annual meeting of the American Academy of Orthopedic surgeons in 2009. This is the first time that the toothed plate was used for fixation of distal femoral varus osteotomy in cases of genu valgum. The hypothesis was that using the toothed plate will facilitate the technique, and the surgery will be less invasive due to the low profile of the plate. In this work the early results of using this plate was studied. Methods: Distal femoral lateral wedge opening osteotomy was done in 16 knees for 9 patients (7 had bilateral genu valgum deformity). Six patients were females and three were males. The age of patients ranged between 14 and 22 years (mean 18 years). All the patients had Genu Valgum deformity which ranged between 16 and 22 (average 18 valgus). Lateral distal incomplete open wedge femoral osteotomy was done under control X-rays. The toothed plate was used for fixation of the osteotomy in all the cases after bone graft insertion. The patients were followed up clinically and radiologicaly for a period ranged from 6 to 18 months. Results: The osteotomy healed in all cases in a period of 6–12 weeks average 9 weeks. The mean tibio-femoral angle was 18.6 ± 2.77 preoperatively, and the mean correction angle was 10.5 ± 2.66 immediately postoperatively. None of the cases lost any correction at the final follow up measurements. Two cases had medialization and flexion of the distal femoral fragment due to fracture of the medial cortex at the time of osteotomy. That was corrected in these cases by revision surgery after three weeks. Clinically full range of motion was regained in 15 knees (93.75%) by 4–8 weeks. One case (6.25%) had fixed extension lag of 10, and one case 5.

S299 Conclusions: Given the findings of solid union in all cases with no loss of correction it can be an effective method of fixation with less invasive and less demanding surgical technique.

P29-758 Simultaneous anterior cruciate ligament (ACL) reconstruction and medial opening wedge high tibial osteotomy (OWHTO) G. Cox1, N.P. Thomas2, A. Wilson1 1 North Hampshire Hospital, Department of Orthopaedics, Basingstoke, United Kingdom, 2Wessex Nuffield Hospital, The Wessex Knee Unit, Eastleigh, United Kingdom Objectives: The management of chronic anterior cruciate ligament (ACL) deficiency in combination with symptomatic medial compartment osteoarthritis is challenging, especially in the younger patient. This combination of instability and medial knee pain frequently requires multiple operations, each requiring long rehabilitation periods. When the alignment is unfavourable, in the coronal or sagittal plane, a combined ACL reconstruction and tibial/ femoral osteotomy can be performed as a single procedure. Methods: Simultaneous arthroscopic ACL reconstruction (hamstrings graft) and medial opening wedge high tibial osteotomy (OWHTO) using Tomofix plate fixation (Synthes) was performed in 10 male patients with a mean age of 38.8 years (27–46). The mean tourniquet time was 94 min (81–118) at 300 mmHg. In one patient the osteotomy was grafted with HATriC bone substitute (Arthrex) in the remaining 9 patients the osteotomy site was not grafted. Patients were mobilised fully-weight-bearing and underwent a standard ACL rehabilitation programme. They were scored in a dedicated physiotherapy clinic, prior and after surgery using the KOOS and Oxford knee scores. Mean follow up was 71 (30–135) months. Results: Mechanical axis angle was corrected from a mean of 4.7 varus (0.3–10.1) to 2.3 valgus (-0.1–4.3) with adjustment of weight bearing axis from 20% (0–38) to 53.6% (50.3–59) of the tibial plateau. Tibial slope pre-operatively was -8 (0 to -12) and 7.6 (-2 to -13) post-operatively. Pre-operative Oxford and KOOS scores were 33 (27–44) and 60.3 (48–76) respectively, with post-operative scores being 43.7 (42–48) and 78.3 (67.7–93.8). No complications were identified by time of final follow-up. Conclusions: Combining realignment osteotomy with ACL reconstruction provides technical challenges but is a surgical procedure that is safe, efficacious and avoids multiple procedures and periods of rehabilitation.

P29-759 Osteotomy around the knee and its effect on joint line obliquity at the ankle: a radiographic study G. Cox1, E. Robin1, N.P. Thomas2, A. Wilson1 1 North Hampshire Hospital, Department of Orthopaedics, Basingstoke, United Kingdom, 2Wessex Nuffield Hospital, The Wessex Knee Unit, Eastleigh, United Kingdom Objectives: When planning realignment osteotomy of the knee it has been previously noted that in addition to correcting deformity, in coronal and sagittal planes, attention must be made to an osteotomy’s effect on joint line obliquity. Whilst historically this has concentrated on the tibio-femoral joint, we sought to investigate the changes the surgery around the knee had on joint line obliquity at the ankle. Methods: A retrospective analysis of 145 knees (138 patients, 91 males) undergoing single bone, realignment osteotomy (134 tibia, 11 femur) for the treatment of osteoarthritis. There were 137 osteotomies to correct a varus deformity, including 128 medial opening (128 tibial, 1 femoral) and 7 lateral closing (3 tibial, 4 femoral) osteotomies. There were 9 closing medial osteotomies (3 tibial, 6 femoral) to correct valgus deformities. Standardised pre/post-operative long-leg

123

S300 alignment films were examined and the tibial plafond angle compared with the horizontal using PACS software (Centricity GE). Results: Joint line obliquity at the ankle, for patients with a varus alignment at knee, showed a mean valgus deformity of 0.41 (SD 4.89) pre-operatively and mean valgus deformity of 5.79 (SD 5.86) (p \ 0.02) post-operatively (valgus change of 6.72 (SD 5.09)). Patients with a valgus deformity at the knee, had a mean joint line obliquity of 8.14 (SD 4.85) valgus pre-operatively and 2 (SD 5.32) valgus post-operatively (p \ 0.0001), giving a mean varus change of 4.56 (SD 5.00). The relationship between osteotomy angle and degree of change in joint line obliquity did not reach significance when investigated using linear regression analysis for either bone/type of osteotomy. Conclusions: Realignment osteotomy around the knee changes joint line obliquity at the ankle, with valgising and varising knee osteotomies producing valgus and varus effects respectively. The clinical implications of this observation are as yet unknown, although we note a recent cadaveric study has shown that varus and valgus deformity of the distal tibia caused significant changes in the contact area of the tibiotalar joint and one patient in our series (tibiotalar valgus alignment of 12) developed post-operative ankle pain. We recommend that when planning a realignment osteotomy that full deformity analysis is carried out which also includes ankle joint obliquity.

P29-819 The weight-bearing scanogram technique provides better coronal limb alignment than the navigation technique in open high tibial osteotomy J.I. Kim1, K. Nha2, J.H. Kwon1, H.D. Lee2, B.H. Jang1 1 Inje University, Ilsanpaik Hospital, Koyang, Republic of Korea, 2 Inje University, Ilsanpaik Hospital, Koyangsi, Republic of Korea Objectives: Successful outcomes following high tibial osteotomy (HTO) require precise realignment of the mechanical axis of the lower extremity. We hypothesized that weight bearing limb scanogram (WBS) technique may show more accurate mechanical axis realignment than nonweight-bearing computer assisted (CAS) technique in OHTO. The purpose of this study was to compare coronal limb alignment using computer-assisted versus weight-bearing scanogram (WBS) technique in open high tibial osteotomy (OHTO). Methods: Forty OHTOs performed using the WBS technique were prospectively compared with 40 OHTOs performed using the CAS technique for postoperative coronal limb alignment using the weightbearing line (WBL) ratio on full length standing hip-to ankle radiographs. Results: The mean postoperative WBL ratio in the WBS group was significantly greater (p = 0.001) compared to the CAS group implying that the mean WBL ratio in WBS group was significantly closer to the postoperative WBL ratio target of 62% when compared to the CAS group. In the CAS group 10 limbs (25%) were undercorrected (i.e. with a postoperative WBL of B50%) compared to none in the WBS group. Conclusions: Weight bearing limb scanography (WBS) technqiue may be more reliable and accurate for restoration of coronal leg alignment in OHTO. Navigation techniques for HTO may need to factor in weight-bearing to minimize deviations of limb alignment after OHTO.

P29-959 Outcome of distal femoral osteotomy: a surgeon’s experience! A. Desai1, M.J. Dawson2 1 Cumberland Infirmary, Orthopaedics, Edinburgh, United Kingdom, 2 Cumberland Infirmary, Orthopaedics, Carlisle, United Kingdom

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Distal femoral varus osteotomy is performed for the treatment of lateral compartment osteoarthritis of the knee as well as for correction of the associated valgus deformity. It is also indicated in cases of complex instability in less than 60 years of age. The aim of our prospective study was to evaluate the outcome of the Distal femoral osteotomy performed for such reasons. Methods: Twenty-one patients underwent distal femoral osteotomy between 2006 and 2010. Of the 21 knees, initial nine cases were of lateral open wedge and the rest twelve underwent medial close wedge Biplanar osteotomy. In initial seven cases Puddu plate was used and in the rest Tomofix plate was used for fixtaion. Bone grafting was used in 10 cases. Radiological outcome was assessed by measuring the tibio-femoral and congruence angles. Functional outcome was assessed pre and post operatively by Oxford knee score, Lysholm score and Kujala score. Results: The average of the patients was 39.5 (18–57). Of the 21 knees, there were 11 male and 12 female patients. Seventeen patients underwent arthroscopy, five had associated microfracture treatment. Associated procedures like MPFL, ACL reconstruction and meniscal resection were carried out in 15 patients along with osteotomy. The average follow up was 24 months. Complications included one case of non-union and 2 cases of cellulitis. There were no cases of deep infection. Radiological evaluation showed statistically significant improvement in the correction of both tibio femoral angle and overall long leg alignment (p \ 0.01). Functional evaluation by the OKS, kujala and Lysholm scores showed statistically significant improvement (p \ 0.01). Conclusions: We conclude Distal femoral osteotomy is a safe and effective procedure for gonarthrosis associated with valgus deformity in both young and active older patients. Medial close wedge biplanar osteotomy with a rigid Tomofix plate provided immediate stability, satisfactory healing of the osteotomy site without the need for bone graft or bone substitutes.

P29-1229 Accelerated bone healing in rabbit tibia using a complex of b-tricalcium phosphate and hepatocyte growth factor K. Goshima1, J. Nakase1, K. Matsumoto2, H. Tsuchiya1 1 Kanazawa University, Orthopaedic surgery, Kanazawa, Japan, 2 Kanazawa University, Cancer Research Institute, Kanazawa, Japan Objectives: Medial opening high tibial osteotomy (HTO) has become popular for treatment of knee osteoarthritis. b-tricalcium phosphate (b-TCP) is widely used to fill the opened defect in the medial opening procedure because of its excellent biocompatibility and osteoconductive properties. However, b-TCP is in adequate mechanically for weight-bearing sites until bone incorporation occur. Hepatocyte growth factor (HGF) was reported to induce angiogenic, morphogenic, and antiapoptotic activity in various tissues. We hypothesized HGF combined with b-TCP and collagen matrix would promote bone healing and increase biomechanical strength of bone defect. This study was performed to evaluate the effect of HGF in a complex of btricalcium phosphate (b-TCP) and collagen in a rabbit osteotomy model. Methods: Segmental bone defects of 5 mm in length were created in the middle of the tibial shafts of rabbits. The defect was stabilized with external fixators and implanted with a complex of b-TCP granules and collagen, with or without 100 lg of recombinant human HGF. Biweekly, bone regeneration and b-TCP resorption were assessed radiographically and histologically. At 4 and 8 weeks, bone regeneration was evaluated by micro-computed tomography and mechanical tests. Results: Compared to the bone tissue treated with b-TCP and collagen, mineralization, angiogenesis, new bone formation, and absorption of b-TCP were promoted 4 weeks postoperatively by

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 treatment with HGF in the b-TCP and collagen group. These changes were associated with promoting biomechanical regeneration. By 8 weeks, the formation of bone marrow in newly generated bone and the absorption of the b-TCP granules were completed in a shorter period by combining HGF with b-TCP and collagen, compared to tissues without HGF. Conclusions: The combined application of HGF in a b-TCP and collagen matrix promoted histological bone healing and augmented mechanical strength of the healing bone, particularly at the early stages. The combined use of HGF and b-TCP could shorten the rehabilitation time for medial opening HTO.

P29-1244 Wedge volume and osteotomy surface depend on surgical technique for high tibial osteotomy D. Pape1, K. Dueck2, M. Haag2, O. Lorbach3, R. Seil4, H. Madry5 1 Orthopa¨dische Klinik, Olympic Medical Center, Centre Hospitalier, Luxemburg, Luxembourg, 2Centre Hospitalier de Luxembourg, Luxembourg, Luxembourg, 3University Hospital Saarland, Homburg/ Saar, Germany, 4Centre Hospitalier Luxemburg, Clinique d’Eich, Olympic Medical Center, Orthopedic Department, Luxemburg, Luxembourg, 5Universita¨tsklinikum des Saarlandes, Klinik fu¨r Orthopa¨die und Orthopa¨dische Chirurgie, Homburg, Germany Objectives: Biplanar open wedge high tibial osteotomy (HTO) is thought to promote rapid bone healing due to the increased cancellous bone surface compared to other HTO techniques. However, precise data on the bone surface area and wedge volume resulting from both open and closed wedge HTO techniques remain unknown. We hypothesized that biplanar, rather than uniplanar HTO better reflects the ideal geometrical requirements for bone healing: a large cancellous bone surface combined with a small wedge volume. Methods: Tibial saw bones were assigned to the following 4 different groups of valgisation high tibial osteotomies: group 1: open wedge uniplanar HTO; group 2: open wedge biplanar HTO with ascending frontal cut; group 3: open wedge biplanar HTO with descending frontal cut (retrotubercule osteotomy technique) and group 4: closed wedge uniplanar HTO. Bone surface areas of all osteotomy planes were quantified. Wedge volumes were calculated prismbased on standardized wedge heights of 5, 10 and 15 mm. Results: The open-wedge biplanar osteotomy with a descending frontal cut (group 3) created significantly larger bone surfaces compared to the ‘‘classic’’ biplanar technique with an ascending frontal cut (group 2) and compared to all uniplanar techniques. Bone surfaces after the classic open wedge technique (group 2) were slightly larger compared to all uniplanar techniques (group 1 and 4). No significant differences of wedge volumes were found between the retrotubercle (group 3) and classic open wedge techniques (group 2). Wedge volumes were significantly higher in the uniplanar open wedge technique (group 1) compared to the biplanar open wedge techniques (group 2 and 3). Conclusions: Bone geometry following HTO suggests that the biplanar open wedge techniques simultaneously create smaller wedge volumes and larger bone surface areas compared to the uniplanar open wedge techniques. The recently neglected closed wedge technique still offers the best theoretical healing potential with an almost absent wedge volume and a high amount of solid bone-tobone contact area. Although this idealized geometric view on bony geometry excludes all biologic factors that influence bone healing, the current data suggest a general rule for the applied standard osteotomy techniques and all of their surgical modifications: reducing the amount of slow gap healing and simultaneously increasing the area of faster contact healing will be beneficial for osteotomy healing.

S301 P29-1411 Clinical and radiological results of medial gonarthrosis treated with closed wedge high tibial osteotmy N.N. Stegemann1, F. Erdogan2, A. Can1, I.A. Sankaya2 1 Istanbul Cerrahi Hastanesi, Istanbul, Turkey, 2Istanbul Cerrahpasa University, Istanbul, Turkey Objectives: Arthroplasty has become an increasingly popular method for the treatment of gonarthtrosis and frequency of high tibial osteotomy decreased. In this study clinical and radiographic midterm results of high tibial osteotomy for unicompartment osteoarthtrosis treatment were retrospectively evaluated. Methods: 38 patients who underwent lateral closing wedge valgus osteotomy for primary degenerative gonarthtrosis (33 female, 5 male) were included in the study. The mean age at the time of surgery was 52 years (range 37–74 years). The patients were assessed clinically with range of motion and instability. Lysholm Knee Scoring Scale, Hospital for Special Surgery (HSS) knee score, Hokkaido knee score were used for clinical and anterior-posterior radiographs of lower extremity were used for radiographic evaluation Mean follow-up was 5.5 years (range 3–10 years). Results: Postoperative mechanical axis improved significantly with regard to preoperative period (p \ 0.05). HSS knee score improved significantly in the postoperative period. We were unable to show a relationship between postoperative mechanical axis or knee scores and BMI [ 25, age [ 50 at the time of surgery, gender, follow-up period of 5 years. 26 (68.4%) patients were highly satisfied with their surgical results but 11 (28.9%) were not satisfied, 1 (2.6%) patient was complaining about the results. Conclusions: Successful clinical and radiographic results were obtained with lateral closed wedge osteotomy in the treatment of unicompartmental osteoarthritis caused by improper alignment. Obesity, age at surgery, gender and severity of deformity did not affect the success of treatment.

Knee - total joint replacement I P30-21 Contribution of femoral & tibial component position to the overall coronal limb alignment in conventional instrumentation total knee arthroplasty: a comparison with computer assisted total knee arthroplasty K.Y. Chan1, S.S. Sathappan2, Y.H. Teo2 1 Penang General Hospital, Orthopaedics & Traumatology, Penang, Malaysia, 2Tan Tock Seng Hospital, Orthopaedic Surgery, Singapore, Singapore Objectives: Long term survival of the total knee arthroplasty (TKA) is dependent on accurate restoration of the lower limb mechanical axis. This is dependent on accurate alignment of both the femoral and tibial component. The objective of this study is to assess the positioning of both components in conventional instrumentation (CI) TKA versus computer assisted (CAS) TKA and its contribution to the overall alignment. Methods: 52 CAS TKA and 51 CI TKA were retrospectively compared in terms of their postoperative alignment and individual femoral and tibial component positioning by evaluation of postoperative lower limb standing radiographs. Acceptable overall alignment is taken as within 3 valgus or varus to the lower limb mechanical axis. The femoral component alignment is taken as the angle of deviation (valgus or varus) from the ideal perpendicular to the femoral mechanical axis while the tibial component alignment is taken as the angle of deviation (valgus or varus) from perpendicular to the tibial mechanical axis. Results: The CAS group showed a significantly greater percentage (88.5%) of knees within the acceptable overall alignment compared to

123

S302 only 49.0% in the CI group. The mean deviation from the mechanical axis was also significantly less (1.58) in the CAS group compared to 3.28 in the CI group. The main contributor to overall malalignment in the CI group was the femoral component with the CI group showing a mean femoral component deviation of 2.32 which was significantly greater than the CAS group (1.25). The tibial component alignment deviation did not differ significantly between both groups. Conclusions: The main contributor to malalignment in CI TKA is the femoral component position. CAS results in more accurate positioning of the femoral component of the TKA and correspondingly better overall coronal alignment of the lower limb compared to CI. We believe that CAS should result in better long term survival of the TKA, but this will require long term outcome comparative studies with CI to prove.

P30-27 Extensor mechanism allograft reconstruction in TKA for ruptures or ankylosis A. Baldini1, L. Manfredini1, P. Summa1, F. Traverso2, G. Grappiolo2 1 IFCA, Ortopedia, Firenze, Italy, 2IRCCS Istituto Clinico Humanitas, Rozzano Milano, Italy Objectives: Extensor mechanism disruption in total knee arthroplasty (TKA) occurs infrequently but often requires surgical intervention. We compared two cohorts undergoing extensor mechanism allograft reconstruction, one group had an extensor mechanism rupture, and the other had a recurrent ankylosed knee. Methods: Twenty-four consecutive patients with extensor mechanism disruption or ankylosis after TKA were treated. Two different types of extensor mechanism allografts were used: total extensor mechanism, and Achilles tendon allograft. The former graft was chosen when the patellar bone was deficient or replaced and the latter when the patellar bone was present with a sufficient bone stock. Demographic factors, diagnosis at extensor failure, Knee Society clinical rating scores, radiographs, and patient satisfaction were recorded. In the ruptured group (15 patients) the average time from extensor mechanism disruption to surgery was 6.6 months (range, 1–24 months). In the ankylosis group (nine patients) two or more surgical failed interventions to gain motion were already performed, and mean preoperative flexion was 20 (range 0–45). Results: At a mean follow-up of 40 months (range, 6–70 months), all patients were community ambulators. None of the patients showed a postoperative extensor lag [5. Average postoperative maximum flexion was 97 (90–115) for the ruptured group, and 85 (55–95) for the ankylosed group. All patients thought their functional status had improved, and 87% were satisfied with the results of the allograft reconstruction. One patient in the ruptured group had allograft failure due to recurrent infection after re-revision for sepsis. One patient of the stiff group had recurrent stiffness (maximum flexion: 55) at follow up. Conclusions: The total extensor mechanism allograft and Achilles tendon allograft both were successful in the treatment of the failed extensor mechanism and showed promising results for the treatment of the ankylosed knee. Surgical technique and postoperative protocols should be tailored to the indication for the extensor mechanism reconstruction, and to the status of the quadriceps muscle belly.

P30-51 Behaviour of the synovial fluid cytokines after a total knee replacement in osteoarthritis D. Popescu1, L. Lozano2, O. Ares3, J.C. Martinez-Pastor4, J.M. Segur4, S. Suso5 1 Hospital Clinic i Provincial, Barcelona, Spain, 2Hospital Clinic de Barcelona, Knee Unit, ICEMEQ, Barcelona, Spain, 3Hospital de

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Viladecans, Viladecans, Spain, 4Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain, 5Hospital Clinic i Universitari de Barcelona, Knee Unit. Orthopaedic Surgery Department (ICEMEQ), Barcelona, Spain Objectives: To determine the behaviour of the synovial fluid parameters (IL-1b, TNF-a and IL-6) after a TKR. Methods: A prospective study was performed to evaluate the outcomes in thirty-two patients who had undergone sequential bilateral primary total knee replacement (TKR). 11 patients were excluded for the follow-up due to different adverse events: 2 acute infections, 2 instabilities, 2 stiffness’s, 1 severe health problems, 2 insatisfactions after the first TKR and 2 patients who were satisfied with their knee and didn’t want the contralateral TKR. The total number of patients for the final assessment was 23 (46 TKR). There were 8 men and 15 women; the mean age at the time of surgery was 65.3 years. The mean duration of follow-up was 1 year. Clinical, radiographic and synovial parameters evaluations were performed preoperatively and at 6 and 12 months after surgery. Two Scores were used for the clinical assessment: (KSS) for knee and function and WOMAC Score. The radiographic evaluation included the preoperative grade of osteoarthritis, alignment of the components, signs of loosening or other complication. The synovial fluid parameters were evaluated with Elisa. Results: We found a significant decrease in TNF-a and IL-1b values at 1-year after the surgery. Also we found a correlation between both preoperative and postoperative TNF-a and IL-1b values and their respective WOMAC scores. Conclusions: Cytokines like TNF-a and IL-1b seem to be implicated in joint deterioration in osteoarthritis, there values diminishing after TKR, with consequent clinical improvement.

P30-147 Floating platform total knee replacement. Prospective evaluation after 5–8 years J.-Y. Jenny1, R.K. Miehlke2, D. Saragaglia3 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France, 2The Rhine-Main Center for Joint Diseases, Wiesbaden, Germany, 3Grenoble South Teaching Hospital, Orthopaedic and Sport Traumatology, E´chirolles, France Objectives: Polyethylene wear is one of the reasons for failure of total knee replacement (TKR). There are several reasons for wear, and the femoro-tibial contact area is an important factor. Mobile bearing, highly congruent prostheses might be more resistant to polyethylene wear than fixed bearing, incongruent prostheses. We evaluated the 5to 8-year experience of three university departments by using an original system with following highlights: implantation with a navigation system, extended congruency up to 90 of flexion, floating polyethylene component with non-limited movements of rotation, antero-posterior translation and medio-lateral translation. Methods: 347 patients have been operated on in the three participating departments with this new prosthesis system between 2001 and 2004, and have been prospectively followed with clinical and radiologic examination with a minimal follow-up time of 5 years. There were 246 women and 101 men, with a mean age of 67 years. Clinical and functional results have been analyzed according to the Knee Society scoring system. Accuracy of implantation has been assessed on post-operative long leg antero-posterior and lateral X-rays. Survival rate up to 8 years has been calculated according to Kaplan and Meier, with mechanical revision or any revision as end-points. Results: Complete patient history was obtained by 319 cases (92%). The mean clinical score was 93 points. The mean pain score was 47 points. The mean flexion angle was 118. The mean functional score was 87 points. An optimal correction of the coronal femoro-tibial axis was obtained in 94% of the cases. Survival rate after 8 years was 98.8% for mechanical revisions and 95.5% for all revisions.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S303

Conclusions: We confirmed the influence of the navigation system on the accuracy of implantation. The clinical and functional results after 5–8 years are in line with the better results of the current literature after conventional implantation of non-congruent prostheses. 5- to 8-year survival rate is comparable to the current standards. The influence of the design on polyethylene wear will need a longer follow-up.

Results: A linear correlation between the angle between the lateral facet of the fibular head and tibial tuberosity as well as the tibial tuberosity and the geometric centre of the tibia was found (R = 0.73, p \ 0.001). Conclusions: The position of the fibular head determines femorotibial rotation and may be a helpful new landmark for establishing the rotation of the tibial component.

P30-160 Computer navigated versus conventional total knee arthroplasty. Retrospective cohort study on prosthesis alignment and functional outcome in 100 LCS rotating platform TKA at mid term follow-up J.J. Tolk1, H.W.J. Koot1, R.P.A. Janssen1 1 Maxima Medical Center, Orthopaedic Center Maxima, Eindhoven, The Netherlands Objectives: In total knee arthroplasty (TKA), reconstruction of a neutral mechanical axis is of great importance. The main goal of this study was to compare the accuracy of alignment after conventional versus computer assisted TKA. Additionally the effect of computer assisted surgery (CAS) on functional outcome was analyzed. Methods: Out of a consecutive series, 50 conventional TKA were compared with 50 computer assisted TKA. Except for the use of CAS, all peri- and postoperative interventions were comparable. Radiological outcome was analyzed on standardized standing long-leg radiographs. Functional outcome was assessed using the Oxford Knee Score (OKS) and the Knee Society Score (KSS). Results: No significant difference in mean mechanical axis alignment between the two groups was found. The number of outliers deviating more than 3 from the mechanical axis was significantly reduced by using CAS, with 50% outliers in the conventional group and 26% outliers in the CAS group (P = 0.023). At mid term follow-up, the OKS and KSS knee and function scores did not show statistical difference between the two groups. Conclusions: The present study showed a reduction of the number of outliers exceeding 3 from the tibiofemoral mechanical using CASTKA. This improvement in accuracy did not result in improvement of the functional outcome at mid term follow-up.

P30-386 Improvement of health-related quality of life after TKA. Best results after bilateral TKA in one time V.J. Leo´n Mun˜oz1, A.J. Liso´n Almagro1, C.H. Herna´ndez Garcı´a1 1 Hospital de la Vega Lorenzo Guirao, Department of Orthopedics, Cieza (Murcia), Spain Objectives: Total knee arthroplasty (TKA) is well accepted as reliable and suitable surgical procedure to return patients to function. Helped by the COOP/WONCA questionnaire, our aim is to determine if total knee replacement can improve the health-related quality of life (HRQOL) in a group of knee osteoarthritis (OA) affected patients. Methods: Design of the study: cross-sectional study. Data collected from January to June 2010. Studied population: coming from the Knee Surgery Unity of our Orthopedic Surgery Department. The variable HRQOL was measured through COOP/WONCA charts of quality of life (9 dimensions), culturally validated for Spanish people. Scores range from 1 to 5 for each dimension, with 1 representing the best and 5 indicating the worst level of functioning or well-being. Other variables included age, sex, body mass index (BMI), and number of comorbidities attending to the Elixhauser comorbidity index. Results: 163 patients (76.7% female and 23.3% male). Mean age: 68.56 ± 8.5 years. Mean BMI: 32.83 ± 5.01 kg/m2. Mean follow-up of the 91 TKA cases: 38.01 ± 27.19 months (ranging from 6 to 120 months). Mean number of comorbidities (excluding the obesity, since we have considered the BMI like another variable): 1.74 ± 1.15 comorbidities. Of the multiple diagnostic combinations, attending to the particular affectation of each knee, we have established 9 diagnostics groups. We have obtained a statistically significant difference in the comparison among BMI (female 33.53 ± 5.18 kg/m2 and male 30.52 ± 3.59 kg/m2) and age (female 69.91 ± 7.93 years and male 64.13 ± 8.94 years) (p = 0.000) attending to gender. Attending to the gender, no significant difference among global results of the COOP/WONCA questionnaire was obtained (female 28.16 ± 6.74 points and male 26.08 ± 7.02 points). In the comparison of the COOP/WONCA punctuation related to the different diagnostics groups, we have obtained significant differences in the patients with TKA with regard to the rest of groups. It stands out a mean punctuation of 22.07 ± 7.2 points in patients who had undergone bilateral TKA in one time for OA of both knees (group with better HRQOL). No significant difference was obtained when the patient presents one side with TKA and has been programmed for surgery of the second side (31.25 ± 8.14 points). We have obtained a statistically significant difference (p = 0.05) in the comparison among patients with knee OA that we have consider candidates to conservative management (27.69 ± 5.28 points) and those that we have consider candidates to surgery and have programmed for total knee replacement (32.56 ± 4.41 points). Conclusions: We can affirm that the COOP/WONCA questionnaire has discriminated the HRQOL in the population of patients studied. Patients with TKA presented better HRQOL than patients with knee OA. TKA achieves the aim to improve the health-related quality of life of the patient with remarkable results if bilateral surgery in one time is done.

P30-205 Femorotibial rotation is influenced by the position of the fibular head after implantation of a total knee prosthesis T. Pfitzner1, G. Matziolis1, C. Perka1 1 Charite´, University medicine Berlin, Orthopaedic Department, Berlin, Germany Objectives: A gold standard for the correct rotation of the tibial component has yet to be established in total knee arthroplasty. The target parameter of correct rotation is the facilitation of femorotibial rotation over the entire range of movement without an implant overhang. Although the origin of the lateral collateral ligament is a recognised landmark for determining the rotation of the femoral component (epicondylar axis), the attachment of the lateral collateral ligament has not been taken into consideration for adjusting tibial rotation up to now. Methods: Seventy patients who had undergone total knee arthroplasty were enrolled in this retrospective study. 6 months after surgery, a CT of the operated knee was performed in all cases and the position of the lateral facet of the fibular head, the tibial tuberosity, the geometric centre of the tibia and the femoral epicondyles were determined.

123

S304 P30-454 CT based patient-specific cutting blocks for total knee arthroplasty: technique and preliminary radiological results P.P. Koch1, D.A. Mu¨ller1, S.F. Fucentese1 1 University of Zurich, Orthopaedic Department, Zurich, Switzerland Objectives: Accuracy in component positioning for total knee arthroplasty (TKA) remains a major concern. Many studies show that computer-assisted surgery (CAS) improves the precision significantly compared with standard manual techniques. However, computer navigation has limitations such as investment costs, longer operation time and additional complication risks. In the last 2 years, the technology of polyamide laser sintering to create patient-specific orientation tools according to preoperative CT-or MRI-data has been emerging. We present our experience with the MyKnee technique (Medacta International SA), which combines the guidance and cutting block in one. Methods: Consecutive data of 70 patients were preoperatively sampled and prospectively analized for radiological precision of component implantation. With the CT scan a tridimensional bone model of the patient’s knee with its hip-knee-ankle axis is created. This bone modeling acts as the base creating the anatomical cutting blocks that can fit a patient’s knee morphology. The surgeon can plan his preferred landmarks and define the implant size, resection levels, femoral rotation and the amount of tibial posterior slope. After a standard surgical approach the cutting blocks are mounted to the tibial plateau and the distal femur, adapted to unambiguous bony landmarks. The cuts are performed directly through that block. Further surgical steps are following according standard techniques. Pre- and post-operative long-standing X-ray were available for analysis, planned and implanted component size were compared. Results: The postoperative mechanical axis were between 4.5 of valgus and 3.1 of varus, including 7 outliers (deviation [ 3 from neutral or planning, corresponding 10%). Posterior tibial slope varied between 0 and 10 (15 outliers, 21.4%), comparing the first 35 cases (25.7% outliers) with the second 35 cases (17.1% outliers) a clear improvement was found. Femoral component flexion ranged between 0.2 and 6.4 (8 outliers, 11.4%). Compared to the planning 7 out of 140 components (5%) had to be changed in the size, 6 on the tibial side and two on the femoral side. Conclusions: Compared to literature the radiological results of component implantation with patient matched instrumentation are comparable to the most precise technique of CAS TKA. The number of outliers are similar in all measured component positions, but more data will be needed to define significance. Our preliminary experience indicates that the MyKnee technology of CT-based patient-specific cutting blocks represents a reliable, efficient and precise technique.

P30-523 A new spacer-guided PCL balancing technique results into good kinematics of an anatomically designed cruciate-retaining total knee arthroplasty P. Heesterbeek1, L. Labey2, P. Wong2, B. Innocenti2, A. Wymenga3 1 Sint Maartenskliniek, RD&E, Nijmegen, The Netherlands, 2Smith & Nephew, European Centre for Knee Research, Leuven, Belgium, 3 Sint Maartenskliniek, Orthopedic Surgery, Knee Reconstruction Unit, Nijmegen, The Netherlands Objectives: After total knee arthroplasty (TKA) with a PCL-retaining implant the location of the tibiofemoral contact point (CP) should be restored in order to obtain normal kinematics. The difficulty during surgery is to control this location since the position of the femur on the tibia cannot easily be measured from the back of the joint. Therefore, we developed a simple ‘‘spacer technique’’ to check the CP indirectly in 90 flexion after all bone cuts are made by measuring the step-off between the distal cut of the femur and the anterior edge of

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 the tibia with a spacer in place. The goal of this experiment was to investigate whether this new PCL balancing approach with the spacer technique created the correct CP location. Methods: Nine fresh-frozen full leg cadaver specimens were used. After native testing, prototype components of a new PCL-retaining implant were implanted using navigation and a bone-referenced technique. After finishing the bone cuts of tibia and femur, the spacer was inserted in flexion and positioned on the anterior edge of the bony surface to measure the step-off. If necessary, an extra cut was made to correct the step-off and thus balance the PCL. The specimen was mounted on the knee kinematics rig and a squat with constant vertical ankle force (130 N) and constant medial and lateral hamstrings forces (50 N) was performed between 30 and 130 of knee flexion. The trajectories of the reflective tibial and femoral markers were continuously recorded using six infrared cameras. CP (native and implant) were calculated as the projections of the femoral condylar centers on the horizontal plane of the tibia. Results: Of the 9 specimens, the calculated step-off was correct in 7 after finishing the bone cuts and in 2 specimens an additional tibia cut with 2–3 more slope was sufficient to achieve the correct step-off. No lift-off of the tibial tray occurred during the tests. The patterns of the kinematics of the native and replaced knee showed quite good similarity. The medial CP of the knee implant is at the same position as the medial CP of the native knee. No paradoxical roll forward is seen in the knee implants, showing that the PCL balancing apparently seems to work quite well. The lateral CP of the knee has a similar kinematic pattern in flexion before and after TKA. The CP of the implant shows a slightly more anterior location near extension but this is only marginal. Conclusions: The kinematics of the PCL-retaining implant are on average comparable to the kinematic pattern of the native knee. Apparently, the joint surfaces of the anatomic knee designed with a dished medial insert surface and a convex lateral insert surface and a 3 varus of the joint line is guiding the motion towards that of a normal knee joint. We feel that correct balancing of the PCL during implantation is of major importance in achieving these results. The spacer technique to balance the PCL seems to work well in this experiment.

P30-576 A novel reference axis that indicates axial alignment of distal tibia in total knee arthroplasty H. Enomoto1, T. Nakamura2, H. Shimosawa1, Y. Niki1, Y. Toyama1, Y. Suda1 1 Keio University, Department of Orthopaedic Surgery, Tokyo, Japan, 2 Johnson & Johnson Japan, Tokyo, Japan Objectives: Appropriate alignment is essential for successful clinical outcome and the implant longevity after TKA. In proximal tibial osteotomy with an extra-medullary guide, we usually set the instrument with reference to tibial tuberosity proximally and 1st or 2nd metatarsus distally. However, especially in case of ankle deformity, we occasionally feel dilemma how to align the instrument distally, considering the metatarsus and/or trans-malleolar axis. Here we introduce a novel reference axis indicating axial rotation, which we defined specifically for this study. Methods: 3D-CAD models of 20 tibiae from OA patients (73.8 ± 6.9 y/o) were reconstructed from CT data using Mimics (Materialise). Tibial Coordinate System; Our system is mid-sagittal plane based algorithm defined by an apex of the tibial plafond, PCL enthesis, and tibial tubercle. The origin was projected midpoint of bilateral eminences on the sagittal plane. Then the Z (vertical) axis was defined as the line between the origin and the apex of the tibial plafond. The normal vector of the sagittal plane was assigned as the Y axis (ML).

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 The X axis (AP) was then determined as a cross product of the Z and the Y axis, contained within the sagittal plane. Trans-malleolar axis, and Plafond axis; The trans-malleolar axis is a conventional reference line that connects the tips of bilateral malleolus. The plafond axis was a line that connects both of midpoints of medial and lateral margins of the tibial plafond facet. Measurements of axial rotation of distal tibia; The axial alignment of distal tibia was examined by measuring the projected angle of the distal reference axes that we determined above relative to proximal ML axis in the 3D-coordinate system respectively. Results: In defining the trans-malleolar axis, interobserver errors among 3 surgeons were 3.14 ± 0.47 mm (medial) and 4.88 ± 0.59 mm (lateral). As for the Plafond axis, those were 1.97 ± 0.53 mm (medial) and 2.02 ± 0.44 mm (lateral) respectively, both of which were significantly lower compared to the trans-malleolar axis (p \ 0.05). ICC (2.1) of y-value to be pointed out as midial and lateral reference points for the Plafond axis was higher than that for the transmalleolar axis. The eventual axial alignments of distal tibial with reference to the plafond axis and to the trans-malleolar axis were 9.8 ± 8.4 (-6.0–26.2), and 16.3 ± 6.3 (-0.1–27.9) respectively. Conclusions: This is the first study to demonstrate that the Plafond axis is also a reproducible reference axis in the 3D-algorythm. In terms of interobserver errors and ICC (2.1), the Plafond axis was expected to be more accurate and more reproducible indicator of axial alignment compared to the trans-malleolar axis. We consider that the plafond axis is worth to be applied in preoperative planning as another reference axis in determining mid-sagittal plane distally more precisely.

P30-587 Conservative treatment of the spontaneous avascular necrosis of the knee in middle-agedpatients G.M. Marcheggiani Muccioli1, A. Grassi1, T. Bonanzinga1, M. Nitri1, S. Zaffagnini1, M. Marcacci1 1 Istituto Ortopedico Rizzoli, University of Bologna, Biomechanics Laboratory, Bologna, Italy Objectives: Pulsed electromagnetic field (PEMF) treatment is indicated in the early stages of osteonecrosis of the femoral head [1]. The aim of the present study is to investigate the effect of PEMF treatment in patients suffering from spontaneous avascular necrosis (AVN) of the knee. Methods: Twenty-eight pts (19 M, 9 F, mean age 49.8 ± 16.4 years, BMI 26.0 ± 3.9) suffering from spontaneous AVN of the knee were prospectively enrolled in this study. Inclusion criteria: presence of a sympthomatic (acute and progressive pain) spontaneous AVN of the knee (a non-traumatic Bone Marrow Lesion (BML) of this joint [2]) in a middle-aged or elderly patient, without ligamentous knee laxity. Exclusion criteria: previous knee surgery or presence of total hip replacement on affected or contra-lateral limb, varus or valgus knee deformity exceeding 10, BMI [ 30 kg/m2, infection, rheumathoid arthritis, autoimmune diseases, sistemic diseses, tumors, use of steroids, alchool or nicotine abuse. The patients were treated with PEMF (I-ONE therapy—IGEA: magnetic field 1.5 mT, frequency 75 Hz and duty-cycle of 10%) for at least 6 h daily for 1 month. The coil was placed over the knee, not in direct contact with the skin. They were clinically evaluated before and 6 months after treatment by means of: 100 mm VAS for pain, Knee Society Score (KSS). Tegner and EQ5D scales. MRI studies were also performed at the same times, measuring BMLs areas (by Osirix MD Software) and grading these lesions by means Whole-Organ Magnetic Resonance Imaging Score (WORMS) for marrow abnormality [3]. Patients were followed recording failures (number of patients undergoing unicompartmental or total knee artroplasty surgery) until 2-year follow-up.

S305 Results: Pain was significantly reduced after 6 months (mean VAS from 73.2 ± 20.7 to 29.6 ± 21.3) (P \ 0.0001). Clinical scores significantly improved at 6 months: mean KSS (from 34.0 ± 13.3 to 76.1 ± 15.9) (P \ 0.0001), median Tegner (time zero 1, range 1–1; follow-up 3, range 3–4) (P \ 0.0001) and mean EQ-5D (from 0.32 ± 0.33 to 0.74 ± 0.23) (P \ 0.0001). MRI evaluation showed a significant reduction of BMLs areas for femoral lesions, especially in the sagittal plane (P \ 0.005). This area reduction is strongly correlated to WORMS grading (P \ 0.005). Four patients required knee arthroplasty surgery (2 UKA and 2 TKA). PEMF therapy preserved 85.7% of treated knees. Conclusions: The results of this study confirm that PEMF can have a role in the treatment of early stages of spontaneous AVN of the knee. PEMF stimulation was able to either reduce knee pain and to increase knee function or delay the time until prosthetic surgery. However randomized controlled studies are needed on this topic. References: 1. Massari L et al.; JBJS Am 2006;88:56–60 2. Roemer FW et al.; Osteoarthritis and Cartilage. 2009;17:1115–31 3. Peterfy GC et al.; Osteoarthritis and Cartilage. 2004;12:177– 90

P30-609 Analysis of different stem lengths and fixation techniques in hinged total knee arthroplasty S. Fuchs-Winkelmann1, B.F. El-Zayat2, T.J. Heyse1, N. Fanciullacci3, C. Yan4, B. Innocenti5 1 Universita¨tsklinikum Gießen und Marburg, Standort Marburg, Klinik fu¨r Orthopa¨die und Rheumatologie, Marburg, Germany, 2 Universita¨tsklinikum Giessen und Marburg, Standort Marburg, Klinik fu¨r Orthopa¨die und Rheumatologie, Marburg, Germany, 3 Department of Mechanics and Industrial Technology, Florence, Italy, 4 European Centre for Knee Research, Leuven, Belgium, 5Smith & Nephew, European Centre for Knee Research, Leuven, Belgium Objectives: To improve component stability in hinged Total Knee Arthroplasty (TKA), different stem lengths could be included during surgery. The stem, which can be either cemented or uncemented, transfers stress from the damaged proximal bone surface to the distal cortical bone. However, no evidence-based guidelines are available to help surgeons decide on the length of stem and whether to cement it or not. A numerical model was developed to compare different fixation techniques and stem lengths in a hinged TKA during a lunge and a squat. Methods: A physiological 3D tibia model was created from Computed Tomography images of a left mechanical-equivalent Sawbone tibia. A hinged TKA (RT-PLUS, Smith&Nephew, Memphis, TN) was selected for the study. Four different configurations were considered: a short cementless stem, a long cementless stem, a short cemented stem and a long cemented stem. The short and long stems had a length of 95 and 160 mm respectively. Stem lengths and sizes were selected based on experimental tests. Loading conditions for squat and lunge motions were calculated using a validated musculoskeletal model and were applied in the finite element model. For all the movements and configurations stresses in selected regions of interest and micromotions between the implant and the bone were computed and compared. Results: Figure 1 illustrates the distribution of the average compressive stress for the four analyzed configurations in the squat movements as a function of the distance from the tibial cut. For both movements, the most stressed regions were situated around the stem tips and the presence of cement reduced the stresses along the bonecement interface compared to the cementless configuration.

123

S306

Fig. 3 Average principal stress (compressive) in the 20 region of interest analysed during the lunge. The two vertical black lines in the figure represent the region of interest corresponding to the short stem tip (dotted line) and to the long stem tip (continuous line)

The maximal average compressive stress was higher for the cementless long stem configuration (squat 18.2 MPa, lunge 17.7 MPa) and lower for the cemented long stem configurations (squat 11.5 MPa, lunge 10.1 MPa). Also, for the short stem, the cementless configuration showed a higher average compressive stress (squat 13.3 MPa, lunge 14.6 MPa) compared to the cemented configuration (squat 9.5 MPa, lunge 10.4 MPa) in the region situated around the stem tips. However, cemented and cementless short stems showed similar maximal stresses in a region below the stem tip. Cementless stems show higher micromotions compared to cemented stems (*50%). Long cemented stems result in lower micromotions (*50 lm) compared to short cementless stems (*120 lm). Conclusions: The presence of cemented stem induces lower stresses in the tibial bone-stem interface and lower micromotions between implant and bone compared to cementless stem. A short stem shows similar maximal stresses in a region below the stem tip.

P30-676 Hip abductors strength: reliability and association with physical function after unilateral total knee arthroplasty A.H. Alnahdi1, J. Zeni2, L. Snyder-Mackler2 1 University of Delaware, Biomechanics and Movement Science Program, Newark, United States, 2University of Delaware, Department of Physical Therapy, Newark, United States Objectives: The aims of this study were to 1. examine the test–retest reliability of using a hand-held dynamometer to measure strength of the hip abductor muscles and 2. examine the relationship between hip abductors strength and physical function in patients with unilateral Total Knee Arthroplasty (TKA). Methods: Patients with primary unilateral TKA for knee osteoarthritis were recruited for this study. The maximum isometric strength of the hip abductors was tested in a side-lying position using a handheld dynamometer (Lafayette Manual Muscle Test System Model 01163, Lafayette Instrument, Lafayette, IN) and a stabilization strap. Patients also underwent an isometric strength test for the quadriceps femoris muscle using an isokinetic dynamometer (Kincom, Chattecx Corp, Harrison, TN). Physical function was assessed using three performance-based measures, Times Up and Go Test (TUG), Stair Climbing Test (SCT), and six-Minute Walk Test (6 MW). Intra-rater test–retest

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 reliability of the hip abductor muscle strength was assessed using intraclass correlation coefficient model 3.1. Standard Error of Measurement (SEM) and Minimal Detectable Change at 90% confidence interval (MDC90) for hip abductor strength were also computed. The relationship between hip abductors strength and physical function were assessed using a hierarchical linear regression. We ran three separate regression models to examine predictors of performance in TUG, SCT and 6 MW separately. We entered age and BMI as the first step. Quadriceps strength on the operated side was entered as the second step, followed by quadriceps strength on the nonoperated side and finally strength of the hip abductors on the operated side. Statistical analyses were performed using IBM SPSS 19. Results: 160 subjects (54% women) participated in the study with age (68.1 ± 7.7 year), BMI (31.3 ± 5.9), time from surgery (12.8 ± 8 months). A subset of 15 subjects were used for test-rest study of the hip abductors which yielded an ICC (3.1) of 0.91 (95% CI: .83–.95), SEM = 0.76 N/BMI, and MDC90 = 1.8 N/BMI. After accounting for age and BMI, quadriceps strength on both limbs accounted for 12.5% (operated side) and 3% (nonoperated side) of variance in values for the TUG and 6 MW. Adding the strength of the hip abductors did not significantly improve the model. After accounting for age and BMI, quadriceps strength on both limbs accounted for 13% (operated side) and 2% (nonoperated side) of variance in the values for the SCT. Adding the strength of the hip abductors explained an additional 3.3% of the variance (p = 0.003). Conclusions: Testing the strength of the hip abductors in patients with TKA using the method described has excellent reliability. While this study provides evidence that the strength of the hip abductors muscles can be reliably measured, no contribution to ADLS other than a small contribution to stair climbing was demonstrated. The value of strengthening exercises for the hip abductors in rehabilitation programs for patients after TKA needs further study.

P30-703 Computer assisted UKR: a prospective randomized study at a short term follow-up N. Confalonieri1, A. Manzotti1 1 CTO Hospital, 1st Orthop Dept, Milan, Italy Objectives: Despite clear clinical advantages Unicompartimetal knee replacement still remain an high demanding surgical procedure. Different Authors in literature pointed out how coronal tibial malalignment beyond 3 as well as tibial slope beyond 7 increase the rate of aseptic failure. Likewise overcorrection in the coronal plain is a well recognized cause of failure. Furthermore it has been shown how even with short narrow intramedullary guide this can causes errors in both coronal planes. Computer assisted surgery has been proposed to improve implant positioning in joint replacement surgery with no need of intramedullary guide. Aim of the study is to present a prospective randomized study comparing 2 groups of UKR s using either a computer assisted technique or a free-hand technique. Methods: Since January 2010 44 patients undergoing UKR with the same implant have been enrolled in the study prospectively. Patients were alternatively assigned to either the traditional or computerassisted alignment group. In the group A (22 knees) the implant was positioned using a CT-free computer assisted alignment system specifically created for UKR surgery while in group B (22 knees) a traditional technique based on an extramedullary tibial guide was adopted. The duration of surgery was documented in all cases. Six months after surgery each patient had long-leg standing anteriorposterior radiographs and lateral radiographs of the knee. The radiographs were assessed to determine the frontal femoral component angle (FFC), the frontal tibial component angle (FTC), the hipknee-ankle angle (HKA) and the sagital orientation (slope) of both tibial and femoral component. The number and percentage of

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 outliners for each parameter was determined. In addition the percentage of patients from each group with all 5 parameters within the desired range was calculated. Results: At a minimum follow-up of 6 months there non differences in the clinical outcome. The mean surgical was longer in the navigated group of a mean of 8.3 min. The mechanical axes, tibial slope the FTC angle were significantly better aligned in the navigated group. A statistically significant higher number of outliners was seen in the fre-hand group. The number of implants with all 5 radiological parameters aligned within the desired range was statistically higher in the navigated group. All the implants in the navigated group were correctly aligned in all the planned parameters. Conclusions: Despite a not significant longer surgical time, the results demonstrated an improved implant alignment with a statistically significant reduction in the number of outliners in the computerassisted technique. The Authors believe navigation even more helpful in UKR than in TKR because of a higher demanding and less forgiving surgery.

P30-856 Accuracy of an MRI based patient matched cutting jigs technology in TKA: evaluation with navigation A. Ferretti1, F. Conteduca1, R. Iorio1, L. Caperna1, D. Mazza2 , P. Di Sette3 1 St.Andrea Hospital, University of Rome ‘La Sapienza’, Rome, Italy, 2 Ospedale Sant’Andrea, Fiumicino, Italy, 3Ospedale Sant’Andrea, Rome, Italy Objectives: The aim of this study is to evaluate the accuracy of VISIONAIRE Patient Match Technology (Smith & Nephew, Inc, Memphis, Tenn) by analysing the jigs data as detected by intraoperative use of navigation software from BrainLAB (Redwood City, Calif). Methods: Between February 2011 and May 2011, 15 patients with primary gonarthrosis were selected for unilateral TKR. The first three patient were excluded from this study, as they were considered as a warm up to set up the procedure. Preoperatively all patient underwent a full-length weight-bearing radiograph in A-P and an MRI according to the protocol approved by the manufacturer. All patients were operated with cemented posterior stabilized prosthesis cruciate ligament sacrifing (Journey BCS, Smith & Nephew, Inc, Memphis, Tenn) by the same surgeon using the VISIONAIRE jigs. During surgery, once the guides were placed and fixed, the orientation was checked by the navigator. The following parameters were evaluated: seize of the implant, level of resection, alignment in coronal and sagittal plane. In case of great deviation from the planned resection a recut was performed. An unsatisfactory result was considered an error C 2 in both plane for each component as a possible error of 4 could result in aggregate. Results: On the coronal plane the mean deviation of the tibial guide from the ideal alignment was 1.29 ± 1.55 (range 0–5) and in the sagittal plan was 3.83 ± 2.46 (range 0–7.5). On the coronal plane the mean deviation of the femoral guide from the ideal alignment was 1.29 ± 0.68 and in the sagittal was 3.75 ± 2.05. The size of the custom cutting blocks were correct in all the patients. No recuts were necessary. Conclusions: The theoretical advantage of this technology is possible reduction of operating time, costs due to storage and transposition of various size for each patient, and storage and sterilization of several instrumentation boxes. A previous study documented inaccuracy of a similar method (based on TC scan) as evaluated by post-op X-ray. The advantage of this method is use of an AP standing X-ray of lower limbs and an MRI with no additional X-ray exposure.

S307 The main merit of this study is the use of a navigation system to evaluate the accuracy of the system. In fact the final alignment of the implant can be a result of several factors (positioning of the guide, actual plane of resection as deviation of the blade, uniform distribution of cement mantle, symmetry of the final impaction), the only way to actually check the accuracy of the guide is to use of the navigation once the guide is placed in situ. The results of this preliminary study documented a only fair accuracy of the method with a consistent risk of error of more than 3 wether the errors of tibia and femur are considered in aggregate, especially in the sagittal plane. We could speculate that the great error found in the tibial slope and femoral flexion is due to the lack of a preoperative radiological study of the overall lower limb in lateral view.

P30-863 Comparison of accuracy between a MRI based patient matched cutting jigs and the EM tibial guides F. Conteduca1, A. Ferretti1, R. Iorio1, D. Mazza2, L. Caperna1, P. Di Sette2 1 St.Andrea Hospital, University of Rome ‘La Sapienza’, Rome, Italy, 2 Ospedale Sant’Andrea, Rome, Italy Objectives: The aim of this study is to evaluate the accuracy of VISIONAIRE (Smith & Nephew, Inc, Memphis, Tenn) Patient Matched Cutting tibial jigs in comparison with EM tibial instrumentation by analysing data as detected by intraoperative use of navigation system. Methods: Between February 2011 and May 2011, 15 patients with primary gonarthrosis were selected for unilateral TKR. The first three patient was excluded from this study, as they were considered as a warm up to set up the procedure. Preoperatively all patient underwent a full-length weight-bearing radiograph in A-P and a MRI according to the protocol approved by the manufacturer. During surgery, once the extra-medullar guides were placed and fixed on the tibia, the orientation on coronal and sagittal plane was checked by the navigator and then he was compared with the data obtained by measuring the orientation of VISIONAIRE tibial jigs. All patients were operated with cemented posterior stabilized prosthesis cruciate ligament sacrifing (Journey BCS, Smith & Nephew, Inc, Memphis, Tenn) by the same surgeon using the VISIONAIRE technology. In case of great deviation of the tibial guide from the planned resection a recut was performed. An unsatisfactory result ws considerer an error C 2 in coronal and sagittal plane. The results were analyzed using t-Student test for statistical significance (P \ 0.05). Results: On the coronal plane the mean deviation of the EM tibial guides from ideal alignment was 0.7 ± 0.39 (range 0–1.5) and no patients exceeded 1.5. Regarding the positioning of Patient Matched Cutting jigs the mean deviation was 1.29 ± 1.55 (range 0–5) with 2 cases exceeding 1.5. (P = 0.22). On the sagittal plane the mean deviation of the EM tibial guides from ideal alignment (3 of posterior slope) was -1.62 ± 1.78 (range 5; +7) and 4 patients exceeded 1.5. Regarding the positioning of Patient Matched Cutting jigs the mean deviation was +1.16 ± 4.29 (range -4.5; +9) with 9 cases exceeding 1.5 (P \ 0.05). Negative values indicate a more posterior slope from the ideal, positive values an anterior slope. Conclusions: The theoretical advantage of this technology is a reduction of operating time due to the fact that the surgeon is able to perform the cuts without using other instrumentation. From the data obtained we can affirm that we obtained a better alignment with the manual instrumentation especially on the coronal plane. The results of this preliminary study documented only a fair accuracy of the method with a consistent risk of error of more than 3 especially on the

123

S308 sagittal plane. We could speculate that the problem on the sagittal plane depends on the fact that the preoperative protocol does not include a lateral X-ray projection of the knee and only include AP standing X-ray of the straight-leg and MRI. We believe that it is not really possible to determine the real sagittal anatomy without a lateral X-ray. We believe that the system has to be improved to be at the standards of an experienced surgenon.

P30-902 Does total knee arthroplasty design affect proprioception, balance, strength and clinical outcome: a comparative study between posterior cruciate-stabilizing and posterior cruciateretaining TKA P.-J. Vandekerckhove1, R. Parys2, A. Dhollander1, N. Stevens1, L. Van den Daelen3, P. Verdonk1 1 Gent University Hospital, Orthopaedic Surgery and Traumatology, Gent, Belgium, 2KU Leuven, Physical Therapy, Leuven, Belgium, 3 Stedelijk Ziekenhuis Roeselare, Orthopaedic Surgery, Roeselare, Belgium Objectives: The appropriate role for the posterior cruciate ligament in total knee arthroplasty continues to be debated. The goal of this study was to examine its function by comparing posterior stabilized (PS) TKAs versus posterior cruciate retaining (CR) TKAs versus the normal contralateral knee and to investigate knee-joint proprioception, balance, strength and clinical outcome 1 year after surgery. Methods: We evaluated 45 well-functioning TKA patients of which 18 were operated with a PS design and 27 with a CR design implant. The mean age in the CR group was 70 years, in the PS group 67. All surgeries were performed by a single experienced surgeon using the Scorpio CR and PS design (Stryker, Kalamazoo, Michigan, USA). In order to adequately compare both groups we evaluated only patients with a unilateral TKA that had no other orthopaedic surgery performed in the lower legs or spine. None of the patients had a history of neurological problems, which could influence the outcome. Both study groups underwent the Neurocom Balance Master (Natus Medical Incorporated, San Carlos, USA) and Biodex Dynamometer (Biodex Medical Systems Incorporated, Shirley, New York, USA) tests of both legs in order to analyze static balance, proprioception and strength. A clinical examination and clinical scores (KOOS, SF36, VAS score for pain, HSS, Lysholm and Range of Motion) were also taken. The results in terms of balance/proprioception and strength were compared between the CR and PS group while the controlateral nonoperated knee served as control. Results: There was no significant difference in strength, balance or proprioception between the CR and the PS groups, nor was any difference found between the operated and non-operated leg. In addition, our results show no difference in clinical outcome between both groups. Conclusions: The current findings suggest that the preservation or substitution of the PCL in TKA does not affect knee-joint proprioception, balance, strength and subsequently may not influence the clinical outcome after TKA.

P30-911 Clinical, functional, and radiological outcomes in computerassisted total knee replacement: comparison between two prosthetic designs F. Catani1, V. Digennaro1, G. Grandi1, R. Mugnai1, A. Marcovigi1 1 Modena University Hospital, Modena, Italy Objectives: The aim of this study was to evaluate clinical, radiographical and functional outcomes of two different oprosthetic design in 219 consecutive cases.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: Between 2007 and 2009, 227 computer-assisted primary TKA were performed in 219 consecutive patients. 19 patients were ineligible to enter the study as they didn’t met the inclusion criteria. Two different TKA designs were evaluated in this study: the Scorpiorestrictive geometry (NRG) knee system and the Journey bi-cruciate stabilized knee system (BCS). All prostheses were implanted with the aid of surgical navigation (Stryker Knee Navigation System). The patella was resurfaced in all cases. The following clinical parameters were assessed: flexion–extension range of motion (ROM) and functional outcome, measured with the Knee Injury and Osteoarthritis Outcome Score (KOOS). Results: At a mean 29 months follow-up the Journey group obtained higher mean KOOS scores in all subscales and a greater ROM with respect to the NRG group. This difference proved to be statistically significant regarding pain (p = 0.007) and knee-related quality of life (p = 0.045) KOOS subscales, and the postoperative ROM (p = 0.018). Complications observed in 6.9% of patients were: knee stiffness in 3.2%, frontal plane instability in 1.1%, anterior knee pain in 2.1%, and synovitis pain in 0.5%. Conclusions: In conclusion we observed a significant correlation between the type of prosthesis and the KOOS’ domains Pain and Quality of Life, and the postoperative ROM, suggesting a better impact on outcome for the patients operated on Journey TKR. In the present research the mean postoperative KOOS score was found higher than those reported in recent studies on primary non-navigated TKA, suggesting that with the aid of surgical navigation better clinical outcomes can be achieved in TKA.

P30-912 Five to nine years results of high flexion fixed bearing total knee arthroplasty: how many knees survived? M.C. Lee1, D. Kim1, S. Lee1, J. Jang1, S.H. Chun1, S.C. Seong1 1 Seoul National University College of Medicine, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: There is concern whether deep flexion will increase early loosening. The aim of this study was to evaluate the mid and long term survival rate, clinical and radiological results of a fixed bearing high flexion posterior stabilized (PS) total knee arthroplasty (TKA). Methods: Between July 2001 and Feb 2006, 399 high flexion PS TKAs (335 patients) were performed and 339 knees (284 patients, 84.8%) had been followed up for mean 6.1 years (range, 5–9.2 years) clinically and radiographically. We evaluated degree of knee flexion, Knee rating system of the Hospital for Special Surgery (HSS) and Knee Society (KS) scoring system, and survival rate. Results: The mean flexion improved from 125.0 preoperatively to 130.6 at the latest follow-up. Two hundred seventy-four knees (80.8%) and 96 knees (28.3%) showed more than 125 and 140 of flexion. The mean KS clinical and function score improved from 39 to 93 points (p \ 0.01) and from 40 to 85.4 points (p \ 0.01). The mean HSS score improved from 41.2 to 86.3 points (p \ 0.01). The estimated survival rate at 9 years with revision for any reason and for aseptic loosening was 89.4 and 91.1%. Conclusions: The high flexion PS TKA showed satisfactory clinical results with high degree of knee flexion. The relatively long term survival was about 90% in this series.

P30-929 Analysis of the knee joint kinematics and muscle activity during gait in patients with total knee replacement G. Hajduk1, G. Sobota2, K. Nowak2, B. Bacik2, D. Kusz1, K. Kopec1 1 Medical University of Silesia, Department of Orthopaedics and Traumatolgy, Katowice, Poland, 2Institute of Biomechanics, The Jerzy Kukuczka Academy, Katowice, Poland

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Usually, there is an excellent improvement in gait pattern after Total Knee Replacement (TKR). Despite a very good clinical outcomes, some kinematics and muscle function abnormalities persist and they might have implication in long term prosthesis failure. The aim of this study was to asses the gait pattern and muscle activity in patients with knee osteoarthritis before and after TKR, as well as their relationship with knee biomechanics. Methods: 14 patients with severe knee osteoarthritis were evaluated preoperatively and at 12 months after TKR by means of a threedimensional kinematic gait analysis (BTS Smart—500D) and clinical assessment including Knee Society Score (KSS) and WOMAC questionnaire. Additionally, surface EMG from lower limb muscles was registered. All of the patients received the same implant with posterioly stabilized insert. Patella was not replaced. The control group consisted of 13 healthy, age-matched individuals Results: Analysis of temporo-spatial parameters in both study groups before and after TKR showed a distinct decrease in values of a number of variables such as velocity, cadence, step length. Stance phase were elongated. The operated knee has limited flexion both during the stance and the swing phase. Examined knee muscle activity during walking following TKR surgery report prolonged muscular co-contraction stance phase. Gait pattern and muscle activity in patients with TKR became closer to group with knee OA before surgery when compared to the analysis performed in healthy individuals. Conclusions: Prolonged quadriceps contraction and limited knee flexion in patient with TKR could be a compensatory mechanism aimed at providing greater control of the knee joint during stance.

P30-1020 Genu recurvatum versus fixed flexion deformity after total knee arthroplasty: which is the lesser evil? K. Koo1, S. Amila1, P.L. Chin1, S.-L. Chia1, N.N. Lo1, S.J. Yeo1 1 Singapore General Hospital, Orthopaedic Surgery, Singapore, Singapore Objectives: The goal of total knee arthroplasty (TKA) is to provide a knee that has a functional range of motion. Although a means to achieving this goal is to bring the knee to full extension (i.e 0), postTKA the knee can either end up in genu recurvatum (GR) or with a fixed flexion deformity (FFD). To date, there is no study comparing clinical outcome between post TKA GR and FFD. The aim of this study is to provide knowledge to clinicians when deciding which side (i.e. in GR or in FFD) to err on when neutral extension is not achieved. Methods: We retrospectively reviewed all prospectively collected data on primary TKAs done at a single centre between 2004 and 2008. Patient biodata, post-TKA range of motion as well as clinical scores (Knee Society Clinical Rating or KSC, Oxford Knee and SF-36 scores) were reviewed at 6 months and at 2 years follow-up. Clinical outcome and scores of patients with post-TKA GR were compared with patients with post-TKA FFD. Results: 2,136 primary TKAs were reviewed at both 6 months and 2 years post TKA. At 6 months, 254 knees resulted in GR, while 1,882 knees had a FFD. At 2 years, there were 228 GR knees and 1,332 with FFD. At 6 months, the degree of FFD ranged from 1 to 50 (mean 8.26, SD 5.89) while the degree of GR ranged from 1 to 20 (mean 4.85, SD 2.62). At 2 years, FFD ranged from 1 to 30 (mean 6.25, SD 3.41) while GR ranged from 1 to 30 (mean 5.00, SD 2.86). There was no statistical difference in mean KSC and Oxford Knee scores between the 2 groups at 6 months and 2 years. However, SF2 (Role Functioning—Physical) and SF 3 (Bodily Pain) scores were significantly better in FFD compared to GR at 6 months (p \ 0.001) and 2 years (p \ 0.001). Majority of our patients fall within the -10

S309 (i.e GR) to +10 (FFD) range. 89.1 and 94.5% at 6 months and 2 years respectively. For this subset of patients, KSC (Knee and Function) Scores at both 6 months and 2 years as well as Oxford knee scores at 6 months were significantly better in FFD between 0 and 10 as compared to GR of 0–10. (p values all \ 0.05). 1,785 out of 1,882 (94.8%) knees with fixed flexion showed improvement in their deformity from 6 months to 2 years. In contrast, only 78 out of 254 (30.7%) knees with genu recurvatum showed an improvement in their deformity. At 6 months, knees with -4 to +6 have Oxford Knee Score and KSC (Function and Knee) scores that is not significantly different from that of patients who attain neutral. (p [ 0.05). Conclusions: SF-2 and 3 scores were significantly better in FFD compared to GR at 6 months and 2 years. However, on removing outliers and analyzing only those within -10 to +10, KSC (Function and Knee) scores at 6 months and 2 years as well as Oxford scores at 6 months were signifcantly better in knees with FFD. Knees in FFD also show greater improvement in deformity over time as compared to GR. Based on this study, we conclude it is better to err on the side of FFD rather than GR.

P30-1050 Assessment of joint gap for mid-flexion (45) and hyper-flexion (120) during navigation-assisted TKA J.-H. Yang1, J.-R. Yoon1, K.J. Oh2, C.-H. Oh1, H. Lim1, Y.-C. Kim1 1 Seoul Veterans Hospital, Department of Orthopaedic Surgery, Seoul, Republic of Korea, 2KonKuk University Medical Center, Department of Orthopaedic Surgery, KonKuk University School, Seoul, Republic of Korea Objectives: The aim of gap balancing during total knee arthroplasty (TKA) is to achieve rectangular flexion and extension gaps. However, assessment of gap in midflexion (45) and deep flexion (120) is obscure. The hypothesis of this study was that if the mediolateral gap is assessed in different knee flexion angles (0, 45, 90, 120), gap difference would be categorized into several groups with clinical difference between them. Methods: Fifty knees operated by TKA using a navigation-assisted gap balancing technique with a minimum of 2-year follow-up were evaluated. Final mediolateral gap in each flexion angle (0, 45, 90, 120) was measured after tibia/femoral bone resections with femoral prosthesis in situ. Any gap difference of more than 3 mm was considered significant. Clinical outcomes were assessed at 3, 6, 12 and 24 months postoperatively using the Knee Society Score (KSS) and Western Ontario MacMaster (WOMAC) score. Preopand postop- range of motion (ROM) and mechanical axis was obtained at each visit. Analysis of variance (ANOVA) and Pearson correlation analysis was performed. Statistical significance was considered in p \ 0.05. Results: All cases were managed to have rectangular gap at 0 and 90. The final mechanical axis were within 3 in all cases. Patients were divided into 4 groups. Group 1: no gap difference, Group 2: Lax in midflexion, Group 3: Lax in hyperflexion and Group 4: Lax in both midflexion and hyperflexion. Number of patients in each group was as follows: Group 1; n = 32 (64%) Group 2; n = 10 (20%), Group 3; n = 4 (8%) and Group 4; n = 4 (8%). All of the joint gaps with significant difference ([3 mm) were in trapezoidal shape with a wider lateral side. However, no clinical differences were observed between groups (p [ 0.05). Correlation between the incidence of midflexion/ hyperflexion laxity and preoperative mechanical alignment was not found. Conclusions: This study shows significant proportion (20%) of cases with lax midflexion (45) even in rectangular extension (0)-flexion

123

S310 (90) gap. However, clinical significance was not found in this short term follow-up study.

P30-1054 Comparison of total knee arthroplasty using patient specific instruments versus conventional instruments M.C. Lee1, S. Lee1, J. Jang1, S.H. Chun1, Y.W. Roh1, S.C. Seong1 1 Seoul National University College of Medicine, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: Component alignment is an important factor affecting the long-term outcome of total knee arthroplasty. Patient specific instrument was developed to improve the accuracy of implant placement without negative aspects of navigated total knee arthroplasty. The purpose of this study was to compare the alignment of TKA using patient specific instruments (PSI) to TKA using conventional instruments. Methods: Fifty primary osteoarthritis knees were prospectively randomized to undergo total knee arthroplasty with patient specific instruments (PSI group; 30 knees) or conventional instruments (Conventional group; 30 knees). Patient specific guides for PSI group were generated from CT scan. We analyzed postoperative mechanical axis, coronal and sagittal alignment with plain radiographs and 3-dimensional CT scan. Outliers in coronal alignment were defined as [3. Results: There was no difference in preoperative demographics, clinical and radiographic data between the groups. The mechanical axis of the leg was varus 0.6 in PSI group and valgus 0.9 in conventional group (p [ 0.05). Femoral coronal alignment (varus 1 vs. varus 0.1; p [ 0.05) and tibial coronal alignment (varus 0.5 vs. valgus 0.2; p [ 0.05) were similar between two groups. Femoral sagittal alignment (3.5 flexion vs. 3.1 flexion; p [ 0.05) and tibial sagittal alignment (3.1 posterior slope vs. 3.6 posterior slope; p [ 0.05) were also similar between the groups. The prevalence of outliers in the mechanical axis of the leg was 16% in PSI group and 12% in conventional group (p [ 0.05). The percentage of outliers in the femoral (4 vs. 4%; p [ 0.05) and tibial coronal alignment (0 vs. 0%; p [ 0.05) was also similar between the groups. Conclusions: Total knee arthroplasty with patient specific instruments provide accurate implant positioning and alignment without breaching the intramedullary canal.

P30-1085 Improved metaphyseal fixation of revision total knee arthroplasty with metal modular augments fixed to the bone instead of the tibial baseplate A. Baldini1, L. Manfredini1, F. Traverso2, G. Grappiolo2 1 IFCA, Ortopedia, Firenze, Italy, 2IRCCS Istituto Clinico Humanitas, Rozzano Milano, Italy Objectives: High rate of radiolucent lines (RLL) at the level of the tibial baseplate modular augments bone-cement interface after revision total knee arthroplasty (RevTKA) has been reported. Sclerotic bone margins under a failed TKA, and stress-shielding from diaphyseal fixation with stems are leading causes of bone-implant RLL and suboptimal metaphyseal fixation. The aim of our paper is to verify if cementless tantalum augments directly fixed to bone with cancellous screws could improve metaphyseal fixation and reduce the rate of RLL at follow up. Methods: Twenty consecutive RevTKA performed for septic or aseptic loosening with a medial tibial uncontained defect of grade 2 AORI were enrolled in a prospective randomized study. Patients were randomized to receive the bone defect treatment either with a cemented titanium augment fixed to the tibial baseplate (group I) or a tantalum augment fixed to the baseplate with cement and to the defect

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 with cancellous screws (group II). All the tibial components were fixed with cement to the lateral plateau and stem and cementless diaphyseal-engaging stem extension. Use of a semiconstrained polyethylene was comparable in the two groups. Results: At an average follow-up of 16 months (range: 12–22 months) RLL were evaluated using fluoro-aligned images according the Knee Society Roentgenographic System. A line in at least one tibial zone was found in 6 of the eleven patients in group I and in none of the nine patients of group II. All trabecular metal augments showed signs of bone integration. No significant clinical difference was found between the two groups. Conclusions: Cementless metaphyseal fixation with trabecular metal augments directly secured to bone defects may improve implant fixation in RevTKA. Additional follow-up is needed to evaluate possible clinical outcome differences at mid-to-long term.

P30-1100 Health economics and adoption of a new technology in orthopaedics A. Toms1, D. Coorey2, R. Latimore3, E. Garling4 1 Princess Elizabeth Orthopaedic Centre, Exeter Knee Reconstruction Unit, Exeter, United Kingdom, 237point5 Limited, Warton, United Kingdom, 3Stryker Europe, Newbury, United Kingdom, 4Stryker SA, Montreux, Switzerland Objectives: The trend towards not just clinical effectiveness but comparative effectiveness and efficiency is being demanded on the introduction of new medical devices. Orthopaedic knee replacements are a particular focus for scrutiny in some health economies like the UK. Especially since patient satisfaction following total knee replacement (TKR) remains a challenge. The recent introduction of the OtisMed CustomFit technology (Stryker, Mahwah, US) is aiming to improve patients’ outcome and to improve the total knee pathway efficiency resulting in lower costs. The aim of the current study was to assess the economic benefits of using the OtisMed CustomFit patient pathway. Methods: OtisMed CustomFit technology uses the Triathlon Knee prosthesis, Stryker’s Precision instrumentation and MRI or CT data to prepare an accurate 3D pre-operative plan where each patient’s prearthritic anatomy and alignment is recreated. In total, six hospitals in the UK were selected to participate in the study; three private and three public health service hospitals. Extensive pathway mapping was conducted at all sites and Triathlon CustomFit Knee was compared to existing pathways to determine positive and negative influences. Using data gathered from stakeholder interviews and theatre list observations, each segment of the pathway was mapped via a swim lane process map to allow identification of both the process itself and interaction across departments. Results: The study found that the net effect of Triathlon CustomFit Knee Precision disposable instruments is a £950 surplus, including potential savings released by reduced length of stay. This surplus must be cash releasing i.e. the service must either increase volume or reduce staff costs to realise this saving. The use of Precision disposable instrumentation allows the maximum reduction in trays (from 5 to 1 core tray) allowing both the maximum sterilization savings and time releases during interface time (15 min per case). This time release coupled with the release in time during operating room setup, narrowing in the variability of anaesthetic times and shorter intra operative time (10–15 min) as reported by (Howell 2008) with similar OtisMed technology, is critical to releasing enough time to do a single extra case. Conclusions: The assessment of the economic impact of new technologies requires a complete understanding of the full healthcare pathway in which the technology is to be placed and the environment and associated limitations in which the pathway is being executed.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Triathlon CustomFit Knee with OtisMed ShapeMatch Technology offers those healthcare providers who have the ability to effect the execution of their orthopaedic pathway, the chance to access the potential clinical and financial benefits for the patients they serve. The financial model suggests an affordability model exists for healthcare providers to gain access to this technology and savings may in fact be made when correct patient flow pathway is followed.

P30-1127 Cemented versus cementless fixation in Oxford unicompartmental knee arthroplasty at 5 years: a randomised controlled trial C.A. Dodd1, H. Pandit2, A.D. Liddle3, A.J. Price3, H.S. Gill3, D.W. Murray3 1 Nuffield Orthopaedic Centre, Nuffield Department of Orthopaedic Surgery, Oxford, United Kingdom, 2Nuffield Orthopaedic Centre, NDORMS, Oxford, United Kingdom, 3University of Oxford, NDORMS, Oxford, United Kingdom Objectives: The Oxford Unicompartmental Knee Arthroplasty (OUKA) has been used for over 30 years as a cemented implant. It has been shown to have a low rate of wear in vivo and in simulation studies. The majority of early revisions may be attributable to errors in cementation. Failure by aseptic loosening is considered to be rare but varies between published series. Using joint registry data for all designs of unicompartmental knee arthroplasty, a significant proportion of all revisions is for aseptic loosening. In between 70 and 96% of cases, a thin radiolucent line is present in association with the OUKA on follow-up radiographs. This is not associated with loosening but may precipitate revision in cases of unexplained pain. Modification of the OUKA implant to allow cementless fixation has been shown to reduce the incidence of radiolucencies at 1 year with no effect on clinical outcome. The objective of this study was to compare the clinical outcomes and radiological appearances of cemented and cementless OUKA at 5 years. Methods: A randomised controlled trial was established with 62 knees (61 patients) being randomised to receive either cemented or cementless OUKA. 32 patients received cemented Phase III OUKA implants; 30 patients received a similar implant, modified by coating the implants with porous titanium and hydroxyapatite to allow cementless fixation. Groups were well matched for age, gender, body mass index and preoperative functional scoring. Patients were followed-up at 6 months, one and 5 years with functional scoring and fluoroscopically-controlled radiographs. Results: Four patients (three cementless and one cemented) could not be reviewed either due to death unrelated to surgery (three patients) or development of dementia (one patient). One further patient was removed from the trial after randomisation as the cementless component was unavailable. There have been no major complications and no revisions. Mean Oxford Knee Score (OKS) was 41.0 (standard deviation 9.9) in the cemented group and 39.4 (SD 9.1) in the cementless group. The functional component of the American Knee Score (AKS) was 81.9 (SD 18.5) in the cemented group and 94.1 (SD 10.3) in the cementless group. The mean objective component of the AKS was 81.5 (SD 19.0) in the cemented group and 78.5 (SD 14.1) in the cementless group. There was no significant difference in OKS or the objective component of the AKS between the two groups. Functional component of the AKS was significantly better in the uncemented group (p = 0.015). In the cemented group, there were 13 patients with radiolucencies, 3 partial and 10 complete. In the cementless group, there were 4 radiolucencies overall, 4 partial and 0 complete. The incidence of radiolucencies was significantly lower in the cementless group (P = 0.012).

S311 Conclusions: Cementless fixation is associated with significantly fewer radiolucencies at 5 years when compared with cemented fixation in OUKA whilst maintaining equivalent clinical outcomes.

P30-1149 Fitness of the femoral components of total knee arthroplasty with distal femoral cut surfaces in Korean women S.D. Cho1, Y.S. Youm1, J. Eo1, K.J. Lee1 1 Ulsan University Hospital, Department of Orthopedic Surgery, Ulsan, Republic of Korea Objectives: To determine the intra-operative anatomical fitness between the distal femur and the femoral implant and to evaluate the relationship of aspect ratio and width ratio between the distal femur and the femoral implant in total knee arthroplasty (TKA) for Korean women. Methods: Two hundred twenty-seven women (321 knees) who underwent TKA using NexGen LPS (162 knees) and PS ADVANCE MPK (159 knees) implants, were analyzed prospectively. The mean age was 67.8 (range 52–87) years. The femoral component size was determined by considering both anteroposterior dimension and the amount of posterior condylar resection. After distal femoral resection, the mediolateral (ML) width was measured at four points (anterior (Ant), distal anterior (DA), distal posterior (DP) and posterior (Post)) and compared with the ML width of the implant respectively. The aspect ratio (AR; ML/AP ratio) and width ratio (WR, anterior/distal posterior ML width; Ant/DP ratio) were calculated. The relationship of aspect ratio and width ratio between the distal femur and the femoral implant were analyzed. Results: For LPS conventional/gender designs, the ML width at Ant of femoral component was overhang in 22/2, optimal in 98/46 and undersize in 42/114 knees; 8/0, 116/26 and 38/136 at DA; 1/0, 89/34 and 72/128 at DP; 1/0, 85/30 and 76/132 at Post, respectively. For conventional MPK design, those at Ant was overhang in 67, optimal in 89 and undersize in 3 knees; 52, 102 and 5 at DA; 5, 114 and 40 at DP; 3, 92 and 64 at Post, respectively. For the size 2 and 3 of MPK conventional/gender designs, those at Ant was overhang in 61/3, optimal in 65/30 and undersize in 3/96 knees; 51/2, 76/30 and 2/97 at DA; 5/0, 94/12 and 30/117 at DP; 3/0, 77/8 and 49/121 at Post, respectively. For LPS and MPK conventional/gender designs, only the AR of conventional MPK size 3 was larger than that of distal femur. For LPS and MPK conventional designs, the WR was larger than that of distal femur in all cases. For both gender designs, that of MPK size 3 was larger than that of distal femur. Conclusions: Conventional Nexgen LPS component had more anatomical fitness of the femoral component than conventional PS ADVANCE MPK or gender design of each implant. Overhang on the anterior cut surface could occur though optimal on the distal cut surface, which is reflected by the width ratio.

P30-1150 Effect of accompanying diseases on the quality of life results after total knee arthroplasty A. Murat1, B. Bostan1, E. Bilgic1, F. Tas1, T. Gunes1, U. Erkorkmaz2 1 Gaziosmanpasa University Medical School, Orthopaedics and Traumatology, Tokat, Turkey, 2Gaziosmanpasa University Medical School, Tokat, Turkey Objectives: Total knee arthroplasty is one of the leading major operations of the current orthopaedic practice. It is generally performed on elderly patients and results are usually satisfactory. In this age group most of the patients have accompanying diseases. The

123

S312 effect of these accompanying diseases on the results of quality of life after total knee arthroplasty is investigated. Methods: Between 2003 and 2011, 217 total knee arthroplasties were performed in our department. 112 of these patients were evaluated by using Charlson comorbidity index and SF-36 scoring systems retrospectively. Charlson comorbidity index was acquired from the medical records of the patients. Pre- and postoperative SF-36 scores were compared within and among the groups of Charlson comorbidity index. Results: In regard to Charlson comorbidity index patients were divided into four groups. Patients without an accompanying disease were called Charlson 0 and there were 18 patients in this group. There was 40 patients with one accompanying disease and this group is named as Charlson 1. The number of patients with two accompanying diseases, Charlson 2, and three accompanying diseases, Charlson 3; were 28 and 26 retrospectively. All parameters of SF-36 showed statistically significant increase in the postoperative period. Patients with 3 or more accompanying diseases had statistically significant improvement in respect to other groups on the mental health and general health parameters of SF-36 (p = 0.0001). Conclusions: A significant increase in the quality of life and functional results can be obtained after total knee arthroplasty. The increase in mental health and general health parameters of SF-36 in patients with 3 or more accompanying diseases were attributed to improvement in their depressive status, caused by multiple accompanying diseases, as a result of motivation supplied by mobilization without pain.

P30-1180 Comparison of the changes in the patellar tendon thickness with flexion between meniscal bearing and rotating platform design prosthesis M. Takeda1, Y. Ishii1, H. Noguchi1, J. Sato1 1 Ishii Orthopaedic and Rehabilitation Clinic, Gyoda, Japan Objectives: The low-contact stress (LCS) knee prosthesis is a mobilebearing design with modifications to the tibial component that allow for meniscal-bearing (MB) or rotating-platform (RP). The MB design had nonconstrained anteroposterior and rotational movement, and the RP design has only nonconstrained rotational movement. The anterior soft tissues, including patellar tendon (PT), prevent anterior dislocation of the MB. Therefore, we hypothesized that the PT thickness with flexion in MB revealed less changes than those of RP due to degeneration of the PT induced by much mechanical stress of the MB movement, and their changes might affect on maximum flexion angle. To confirm this hypothesis, we analyze the changes in the PT thickness with flexion induced by mobile-bearing inserts. Sixty-six LCS prostheses in 33 patients were analyzed. The average follow-up time was 61 months. MB prosthesis was used on one side of the knee and RP prosthesis was used on the contralateral side of the knee. All patients were chosen from group with no clinical complication, and all had achieved passive full extension and at least 90of flexion. The average Hospital for Special Surgery Score was 94.6 ± 2.7. Methods: We measured the thickness of PT at joint line level, which were confirmed by sagittal section using ultrasound between extension (EX) and 90 flexion (FX) and compared the patellar tendon thickness ratio (PTTR), which calculated (FX-EX) 9 100/EX (%), between MB and RP design prosthesis. Simultaneously we investigated the correlation between maximum flexion angle and PTTR. Results: The mean PTTR (SD) was -2.6% (12.1) with MB and 3.6% (16.5) with RP design prosthesis. No significant differences (p = 0.79) were found between both groups. There was no significant

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 correlation between maximum flexion angle and PTTR in both groups (p = 0.22 in MB) (p = 0.52 in RP). Conclusions: The current results rejected 2 our hypotheses. One is that PT thickness with flexion in MB reveals less change than those of RP due to degeneration of the PT induced by much mechanical stress of the MB movement. The other is that their changes might affect on maximum flexion angle. The possible reasons are the following two: (1) We did not remove infra-patellar fat pad, which might play shock absorber of mechanical stress from MB, and prevent from significant degeneration of PT and (2) MB inserts did not move so as to cause degeneration in PT.

P30-1183 Effects of component malpositioning and soft-tissue balancing in total knee arthroplasty kinematics S. Pianigiani1, Y. Chevalier2, L. Labey2, W. Pascale1, B. Innocenti2 1 I.R.C.C.S. Istituto Ortopedico Galeazzi, Milano, Italy, 2European Centre for Knee Research, Smith & Nephew, Leuven, Belgium Objectives: Clinical outcomes in total knee arthroplasty (TKA) can be affected by implant alignment or ligament balancing. Component malpositioning was already suggested to alter tibiofemoral (TF) and patellofemoral maximum contact forces. However, how component malpositioning and soft-tissue balancing affect TF kinematics has not yet been fully investigated. Our goal was to evaluate how such factors can alter TF kinematics for TKAs knee. Methods: A validated, dynamic, musculoskeletal model based on computer tomography of a cadaver leg was used to simulate a squat motion up to 120 with a 200 N vertical hip load. The model was implanted with four different cruciate sacrificing TKA types: Type I) a fixed bearing, posterior stabilized design (Genesis II PS); Type II) a fixed bearing, high flexion design (Journey BCS); Type III) a hinge design (RT-PLUS); and Type IV) a mobile bearing design (EPP). All prostheses are from Smith & Nephew, Memphis, TN. Modifications were done to simulate several component malpositions or ligament insertion shifts (balancing) (Table 1). Internal-external (IE) and abduction–adduction (AA) rotations and antero-posterior (AP) translations of the femoral component with respect to the tibial component were finally evaluated for each model during the descent phase of the squat. Results: Figure 1 shows the maximal effects for malposition scenarios and ligament insertion shifts on the ranges of motion in IE and AA

Table 1 Simulated TKA component malpositions and ligament balancing Tibial component: Antero/posterior translation

±3 mm

Medio/lateral translation

±3 mm

Slope

±3

Abduction/adduction rotation

±3

Internal/external rotation

±5

Lateral collateral ligament: Antero/posterior translation

±5 mm

Medial/lateral translation

±5 mm

Proximal/distal translation

±5 mm

Medial collateral ligament: Antero/posterior translation

±5 mm

Medial/lateral translation

±5 mm

Proximal/distal translation

±5 mm

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S313

Fig. 1 For each type of TKA, maximal effects for malposition scenarios (blue) and ligament insertion shifts (green) on the ranges of motion in: a IE rotations, b AA rotations and c AP translation compared to the reference configuration (red line)

rotations and AP translation compared to the reference configuration (red line). Maximal effects on IE rotations were observed for changes in ligament balancing for some of the designs but malpositioning still affected the measured range of motion (Fig. 1a). Tibial component malpositions affected the ranges of AA rotation and AP translation more than ligament balancing for all designs (Fig. 1b, c). Conclusions: This numerical study was done not to compare different designs, but to demonstrate that, irrespective of TKA implant design, TF kinematics can be altered by changes in implant positioning and changes in collateral ligament insertions simulating balancing. The predicted effects suggest that both an adequate positioning of the component and balancing of the knee might be important factors to influence clinical outcomes. This study will help guide future research on implant behavior and surgical procedures, ultimately leading to improved TKAs life expectancy.

P30-1218 Five RSA studies to study the safety of the triathlon TKA system: a case for phased innovation in orthopaedics M. Molt1, S. Toksvig-Larsen1, E. Garling2 1 Ortopedkliniken Sjukhuset Ha¨ssleholm, Ha¨ssleholm, Sweden, 2 Stryker SA, Montreux, The Netherlands Objectives: Market forces continuously apply pressure to the healthcare sector to introduce new orthopaedic devices. However, decisions about any medical treatment should be based on a careful appraisal of the best evidence available. A step-wise introduction is necessary to increase the use of evidence-based decision making in the implementation of new surgical techniques and implants while exposing as few patients as possible to the potential risk of failure (Malchau 2000). RSA is a methodology that should and can be used in the first clinical introduction phase of this Phased Innovation Model (PIM).

With an RSA study, a small number of patients identify faulty designs after 2 years of follow-up preventing numerous patients being exposed to inferior prostheses or fixation methods. In this regard, we tested the variety of implants of the Triathlon TKA system introduced in 2005 (Stryker, Mahwah, USA). Methods: In 5 prospective randomized single center RSA studies with a total of 300 patients the short term fixation after 2 years of followup as an indicator for future mechanical loosening was assessed. Knee Society Scores and KOOS clinical scores for all patients were collected as well as normal clinical radiographic evaluations. Results: The 1st study compared the cemented version of the Triathlon with its predecessor the Duracon total knee. There were no significant differences in the RSA 2-year follow-up data nor in the clinical data (p \ 0.05), which suggests the Triathlon knee system may replicate the excellent long-term clinical results achieved by the Duracon knee system. This outcome is confirmed by step 4 of the PIM: the Finnish and NJR registries report more than 97% survival after 5 years. The 2nd study compared the CR and PS version and again no significant differences in clinical or radiographic outcomes were found. The 3rd study compared the uncemented PA coated Triathlon with the Pressfit Triathlon. Minimal micromotion of the PA coated tibial components was observed, but higher variability of the Pressfit components. The 4th study showed that the short stem Triathlon has the same migration pattern as the normal stemmed tibia component. The 5th study is comparing the cemented version with the uncemented Triathlon (results expected Q1, 2012). The exposure of this relatively small patient group to the new Triathlon Knee system has showed that the Triathlon Knee system is safe. Conclusions: A step-wise introduction is necessary to increase the use of evidence-based decision making in the implementation of new surgical techniques and implants while exposing as few patients as possible to the potential risk of failure. RSA is a methodology that should and can be used in the first clinical introduction phase of this

123

S314 Phased Innovation Model exposing as few patients as possible to the potential risk of failure. Regulatory bodies should adopt the PIM providing reliable data for better evidence based decisions about innovations and new or improved medical devices.

P30-1299 Effect of polyethylene crosslinking and bearing design on wear of unicompartimental arthroplasty R. Streicher1, J. Netter2, J. Hermida2, J. D’Alessio3, M. Kester3, D. D’Lima2 1 Stryker SA, Montreux, Switzerland, 2Scripps Clinic, La Jolla, United States, 3Stryker Orthopaedics, Mahwah, United States Objectives: Wear and polyethylene damage continue to be important factors affecting outcomes, and have been implicated in up to 22% of revision surgeries after unicompartmental knee replacement. This study was designed to determine if a highly crosslinked fixed-bearing design would increase wear resistance. A secondary objective was to quantify the relative contribution backside wear in unicompartmental mobile-bearing designs. Methods: Gravimetric wear rates were measured for two unicompartmental implant designs: Oxford unicompartmental (Biomet) and Triathlon X3 PKR (Stryker) on a knee wear simulator (AMTI) using the ISO recommended standard. A finite element model of the AMTI wear simulation for both designs was constructed to replicate experimental conditions and to compute wear. This approach was validated using experimental results from previous studies. The wear coefficient obtained previously for radiation sterilized low crosslinked polyethylene was used to predict wear in Oxford components. The wear coefficient obtained for highly crosslinked polyethylene was used to predict wear in Triathlon X3 PKR components. To study the effect design and polyethylene crosslinking, wear rates were computed for each design using both wear coefficients. Results: FEA predicted wear rates were very close to those measured experimentally for both Oxford and Triathlon X3 PKR designs, validating our model assumptions. Our FEA wear penetration rates (0.024 mm/million cycles) also compare well to in vivo studies, which reported 0.022 mm/year for Oxford bearings. Over 40% of the total wear occurred at the ‘‘backside’’ of the mobile bearing, for low and high crosslinked polyethylene. Overall, wear in the Triathlon X3 PKR design was 80% less than that in the Oxford design. Major factors that affect wear are the level of crosslinking in polyethylene and the implant design. Conclusions: We used a combined experimental and computational approach to quantify factors contributing to polyethylene wear after unicompartmental knee arthroplasty. To isolate the effect of crosslinking level and mobile-bearing design, we computed wear rates for both designs using wear coefficients obtained for both low and highly crosslinked polyethylene. Wear rate in a highly crosslinked Oxford insert was reduced although still higher than that in a highly crosslinked Triathlon X3 PKR. The level of crosslinking and backside wear are major factors contributing to the difference in wear: backside wear (46%) and increased crosslinking (43%). A validated computer model is extremely valuable for efficient evaluation of wear performance and design development. Increasing conformity to increase contact area and reduce contact stress may not be the sole predictor of wear performance.

P30-1488 Effect of posterior condyle of femur on extension gap during primary TKA S.-S. Seo1, C.-W. Kim2, J.-J. Kim2, D.-W. Jung2 1 Busan Paik Hospital, Inje University, Orthopaedics, Busan, Republic of Korea, 2Inje University, Busan, Republic of Korea

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: It is important to obtain the equal flexion–extension gap on performing TKA with gap technique. We hypothesized that resection of posterior femoral condyle would affect extension gap. The purpose of this study is to evaluate the difference of extension gap between before resection and after resection of posterior condyle of femur and to elucidate its clinical implication. Methods: 40 knees with primary OA were enrolled for this study. The knees with uncorrectable, contracted collateral ligament and large osteophyte on posterior femoral condyle before surgery were excluded. After proximal tibial resection an appropriate femoral AP cutting block was applied and anterior femur was removed. The posterior femoral condyle was incompletely cut and measured extension gap with torque wrench in force of 89 N and 178 N. And then the posterior femoral condyle was totally removed and repeatedly measured extension gap in same way. Results: Under 89 N force the mean value of extension gap was 13.3333 (12.7543–13.9124 95% CL) before and 13.8182 (13.2614–14.3750 95% CL) after removal of posterior femoral condyle. Mean difference was 0.4824 ± 1.0038 (Mean ± SD) and was significant (p = 0.0091). Under 178 N force the mean value of extension gap was 15.1222 (14.5986–15.6438 95% CL) before and 16.0606 (15.4713–16.6493 95% CL) after removal of posterior femoral condyle. Mean difference was 0.9394 ± 0.9417 (Mean ± SD) and was significant (p \ 0.0001). Both experimental tension force showed less than 1 mm of mean difference of extension gap. Conclusions: Mean extension gap difference between before and after resection of posterior femoral condyle was less than 1 mm which was statistically significant. But commonly used modular TKA implants provide a different thickness of 2 mm polyethylene insert. But less than 1 mm extension gap difference would not have clinical implication.

Knee - total joint replacement II

P31-142 Intra-operative kinematic analysis of the knee during total replacement. A prognostic information? J.-Y. Jenny1, L. Wasser1 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France Objectives: We wanted to assess the possible correlation between the intra-operative kinematics of the knee and the clinical results after total knee replacement (TKR). Methods: 187 cases of TKR have been prospectively analyzed. There were 127 women and 60 men, with a mean age of 71.4 years. Indication for TKR was osteoarthritis in 161 cases and inflammatory arthritis in 26 cases. A floating platform, PCL preserving, cemented TKR was implanted in all cases. A non-image based navigation system was used in all cases to help for accuracy of bone resections and ligamentous balancing. The standard navigation system was modified to allow recording the three-dimensional tibio-femoral movement during passive knee flexion. Two sets of records have been performed: before any intra-articular procedure and after final implantation. Only antero-posterior femoral translation (in mm) and internalexternal femoral rotation (in degrees) have been recorded. Kinematic data have been analyzed in a quantitative manner (total amount of displacement) and in a qualitative manner (restoration of the physiological posterior femoral translation and femoral external rotation during knee flexion). Clinical and functional results have been analyzed according to the Knee Society scoring system with a minimal follow-up of 1 year. Statistical links between kinematic

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 data and Knee Society scores have been analyzed with an ANOVA test and a Spearman correlation test at a 0.05 level of significance. Results: 101 knees had a posterior femoral translation during flexion before and after TKR. 18 knees had a paradoxical anterior femoral translation during flexion before and after TKR. 51 knees had the preTKR paradoxical anterior femoral translation corrected to posterior femoral translation after TKR. 14 knees had the pre-TKR posterior femoral translation modified to a paradoxical anterior femoral translation after TKR. 91 knees had a femoral external rotation during flexion before and after TKR. 34 knees had a paradoxical femoral internal rotation during flexion before and after TKR. 50 knees had the pre-TKR paradoxical femoral internal rotation corrected to a femoral external rotation after TKR. 9 knees had the pre-TKR femoral external rotation modified to a paradoxical femoral internal rotation after TKR. There was a moderate statistical link between the reconstruction of a physiological kinematics after TKR and the Knee Society scores, with higher scores in the group of physiological kinematics after reconstruction. There was no correlation between the quantitative data and the Knee Society scores. Conclusions: To record the knee kinematics during TKR is feasible. This information might help the surgeon choosing the optimal reconstruction compromise. However, it is not well defined how to influence final kinematics during knee replacement. The exact influence of the quality of the kinematic reconstruction measured during surgery on the clinical and functional results has to be investigated more extensively.

P31-158 Leg alignment and tibial slope after minimal invasive total knee arthroplasty: a prospective randomized radiological study of intramedullary versus extramedullary tibial instrumentation (doi.org/10.1016/j.knee.2011.04.007) R.P.A. Janssen1, K.E. Kroon de1, S. Houterman2 1 Ma´xima Medical Center, Orthopaedic Surgery and Traumatology, Veldhoven, The Netherlands, 2Ma´xima Medical Center, Statistics, Veldhoven, The Netherlands Objectives: The purpose of the study was analysis of leg alignment and tibial slope comparing intramedullary versus extramedullary tibial instrumentation in the Genesis II MIS-TKA (Smith & Nephew, Memphis, USA). Methods: A prospective randomized study was performed according to the CONSORT guidelines. All patients (56 patients) for MIS-TKA were included. Randomization was performed by envelope selecting intra- or extramedullary tibia MIS instrumentation. All patients were operated by, or under supervision, of one experienced knee surgeon. Measurements of leg alignment and tibial slope were made on standardized long leg standing X-rays and lateral knee X-rays performed preoperatively and 4–12 months post surgery. Leg alignment was defined as being within or outside the range of 3˚ varus-valgus on the mechanical leg. The tibial slope was compared pre- and post-surgery. Intraclass correlation and inter-observer reliability was measured for leg alignment and tibial slope between examiners. Statistical significance was set at p \ 0.05. Results: Leg alignment In the intramedullary tibial outlining group, 3 patients had a persistent knee flexion contracture. They were excluded from final analysis for leg alignment. This occurred in 5 patients in the extramedullary tibial outlining group. Thus, 42 patients were included in the final leg alignment analysis: 23 patients in the intramedullary group, 19 patients in the extramedullary group. Postoperatively, the mean of leg alignment in the intramedullary group was 2.4 (sd 2.1) and in the extramedullary group 2.7 (sd 2.1). No significant difference in leg alignment was found between intramedullary or extramedullary tibial instrumentation in MIS-TKA using a mini-midvastus approach (p = 0.58). The intraclass coefficient (ICC)

S315 for alignment intra-observer reliability was 0.99. The ICC for alignment inter-observer reliability was 0.98. Tibial slope Posterior tibial slope was significantly better restored (in reference to pre-operative values) with use of extramedullary tibia instrumentation versus an intramedullary method (p = 0.001). The ICC for slope intra-observer reliability was 0.99. The ICC for slope inter-observer reliability was 0.98. Conclusions: There was no difference in leg alignment after MISTKA comparing intramedullary versus extramedullary tibial instrumentation. Restoration of tibial slope was significantly better with use of the extramedullary tibial instrumentation.

P31-179 Is chondromalacia patella contraindicated to unicompartmental knee arthroplasty? M.-H. Song1 1 Daedong General Hospital, Orthopaedic Surgery, Busan, Republic of Korea Objectives: The frequency of unicompartmental knee arthroplasty (UKA) is gradually rising together with advancement of minimally invasive technique and surgical instruments. Because of difficulties to define the exact indications for UKA, surgeons are afraid of the results of the operation. It is even classified as a contraindication of the surgery especially if the patient have chondromalacia patella. We report the results of pain and function of patello-femoral joint after UKA by analyzing prospectively. Methods: The target was 62 cases that were able to be followed up over 2 years and didn’t have any complications or underlying disease among 71 cases using Oxford Knee Phase 3 (Biomet, South Wales, UK) from May 2006 to March 2007. They were 55 cases of female and 7 cases of male and the mean age was 65 (49 * 78). For the clinical evaluation, we analyzed the pain and functional scores using Lonner’s Patellofemoral Scoring System before operation and 3 months, 6 months, 1 year and 2 years after operation. Results: The average preoperative pain score of patello-femoral joint was 19.9 and the score was improved 51.2 at 2 years follow ups. the average function score was improved from 14.2 to 38.3 at 2 years after the operation. The total score was improved from 34.2 preoperatively and changed to 89.4 at 2 years after the operation (P \ 0.01). At the last follow-up, the grade of the total scores were 36 cases (58%) of Excellent, 16 cases (26%) of Good, 10 cases (16%) of Fair and none of Poor. Conclusions: When the patient with osteoarthritis of knee is accompanied by chondromalacia patella, they can be classified as a contraindication for UKA. However, the authors prospectively analyzed changes of pain and function of patello-femoral joint after UKA, the result in short term follow up was satisfactory even in the case with chondromalacia patella. However, authors believe continuous and long term follow up would be necessary.

P31-288 New high flexion design does not improve maximal flexion in total knee arthroplasty: a randomized controlled trial J. Schimmel1, A. Wymenga2, K. Defoort2, G. van Hellemondt2 1 Sint Maartenskliniek, RD & E, Nijmegen, The Netherlands, 2Sint Maartenskliniek, Department for Orthopaedic Surgery, Knee Reconstruction Unit, Nijmegen, The Netherlands Objectives: An important factor in the functional results after total knee arthroplasty (TKA) is the achieved maximal flexion. To date, a TKA still provides dissimilar flexion capabilities compared to the healthy knee, which could be due to the mismatch between the geometric properties of the normal knee and the implant. The implant design of a new bi-cruciate stabilized knee system aims to replicate the normal knee function and intends to accommodate high flexion.

123

S316

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 either the high flexion or the conventional prosthesis. The primary outcome was maximum knee flexion after 1 year measured on a lateral X-ray with the patient supine and using manual force bending the knee. Secondary outcomes were: active flexion (supine and standing), Knee Society Score (KSS), Patella Scoring System (PSS), adverse events (AEs) and satisfaction until 2 years post-operative. Changes in KSS and PSS were calculated as: DKSS = KSS1yearKSSpre-op, and DPSS = PSS1year-PSSpre-op. Two-sided t tests and non-parametric alternatives were used to compare groups. Results: Demographic and flexion characteristics were comparable between the two groups at baseline. No significant differences were observed at 1 year in maximal flexion and active flexion between both groups (Fig. 1). Median maximal flexion on X-ray was 127 (range 83–150) for the high flexion system and 125 (range 81–145) for its predecessor. Active flexion was lower (p \ 0.001) than flexion on X-ray (both supine and standing, median 120). The DKSS and DPSS scores improved significantly more for the conventional group (p = 0.018 and p = 0.005, respectively). In the high flexion group, 40 AEs in 28 patients were reported, including 14 manipulations under anaesthesia (MUA), 10 persistent pain, 3 insert exchanges, and 2 total system revisions. In the conventional group, 13 AEs were observed in 12 patients, including 6 MUA, 3 persistent pain, and 1 insert exchange. Patients were equally satisfied in both groups. The 2-year follow-up will be complete at the time of the congress. Conclusions: The new high flexion design did not increase maximal flexion. In both groups, patients did not use the maximum of their flexion ability. Less improvement in clinical and functional outcomes and a higher number of AEs were observed in the high flexion group. The new design does not result in the desired increased flexion and better outcomes. It remains unclear if the disappointing clinical results were caused by a surgical or a design issue.

This study investigated maximal knee flexion of the high flexion system compared to the conventional system. Methods: In a prospective randomized controlled trial, a total of 124 patients presenting with non-inflammatory osteoarthritis received

123

P31-459 Is it safe to use unconstrained TKA in UKA revision surgery? P.E. Mu¨ller1, M.F. Pietschmann1, L. Wohlleb1, P. Weber1, V. Jansson1 1 Ludwig-Maximilians University, Campus Grosshadern, Department of Orthopaedic Surgery, Munich, Germany Objectives: The advantages of UKA in medial osteoarthritis include a faster and better functional recovery than total knee arthroplasty and less morbidity. Therefore a younger and more active population of patients has been undergoing UKA surgery in the last two decades. Despite these advantages, there remains an inevitable failure rate of UKA due to PE wear, loosening, and progression of osteoarthritis. The purpose of this study was to present the results for revision surgery from medial UKA to total knee arthroplasty (TKA) using a special technical strategy for dealing with tibial bone defects. Methods: Between 2000 and 2008, 30 pts. were treated in our institution with a TKA after failed UKA. In 28 pts. (93%) we used an unconstrained resurfacing TKA, in two cases (7%) a constrained revision implant was needed. We could include 21 pts. (75%) out of the 28 with an unconstrained TKA in our study (4 pts. died during follow up and 3 were lost to follow up). After a mean of 5.2 years we asked for patient’s satisfaction and pain. The WOMAC and Oxford Knee Score (OKS) were used to assess postoperative knee symptoms and function. Results: Reasons for failure were as follows: aseptic loosening in 22 pts (73%), progression of OA in 4 pts. (13%), a traumatic dislocation of the insert in 2 pts (7%), a instability (3%) and breakage of the implant (3%) in one patient each. We had to fill a bone defect in 18 tibial plateaux (64%). These defects were filled using a wedge of

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 cancellous bone from the lateral resection site. After revision surgery using a resurfacing TKA the mean flexion was 104 [range 80–135] while the two revision implants only showed a flexion of 80 and 90. After resurfacing TKA the patient’s satisfaction was excellent in 15 pts., in 4 pts. fair and in 2 pts. poor. The mean VAS for pain at rest was 1.7 and under weight bearing 3. The average WOMAC score for these patients 66.3 points and the OKS 32.1. In three cases another revision was needed. Conclusions: The use of an unconstrained resurfacing TKA for revision of failed medial UKA is a safe procedure, if planned thoroughly. The technique described in our study for filling medial tibial bone defects by using a cancellous bone wedge harvested from the lateral tibial resection is safe and reproducible. Early loosing during follow up was not found in our series. The advantages of this approach are a better functional outcome than with constrained TKA and more options for a possible future revision. We therefore recommend the use of unconstrained resurfacing TKA for revision of medial UKA with the use of autologous bone grafting.

P31-626 Is navigation a useful tool in unicompartmental knee arthroplasty? A meta-analysis of the current literature P. Weber1, A. Crispin2, S. Utzschneider1, M. Pietschmann1, V. Jansson1, P. Mu¨ller1 1 University Hospital Munich, (Ludwig-Maximilians-University), Department of Orthopedics, Munich, Germany, 2LudwigMaximilians-University Munich, Department of Medical Informatics, Biometry, and Epidemiology, Munich, Germany Objectives: One important factor for the long-term survival of unicompartmental knee arthroplasty (UKA) is the accuracy of implantation although controversy remains about the optimal implantation angles. Inaccurate implantation rates of up to 30% have been reported in cases using the conventional technique. Up to now several studies have investigated the role of navigation in UKA, all of them however with a limited number of patients and with inconsistent results. The aim of this meta-analysis was to compare risks of unsatisfactory outcomes in patients with navigated and conventional technique. Methods: An electronic search was performed in PubMed (1966 to October 2011) with the search terms unicondylar or unicompartmental or UKA and navigation. We identified a total of 55 papers, after the analysis of the title or the abstract, 38 papers were considered irrelevant. Of the remaining 17, 10 papers were finally eligible and included in the metaanalysis, with a total of 258 prostheses implanted with the navigated technique and 295 with the conventional one. One of these studies was a randomized controlled trial; all the others were observational cohort studies, two of which used a historical control group. The following items were analyzed: radiological positioning of the femoral and the tibial component in the a.p. and lateral view, radiological analysis of the tibiofemoral mechanical axis and the difference in operating time between the two groups. Relative risks (RR) were calculated from the reported percentages of implants outside the optimal ranges defined by the manufacturers or the study groups. Log relative risks were pooled by means of random effects models using the using the package ‘‘meta’’, version 1.6–1, under R, version 2.12.2 for Windows (R Foundation for Statistical Computing). Results: For all the analyzed radiological parameters, the RR of measurements outside the optimal ranges were less than 1 in the navigation group: the RR for suboptimal values for the mechanical axis was 0.39 (95% CI: 0.17–0.91), for the a.p. positioning of the femoral component 0.31 (95% CI: 0.16–0.59), for the lateral alignment of the femoral component 0.49 (95% CI: 0.28–0.86), for the a.p. positioning of the tibial component 0.50 (95% CI: 0.29–0.87) and for the lateral alignment of the tibial component 0.40 (95% CI: 0.23–0.69). The mean difference in operating times between the navigated and the conventional group was 15.4 min (95% CI: 10.2–20.6).

S317 Conclusions: This analysis showed that while the navigation technique results in longer operation times, UKA positioning is more precise. The limits defined by the manufacturers for an optimal positioning are not consistent and further studies regarding the optimal positioning of a UKA need to be performed. Our finding of a more precise UKA positioning needs to be confirmed by randomized controlled trials. P31-651 No effect of knee joint icing on knee-extension strength or knee pain shortly after total knee arthroplasty. A randomized crossover study B. Holm1, H. Husted2, H. Kehlet3, T. Bandholm4 1 Copenhagen University Hospital, The Lundbeckcenter for Fast-track Hip and Knee Arthroplasty, Hvidovre, Denmark, 2Copenhagen University Hospital, Department of Orthopaedic Surgery, Hvidovre, Denmark, 3Copenhagen University Hospital, Rigshospitalet, Section for Surgical Pathophysiology (4074), Copenhagen, Denmark, 4 Copenhagen University Hospital, Clinical Research Centre (136), Hvidovre, Denmark Objectives: The purpose of this study was to investigate the acute effect of knee joint icing on knee-extension strength and knee pain in patients shortly after total knee arthroplasty (TKA), as patients experience a great loss of knee-extension strength after surgery. Methods: Twenty patients (10 women) scheduled for primary unilateral TKA having a mean (SD) age, body mass, and height of 66 ± 12 years, 87.0 ± 16.6 kg, and 169.1 ± 17.1 cm, respectively, were consecutively recruited from a special unit for fast-track arthroplasty operations at a university hospital. Using a randomized cross-over study design, all patients were investigated on two different days—separated by 2 days—in the first week after the operation. One day they received 30 min of knee joint icing (active treatment) and the other day 30 min of elbow joint icing (control treatment). The order of treatments was randomized. Before and after both treatments we measured knee-extension strength (primary outcome) using a fixed hand-held dynamometer, and knee joint pain at rest and during maximal contractions using a 100-mm VAS scale. Both variables were normally distributed, and parametric tests were applied. Paired Samples T Tests were used to examine between-treatment differences in knee-extension strength and knee joint pain. Results: Knee joint icing had no acute effect on knee-extension strength (Pre knee ice: 0.24 Nm/kg, post knee ice: 0.23 Nm/kg, pre elbow ice: 0.27 Nm/kg, post elbow ice: 0.24 Nm/kg, P = 0.501). Knee joint pain at rest decreased after both interventions (P \ 0.018), but the decrease was not different between treatments (Knee icing decrease: 11 mm, elbow icing decrease: 8 mm, P = 0.749). Knee pain during maximal contractions showed no change after both interventions. Conclusions: Cooling the knee joint for 30 min in the first week after TKA, where knee pain, swelling, and inflammation are pronounced, has no acute effect on knee-extension strength, which is contrary to that observed in experimental knee effusion models. Cooling the knee showed no effect on knee pain, but the similar reduction in knee pain at rest after both interventions may indicate that the maximal contractions performed during the knee-extension strength measurements at both days is modulating knee pain.

P31-667 Joint line reconstruction in medial unicompartmental knee arthroplasty: development of a measurement method and comparison of 2 prostheses with different gap management P. Mu¨ller1, P. Weber1, C. Schro¨der1, S. Utzschneider1, M. Pietschmann1, V. Jansson1

123

S318 1

University Hospital Munich, (Ludwig-Maximilians-University), Department of Orthopedics, Munich, Germany Objectives: The UKA Type Oxford (Biomet Warsaw, USA) in the mobile-bearing- design has 10-years results of 98%. The minimally invasive implantation leads to a quicker rehabilitation and less pain. However the experience with UKA is less than with the TKA and especially improvements in the area of biomechanics can be obtained. The conservation of the joint line in UKA is an important factor as a distalisation of the medial joint line with conservation of the lateral leads to a considerable change in knee kinematics. The newly designed UKA ‘‘Univation’’ (Aesculap, Tuttlingen, Germany) in the mobile-bearing design is similar to the Oxford UKA in a lot of areas with a spheric condylar surface. As a difference the gap management is performed with changing in the tibial slope. The femoral implant is positioned anatomically. This should allow a better conservation of the joint line. The aim of this study was to evaluate radiologically the changings in the joint line in both UKAs. As there is no established method to evaluate the changing of the joint line radiologically we introduced two methods and correlated them. Methods: We analyzed retrospectively the X-rays of patients with a medial UKA type Univation and type Oxford. There were 30 patients in each group. The changing of the medial joint line from pre- to postoperative was measured with two different methods. First we evaluated the correlation of the two methods in each group. In a second step we compared the changings of the joint line between the 2 groups (Oxford/Univation). Results: The two methods showed a good correlation for the two groups (Univation: 0.87, p \ 0.0001, Oxford: 0.82, p \ 0.0001, Wilcoxon signed rank test). The joint line in the Univation was distalised in the mean by 2.08 mm (method 1: 2.13 mm, method 2: 2.03 mm, range each: 0–5 mm), in the Oxford the mean distalisation was 4.53 mm (methode 1: 4.43 mm, methode 2: 4.62 mm, range each 2–7 mm). The difference was statistically significant (p \ 0.0001, Mann–Whitney-Test). Conclusions: We developed two methods to measure the change in the joint line after UKA and showed that a different gap management lead to a different joint line reconstruction. Further studies need to show if this leads to a better kinematics with better long-term results, a lower load in the lateral compartment and on the ligaments.

P31-896 Lessons learned with 226 consecutive journey knee replacements - learning curve and complication rate B. Christen1, E. Aghayev2, B. Rieger3, M.S. Neukamp2 1 Orthopa¨dische Klinik Bern, Salemspital, Bern, Switzerland, 2 MEM Research Center, Bern, Switzerland, 3Kantonsspital Bruderholz, Orthopa¨dische Klinik, Bruderholz, Switzerland Objectives: New designs in total knee replacement are expected to lead to better function including more flexion due to better biomechanics. Whereas it is still not proven the question remains if more aggressive kinematics in TKR could lead to more complications and a flat learning curve. Methods: In 196 patients (126 female, 70 male) with a mean age of 67.7 years with gonarthritis 226 Journey BCS total knee arthroplasties (Smith and Nephew, Memphis, USA) were implanted between December 2006 and May 2011. All operations were performed by one single surgeon. The first 102 cases were operated with a conventional instrumentation (group 1). Tibial aiming was performed extra-, femoral one intramedullary. The following 124 knees (group 2) were operated with a CT-less computer navigation (PI Galileo, Smith and Nephew, Aarau, Switzerland) for the tibial and distal femoral cut. In both groups the femoral rotation was defined with a tensioner taking

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 in account the classical bony landmarks. All knees were controlled clinically and radiologically 2, 4 and 12 months postoperatively. Results: The mean flexion increased from in mean 111 to 126.2 and 130.3 2, 4 and 12 months respectively. The mean KSS was 115.6 points preoperatively and 174 points after 1 year. In group 1 18 (17.6%) complications had to be registered, in group 2 11 (8.9%). These consisted in 7 friction of the iliotibial band, 5 stiff knees (flexion \ 90), 3 knee dislocations, 2 loosenings of the patellar button, 2 infections, 1 periprosthetic femur fracture, 1 avulsion of the tibial tuberosity and 1 wrong side polyethylene. Of the complications 17 and 9 knees respectively had to be revised. In 6 cases in group 1 and 3 in group 2 major revisions defined as hardware revision were necessary. Complication rate was overall 4.0% in mean 2.3 years or 1.73/100 TKR years. Conclusions: With a Journey BCS knee one can expect a mean flexion of 130 1 year after surgery. The complication rate overall was 12.8%, and could be almost bisected in group 2. The rate of major (hardware) revision remained unchanged, which is comparable to the literature and higher than in conventional TKR systems. CT-less navigation seems not to improve complication rate in BCS TKR. The theoretically excellent Journey BCS knee kinematics seems to lead to a less forgiving system which leads to a flat learning curve and an increased complication rate.

P31-1083 Is indication for revision TKA a determinant for outcome? A cohort of 224 patients with a minimum follow-up of 2 years J. Schimmel1, R. van Kempen2, H. Vandenneucker3, A. Wymenga2, G. van Hellemondt2, K. Defoort2 1 Sint Maartenskliniek, RD&E, Nijmegen, The Netherlands, 2 Sint Maartenskliniek, Department for Orthopaedic Surgery, Knee Reconstruction Unit, Nijmegen, The Netherlands, 3UZ Pellenberg KU Leuven, Department of Orthopaedics, Knee and Sports Medicine, Pellenberg-Lubbeek, Belgium Objectives: Worldwide, ever growing amounts of revision arthroplasties of the knee are performed. It is well known that only when a treatable cause for a malfunctioning total knee arthroplasty (TKA) is found, a revision TKA procedure can be performed. Several indications for revision TKA can be distinguished. Little is known about the functioning, pain and satisfaction after revision TKA in the different indication groups. Therefore, we started a prospective cohort study to analyse the outcome of revision TKA. Methods: We included all patients that received a revision TKA, using the Genesis II or Legion revision system, and were treated at the Sint Maartenskliniek or the University Hospital Pellenberg since 2004. Patients with prosthetic joint infections were treated two-staged with an antibiotic loaded spacer, all others were treated in a one stage procedure. Peroperatively the main indications were determined: aseptic loosening, malposition, infection, instability and arthrofibrosis. We evaluated pre-, and postoperatively at predetermined intervals KSS scores, VAS pain, VAS patient satisfaction and adverse events (AE). For this presentation, we evaluated all patients with a minimal followup of 2 years (n = 217) and a cohort (n = 51) with 5-year follow-up. Descriptive statistics were performed for the indication groups. Results: Of the 217 included patients, 204 reported a complete 2-year follow-up (94%). KSS clinical and functional scores showed an improvement of 25 points on average and the VAS pain was reduced with at least 12 points for the total group. Patients were highly satisfied (median score [ 73 in all groups), except the arthrofibrosis group (median 48.5 (15–98)). The arthrofibrosis group performed significantly worst on KSS clinical and VAS pain score. The aseptic

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 loosening group showed better outcomes in comparison to the instability, malposition and infection group. After 5 years, the KSS scores remained at the same level, while the VAS pain score did show a small improvement compared to 2-years. We reported 178 device related AEs in 96 patients, of whom five patients underwent a rerevision, two patients an arthrodesis and one patient an amputation. Conclusions: Further evaluation is still in progress and we hope to have 250 patients with a 2-year and 80 with a 5-year follow-up by the time of the congress. We can conclude that for the whole revision TKA group, an improvement in clinical scores after surgery can be expected. The arthrofibrosis group performed worse on all outcome measures, while the aseptic loosening group performed best. Reoccurrence of the primary reason for revision was responsible for most of the re-interventions in the different indication groups. The relation between indication and clinical outcome after revision TKA gives the surgeon the opportunity to better estimate the outcome of a revision TKA in a certain diagnosis group, thereby providing better patient information and counselling.

P31-1233 Malalignment: can thin sequentially cross-linked tibial inserts pass the test? R. Streicher1, L. Korduba2, A. Essner2, R. Delanois3, M. Mont3, S. Sayeed3 1 Stryker SA, Montreux, Switzerland, 2Stryker, Mahwah, United States, 3Sinai Hospital of Baltimore, Baltimore, United States Objectives: Historically, there has been concern with high stresses placed on thin polyethylene bearings used in total knee arthroplasties. An 8-mm thick bearing has been reported to be the thinnest tibial bearing that can be used which maximizes conformity, thickness, and material. Sequentially cross-linked and annealed polyethylene has been shown to reduce the wear rate significantly compared to conventional ultra high molecular weight polyethylene. Compared to well aligned tibial components, varus placement of tibial components of some devices has been shown to increase the wear of polyethylene bearings and lead to early failure of total knee arthroplasties. The purposes of this study were to analyze the effect of tibial insert thickness, subjected to malaligned conditions, on wear performance and to evaluate fatigue failure during cyclic testing. Methods: A 6-station MTS knee simulator was used. Cobalt-chromium femoral and tibial components were mated against polyethylene bearings. The polyethylene inserts were manufactured from GUR 1020 UHMWPE that was sequentially annealed and irradiated three times and then gas plasma sterilized (X3, Stryker Orthopaedics). The insert thicknesses tested were either company labeled 7- or 9-mm, which corresponded to true nominal polyethylene thicknesses of 4- and 6-mm, respectively. Normal gait kinematics, at 5 of joint line malalignment to achieve unicondylar loading, were used for testing following ISO 14243-3. The lubricant used was Alpha Calf Fraction serum diluted to 50%. The serum solution was replaced and inserts were weighed for gravimetric wear at least every 0.5 million cycles. Soak control specimens were used to correct for fluid absorption with weight loss data converted to volumetric data. Results: The total volume loss at 1 million cycles for polyethylene thicknesses of 4- and 6-mm was 11.3 ± 3.9 mm3 and 9.6 ± 2.4 mm3, respectively. The volumetric wear rate for polyethylene thicknesses of 4- and 6-mm was 12.1 ± 4.8 mm3/106 cycles and 10.0 ± 3.4 mm3/ 106 cycles, respectively. There was no statistical difference in wear between the two polyethylene thicknesses (p value = 0.121). Visual inspection of the inserts, utilizing the quadratic grading system, showed wear scars in areas of contact including burnishing, striations and occasional scratches. There was no evidence of functional failure after 1 million cycles.

S319 Conclusions: The results of this study demonstrate that malaligned, sequentially cross-linked and annealed polyethylene provides the opportunity for reducing thickness while maintaining desired material properties, such as wear performance. The differences in the volumetric wear rate for 4-mm thick polyethylene have been shown to not be statistically significant from 6-mm thick polyethylene during malaligned testing. In addition, no fatigue failure of the polyethylene was observed. The results of this study may initiate the discussion into what the ‘‘magic number’’ for polyethylene thickness really is.

P31-1258 Is the posterior offset of femoral condyles related to the sagittal tibial slope? G. Cinotti1, P. Sessa1, F.R. Ripani1, A. Della Rocca1, W. Salustri1 1 Orthopaedic Clinic University La Sapienza, Rome, Italy Objectives: Aims of the study were to assess the sagittal tibial slope and posterior offset of femoral condyles and whether any correlation between the two is present. In particular, we hypothesized that the degree of posterior tibial slope is related to the posterior offset of femoral condyles to achieve adequate joint motion and stability during the flexion of the knee. Methods: We evaluated the magnetic resonance studies of the knees of 80 subjects, 45 males and 35 females, with an average age of 38.9 years, to assess the variability of posterior condylar offset and sagittal tibial slope and whether any correlation between the two is present. Measurements were performed by two independent observers using an imaging visualization software. The tibial slope was calculated on the medial and lateral side as the angle between the sagittal longitudinal axis and a line tangent to the medial and lateral plateau. The posterior condylar offset was calculated as the distance between the posterior femoral cortex and the most posterior point of the femoral condyle. Results: The tibial bone slope averaged 8 and 7.7, on the medial and lateral side, respectively, without significant difference between the two. The posterior offset of femoral condyles averaged 27.4 and 25.2 mm on the medial and lateral side, respectively (p = 0.001). The variation coefficient of posterior condylar offset and sagittal tibial slope was 11.5% and 38%, respectively. A significant relationship was found between the posterior condylar offset and sagittal tibial slope, but on the medial side only. Conclusions: In planning a cruciate retaining TKA, it may be helpful to take into account the posterior condylar offset and tibial slope concomitantly to achieve a satisfactory functional outcome.

P31-1267 Kinematic analysis of a three condyles prosthesis design F. Colle1, S. Bignozzi2, N. Lopomo3, S. Zaffagnini4, M. Marcacci4 1 Istituto Ortopedico Rizzoli, University of Bologna, Laboratorio di Biomeccanica e Innovazione Tecnologica, Bologna, Italy, 2Rizzoli Orthopaedic Institute, Biomechanics Lab, Bologna, Italy, 3Istituto Ortopedico Rizzoli, Laboratorio di Biomeccanica, Bologna, Italy, 4 Istituto Ortopedico Rizzoli Bologna, Bologna, Italy Objectives: To understand the effect of the third condyle of this prosthesis design on passive kinematics of the knee. Methods: This work analyses the effect of two different prosthesis design on the kinematic behaviour of the knee with particular attention to the antero-posterior (AP) translation and the intra-extra (IE) rotation of the tibia during a range of motion between 0 and 120. The two prosthesis under analysis are the Link Gemini, cruciate retaining design, and the Tornier Noetos with the third condyle innovation. Kinematic analysis was performed with a surgical navigation system and data were collected intraoperatively. Results: Both prosthesis design didn’t show any statistical differences between preoperative and postoperative values both in AP translation

123

S320 and IE rotations but the postoperative curves of the Tornier Noetos design showed more similarity with preoperative curves. Conclusions: The two different types of prosthesis design showed similar postoperative AP translation and IE rotation curves. Despite both prostheses didn’t show significant differences with respect to preoperative conditions, the Tornier design seemed to have a kinematic curve more similar to preoperative one. This may indicate a trend to sustain the proper kinematic features of the patient.

Knee - total joint replacement III P32-37 Outcomes in the elderly following total knee arthroplasty S. Miyamoto1, M. Kosugi1, S. Sasaki1, H. Kojin1, J. Suga1, T. Hamahata1 1 Kohsei Chuo General Hospital, Orthopaedic Surgery, Tokyo, Japan Objectives: The average life expectancy is rising in Japan and the subsequent increase in the elderly has resulted in a relative increase in the age at which total knee arthroplasty (TKA) is considered appropriate. At our hospital, we used tourniquets during TKA procedures until March 2010. Since May 2010, we have prevented postoperative hemorrhage by administering without using a tourniquet. In this comparative study, we investigated TKA outcomes from the perioperative period until hospital discharge, including whether a tourniquet was used, in patients aged around 76 years at the time of surgery. Methods: The study involved 114 knees replaced through TKA, which were divided into group A comprising 64 knees in subjects aged \ 76 years [(a) 37 knees replaced with TKA + tourniquet and (b) 27 knees replaced with tourniquet-less TKA] and group B comprising 50 knees in subjects aged C 76 years [(a) 27 knees replaced with TKA + tourniquet and (b) 23 knees replaced with tourniquet-less TKA]. The average age of the study subjects was 68.9 years in group A and 80.3 years in group B. We investigated the following parameters: duration of surgery; blood loss volume during surgery; differences in hemoglobin (Hb) and hematocrit (Ht) levels before surgery, immediately after surgery, and at 1 and 7 days after surgery; range of motion (ROM) and Knee Society Score (KSS) before surgery and at hospital discharge; number of days before starting stairclimbing exercises; and length of hospitalization. Results: Duration of surgery was 96.6 min in group A and 95 min in group B. Blood loss volume during surgery was 222.6 mL in group A and 190.5 mL in group B. Hb levels were 1.3 g/dL in group A and 1.4 g/dL in group B before and immediately after surgery, 1.9 g/dL in group A and 2.1 g/dL in group B at 1 day after surgery, and 1.7 g/dL in group A and 2 g/dL in group B at 7 days after surgery. Ht levels were 4% in group A and 4.8% in group B before and immediately after surgery, 5.4% in group A and 7.5% in group B at 1 day after surgery, and 5.2% in group A and 4.5% in group B at 7 days after surgery. ROM in group A was 7.7/121.5 before surgery and 2.5/ 108.4 after surgery, while ROM in group B was 7.1/119.3 before surgery and 2/110 after surgery. KSS in group A was 53.3 points before surgery and 75.1 points at hospital discharge, while KSS in group B was 49.9 points before surgery and 73.4 points at hospital discharge. The number of days before starting stair-climbing exercises was 18.3 days in group A and 19 days in group B. The length of hospitalization was 34 days in group A and 36.8 days in group B. Conclusions: In both groups, the ROM and KSS before surgery and at hospital discharge showed an improvement after surgery, with a significant improvement observed in study subjects aged 76 years or older where the knee was replaced with tourniquetless TKA. TKA in the elderly is appropriate at higher ages if a

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 cautious approach is taken and a tourniquet-less procedure is also possible.

P32-76 Residual varus alignment does not compromise results of TKAs in patients with preoperative varus E. Servien1, R. Magnussen2, S. Lustig3, G. Demey4, P. Neyret5 1 Hopital de la Croix-Rousse, Centre Albert Trillat, Hospices Civils de Lyon, Lyon University, Lyon, France, 2Hopital de la Croix-Rousse, Centre Albert Trillat, Lyon, France, 3Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 4Centre Albert Trillat, Lyon-Caluire, France, 5Hopital Croix-Rousse, Centre Livet, Chirurgie Orthopedique, Caluire-et-Cuire, France Objectives: Postoperative varus alignment has been associated with lower IKS scores and increased failure rates. Appropriate positioning of TKA components therefore is a key concern of surgeons. However, obtaining neutral alignment can be challenging in patients with substantial preoperative varus deformity and it is unclear whether residual deformity influences revision rates. We asked: (1) in patients with preoperative varus deformities, does residual postoperative varus limb alignment lead to increased revision rates or lower IKS scores compared with correction to neutral alignment, (2) does placing the tibial component in varus alignment lead to increased revision rates and lower IKS scores, (3) does femoral component alignment affect revision rates and IKS scores, and (4) do these findings change in patients with at least 10 varus alignment preoperatively? Methods: From a prospective database, we identified 553 patients undergoing TKAs for varus osteoarthritis. Patients were divided into those with residual postoperative varus and those with neutral postoperative alignment. Revision rates and International Knee Society (IKS) scores were compared between the two groups and assessed based on postoperative component alignment. Survival analysis was conducted with revision as the endpoint. The analysis was repeated in a subgroup of patients with at least 10 preoperative varus. Minimum followup was 2 years (median, 4.7 years; range, 2–19.8 years). Results: The two groups had similar survival rates to 10 years and similar IKS scores. Varus tibial component alignment and valgus femoral component alignment were associated with lower mean scores. Revision rates and scores were similar in a subgroup of patients with substantial preoperative varus. Conclusions: Our data suggest residual postoperative varus deformity after TKA does not increase survival rates at medium-term in patients with preoperative varus deformities, providing tibial component varus is avoided. Tibial component varus negatively influences IKS score.

P32-83 The accuracy of lower extremity alignment in total knee arthroplasty using navigation system: data analysis of 661 cases H.G. Park1, J.-H. Lee1 1 Dankook University Hospital, Orthopedic Surgery, Cheon An, Republic of Korea Objectives: To evaluate the accuracy of the alignment of lower extremity in 661 cases of total knee replacement arthroplasty (TKA) using navigation system. Methods: From June 2006 to September 2008, 661 cases (431patients) of TKA using navigation system were operated. To analyze the mechanical axis, the weight bearing full length lower extremity radiographs were taken preoperatively and 3 weeks after the operation. The results from a well- experienced surgeon (423 cases) were

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 compared with those from a less-experienced surgeon (238 cases), and they both used the navigation. Results: The mean of mechanical axis was -13.3 (range: -33.3–10.6) preoperatively, but it was corrected as -2.0 (range: -14.3–7.5) after TKA using navigation. There was no significant difference between the mean, -1.8 (range: -13.4–6.8) by a well-experienced surgeon and the mean, -2.2 (range: -14.3–7.5) by a less-experienced one. Conclusions: According to the radiologic results, the navigation is beneficiary for the accuracy of mechanical axis in TKA. The navigation system helps a less-experienced surgeon increase the accuracy of lower extremity alignment.

P32-109 Outcomes of unicondylar knee athroplasty in patients with severe deformity: a minimum 2 year follow up study C. Seng1, D. Ho1, P.L. Chin2, N.N. Lo1, S.-L. Chia1, S.J. Yeo1 1 Singapore General Hospital, Orthopedic Surgery, Singapore, Singapore, 2Singapore General Hospital, Singapore, Singapore Objectives: The classical indications for UKA include fixed flexion deformity (FFD) \15 and varus deformity \15. Our study is to investigate if patients with more severe angular deformity were able to achieve satisfactory functional and quality of life outcomes with UKA. This would help broaden the indications for UKA, and better justify its use as an alternative to HTO or TKA in patients with medial unicompartmental knee osteoarthritis. Methods: 647 patients underwent UKA from 2004 to 2008 from our hospital joint registry. There were 40 patients with FFD [ or equals 15, 13 patients with varus deformity [ or equals 15 and 594 patients in the control group (neutral or varus/FFD \ 15). We compare the flexion range, Knee society scores (KSS), Oxford knee scores (OKS) and SF-36 Quality of life (QOL) scores preoperatively to 2 years. The Preservation UKA implant and MillerGallante implant were utilised in this study. Results: All 3 groups were well matched in terms of age, sex, BMI, site of surgery and type of implants. Patients with FFD or varus B15 reported lower flexion angle post-operatively compared to the control group which is not clinically significant (123.7 vs. 129.5, p = 0.009; 125.2 vs. 129.5, p = 0.109). Patients with FFD [ or equals 15 have poorer social and role functioning scores of SF-36 that is statistically significant. They have comparable outcomes to the control group in the KSS, OKS and rest of SF-36. Patients with varus [ or equals 15 have comparable outcomes to the control group in the KSS, OKS and all 8 components of SF-36 (QOL). There are 8 revisions in the control group at 34 months postoperatively and none in the patients with severe deformity (survivorship 98.76%). Conclusions: Carefully selected patients with severe deformity that underwent UKA, fared well post operatively at a minimum of 2 years follow up, with results that were largely comparable to the control group of patients. Hopefully our current results will lead to a more liberal approach towards considering UKA for patients with more severe deformity, and eventually lead to bigger and newer studies that can give us a more definitive conclusion on the long term outcomes of patients with severe deformity who undergo UKA.

P32-143 Oxford knee questionnaire. Is information identically informative before and after knee replacement? J.-Y. Jenny1, P. Louis1 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France Objectives: Evaluation of the results after knee replacement may be performed with quality-of-life questionnaires, which may be generic

S321 or knee-specific. The Oxford knee questionnaire has been validated as a knee-specific, self-administered questionnaire for patients with osteoarthritis waiting for knee replacement. The validity of a questionnaire (understanding and pertinence of questions and answers, internal and external validity) and its discriminating power (ceiling and floor effects) must be assessed independently before and after the surgical procedure. Methods: 200 patients have been analyzed: 100 patients on the waiting list for total knee replacement (TKR) and 100 patients already operated for TKR with more than 1 year follow-up. There were 139 women and 61 men with a mean age of 69.7 years. The questionnaire was given to the patient during the pre-surgery visit or during a post-operative control visit, and was filled up by the patient independently from the medical staff. The clinical and functional Knee Society scores were also assessed by a member of the medical staff at the same time. Data were analyzed independently for pre-operative and post-operative visits. The Oxford score was compared to the Knee Society scores on an individual basis. The same analysis was performed for the sub-scores ‘‘pain’’, ‘‘gait’’ and ‘‘knee flexion’’ of both evaluation tools. The comparison was performed by calculation of the linear correlation coefficient and of the Spearman correlation coefficient. Internal validity was assessed by calculation of the alpha Cronbach coefficient. Presence of ceiling effect (for poorer results) and floor effect (for better results) was assessed by the number of cases with the maximal score minus one SD (ceiling) or the minimal score plus one SD (floor). Results: The Cronbach coefficient was 0.88 before surgery and 0.66 after surgery, showing a satisfactory internal consistency. There was a significant negative correlation between the Oxford score and clinical and functional scores of the Knee Society for a given patient, either before or after surgery. This correlation was stronger before surgery than after surgery. Before surgery, there was no floor effect (no score less than 19); there was a moderate ceiling effect (7 scores over 53). After surgery, there was an important floor effect (36 scores less than 19); there was no ceiling effect (no score over 53). Conclusions: The Oxford knee questionnaire is a valuable evaluation tool for knees before and after knee replacement. Before surgery, the absence of any ceiling and floor effects allows an excellent discrimination power. After surgery, the discriminating power is lower for the better results because of the presence of a significant floor effect.

P32-173 Targeted thromboprophylaxis in knee arthroplasty: an audit of 5,544 patients G. Cox1, A. Pearce1, J. Britton1, N. Thomas1, A. Wilson1, N. Rossiter1 1 North Hampshire Hospital, Department of Orthopaedics, Basingstoke, United Kingdom Objectives: Thromboprophylaxis in knee arthroplasty is topical, controversial and extremely varied. It is our practice to employ a targeted strategy with all patients receiving mechanical thromboprophylaxis (A-V Impulse SystemTM, INTAVENT ORTHOFIX LIMITED, Maidenhead, Berkshire, UK), unless contraindicated and only high-risk patients (10%) receiving additional chemical treatment. Our objective was to audit this protocol with respect to venous thrombotic event (VTE) rate in 5,544 patients undergoing knee arthroplasty. Methods: Eleven year data was extracted from our local database (01/ 10/1999–01/10/2010). This included 5,265 primary operations (4,383 total-knee-replacements (TKRs), 754 unicondylar-knee-replacements (UKR), 128 patella-femoral-joint-replacements (PFJR)) and 279 revision TKRs. Patients were followed up in a dedicated joint review clinic at 2, 6, 12, and 52 weeks post-arthoplasty and then yearly. Outcomes measured were;

123

S322 1. symptomatic DVT; 2. symptomatic PE; 3. post-operative complication. Results: Our VTE rates were as follows; symptomatic DVT 1%; symptomatic PE 0.49%; and fatal PE 0.02%. Post-operative haematoma rate requiring surgical evacuation was 0.40%; and 0.07% of patients required post-operative blood transfusion. There were no statistically significant differences in these rates when patients were stratified for type of arthroplasty/level of risk. Conclusions: The VTE rate in our series is comparable to recent studies where all patients have received chemical thromboprophylaxis. We believe that our targeted strategy, where mechanical thromboprophylaxis alone is used in the majority of our patients is effective, results in lower morbidity and at reduced cost.

P32-200 Physical activity and outcomes thirty months after revision total knee replacement M. Nu´n˜ez1, E. Nun˜ez2, S. Sastre3, L. Lozano3, J.M. Segur3, A. Saulo´4 1 Hospital Clinic i Universitari de Barcelona, Rheumatology Department. IDIBAPS. Knee Unit, Barcelona, Spain, 2Institut Catala` de la Salut, SAP Suport al Diagno`stic i al Tractament, Barcelona, Spain, 3Hospital Clinic i Universitari de Barcelona, Knee Unit, Barcelona, Spain, 4Hospital Clinic i Universitari de Barcelona, Rheumatology Department, Barcelona, Spain Objectives: Common outcome variables after total knee replacement (TKR) are decreased pain, improved limb function and the patient’s return to normal life, including physical activity. Our hypothesis was that although functional outcomes improve after revision TKR, leading an active life, including regular exercise and remaining seated for less time leads to greater improvements. The objectives of this study were to assess health outcomes in OA patients after revision TKR and identify the influence of physical activity on functional capacity. Methods: Prospective study with 30 months follow-up. Sociodemographic and clinical variables were recorded. Outcome measures were the Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC], the Short Form 36 [SF-36]). The WOMAC function and SF-36 physical function dimensions were used to assess functional capacity. Physical activity, including regular physical exercise C3 times per week for C30 min and remaining seated B6 h per day were recorded. Patient satisfaction with the results after TKR, including improvements in pain and function, was collected. Linear regression models were employed to analyze associations. Results: 60 patients (72% female, mean age 71.8 [SD 7.5] years, BMI 30.8 (SD 5.1), number of comorbidities 4.8 (SD 2.2)) completed follow-up data. 32% were physically active. There were significant differences between pre- and postoperative total WOMAC scores (p \ 0.001), the SF-36 physical component summary (PCS) (p = 0.006) and mental component summary (MCS) (p = 0.006). The effect size was 1.75 for total WOMAC scores, 0.63 for the PCS and 0.39 for the MCS. Being physically active was significantly associated with better WOMAC function dimension scores (dependent variable) (coefficient -22.2 [p \ 0.001]) and better SF-36 physical function dimension scores (dependent variable) (coefficient 25.3 [p \ 0.001]). In the two models, which included age, sex, and the number of comorbidities as potentially confounding variables, physical activity explained 37% and 27% (R2 change), respectively, of the variability. A total of 17 of patients said they would not submit to further surgery due to unsatisfactory results or excess suffering. Conclusions: After revision TKR, physical activity and exercise were significantly associated with better function. The WOMAC questionnaire showed greater sensitivity in detecting improvements in health than the SF-36 questionnaire.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P32-275 Patellar tilt and thickness in a high-flex design total knee arthroplasty: do they influence the postoperative flexion? P. Verdonk1, A. Dhollander2, D. Bassens2, J. Victor3 1 Gent University Hospital, Orthopaedics, Gent, Belgium, 2Gent University Hospital, Orthopaedic Surgery and Traumatology, Gent, Belgium, 3AZ St Lucas Brugge, Department of Orthopedics, Brugge, Belgium Objectives: The purpose of this short-term study was to determine whether patellar thickness and tilt influence the postoperative flexion in a high-flex design total knee arthroplasty (TKA). Methods: Between 2007 and 2009, 113 patients underwent surgery for TKA using RP-F prosthesis (DePuy, Warsaw, Ind). All patients were evaluated preoperatively and at 12 months of follow-up by determination of the range of flexion. Standard standing antero-posterior and lateral weight-bearing radiographs were made. The patellofemoral joints were evaluated in skyline views taken with the knees at approximately 60 of flexion. Patellar thickness and tilt were measured. Results: The mean flexion observed before surgery was 124.3 ± 18.1 and after 1 year was 127.7 ± 12.9. The mean patellar thickness preoperatively was 24.4 ± 2.8 mm and 26.0 ± 3.4 mm at 12 months after surgery. The mean patellar tilt before the procedure was 2.66 ± 4.43 and after 12 months of follow-up was -0.98 ± 5.07. No significant correlation was found between postoperative patellar tilt/thickness and postoperative flexion. Conclusions: Patellar tilt and thickness after TKA are factors that depend on the surgery. The resection of the patella can influence both patellar thickness and patellar tilt. By developing adequate tools, it would be possible to avoid the occurrence of an exaggerated patellar tilt or a major difference in patellar thickness. However, these two factors did not seem to influence the postoperative flexion in a highflex design TKA, which can be seen as one of the most important outcome factors after TKA.

P32-368 The relationship between the posterior condylar offset and the flexion balance in TKA V.J. Leo´n Mun˜oz1, A.J. Liso´n Almagro1, J.M. Gonza´lez Garro2 1 Hospital de la Vega Lorenzo Guirao, Department of Orthopedics, Cieza (Murcia), Spain, 2Hospital de la Vega Lorenzo Guirao, Department of Radiology, Cieza (Murcia), Spain Objectives: Total knee arthroplasty (TKA) is performed primarily to relieve pain, but the post-operative range of movement is an important aspect of outcome. High-flexion posterior-stabilized prostheses have been introduced with increased offset of the posterior femoral condyle intended to allow more flexion. The aim of the study is to determine the relationship among the posterior condylar offset (PCO) and the flexion balance in different types of TKA. Methods: Postoperative PCO was evaluated by two independent observers on true lateral radiographs by measuring the maximal thickness of the posterior condyle, projected posteriorly to the tangent of the posterior cortex of the femoral shaft in 220 TKA radiographic controls (66 ultracongruent (UC) implants, 60 standard posteriorstabilized (PS) prostheses and 94 high flexion PS TKA). The posterior condylar offset ratio (PCOR), which is the ratio of the PCO to the diameter of the femur, 2.5 cm above the termination of the supracondylar flare on true lateral views was also determined. The range of movement was determined using standard (38 cm) clinical goniometers before operation and at review. Results: UC TKA: PCO observer 1: 29.52 ± 3.68 mm; PCOR observer 1: 0.888 ± 0.12; PCO observer 2: 28.73 ± 3.35 mm; PCOR observer 2: 0.879 ± 013. Standard PS TKA: PCO observer 1: 28.83 ± 315 mm; PCOR observer 1: 0.845 ± 012; PCO observer 2:

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 28.72 ± 2.96 mm; PCOR observer 2: 0.851 ± 0.11. High-flexion PS TKA: PCO observer 1: 31.65 ± 3.62 mm; PCOR observer 1: 0.973 ± 0.13; PCO observer 2: 31.24 ± 3.47 mm; PCOR observer 2: 0.972 ± 0.14. Active maximal flexion (12 months after implantation): UC TKA 112.14 ± 10.54, Standard PS TKA: 111.96 ± 10.86 and High-flexion PS TKA: 112.31 ± 11.36. We have obtained a statistically significant difference in the comparison between the postoperative PCOR attending to gender. We have obtained a statistically significant difference in the comparison between the postoperative PCO and PCOR of high-flexion PS implants and UC and standard PS implants. No significant difference among UC and conventional PS knees was obtained. Attending to the active maximal flexion, no significant difference among UC, conventional PS knees and high-flexion PS implants was obtained. No significant correlation among the postoperative PCO and PCOR values and the active maximal flexion was obtained. Reproducibility among the two observers was assessed according to Bland and Altman’s statistical method. Conclusions: Changes in posterior femoral condylar offset are proposed to have an influence on range of motion. Our analysis suggested that high-flex knee prostheses do not increase the post-operative maximum knee flexion compared with conventional PS implants or UC implants and no significant correlation among the postoperative PCO and PCOR values and the active maximal flexion was obtained.

P32-369 The influence of the instrumentation system in the postoperative posterior condylar offset of TKA V.J. Leo´n Mun˜oz1, A.J. Liso´n Almagro1, A´. Esca´mez Pe´rez1, J.M. Gonza´lez Garro2 1 Hospital de la Vega Lorenzo Guirao, Department of Orthopedics, Cieza (Murcia), Spain, 2Hospital de la Vega Lorenzo Guirao, Department of Radiology, Cieza (Murcia), Spain Objectives: Changes in posterior femoral condylar offset are proposed to have an influence on range of motion. The aim of the study is to determine the influence of the instrumentation system employed in the postoperative posterior condylar offset (PCO) obtained. Methods: We studied 101 consecutive patients who had undergone an ultracongruent TKA for primary osteoarthritis of the knee. Preoperative and postoperative PCO (maximal thickness of the posterior condyle, projected posteriorly to the tangent of the posterior cortex of the femoral shaft) and PCO ratio (PCOR) (ratio of the PCO to the diameter of the femur, 2.5 cm above the termination of the supracondylar flare) was evaluated on true lateral radiographs and correlated with the instrumentation system used (47 cases with conventional mechanical instrumentation, 35 cases with computer assisted surgery system (CAS) and 19 cases with patient matched technology cutting blocks produced after the 3D reconstruction of the joint and the pre-operative 3D planning). Statistical analysis was performed using SPSS for Windows statistical package (v 15; SPSS, Chicago, IL, USA). Results: Preoperative PCO: 28.21 ± 2.94 mm; Preoperative PCOR: 0.83 ± 0.11; Postoperative PCO: 30.21 ± 3.46 mm; Postoperative PCOR: 0.9 ± 0.12. Conventional mechanical instrumentation: Postoperative PCO: 31.04 ± 3.6 mm; Postoperative PCOR: 0.92 ± 0.14. Computer assisted surgery: Postoperative PCO: 28.77 ± 3.26 mm; Postoperative PCOR: 0.85 ± 0.09. Patient matched technology cutting blocks: Postoperative PCO: 30.79 ± 2.64 mm; Postoperative PCOR: 0.96 ± 0.88. (The values are presented as the mean and the standard deviation.) We have obtained a statistically significant difference in the comparison between the postoperative PCO of the mechanical instrumentation system and the postoperative PCO of the CAS TKA. No significant difference among the postoperative PCO of mechanical

S323 instrumentation and patient matched technology was obtained. No significant difference among the postoperative PCO of CAS and patient matched technology was obtained. We have obtained a statistically significant difference in the comparison between the postoperative PCOR of the CAS TKA and the PCOR of the mechanical instrumentation and patient matched technology implanted TKA. No significant difference among the postoperative PCOR of mechanical instrumentation and patient matched technology was obtained. Conclusions: A tendency to decrease the PCO with CAS has been observed, but we have doubts about the described impingement of the posterior tibial insert against the femur, as responsibly of aberrant kinematics with anterior sliding of the femur during flexion. Further research is needed to document the relationship between the PCO and the flexion balance of TKA.

P32-387 Unicompartmental knee replacement. Mid-term results with ZUK prosthesis A. Schiavone Panni1, M. Vasso1, S. Cerciello1, A. Felici1 1 University of Molise, Science for Health, Campobasso, Italy Objectives: Progressive surgical technique and instrumentation improvement associated with materials and design evolution have increased effectiveness and popularity of unicompartmental knee replacement in the last 10 years. Unicompartmental knee arthroplasty preserves uninvolved osteocartilagineous and sosf tissue structures, so allowing easier and faster rehabilitation and a more physiologic functional recovery. The aim of this study was to report the 5-years results of ZUK unicompartmental metal-back prosthesis. Methods: Between February 2005 and December 2007, we implanted 80 ZUK prostheses in 80 patients for unicompartmental osteoarthritis or avascular necrosis. Patients were assessed clinically using the International Knee Society knee and function scores. Postoperative values of mechanical axis were calculated 12 months after surgery and compared to preoperative ones. Mean follow-up was of 4 years (2–5). Results: The mean International Knee Society knee and function scores improved respectively from 46 and 54 points pre-operatively to 82 and 94 at the last follow-up (p \ 0.0001). Average flexion increased from 110 to 127 (p \ 0.001). No patient had perioperative complications. We did not report ipo- and iper-correction of the frontal deformity in any cases. At the last follow-up, no knee required revision. Conclusions: High success rates of modern unicompartmental knee implants could derive from materials and design evolution, improvement of instrumentation and surgical technique, and strong restriction of indications. ZUK prosthesis represents the modern evolution of Miller-Galante unicompartmental knee system; we report our personal experience with this type of prosthesis implanted with a minimally invasive approach.

P32-423 Vitamin C plasma values reduced after orthopaedic surgical intervention in patients with total knee arthroplasty and prosthesis revision operation H. Lengnick1, K. Giesinger1, M. Kuster1, H. Behrend1 1 Kantonsspital St.Gallen, Orthopa¨die, St.Gallen, Switzerland Objectives: Vitamin C is the specific agent in the formation of intercellular substance and is needed in the collagen synthesis activating the enzyme prolyl-hydroxylase. It is involved in the process of wound and bone healing. Its deficiency is linked to wound healing complications, diseases of connective tissues and hypothetically to arthrofibrosis after total knee arthroplasty (TKA). To our knowledge

123

S324 no data exist concerning perioperative vitamin C concentrations in orthopedic patients. The aim of this study was to investigate whether a decrease in vitamin C plasma levels can be observed in patients undergoing orthopedic surgery depending on the extent of the surgical intervention. Methods: Twenty patients were divided into four same-size groups. Group A underwent ORIF for an ankle fracture, group B unterwent primary TKA, group C had hip or knee revision surgery for aseptic loosening. Five healthy volunteers (group D) without any surgical intervention were sampled as reference group. Vitamin C plasma values were measured 1 day preoperatively (d-1) and on day 1, 3 and 7 (d1,3,7) postoperatively. Results: Mean age and operation time was 54 years and 34 min in group A, 73 years and 94 min in group B, 77 years and 122 min in group C. Healthy volunteers were on average 52 years old. Patients undergoing primary TKA showed significant lowering of vitamin C levels in the postoperative course compared to preoperative values (d1: 10.0 ± 6.7 mg/l, d7: 1.7 ± 2.6 mg/l, p value = 0.007). A strong tendency of vitamin C depletion of almost 80% was observable in patients after revision intervention (d-1: 5.5 ± 3.0 mg/l, d7: 1.1 ± 0.3 mg/l, p = 0.05). Patients with ankle fractures already reached the preoperative level of vitamin C (d-1: 4.1 ± 2.1 mg/l, d3: 3.9 ± 1.4 mg/l, p [ 0.05) on day 3. Healthy volunteers did not show any lowering during the examination period. Conclusions: Vitamin C values were significantly lowered in patient groups with large operative interventions and longer operation time (TKA, revision surgery). The results of this study show an increased vitamin C depletion in patients after extended orthopedic surgical intervention. Hypothetically this is associated with accumulated oxidative stress and higher metabolic activity. Further studies are necessary to examine the role of vitamin C in the healing and rehabilitation process and its effect on improved postoperative function of musculoskeletal system in orthopedic patients.

P32-428 Patellar height after reconstruction of the extensor mechanism with allograft in total knee arthroplasty A. Valentı´1, R. Llombart1, P. Diaz de Rada1, G. Mora1, J.R. Valentı´1 1 Clı´nica Universidad de Navarra, Orthopaedic Surgery, Pamplona, Spain Objectives: Patellar tendon rupture after total knee replacement is a very limiting injury ranging between 0.1 and 2.5%. The etiology is multifactorial and among the risk factors are the number of previous surgical procedures, comorbidities and the surgical technique. The purpose is to share our experience in patellar tendon reconstruction with fresh-frozen allograft after primary or revision arthroplasty. Methods: We present a series of seven patients with patellar tendon rupture treated with fresh-frozen allograft (anterior tibialis and Achilles tendon) reconstruction after total knee replacement (4 revision and 3 primary). All patients were followed up clinically and radiologically in an average of 23 months (20–31 months). Results: The mean patient age was 76 years (61–83 years) with a time of evolution from arthroplasty and the tendon injury of 2.9 months (0–12 months) with an average of 2.9 previous surgeries on the knee (2–5). The functional assessment using the ‘‘knee and functional score’’ raised from 26 and 16 to 82 and 55 postoperatively. The mean extension lag was 4 (0–15) with a range of motion of 99. The comparative radiographs of the immediate postoperative and the last control showed a rise of the patella of 13 mm. Conclusions: The use of fresh-frozen allografts as a therapeutic solution to the patellar tendon ruptures after total knee replacement seems to have acceptable results given the commitment of the tissues

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 and the difficulties of comorbidity associated with these knees. An increased patellar height does not seem to affect functionality.

P32-497 Use of Tantalum in total knee arthroplasty in severe and morbidly obese patients (BMI [ 35 kg/m2) L.M. Lozano1, D. Popescu2, S. Sastre3, J.M. Segur1, J.C. Martinez-Pastor4, F. Macule4 1 Hospital Clı´nic. University of Barcelona, Knee Unit, Orthopaedic Surgery Department, Barcelona, Spain, 2Hospital Clinic i Provincial, Barcelona, Spain, 3Hospital Clinic i Universitari de Barcelona, Knee Unit, Barcelona, Spain, 4Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain Objectives: Study our first results with the use of trabecular metal tibial implant in Total Knee Artroplasty in Severe and morbid obese patients. Methods: Patients with BMI [ 35 kg/m2 diagnosed with knee osteoarthritis. Study initiated in November 2008. We used a Nexgen (Zimmer) total knee arthroplasty with a trabecular metalTM tibial tray. Functional outcomes evaluated using WOMAC and KSS R, KSS F and KSS radiological. Results: 39 prostheses were studied. Follow-up 2 years. 38 women and 1 male. Mean age 69.43 years. Mean BMI 39.37 kg/m2 (35–55). There were two deep infections. There was a significant improvement in WOMAC and KSSR and KSSF scores (p \ 0.0001) at the final follow-up (average 34 months). There where 16 patients (44.4%) with radiolucent lines at 1 year at tibial component and only 9 patients presented radiolucent lines at final follow up. In all patients the Radiological Knee Society score was \ 4. Conclusions: Most tibial implants present tibial subsidence during the first three months without any lift-off. Complete stabilization of trabecular metal implants occurs at 12 months. Subsidence affects the entire implant. This is important since it means no lift-off. The reason may be the low rigidity of trabecular metal. The subsidence is about 0.8 mm at 6 weeks and 3 months. Complete stabilization of the trabecular metalTM implants at 12 months may indicate the possibility of bone growth within the trabecular bone. Namba drew attention to the high incidence of infection after total knee arthroplasty in patients with a BMI [ 35 kg/m2 The risk of infection (odds ratio) was 6.7 times higher in this group of patients. Aretrospective review of 8,494 hip and knee arthroplasties by Malinzak in 2009 found that a BMI of 50 kg/m2 increased the risk of deep infection by 21.3 (p \ 0.0001). An Australian study of 1,214 total knee arthroplasties by Dowsey in 2009 found that factors correlating with prosthetic infection were morbid obesity (OR 8.96) and diabetes (OR 6.87). The biomechanical advantage of the tantalum tibial surface is the virtual absence of liftoff, a cause of prosthetic loosening. The initial advantage found was the high flexion achieved by these patients and the significant improvement in WOMAC and KSS scores.

P32-535 Posterior condylar offset and tibiofemoral contact point do not influence outcome and range of motion in a mobile bearing total knee arthroplasty G.J.P. Geysen1, G. van Stralen2, P.G. Anderson3, P. Heesterbeek3, A. Wymenga4 1 Sint Maartenskliniek, Orthopedic surgery, Den Bosch, The Netherlands, 2Ziekenhuis Nij Smellinghe, Orthopedic surgery, Drachten, Netherlands, 3Sint Maartenskliniek, RD&E, Nijmegen, The Netherlands, 4Sint Maartenskliniek, Orthopedic Surgery, Knee Reconstruction Unit, Nijmegen, The Netherlands Objectives: The posterior condylar offset and the tibiofemoral contact point have been reported as important factors that can influence range of motion and clinical outcome after Total Knee Arthroplasty (TKA)

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 A mobile bearing knee implant with an anterior posterior gliding insert would in theory be more sensitive for changes in posterior condylar offset and contact point. For this reason we analysed the posterior condylar offset and tibiofemoral contact point and the relation with outcome and range of motion in 132 patients from a prospectively documented cohort in this type of implant. Methods: The prosthesis we used was a posterior cruciate retaining AP gliding mobile bearing total knee replacement (SAL II Sulzer Medica, Switzerland). 132 knees were evaluated by means of pre-and postoperative posterior condylar offset (PCO), and postoperative tibiofemoral contactpoint (CP). Measurements were done on X-rays of the knee taken in 90 of flexion and with less than 3 mm rotation of the femur condyles. The outcome parameters; range of motion (ROM) and the Knee Society Score (KSS) for each knee were determined preoperatively and at 1, 3 and 5 years follow up. Results: The mean KSS improved from 91 to 161 at 5 year follow up, a statistically significant improvement (p \ 0.001). The mean ROM improved significantly (p \ 0.05) from 102 to 108. No difference was found between the pre-operative PCO (28.9, SD 3.5 mm) and the post-operative PCO (28.9, SD 3.1 mm). The tibiofemoral contact point ratio had an average of 53.9% (SD 5.5, n = 132). Scatter plots showed uniform clouds of values; increase or decrease in PCO and CP had no significant influence on ROM or KSS. Conclusions: The posterior condylar offset and the position of the tibiofemoral contact point do not significantly influence the range of motion, nor the KSS in a cruciate retaining AP gliding mobile bearing knee implant in an older population. Suboptimal placement and balancing of the prosthesis and the posterior cruciate ligament still gives a good functional and clinical outcome in this patient group.

P32-705 Problems, obstacles and complications in more than 800 computer assisted knee arthroplasties A. Manzotti1, N. Confalonieri1 1 CTO Hospital, 1st Orthop Department of Milan, Italy Objectives: Aim of the study was to difficulties and adverse effects in more than 800 CAS knee replacement performed in one center since 1999 to assess their incidence and determine if computer assisted knee replacement has to be considered as a demanding surgical technique because of an higher number of adverse consequences compare to traditional techniques. Methods: Among more than 1,000 computer assisted TKRs were performed in our department, 816 TKRs were followed for at least 6 months postop. and included in the study. All the cases were divided in 3 series according to when the surgery had been performed to consider the evolution of the navigation systems and the familiarity of the surgeons with this technological improvements (group A ranging from 1999 to 2001, group B from 2002 to 2005 and group C from 2006 to 2009). The mean surgical time was assessed from the surgical charts. All intraop. problems (difficulties that required no operative intervention to resolve or without any consequence on the navigation process), intraop. obstacles (difficulties that required operative intervention or that caused a failure of the navigation process) and complications (intra-operative injuries and all the problems following in the first 6 months postop.) were registered. Adverse facts not directly caused by the surgical but derived by other conditions were excluded from the study. Results: There were 210 cases in group A, 272 in group B and 334 in group C. We did not registered any failure of the implants because malalignment or instability within 6 months postop. There were no differences in number of total problems and complications among the 3 groups. Obstacles were statistically higher in group A where we observed an higher incidence of navigation failures in association both to the first versions of the softwares and to a minor surgical

S325 experience. We did not registered any statist. significant difference in incidence of superficial pin site infections (complication) between the 3 groups. In 2 cases in group B we registered a pin bending (problem) because too close to the tibial stem without any problem at the removal. We registered a cases of intraop. tibial fracture in group C. No abnormal intra op. or post-op bleeding was registered because of the surgical technique in all 3 groups. No statistical difference in clinical evident DVT among the 3 groups was detected in the first 6 months postoperatively. Surgical time was longer in group A with a statist. significant difference compared both to group B and C. Conclusions: In this study the authors registered both a higher rate in navigation failure and longer surgical time in the first cases even because of less advance navigation systems and lower experience with CAS. However in a series of more than 800 CAS TKRs the authors could demonstrated no increased rate of complications compared to traditional techniques despite of a at least radiological better alignment already demonstrated in literature.

P32-778 Unicompartmental knee resurfacing versus segmental bicompartmental knee resurfacing versus total knee arthroplasty: a comparison of clinical outcomes at 2 years follow-up O. Al-Dadah1, G. Hawes1, P.J. Chapman-Sheath1, J.W. Tice1, D.S. Barrett1 1 University Hospital Southampton, Trauma and Orthopaedic Surgery, Southampton, United Kingdom Objectives: To compare the clinical outcome of patients who have undergone unicompartmental knee resurfacing, segmental bicompartmental knee resurfacing and total knee arthroplasty up to 2 years post-operatively. Methods: The study comprised a total of 133 subjects. This included 30 patients who underwent medial unicompartmental knee resurfacing (UKR group), 53 patients who underwent combined segmental medial unicompartmental knee resurfacing with patellofemoral joint resurfacing (BKR group) and 50 patients who underwent conventional total knee arthroplasty, including patella resurfacing (TKR group). All subjects were evaluated using the Oxford Knee Score (OKS), Western Ontario and MacMaster Universities Osteoarthritis Index (WOMAC) and Knee Injury and Osteoarthritis Outcome Score (KOOS). Patients in each group were assessed using all 3 scoring systems pre-operatively and 6 months, 1 year and 2 years postoperatively. Results: Statistical analysis revealed a significant improvement of the mean pre-operative OKS for UKR (22.7), BKR (22.6) and TKR (16.6) as compared to 6 months post-operative results UKR (38.1, p = 0.046), BKR (36.8, p \ 0.001) and TKR (34.5, p \ 0.001). For WOMAC Pain, Function and Stiffness subscores pre-operatively for UKR (49.3), (54.1), (51.0) respectively, BKR (46.0), (51.8), (46.8) respectively and TKR (42.6), (40.8), (44.2) respectively compared to 6 months post-operatively UKR (65.0, p = 0.049), (69.6, p = 0.178), (56.5, p = 0.184) respectively, for BKR (79.0, p = 0.001), (76.0, p \ 0.001), (70.8, p = 0.009) respectively and for TKR (79.4, p \ 0.001), (77.5, p \ 0.001), (70.1, p \ 0.001) respectively. For KOOS Pain, Symptoms, Activities of Daily Living, Sport and Recreation and Quality of Life subscores pre-operatively for UKR (43.0) (59.5), (52.2), (19.2), (27.2) respectively, for BKR (40.7), (47.2), (50.8), (27.5), (24.5) respectively, for TKR (31.3), (36.1), (37.3), (15.7), (17.9) respectively compared to 6 months post-operatively UKR (80.4, p = 0.117), (84.0, p = 0.135), (87.4, p = 0.150), (67.0, p = 0.251), (64.0, p = 0.399) respectively, for BKR (69.9, p = 0.038), (68.3, p = 0.147), (65.2, p = 0.089), (41.7, p = 0.525), (57.1, p = 0.006) respectively and for TKR (75.8, p = 0.008), (72.5, p = 0.077), (78.1, p = 0.036), (35.8, p = 0.219), (62.1, p = 0.060) respectively. After 6 months there was no further statistically

123

S326 significant improvement in any of the outcome scores in any of the groups up to the 2 year follow-up results. There was no significant difference in any outcome score post-operatively between the 3 groups. Conclusions: The magnitude of clinical improvement following knee arthroplasty is greatest at 6 months, thereafter only modest improvements continue to occur. Segmental bicompartmental knee resurfacing performs satisfactorily compared to current knee arthroplasty techniques and is an important adjunct in knee resurfacing options for younger patients with significant arthritis.

P32-811 Semimembranosus release as second stage of soft tissue balancing in varus total knee arthroplasty Y. In1 1 Seoul St. Mary’s Hospital, The Catholic University of Korea, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: There are limited clinical data on effect of semimembranosus release in patients with varus knees having TKA. The purpose of this study was to determine the effect of semimembranosus release and to evaluate the effectiveness of our medial release technique based on three step algorithm. Methods: One hundred four consecutive varus knees performing posterior stabilized TKA were included in the study. After deep medial collateral ligament (MCL) release and bone cutting, 31 knees (29.8%) were not balanced. To determine the effect of semimembranosus release, medial and lateral gaps were measured using a distractor before and after the semimembranosus release in these 31 knees with the knee full extension, 45 flexion, and 90 flexion position. The gap changes were calculated and statistical analysis was performed. Seven patients (6.7%) still showed medial tightness even after the semimembranosus release. Medial tightness was relieved by using the piecrust release of the superficial MCL in these knees. Results: Medial gaps were significantly increased after the semimembranosus release compared to the pre-release status (gap increment, full extension; 1.45 ± 1.60 mm, 45 flexion; 2.00 ± 1.57 mm, and 90 flexion; 2.25 ± 1.29 mm) (P = 0.000). Lateral gaps were also significantly increased after the semimembranosus release except full extension position (gap increment, full extension; 0.51 ± 1.82 mm (P = 0.126), 45 flexion; 1.06 ± 2.06 mm (P = 0.07), and 90 flexion; 1.41 ± 1.58 mm (P = 0.000)). Soft tissue balancing was obtained in all patients with medial release technique based on three step algorithm. Conclusions: Semimembranosus release has its desired effect on gap balancing in varus knees. Algorithmic release approach to the medial side focusing on conservation of superficial MCL is considered to be effective in correcting varus deformities of the knees having TKA.

P32-849 ‘Two radius area contact’: posterior stabilized total knee arthroplasty. Short and long term results A. Verdoodt1, P.-P. Casteleyn1 1 Vrije Universiteit Brussel, Orthopeadic surgery, Brussel, Belgium Objectives: The Two Radius Area Contact (TRAC) posterior stabilized prosthesis tried to minimize the bearing and interface stresses to reduce the risk of polyethylene wear and loosening, by maximizing the contact area from full extension to full flexion. We recently started a short- and long- term review of these cases, in order to see if the theoretical benefits concerning the functional results could be confirmed. Methods: The population consists of 85 patients. 43 had died of unrelated causes and 3 were unable to be examined because of extremely poor physical or mental health. Five patients were lost to

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 follow up. 34 patients could be clinically evaluated after a mean follow-up period of 13 years (11–15 years). A radiological follow up with a mean of 10 years (10–14 years) could be performed for 47 patients. For the clinical examinations, the WOMAC score, the SF-36 and the KSS, as well as the VAS were used. For the radiological evaluation component position, alignment and radiolucency’s were measured using the Knee Society Ro¨ntgenographicScoring-System. Results: The short term results were characterized by 6 cases of AP-luxation’s, which could result from design of the prosthesis, (dual track and low jump height). The luxations could all be reduced closed. A PE exchange was only needed in 1 case. The early engagement of the cam-post mechanism during flexion also increased the load on the extensor mechanism, resulting in 3 cases of patellar component revisions. Another early problem was pain due to synovial tissue impingement between the femoral component and the PE, in the lateral compartment. In 5 patients an intervention was performed. The mean WOMAC score is 20.6% (2.3–56.2). The mean score for pain is 6% (0–22), stiffness 21% (0–62.5) and for loss of function 27% (0–97). The mean Knee Society Score is 86.8 clinical (59–99) and 44% functional (0–90). The mean range of flexion is 105. The mean score on the SF-36 questionnaire is 60% (22–87). The pain at the time of examination was 6% (0–30), the maximal pain 47% (0–100) and the minimal pain 6% (0–50). Radiolucent lines are seen in 4 patients. Two revisions are performed, at 4 and 12 years, because of loosening of the TKA. Using revision for all causes as the endpoint, the cumulative survival rate at 10 years is 98.68% (95% CI 93.2–99.8). Conclusions: The long- term results are similar to other TKA-designs. The 34 patients, who are clinically evaluated, show limited problems in daily activities. A majority of these patients are satisfied of the functional result after a mean time of 13 years. The radiological evaluation shows well positioned components and little loosening. The short term results shows a relative big amount of complications (19%). Except for the synovial tissue impingement between the femoral component and the polyethylene mobile bearing, these complications also occur with other TKA designs. Because of this intra-articular impingement of synovia, the use of the TRAC-PS prosthesis was abandoned.

P32-887 Patient-specific cutting block in TKR: comparison between CT and MRI 3D planning G. Messerli1, D. Fritschy1 1 Hoˆpitaux Universitaires de Gene`ve, Service d’Orthope´die et Traumatologie de l’Appareil Moteur, Gene`ve, Switzerland Objectives: The last innovation in the field of knee joint replacement technique is the use of customized cutting blocks which are produced individually for each patient. The production of these guides is based on MRI or CT data. The purpose of this study is to compare the precision of MRI and CT in the design and production of patient-specific cutting blocks. Methods: A first experiment has been realized with 3 human specimens on which MRI and CT images were obtained for the production of femoral and tibial cutting guides. The surgical precision of both cutting guides series was assessed clinically and by computer assisted navigation. 10 patients, on the list to undergo a total knee replacement, accepted to be part of this prospective study to compare MRI and CT production parameters. Each patient entered a protocol of MRI and CT images providing data of hip, knee, ankle joints and lower limb alignment. These parameters allowed, after acceptance of the surgeon, the construction of 3D patient-specific femoral and tibial guides. Long standing X-rays were also performed, as usually.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 During surgery, positioning of both MRI- and CT-based cutting blocks was controlled by computer assisted navigation and conventional cutting guides. Post-op long standing X-rays were obtained in the following weeks. 10 patients, operated on during the same period with classical intramedullary guided technique, provide a control group with postop long standing X-rays. Results: Slope and distal femoral rotation are respected with good correlation between planning and navigation measurements. Tibial and distal fe´moral cuts are also respected with a better pre´cision for CT guide versus MRI (0–1 mm vs. 2–3 mm). Within a limit of 2 (+ and -1), the mechanical axis of the lower limb (HKA) was between 178 and 180 in all cases with both MRI and CT customized cutting guides intra-operatively. The surgical measurements and navigation data were partially confirmed by the post-operative long standing X-rays views. The variation was from 0 to 5, always in valgus with a mean value of 2. These results are in the limits of the control group (m. = 2, 0–6). Conclusions: The precision of the 3D patient-specific cutting guides is slightly better with CT based options, in our experience. Therefore, we think that CT images can be used in the design of patient-specific cutting blocks for total knee replacement.

P32-895 Periprosthetic fractures of the distal femur: results of operative treatment in a single center O. Leino1, L. Lempainen1, J. Mokka1, M. Seppanen1, P. Virolainen1, K. Makela1 1 Turku University Hospital, Department of Orthopaedic Surgery and Traumatology, Turku, Finland Objectives: Periprosthetic fractures (PF) of the distal femur are a serious complication following total knee arthroplasty (TKA). Current operative methods include locking plates with or without bone graft augments, revision arthroplasty with a long-stemmed femoral component or retrograde intramedullary nailing. Our main objective was to study operative treatment and patients’ ability to return to preceding state of living after injury. We also wanted to analyze post-operative complications in our patient material. Methods: During 2000–2010 a total of 70 patients (60 women (86%), 10 men) with a PF of the distal femur were treated operatively at our center. The data used in this study was retrospectively collected from our electronic patient archives. Variables of interest for this study were time from TKA to PF, chosen operative method, fracture type using Rorabeck classification, possible early and late complications and revision surgeries. Death or last record of orthopedic appointment was chosen as termination points of follow-up time. Results: The mean age at time of PF was 78.9 years (range 43–95, median 81, SD 10.3) and the average time interval between TKA and PF was 7 years (range 0–22, median 5, SD 6). The average follow-up after the PF was 9 months (range 0–120). Operative methods used were plating (n = 39, 56%), revision arthroplasty (n = 28), combined plating and revision arthroplasty (n = 1) and retrograde nailing (n = 1). Bone grafts were used in 37 cases; in 33 platings and in 4 revision arthroplasties. According to Rorabeck classification all type I fractures (n = 6) were treated by plating and all type III fractures (n = 15) were treated by revision arthroplasty. Three cases of perioperative complications were reported: one inadequate reduction during plating and two perioperative fractures during revision arthroplasty. Later complications included infections (9 cases, of which all but one required lavation), prosthesis luxation

S327 tendency after revision arthroplasty (3 cases) and non-union after plating (n = 6). Seventeen (26%) patients underwent revision surgery during follow-up. Of these cases 7 were after treatment by revision arthroplasty, 9 cases after treatment by plating and one case after retrograde nailing. Fifteen (21%) patients had not lived at home before time of PF and none of these patients returned home after treatment. Fifty-five (79%) patients had lived at home and only four (7%) of these patients weren’t able to return home after treatment. Conclusions: These fractures mostly concern elderly people with multiple comorbidities. Concerning the fact that major surgery is a great risk for extended hospitalization for sedentary/old patients, we gladly noticed that so many of our patients were eventually able to return to their preceding state of living. According to our results it seems that regardless of the reported complications in this group, operative treatment generally delivers good results.

P32-904 Radiological evaluation of 226 consecutive journey knee replacements - conventional versus computer navigated surgery B. Christen1, B. Rieger2, E. Aghayev3, M.S. Neukamp3 1 Orthopa¨dische Klinik Bern, Salemspital, Bern, Switzerland, 2 Kantonspital Bruderholz, Orthopa¨dische Klinik, Bruderholz, Switzerland, 3MEM Research Center, Institute for Evaluative Research in Orthopaedic Surgery, Bern, Switzerland Objectives: The value of computer navigation in total knee replacement (TKR) is still under discussion concerning precision and clinical outcome. Major interest would focus on the restoration of axis and a correct tibial slope as the rotational alignment of the femoral component is still debated. Methods: In 196 patients (126 female, 70 male) with a mean age of 67.7 years with gonarthritis 226 consecutive Journey BCS total knee replacements (TKR) were implanted (Smith & Nephew, Memphis, USA) between December 2006 and March 2011 in a nonrandomized manor. All operations were performed by one single surgeon. The first 102 cases were operated with the conventional instrumentation. Tibial aiming was performed extra- the femoral one intramedullary. The following 124 knees were operated with a CT-less computer navigation (PI Galileo, Smith& Nephew, Aarau, Switzerland) for the tibial and distal femoral cuts. In both groups femoral rotation was defined by combining bony landmarks with the use of a tensioner. Alpha-, Beta-, Gamma- and Delta-angles were measured postoperatively according to Ewald (CORR 1989, 248:9–12). In addition preoperative femoro-tibial (FT) angle was compared to the postoperative value. Results: Mean a (valgus of femoral component) was 98.4 in the nonnavigated and 99.2 in the navigated group. Mean b counting for the ap position of the tibial component was 89.1 and 88.7 respectively. Mean c (flexion of femoral component) was 91.4 and 88.5 respectively. Mean d finally (posterior slope of tibial component) was 90.2 and 90 respectively. The differences in all groups were not significant. Outliers in FT-angle ([190 and \185) were significantly less in the navigated group and could be reduced for varus and valgus deformities. Conclusions: In this study the radiological results for the tibial and the femoral component in TKR did not differ significantly between the non-navigated and navigated group. Mainly the accuracy of the important tibial slope could not be improved by using the navigation. Nevertheless the outliers for varus or valgus alignment could be reduced in the navigated group. The conventional instrumentation with extramedullary alignment at the tibia and intramedullary at the femur seem to give consistently accurate results which could not be improved essentially by the use of a time consuming and expensive CT-less navigation.

123

S328 P32-936 Retrograde femoral nailing of periprosthetic femoral fractures after total knee joint endoprosthesis replacement A. Fechner1, O. Meyer1, G. Godolias1 1 St. Anna-Hospital, Center for Orthopaedics and Traumatology, Herne, Germany Objectives: The number of patients being treated with knee joint replacement is growing. A growth of peri- and postoperative complication rate is expected. The treatment of a periprosthesis femoral fracture of a stable knee joint prosthesis causes many problems and controversial discussions. In these cases we are often dealing with multimorbid patients, low patient compliance and osteoporosis, factors which cause many problems when treating this kind of fractures. The indications of treatment via retrograde femoral nailing in these cases are fractures of type I and II in the Lewis-Rorabeck classification. Of course this kind of treatment requires a prosthesis design which allows the placement of the retrograde nail. Methods: From 2003 to 2008 21 patients were treated via retrograde femoral nailing (supracondylar nail (SCN) T2, Stryker) while suffering from a distal periprosthesis femoral fracture. All of this fractures where classified as Type II fractures in the Lewis-Rorabeck classification. The age of the patients averaged at 78.5 years (61–89 years) and all of the patients suffered a trauma of the fractured side. The average time between the primary prosthesis implantation and the traumatic incident accounted 5.7 years. All patients received the same postoperative treatment. The average postoperative treatment duration accounted 12.5 (6–24) months. Radiological examinations including a CT-scan where performed prior to surgery. Postoperative only native radiological examinations where conducted. An evaluation of pain-level, activity-level and range of joint motion was performed. Results: In the follow-up examination 16 patients were satisfied with the postoperative result, 14 patients reached the prior to the trauma described activity level. The average range of joint motion accounted 95 for flection (80–120) and 0 for extension (0–5), while in many cases no preoperative motion values could be obtained. 5 patients suffered of loadbearing pain. The average operation time accounted 77 min. No occurrence of significant rotation malpositioning was observed. In 1 case a skin infection was observed and treaded conservatively. One patient suffered a second trauma 4 months after surgery and was then treated via osteosynthesis using a fixed angle plate. 2 cases showed a dorsal tilt of the distal fragment despite correct positioning of the retrograde nail. In 1 case the radiological follow-up examination showed a valgus misalignment, which had no clinical affection on the patient. Conclusions: Retrograde femoral nailing is an effective method for treating distal femoral periprothesis fractures after knee joint replacement surgery. The morbidity caused by the operative approach is low. The achieved stability allows an early functional postoperative treatment. The difficulty in this procedure is caused due to the need of correct axis alignment of the fracture.

P32-983 Unicompartmental knee arthroplasty for sSpontaneous osteonecrosis of the knee: 10-years survivorship and long term clinical results D. Bruni1, F. Iacono2, G. Raspugli2, M. lo Presti2, S. Zaffagnini3, M. Marcacci3 1 Rizzoli Orthopaedic Institute, Biomechanics Lab, Bologna, Italy, 2 Rizzoli Orthopaedic Institute, University of Bologna, Bologna, Italy, 3 Istituto Ortopedico Rizzoli Bologna, Bologna, Italy

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: The purpose of the study is to examine a long term survivorship of UKA’s in a larger group of patients with SPONK, along with their subjective, symptomatic and functional outcome; to determine the percentage of failures and the reasons for the same in an attempt to identify relevant indications, contraindications, and technical parameters in treating SPONK with a modern implant design. Methods: We retrospectively evaluated 84 patients with late-stage spontaneous osteonecrosis of the knee who had a medial UKA from 1998 to 2005. All patients had preoperative MRI to confirm the diagnosis, exclude metaphyseal involvement, and confirm the absence of major degenerative changes in the lateral and patellofermoal compartment. Mean age at surgery was 66 years and mean body mass index was 28.9. A Kaplan–Meier survival analysis was conducted using revision for any reason as the end point. The minimum followup of 63 months (mean, 98 months; range, 63–145 months). Results: The 10-year survivorship was 89%. Ten revisions were performed and the most common reasons for revision were subsidence of the tibial component (four) and aseptic loosening of the tibial component (three). No patient underwent revision for osteoarthritis progression in the lateral or patellofemoral compartment. There was a statistically significant difference between postoperative VAS, KSS, WOMAC, Oxford, Range of Motion and Tibial Slope (overcorrected by 3.7, p \ 0.0023) between the survivors and the failures. Conclusions: SPONK may be an optimal indication for UKA provided secondary osteonecrosis of the knee is ruled out; pre-operative MRI is performed to document involvement of other compartments, status of the ligaments and depth of lesion; and there is no overcorrection in any plane.

P32-1051 One-stage versus two-stage bilateral unicompartmental knee arthroplasty: a comparison of post-operative complications and functional outcome J. Chen1, N.N. Lo2, S.J. Yeo2 1 Singapore General Hospital, Department of Orthopaedic Surgery, Singapore, Singapore, 2Singapore General Hospital, Department of Orthopedic Surgery, Singapore, Singapore Objectives: Unicompartmental knee arthroplasty (UKA) is a popular treatment of choice for patients diagnosed with unicompartment osteoarthritis of the knee. Compared to total knee arthroplasty (TKR), UKA offers quicker functional recovery, shorter hospitalization stay, fewer failures at long term follow up and lesser post-operative complications owing to the less invasive approach and shorter surgery duration. While there are many literatures comparing one- and two-stage TKR, there are very few published data on the safety and functional outcome of one-stage bilateral UKA. Two recently retrospective studies comparing one- and two-stage bilateral UKA have shown conflicting results on the safety of one-stage bilateral UKA. We aim to investigate our centre’s experience on the safety and functional outcome of one-stage versus two-stage bilateral UKA. Methods: We reviewed 171 patients who underwent bilateral UKA at our centre. 124 patients underwent a one-stage procedure while 47 patients underwent a two-stage procedure with the two surgeries done within a year apart. They were prospectively followed up for 2 years. Post-operative complications including proximal deep vein thrombosis, pulmonary embolism, cardiac events, periprosthetic fracture and infection were recorded for both groups. Range of movement and functional outcome were assessed at 6 months and 2 years post-surgery using American Knee Society Score, Oxford Knee Score and SF36. Data analysis was done using student’s unpaired t test with SPSS 15.0. Post hoc power analysis revealed sufficient power to detect a

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 clinically significant difference in all variables except post-operative complications. Results: Both groups of patients shared similar demographic and disease profiles. Patients in the one-stage group showed significant shorter cumulative hospitalization stay of 6.0 days compared to 8.5 days in the two-stage group (p \ 0.001). There was no significant difference in post-operative complications at 2 years follow up (p [ 0.05). At 6 months post-surgery, the range of movement was 3.38–127.20 and 3.53–125.88 for one- and two-stage procedure respectively (p [ 0.05). At 2 years post-surgery, the range of movement improved to 2.20–129.83 and 2.87–128.19 for oneand two-stage procedure respectively (p [ 0.05). While both groups showed significant improvement in American Knee Society Score, Oxford Knee Score and SF-36 at 6 months and 2 years post-surgery (p \ 0.05), there was no significant difference in the scores between the two groups at both 6 months and 2 years post-surgery (p [ 0.05). Conclusions: While one-stage bilateral UKA offers a shorter cumulative hospitalization stay, there was no significant difference between one- and two-stage bilateral UKA in terms of safety and functional outcome at 2 years post-surgery.

P32-1057 The reliability and accuracy of knee implants sizing predicted by digital templating S. Zadoroznijs1 1 VSIA ‘Traumatologijas un Ortopedijas Slimnica’, Latvia, Latvia Objectives: Digital templating is becoming more and more popular in total knee replacement generally cause of the latest trend to digitalize medical records as well as surgery. We aimed to determine the reliability and accuracy of digital templating in the pre-operative workup for total knee arthroplasty. Methods: A retrospective study was done to assess the accuracy of the knee implant sizing predicted by digital images in 105 caucasian adults, who had osteoarthritis of the knee. Digital templating was performed using a calibrating 25-mm metallic ball and Agfa Orthopaedic Tools digital software package by a surgeon not involved with the operation, who was blinded to the size of the implant inserted. The Press Fit Condylar Sigma Knee system was used in all the patients. Digital anteroposterior and lateral radiographs of the knee were used in measuring the implant size. The results from digital images were compared with the size of actual femoral and tibial implants used at the time of surgery. Results: The correct size of the implant was predicted in 73 of 105 (69.524%) of the femoral and 70 of 105 (66.667%) of the tibial components. The correct size of the whole system was predicted in 58 of 105 (55.238%) cases. The digital preoperative planning predicted 104 of 105 (99.048%) femoral and tibial implants and 103 of 105 (98.095%) whole systems to within one size. There were 2 cases in which the predicted implant (1 case—femoral, other— tibial) appeared to be undersized from the final component by 2 sizes. The tibial component appeared to be more often undersized25 of 105 (23.810%) versa 22 of 105 (20.952%) in femoral component. The rate of femoral and tibial components to be oversized on the preoperative radiographs appeared to be the same—10 of 105 (9.524%). There were no cases of components to be oversized by 2 sizes. Conclusions: We conclude that digital templating using a calibrating 25-mm metallic ball and Agfa Orthopaedic Tools digital software is a reliable method of predicting the implant to within one size. There was a trend toward implants to be undersized in digital templating, which can be explained by the will of the templating surgeon to select the implant not overhanging the bone and by the desire of the operating surgeon to preserve as much bone as possible. Future prospective

S329 studies are needed to determine whether preoperative digital templating by the operating surgeon impacts his choice thus improving the accuracy of knee implants sizing.

P32-1081 Testing high performance after knee arthroplasty: a new objective functional score A. Baldini1, L. Manfredini1, P. Ceruli Mariani2, A. Pisaneschi3 1 IFCA, Ortopedia, Firenze, Italy, 2Ospedale Misericordia e Dolce, Prato, Italy, 3Villanova Clinic, Firenze, Italy Objectives: Functional recovery assessment after total knee arthroplasty requires specific evaluation tools. The objective of our study was to develop and validate a new objective functional scoring system which detect the ability of the patients to perform high activity level tasks. A secondary objective was to assess the activity level of patients with a total and unicompartmental knee arthroplasty compared to their peers. Methods: Four matched groups of fifty patients each, below 65 years old, were recruited. One group were normal patients, one were patients with an osteoarthritic knee, one group were patients who underwent unicompartmental knee arthroplasty, and one group were patients who underwent posterior-stabilized total knee arthroplasty. All arthroplasty patients had the following inclusion criteria: Charnley category A, 18 months minimum follow-up, and excellent score at the Knee Society score. Examinations were performed by two blinded examiners. Outcome measures include the Knee Society Scoring System, the Knee Osteoarthritis Outcome Score (KOOS), and the new Knee Performance Score. The new score was developed to be administered in few minutes in every medical office setting, and without special instruments. It included simple tasks which explore the knee strength, flexibility, agility, and propioception. Results: Patients with a well functioning posterior-stabilized total knee arthroplasty performed similarly to their healthy peers and to those patients with a unicompartmental arthroplasty. For few propioceptive and agility tasks the healthy group and the unicompartmental one performed statistically better then the total knee group. Patients with a replaced or normal knee who rated higher in the strength tasks had a higher overall performance score. The new score showed high intra and interobserver reliability. Distribution of data did not show ceiling effect within excellent results. Conclusions: The study has contributed further evidence to understand the level of functional recovery after knee arthroplasty. Subtle differences between different arthroplasty designs and patients categories may be evidenced using a more sensitive functional scoring system.

P32-1092 Total joint replacement in the past does not relate to a deteriorated functional level and health status in the oldest old W. Verra1, A.J. de Craen2, C.C. Jaspars1, A.B. Maier2, R. Nelissen3 1 Leiden University Medical Center, Orthopaedic Surgery, Leiden, Netherlands, 2Leiden University Medical Center, Geriatrics and Gerontology, Leiden, The Netherlands, 3Leiden University Medical Center, Department of Orthopedic Surgery, ZA Leiden, The Netherlands Objectives: Total hip or knee replacement has been proven effective in terms of improvement of joint function, quality of life and pain reduction. However, in earlier studies oldest old people with total joint replacements have been underrepresented with respect to long term outcome results. In this study the health related and functional characteristics of oldest old subjects with and without total joint replacement are compared. Methods: Data originates from the Leiden 85-plus study; a large prospective cohort study that included and characterized subjects when they reached the age of 85. Participants, all aged 85 years, were

123

S330 divided into a group who had received a total hip or knee replacement in the past and a group of participants without a joint replacement. Characteristics recorded were co-morbidity, physical functioning, daily living activities, joint complaints, quality of life and mortality. Results: Thirty-eight out of 599 participants (6.3%) received a total joint replacement in the past (mean age 78.2 (SD 4.7) years). The prevalence of co-morbidities, such as pulmonary disease and Parkinson’s disease, was slightly lower in the group participants with a total joint replacement. Participants with a total joint replacement walked 0.11 m/s slower (p = 0.006). Forty percent of those with a joint replacement complained about hip pain as opposed to 16% of those without a total joint replacement (p = 0.007). All other characteristics were comparable between both groups. Mortality was lower during the first 8 years of follow-up in the group with joint replacement compared to the control group (p = 0.04). After final follow-up of 12 years, a difference in mortality of participants with and without a joint replacement was no longer found (p = 0.41). Conclusions: Subjects with a hip or knee replacement performed equally well compared to those without a joint replacement, besides the lower gait speed and the higher amount of hip pain complaints. Having a total hip or knee replacement is not associated with poorer functional level and health status.

P32-1118 Utilisation of finite element analysis to predict elevated bone metabolic activity in patients with painful total knee arthroplasty R.T. Keller1, M.T. Hirschmann2, N.F. Friederich2, U.N. Hansen1, A.A. Amis1 1 Imperial College London, Mechanical Engineering Department, London, United Kingdom, 2Kantonsspital Bruderholz, Department of Orthopaedic Surgery and Traumatology, Bruderholz, Switzerland Objectives: Unexplained knee pain after TKA is a significant clinical issue. Understanding how patient-specific joint geometry and implant placement may contribute to abnormal loadings and thus pain may enable clinicians to provide more effective treatments. To better understand why a TKA procedure may result in an unsatisfactory outcome, we created patient-specific FEA models of knee joints from Single Photon Emission Computer Tomography (SPECT)/Computer Tomography (CT) scans and analyzed stresses and strains at the bone-implant interface. Methods: SPECT/CT scans were collected from 100 patients undergoing TKA. Patients were injected with 700 MBq 99mTc-HDP and scanned using a Symbia T16 (Siemens AG) hybrid scanner. SPECT/CT was performed 2 h after injection during the delayed metabolic phase. Bone metabolic activity was classified using a validated localization scheme. CT images were rendered into 3-dimensional surface models using Avizo, a visualization software package (VSG). Surface models were then imported into Marc.Mentat (MSC Software), an FEA software package, and the model was re-meshed into a solid model. Using a custom subroutine, heterogeneous material properties were assigned on an element-by-element basis. Relevant physiological forces were applied to the model, simulating the peak forces during gait. Results: We have created patient-specific FEA models based on CT scans after TKA. Models contained approximately 100,000 elements for the tibial bone section and 30,000 elements for the implant. Meshes were composed of four-node tetrahedral elements *1.8 mm in edge length. A typical mesh is shown in Fig. 1. Stresses at the bone-implant interface were simulated given normal gait cycle loadings. Models predicted areas of potential increased stresses and were subsequently correlated with SPECT outcomes. FE mesh of the tibia (blue) and implant component (green). Body weight forces are represented by purple arrows. Conclusions: Our FEA model provides a unique opportunity to selfvalidate patient-specific outcomes of TKA. Increased nuclide uptake has been associated with increased bone metabolic activity and bone remodelling and may be an indication of increased stress in a

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

Fig. 1 The graph shows the trend of the PCCF for the squat movement for the Genesis II primary design. a is the post-cam engagement, b is the PCCF for the maximum flexion angle. In the tabke are shown the maximum PCCF for each design for all analysed motor tasks. Maximum PCCF differ between motor tasks with higher flexion motor tasks showing increased maxima: lowest PCCF are observed for walking (287–311 N) while squatting resulted in the highest forces (841–902 N)

particular area. Increased nuclide uptake is also associated with patient reports of pain and may be a predictor of future poor outcomes. Imperfect anatomical implant alignment and patient-specific differences in joint anatomy may affect joint kinematics and loadings, causing pain, and potentially increased uptake of radionuclides. By modeling such effects in patients with unexplained pain post TKA by using FEA, revision surgeries could potentially take such differences into account and prevent future poor outcomes.

P32-1147 Usefulness of medial tibial condylar resection in total knee arthroplasty S.D. Cho1, Y.S. Youm1, J. Eo1, K.J. Lee1 1 Ulsan University Hospital, Department of Orthopedic Surgery, Ulsan, Republic of Korea Objectives: To demonstrate the usefulness of medial tibial condylar (MTC) resection in total knee arthroplasty (TKA) for varus osteoarthritis. Methods: Between September 2008 and February 2009, patients who underwent primary TKA for varus osteoarthritis and followed up more than 2 years were analyzed prospectively. After preliminary medial release and femoral-tibial bone resection, medio-lateral balance was checked. If there was narrow medial extension-flexion gap, tibial trial component was positioned on Akagi line and uncovered bony surface of medial tibial condyle was resected in 70 knees (50 patients). Medio-lateral balance was reassessed after MTC resection and patients were classified into group I with appropriate balance after only MTC resection and group II with persistent tight medial gap after MTC resection, so requiring additional medial release. Preoperative radiographic alignment between two groups was compared. Results: Mean age was 68 years (range, 53–81 years) and there were 5 men and 45 women. Preoperative mean varus deformity (tibiofemoral angle) was 10.5(range, 2–27). The kind of implants were 35 NexGen LPS and 35 Advanced Medial Pivot Knee system. Group I and II included each 35 cases (50%). Preoperative mean

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 varus deformity was 9.8(range, 2–22) in group I and 11.2(range, 3–27) in group II (p [ 0.05). There was no complication associated with MTC resection. Conclusions: In TKA for varus deformity, medio-lateral balance was obtained in 50% after MTC resection without downsizing. Preoperative varus deformity seemed to be severe in group II (additional release group) but there was no statistical significance. This MTC resection technique reduced the amount and incidence of soft tissue release required to balance the knee, thus minimizing the risk of medial over-release.

P32-1194 Preoperative assisted navigation for total knee arthroplasty (TKA): preliminary results V.C. Sansone1, N. Ursino1, C. Bonora1 1 Universita’ degli Studi di Milano, Orthopaedics and Traumatology, Istituto Ortopedico Galeazzi IRCCS, Milano, Italy Objectives: To achieve a successful and durable Total Knee Arthroplasty (TKA) the mechanical axis, with balanced flexion–extension gaps, must be restored. Preoperative assisted navigation aims to avoid the pitfalls associated with conventional alignment techniques and computer navigation, whilst maintaining the accuracy of computer assisted surgery. MRI scans generate a 3D model of the patient’s knee, on which patient-specific resin cutting jigs are moulded. These enable accurate pin placement for the standard resection instrumentation. The aim of this study was to evaluate early clinical, functional and radiographic results of TKA performed with preoperative assisted navigation. Methods: From January 2010 to date, 30 osteoarthritis patients, candidates for TKA, agreed to undergo the pre-operative MRI protocol. There were 21 women (70%) and 9 men (30%), average age of 72.5 years (range 55–82 years) and body mass index of 29.12 (range 26.2–33.5). Preoperative alignment was normal in 7 knees (23%), varus in 18 knees (60%), valgus in 5 knees (17%). Preoperative flexion contracture was observed in 14 knees (46.5%). All procedures were performed by the 2 senior Authors. A cemented posterior-stabilised prosthesis (Vanguard, Biomet Inc.) was implanted. We recorded the blood-loss (intra-operative, at 24 h, at 3 and 7 days), requirement for blood transfusions, operative time, and use of analgesic in the post-operative period. Any changes to the preoperative planning (e.g bone cuts, component sizes) were noted. Functional recovery was evaluated using the Functional Independence Measure (FIM) and the Barthel Index at 15 and 45 days and with the Knee Society function score at 45 days and at final followup. Radiographic and clinical results were evaluated with the Knee Society clinical score at 45 days and at final follow-up. Results: The average operative time was 73 min. (range 57–86). The peak of haemoglobin loss was at 3 days and averaged 3.7 gr (range 2.6–5.7). At 15-day follow-up, FIM score was 96.2 (range 82–100) and Barthel Index 120.3 (range 97–126). In all cases, radiographs showed satisfactory component position and restoration of normal alignment (2–7 valgus). In 5 cases, a 1-size smaller femoral component was implanted. Conclusions: Our preliminary results in 30 TKAs show good restoration of limb alignment and component rotation in all cases. Component sizing seems to be less accurate, although the choice of a 2 mm-smaller femoral component may be highly subjective. Operative time is similar to that of conventional technique, although the first cases were affected by a learning curve and the prudence of the surgeon who double-checked the positioning of the jigs using conventional means. Further larger controlled studies, with longer followup and a proper analysis of cost-effectiveness will be necessary to

S331 validate this new system, although our initial results seem to be encouraging.

P32-1237 Should the presence of menisci be considered when measuring the posterior tibial slope in total knee arthroplasty ? G. Cinotti1, P. Sessa1, F.R. Ripani1, A. Della Rocca1, W. Salustri1 1 Orthopaedic Clinic University La Sapienza, Rome, Italy Objectives: In planning a total knee arthroplasty, the sagittal tibial slope is commonly measured on a lateral view radiograph. Such a measurement does not include cartilage and menisci, although these structures have a relevant role in loading distribution on the tibial plateau. Aims of this investigation were to assess whether the posterior tibial slope is affected by the presence of cartilage and menisci and whether any correlation exists between the degree of tibial bone slope and cartilage and meniscal thickness. Methods: Magnetic resonance studies of the knees of 80 subjects, 45 males and 35 females, with an average age of 38.9 years, were analysed. Using an imaging visualization software, the sagittal anatomical axis of the tibia was identified. The angle between the latter and a line tangent to the bone profile of the tibial plateau (bone slope) and a line tangent to the superior border of the anterior and posterior portion of the menisci (meniscal slope) were calculated. Results: The tibial bone slope averaged 8 on the medial side and 7.7 on the lateral side. The meniscal slope averaged 4.1 on the medial side and 3.2 on the lateral side. The meniscal slope was significantly reduced compared to the bone slope, on both the medial and lateral sides. A significant correlation was found, on the lateral side, between the severity of the posterior bone slope and the difference between the bone and meniscal slope. Conclusions: The presence of cartilage and menisci leads to a significant reduction in the posterior tibial slope measured on the bone profile. On the lateral side, the greater the posterior bone slope, the larger the difference between the bone and meniscal slope.

P32-1257 Quadriceps inhibition and physical function after unilateral total knee arthroplasty A. Alnahdi1, J. Zeni2, L. Snyder-Mackler2 1 University of Delaware, Biomechanics and Movement Science Program, Newark, United States, 2University of Delaware, Department of Physical Therapy, Newark, United States Objectives: The aim of this study was to determine if patients with quadriceps activation deficit (inhibition) are more functionally limited compared to patients without quadriceps inhibition after unilateral Total Knee Arthroplasty (TKA). Methods: Patients with primary unilateral TKA for knee osteoarthritis were recruited for this study. The burst superimposition technique was used to test quadriceps strength and activation. Subjects were seated on an isokinetic dynamometer (Kincom, chattecx Corp, Harrison, TN) with the knee flexed to 75. Two self-adhesive electrodes were placed over the quadriceps muscle. Subjects were instructed to produce their maximal isometric contraction, and then a stimulator delivered a supramaximal burst of electrical stimulation. The peak volitional force generated before the delivery of the burst was used to quantify quadriceps strength. Voluntary activation was quantified using the central activation ratio (CAR). CAR was calculated by dividing the peak volitional force by the peak force produced by the combination of volitional effort and the superimposed burst. CAR of 1 indicates full activation of the muscle. Physical function was assessed using the Knee Outcome Survey-Activities of Daily Living Scale (KOS),

123

S332 Times Up and Go Test (TUG), Stair Climbing Test (SCT), and sixMinute Walk Test (6 MW). Subjects were divided into two groups based on quadriceps activation level. Subjects with CAR C 0.95 were considered not to have activation deficit and were assigned to ‘‘Full Activation’’ (FA) group. Subjects with CAR \ 0.95 were considered to have activation deficit and were assigned to ‘‘Incomplete Activation’’ (IA) group. Independent t test was used to examine between group differences in self-report and performance-based measures of physical function. Results: 159 subjects (48% women) participated in the study with age (67.4 ± 7.6 yr), BMI (31 ± 5.6), time from surgery (12.6 ± 7.6 months). 40% (N = 63) of the sample had full quadriceps activation while 60% (N = 95) had an activation deficit. Mean quadriceps activation level was 0.97 ± .02 for the FA group and 0.84 ± .09 for the IA group. The IA group exhibited a 25% quadriceps strength deficit compared to the FA group (P \ 0.001). Age, BMI, time from surgery, and sex proportions were not different between the FA and IA groups. Self-reported outcome scores were not different between the FA (85% ± 12%) and IA (85% ± 12%) groups (P = 0.93). The IA group needed 14% more time to complete the TUG (P \ 0.001) and SCT tests (P = 0.015) compared to the FA group. In the 6 MW test, the IA group covered 10% less distance compared to the FA group (P \ 0.001). Conclusions: After unilateral TKA, the majority of patients who are considered fully recovered still demonstrate quadriceps activation deficit. Patients who are unable to fully activate the quadriceps are more limited in physical performance-based functional tests compared to patients who are able to fully activate. Self-reported function does not seem to be influenced by the activation level.

P32-1330 The relationship of joint line and flexion/extension axes of the knee to the mechanical axis in the coronal plane R. Streicher1, J. D’Alessio2, A. Patel2, M. Kester2, D. Jacofski3 1 Stryker SA, Montreux, Switzerland, 2Stryker Orthopaedics, Mahwah, United States, 3CORE Institute, Phoenix, United States Objectives: Classical and Anatomic Alignment rely on various axes for component orientation in TKA. Studies indicated that the knee has a single flexion/extension axis but debated the location of this axis in the femur. The relationship of the flexion/extension axis in the coronal plane to the mechanical axis has received little attention. The purpose of this study was to investigate the relationship of the various axes and references with respect to the mechanical axis in the coronal plane utilizing 3D virtual models of human subjects, along with their relationship or variation to a neutral distal resection. Methods: Subjects were prospectively scanned into a Virtual Bone Database (Stryker Orthopaedics, Mahwah, NJ), which is a collection of body CT scans from subjects collected globally. CT Scans of the Left Knees (n = 263), Right Knees (n = 256), and combined left/ right knee (n = 519) were collected for this study. Results: The combined (left/right) mean femoral joint angle was 86.1 ± 2.0 (Range: 80.2–92.2). The combined (left and right) mean Femoral TransepicondylarAxis (TE) angle was 88.8 ± 2.5 (Range: 81.7–98.4). The combined (left/right) mean Femoral Posterior Cylindrical (PC) angle was 87.9 ± 2.2 (Range: 81.8–94.0). The average deviations from a neutral resection were 3.8, 1.2 and 2.1 for the Femoral Joint Angle, Femoral TE Angle, and Femoral PC angle respectively. The mean Femoral Joint angle had the lowest variability, while the mean Femoral TE angle showed the largest. Conclusions: On average, the transepicondylar axis and the posterior cylindrical axis were approximately perpendicular to the mechanical axis in the coronal plane, although distances to the insertion of the collateral ligaments (epicondyles) were not equal to the distal and posterior surfaces throughout the knee range of motion. Although

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 surgeons do not normally attempt to align components in the coronal plane specifically to either axis, this data suggests that the average value is within the accepted ± 3 range reported in similar studies for overall positioning. The Femoral PC values are closer to the values of the femoral joint line when compared to the Femoral TE. The Femoral PC may be a more reproducible landmark as it may be determined by either preoperative imaging or intraoperatively from instrumentation that references the distal and posterior surfaces. The use of the posterior cylindrical axis may have substantial implications for optimizing flexion, midflexion, and extension balance in TKA, especially with single radius knee designs. The use of this axis may allow for more biomechanically symmetric gaps throughout the range of motion of the knee.

P32-1347 Periprosthetic fractures following total knee replacement. A 17 cases review O. Ares1, S. Prat1, A. Carren˜o1, P. Camacho1, R. Garcia1, A. Zumbado1 1 Hospital Clinic de Barcelona, University of Barcelona, Orthopedic Department, Barcelona, Spain Objectives: Periprosthetic fractures following total knee arthroplasty is a problem of growing concern given that these associated fractures have been reported to range between 0.3 and 2.5%, and total knee arthroplasty is one of most common surgeries in the developed world. Its treatment and correct management may be challenging for orthopedic surgeons, and several different therapeutic options have been studied. Our aim is to evaluate osteoporotic risk fracture through Fracture Risk Assessment Tool (FRAX) algorithm and postoperative results regarding quality of life, which we measured with Knee Society Score (KSS). Methods: We retrospectively reviewed 17 cases of knee periprosthetic fractures admitted to our hospital between 2005 and 2010. We examined the sample assessing in greater osteoporotic risk fracture index and osteoporotic hip risk fracture. For this evaluation we used FRAX algorithm, whereas functional outcomes were measured using KSS. Type and situation of the knee prosthesis, fracture management, consolidation time, reoperation rate and complications were recorded. Results: From the initial sample of 17 patients, 16 cases were fixed with locking plates and 1 case required revision of the femoral component. The average consolidation time was 8.2 months. Five cases (35.71%) required some type of reoperation. The final review for these 17 patients was 1 death, 1 changed of city, 1 on vacation, 8 were not found on their phone and 6 were the total patients that completed the questionnaires. According to the FRAX algorithm, at the time of fracture, the 10-year probability of osteoporotic fractures in our patients was a mean of 10.19% and hip risk fracture was 4.9%. After treating the periprosthetic fracture, KSS for objective evaluation was 47.43 (SD 12.38), whereas functional evaluation was 34.29 (SD 25.07). Total KSS was 76 (SD 44.78). Conclusions: Knee periprosthetic fractures occur in a population with high-risk major osteoporotic fracture associated ([10%). Given a correct surgical management of the fracture, good functional and clinical results are seen in these patients.

P32-1489 Novel antomical landmark for rotational alignment of tibial component in TKA S.-S. Seo1, C.-W. Kim2, Y.-J. Kim2, D.-W. Jung2 1 Busan Paik Hospital, Inje University, Orthopaedics, Busan, Republic of Korea, 2Inje University, Busan, Republic of Korea Objectives: To investigate new useful bony landmark of the anterior and posterior tibial axis in total knee replacement arthroplasty.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: From October 2006 to July 2008, we took computed tomography of 132 people who have a no deformity of a hip and ankle joints, then computerized them. We defined medial–lateral axis by femoral transepicondylar axis and transferred to tibial plateau surface. From that line, the line runs through posterior notch defined as anterior-posterior axis. We measured a (the angle between AP axis and tibial tuberosity medial one-third space), b (the angle between tibial anterior border at the tibial mid-shaft area) c (the angle between anterior tibial border at the superior surface of ankle joint) as degree of curl. Results: Included angle between the line passes through posterior notch 1/3 point of tibial tuberosity and the anterior and posterior tibial axis (a) checked 19.1 ± 4.4(8.3 * 28.7) on average and rotation degree was 25 on average in 110 people (83.3%) in computerized tomography. The angle between the line runs from posterior notch to tibial axis in the middle part of tibial shaft and the front and rear tibial axis (b) was measured 9.5 ± 3.4(1.6 * 18.5). It appeared a significant difference in Wilcoxon signed rank test. (p \ 0.001), It showed no statistical meaning that the included angle, between the line runs from the posterior notch to the anterior border of tibia and the anterior and posterior tibial axis which measured from the superial part of ankle joint (c) was measured 2.5 ± 6.9(-5.8 * 14.3). Conclusions: In case of TKRA, the middle point of anterior border of tibial axis is a useful indicator compared to the medial one-third of tibial tuberosity. This will be a good indicator because it’s easy and simple to measure.

P32-1497 Rehabilitation after total knee arthroplasty: an evaluation of different regimes C. Stukenborg-Colsman1, M. Thomson2, P. Virolainen3, E. Guerado4, P. Hinarejos5, K. Radtke1 1 Orthopa¨dische Klinik der Medizinischen Hochschule, Hannover, Germany, 2DRK Krankenhaus Baden–Baden, Baden–Baden, Germany, 3Turku University Central Hospital, Turku, United Kingdom, 4Hospital Costa del Sol, Marbella, Spain, 5Parc de Salut Mar. Universitat Auto`noma de Barcelona, Barcelona, Spain Objectives: In the literature a high variance in rehabilitation regimes after Total Knee Arthroplasty (TKA) is described, but the number of prospective randomized controlled trials evaluating the outcome after different postoperative management are limited. Early effects (e.g. shorter length of hospital stay, higher ROM, better function) were reported after functional training, but all effects did not persist until 1 year. The aim of the presented study is to evaluate the outcome after six different rehabilitation regimes in six European study centers performing a multi-center study after TKA. Methods: 188 knee arthroplasties with the Triathlon knee prosthesis system (Stryker, Limerick, Ireland) were evaluated in six European study centers (A-F) in the Triathlon CR International Multicenter Clinical Study. For all study centers the postoperative management strategies were evaluated. Duration, intensity and start of the programs differed between all centers. Follow-up was at three months and 1 year postoperative for all patients in the study. At every followup outcome measures were evaluated with range of motion (ROM) as the most important outcome parameter. Results: At 1 year postoperative follow-up, two patients had undergone revision surgery; The results of 145 patients after 1 year followup were evaluated. The mean age was 66 years (range 44–80 years) and mean body-mass-index (BMI) was 30.0. The mean preoperative ROM was 104.5 (SD 15.6). The mean postoperative ROM was 104.6 (SD 13.7) three months after TKA, and 109.6 (SD 13.5) at 1 year postoperative. There was a significant difference in the Length of hospital Stay (LOS) between the centers due to the different rehabilitation strategies and release criteria. LOS was significantly

S333 correlated (p \ 0.05) with KSS pain scores after 3 months and 1 year of follow-up as well as with the KSS function and the WOMAC scores after 1 year of follow-up. Conclusions: The postoperative range of motion is one of the indicators for the success of Total Knee Arthroplasty. In this study the mean postoperative ROM was not different between centers and showed a low variance. This indicates good clinical results in all study centers taking ROM as the main outcome parameter. However, LOS was related to short term pain and medium term function scores of patients after TKA in this study. The reported results indicate that a complex treatment can lead to excellent results in some patients; meanwhile others seem to benefit from a more restrictive management. The aim of following studies should be to identify those patients.

P32-1502 Two-stage revision of infected total knee arthroplasty: mid-term results D. Celik1, O¨. Kilicoglu1, R. To¨zu¨n2, O¨. Yazicioglu1 1 Istanbul University, Faculty of Istanbul Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 2Acibadem Hospital Orthopedics and Traumatology, Istanbul, Turkey Objectives: Two stage revision arthroplasty is the gold-standard for treatment of infected total knee arthroplasty. This study questions the efficacy of this method in treating infection and evaluates the functional results in a mid-size case series. Methods: Twenty-two knees of 21 patients [mean age 74, (range 49–88)] treated with two stage revision for infected total knee arthroplasty were evaluated retrospectively after a mean follow-up of 82 months (range 12–203). Custom made antibiotic spacers loaded with gentamicin, teikoplanin or both were used. Mean interval between two stages was 48 weeks (range 3–156). All femoral and 9 tibial components were implanted using stems at revision surgery. Five femoral blocks and 4 tibial wedges were used. Tibial tubercle osteoteomy was required in four cases. One patient was revised with constrained knee prosthesis. Last available erythrocyte sedimentation rate (ESR) and C-Reactive Protein (CRP) levels of 22 patients were compared with their preoperative values. Implant alignment was evaluated on the early postoperative radiographs and findings of loosening on the last available radiographs. Ten knee joints of 9 patients were also reevaluated clinically. Functional status of these patients was evaluated with WOMAC (Western Ontario and McMaster Universities Osteoarthritis Index) and HSS (Hospital for Special Surgery) scores, pain with visual analogue scale (VAS) and range of motion (ROM) with goniometric measurements. Results: Three knee joints (13.6%) were revised with knee fusion procedure due to infection recurrence at an average of 161 months. Remaining 19 knees were found to be free of infection findings at their last follow up. Four patients died because of non-infectious causes, and 2 were living bedridden due to poor health conditions. Three were contacted with phone, and no complaints related to infection were noted. The remaining 10 knee joints of 8 female and 1 male patients (mean age 71; range 63–87) were re-evaluated after an average of 92 months (range 12–203 months). Average WOMAC score was 69.4, HSS knee score was 53.7 and HSS function score was 52.7. Mean pain level, assessed with VAS was 1.6 at rest and 2.8 with activity. Average total ROM was measured as 60 with a mean extension deficit of 2.2. Radiological findings of loosening were observed in two components, without accompanying clinical symptoms. Average CRP and ESR levels prior to first stage were 68 mg/L (range 11–199 mg/L) and 71 mm/h (range 17–150 mm/h), respectively. Immediately before the second stage, these values were 21 mg/L (range 2–93 mg/L) and 43 mm/h (range 9–98 mm/h), and at the last follow up 49 mg/L (range 4–68 mg/L) and 48 mm/h (range 17–78 mm/h), respectively.

123

S334 Conclusions: An infection free survival can be expected in 6 of 7 patients following two-stage revion of infected knee arthroplasty within 82 months. Although functional results are inferior compared to published primary knee arthroplasty results, high satisfaction rate of patients can be explained with significantly lower pain levels.

P32-1503 Two-stage revision of infected total hip arthroplasty: mid-term results D. Celik1, H. Cil1, O¨. Kilicoglu1, O¨. Yazicioglu1, R. To¨zu¨n2, B. Gu¨lenc1 1 Istanbul University, Faculty of Istanbul Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 2Acibadem Hospital Orthopedics and Traumatology, Istanbul, Turkey Objectives: Two stage revision arthroplasty is the gold-standard for treatment of infected total hip arthroplasty. This study questions the efficacy of this method in treating infection and evaluates the functional functional results in a mid-size case series. Methods: Thirty-seven patients [mean age 65, (range 35–86)] treated with two stage revision for infected total hip arthroplasty were evaluated retrospectively after a mean follow-up of 91 months (range 12–204). Custom made antibiotic spacers loaded with gentamicin, teikoplanin or both were were used. Mean For 5 cases, ring was used for acetabular revision. Three cases required trochanteric grip, 2 cases required plate fixation and 1 case required allograft for femoral revision. One case was revised with trochanteric osteotomy. Last available erythrocyte sedimentation rate (ESR) and C-Reactive Protein (CRP) levels of 22 patients were compared with their preoperative values. Implant alignment was evaluated on the early postoperative radiographs and findings of loosening on the last available radiographs. Ten hip joints of 22 patients were also reevaluated clinically. Functional status of these patients was evaluated with Harris Hip Score, pain with visual analogue scale (VAS) and range of motion (ROM) with goniometric measurements. Results: Eight hips (21.6%) required girdlestone procedure due to infection recurrence at an average of 77 months. One patient died because of infection related conditions after the first year of surgery. Four patients died due to non-infectious causes and 2 patients were living bedridden due to poor health conditions. The remaining 22 hip joints of 14 female and 8 male patients (mean age 62; range 35–85) were re-evaluated after an average of 96 months (range 24–204 months). Average Harris Hip Score was 69.4. Pain was assessed with VAS (Visual analogue scale); average score at rest was 1.6 and with motion was 2.8. Average total ROM was measured as; 75.8 flexion, 36.9 abduction, 21.6 external rotation and 20.2 internal rotation. Radiological findings of loosening were observed in two cases prior to 1st stage surgery, however none of the re-evaluated cases has loosening signs at the last follow up. Average CRP and ESR levels prior to 1st stage were 68 mg/L (range 11–199 mg/L) and 71 mm/h (range 17–150 mm/h), respectively. Immediately before the second stage, these values were 21 mg/L (range 2–93 mg/L) and 43 mm/h (range 9–98 mm/h), and at the last follow up 49 mg/L (range 4–68 mg/L) and 48 mm/h (range 17–78 mm/h), respectively. Average C-reactive Protein (CRP) and eritrocyte sedimentation rate (ESR) were 23 mg/L and 66 mm/h respectively prior to 1st stage. After the 1st stage they were calculated as 24 mg/L and 47 mm/h, and the values measured at the last follow up of the patients were 7 mg/L and 32 mm/h respectively. Conclusions: An infection free survival can be expected in 78% of the patients following two-stage revion of infected hip arthroplasty within 91 months. Although functional results are inferior compared to published primary hip arthroplasty results, high satisfaction rate of patients can be explained with significantly lower pain levels. Keywords: Total hip arthroplasty, Iinfection, Revision

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P32-1505 Prospective comparison of tensioned gaps versus posterior condylar referencing in determining femoral component rotation in TKA T. Luyckx1, J. Victor2, T. Peeters1, H. Vandenneucker3, J. Bellemans1 1 UZ Leuven, Orthopaedic Surgery, Pellenberg, Belgium, 2UZ Gent, Orthopaedic Surgery, Gent, Belgium, 3UZ Leuven, Orthopaedic Surgery, Pellenberg-Lubbeek, Belgium Objectives: Obtaining a balanced flexion gap with a correct femoral component rotation is one of the prerequisites for successful outcome after total knee arthroplasty (TKA). Different techniques for achieving this have been described. In this study, we compared a tensor device versus posterior condylar referencing in terms of reliability and accuracy towards femoral component rotation in TKA. To our knowledge, this is the first study to compare these techniques in a prospective controlled trial. Methods: We prospectively studied 96 consecutive primary TKA cases at our department. In 48 patients, a tensor device was used to determine rotation of the femoral component. In the second group of 48 patients, a posterior condylar referencing technique was used. Pre- and postoperative radiographs as well as CT-scan measurements were performed. Pre- and postoperative mechanical alignment, native rotational geometry of the distal femur and rotation of the femoral component were determined. All measurement were performed twice by three independent observers. For all 96 cases, the need for per-operative soft tissue releases was noted. Power analysis showed us a 80% power to detect a rotational difference of 1.35 between both groups. To detect a difference of 3 (defined as ‘‘clinically relevant’’) the power of the current study exceeded 99.9%. Results: Both groups systematically reproduced a similar external rotation of the femoral component. The mean postoperative external rotation of the femoral component was 1.68 (SD 0.30) in the posterior condylar referencing group. In the tensioned gap group, the mean external rotation was 2.39 (SD 0.36). There was a tendency towards more external rotation in the second group. However, this difference was not statistically significant. The Spearman correlation coefficient showed a strong influence of the pre-operative rotational geometry of the distal femur on the femoral component rotation in both groups. This correlation is stronger in the posterior condylar referencing group (0.63; p \ 0.0001) than in the tensioned gaps group (0.45; p \ 0.001). This correlation was not significantly different between the two groups. In the posterior condylar referencing group but not in the tensioner group, there was also a significant influence of the pre-operative coronal alignment (-0.37; p \ 0.01). There was no statistical significant difference in the number of outliers between the two groups. For none of the ligamentous releases, there was statistical evidence that the percentage of releases differed between the two groups. Conclusions: Both tensioned gaps and posterior condylar referencing techniques are equally reliable and accurate in determining femoral component rotation in TKA. There was a tendency towards more external rotation in the tensioned gaps group. However, this difference was small (0.71) and not statistically significant.

P32-1512 Ten year survival after unicompartmental knee arthroplasty G. Bontemps1, K. Schlu¨ter-Brust2 1 Fabricius Klinik, Orthopa¨die, Remscheid, Germany, 2Klinikum der Universita¨t zu Ko¨ln, Klinik und Poliklinik fu¨r Orthopa¨die, Ko¨ln, Germany

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: A prospective, long-term follow-up study: 10 Year Survival After Unicompartmental Knee Arthroplasty. Methods: The following prospective study describes the clinical outcome and the 10 years survival rates in a series of 242 cemented and uncemented medial Unicompartmental Knee Arthroplasty (UKA) cases performed in 236 patients for medial osteoarthritis. Details from patients with unicompartmental knee arthroplasty were recorded and updated on an annually base from 1991 up to the present. Patients were assessed by an independent clinical observer using the Knee Society Rating System as a validated outcome measure. Kaplan– Meier analysis was used to calculate the 10 year survival rates using the endpoint of revision for any cause. Results: The mean elapsed time since the day of surgery was 10.4 years (maximum 18.7 years). There were no failures due to progression of lateral osteoarthritis, aseptic loosening of the femur component or due to polyethylene wear. There had been thirteen surgeries for revision because of failures for any reason and 41 patients had been withdrawn because they had died, giving an all over cumulative survival rate at 10 years (knees at risk = 201) of 94.07%. The Knee Society Rating System (KSRS) showed a significant improvement. The knee (function) score showed an increase from pre 33.5 (54.7) to post operative 94 (83.6) Points. The Range of Motion gained in average from 106.8 to 122.3. We could not detect a significant difference in the 10 year survival rate of patients with a BMI \ 30 (95.59%), BMI from 30 to 36 (92.39%) and a BMI [ 36 (100%). Conclusions: Given strict indication criteria’s and appropriate surgical expertise, UKA has a high survival rate comparable to TKA and shows a significant improvement in knee and function scores. The results indicate that there is no relationship between BMI and 10 year survival rate in this cohort.

P32-1516 Medial unicompartmental arthroplasty with a mobile bearing implant: Clinical and radiographic results of 463 knees at a Canadian center R.S.J. Burnett1, R. Nair1, D. Jacks1, C. Hall1, M. McAllister1 1 Vancouver Island Health, South Island, Division of Orthopaedic Surgery, UBC/UVIC Island Medical School, Victoria, Canada Objectives: Unicompartmental knee arthroplasty (UKA) is frequently performed in the treatment of isolated medial compartment osteoarthritis (OA) in patients with anteromedial OA and an intact ACL with favourable results. The results of this implant and technique have been reported by the implant designers and by non-implant -design centers with different clinical outcomes in the literature utilizing similar indications and surgical technique. A retrospective study was undertaken at a Canadian center to assess the clinical and radiographic outcomes of 463 consecutive Oxford Phase 3 mobile bearing medial UKA. This study is the largest non-implant designer cohort reported to date. Methods: A retrospective review of 463 consecutive Oxford medial Phase 3 UKA’s (382 patients) was performed at a single Canadian tertiary care center over a 10-year period. Patients were followed clinically and radiographically and no patients were lost to follow-up. All patients were evaluated at most recent follow-up by an independent surgeon other than the treating surgeon. Outcome measures included: Knee Society Clinical Rating Score (KSCRS), Oxford Knee Score, WOMAC, SF-12, reasons for revision and reoperation, radiographs including mechanical axis alignment, and patient satisfaction. A comprehensive statistical analysis including predictors of failure and reoperation were performed by a statistician. Results: The mean age at surgery was 69 years (range, 38–88). There were 218 women (57%) and 164 men (43%). At a mean follow-up of 3.9 years (range, 0.1–8.3), the final WOMAC score was 82, SF-12

S335 physical score was 47 and mental score was 51. The overall reoperation rate was 6.7% (31 knees). Twenty-three knees were revised to TKA at a mean of 35.6 months (range, 7–92) most commonly for progression to lateral compartment OA (10 knees). Eight knees were revised for tibial loosening, four for femoral loosening, and one for PCL failure. Five further knees were deemed a failure and are pending revision to TKA for an overall revision rate of 6.1% (28 knees). Three knees underwent polyethylene liner exchange for bearing dislocation and two knees had further arthroscopic procedures. Improvements in preoperative compared to postoperative scores for ROM KSCRS Pain and Function, and Oxford Knee Scores were significant (p \ .05). Overall 92% of the patients were extremely or very satisfied with the outcome and 96% would have the procedure again. Conclusions: UKA surgery with the medial Oxford Phase 3 implant is associated with high patient satisfaction and a favourable revision rate at medium term. Progression to lateral compartment OA was the most common reason for revision. Predictors of failure included: over correction to a valgus mechanical axis, thickness of the polyethylene, and thickness of tibial bone resection. The results of this study are similar to those reported by the implant designers utilizing the same selection criteria and implant design.

Leg/ankle

P33-16 A hybrid anatomic lateral ankle reconstruction: ATFL autograft with plication of the CFL C. Murawski1, J. Kennedy1 1 Hospital for Special Surgery, Orthopaedic Foot and Ankle Surgery, New York, United States Objectives: The lateral ankle sprain is the single most common sports injury worldwide. Surgical strategies for addressing lateral instability include anatomic reconstruction and checkrein procedures. Concerns over inadequate reparative tissue, scarring and over tightening of the subtalar joint have prompted the introduction of a hybrid reconstruction. Using a peroneal tendon autograft fixed to the isometric points of the ATFL, and plicating rather than substituting the CFL provide the benefits of both techniques while reducing the drawbacks of both. The current study evaluates a hybrid of the lateral ankle ligament complex. Methods: Between 2006 and 2009, 57 patients underwent a hybrid lateral ankle ligament reconstruction technique. Each patient failed a 3-month conservative triple-phase therapy program. All patients were followed for a minimum of one-year following surgery and were treated in identical fashion. Surgery included substituting the native ATFL with a 4 cm split peroneus longus autograft in addition to a vest-over-pants plication of the CFL. All patients had pre- and postoperative Foot and Ankle Outcome Scores (FAOS) and Short Form36v2 scores. Pre- and post-operative MRIs were compared to evaluate ankle and subtalar arthrosis. Results: FAOS scores increased significantly pre- to post-operatively from 58 to 89 points. SF-36v2 scores also increased significantly from 67 points pre-operatively to age adjusted normal levels. Two of 57 patients had pre-operative grade II cartilage loss in the posterior facet of the subtalar joint. In one case, this had advanced to grade 3 at 2 years follow-up. Two additional patients had grade I changes in the subtalar joint at 2 years and one patient demonstrated grade II changes in the ankle joint at 1-year follow-up. All patients reported competing at some level of athletic sport prior to surgery. 5 of 57 patients did not return to pre-operative sporting levels. All 5 patients had mechanical stability but all had

123

S336 functional instability. The incidence of functional instability was 22% overall and persistence of functional instability was a predictor of failure to return to sports. Complications included a painful hypertrophic peroneal tendon, two cases of superficial peroneal nerve neurapraxia and a post-operative sinus tarsi syndrome. Conclusions: Traditional anatomic reconstructions of the lateral ligament complex have demonstrated good outcomes. The hybrid procedure described in the current study may provide an alternative to the Brostrom procedure when inadequate reparative tissue limits a direct repair. Functional stability training is critical to facilitate a full return to sports.

P33-17 Functional and quantitative T2-mapping results of autologous osteochondral transplantation in 72 patients C. Murawski1, J. Kennedy1 1 Hospital for Special Surgery, Orthopaedic Foot and Ankle Surgery, New York, United States Objectives: Osteochondral lesions of the talus are common injuries following acute and chronic ankle sprains and fractures, the treatment strategies of which include both reparative and restorative techniques. Recently, restorative techniques (i.e., autologous osteochondral transplantation) have been become increasingly popular as a primary treatment strategy, in part due to the potential advantages of replacing ‘‘like with like’’ in terms of hyaline cartilage at the site of cartilage repair. The current study examines the functional results of autologous osteochondral transplantation of the talus in 72 patients. Methods: Between 2005 and 2009, 72 patients underwent autologous osteochondral transplantation under the care of the care of the senior author. The mean patient age at the time of surgery was 34.19 years (range, 16–85 years). The mean follow-up time was 28.02 months (range, 12–64 months). Patient-reported outcome measures were taken pre-operatively and at final-follow-up using the Foot and Ankle Outcome Score and Short-Form 12 general health questionnaire. Quantitative T2-mapping MRI was also performed on select patients at 1-year post-operatively. Results: The mean FAOS scores improved from 52.67 points preoperatively to 86.19 points post-operatively (range, 71–100 points). The mean SF-12 scores also improved from 59.40 points pre-operatively to 88.63 points post-operatively (range, 52–98 points). Three patients reported donor site knee pain after surgery. Quantitative T2mapping MRI demonstrated relaxation times that were not significantly different to those of native cartilage in both the superficial and deep halves of the repair tissue. Conclusions: Autologous osteochondral transplantation is a reproducible and primary treatment strategy for large osteochondral lesions of the talus and provides repair tissue that is biochemically similar to that of native cartilage on quantitative T2-mapping MRI. This may ultimately allow the ankle joint to function adequately over time.

P33-18 Chevron-type medial malleolar osteotomy: a functional, radiographic and quantitative T2-mapping MRI analysis J. Lamb1, C. Murawski2, T. Deyer3, J. Kennedy2 1 Hospital for Special Surgery, New York, United States, 2Hospital for Special Surgery, Orthopaedic Foot and Ankle Surgery, New York, United States, 3East River Medical Imaging, New York, United States Objectives: The purpose of this study was to retrospectively evaluate a large series of patients having undergone a Chevron-type medial malleolar osteotomy with parallel and transverse screw fixation.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Specifically, we sought to quantitatively evaluate the cartilage infill at the tibial interface of the osteotomy using T2-mapping magnetic resonance imaging (MRI) and compare it to normal cartilage. Methods: Between 2005 and 2009, 62 patients underwent a Chevrontype medial malleolar osteotomy. The mean follow-up time period was 30.2 months (range, 24–71 months). Standard digital radiographs were used to determine bony union as the presence of continuous trabeculation at the osteotomy site. The angle of the osteotomy was also defined relative to the longitudinal axis of the tibia by two observers using previously established methods. MRI with quantitative T2-mapping was performed on 32 patients. T2 relaxation values were measured in both the superficial and deep chondral layers at the site of the osteotomy and compared to adjacent normal cartilage over the tibial plafond. Results: At final follow-up, fifty-eight patients (94%) reported being asymptomatic at the site of the medial malleolar osteotomy. Postoperative radiographs demonstrated that each patient was anatomically reduced with no evidence of shift in fixation or migration of the osteotomy fragment. The mean time to healing on standard radiograph was 5 weeks (range, 4–6 weeks). The mean osteotomy angle relative to the long axis of the tibia for the overall population was 31.70 ± 6.91. The Interclass Correlation Coefficient of variability between observers was good (0.86). Quantitative T2-mapping MRI analysis demonstrated that the deep half of interface repair tissue had relaxation times that were not significantly different from normal tibial cartilage. In contrast, interface repair tissue in the superficial half demonstrated significant prolongation from normal relaxation time values, indicating a more fibrocartilaginous repair tissue. Three patients (5%) reported persistent hardware-related pain three months after surgery that were easily addressed. A fourth patient developed a subchondral cyst on the distal tibia at the interface of the osteotomy. This patient received an injection of platelet-rich plasma and reported being asymptomatic after 12 weeks but failed to demonstrate resolution of the cyst on MRI. Conclusions: A Chevron-type medial malleolar osteotomy with transverse and parallel screw fixation demonstrates satisfactory healing and fixation without evidence of migration of the osteotomy fragment. The prolongation of T2 stratification in the superficial layer is likely the result of chondral injury following the osteotomy. Given the normalization of T2 values in the deep layer, it would be interesting to see whether this cohort followed longitudinally has normalization of the superficial values, suggesting continued healing of the cartilage over the osteotomy.

P33-31 A single platelet-rich plasma injection for the treatment of chronic insertional Achilles tendinopathy C. Murawski1, H. Newman2, J. Kennedy1 1 Hospital for Special Surgery, Orthopaedic Foot and Ankle Surgery, New York, United States, 2Hospital for Special Surgery, New York, United States Objectives: Chronic insertional Achilles tendinopathy is a common problem among weekend warriors and athletes alike. Common conservative treatment strategies include a combination of rest, physical therapy and/or bracing. However, if these modalities fail, surgery with a prolonged recovery time is often recommended. The current study retrospectively evaluated the treatment of chronic insertional Achilles tendinopathy with a single platelet-rich plasma (PRP) injection. Methods: Between August 2009 and February 2010, 20 patients received a PRP injection for the treatment of chronic insertional Achilles tendinopathy after having failed a 3-month conservative treatment plan. All patients completed pre and post-injection Foot and Ankle Outcome Scores (FAOS) and the SF-12 general health

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 questionnaire at 6 months. All patients also had pre- and post-injection MRI at 6 months. Following injection, patients were placed in a CAM walker for two weeks and then underwent physical therapy including functional isometric training for a further two months. Results: 10/19 patients (53%) reported a complete resolution of symptoms with an increase in mean FAOS and SF-12 outcome scores pre- to post-injection of 44–83 points and 59–86 points, respectively at 6 months follow-up. The remaining 9 patients (43%) did not improve or got worse and ultimately required surgery in the form of a Haglund’s resection and tendon debridement. Post-PRP MRI findings did not correlate with patient symptoms (r2 = 0.29). There was no correlation between age or gender and patient outcome (r2 = 0.16). Conclusions: The treatment of chronic insertional Achilles tendinopathy with a single injection of PRP appears to have a beneficial effect on approximately half of those undergoing treatment. Chroncity, gender and age, however, are not found to be predictors of who will benefit from PRP. Prospective randomized controlled trials are of paramount importance to establish the clinical efficacy of PRP.

P33-124 Simultaneously performed ankle arthroscopy and hindfoot endoscopy for the treatment of severe ankle impingement J. Sasahara1, M. Takao2, T. Matsushita2 1 Teikyo Institute of Sports Sciences and Medicine, Tokyo, Japan, 2 Teikyo University, Orthopaedic Surgery, Itabashi, Japan Objectives: Ankle arthroscopy for anterior ankle impingement and hindfoot endoscopy for posterior ankle impingement provide the surgeon with good results, respectively. However, the efficacy of ankle arthroscopy and hindfoot endoscopy for severe ankle impingement combined anterior and posterior osteophytes (McDermott’s classification: grade IV) remains unclear. The purpose of this study was to evaluate the short-time results of simultaneously performed ankle arthroscopy and hindfoot endoscopy for the treatment of severe ankle impingement. Methods: Twelve patients (15 ankles) with severe ankle impingement who had simultaneously performed ankle arthroscopy and hindfoot endoscopy were included in this study, including 11 men and 1 woman with a mean age of 30.9 years. The time to return to sports activities and complications were recorded. The American Orthopedic Foot and Ankle Society (AOFAS) score and the range of motion of the ankle were obtained before surgery and at the final follow-up (mean, 14.3 months). Results: All 15 ankles had anterior talofibular ligament (ATFL) injuries and osteoarthritic changes. Seven ankles underwent anatomical ATFL reconstruction with a gracilis autograft, and 3 ankles underwent repair of ATFL. They were allowed to return sports activities at 8 weeks post-surgery, and the average time to return to sports activities was 12.3 weeks. Two ankles had osteochondral lesions (OCL). One patient was treated with arthroscopic antegrade drilling for OCL of the talus, and another patient was treated with arthroscopic retrograde cancellous bone plug transplantation from the iliac crest for OCL with large subchondral lesion of the talus. They were allowed to return sports activities at 6 months post-surgery, and both of them were able to return to football at 6 months post-surgery. Among the remaining 4 patients who had no additional treatment for ATFL or OCL, the average time to return to sports activities was 3.0 weeks. No vascular or infectious complications occurred. One patient showed temporary numbness of the toes probably due to intraoperative traction and the numbness resolved completely within 8 weeks. No patients require reoperation, but asymptomatic osteophyte recurrences were observed radiographically in 2 ankles.

S337 Mean AOFAS score improved from 64.1 to 84.3 (p \ 0.001). Mean dorsiflexion angle improved from 12.3 to 14.2 (p \ 0.01) and mean plantar flexion angle from 38.9 to 45.9 (p \ 0.001). Conclusions: Scranton et al. concluded grade IV patients were not suitable candidates for an arthroscopic debridement. However, they performed arthroscopic or open debridement for only anterior ankle impingement, but no treatment for posterior ankle impingement. We believe this is the first report of simultaneously performed ankle arthroscopy and hindfoot endoscopy for the treatment of severe ankle impingement with satisfactory early outcomes.

P33-221 Bipolar fresh total osteochondral allograft: why, where, when S. Giannini1, R. Buda1, M. Cavallo1, A. Ruffilli1, S. Neri2, F. Vannini1 1 Istituto Ortopedico Rizzoli, Bologna University, VI Department of Orthopaedics and Traumatology, Bologna, Italy, 2Istituto Ortopedico Rizzoli, Bologna University, Bologna, Italy Objectives: Bipolar fresh total osteochondral allograft (BFTOA) is a fascinating option for the treatment of end stage arthritis. Although there is general agreement on the validity of partial allograft, BFTOA have been proposed to date only in the ankle joint with controversial results. Aim of this study is to describe the application of BFTOA to shoulder, knee, ankle and Ist metatarsophalangeal joint and provide the results at 4 years. Methods: 71 patients (37.3 ± 10.2 years), affected by end stage arthritis of either shoulder, knee, ankle or Ist metatarsophalangeal joint, underwent BFTOA. Patients evaluation included clinical (100 points scores: Constant, IKDC, AOFAS), X-rays and MRI evaluation at established follow-up. Bioptic samples harvested during II look were evaluated by histological, immunohistochemical analyses and DNA genetic typing by microsatellite. Results: Clinical score increased from 30.3 ± 10.6 pre-operatively to 72.3 ± 13.9 at 47.3 ± 17.5 months follow-up (p \ 0.0005). 15 BFTOA failed (6 knees, 8 ankles and 2 Ist metatarsophalangeal joint). X-ray evaluation showed arthritis recurrence in all the cases, but ns relationship was found between this finding and the clinical outcome. Histological and immunohistochemical analyses showed viable transplanted cartilage with a mixed DNA profile, indicating the presence of both donor and host cells. Conclusions: BFTOA was capable to give satisfactory outcomes in selected cases particularly in the ankle, shoulder and Ist metatarsophalangeal joint. BFOA otherwise failed in all the knees, with exception of one case of desarthrodesis which is one of the most satisfactory of the series. The prevailing presence of host DNA suggests the ingrowth of host cells into transplanted cartilage.

P33-352 Computertomographic evaluation of frontal talar edge configuration for osteochondral plug transplantation M. Wiewiorski1, S. Ho¨chel1, A.M. Nowakowski1, V. Valderrabano1 1 University Hospital of Basel, Orthopaedic Department, Basel, Switzerland Objectives: One of the current research topics is the aim to produce tissue engineered osteochondral grafts for future treatment of osteochondral lesions (OCL) of the talus. For the exact anatomic reconstruction, the dimensions of the medial and lateral talar dome must be considered. Sparse data is available regarding the normal anatomic talar dimensions on standard radiographs of ankle joints [1, 2]. The purpose of this study was to describe normal anatomy of different sections of the talar dome on computertomographic (CT) images.

123

S338 Methods: CT data sets (Somatom 10, Siemens Erlangen, Germany) of 82 patients (86 ankles) (28 female, 54 male; average age 41.9 years (range 15–76 years)) without talar pathologies were included. Measurements were performed with a geometry analysis software (VGStudio MAX 2.0, Volume Graphics, Heidelberg, Germany). To assure measurement reproducibility, the reference planes were defined in a first step. To measure the radius, circles were fitted into the medial and lateral talar edge on frontal planes. To allow measurement of different segments of the talus, the frontal plane was tilted through the center of the talus (defined as a circle fitted to the talus on sagittal view) at 15 and 30 anteriorly and posteriorly. Results: The talar edge radius in the frontal plane at 0 wa s 4.9 mm medially (3.0 mm laterally), at 15 ant. 4.2 mm (3.1 mm), at 30 ant. 4.6 mm (3.1 mm), at 15 post. 4.5 mm (3.9 mm), and at 30 post 4.1 mm (6 mm). There was a significant difference (p \ 00.1) between the mean medial and lateral talar edge radius at all angles. Conclusions: This study shows a significant difference between physiological medial and talar edge configuration at different parts of the talar dome. The assessed data provides important aid for engineering of pre-formed, pre-sized osteochondral grafts. Such preshaped grafts could help restoring the physiological joint surface by matching exactly into the lesion and consequently achieving the recovery of the physiological joint biomechanics and prevention of secondary degenerative disease. Refereces: 1. Leumann, A., et al., Radiographic evaluation of frontal talar edge configuration for osteochondral plug transplantation. Clin Anat, 2009. 22(2): p. 261–6. 2. Riede, U.N., P. Heitz, and T. Ruedi, [Studies of the joint mechanics elucidating the pathogenesis of posttraumatic arthrosis of the ankle joint in man. II. Influence of the talar shape on the biomechanics of the ankle joint]. Langenbecks Arch Chir, 1971. 330(2): p. 174–84.

P33-355 Comparison of active and passive measurements of ankle proprioception in healthy individual M. Unal1, M. Baydar2, S. Gulbahar2, E. Akalin2, D. Akseki3, H. Pinar4 1 Isparta Sifa Hospital, Orthopaedics and Traumatology, Isparta, Turkey, 2Dokuz Eylu¨l University, Physical Therapy and Rehabilitation, Izmir, Turkey, 3Balikesir University, Orthopaedics and Traumatology, Balikesir, Turkey, 4Dokuz Eylu¨l University, Orthopaedics and Traumatology, Izmir, Turkey Objectives: Importance of proprioception in the treatment and prevention of sports injuries has become increasingly clear. There are a lot of measurement methods of proprioception, but none of them are gold standard. Active or passive testing of proprioception are believed to give different results in previous reports. The purpose of this study was to determine the differences between active and passive measurements of ankle proprioception in healthy volunteers. Methods: Seventeen women, 17 men, total 34 healthy volunteers with no history of ankle pathologies, trauma or surgery with ages between 20 and 30 (av. 24.8) were included in the study. Proprioception was measured on dominant ankles with the technique of both active and passive joint position sense by a isokinetic dynamometer. Passive measurements were performed first. Target angles were 10 and 20 of inversion. Statistical analyses were done by Wilcoxon Ranked Sign test. Results: No significant differences were found between reproduction errors of target angles between passive and active measurements at

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 10 inversion (p [ 0.05). Also there were no significant differences found at 20 inversion measurements (p [ 0.05). Conclusions: Our results showed that active or passive testing of proprioception gives similar results in ankles of healty volunteers. Further studies are needed to clarify that both techniques give similar results in different diseases or traumatic conditions.

P33-400 Comparison of clinical and radiological results in minimally invasive plate osteosynthesis for distal tibial fractures: conventional versus anatomically percutaneous wiring reduction technique Y.-M. Kim1, Y.-B. Joo1, T.-H. Kang1, Y.-S. Jeon1, J.-M. Hwang1 1 School of Medicine, Chungnam National University, Department of Orthopaedic Surgery, Daejeon, Republic of Korea Objectives: To report the results of comparing the clinical and radiological results of conventional minimally invasive internal fixation and the new anatomical percutaneous wire reduction method for distal tibia fracture. Methods: Patients were followed up for at least 1 year after performance of MIPO using a periarticular locking plate for distal tibia fracture: 27 patients were in group 1, who underwent conventional treatment from May 2004 to December 2005, and 17 patients were in group 2, who underwent anatomical percutaneous wire reduction from August 2007 to February 2010. Mean patient age was 55.9 years in group 1 and 50.0 years in group 2. Mean follow-up was 21.0 months in group 1 and 20.0 months in group 2. We investigated the period until bone union was achieved, the degree of angulation angle, and complications. For postoperative evaluation, the OlerudMolander score in daily living was calculated. Results: Anatomically percutaneous wire reduction was significantly superior with regard to operation time, AP angle, and varus/valgus angle. The Olerud-Molander score was not statistically significant. Conclusions: Anatomical percutaneous wire reduction can achieve anatomical reduction more easily, and the plate could be simply fixed without additional manual reduction and maintenance, it shows less angulation deformity and shorter operation time than MIPO.

P33-414 Histologic evaluation of anterior talofibular ligament diagnosed as chronic lateral injury of ankle joint and comparison to arthroscopic finding W. Miyamoto1, M. Takao2 1 Teikyo University of Science, Tokyo Physical Therapy, Tokyo, Japan, 2Teikyo University School of Medicine, Orthopaedic Surgery, Tokyo, Japan Objectives: Recently efficacy of early accelerated rehabilitation after injuries of soft tissue including ligament has been reported. Although the most popular surgery for chronic lateral ligament injury of ankle joint seems to be a repair of anterior talofibular ligament (ATFL) using its remnant, there is unclear whether the remnant tissue has still had sufficient strength for early accelerated rehabilitation. The purpose of this study was to investigate histologic change of ATFL remnant after chronic lateral injury of ankle joint and correlation between histologic change and macroscopic finding of remnant under arthroscopy. Methods: Between 2007 and 2009, we performed ATFL reconstruction using gracilis tendon for 34 patients who were diagnosed as chronic lateral ligament injury of ankle joint. At surgery, all patients

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 underwent arthroscopy to evaluate continuity, irregularity of fiber and whether there was synovium coverage of ATFL remnant. Resected remnant at reconstruction was obtained in all patients and content of type I and III collagen was assessed by immunostaining. The contents of type I and III collagen was compared with those in normal ligament. Furthermore, correlation between arthroscopic finding and results of the histologic findings were assessed. Results: Arthroscopic findings revealed 19 patients with continuity, 13 patients with no continuity of ATFL, and 13 patients with mild, 11 patients with moderate, 8 patients with sever irregularity of fiber respectively. Coverage of synovium was confirmed in 30 patients (15 patients: red colored, 15 patients: white colored), but 2 patients did not revealed synovium coverage on their ATFL. In all patients, staining of ATFL remnant revealed increased type I collagen and decreased type III collagen compared with staining of normal ligament. Furthermore, irregularity of ATFL fiber in arthroscopic assessment was correlated with increased staining of type I collagen in histologic assessment. Conclusions: Our study revealed the possibility that ATFL remnant did not have enough strength for early accelerated rehabilitation after repair of ATFL for chronic lateral ankle injury because of their histologic change. In arthroscopic assessment, irregularity of ATFL fiber reflected histologic change its remnant.

P33-481 CT assessment of the fusion rate after posterior subtalar arthrodesis M. Thaunat1, X. Bajard2, P. Boisrenoult3, P. Beaufils4, P. Oger2 1 Hopital Pitie´ Salpe´trie`re, Paris, France, 2Paris Ouest, Le Chesnay, France, 3Hopital Andre Mignot, Chirurgie Orthope´dique, Versailles, France, 4Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France Objectives: The purpose of the study was to assess the fusion rate on CT examinations and to correlate clinically the functional result with the degree of bone fusion in the subtalar joint after posterior arthroscopic subtalar arthrodesis (PASTA). Methods: Fourteen cases, from 36 to 84-years-old, were retrospectively followed up for minimum of one-year (range 12–92 months). A CT scan had been systematically performed at the 6 months followup visit. The CT scan were examined in sagittal 2-mm-thick reformatted slices, measuring on each image the length of the joint surface and the length of the fused portion of the joint space. Results: At 6 months, the average fusion ratio was 39 ± 19% (range 0–69%). Fusion defined by a fusion ratio superior or equal to 33% on the CT scan was observed in 11 cases. One patient had a delayed union and required a revision of fixation. One patient had a bilateral nonunion. Mean average AOFAS score improved from 51 ± 10 to 77 ± 9 at last follow up. Conclusions: Compared to open procedures, the posterior arthroscopic fusion seems to offer an interesting alternative. However, our results suggest that the fusion rate following PASTA is not as favorable as reported in previous studies. Factors such as adequate compression and stable fixation provided by the screws as well as the surgeons’ experience with this demanding technique are of the utmost importance. A 33% CT fusion ratio threshold could accurately discriminate between clinical stability and instability.

P33-635 Cartilage reconstruction in the foot and ankle H. Thermann1, F. Suezer1, S. Feil1 1 ATOS Clinic Center, Center for hip-, knee- and foot surgery, Heidelberg, Germany Objectives: This prospective study was designed to describe the details of the fully arthroscopical surgical technique with autologous matrix-induced chondrogenesis (AMIC) for cartilage reconstruction

S339 in the foot and ankle. Short-term results evaluated with Hannover Scoring System (HSS) and VAS for pain, function and satisfaction will be presented. Methods: We performed 64 arthroscopic AMIC procedures in the foot and ankle in 2009–2010. 42 patients were included in this ongoing study in different locations, ankle anterior n = 26, ankle posterior n = 4, Subtalar n = 4, talonavicular n = 2 (arhtoscopic assisted), talo-calcaneal coaliton n = 2, MTP1 n = 2 (open). We invited the patients for a follow-up examination prospectively at 6 weeks, 12 weeks, 6 months and 12 months and evaluated them with the VAS Score for pain, function and satisfaction and the Hannover Scoring System (HSS). The operation was performed in all patients by one surgeon (HT). We debrided the local (osteo)chondral lesion first, afterwards we performed a microfracturing with a special awl. The ,,microfracture holes’’ of the subchondral bone are filled with platlet rich plasma (PRP). The defect is covered with a collagen matrix soaked with PRP and then sealed with fibrin glue. In the aftertreatment, the patients received a plaster for 3 days, physiotherapy program for 8–12 weeks, partial weight bearing for 8 weeks. We recommend jogging and impact sports after 8–12 months. Results: At this stage scientific reliable data could be delivered in a short-term follow-up up to 1 year. In the ankle joint, the HSS improved from 55 (± 12.18) preop to 6 weeks 62.1 (±15.41), 3 months 65.2 (±15.89); 6 months 72.8 (±17.23); 1 year 82.7 (±14.64). VAS pain: 4.4 (±2.74) preop to 6 weeks 6.4 (±2.78); 3 months 7.2 (±2.61); 6 months 7.6 (±2.14); 1 year 9.1 (±0.73). VAS function: 3.9 (±2.42) preop to 6 weeks 4.3 (±2.99); 3 months 6.1 (±2.29); 6 months 6.6 (±3.02); 1 year 7.7 (±2.78). VAS satisfaction: 2.7 (±2.67) preop to 6 weeks 5.3 (±2.69); 3 months 6.2 (±2.89); 6 months 6.4 (±3.43); 1 year 7.7 (±2.91). Subtalar joint and talonavicular joint were painfree after 1 year with minor limitations in eversion and inversion. Separations of the talo-calcaneal coalitions were still open and painfree after 2 years. Conclusions: AMIC procedure is a reliable therapy for osteochondral lesions of the foot and ankle. We observed promising early results in a short term follow-up.

P33-839 The status of ankle joint at the time of implant removal after malleolar fractures. An arthroscopic study V. Chouliaras1, C. Christogiannis1, S. Plakoutsis1, S. Kopsidas1, A. Grestas1 1 General Hospital of Arta, Orthopaedic Department, Arta, Greece Objectives: The purpose of this study was to evaluate the status of ankle joint arthroscopically, in patients with maleollar fractures at the time of implant removal. Methods: Arthroscopic examination was performed at the time of implant removal after an average of 12.8 months (range 8–16 months) in a series of 35 consecutive patients. There were 15 women and 20 men with an average age of 52.8 years (range 17–63 years old). The Danis-Weber classification system was used for all fractures. All the patients were evaluated by the AOFAS Clinical Rating System before and after implant removal. Results: In all the patients the fractures were healed without need of a second operation. Synovitis was present in all the patients and was located antero-laterally in 23 patients. Articular cartilage damage was noted in 16 patients, among whom 9 were treated by shaving and 7 were treated by the microfracture technique. In 9 patients we removed loose bodies. The lesions tended to be worse in patients over 60 years of age. The frequency and severity of the lesions also increased in type-B and type-C fractures. Conclusions: There is clear evidence that despite anatomic and appropriate reduction postoperative results of malleolar fractures are

123

S340 not free of complications. Synovitis is a common finding, and chondral lesions are commonly found associated especially with the more severe ankle fracture pattern. Our findings show that arthroscopy at the time of implant removal is useful in identifying and in treating intra-articular ankle lesions.

P33-865 Impairment of sensory motor control in functionally instable ankle joint F. Pisanu1, V. Verderosa2, A. Di Monda2, A. Santandrea2, V. Marcialis2, F. Deriu3 1 Universita` di Sassari, Clinica Ortopedica, Dipartimento di Scienze Biomediche, Sassari, Italy, 2Universita` di Sassari, Clinica Ortopedica, Sassari, Italy, 3Universita` di Sassari, Dipartimento di Scienze Biomediche, Sassari, Italy Objectives: To present some of the current concepts that form a comprehensive model of sensory-motor control impairment in functionally instable ankle. Methods: A literature search of journal articles in English was conducted using the PubMed database from inception to 2011 and relevant studies dealing with sensory-motor factors associated with human ankle stability were identified. Results: Findings from individual studies on the role of sensory-motor function in ankle stability yield to conflicting conclusions. The current body of literature as well as recent systematic reviews quite unanimously show: (1) an immediate impairment in ankle proprioception after ankle ligament injury, without any direct evidence of a local mechanoreceptor deficit; (2) no consistent experimental evidence of impairment in muscle reaction time and in muscle strength associated with functional instable ankle; (3) evidence of a bilateral response pattern, at both the ankle and proximal joints, after unilateral foot and leg displacement; (4) the occurrence of a bilateral impairment of postural and balance control, following unilateral injury. Conclusions: Since the mid-1960 sensory-motor control has been considered an important factor in ankle stability. In the clinical condition known as functional instable ankle, in the specific topographic origin of proprioception deficits has yet to be fully explained. Furthermore, the general belief that peroneal muscle weakness may contribute to lateral ankle instability is not supported by any evidence of impairment in muscle reaction time and in muscle strength. Based on the extremely short time periods required to react to forces during ankle functional tasks, the role played by sensory-motor integration in preparatory muscle activity seems to be more important in ankle stability. The bilateral sensory motor deficit at the entire kinetic chain, despite unilateral injury, provides further evidence that central processing of peripheral sensory information may have a crucial role in ankle stability. This assumption well support the hypothesis that functional ankle instability could be restored through global coordination training programs mediated by central nervous mechanisms.

P33-905 Metaphyseal locking compression plate as an external fixator for the distal tibia S. Tulner1, P. Kloen2 1 Tergooi Hospitals, Orthopaedic Surgery and Traumatology, Hilversum, The Netherlands, 2Academic Medical Center, Orthopaedic Surgery, Amsterdam, The Netherlands Objectives: Infection of the distal tibia after treatment for open fractures is relatively common and remains a challenging problem. It is common practice that after initial debridement and hardware

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 removal, temporary bony stabilization is provided by external fixation. As most external frames for the lower leg are bulky and cumbersome, we have successfully used a metaphyseal locking compression plate (LCP) as external fixator for the distal tibia. Methods: In 5 patients with infect/defect non-unions of the distal tibia a metaphyseal LCP was used as temporary external fixator after initial thorough debridement and hardware removal. Postoperatively, we let the patient toe-touch or partially weight bear depending on the intrinsic stability of the reconstruction. All patients were treated by systemic antibiotic treatment until infection was eradicated followed by definite treatment. Results: In the five patients, the external fixator was in place for an average of 11 months (range 5–18). In three patients the plate was left in place until full bony healing was obtained. In the other two patients the plate was removed and replaced by formal internal plate fixation. We have not seen any significant pin tract infections or any loosening of the hardware and the skin seemed to tolerate the titanium screws well. In a mean follow-up of 10 months (range 3–21), all patients achieved union and there were no signs of infection. Conclusions: External fixation with a metaphyseal LCP imparts a much lower profile as the standard and circular external fixators, which are bulky and uncomfortable. Advantages of the LCP as external fixator are that it is easily concealed under regular clothing. There is much less tendency for the frame to strike the contralateral lower leg in the swing-thru phase of either leg. The multiple 3.5 mm locking holes distally provide many options for very distal fixation rather than 1 or 2 large external fixator pins. Despite its low profile the strength of fixation seems strong enough to withstand the forces acting on the distal tibia. Obviously, the group of patients is relatively small and the indications are limited. However, we feel the use of this plate has been a useful addition to the reconstructive efforts in these challenging problems.

P33-909 The ability to perform a single heel-rise is significantly related to patient reported outcome 12 weeks after an Achilles tendon rupture N. Olsson1, J. Karlsson1, B. Eriksson1, M. Lundberg2, K.G. Silbernagel3 1 Institute of Clinical Sciences, Department of Orthopaedics, Sahlgrenska Academy at Go¨teborg University, Go¨teborg, Sweden, 2 Institute of Clinical Sciences, University of Gothenburg, Division of Occupational Orthopaedics, Go¨teborg, Sweden, 3University of the Sciences, Philadelphia, Department of Physical Therapy, Samson College of Health Professions, Philadelphia, United States Objectives: The purpose of this study was to evaluate the recovery of calf muscle strength 12 weeks after an Achilles tendon rupture and study how this relates to the patient-reported outcomes with regard to the lower limb function as well as general health and quality of life. The secondary purpose was to evaluate if the degree of patients’ fear of movement was related to function and patient reported outcome. Methods: This was a prospective study of patients treated surgically or non-surgically with early active rehabilitation after Achilles tendon rupture. Eighty-one patients (69 males and 12 females) with a mean (SD) age of 40 (10) years were included. Patients’ ability to perform a one-legged heel-rise, physical activity level (Physical Activity ScalePAS), patient reported symptoms (using Achilles tendon Total Rupture Score—ATRS and Foot and Ankle Outcome Score—FAOS), general health (EQ-5D) and fear of movement (Tampa Scale for Kinesiophobia-TSK) were evaluated at a mean (SD) 12 (0.9) weeks after injury. At the time of inclusion (baseline) PAS and EQ-5D were measured. The ability to perform a single heel-rise was evaluated by having the participants stand on a box with ankle in neutral position. The patients were classified as being able to perform a one-legged

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 heel-rise if they were able to lift the heel at least 2 cm once while keeping the knee straight. Results: The heel-rise test showed that 40 of 81 (49%) patients were unable to perform a single heel-rise 12 weeks after injury. There was no statistical difference (p = 0.269) between surgical and non-surgical group according to heel-rise ability. We found that patients who were able to perform a single legged heel-rise were significantly younger (p = 0.005), more often of male gender (p = 0.013) and had a higher score in ATRS (p = 0.002) and FAOS (p = 0.001–0.017) and had a higher degree of physical activity (p = 0.022) at 12 weeks. There were significant (p = 0.001–0.032) negative correlations (r = -0.239-0.376) between the TSK score and all the patientreported outcomes and the physical activity level. Conclusions: The heel-rise ability appears to be an important early achievement which influences patient reported outcome and physical activity. Future treatment protocols focusing on regaining strength early therefore seem of great importance. The heel-rise test can be recommended as part of the evaluation protocol. Fear of physical activity and movement need to be addressed early during the rehabilitation process.

P33-1005 Traumatic osteochondral lesions of the talus treated arthroscopically with microfactures and arthroscpic four step treatment for ankle instability A. Ventura1, C. Terzaghi1, C. Legnani2, J. Acquati Lozej2, E. Borgo1 1 Gaetano Pini Orthopaedic Institute, CAM, Minimally Invasive Surgery Unit, Milano, Italy, 2Scuola di specializzazione, Universita` degli Studi di Milan, Milano, Italy Objectives: Osteochondral lesions (OCLs) of the talus occur usually following one or multiple ankle sprains [1, 2] and they are frequently associated with chronic lateral ankle instability [3]. Operative treatment depends on the size and site of the lesion; debridement and bone marrow stimulation of the subchondral bone with arthroscopic microdrilling/micropicking currently represents the primary surgical treatment of most OCLs of the ankle [4]. Open procedures, such as osteochondral autograft transfer and autologous chondrocyte transplantation, represent also a popular treatment option in selected cases [5, 6]. The purpose of this prospective study is to assess the outcomes of arthroscopic treatment of ankle-sprain related OCLs by associating concomitant microfracturing and ligamental retension in the same sitting. Methods: 40 patients with at least one MRI-documented OCL of the ankle were treated at our Department from 2001 to 2010. Mean age at surgery was 39 years (range: 15–62). All patients underwent combined microfractures and arthoscopic four step treatment for ankle instabilty: synoviectomy, debridement of ATLF lesion border, capsular shrinkage. Rehabilitation protocol included immobilization and non weightbearing for 21 days followed by gradual weightbearing, complete active and passive ankle range of motion exercises and regaining of proprioception. Results: Patients were examined preoperatively and follow-up prospectively after an average time of 50 months (range 20–120 months). Clinical assessment included the AOFAS ankle and hindfoot scoring system, Karlsson and Peterson scale, Tegner activity level, Sefton articular stability scale and objective examination comprehending range of motion (R.O.M.), anterior drawer sign and talar tilt test. Paired t test (two sided test and a = 0.05) was utilized to compare the preoperative and follow-up status. Differences with a p value \ 0.05 were considered statistically significant.

S341 Table 1 Pre-operative

Post-operative

t test

AOFAS scale

63.51 (sd: 8.18) 86.63 (sd: 6.93) (p \ 0.001)

Karlsson score

61.81 (sd: 11.07)

89.12 (sd: 8.81) (p \ 0.001)

Tegner score

3.59 (sd: 1.23; median: 3) 4.00 (sd: 0.00; median: 4)

4 (sd: 0.94; median: 4) 1.6 (sd: 0.83; median: 2)

(p \ 0.001)

5 (5.7%)

(p \ 0.001)

Sefton scale

Anterior drawer sign 88 (100%) (positivity)

(p \ 0.001)

Objective examination documented significantly reduction of anterior drawer sign and talar tilt test and complete ROM (p \ 0.001) Mean point scales rating improved significantly from baseline (p \ 0.001) (Table 1). Conclusions: In the young and active population, OCLs following sports-related injuries are usually associated with the involvement of the external ligamental complex, and thus both aspects should be addressed during the treatment in order to maximize the outcomes. On the bases of our findings, we propose combined microfractures and ligamental shrinkage as an effective method to treat chronic ankle instability in patients with sports-related OCLs of the ankle.

P33-1154 All arthroscopic Autologous Matrix Induced Chondrogenesis cartilage reconstruction open technique in the ankle - early results K. Ciemniewska-Gorzela1, T. Piontek1, A. Szulc2, J. Naczk3 1 Rehasport Clinic, University of Medical Sciences, Clinic of Pediatric Orthopedic Surgery, Poznan, Poland, 2University of Medical Sciences, Clinic of Pediatric Orthopedic Surgery, Poznan, Poland, 3 Rehasport Clinic, Poznan, Poland Objectives: One of the methods of cartilage ankle reconstruction is Autologous Matrix-Induced Chondrogenesis (AMIC). AMIC is a variant of the microfracture method. It makes use of a collagen membrane that serves as a scaffold for new cells and allows effective reconstruction of even large fragments of damaged cartilage surface. Currently, such procedures are performed by means of surgical opening of the ankle joint. The goal of the study is to present an arthroscopic technique for reconstructing damaged fragments of talus and tibiae cartilage using the AMIC technique with collagen matrix and present 1 year results after operation. We do not use any special tools. Methods: We operated 7 patients all arthroscopic AMIC technique. We control patients before and after 6, 12 month after operation using MRI scans. We used AOFAS, Povacz and VAS skale before and after operation. Results: We observe in MR scans good covering of the cartilage by the new tissue and in modificate MOCART scale 75–100% good response for the treatment after 12 month. In AOFAS scale improvement from 22 points before to 41 after 1 year, Povacz from 15 points to 23 after, VAS from 4 points before to 8 after 1 year. Conclusions: We have presented a simple, entirely arthroscopic technique for reconstructing extensive cartilage defects without bone defects. Results of treatment are good promising for future investigations.

123

S342

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

P33-1273 Lower limb muscle activity during high-heeled versus low-heeled gait A. Arnadottir1, I.H. Kjartansdottir1, S.K. Magnusdottir1, K. Briem2 1 University of Iceland, Physical Therapy, Reykjavik, Iceland, 2 University of Iceland, Research Centre of Movement Science, Reykjavik, Iceland Objectives: High-heeled shoes are prominent in today’s society and are worn on special occasions as well as casually. Certain known kinematic changes occur when walking in shoes with high heels versus low-heels as well as a reduction in the base of support. Concurrent changes in muscle activation patterns have, however, not been investigated extensively. The purpose of the study was to identify changes in muscle activity when walking in high-heeled shoes of varied height versus flats. Methods: 24 women participated in this cross-sectional, controlled laboratory study. Each participant was identified as a regular- versus non-user of high-heeled shoes and walked in trainers and in dress shoes with heels of 3, 6 and 8 cm on a treadmill with a pressure measuring platform at a constant speed. Muscle activity was sampled during 10 gait cycles for each shoe condition at 1,600 Hz with wireless surface electrodes. Data from 5 cycles were integrated to represent the average energy of each muscle’s signal during gait. The target muscles were: fibularis longus (FL), tibialis anterior (TA), soleus (S), gastrocnemius medialis (GM) and lateralis (GL), biceps femoris (BF), vastus lateralis (VL) and rectus femoris (RF). Temporospatial parameters were also observed. ANOVA were used to analyze data across conditions (within-subjects factor) between regular- and non-users (between-subjects factor). Results: A main effect of shoe condition was found for muscle activity of the FL, Sol, GM, GL, and VL (p \ 0.01). Both GM and GL demonstrated an immediate increase in activation levels from flats to 3 cm heels, while Sol, FL and VL did not increase activation levels until at the 6 cm heel (Table 1). A main effect of shoe condition was found in push-off force (p \ 0.001), due to a significant reduction during gait in the 6 and 8 cm heels. Although step width and foot progression angle generally decreased as heel height increased (p \ 0.001), regular users of highheeled shoes demonstrated an overall narrower step-width and a greater progression angle (p \ 0.05) than non-users. Conclusions: With increased heel height the ankle is put in a more plantarflexed position, which offers less passive stability at the ankle and active insufficiency of the plantarflexors. At a relatively low heelheight, greater activation levels of plantarflexors effectively maintained push-off force at previous levels and no increase in FL was seen. As heel height increased, greater activation of the plantarflexors was ineffective in maintaining push-off force and a greater demand for dynamic stability was met by increased FL activation. Differences in strategy may exist between those who regularly use high-heeled shoes and those who do not.

Table 1 Mean (SD) muscle activation (mV) Muscle

Flats

3 cm heels 6 cm heels 8 cm heels

Fibularis longus

9.6 (9.4) 13.2 (13.3) 23.3 (24.9) 30.5 (31.4)

Soleus

7.1 (2.8)

Gastrocnemius medialis

7.8 (3.2)

10.6 (5.4)

12.4 (5.9)

13.1 (5.1) 16.7 (9.4)

16.6 (8.7)

20.0 (12.2)

Gastrocnemius lateralis

5.4 (4.2)

6.8 (5.2)

Vastus lateralis

1.5 (1.5)

2.4 (3.3)

123

11.0 (11.0) 12.5 (7.8) 2.9 (1.9)

3.9 (2.7)

P33-1391 Microfracture in the treatment of osteochondral defects of the talus: clinical results at an average follow-up of 9 years with T2-mapping at 3 Tesla C. Becher1, D. Zuehlke1, C. Plaass1, M. Ewig2, C. StukenborgColsman1, H. Thermann3 1 Hannover Medical School, Orthopaedic Department, Hannover, Germany, 2Radiologie am Raschplatz, Hannover, Germany, 3ATOS Clinic Center, Department for Hip, Knee & Foot Surgery/Sports Trauma, Heidelberg, Germany Objectives: Microfracture is an established method in the treatment of osteochondral defects of the talus. The application of T2-mapping in the ankle at 3 Tesla was recently introduced as a non-invasive modality to assess the repair tissue. The aim of this study was to evaluate patients with a mid to long term follow-up after microfracture treatment for an osteochondral defect of the talus with common clinical scores and T2-Mapping at 3 Tesla. Methods: Patients from our database with a minimum follow-up of 7 years after microfracture for an osteochondral defect of the talus were invited for clinical and MRI evaluation. The Hannover Scoring System for the ankle and the American Orthopaedic Foot and Ankle Society (AOFAS) score were used to evaluate the clinical outcome. Cartilage repair tissue assessment by MRI was done on a 3 Tesla MR unit (Discovery MR750, GE Healthcare, Munich, Germany) using isotropic 3-dimensional (3-D) gradient echo (true FISP), proton density fat-suppressed turbo spin echo (PD-FSTSE) and T1-weighted spin echo sequences. Images were evaluated by three independent examiners. The defect size and mean T2-values were measured at the repair tissue (RT) and at the native adjacent reference cartilage (RC) at two different coronal and sagittal images. Overall mean values were taken for statistical analysis using a 2-sided independent t test. Correlations between T2-values and age, follow-up time and the clinical score results were evaluated with the Pearson correlation coefficient. Results: To date, 12 patients (7 female, 5 male) with an average age of 39 ± 17 years at the time of surgery participated in the study. Average follow-up was 9 years (7–13 years). Clinical evaluation revealed 91 ± 11 points at the AOFAS score and 89 ± 6 points at the Hannover Scoring System. Average defect size was 8 9 11 mm. No significant differences were found between the repair tissue and the native adjacent reference cartilage. RT T2 was 40.9 ± 8.2 ms, and RC T2 was 38.5 ± 5.9 ms (P = 0.753). No correlations were found between the T2-values and age, follow-up time as well as the clinical score results. Conclusions: In a small patient group, the clinical results for microfracture in the treatment for articular cartilage defects of the talus were excellent at a mean follow up at 9 years. The repair tissue of these patients appears to have similar T2 properties compared to those of the native adjacent cartilage.

P33-1409 Bone volume of an osteochondral talar defect after debridement and microfracturing: a new method to quantify bone volume changes M. Reilingh1, R. Gerards1, G. Streekstra2, R. Jonges2, L. Blankevoort1, N. van Dijk1 1 Academic Medical Center, University of Amsterdam, Orthopaedic Surgery, Amsterdam, The Netherlands, 2Academic Medical Center, University of Amsterdam, Biomedical Engineering and Physics, Amsterdam, The Netherlands Objectives: Debridement and microfracturing for isolated osteochondral talar defects is the primary treatment for lesions of up to 15 mm in diameter. The aim of this procedure is to induce subchondral bone revascularization and subsequently to accomplish new

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 bone and stable fibrocartilage formation. However, to our best knowledge there is no reliable method to measure bone volume changes. The purpose of this study was to evaluate the reliability of bone volume measurements of an osteochondral talar defect after debridement and microfracturing on CT imaging using a 3D segmentation model. Methods: Computed tomography (CT) scans of 38 ankles were obtained preoperatively and two weeks postoperatively. Postoperative bone volume of the osteochondral defect was determined on CT using 3D segmentation software. A sphere was drawn outlining the defect to select the region of interest. Within this sphere the amount of the voxels was calculated on a fixed global threshold to determine the bone volume postoperatively. The remainder of the talus was segmented and used as a template to match the talus on the preoperative CT. After matching the diameter and position of the sphere should be the same as drawn on the postoperative talus. Next, the amount of voxels within the sphere was calculated to determine the bone volume preoperatively. The removed bone volume after debridement and microfracturing can then be calculated. Measurements were performed by two different examiners and twice by one examiner to calculate the intra- and interobserver reliability. Results: The inter observer intraclass correlation coefficient (ICC) was 0.997. The intra observer intraclass correlation coefficient was 0.998. The average bone volume removed is 145.7 mm3.

Pre- postoperative segmented talar bone Conclusions: The use of this 3D segmentation model to quantify bone volume changes after debridement and microfracturing is reliable.

P33-1458 Loose bodies are consistently present after scopic debridement and microfracturing of osteochondral talar defects M. Reilingh1, C. van Bergen1, R. Gerards1, N. van Dijk1 1 Academic Medical Center, University of Amsterdam, Orthopaedic Surgery, Amsterdam, The Netherlands Objectives: Debridement and microfracturing for isolated osteochondral talar defects is the primary treatment for lesions of up to 15 mm in diameter. However, loose bony particles have been reported as a potential pitfall in the arthroscopic microfracturing technique. The purpose of this study was to evaluate the presence of loose bony particles after arthroscopic debridement and microfracturing of osteochondral talar defects. Methods: Computed tomography (CT) scans of 23 ankles were obtained preoperatively and two weeks and 1 year postoperatively. 3D reconstructions were made of all CT scans using a fixed global threshold to extract the mineralized bone phase. Loose bony particles were counted on the two-week’ and one-year’ CT reconstructions and compared too preoperatively. Furthermore, the AOFAS was scored preoperatively, and 1 year postoperatively. The correlation was analysed with the Spearman’s rho test.

S343 Results: One or more (median 4.5, range 1–25) additional loose bony particles were found in 20 out of 23 cases (87%) 2 weeks postoperatively (Figs. 1, 2). In 16 out of these 20 cases (80%) all loose bony particles were resolved after 1 year. In the other four cases a decrease in amount of loose bony particles was found. The median AOFAS improved from 72 (range, 43–90) preoperatively to 90 (range, 67–100) at 1 year (p \ 0.01). No correlation (r = 0.3, p = 0.16) was found between the presence of additional loose bony particles and the clinical outcome. Conclusions: Loose bony particles are seen in 87% of patients after debridement and microfracturing of osteochondral talar defects. However, in all patients the loose bony particles resolve or decrease in amount and no correlation was found between the clinical outcome and the presence of additional loose bony particles postoperatively.

Foot

P34-354 Osteochondral plug transplantation for treatment of stage III or IV Freiberg disease S. Miki1, M. Takao1, W. Miyamoto1, Y. Yasui1, K. Innami1, T. Matsushita1 1 Teikyo University School of Medicine, Orthopaedic Surgery, Tokyo, Japan Objectives: There is no gold standard of the treatment for stage III or IV Freiberg disease according to Smillie’s classification. The purpose of this study was to investigate the clinical, radiological and arthroscopic results of osteochondral plug transplantation for stage III or IV Freiberg disease. Methods: We examined 10 consecutive cases (all female with an average age of 13, right: 7 cases, left: 3 cases) who were diagnosed as stage III or IV Freiberg disease according to Smillie’s classification. Osteochondral plug transplantation was performed, harvesting from a non weight bearing site of the upper lateral femoral condyle of the ipsilateral knee. Clinical evaluation using the American Orthopaedic Foot and Ankle Society (AOFAS) score was performed before surgery and at the final follow up. Magnetic resonance imaging (MRI) was performed before and at 6 and 12 months after surgery. In 6 patients, arthroscopic evaluation was performed at 12 months after surgery. The mean follow up was 37 months after surgery (range 12–72). Results: At the final follow up, the average AOFAS score improved 69.6 ± 3.0 preoperatively to 96.8 ± 3.2 (p \ 0.0001). In MRI evaluation, healing of the osteochondral plug was confirmed at 12 months after surgery in all patients. Furthermore, arthroscopic evaluation revealed the score more than nearly normal according to International Cartilage Repair Society Cartilage Repair Assessment Score in all 6 patients. Conclusions: Osteochondral plug transplantation for stage III or IV Freiberg disease yielded satisfactory results upon clinical, radiological and arthroscopic evaluations.

P34-502 New endoscopic approach for the treatment of plantar fasciitis F. Komatsu1, M. Takao1, K. Innami1, W. Miyamoto1, K. Nakajima1, T. Matsushita1 1 Teikyo University School of Medicine, Orthopaedic Surgery, Tokyo, Japan

123

S344 Objectives: The purpose of this study was to determine the clinical results of new endoscopic approach, established deep-fascial medial and lateral portals, in performing endoscopic surgery for plantar fasciitis. Methods: Between 2005 and 2010, 17 feet in 15 patients who were treated conservatively for more than 6 months with failure to relieve their symptoms, endoscopic surgery was performed. The patient is placed in the supine position to elevate the operative foot approximately 15 cm with a leg holder. To prevent a risk of injuring the first blanch of the lateral plantar nerve, a medial portal was made 5 mm deep to the plantar fascia and 10 mm anterior to its origin on the calcaneus under fluoroscopy. The lateral portal was established by placing a blunt trocar deep and perpendicular to the plantar fascia. A 2.7-mm-diameter arthroscope is passed through the deep-lateral portal, and the surgical devices are inserted through the deep-medial portal. A motorized shaver is used for making a working space to excise the fat tissue and a plantar portion of the flexor digitorum brevis muscle, as minimally as possible, to obtain good visualization. First, the plantar surface of the calcaneus and the calcaneal attachment of the plantar fascia should be identified for a landmark. If there was a heel spur, it was resected to establish a clear view of the plantar fascia using an arthroscopic burr. After exposure of the plantar fascia, its width is measured with a probe, and an area of less than the medial half of the plantar fascia is resected with an Arthro-Knife (ConMed Linvatec, Largo, FL). The plantar fascia should be removed until the plantar fat tissue is exposed, which is the sign that the plantar fascia has been resected completely toward its superficial layer. Active range-of-motion exercise of the foot and ankle is performed 1 day after surgery. Partial weight bearing is allowed 3 days after surgery and gradually increases to full weight bearing in accordance with patient tolerance. Results: The mean score on the American Orthopedics Foot and Ankle Society Ankle Hindfoot Scale was 64.2 ± 6.3 points before surgery and 92.6 ± 7.1 points at final follow up (P \ .0001). Patients’ subjective assessment of the operation was 76% of patients were very satisfied, 18% were satisfied and 6% were moderately satisfied. The mean duration to full weight bearing after surgery was 13.9 ± 8.4 days. All patients returned to full athletic activities by a mean of 10.7 ± 2.6 weeks. Conclusions: Endoscopic surgery for plantar fasciitis through a deepfascial approach allows a wide field of vision and working space, permitting reliable resection of the plantar fascia and heel spur.

P34-584 Comparison of screw-fixation stabilities of first metatarsal shaft osteotomies: a biomechanical study M. Unal1, O. Baran2, B. Uzun2, A. Turan3 1 Isparta Sifa Hospital, Orthopaedics and Traumatology, Isparta, Turkey, 2Dokuz Eylu¨l University, Izmir, Turkey, 3Siverek Government Hospital, Sanliurfa, Turkey Objectives: Although metatarsal shaft osteotomies have become popular in the surgical treat- ment of moderate or advanced hallux valgus owing to better reduction and stability, they present fixation problems as the angular correction increases. The purpose of this biomechanical study was to evaluate the effects of widely used metatarsal shaft osteotomies and a newly defined osteotomy modification on the stability of screw fixation at greater angular corrections. Methods: Upon evaluation of known problems of shaft osteotomies, a new osteotomy type was designed that might provide an adequate contact area while allowing a greater angular correction, increased stability, and safer osteosynthesis. In our new modification of the Mau osteotomy, the proximal plantar notch that was defined for the Sammarco’s modification to increase the contact area was created more proximally making an angle of about 50 with the osteotomy,

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 and the osteotomy was extended until 5 mm to the distal joint surface, aiming to increase the contact area and intrinsic stability. For biomechanical analysis, 30 standard metatarsal bone models (Sawbones) were divided into five groups equal in number for the following osteotomy methods: Myerson’s modification of the Ludloff osteotomy, Mau osteotomy, scarf osteotomy, offset V osteotomy, and the new modification of the Mau osteotomy. Osteotomies were performed with a standard correction of 10 in the intermetatarsal angle, followed by appropriate osteosynthesis with fixation by two Acutrak compression screws. The relationship between osteotomies and osteosynthesis in terms of stability was assessed by the three-point bending test. Results: The mean stiffness of the Ludloff osteotomy was significantly lower than all the other osteotomy groups (p \ 0.05). Stiffness of the Mau group was significantly greater than three groups (p \ 0.05), but the difference from the offset V group did not reach significance. Stiffness of the new Mau modification was significantly greater than the scarf group (p = 0.016), but did not differ significantly from the offset V group. Osteotomy groups with and without notching had similar stiffness values (p = 0.582), whereas single notching was associated with a significantly greater stiffness compared to double notching (p = 0.031). Conclusions: Our findings suggest that the new modification to the proximal shaft osteotomies moves the center of rotation of angulation more proximally and provides sufficient stability of screw fixation.

P34-746 Comparing unstable shoes versus low dye taping and insoles as treatment for plantar fasciitis B. Ru´narsson1, K. Briem1, R. Magnu´sson2, A´. A´rnason1 1 University of Iceland, Research Centre of Movement Science, Department of Physiotherapy, Reykjavik, Iceland, 2Atlas Physiotherapy, Reykjavik, Iceland Objectives: Plantar fasciitis is one of the most common causes of heel pain, usually located at the inferior medial aspect of the calcaneus. The purpose of this study was to compare the efficacy of two treatment forms for plantar fasciitis. The aim was to see if either Masai Barefoot Technology (MBT) shoes or low dye taping and insoles, or both treatment forms were beneficial and to see if one was superior to the other. Methods: A total of 28 eligible subjects from Reykjavik, Selfoss and surrounding region agreed to participate in this prospective study. Participants were randomly assigned to one of the two treatment groups, 14 to the MBT group (MBTG) and 14 to the low dye tape and insole group (TIG). Participants in the MBTG received MBT shoes for a period of 12 weeks. Participants in the TIG were taped by the same physiotherapist during the first 4 weeks, after that they received insoles for the next 8 weeks. All participants filled out a diary about their use of MBT shoes and insoles and they were instructed to wear them for an average of 2 h per day the first week and 4 h per day after that. Four individuals dropped out of the study leaving 24, 19 women and 5 men, whose data were used for statistical analysis. Interventions lasted for 12 weeks and data were collected in the beginning, after 4 weeks and at the end of the 12 weeks. Outcome measures included morning pain (VAS 0–100 mm), function using Foot and Ankle Ability Measure (FAAM) questionnaire, and pressure pain threshold using algometry. The heel rise test, which involved raising the heels five times and lowering them back slowly to the floor, was also used to evaluate if it could be used to help diagnose plantar fasciitis and to evaluate treatment response. After the test a VAS (0–100 mm) was used to evaluate

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Table 1 Outcome measures Outcome measures

MBTG 0 weeks

Morning

8.1 (0.8)

TIG 4 weeks 4.4 (2.9)

12 weeks 2.3 (2.4)

0 weeks 7.8 (1.5)

4 weeks 6.2 (2.3)

12 weeks 2.9 (3.4)

pain (VAS) Questionnaire

61 (12)

75 (13)

85 (13)

65 (15)

73 (11)

88 (11)

293 (160)

346 (186)

445 (179)

314 (103)

352 (139)

451 (220)

(FAAM) Pressure pain (kPa) Heel rise

3.5 (3.6)

1.6 (1.5)

0.8 (0.6)

2.4 (2.9)

1.2 (1.8)

0.6 (1.1)

test (VAS) Values: mean (SD)

pain. Analyses of variance were used to evaluate outcome measures across time (within-subjects factor) between the two groups. Results: There was no significant difference between baseline measures of MBTG and TIG with the exception of duration of symptoms, where the TIG had experienced longer symptom duration than the MBTG. Overall, morning pain decreased significantly (p \ 0.001), as did pressure pain (p = 0.004), and function (measured with the FAAM questionnaire) improved (p \ 0.001), while no significant difference was found between MBTG and TIG. In evaluation of the heel rise test during baseline testing, half of the participants had VAS score between 0 and 1, which indicates that the test is not an effective tool for diagnosing plantar fasciitis. Results from those who scored high during baseline testing showed decrease of VAS score during the study period, indicating that the test may be of use for evaluating treatment response (Table 1). Conclusions: The results from this study indicate that both treatment forms, MBT shoes on the one hand and low dye taping and insoles on the other, improve symptoms and function in people with plantar fasciitis by decreasing pain and improving function.

P34-1052 The expectation of a return to play in 6 weeks following fifth metatarsal fracture in professional footballers: clinical reality or media inaccuracy? H. Tucker1, N. Jain2, D. Murray2, J. Calder3 1 University of Manchester, School of Medicine, Manchester, United Kingdom, 2North West Deanery, Trauma and Orthopaedics, Manchester, United Kingdom, 3The Lister Hospital, Clinic for Foot and Ankle Surgery, London, United Kingdom Objectives: Fifth metatarsal fractures are a common injury suffered by professional footballers. It is speculated that they have increased in frequency since the introduction of bladed footwear and playing on harder surfaces. It is frequently reported in the media that a professional footballer will be able to return to play 6 weeks following such an injury. This is rarely the case and leads to unrealistic expectations from players and clubs with increased pressures on the treating medical staff. The purpose of this study was to assess frequency of media reporting of fifth metatarsal fractures, the time that is predicted by the media before the player will return to soccer and the actual time taken for the player to return to play. Methods: Internet search engines identified 40 professional footballers that suffered 49 fifth metatarsal fractures between 2001 and 2011. Information was collected from various media and team websites,

S345 match reports as well as photography and video evidence to provide data regarding the mechanism of injury, playing surface conditions, frequency of fractures in each season, fracture treatment, estimated amount of time to be missed due to the injury and time taken to return to play. The actual time to return to play was defined as the first time after the injury that a player was selected in a first team match squad. Results: 49 fractures were identified in the 40 players. 6 players suffered a recurrence of a previous fifth metatarsal fracture. 3 players sustained fifth metatarsal fractures in their contralateral foot. There was a trend of increased reporting of fifth metatarsal fractures over the course of the decade. 43% of fractures were sustained without contact with another player, 24% were suffered in a tackle. 77% of fractures were sustained in dry conditions. 90% of fractures were treated surgically. The mean return to play time was 14.6 weeks (range 5–34 weeks). The mean estimated absence in the media was 7.8 weeks (range 2–16 weeks) with a median of 6 weeks. Conclusions: Fifth metatarsal fractures are commonly reported in professional footballers with an increasing trend. Most are treated surgically. It would appear that the commonly quoted period of 6 weeks before return to play is unrealistic and it should be expected that a player would be unavailable for 3 months.

P34-1189 A Multi-segment model study to evaluate the sagittal plane blockade in patients with Functional Hallux Limitus (FHL): results after endoscopic tenolysis of the Flexor Hallucis Longus (Fhl) at the retrotallar pulley J. Vallotton1, S. Diehl1, K. Boulos1, H. Rouhani1, K. Aminian1, Functional Hallux Limitus Study Group 1 Medicol Sa`rl/Clinique Bois Cerf, Lausanne, Switzerland Objectives: Evaluate the functional results in the sagittal plane biomechanics after release of the Flexor Hallucis Longus (Fhl) tendon at the retrotalar pulley, by assessing the kinematics and kinetics of the foot using a multi-segment model to describe a pathological gait pattern that is restored after endoscopic surgical release. Methods: A prospective cohort of 20 patients with FHL were analyzed before and after surgical endoscopic tenolysis of the Fhl and compared with 10 healthy subjects control group. The cohort was taken from a clinical setting and it was assured by a clinical examination that no other co-morbidities were present. FHL was diagnosed if patients had a positive stretch test. The control group was chosen randomly from a local sports and active healthy population with a negative stretch test. A complete orthopaedic clinical examination was performed looking for muscular; ligament or bony alterations. Foot posture was determined with the Foot Posture Indexa˜ and function was measured before and after surgery with the AOFAS Ankle-Hindfoot and AOFAS Midfoot validated scales. The multi-segment model consists of gait and joint angles analysis with the use of inertial sensors (Physilog, BioAGM, CH), plantar pressure analysis with the use of pressure insoles (Pedar, Novel, DE) and surface electromyography analysis with the use of EMG (Physilog, BioAGM, CH). The three data-loggers recorded synchronously while subjects walked in flat corridor and on an inclined treadmill at a fixed speed. Spatio-temporal parameters of gait, foot joint range of motion (ROM), and peak of angular velocity of segments in three anatomical planes; Pressure parameters, e.g., peak pressure, maximum force, time of occurrence, and contact time in different foot sub-regions and duration of activation, normalized EMG amplitude, and frequency content of selected muscles of shank were calculated. Results: With respect to the control group, results show an alteration of the gait pattern in the sagittal plane by increasing the flexion moment at the knee and ankle. A diminished dorsal flexion of the great toe at push-off, explaining the failure of the windlass

123

S346 mechanism, was also evident, which suggests a sagittal plane blockade when FHL is present. Results also showed an alteration of the plantar pressure distribution along the stance phase associated to a mistimed supination to pronation motion, necessary to gain instability before heel-strike and push-off. Conclusions: FHL causes a sagittal plane blockade, which is usually present at the retrotalar pulley and causes an asynchronic and mistimed gait that changes the biomechanics of the lower limb, specially the foot. Correction of this condition by an endoscopic retrotalar release restores normal gait biomechanics preventing people from irreversible conditions and painful syndromes.

P34-1390 Predictive variables of clinical outcome of osteochondral autologous transplantation in osteochondritis dissecans of the talus J. Woelfle1, H. Reichel1, M. Nelitz1 1 Ulm University, Department of Orthopaedic Surgery, Centre of Musculoskeletal Research Ulm, Ulm, Germany Objectives: Osteochondral autologous transplantation (OATS) from the ipsilateral femoral lateral condyle in osteochondritis dissecans (OD) of the talus has shown good clinical results in the past. However, little research has been done on limitations and indications of OATS. Methods: In this study, we evaluated the clinical outcome of OATS of 32 patients (median follow-up 23 months) by means of the AOFAS Ankle-Hindfoot Scale, ankle pain on the visual analogue scale (VAS), and HSS Patella Score. We conducted explorative statistical analysis to identify potential predictive variables such as age, pre-existing osteoarthritis, size of the lesion, necessity of malleolar osteotomy, localization of the lesion, or number of previous surgeries, which might be associated with limited success of the procedure. Results: Median AOFAS score was 86, median ankle pain on VAS was 2.0, and median HSS Patella score was 95. Patients above the age of 40 years had significantly higher donor-site morbidity when compared to younger patients. Apart from that, no predictive variables of poor clinical outcome such as pre-existing osteoarthritis, number of grafts, failed previous surgery, localization of the lesion, or necessity of malleolar osteotomy could be identified. Conclusions: OATS in OD of the talus is a safe procedure with good clinical results. As patients above the age of 40 years are at risk of high donor-site morbidity, osteoarthritis of the knee should be carefully excluded before performing OATS in these patients.

Muscle

P35-648 Low molecular weight dextran sulfate improves human myoblasts survival after intramuscular transplantation in mice T. Laumonier1, P. Hoffmeyer1, J. Menetrey1 1 University Hospital of Geneva, Orthopedic Surgery, Geneva, Switzerland Objectives: Myoblast transplantation remains a promising therapeutic approach in the treatment of several muscular disorders and severe muscle injury. Nevertheless, even in autologous combination, such therapies are limited by a massive early cell death, limited migration and rejection of transplanted myoblasts. Low molecular weight dextran sulfate (DXS), a sulfated polysaccharide, has been reported to act as a cytoprotectant in vitro and in vivo. In the present study, our aim was to test if DXS pretreatment can enhance the success of human myoblast transplantation.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: All research with human samples and mice was performed in compliance with the institutional guidelines, national as well as international guidelines. Clonal cultures of human myoblasts were prepared from single satellite cells and expanded as described previously. Cells were 100% positives for the myogenic markers CD56 and desmin. To follow human myoblasts in vivo, we have generated a bioluminescent lentiviral vector using the Renilla luciferase bioluminescence marker gene (Rluc). Mice were anesthetized using an isoflurane inhalation system. Rluc human myoblasts were injected intramuscularly in the Tibialis Anterior. Rluc substrate (Coelenterazine,1 mg/kg) was administered i.v and a whole body image was acquired using the IVIS 200 system. Results: In vitro, DXS binds dose dependently to human myoblasts and significantly inhibits staurosporine mediated apoptosis and necrosis. DXS pretreatment also protects human myoblasts from natural killer cell-mediated cytotoxicity in a dose dependent manner. Non-invasive bioluminescence imaging was used to quantify survival of engrafted human myoblasts after transplantation. At a MOI of 1, 80% of the cells expressed the renilla luciferase transgene after lentiviral infection. In vivo, we observed a two-fold improvement in survival of human myoblasts that were pretreated with DXS as compared to untreated cells, at 1 day and at 3 days after transplantation (77.9 ± 10.1% vs. 39.4 ± 4.9%; p = 0.0009 and 38.1 ± 8.5% vs. 15.1 ± 3.4%; p = 0.01, respectively). Conclusions: Our results indicate that DXS might represent a useful therapeutic reagent to improve the success of myoblast/myogenic precursor cell transplantation. DXS has been injected in human with little side effects. More experiments are needed to evaluate if combination of DXS pretreatment of human myoblasts with systemic DXS administration of the recipient may be envisaged to enhance myoblast transplantation success.

P35-801 Operative treatment of complete rectus femoris muscle ruptures in athletes L. Lempainen1, J. Sarimo2, K. Mattila3, J.T. Heikkila¨4, S. Orava2 1 Turku University Hospital, Department of Orthopaedic Surgery and Traumatology, Turku, Finland, 2Hospital Neo, Turku, Finland, 3 Medical Imaging Centre of Southwest Finland, University Hospital of Turku, Turku, Finland, 4Mehila¨inen Sports Trauma Research Center, Hospital and Sports Clinic, Turku, Finland Objectives: Rectus femoris muscle injuries are common in sports. Most of these injuries are mild strains or contusions which can be treated conservatively with good results. There are, however, also more severe injuries in which the decision of optimal treatment method is not so evident. The main aim of this study was to evaluate the effect of surgery on complete mid-substance rectus femoris muscle tears in athletes. Methods: During the years 1997–2011 a total of 27 patients with a complete rectus femoris muscle rupture were treated operatively at our center. The mean age of the patients was 29 years. All of the patients were athletes and the injuries were sports related. In fourteen cases the injury mechanism was falling with overstretching the rectus femoris muscle. Other common injury mechanisms were maximal sprinting or hard kicking while playing soccer. In three cases a complete rectus femoris rupture was a result of a direct contusion. The diagnosis was confirmed preoperatively using an MRI. In eight cases the operation was performed acutely (\ 4 weeks) after injury. In thirteen cases the initial early diagnosis was delayed and the operation was performed between one and four months from the injury. The rest six patients were first treated conservatively (between 6 and 12 months) after injury and the operation was done because of unsuccessful result after conservative treatment. A four category

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 rating system was used to evaluate the overall result. Retrospective study protocol. Results: The mean length of the follow-up was 30 months. Of the 27 patients, 20 had excellent or good results (free of symptoms and complete return to pre-injury level of sports or minor symptoms in strenuous athletic activity) and they were able to return to their former level of sport after an average of five months. In seven patients, all operated after three months from the injury, the result was classified as moderate (weakness and atrophy in the rectus femoris muscle, inability to do strenuous activities). All 27 patients felt that they had benefited from the surgery, and their performance as well as the strength of the operated anterior thigh had improved after the operation. Postoperatively there were four hematoma formations which were drained percutaneously and one superficial wound infection that was successfully treated with oral antimicrobial drugs. One patient had to undergo a reoperation during the follow-up period because of the rerupture of the operated rectus femoris muscle. Conclusions: According to our results, it seems that excellent or good results may be often expected after early surgical repair of complete rectus femoris ruptures. When choosing a suitable treatment method, it should be remembered that a complete rectus femoris muscle injury can be a career ending injury for an athlete. Further studies are required to evaluate the role of conservative treatment in these injuries.

P35-807 The influence of mouthguards on muscular performance T. Machado1, R. Pereira1, M. Cervaens1 1 Universidade Fernando Pessoa, Porto, Portugal Objectives: The objective of this experimental and prospective study was to verify the influence of mouthguards on quadriceps and hamstring performance of senior indoor soccer athletes. The sample consisted of 15 healthy male athletes with an average age of 23 years. Methods: Subjects underwent three different moments of evaluation, using custom-made mouthguards, boil-and-bite mouthguards and without any type of mouthguards, during three consecutive days. To eliminate learning bias, the use of the different mouthguards was randomly assigned. The trials were tested on an isokinetic dynamometer Biodex System 4-Pro (New York, USA). The isokinetic protocol consisted of two different speeds: 60/s and 300/s, and the variables analyzed were: peak torque, total work, work fatigue and agon/antag ratio. The collected data was analyzed using the 19.0 version for Windows of the Statistical Package for the Social Sciences (SPSS). Initially, it was used the Shapiro–Wilk test to ensure the normality of the sample. Afterwards, to compare the use of different mouthguards and different parameters, the analysis of the average was performed through the T test—paired samples. Differences were considered statistically significant when p \ 0.05. Results: The results revealed a significant difference by using the different types of mouthguards. Custom-Made mouthguards obtained significant results at 60/s, in peak torque of quadriceps (p = 0.014) and hamstrings (p = 0.015) and at 300/s in work fatigue of the hamstrings (p = 0.039), while Boil-and-Bite mouthguards, obtained significant results, at 60/s, in hamstring peak torque (p = 0.020) and at 300/s in peak torque (p = 0.042) and total work (p = 0.025) of the same muscle, however there were no differences in quadriceps. In relation to agon/antag ratio there were no differences with any protector at any speed. Conclusions: This study verified that both mouthguards had influence on the performance of these athletes specifically on the values of peak torque, total work and work fatigue, although there were no differences in agon/antag ratio.

S347 Future research is needed to obtain guidelines for improving the usefulness and developing the mouthguards.

P35-1192 Experimental study on the influence of Intramuscular injection of Plasma Rich Growth Factors in serum Insuline like Growth Factors-I (IGF-I) and C-Reactive Protein (CRP) in the DOG M. Rubio1, E. Damia2, J.J. Ceron3, R. Cugat Bertomeu4, M. Garcia-Balletbo5, J.M. Carrillo1 1 CEU-Cardenal Herrera University, Animal Medicine and Surgery, Moncada, Spain, 2Global Veterinaria, Moncada, Spain, 3University of Murcia, Animal Medicine and Surgery, Espinardo, Spain, 4Hospital Quiron Barcelona, Orthopaedics and Traumatology, Mutualidad Catalana de Futbolistas, Barcelona, Spain, 5Hospital Quiron, Orthopaedic Surgery, Barcelona, Spain Objectives: The use of Plasma Rich in Growth Factors (PRGF), in muscle injury was prohibited by WADA in January 2010. A possible reason for this prohibition was that the PRP would give an increase in insulin growth factor (IGF-I) systems, which could have beneficial and advantageous effects on muscle performance for the athlete. To the author’s knwoledge, there are no studies in which IGF-I and C-reactive protein (CRP) have been measured after PRGF or PRP intramuscular application. So, the aim of this work was to study the influence of local PRGF injection in healthy muscle in the canine species, looking at serum levels of IGF-I and CRP and their relationship. Methods: Prospective crossover study design. The study was conducted with 8 Beagle dogs, infiltrated with 3 different protocols into healthy muscle (L5 level) with a month free period between them. • Group 1 (PRGF): 1 ml of PRGF with 0.05 ml of 10% Ca2Cl. • Group 2 (PCB): 1 ml of SSF + 0.05 ml of 10% calcium chloride (Ca2Cl). • Group 3 (HPRGF): 3 ml of PRGF with 0.15 ml of 10% Ca2Cl. Blood samples were taken: baseline and at the following times postinjection: 10 , 150 , 300 , 1, 6, 12, 24 h, 7 and 14 days. Blood samples were obtained by venipuncture directly into the jugular vein. To measure the growth factor IGF-I and CRP, serum tubes were used with a volume of 5 ml. The IGF-I was analyzed by automated immunoassay system previously validated for optimal use in the canine model. The concentration of CRP was determined through an immunoturbidimetric assay validated for use in the canine model. The data were analyzed using SPSS 15.0 for Window. The result of each parameter was evaluated with a nonparametric test KolgomorovSmirnov test for normality, and log transformed if necessary. Platelets and IGF-I in blood, PRGF and Platelets Poor Plasma (PPP), were compared with repeated measures. A descriptive study of the levels of IGF-I and CRP over time was performed, and a comparison of means over time in each protocol was analyzed with an ANOVA and posthoc Tukey test to assess differences within groups. A correlation between both variables was performed to assess their behaviour with a Kendall’s Tau-b correlation test. Results: When compared blood, platelets poor plasma and PRGF, between protocols, there were no differences in the level of platelets and IGF-I, so the infiltration was similar in all groups. When IGF-I and CRP were compared throughout the procedure, there were no differences between times in the three protocols, and there was a significant correlation between them. Conclusions: We were able to conclude with the study of our 8 patients that IGF-I does not increase at a systemic level after local injection of PRGF. So, if systemic levels of IGF-I were maintained after local application of PRGF, athletes could still benefit from the advantages of it use, resulting in economic and temporary benefit. In fact WADA retracted its decision in December 2010.

123

S348

Tendon

P36-499 The effect of deep friction massage on the Achilles tendon blood flow N. Mahieu1, E. Pattyn1, A. Maenhout1, E. Witvrouw1 1 Rehabilitation Sciences and Physiotherapy, Ghent University, Ghent, Belgium Objectives: Cyriax’s friction or deep transverse friction massage (DTF) is a specific type of connective tissue massage applied precisely to the soft tissue structures such as tendons. Despite the underlying mechanism is not evidence- based, it is hypothesized in literature that the mechanical action of DTF produces vasodilatation and increases blood flow. This may facilitate the removal of clinical irritants, and increase the transportation of endogenous opiates, resulting in a decrease in pain. But as yet, no experimental study has confirmed the effect of DTF on blood flow in the Achilles tendon. Therefore, the purpose of this study was to investigate whether one session of DTF can influence the Achilles tendon blood flow. Methods: A randomized pre- posttest trial was set up to assess the effect of a one session DTF. Sixty-two healthy volunteers (31.87 ± 13.09 years) were prepared to take part in the study. Each person underwent the following four steps of the procedure: 1. measuring the tendon blood flow (PRE), 2. fifteen minutes of DTF on the Achilles tendon, 3. measuring of the tendon blood flow (POST), 4. follow—up measurement of tendon blood flow after 20 min of rest (POST20). One leg of each person was randomly assigned to deep transverse friction session, the other leg was used as a control leg. For the application of the DTF, the participants were positioned supine with the ankle in 90 of dorsiflexion, in order to bring some tension on the tendon. The clinician stabilized the subject’s foot with one hand while applying the DTF with the other hand. The DTF is a pinching technique with thumb and index finger, reinforced by the middle finger. Friction was applied continuously for a total of 15 min. The microcirculation was determined at 2 and 8 tissue depths at the distal and the proximal midportion of the Achilles tendon. For each level, the capillary blood flow, the tissue oxygen saturation, and the postcapillary venous filling pressure was registered at PRE, POST and POST20. Results: During the PRE-measurement, for all measurement variables, there is no significant difference between the control leg and the friction leg. For all the different variables, the linear mixed model showed a significant effect for time, group and group*time. Therefore, all post-hoc tests of the linear mixed model could be interpreted. The blood flow and the tendon oxygen saturation increased significantly after DTF, while the postcapillary venous filling pressure decreased significantly. After 20 min of rest, all parameters normalized towards their baseline-values. Conclusions: These findings confirm the hypothesis that due to the higher blood flow, DTF is able to increase the tissue metabolism. To what extent the described microcirculatory effects of sole DTF are evident in patients with Achilles tendinopathy should be investigated in future research.

P36-557 Platelet-rich plasma as a treatment for patellar tendinopathy: a double-blind randomized controlled trial J.L. Dragoo1, A.S. Wasterlain1, H.J. Braun1 1 Stanford University, Orthopedic Surgery, Redwood City, United States

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Patellar tendinopathy is difficult to treat. Previous research has shown that in patients with patellar tendinopathy, mean Victorian Institute of Sport Assessment (VISA) scores improved after treatment with dry needling plus platelet-rich plasma (PRP). However, there have been no randomized controlled trials comparing treatment with dry needling and PRP to dry needling alone in patellar tendinopathy. Methods: 11 patients were enrolled in a double-blind randomized controlled trial comparing dry needling with PRP to dry needling alone in patients with patellar tendinopathy on clinical exam and MRI imaging. Treatment groups consisted of standardized eccentric exercises and either dry needling plus ultrasound-guided PRP injection (PRP group, n = 6) or dry needling alone (dry needling group, n = 5). Participants completed the VISA, Tegner, Lysholm and SF12 questionnaires before and at 3, 6, 9 and 12 weeks after the intervention. Results: All survey data were analyzed using two-tailed paired T tests and intention to treat. All patients and researchers remained blinded throughout the 12-week follow-up period. Within the PRP group, we observed clinically and statistically significant improvements by a mean of 33.0 points for VISA (p = 0.02), 3.2 for VAS (p = 0.01), and 35.8 for Lysholm (p = 0.02). Within the dry needling group, scores improved by 27.3 points for Lysholm and by 1.0 points for VAS, but worsened by 1.0 point for Tegner, by 1.8 points for VISA, and by 1.5 points for SF-12 at 12 weeks. Between-groups comparisons showed a clinically and statistically significant 31.3-point greater improvement in VISA scores in the PRP group relative to the dry needling group (p = 0.04), and a clinically significant 3.4-point greater improvement in Tegner scores (p = 0.17).

VISA score of symptom severity Conclusions: Published scientific literature suggests the minimum clinically significant difference is 12 points for VISA and 1 point for Tegner. Thus, our results indicate that a therapeutic regimen of eccentric exercise and ultrasound-guided dry needling with PRP leads to clinically significant improvement in activity, pain, and symptoms at 12 weeks; and that this regimen is both clinically and statistically significantly better than exercises and dry needling alone based on VISA score of patellar tendinopathy symptoms. This supports existing evidence that PRP may accelerate healing time following tendon injuries, and provides additional evidence that PRP treatment leads to better patient-reported clinical outcomes than alternatives. Although previous reports have suggested a positive effect of dry needling, our results suggest that dry needling without PRP did not improve outcomes.

P36-625 Intermittent pneumatic compression enhances metabolic activity in human Achilles tendon repair E. Arverud1, G. Nilsson1, F. Labruto1, D.K.I. Bring1, P.W. Ackermann1

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 1

Karolinska Institutet, Molecular Medicine and Surgery, Stockholm, Sweden Objectives: Achilles tendon healing after rupture is often protracted and associated with a high degree of complications such as deep venous thrombosis. This may partly be due to a sparse blood circulation and a low metabolic activity in the tendon, which are even more restricted during limb immobilization. Basic metabolites including glutamate, glucose, lactate, pyruvate and glycerol, are known to meet biosynthetic and proliferative healing demands. This study assessed whether these metabolites are present and up-regulated in the healing Achilles tendon. We moreover hypothesized that adjuvant intermittent pneumatic compression (IPC), known to increase local blood flow, can up-regulate metabolite concentration at the tendon repair site. Methods: As part of a larger prospective randomized trial on acute Achilles tendon rupture eleven patients were recruited, operated on and subsequently randomized. The control group, five patients, received conventional treatment with two weeks of plaster cast immobilization, while 6 patients received adjuvant foot IPC system (Covidien, New Haven, Conn., USA) beneath the plaster cast. The settings used were 130 mm Hg pressure, 1 s pressure duration, 20 s compression frequency, 6 h daily. At 2 weeks post-operatively microdialysis (CMA 71; CMA Microdialysis AB, Solna, Sweden; 100 kDa: 1.0 lL/min) of the healing and contralateral intact Achilles tendons was performed followed by quantification of metabolites using ISCUS Analyzer (CMA Microdialysis AB, Solna, Sweden). Significance p less than 0.05. Results: The control group exhibited in the healing Achilles tendons increased levels (mM) of glutamate (60 ± 12 vs. 20 ± 12), lactate (1.3 ± 0.4 vs. 0.8 ± 0.3), and pyruvate (81 ± 24 vs. 43 ± 33) compared to the contralateral intact tendons (p = 0.028), whereas the levels of glucose, glycerol and the lactate/pyruvate ratio were not significantly changed. The injured tendons of the IPC versus control group displayed higher levels of glutamate (84 ± 15 vs. 60 ± 12) and glucose (3.4 ± 0.6 vs. 2.5 ± 1.1); (p = 0.043) and produced a trend toward higher concentrations of pyruvate (113 ± 33 vs. 81 ± 24; p = 0.068). The lactate/pyruvate ratio and the levels of lactate and glycerol were however not significantly changed after IPC. Conclusions: This study demonstrates that early human Achilles tendon repair entails and up-regulates local essential metabolites. The reparative metabolic response can be promoted by adjuvant IPC during plaster cast immobilization. The up-regulation of glutamate levels in the healing tendons are in agreement with studies showing nerve ingrowth and subsequent release of several neuronal transmitters including glutamate, considered to be involved in the repair process by enhancing cellular proliferation. It may prove that compression therapies can optimize healing, allow earlier rehabilitation and reduce the time to return to sports after tendon injury.

P36-640 Histological study of the influence of Plasma Rich in Growth Factors (PRGF) in a model of tenotomized Achilles tendon in sheep J.A. Fernandez-Sarmiento1, J.M. Dominguez1, M.M. Granados1, J. Martin de las Mulas2, Y. Millan2, M. Garcia-Balletbo3 1 University of Cordoba, Small Animal Medicine and Surgery, Cordoba, Spain, 2University of Cordoba, Comparative Pathology, Cordoba, Spain, 3Hospital Quiron, Orthopaedic Surgery, Barcelona, Spain Objectives: Acute rupture of the Achilles tendon (AT) is a common injury with a challenging therapeutic management in some cases. The use of Plasma Rich in Growth Factors (PRGF) has been proposed to improve healing after tendon injuries, but there is an intense debate about the effectiveness of this biological therapy. The objective of

S349 this study was to evaluate the histological effects of PRGF on tendon healing in an experimental model. Methods: Twenty-eight adult sheep were anaesthetized and a tenotomy of the AT was performed. Animals were randomly divided in four groups. Two groups received an infiltration of PRGF on the repaired area just after repairing the tenotomy and every week during the following 3 weeks. The other two groups received a placebo injection with saline. One PRGF-treated group and one saline group were euthanized at 4 weeks, and the other two groups were euthanized at 8 weeks. The following parameters were determined by a computer-assisted histopathological examination in order to evaluate the repaired area: morphometry of fibroblast nuclei, fibroblast density and neovascularisation. Arrangement of collagen fibers and inflammatory response were evaluated using a semiquantitative grading scale. Statistical analysis using the non-parametric Kruskas-Wallis test and the Mann–Whitney U test was perform to analyze the significance between groups. Results: Fibroblast nuclei of PRGF-treated tendons were more elongated (p = 0.008) and more parallel to tendon axis (p = 0.016) than fibroblast nuclei of saline group at 8 weeks. PRGF-treated tendons showed more packed (p \ 0.001) and better oriented (p \ 0.001) collagen bundles, both at 4 and 8 weeks. In addition to an increased maturation of the collagen structure, fibroblast density was significantly lower (p = 0.008) in tendons infiltrated with PRGF. PRGFtreated tendons exhibited faster vascular regression than saline groups, as it was shown by a lower vascular density at 8 weeks (p = 0.008). A lower inflammatory cell infiltration was observed in tendon infiltrated with PRGF, both at 4 and 8 weeks. Conclusions: The findings of this study showed that PRGF accelerated the healing process in tendon after a surgically-induced rupture of AT in sheep. PRGF-treated tendons showed an improvement in morphometric features of fibroblast nuclei, suggesting a more advanced stage of healing. Histological exam of tendons treated with PRGF revealed a more mature organization of collagen bundles, a lower vascular density and a decrease in fibroblast density, producing a more advanced histological appearance of healing process than in tendons infiltrated with saline, mainly 8 weeks after injury. Further biomechanical tests should confirm the histological findings reported in this study. These results provide evidences that PRGF infiltration may be a promising supplementary therapy for the treatment of acute AT rupture.

P36-873 A 5 year follow-up study of Alfredson’s heel-drop exercise programme in chronic midportion Achilles tendinopathy J. Tol1, A. Van Der Plas2, H. van Der Heide3, J. Verhaar4, R.-J. de Vos2, S. De Jonge2 1 Aspetar; Qatar Orthopaedic and Sports Medicine Hospital, Doha, Qatar, 2The Haghue Medical Centre, Leidschendam, The Netherlands, 3Leiden University Medical Centre, Department of Orthopaedics, Leiden, The Netherlands, 4Erasmus University Rotterdam, Rotterdam, The Netherlands Objectives: Eccentric exercises have the most evidence in conservative treatment of midportion Achilles tendinopathy. Although short term studies show significant improvement, little is known of the long term ([3 years) results. The aim of this study is to evaluate the 5 years outcome of patients with chronic midportion Achilles tendinopathy treated with the classical Alfredson’s heel-drop exercise programme, Methods: Study design: Part of a 5 years follow-up of a previously conducted randomised controlled trial. 58 patients (70 tendons) were approached 5 years after the start of the heel-drop exercise programme according to Alfredson. At baseline and at 5 year follow-up the validated Victorian Institute of Sports

123

S350 Assessment-Achilles questionnaire (VISA-A score), pain status, alternative treatments received and ultrasonographic neovascularisation score were recorded. Results: In 46 patients (58 tendons) the VISA-A score significantly increased from 49.2 at baseline to 83.6 after 5 years (p \ 0.001) and from the 1 to 5 years follow-up from 75.0 to 83.4 (p \ 0.01) 39.7% of the patients were completely pain free at follow-up, and 48.3% had received one or more alternative treatments. The sagittal tendon thickness decreased; from 8.05 mm (SD 2.1) at baseline to 7.50 mm (SD 1.6) at the 5 years follow-up (p = 0.051). Conclusions: At 5 years follow-up a significant increase of VISA-A can be expected. After the three month Alfredson’s heel-drop exercise programme almost half of the patients had received other therapies.

P36-882 Bioresorbable polylactide (PLA) beads promote tendon healing in the bone tunnel K. Ficek1, J. Wieczorek2, E. Stodolak3, M. Kajor4, Y. Koseniuk5 1 Galen Orthopaedics, Bierun, Poland, 2Institute of Animal Production, Department of Biotechnology, National Research Institute, Balice, Poland, 3University of Science and Technology, Department of Biomaterials, Krakow, Poland, 4Medical University of Silesia, Department of Pathology, Katowice, Poland, 5Institute of Animal Production, Department of Biotechnology, Balice, Poland Objectives: The outcome of reconstructive ligament surgery is dependent on the quality of healing of the transplanted grafts in the bone tunnels. A major part of the transplant anchored inside the bone is suspected to be integrated relying on the biological processes occurring within the tunnels. Improper graft fixation may cause inadequate joint stability. In this study, microporous polylactide (PLA) beads of a controlled pore size were impregnated with peripheral blood and evaluated in bone tunnels of rabbits to assess their potential to enhance tendon graft osteointegration after ligament reconstruction surgery. Methods: The beads with a pore size in the range of 200–400 lm were produced from poly (L/DL-lactide)—PURAC Biochem, Gorinchem, The Netherlands. 24 rabbits (Ethics Committee approval 668/09, July 7, 2009) were divided into 4 groups of 6 animals each: Group 1 (control), no PLA, implantation time 6 weeks; Group 2, PLA, implantation time 6 weeks; Group 3 (control), no PLA, implantation time 12 weeks; Group 4, PLA, implantation time 12 weeks. Under general anesthesia the long digital extensor tendon of the right hind limb at the lateral femoral condyle was exposed and cutt off. The bone tunnel, with a diameter of 2.6–3.5 mm and length 8–12 mm, was drilled in the proximal tibia metaphysis. The detached tendon was implanted in the bone tunnel. At this moment, the implementation, PLA material was distributed around the graft, starting from its medial entrance. After this application, the maneuver was repeated at the lateral outlet of the tibial tunnel. The animals were euthanized 6 or 12 weeks after surgery. Results: Histological scans confirmed the osteoconductive attributes of the biomaterial and their synergism with osteoinductive properties of the cells from the blood and bone marrow. This compound, administered in the tibial, vascular abundant spongy bone, significantly promoted the transplanted grafts’ osteointegration within the bone tunnels, which was aided by proper tendon tension during implementation. Proliferation of the cartilage around the tendon [um2] was statistically significant in both PLA groups after 6 weeks p = 0.034 and after 12 weeks p = 0.014 (P values \0.05 were considered) in post hoc Sheffe test. Conclusions: The success of the methods in this experiment is encouraging. However, non-uniform distribution of the PLA paste in

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 the bone tunnel around the grafted tendon results in a patchy bone formation. The new bone is almost exclusively formed in these areas where PLA paste is in direct contact with both, the tunnel bone and the outer surface of the tendon i.e. the conditions which facilitate bone regeneration. Further study is required to define an optimal handling procedure that will allow for uniform distribution of the PLA-blood paste within the bone tunnels. Acknowledgments: Bioresorbable polylactide implants used in the study were kindly provided by Professor Sylwester Gogolewski, D.Sc., Ph.D.

P36-888 Endoscopic treatment of the tendinopathy of the Achilles tendon I. Zazirnyi1, V. Yevsyeyenko1, V. Rybka2, Y. Grigoreva2 1 Hospital Feofania, Centre of Orthopaedia, Trauma and Sport Medicine, Kiev, Ukraine, 2Hospital Feofania, Histopathology, Kiev, Ukraine Objectives: The aim of our work was to study the efficacy of endoscopic techniques in tendinopathy of the Achilles tendon as in sportsmen of high qualification, and in ordinary patients. Methods: We have treated from March 2008 till May 2011 24 patients, 14 professional athletes (high jumper and sprinter) and 10 patients who never played sports. Men were 16, women—8, age of patients was from 25 till 57 years, average age was 36.7 years. All patients were performed ultrasound examination (US) and magnetic resonance imaging (MRI). AT function and activity were evaluated by assessment system VISA-A (‘‘Sports Assessment—Achilles tendinous’’ Victorian Institute of Sport—A, maximum of 100 points), Nelen Achilles Tendinopaty Scoring System (NATSS)—maximum 100 points, and visual analog scale (VAS)—maximum 10 points. Endoscopic tendon mobilization has been done to all patients. In 10 cases we perform histopathological studies of the tissues in lesion focus. Results: According to estimates of average VISA-A score was 40.9 before surgery and after surgery—85.5. In assessing the operation by NATSS function evaluated at an average 43.1 points, and after surgery—89.1 points. The average VAS score decreased from 7.2 before surgery to 1.5 points after surgery. In all patients we did not observe postoperative complications. According to histopathological studies in the lesion focus is obtained deformation and rupture of collagen fibers, proliferation of blood vessels and nerve damage zone with the development of immature granulation tissue. Conclusions: Use of endoscopic surgical treatment can effectively restore the function of Achilles tendon in cases of tendinopathy.

P36-935 Proprioception impairment after endoscopically guided percutaneous Achilles tendon repair is related to perceived ankle–foot stiffness D. Kaya1, M.N. Doral2, G. Donmez1, E. Turhan3, O. Atay4, J. Nyland5 1 Hacettepe University, Ankara, Turkey, 2Hacettepe University, Faculty of Medicine, Department of Orthopedic and Sports Medicine, Ankara, Turkey, 3Zonguldak Karaelmas University, Zonguldak, Turkey, 4Hacettepe University, Orthopedics and Traumatology, Ankara, Turkey, 5University of Louisville, Orthopaedic Surgery, Louisville, United States Objectives: To evaluate ankle function following endoscopically guided percutaneous Achilles tendon repair. Our hypothesis was that compared to the unaffected contralateral ankle, patients that had undergone this surgery and had successfully completed rehabilitation would still display impaired involved side ankle proprioception.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: After warming up for 5 min at a self-selected intensity on a stationary bike 19 men (age = 45 years, range 32–59, height = 174 cm, range 167–182, bodyweight = 85 kg, 65–100, time postsurgery = 30 ± 17.7 months, range 12–72) were tested for bilateral ankle active angle reproduction at 10 dorsiflexion and 15 plantar flexion, peak concentric isokinetic ankle dorsi- and plantar flexor torque, one-leg hop for distance, single leg vertical jump height, and perceived function using the Foot and Ankle Outcome Score (FAOS). A series of paired sample t tests were used to compare side-to-side differences (P \ 0.05). Pearson product moment correlations were used to determine the strength of relationship between torque, perceived and actual function and ankle proprioception. Results: Active or passive ankle range of motion, peak isokinetic torque, one-leg hop for distance, single leg vertical jump for height and ankle joint position sense at 10 dorsiflexion did not differ between the affected and unaffected side. Ankle joint position sense for active angle replication at 15 plantar flexion revealed a significant side-to-side difference. FAOS results revealed good to excellent results for ‘‘pain’’ 95.0 ± 8.5, ‘‘other symptoms’’ 88.4 ± 13.5, ‘‘function in daily living’’ 95.1 ± 6.4, ‘‘function in sport and recreation’’ 84.3 ± 15.6, and ‘‘quality of life’’ 75.9 ± 20.3. Active ankle angle replication at 15 plantarflexion and the ‘‘other symptoms’’ subscale of the FAOS revealed an inverse, moderate correlation (r = -0.50, p = 0.02). Conclusions: We identified affected side ankle joint position sense impairment at 15 plantarflexion in patients who were a minimum of 1 year post-surgery. This deficit was inversely related to their perception of ankle–foot stiffness at the affected side using the FAOS ‘‘other symptoms’’ subscale. Future studies should evaluate therapeutic interventions designed to improve neuromuscular reactivity, and decrease kinesiophobia among these patients and attempt to determine the appropriate intervention timetable in relationship to known connective tissue healing parameters. Comparisons with other patient groups such as those who have undergone open Achilles tendon surgery are also suggested.

P36-965 The positive effects of different PRP methods on human muscle, bone, and tendon cells K. Beitzel1, M. McCarthy1, M.P. Cote1, R. Russell1, R.A. Arciero1, A.D. Mazzocca1 1 University of Connecticut Health Center, Department of Orthopedic Surgery, Farmington, United States Objectives: Clinical application of platelet rich plasma (PRP) in the realm of orthopaedic sports medicine has yielded variable results. Differences in separation methods and variability of the individual may contribute to these variable results. The purpose was to compare the effects of different PRP separation methods on human bone, muscle and tendon cells in an in vitro model. Methods: Blood collected from 8 subjects (mean age 31.6 ± 10.9 years) was used to obtain PRP preparations. Three different PRP separation methods were used: A single spin (PRPLP) process yielding a lower platelet concentration, a single spin (PRPHP) process yielding high platelet and white blood cell (WBC) concentration, and a double spin (PRPDS) which produces a higher platelet concentration and less WBCs. Human bone, muscle, and tendon cells obtained from discarded tissue samples during shoulder surgery were placed into culture and treated with the 3 PRP preparations, control media (2 and 10%), and native blood. Radioactive thymidine assays were obtained to examine cell proliferation and ELISA testing was used to determine growth factor concentrations. Results: Addition of PRPLP to bone, muscle, and tendon increased compared to the controls proliferation (p \ 0.05). Adding PRPHP to

S351 bone and muscle showed no significant increase in proliferation compared to controls or other separation methods. There was an increase in proliferation for addition of PRPHP to tendon compared to the controls (p \ 0.05). Adding PRPDS to bone increased proliferation compared to all controls as well as PRPHP (p \ 0.05). The addition of PRPDS to tendon increased proliferation compared to all controls (p \ 0.05). Conclusions: The application of each of three PRP separations increased cell proliferation for bone, muscle, and tendon cells compared in vitro compared to control cells cultured in media and native blood however differences between methods of separation in the amount of cell proliferation were observed. The utilization of different PRP separations may result in a potential beneficial effect on the clinically relevant target cells in vitro however it is unclear which platelet concentration or PRP preparation may be optimal for the treatment of various cell types.

P36-1096 Platelet rich plasma injection after arthroscopic rotator cuff tear repair F. Franceschi1, R. Papalia2, E. Franceschetti3, A. Palumbo4, V. Denaro2, N. Maffulli5 1 Campus Biomedico University, Orthopaedic and Trauma Surgery, Rome, Italy, 2Campus Biomedico University, Orthopaedic Surgery, Rome, Italy, 3University Campus Biomedico of Rome, Rome, Italy, 4 University Campus Biomedico of Rome, Orthopaedic and Trauma Surgery, Rome, Italy, 5Queen Mary University of London, Barts and The London School of Medicine and Dentistry, Centre for Sports and Exercise Medicine, London, United Kingdom Objectives: The purpose was to determine whether injection of Platelet-rich plasma accelerates recovery and enhanced structural integrity of repaired tendon after arthroscopic rotator cuff. Methods: Thirty-eight patients with full-thickness rotator cuff tears were included. Patients were informed about the use of PRP and decided themselves whether to have PRP injection after the time of surgery. Seventeen patients underwent arthroscopic rotator cuff repair with PRP (three PRP injections at *10-days intervals) and 21 underwent arthroscopic rotator cuff repair without PRP injection. Outcomes were assessed preoperatively and at 3, 6, 12, and finally at a minimum of 16 months after surgery (at an average of 17.7 ± 1.7 months) with the following scoring systems: Constant system, the University of California at Los Angeles (UCLA) system and a Visual Analogue Scale (VAS) scale. In addition range of motion and strength in all planes were assessed preoperatively and at 3, 6, 12, and at final follw-up. At a minimum of 6 months after surgery, repaired tendon structural integrities were assessed by magnetic resonance imaging. All patients had the same rehabilitation protocol. Results: Platelet-rich plasma gel application after to arthroscopic rotator cuff tear repairs did not accelerate recovery with respect to pain, range of motion, strength, functional scores, or overall satisfaction as compared with conventional repair at any time point. There was no difference between the 2 groups after 3, 6, 12, months and at final follow up. The follow-up MRI showed no significant difference in the healing rate of the rotator cuff tear. In addition, magnetic resonance imaging, at a minimum of 6 months after surgery, demonstrated a retear rate of 23.5% in the PRP group and 19% in the conventional group, there was no statistical significance between the groups (P = .658). Conclusions: The results suggest that PRP application after arthroscopic rotator cuff tear repair did not accelerated recovery clinically or anatomically. Further studies are needed to confirm these results and understand the mechanism of action, and to find other application

123

S352 modalities, with different platelet concentrations and injection timing, which provide better results.

P36-1287 Fokused shockwave therapy for the treatment of chronic plantar fasciitis: a randomized comparative experimental study B.A. Rydberg1, M. Westin2, A. Heijne3 1 Hela Kroppen Sjukgymnastik AB, Stockholm, Sweden, 2Karolinska Institutet, Stockholm Sports Trauma Research Center, Stockholm, Sweden, 3Karolinska Institutet, Institutionen fo¨r Neurobiologi, va˚rdvetenskap och samha¨lle, Sektionen fo¨r Sjukgymanstik, Huddinge, Sweden Objectives: The purpose of this study was to compare focused shockwave therapy combined with stretching of the plantar fascia and achilles tendon with stretching alone in patients suffering from chronic plantar fasciitis. Methods: Thirty-two participants diagnosed with chronic plantar fasciitis were enrolled. They were randomly selected to start a treatment regime where they either received focused shockwave treatment on three occasions (n = 18), once weekly and daily stretching (SW), or were part of the reference group (R) (n = 14), whom only conducted daily stretching of the plantar fascia and achillestendon. Self-estimated foot function was assessed using the Foot and Ankle Outcome Score (FAOS) before and one month after finishing treatment. Pain during the first steps after inactivity and pain in everyday activity was assessed using the Visual Analogue Scale (VAS) before treatment and one month after finishing treatment, as well as during the three treatment sessions. Thickness of the plantar fascia was measured with ultrasound scanning before treatment and one month after finishing treatment. Results: At one month after treatment there was a significant reduction of pain for both groups compared with baseline, estimated by both FAOS and VAS. The SW- group showed a significant improvement in the subgroup Pain (p = 0.05) (FAOS) and for Pain in everyday activity (VAS) (p = 0.003). The reference group showed a significant improvement in the subgroup of Pain (p = 0.007) and in the subgroup Foot related quality of life (FAOS) (p = 0.02). The reference group showed a significant difference before and after the intervention period (p = 0.008) in terms of Pain at first steps after inactivity (VAS). A significant difference, in favor of the reference group, were observed in the subgroup of Pain (FAOS) (p = 0.02) after treatment. Other parameters were not significant either within or between groups. No significant change in plantar fascia thickness was found in either of the two groups at one month after treatment. SWgroup (p = 0.37) and reference group (p = 0.35). Conclusions: The current study could not demonstrate that the shockwave treatment, as a supplement to stretching of the plantar fascia and achilles tendon, was more effective compared to stretching alone in patients suffering from chronic plantar fasciitis. This study is ongoing aiming to enroll totally sixty patients. A longterm follow up is planned 1 year after treatment, therefore part of long term data can be presented in May 2012.

P36-1304 The collagen III expression analysis using subcutaneous ruptured EPL tendon K. Sasaki1, A. Kunamneni1, A.P. Sinai2, S. Kamineni1 1 University of Kentucky, Orthopaedics and Sports Medicine, Lexington, United States, 2University of Kentucky, Microbiology Immunology and Molecular Genetics, Lexington, United States

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Extensor tendon rupture at the dorsum of the wrist is commonly seen in patients with rheumatoid arthritis. Spontaneous ruptures in association with rheumatic disease, therefore account for most of the non-traumatic ruptures. To examine the expression and change of collagen subtype III in tendon may be helpful to know the mechanisms of tendon rupture. The purpose of study is, using histological analysis and western blotting, to study the expression and measure the change of collagen I and III using ruptured extensor pollicis longus (EPL) tendon. Methods: This study was approved by the ethical committee of University of Kentucky and informed consent was obtained from the patients before specimen collection. The case spontaneous rupture of EPL tendon was 76-years old male with the history of rheumatoid arthritis. After EPL was performed to transfer the extensor indicis proprius (EIP), the ruptured end was cut into each 2 mm slices. Each slice was designated as EPL1 through EPL4 from ruptured end to distal. The EIP was also used as control. The expression patterns of collagen I and III in the EPL and EIP tendon were evaluated using picrosirius red staining and western blotting. During surgery, resected each specimen were divided into 2 pieces. One piece was fixed in 10% formalin immediately; the second one was used for western blotting. In histological analysis, specimens were stained with hematoxylin and eosin and stained with picrosirius red (PSR). To visualize the birefringent collagen in PSR staining, polarized microscopy was used. In western blotting, the tissues (20 mg wet weight) were homogenized and centrifuged. Total protein of ten microgram for each sample was resolved by 7% SDS-PAGE and transferred to nitrocellulose membrane using a semi-dry transfer cell apparatus. Rabbit polyclonal anti-collagen I and III antibodies were used. The bands were visualized using ECL detection system. For semi quantification of western blotting bands, the densities of specific antibodies were measured with Image J. Results: In picrosirius red staining under polarized microscopy, the red to yellow fibers (typical of type I collagen) and the green fibers (typical of type III collagen) were seen in all specimens. The EPL tissue showed irregular collagen organization with poorly defined structure. However the arrangement of the collagen visible in EIP specimen maintained normal tissue architecture, with regularly arranged collagen fibers. In western blotting, the level of collagen III were gradually decreased away from rupture side. The ratios of type III collagen to type I were gradually decreased away from rupture side. Conclusions: To our knowledge, only one study has examined the expression of collagen subtype using human tendon in the method of western blotting. Our report showed that the ratio of collagen III to collagen I were gradually decreased away from rupture side. This result may be helpful to find out the mechanisms of tendon rupture.

P36-1337 Substitution of patellar tendon by Achilles allograft in knee arthroplasty O. Ares1, J.M. Segur2, L. Lozano3, D. Popescu4, S. Sastre2, F. Macule2 1 Hospital Clinic de Barcelona, University of Barcelona, Orthopedic Department, Barcelona, Spain, 2Hospital Clinic University of Barcelona, Knee Unit, Barcelona, Spain, 3Hospital Clinic University of Barcelona, Knee Unit, Orthopaedic Surgery Department, Barcelona, Spain, 4Hospital Clinic i Provincial, Barcelona, Spain Objectives: Surgical treatment is required for acute or chronic injuries of the patellar tendon. These are a rare complication in knee arthroplasty, varying from 0.17 to 2.5% according to the literature. Several surgical techniques have been described in order to repair total functionality. The aim is to report the clinical outcomes of five cases where reconstruction of the extensor mechanism was performed with an Achilles allograft.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Methods: All patients had a deficit in active extension, whereas full passive extension was a prerequisite for this surgery. An anterior approach was made with eversion of the patella. We performed a notch at the tibial tuberosity to fix the calcaneus part of the allograft with two cannulated screws. The patella alta must be avoided in order to keep the patella in a correct position. The Achilles tendon is then divided into 2 bundles; the lateral bundle undergoes a transtendon quadricipital position whereas the medial bundle has a medial position. Both bundles are sutured to the soft tissues next to the patella and quadriceps tendon with Ethibond 5. Following surgery, knees are immobilized in extension for two months, and gradual rehabilitation is started getting a right balance, functionality and ability to walk in all cases. Results: All patients achieved near full extension after one-year follow-up. This technique avoids the problems with patella resurfacing or using patellofemoral arhtroplasties. Fixation with two screws helps with the arthroplasty even when a tibial stem was placed. Frozen Achilles tendon is useful for this technique. Conclusions: Deficit in extension is a complex problem of great concern that can be solved with the surgical technique that we present. Both our cases and the reviewed studies provide an overview of the several surgical possibilities and short-term follow-up. Despite the promising initial results, further monitoring should be done in order to reach a conclusion.

P36-1414 The effect of hydrogen peroxide treatment on human tendon allograft E. Gardner1, N. Vonder-Heide2, R. Fisher1, G. Brooker1, P. Yates2 1 Perth Bone and Tissue Bank, Perth, Australia, 2University Western Australia, Fremantle, Australia Objectives: Human tendon allograft is becoming popular for ligament reconstruction. Treatment of tendon allograft with hydrogen peroxide at the beginning of tissue processing may reduce bacterial graft infection. Bacterial growth was found on swab samples from retrieved cadaveric material after harvest in 24% of cases. The most common organism cultured was Staphylococcus epidermidis, which accounted for 62.5% of positive cultures. Side effects of hydrogen peroxide treatment include hydroysis of the collagen and this may alter the mechanical properties of the graft. Methods: We tested 38 cadaveric ankle tendons as matched pairs in a materials testing machine. All tendons are processed according to standard bone bank protocols including irradiation with 15 Gy. A freeze clamp technique with liquid nitrogen was used to secure the tendon. Tendons within the pairs were randomized to hydrogen peroxide treatment or not. Results: Hydrogen peroxide did not significantly decrease the strength of the tendons. Single strand Tibialis anterior and posterior tendons had a mean ultimate tensile strength of 2,337 Newtons. Conclusions: Hydrogen peroxide does not adversely affect the mechanical properties of human allograft tendons. A doubled hydrogen peroxide treated Tibialis allograft may be suitable for ACL reconstruction.

P36-1450 Postoperative functional rehabilitation after repair of quadriceps tendon ruptures: a comparison of two different protocols R. Langenhan1, M. Baumann2, B. Hohendorff3, A. Probst1, A. Badtke2, P. Trobisch4 1 Hegau-Bodensee-Klinikum Singen, Klinik fu¨r Unfall-, Handchirurgie und Orthopa¨die, Singen, Germany, 2 Berufsgenossenschaftliche Unfallklinik, Eberhard-Karls-Universita¨t, Tu¨bingen, Germany, 3Rho¨n Klinikum AG, Klinik fu¨r Handchirurgie,

S353 Bad Neustadt a. d. Saale, Germany, 4Vivantes Klinikum im Friedrichshain, Zentrum fu¨r Wirbelsa¨ulenerkrankungen, Berlin, Germany Objectives: The literature supports that early passive motion after surgical tendon repair facilitates the healing process and leads to improved joint function but most authors still recommend a prolonged period of immobilization and/or non-weight bearing after surgical repair of ruptured quadriceps tendons. The purpose of the present study is to compare the clinical outcome after a more aggressive and after a rather restrictive postoperative mobilization protocol. Methods: In a retrospective study all consecutive patients were included who had a surgical repair of a primary isolated unilateral quadriceps tendon rupture in any of the two participating hospitals and a minimum follow-up of 24 months. While the surgical procedure does not differ between the hospitals, different postoperative mobilization protocols exist. Patients of institution A were only allowed limited flexion and weight bearing while patients from institution B were allowed early functional mobilization with full weight bearing. Clinical outcome was measured with the subjective IKDC form. Fisher’s exact test and student’s t test on SPSS were used for statistical analysis. Significance was defined at p \ 0.05. Results: 66 patients were included in the study. 28 patients (Group A) were treated with a restrictive and 38 patients (Group B) with an early functional postoperative mobilization protocol. The two groups did not differ in terms of demographic characteristics. Clinical follow-up was available for 95% of patients after an average of 4.5 years. No clinical difference was identified with the use of IKDC form. Patients of group A returned to work an average of 10 days later than patients from group B but this difference was not significant. 2 re-ruptures were observed in each group. There was no significant difference in terms of complication quality or quantity. Conclusions: Early functional postoperative mobilization with full weight bearing after primary repair of a quadriceps tendon rupture is safe and will not lead to inferior clinical outcome or an increased complication rate.

Sport specific injuries

P37-188 Ulnar nerve injuries at avulsion fractures internal epicondyle of humerus at young sportsmen (treatment & prognosis) V. Kuksov1 1 Second City Hospital, Traumatological, Samara, Russian Federation Objectives: One of the abundant complications of avulsion fractures internal epicondyle of humerus at young sportsmen is the primary ulnar nerve injuries. Therefore the neurologic examination is obligatory for children with such injuries. Methods: During lust 7 years we were providing the operative treatment at 120 young sportsmen (the age was from 7 till 14 years old) with avulsion fractures internal epicondyle of humerus. At 15 of them on reason of clinical-rontgenological data and neurologic examination were the plural camps of the blocked internal epicondyle and ulnar nerve at the cavity of elbow (typical picture of the ulnar neuritis). For this group of the patients operative treatment was absolutely shown at early data (first three days). Surgical access to elbow was internal lateral. During operative intervention was careful realizing and drawing aside on rubber holder medially strangulated ulnar nerve, and then was extracting and strangulated epicondyle part. We were providing the comparison of the part and fixing it to maternal place by 2 Kirshner wires. Only at 3 patients the ulnar nerve was laying on the same place. For rest 12 patients were providing the

123

S354 additional operation the front transposition of the ulnar nerve. Plaster immobilization was at 14–18 days. After recovering from anesthetization all 15 patients were marking the positive neurologic symptomatology: was disappeared the sense of torpor on the way of innervations ulnar nerve, the pain senses were greatly reducing. To all these patients was providing the complex rehabilitation treatment: magneto therapy; the dibazol with prozerin (by plan of child’s dosage), light massage of hand fingers; then electrophoresis with glutamine acidity, vitamins complex B (by plan); LFC, hydro therapy. The events of ulnar neuritis were in full disappeared through 1, 5 months after trauma. Results: All 15 patients were observing though 2 years after operative treatment. At all anatomic and functional indices are excellent; configuration condyle of humerus is right form, true union internal epicondyle with maternal place. Senses and moving deviations from norm from the way of ulnar nerve are not. Sport prognosis is favorable, successfully are engaging to sports. Conclusions: Provided by us medical tactics with traumatic ulnar nerve injuries was adequate, about it is evidencing favorable results.

P37-199 ‘Popliteus sling’ reconstruction for posterolateral rotatory instability of knee: a prospective study P. Kodkani1 1 Arthroscopy & Sports Medicine Institute, Mumbai, India Objectives: Injuries to the posterolateral complex (PLC) are common in a multiligament injured knee. These injuries may also occur in isolation. Injury to the PLC results in laxity with 2 components—The varus rotation and the posterolateral rotational laxity. Depending on the structures injured it could therefore result in a combined laxity or an isolated varus or rotational laxity. Posterolateral rotational laxity is primarily due to injury to the popliteus or the popliteofibular ligament. This rotational laxity may be symptomatic even when it occurs in isolation. This posterolateral rotational instability was managed by an anatomical popliteus sling reconstruction procedure and results assessed. Methods: 12 knees with posterolateral rotatory instability were treated with popliteus sling reconstruction in the past 3 years and studied prospectively. 2 knees presented with associated injury to the anterior cruciate with posterior cruciate ligament (PCL). 4 knees presented with associated anterior cruciate injury. 1 associated with PCL injury. 5 cases presented with isolated posterolateral rotatory instability (PLRI). All cases presented between 3 to 15 months following the primary injury. The diagnosis of PLRI was based on clinical examination. Anatomical reconstruction was done for the popliteus sling. Double stranded autologous gracilis was used as graft for reconstruction. In the first 6 cases popliteus reconstruction was done through an open technique using a lateral hockey stick incision. In the final 6 cases an all arthroscopic technique was used for the reconstruction. All associated cruciate reconstructions were performed arthroscopically at the same stage. Results: 2 patients were lost to followup after 1 year. The average follow up was for 18 months. There were no complications encountered in these cases. All patients achieved full range of motion except for 2 cases which had tightness in terminal 5 flexion as compared to opposite normal knee. The patient satisfaction score was better in the arthroscopic group as compared to the open reconstruction group. Instrumented quantitative assessment of the rotations could not be made. Posterolateral rotational laxity was compared to the preoperative laxity and to the opposite normal knee in all cases by dial test. All patients achieved rotations compared by dial test equal to the opposite normal knee. IKDC scoring showed improvement to normal or near normal activity levels in all these cases. KOOS score improved to 100 (Normal) in all parameters (Table 1).

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Table 1 Open

Arthroscopic

Total

Popliteus sling

1

4

5

Popliteus sling + ACL

3

1

4

Popliteus sling + PCL

–

1

1

Popliteus sling + ACL + PCL

2

–

2

Total

6

6

12

Conclusions: It is important to address the posterolateral rotatory instability of the knee to achieve normal functioning. The popliteus sling reconstruction offers a good surgical modality of treatment to address this issue. Anatomical popliteus sling reconstruction gives good results. Arthroscopic popliteus sling reconstruction has better patient satisfaction. It is minimally invasive, safe, effective and reliable procedure for treatment of posterolateral rotatory instability of the knee.

P37-300 Experimental research of magnesium alloys resorption of in vitro M. Golovakha1, V. Cherny2, E. Yatsun3 1 Zaporozhje State Medical University, Orthopedics and Sport Trauma Faculty, Zaporozhye, Ukraine, 2Zaporozhye State Medical University, Zaporozhye, Ukraine, 3Zaporozhje State Medical University, Orthopedics & Trauma, Zaporozhye, Ukraine Objectives: One of the most current method of treatment of fractures still have an osteosynthesis with implants made of stainless steel and titanium. However after fracture consolidation repeat surgery intervention for removal of implants is necessary. So do not stop searching the implants, which could metabolized by the body. Polymers became the first commercial biodecomposed and bioabsorbed materials for implantation. More often following biopolymers are applied: «polylactic acid» (PLA) and «poly-dioxanone» (PDS). These materials have insufficient mechanical possibilities for their use at an osteosynthesis and are replaced by scar, and no bone tissue. Methods: Therefore the attention of researchers is involved with magnesium alloys, is completely dissolved in an organism, not causing an intoxication. Mechanical properties of magnesias alloys above, than at polymers, are caused by interest and elasticity and osteoinductive effect. Bone as living tissue, constantly adapts to stress. But if there is an implant, this process can lead to stress protection (fracture). Initial cortical bone elasticity in the range 13–20 MPa. Modulus of elasticity of steel for flawless 200 MPa for the titanium alloy—115 MPa. Magnesium alloys have a modulus of elasticity of about 45 M, which more closely matches the modulus of elasticity the bone. Results: The purpose of the study—explore experimentally the properties of the alloy bioresorptive system Mg-Zr-Nd-modified Ag, in an environment identical to the physiological environments of the human body. To control the mechanical properties, corrosion tests and metallographic analysis was made of cast samples of alloy additives Ag. This has increased the elastic modulus of the material. Corrosion tests were performed on samples Ø10*5 mm. They were weighed, placed in test tubes with 0.9% aqueous NaCl solution in a thermostat at 38 C, 10, 20 and 30 days, removed, re-weighed and counted the loss of mass. Conclusions: Modification of Mg-Zr-Nd alloy doped Ag, the optimal amount possible to achieve a rate of biodegradation experiment, comparable with an average speed of consolidation of bone fractures, and increased his strength and plastic characteristics. This allows us to

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 consider the use of this alloy for production various metals used in fixation.

P37-475 Wheel gymnastic injuries in amateurs and professionals M.D. Kauther1, S. Rummel1, B. Hussmann1, M. Burggraf1, S. Lendemans1, C. Wedemeyer2 1 University of Duisburg/Essen, Department of Trauma Surgery, Essen, Germany, 2University of Duisburg/Essen, Department of Orthopaedics, Essen, Germany Objectives: Wheel gymnastics is an acrobatic sport with a history of nearly 100 years. It is known from gymnastic competitions as well as from circus shows. So far, there have been no studies or case reports about injuries in wheel gymnastics. Methods: This retrospective epidemiological study of 502 wheel gymnasts surveyed 51 professionals, 100 semi-professionals and 352 amateurs by questionnaire. Qualitative variables were analysed by Chi-square test and quantitative variables using the Whitney U test. The correlations of qualitative and quantitative variables were analysed with the test from Spearman and Pearson. Results: 7,925 injuries and 422 overuse syndromes were found in 988,718.9 h of training leading to a loss of 17,310.3 training days. Professionals reported significantly (p \ 0.001) more injuries and overuse syndromes with significantly more injuries of the wrist (p \ 0.001), knee (p \ 0.001), hip/thigh (p = 0.003), ankle/foot (p = 0.013) and elbow (p = 0.033). Significant differences were found in the time per injury rate. Amateurs practised significantly less hours than semi-professionals and professionals. The most frequent pain was found in the region of the wrist, spine, shoulder and ankle. A positive correlation significant at the 0.0001 level was found between the average pain frequency and the average frequency of wearing protective gear of the elbow, wrist, knee and ankle. Conclusions: Professional wheel gymnasts suffer from fewer injuries per time than amateurs. Protective gear is correlated with pain, but no negative correlation of protective gear and injuries could be found. Physicians should be aware of the common risks in this highly acrobatic kind of sport. The severity of injuries in wheel gymnastic should not be underestimated.

P37-606 Peroneal nerve palsy due to compression by a neural cyst: recovery after nerve preserving resection and tibio-fibular joint arthrodesis K. Piasecki1, C. Lo¨cherbach1 1 Centre Hospitalier Universitaire Vaudois (CHUV), Service d’orthope´die et de traumatologie, Lausanne, Switzerland Objectives: The purpose is to present a case with nerve palsy caused by the rare entity of an intraneural ganglion cyst of the common peroneal nerve (CPN). The rate of recurrence is high after simple excision as unique surgical procedure. Following the unified articular theory, additional fusion of the proximal tibio-fibular joint (PTFJ) is evocated as a reliable treatment without recurrence reported in the literature. Methods: A 53 year-old woman presented complaining about pain and dysaesthesia in the territory of the CPN and a progressive footdrop with a stepping gait. Repeated lumbar MRI scans did not show pathologic findings. Appearance of a tender swelling in the region of the proximal head of the fibula motivated an electromyographic (EMG) examination revealing atypical signs of chronic compression of the CPN. Magnetic resonance imaging (MRI) shows an intraneural cystic formation of the CPN with clearly identified communication with the PTFJ. The surgical procedure consisted in excision of the

S355 cyst preserving the different branches of the CPN, followed by the fusion of the PTFJ. Results: At 1 year follow up a full recovery can be stated, with continuity of the nerve conductivity and a regular and complete motor activity. Conclusions: Intraneural ganglion cysts causing a peripheral nerve compression syndrome of the CPN should be considered as differential diagnosis for peroneal nerve palsy with foot drop. EMG and MRI scan allows diagnosis of this disorder. Taking into consideration recent reports in the literature and the evidence of articular origine of the cyst on MRI, simultaneous excision of the cyst and PTFJ fusion was chosen to minimize the risk of recurrence.

P37-614 Arthroscopic treatment of jumper’s knee in professional athletes F. Pascarella1, M.G. Di Salvatore1, C. Latte1, A. Pascarella1 1 Laboratorio di Chirurgia Artroscopica, Isernia, Italy Objectives: This study analyses the outcome of 27 professional athletes with patellar tendinopathy undergoing arthroscopic surgery after failing conservative treatment. Methods: 27 professional athletes with patellar tendinopathy refractory to conservative management underwent arthroscopic debridement of abnormal patellar tendon and excision of the lower pole of the patella. Pre- and post-operative evaluation was undertaken using the IKDC score, Lysholm knee scale and VISA-P scores for all patients at 1 and 3 years. 43 and 29 patients were similarly assessed at 5 and 10 years, respectively, after surgery. No patients were lost to follow-up. The return to sports and rehabilitation was also assessed. Results: The IKDC, Lysholm and VISA-P scores were all significantly improved at 1 and 3 years follow-up. These scores remained significantly better for the patients assessed at 5 and 10 years followup. 19 of the 27 professional athletes returned to sports at the same level, with a further 5 playing at a lower level. These 24 patients were symptom-free. The other 3 athletes returned to sport at a lower level without symptoms for 3 years after the index operation, but then developed symptoms compatible with a Blazina stage I patellar tendinopathy. All patients were able to return to sports by 3 months. Conclusions: Arthroscopic surgery for patients with patellar tendinopathy refractory to conservative management appears to provide early return to sports with significant improvements in symptoms and function, with improvements maintained for at least 3 years up to 10 years.

P37-721 Magnetic resonance imaging of asymptomatic knees in collegiate basketball players: the effect of 1 season of play M. Safran1, G. Pappas2, M. Vogelsong3, E. Staroswiecki4, G. Gold3 1 Stanford University, Orthopaedic Surgery, Redwood City, United States, 2South Carolina Sports Medicine, North Charleston, United States, 3Stanford University, Radiology, Stanford, United States, 4 Stanford University, Stanford, United States Objectives: MRI is a sensitive tool for the diagnosis of patients, including athletes. Some limited data exists regarding the prevalence of asymptomatic findings on the MRI of knees in collegiate and professional basketball players during the pre-season evaluation. The goal of this study is to determine the prevalence of abnormal structural findings using high resolution, 3.0 T MRI, in the asymptomatic knees of male and female collegiate basketball players before and after a season of Division I collegiate basketball. Methods: After IRB approval, 24 previously asymptomatic knees of 24 collegiate Division I basketball players (age 18–22; 12 male, 12 female) were imaged using a 3.0T MRI scanner prior to and within

123

S356 1 month following the end of their 5 month competitive season. Three players (2 men, 1 woman) did not undergo scanning after the season. Images were evaluated for articular cartilage health, bone marrow edema, patellar and quadriceps tendinopathy, and ligament and meniscal injury. Results: Seventeen (71%) of 24 and 17 (81%) of 21 knees showed chondral lesions pre- and post-season, respectively. Of those, 15 (88% of affected knees) involved the patellofemoral cartilage both pre- and post-season. Bone marrow edema was seen in 18 (75%) knees pre-season and 18 (86%) post-season, with patellofemoral joint involvement in 14 (78% of affected) and 16 (86% of affected) knees. Pre- and post-season prevalence rates for patellar tendinopathy were 20 (83%) and 19 (90%), for quadriceps tendinopathy were 18 (75%) and 19 (90%), and for intra-meniscal signal changes were 10 (42%) and 11 (52%). There were no ligament or meniscal tears, though there were 2 pre-existing asymptomatic meniscal capsular injuries and one subject who experienced an acute chondral injury during the season. There were no significant differences between males and females when the chondral injury was excluded. The only significant change from pre- to post-season was seen as an increase in mean meniscal score (0.62 ± 0.74 to 0.86 ± 0.79; p = 0.025). Conclusions: These results show a higher prevalence of structural knee abnormalities in a young population of asymptomatic athletes than previously reported. The higher strength 3.0T magnet used may facilitate improved detection of lesions. High intensity basketball may have particularly damaging effects on the patellofemoral joint. These findings in asymptomatic athletes should caution the clinician to consider clinical findings, rather than imaging alone, when evaluating athletes.

P37-756 Gender based differences in outcomes following ACL reconstruction in soccer athletes from MOON cohort R. Brophy1, L. Schmitz2, R.W. Wright1, K. Spindler3, MOON Group 1 Washington University School of Medicine, Department of Orthopaedic Surgery, Chesterfield, United States, 2Cleveland Clinic, Cleveland, United States, 3Vanderbilt University Medical Center, Orthopaedics/Sports Medicine, Nashville, Tennessee, United States Objectives: ACL injuries are common among soccer athletes, and occur at a higher rate in females compared to males. There is limited information on the outcomes of ACL reconstruction specifically regarding return to participation in soccer athletes. The purpose of this study was to test the hypothesis that there are gender differences in return to play and risk for future ACL injury in soccer athletes. Methods: Subjects enrolled in the prospective MOON cohort, who were identified as soccer athletes, were contacted to review their return to play following ACL reconstruction. Information regarding if and when they returned to play and their current playing status was recorded. If they were not currently playing, they were asked for the primary reason they stopped playing soccer. Any additional ACL surgery following their initial ACL reconstruction was also recorded. Results: Data was collected on a total of 94 soccer athletes (49 male, 45 female) with 81% follow-up in this cohort. Overall 72% of soccer athletes returned to soccer at an average of 12.5 months after surgery. There was no significant difference between males and females in the rate of return (male 80% vs. female 64%, p = 0.1) or time to return to play (males 10.5 ± 7.5 months vs. females 15.1 ± 20.7 months, p = 0.3). Twelve soccer athletes (13%) had undergone further ACL surgery, including 9 on the contralateral knee (10%) and 3 (3%) on the ipsilateral knee. Females were more likely to need additional ACL surgery (27%) than males (6%) (p = 0.04). At an average follow up of 7.0 years, only 36% of soccer athletes who underwent ACL reconstruction are still playing the sport, a significant decrease

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 compared to initial return to play (p \ 0.0001). There was no significant difference in the long term return to play between males (41%) and females (31%) (p = 0.3). However, men were more likely (59%) than women (29%) to attribute their ACL injury as the primary reason they were no longer playing soccer (p = 0.02). Conclusions: One of the goals of ACL reconstruction is to allow the patient to return to desired activities, including participation in athletic activities. Soccer players have a good initial rate of return to play following ACL reconstruction which declines over time. Males may be affected more directly than females by the injury itself as opposed to other factors or life events in terms of their potential return to play. Future research should identify factors that limit return to play initially and over time in these athletes. Female soccer players who undergo ACL reconstruction are at a higher risk of future ACL injury than males and may benefit from targeted injury prevention programs.

P37-809 Early results of management and outcome of open reduction and internal fixation of acute large osteochondral fractures of the patello-femoral joint using bio-absorbable pins K. Veravalli1, A. Davies2 1 Morriston Hospital, Trauma and Orthopaedics, Morriston, United Kingdom, 2Morriston Hospital, Department of Orthopaedics, Swansea, United Kingdom Objectives: The severity of large osteochondral fractures involving the patello-femoral joint is often underestimated by clinicians. Management of these fractures often involves the removal of fragments, leaving a large area of articular surface to be filled with fibrocartilage. The literature review shows emphasis on studies related to osteochondral fracture of the weight bearing area of the knee with very little data-reporting on the fractures involving the patellofemoral joint. We present our experience and results in treating osteochondral fractures of the patellofemoral joint using poly-p-dioxanone (bio absorbable) pins (Orthosorb, DePuy). Methods: Five patients, between the ages of fourteen and sixteen, with a large (4 cm2) osteochondral fracture involving one of the articular surfaces of the patellofemoral joint of the knee were treated with open reduction and internal fixation with use of poly-p-dioxanone (bio absorbable) pins and an adjunct procedure, as required, to improve the stability of the joint. Each patient had a minimum follow up of at least 1 year after the index procedure with a clinical assessment and documentation of the Tegner activity scale. At 6 months and 1 year following the index procedure each patient had a plain radiograph of the knee including skyline view and magnetic resonance imaging scans were acquired. Results: All five patients who underwent open reduction and internal fixation have returned to their previous level of sporting activity, which is reflected in their Tegner activity scale. Range of movement in the knee joint was full with no further episodes of patellofemoral joint dislocation or instability. Magnetic resonance imaging scans of all patients showed the fragment was well taken with no step in the articular surface. Conclusions: Our study shows that open reduction and internal fixation of osteochondral fractures, involving the articular surface of the patellofemoral joint of the knee using poly-p-dioxanone (bio absorbable) pins, is a relatively simple and worthwhile option.

P37-810 Patella tendonitis in the elite professional footballer N. Jain1, S. Kemp2, D. Murray1 1 North West Deanery, Trauma and Orthopaedics, Manchester, United Kingdom, 2Wolverhampton Wanderers Football Club, Physiotherapy Department of Wolverhampton, United Kingdom

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Objectives: Patella tendonitis is common amongst sportsmen. It has an increased incidence in sports that involve specific types of loading. It is suggested that volleyball players have the highest rate of this pathology. There is no published evidence of this injury in elite professional footballers. Therefore the aim of this study was to determine the frequency of this injury in the elite professional footballer, along with the impact of such an injury on the amount of time missed and the outcomes of various treatment options. Methods: Data was collected prospectively for injuries suffered by first team players, development squad players and academy squad players over the 2009–2010 and 2010–2011 English Premier League (EPL) season at one EPL club. Each player’s demographics were recorded. The injury was recorded along with the time that the player was absent because of the injury, the treatment that the player received and whether they suffered any recurrence of the injury. Results: Players sustained a total of 35 knee injuries during the period of this study. This was 21% of all injuries suffered. Ten of these were patella tendonopathies. This provided a frequency of 28.6% of knee injuries and 6% of all injuries suffered were patella tendonitis. Of these 8 were sustained by first team players, which was 53.3% of knee injuries sustained by this squad. Two players within the academy squad suffered such an injury, which was 14.3% of knee injuries within that squad. There were no patella tendonitis diagnoses within the development squad. The first team players had a mean absence of 67 days and the academy players 120 days. Both the academy players underwent surgery as a primary treatment (100%) whereas no first team player had surgery (0%). Four first team players were treated with local injection, 2 with sclerosant and 2 with Platelet Rich Plasma (PRP). The mean absence for these was 110 days. The remaining 4 injuries were treated with physiotherapy with a mean absence of 24 days. There was recurrence in 4 cases, all within the first team (rate of 50%). There were 2 in each of the physiotherapy and injection groups. No recurrence was observed within either of the academy players that underwent surgery. Conclusions: Patella tendonitis is a common knee injury suffered by elite professional footballers. This is a significant injury as there is a prolonged amount of time that a player is unavailable for selection following this injury. There is also a high recurrence rate with this injury although the best treatment option remains debateable.

P37-951 Knee injuries in elite professional footballers: the findings of one English premier league team D.J. Murray1, N.P. Jain1, S. Kemp2 1 North West Deanery, Trauma and Orthopaedics, Manchester, United Kingdom, 2Wolverhampton Wanderers Football Club, Physiotherapy Dept, Wolverhampton, United Kingdom Objectives: Knee injuries are a common occurrence in footballers. I the modern game of association football, the demand on elite players is increasing, and the cost of injury to player and club can be catastrophic. Footballing injuries to the knee range from minor to career threatening, and can lead to prolonged absences from play. Treatment of these injuries should aim to return a player back to competitive games in a safe and timely manner. The aim of this study was to establish the frequency and variation of knee injuries within one English Premier League (EPL) professional football club over the course of two seasons, to assess time taken to return to play, and analyse current treatment regimen for each injury type. Methods: Data was collected prospectively for all knee injuries suffered by professional players between 2009 and 2011, spanning two EPL seasons at one EPL club. Each players demographics were recorded along with various factors concerning or influencing the injury, including playing surface, pitch condition, dominant side, type and side of injury, ability to continue playing, mechanism of injury

S357 and type of footwear. Time taken to return to play, and treatment received following each injury type was recorded. Results: 35 seperate injuries occurred over the course of this study. The commonest injury was to the medial collateral ligament (MCL) in 34%. Patella tendon injuries were seen in 29%, other inury types included meniscus tears, ACL ruptures, and osteochondral defects. All grade II MCL injuries received sclerosant injections. 40% of patella tendon injuries received plasma-rich protein (PRP) injections, and 30% underwent surgery. The mean recovery time following MCL and patella tendon injuries was 44 and 77 days respectively. 60% of all injuries were sustained during training sessions and 40% were suffered in a competitive game. 26% were recurring injuries, recurrent meniscus and patella tendon injuries took twice as long to recover compared to the initial injury. Conclusions: Our findings suggest that MCL and patella tendon injuries are the most common knee injuries amongst professional footballers, and meniscus tears and ligament ruptures are relatively rare in comparison. Injuries appear to occur more frequently during training. A high proportion of injuries received injection therapy in the form of prolotherapy or PRP. The study suggests recurrent injuries can prolong recovery two-fold.

P37-1035 Standardising the assessment of collateral knee ligament laxity W.T. Wilson1, A.H. Deakin1, F. Picard1, P.E. Riches2, J.V. Clarke1 1 Golden Jubilee National Hospital, Department of Orthopaedic Surgery, Glasgow, United Kingdom, 2University of Strathclyde, Department of Bioengineering, Glasgow, United Kingdom Objectives: The assessment of coronal knee laxity is an important clinical manoeuvre for evaluating injuries to the collateral ligaments. Current routine methods are highly subjective with respect to examination technique, magnitude of clinician-applied load and assessment of joint displacement. Scoring systems to grade severity are often based on millimetres of perceived joint opening with applied manual stress, so are highly reliant on clinician judgement. This study aimed to overcome the subjective nature of routine assessment and develop a repeatable, objective stress test for incorporation into standard clinical practice. Methods: Eighteen clinicians were instructed to measure the collateral laxity of the right knee of a healthy volunteer. The noninvasive adaptation of the Orthopilot image-free navigation system enabled real-time measurement of coronal and sagittal mechanical femorotibial angles, and has been validated to an accuracy of ± 1. Collateral knee laxity was defined as the amount of angular displacement during a stress manoeuvre. Knee flexion angle and hand positioning were kept constant during testing. A manual varus and valgus stress was applied up to a perceived endpoint and the maximum angular deviation recorded. A hand-held force application device (FAD) was then used to standardise the applied load. Participants were instructed to repeat the test using the device in the palm of their right hand and to apply the force until the visual display and an auditory alarm indicated that the target had been reached. Results: Clinicians produced a narrow range of laxity measurements demonstrating the successful quantification and standardisation of the technique of knee laxity assessment. Assessment of valgus laxity was more consistent than varus (standard deviation 0.6 vs. 1.3). Mean varus and valgus laxity (degrees ± SD) Varus

Valgus

Initial stress test

5.4 (±1.3)

1.6 (±0.6)

Using FAD

5.3 (±1.2)

2.4 (±0.6)

123

S358 When using the FAD, the standard deviations remained low for both varus and valgus laxity. Introducing the FAD overall produced a significantly greater angulation in valgus (2.4 compared to 1.6, p \ 0.001) but not varus (p = 0.67) when compared to the initial examination, indicating that usually a lower force is applied during valgus stress testing than varus. This was re-enforced by clinicians, one-third of whom commented that they felt they had to push harder for valgus than varus, despite the FAD target being the same. Conclusions: We have successfully standardised the manual technique of coronal knee laxity assessment by controlling the subjective variables. The incorporation of a FAD into assessment did not affect the clinicians’ ability to produce reliable and repeatable measurements, despite removing the manual perception of laxity. The FAD also provided additional information about the actual moment applied. This quantitative information may have a role in improving the management of collateral ligament injuries with regard to initial diagnosis and grading as well as a guide to rehabilitation.

P37-1119 Proximal avulsion of rectus femoris muscle: presentation, treatment and outcome in 19 patients T.E. Fayad1, F.S. Haddad1, S.S. Sarmah1 1 University College London, Orthopaedics, London, United Kingdom Objectives: Proximal avulsion of rectus femoris muscle is an uncommon injury. It is characterised by the avulsion of the anterior proximal tendon (short head) from the anterior inferior iliac spine and/or the posterior proximal tendon (reflected head) from the groove above the brim of the acetabulum. The aim of this study was to retrospectively review the clinical presentation, radiological investigation, treatment and outcome of patients presenting with this injury between 2007 and 2011. Methods: 19 patients were included in this study and subjects were classified as being treated conservatively or surgically. 17 of those were professional footballers, rugby players or runners. 2 had sports related injuries. MR imaging studies, radiology reports, medical records, and operative notes were reviewed including the type and mechanism of injury, patients’ profession, method of treatment, and time to return to activity. Results: A total of 8 patients required surgical intervention whereas 11 patients were treated non-operatively. MR imaging studies showed 12 patients with avulsion of the short head compared to 5 patients with avulsion of the reflected head and 2 patients with injury to the conjoint tendon. Conclusions: Injuries of the origin of the rectus femoris muscle are rare. Surgical intervention was favored towards kickers with injury affecting their dominant leg. Both conservative and surgical treatment resulted in a satisfactory outcome with a mean recovery time of 6–12 weeks.

P37-1136 Occult lateral tibia plateau fractures in low energy knee injury S. Abrassart1, D. Fritschy2, I. Uckay1, P. Hoffmeyer3, J. Menetrey4 1 HUG, Orthopedie, Geneve, Switzerland, 2Hopitaux Universitaires de Geneve, Department de Chirurgie, Geneve, Switzerland, 3HUG, Orthope´die-Traumatologie, Gene`ve, Switzerland, 4University Hospital of Geneva, Department of Orthopaedic Surgery, Geneve, Switzerland Objectives: Our investigation looked at missed fracture in acute knee injuries. Fractures may be associated to anterior cruciate ligament rupture or to patella dislocations and to a direct trauma on athletic

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 knees. Occult fractures lead to persistent pain and may be a source of conflict with the patient, especially with athletes. Methods: We reviewed 16 patients (age range 24–70 years, mean 44), 7 women and 8 men, with low energy knee injury initially seen in our emergency room. All associated lesions or diseases were excluded. Clinical examination and plain radiographs have been routinely performed in the emergency unit, all injuries were considered as benign and all patients were sent home with symptomatic treatment. All patients were reviewed one week after the initial trauma for persistent pain and they underwent an MRI. Treatment including toe-touch weight-bearing, splint, and anti-coagulation was prescribed for 6 weeks. Physiotherapy with mobilization and quadriceps muscle strengthening was immediately performed. All patients were followed-up to 6 months after the accident. Results: Patients were complaining of persistent lateral and/or medial pain. No laxity and no ACL lesion were noted in those patients. All plain radiographs were normal at admission except two that showed intra-articular effusion. Mechanisms of injury were a fall (5 cases), a ski trauma with rotation (3), a motor cycle trauma (3), hazardous jump (1), hyperextension in running (1 case), rotational movement in running (2) and direct impact (1). MRI showed 12 lateral and 4 medial tibial plateau fractures, one medial meniscal tear, two medial meniscosis, three medial collateral ligament lesions, and three bone bruises on the medial femoral condyle. There is a good correlation between mechanism of rotational injury (6 cases) and meniscal pain (10 cases) and the existence of a tibial fracture (Fisher exact with p = 0.011). One week post-trauma, the clinical examination revealed an articular effusion in three patients, positive meniscal maneuvers in 10 patients. Interestingly, lateral fracture provoked medial pain in 4 patients. At 6 months FU, all patients healed evenly with no residual pain, full mobility and return to sport activity. Conclusions: This case series shows the importance of a systematic clinical evaluation one week after a knee trauma, even after a low energy injury mechanism. Indeed, the only predictive factor is the persistence of a localized and sustained pain on one compartment of the knee. Presence/absence of an effusion, positive meniscal maneuvers, and decrease ROM are no discriminative factors. MRI is the investigation of choice.

P37-1143 Systematic video analysis of ACL injury mechanisms in soccer: a gender comparison R. Brophy1, J. Stepan2, H.J. Silvers3, B. Mandelbaum4 1 Washington University School of Medicine, Department of Orthopaedic Surgery, Chesterfield, United States, 2Washington University School of Medicine, Chesterfield, United States, 3Santa Monica Orthopedic Group, Santa Monica, United States, 4 Santa Monica Orthopaedic and Sports Med Research Foundation, Santa Monica, United States Objectives: Soccer athletes, particularly females, are at risk for ACL injury. There have been few studies to date on the mechanisms of anterior cruciate ligament injuries in soccer athletes and no gender based comparison of these mechanisms. The purpose of this video based study was to qualitatively describe the mechanisms of anterior cruciate ligament injury in soccer athletes and to compare injury patterns in female soccer players with those in male soccer players. Methods: Thirty-nine videos of ACL injuries in 21 male and 18 female soccer players at the collegiate and professional level were reviewed. Visual analysis of each case was performed to describe the injury mechanisms in detail (game situation, player behavior, biomechanical characteristics).

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: The majority of ACL injuries occurred when the opposing team had the ball and the injured athlete was defending (77%) with no significant difference between males (83%) and females (71%). The most common playing action was tackling (51%), followed by cutting (15%) and kicking (10%). Just over half of injuries occurred due to a contact mechanism (56%) and there was no significant difference between males (52%) and females (61%). Noncontact ACL injuries usually occurred with the knee in valgus (77%) and within 30 of full extension (70%), and the hip flexed (88%) and abducted (77%). Noncontact injury involved the left knee in most of the females (5/7) but not males (4/10). Conclusions: Soccer players are at greatest risk for ACL injury when defending, especially when tackling the opponent in an attempt to win the ball. Injury prevention programs should focus on this technique in soccer athletes. Non-contact injuries in soccer athletes usually occur in an at-risk position with the hip flexed and abducted and the knee in valgus and at or close to full extension. Females may be at greater risk for non-contact injury in their left lower extremity.

P37-1160 Studies of the myotendinous junction in humans by histology and transmission electron microscopy indicate that this structure undergoes constant remodelling A. Knudsen1, K. Kjær-Hansen1, A. Mackey-Sennels2, K. Qvortrup3, M. Kjaer2, M. Krogsgaard1 1 Bispebjerg University Hospital, Section for Sportstraumatology and Arthroscopy, Copenhagen, Denmark, 2University of Copenhagen, Institute of Sports Medicine, Bispebjerg Hospital, Copenhagen, Denmark, 3Panum Institute, Copenhagen University, Department of Biomedical Sciences, CFIM, Copenhagen, Denmark Objectives: The myotendinous junction (MTJ) transmits force from muscle to tendon, and the junction between these two tissues is structured with numerous interdigitations and invaginations. The MTJ is a site for painful overuse injuries in humans. The details of the micro- and macroscopic structure of MTJ in humans are widely unknown, as well as the MTJ’s reaction to exercise and immobilisation. Aim : Describe the MTJ in humans. Methods: 14 healthy human subjects (age 25 ± 3 years) with an isolated injury of the anterior cruciate ligament, were scheduled for reconstruction with a semitendinosus/gracilis graft. The MTJ was isolated from the grafts and divided in 3 pieces, which subsequently were embedded in mounting medium and frozen in liquid nitrogen cooled isopenthane. Serial transverse sections were cut in 10 um slices using a myotone at -20 Celsius and stained for neural cell adhesion molecule (NCAM) (also staining satellite cells), Tenascin C, laminin, fibronectin, collagens, intregrins, dystroglycans and fibertypes. Samples were also analyzed by transmission electron microscopy (TEM). Results: All stainings except for integrins were successful in all patients. Type I collagen was the main component of tendon, except at MTJ, where collagen type III was dominant. NCAM was only found in muscle fibres close to MTJ, indicating that MTJ is undergoing constant remodelling—also indicated by a high density of satellite cells close to MTJ. Interdigitations between muscle and tendon were observed by TEM. A high density of small electrondense deposits were identified at the junction between muscle and tendon, which could be cadherins, that are important for cell adhesion in MTJ. Conclusions: The basic structure of the MTJ in humans and in rats is the same. This study indicates, that MTJ is undergoing a constant remodelling in humans and thereby has a high adaptive potential.

S359 P37-1216 Pre-patellar friction syndrome: a new clinical entity causing anterior knee pain in high-level cyclists S. Claes1, T. Claes1, J. Bellemans1, T. Claes2 1 University Hospitals Leuven, Department of Orthopaedic Surgery and Traumatology, Pellenberg, Belgium, 2St. Elisabeth Hospital, Herentals, Department of Orthopedic Surgery, Herentals, Belgium Objectives: Anterior knee pain is extremely common among professional cyclists, annually affecting more than a third of these athletes. The complaints are thought to arise from overload of the patellar cartilage, although little is known about the exact etiology. We describe a new clinical entity of anterior knee pain in elite cyclists caused by friction of one of the defined pre-patellar soft-tissue layers on the patellar surface or edges. Methods: A patient database search was performed to identify all elite cyclists diagnosed with pre-patellar friction syndrome (PPFS). Patient records were retrospectively reviewed with specific attention for the typical history and clinical findings, diagnostic tests, treatment modalities and results. Additional cadaveric dissections were performed in order to study the triple-layer soft-tissue anatomy anterior to the human patella and to relate the newly described PPFS to its specific anatomical substrate. Results: A case series of 40 elite cyclists is presented. All patients were male (semi-) professional athletes, 14 contested in the UCI ProTour/Cyclocross World Cup races, 25 participated in the UCI Continental Tour (Europe) and 1 was a recreational triathlon athlete. A trauma tic onset was reported by 11 athletes, while the complaints were triggered by chronic overuse in the remaining 29 subjects. In all cases, the diagnosis was made by clinical examination in combination with the typical history: anterior knee complaints solely provoked during cycling activities, with localized tenderness at palpation of the soft-tissues anterior to the patella or at its edges. Typically, pain was aggravated by increasing pedal stroke frequency and by standing pedaling (riding ‘‘en danseuse’’). Conservative treatment by adjusting saddle height or position, adjusting crank length or corticoid infiltrations showed poor results. Eventually, all patients were treated operatively with partial resection of the damaged and inflamed prepatellar layer using a mini-incision. At a mean of 8, 3 weeks after surgery, all subjects returned to their pre-injury level. By correlating anatomical findings obtained during surgery with careful dissections on intact cadaveric knees, the so-called ‘‘intermediate oblique aponeurotic layer’’ described by Scott Dye was shown to represent the anatomical substrate of this novel syndrome. Conclusions: Although anterior knee pain is extremely common in professional cyclists, the exact etiology has not been described. We present a new clinical entity in high-level cyclists, termed pre-patellar friction syndrome (PPFS), caused by direct traumatic or chronic overuse damage to the triple-layered soft-tissues anterior to the patella. Correct diagnosis is critical and based on the typical history and clinical findings. In these cases, surgical treatment with careful excision of the disrupted area of the ‘‘intermediate oblique aponeurotic ‘‘pre-patellar sof-tissue layer delivers excellent results with all athletes regaining their preinjury performance level.

P37-1242 Snapping hip syndrome: importance of the dynamic echographic study E. Monteiro1, J. Torres1, S. Silva1, A. Moura-Gonc¸alves2, R. Pinto3 1 Hospital S. Joa˜o, Porto, Portugal, 2Hospital de Sa˜o Joa˜o, Orthopaedics, Porto, Portugal, 3S. Joa˜o Hospital, Orthopaedics, Oporto, Portugal Objectives: The Snapping Hip (S.H.) is characterized by a rebound in the hip joint or adjacent tissues. It may have an external cause (ex:

123

S360 rebound of the iliotibial band on the greater trocanther), or an internal cause. It is more common in female patients, 15–40 years of age. Our goal is to present our experience with a simple, non-expensive diagnostic tool, that allows us to confirm a diagnosis wich is mainly clinical. Methods: Twelve snapping hips (8 regular sports practising female patients, 18–43 years old), were treated and followed. Clinical examination, hip X-ray, Dinamic Echography and MRI were done. 4 hips (2 patients with bilateral S.H.) were submitted to physical therapy without improvement. Infiltration with local anesthetic and steroid in the iliotibial band, over the greater trochanter, was done in 2 hips- one improved. 6 patients needed a Z plasty of the iliotibial band (documented with photos). Results: Physical Examination of the 12 hips showed a snap over the greater trochanter with flexion, aduction and internal rotation. The X-ray and the MRIs were normal. Dinamic Echography (documented in video), showed a rebound of the iliotibial band on the greater trocanther in all hips. All patients submitted to conservative treatment were able to maintain their activity, without pain. Patients who underwent surgery recovered completely, without any symptoms. The average Harris Hip Score was 91—Excelent. Conclusions: X-ray and MRI are structural anatomical exams. Clinical diagnosis can only be confirmed by a dynamic, functional exam, like Dynamic Echography. In our experience, this a simple, noninvasive and cost-effective exam. Most published series of surgically treated patients are very small. Despite the fact that conservative treatment should be the initial approach, dynamic echography is an additional tool when considering surgery, in patients with refractory, painful disease.

P37-1245 Biomechanical study of different suspension devices (cortical and cortico-cancellous) used for ACL reconstruction A. Maestro1, T.E. Garcia Suarez2, C. Rodriguez2, L. Rodriguez3 1 FREMAP, Sports Medicine, Orthopaedic Surgery Department, Gijon, Spain, 2University of Oviedo, Construccio´n e Ingenierı´a de Fabricacio´n, Gijon, Spain, 3Hospital Cabuen˜es, Orthopaedic Surgery, Gijon, Spain Objectives: To analyze the biomechanical behavior of different suspension systems used in the ACL reconstruction, eliminating the variability and premature breakage caused by biological grafts using a synthetic graft. Methods: After made 40 femoral pig ACL reconstructions, with a high resistance cord as graft, 4 devices were analyzed: Biosteon Cross and Cross-Pin-Pin ACL, Endobutton CL and XoButton. There were two types of biomechanical tests, static tests of resistance to failure after surgery, and static tests of resistance to fatigue after 1,000 cycles, calculating the linear stiffness, failure load and elongation at that load. An universal testing machine MTS 15 kN dynamic load capacity was used. Results: There were no significant differences between the trials after surgery and after fatigue. The strongest device was ACL Cross-Pin, with a charge of failure after the fatigue of 958.5 N. Attachments of cortico-cancellous suspension (Biosteon Cross-Pin y Cross-Pin ACL) showed higher rigidities than cortical suspension (45 N/mm compared to 35 N/mm), while the lowest elongation was exhibited by Biosteon Cross-Pin (16.9 mm). The failure mode of devices corticocancellous suspension was rupture of the pin, while the cortical suspension devices was a broken bone because of the plate inserted across the tunnel. Conclusions: The accelerated rehabilitation (500 N) does not diminish the mechanical properties of fixations of suspension. From a mechanical standpoint, the fixations of cortico-cancellous suspension have a better response to load.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 P37-1265 Biomechanical analysis of different interference screw used in the ACL reconstruction A. Maestro1, T.E. Garcia Suarez2, C. Rodriguez2, L. Rodriguez3, J. Fernandez Lombardia4 1 FREMAP, Sports Medicine, Orthopaedic Surgery Department, Gijon, Spain, 2University Oviedo, Dpto. Construccio´n e Ingenierı´a de Fabricacio´n, Gijon, Spain, 3Hospital Cabuen˜es, Orthopaedic Surgery, Gijon, Spain, 4Hospital Arriondas, Orthopaedic Surgery, Arriondas, Spain Objectives: To analyze the biomechanical behavior of different screws used in the ACL reconstruction, eliminating the variability and premature breakage caused by biological grafts using a synthetic graft. Methods: The femoral part of ACL reconstruction was made in 30 pig. As graft was used a high resistance cord, and were analyzed 3 devices: two bioabsorbable interference screws Biocryl (PLA) and Biosteon (PLA and HA) and one metal screw Softsilk. Static tests were performed after the resistance to fatigue failure of 1,000 cycles, using a universal testing machine MTS 15 kN load capacity. was calculated the linear stiffness, the failure load and elongation at that load. Results: No significant differences were found between the behavior of the devices. The major results were obtained for the screw Softsilk, with an average charge of failure after the fatigue of 1,134.9 N, a stiffness of 54.9 N/mm and an elongation of 24.1 mm. The mean loads to failure for Biosteon and Biocryl were respectively 761.6 N y 723.5 N, while its stiffness stood at around 50 N/mm. The failure mode was in all cases drag out the graft fixation. The Softsilk showe the more torn out of the cord. Conclusions: The mechanical properties of compression devices (screws) do not decrease after fatigue. The metal screw has a slightly higher mechanical property, but it’s design produces a more tear in the graft.

P37-1270 Incidence and prognosticators for injuries in Belgian soccer players: evolution over the past decade P. Bollars1, L. Vanlommel2, S. Claes3, K. Corten4, J. Bellemans5 1 KU Leuven, Orthopaedic Surgery, Leuven Pellenberg, Belgium, 2 KU Leuven, Orthopaedic Surgery, Pellenberg, Belgium, 3University Hospitals Leuven, Department of Orthopaedic Surgery and Traumatology, Pellenberg, Belgium, 4KU Leuven, Leuven, Belgium, 5 University Hospitals Leuven, Catholic University Leuven, Orthopaedic Department, Pellenberg, Belgium Objectives: Football is the world’s most popular sport with approximately 265 million active players. With ever growing athletic and financial expectations, the nature of the football game may have become more intense with an associated increased risk for injuries. This large retrospective nation-wide study (1) evaluated whether the incidence, type, and severity of football injuries in Belgium have significantly changed over the past ten-year period and (2) evaluated the influence of possible prognosticators for the risk of injury during a football season. Methods: The national Royal Belgian Football Association (Koninklijke Belgische Voetbalband, KBVB) represents 416,000 football players. The KBVB database collects all injuries of the members and this database was searched for all injury reports. The detailed injury data from the season 1999–2000 and season 2009–2010 were vigorously recorded and compared the incidence of injuries, type of injury, affected body part, timing of injury and the respective influence of gender, age and level of performance as variable possibly influencing the abovementioned parameters. Results: 417,462 soccer players (401,976 men and 15,486 women, 162,558 adult and 254,904 youth players) were members of the

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 KBVB in season 1999–2000. A total of 31,493 injuries (7.5%) were reported, with an average of 0.075 injuries per player per season (0.076 and 0.061 for men and women respectively, p \ 0.0001). In season 2009–2010, 415,934 players (394,250 men and 21,684 women, 161,963 adult and 253,971 youth) were member, with a total of 24,280 injuries (5.8%). The average number of injuries per player per season was significantly lower in 2009–2010, i.e. 0.058 (p \ 0.0001), 0.059 for men and 0.045 for the females, a reduction of 23% compared to the first season. Gender was an important risk factor with a significantly lower relative risk in female players. The top level players in the National competition had a significantly lower relative risk compared to amateur level players (0.043 and 0.071, respectively). The mean age of all players in the KBVB was 21.8 years (range 4–90), with a significantly lower risk for youth players compared to adult players (0.044 vs. 0.102 for adult players). The vast majority (65%) of injuries occurred during competition, whereas 23% occurred during training activities. 10.5% and 13.1% of injuries were fractures sustained during both seasons, respectively. Nation-wide preventive measures taken in the second season significantly decreased the risk of injuries during the winter compared to the first season (0.018 vs. 0.011, respectively). Conclusions: Despite the proposed increase in sportive and financial pressure, contemporary football inflicts fewer injuries in Belgian football compared to one decade ago. Possible explanations for this positive trend are a good prevention program (the FIFA 11 + program), better medical care and renewed postponement policy by the KBVB.

P37-1274 Stress fracture in youth elite athletes A. Ta´llay1 1 National Institute of Sports Medicine, Sports Surgery, Budapest, Hungary Objectives: Children are being encouraged to take up sports earlier and earlier, on multiple teams and at greater levels of intensity. The most significant known predictors of stress fractures are high-impact activities, particularly running, basketball, and gymnastics. Similar pattern can lead to stress fracture, even in unexpected sports and under unusual circumstances. Methods: In a 1 year period prospectively all youth elite athletes with the diagnosis of stress fractures were collected in the National Institute for Sports Medicine. More than 500 adolescent and preadolescent elite athletes appeared in the study period, we found stress fractures of different locations in six cases. Average age was 16 (13–19) years. 2 male and 4 female athletes were involved. In our 1 year survey all five elite athletes were treated conservatively. Immediate X-ray in none of the cases was informative. Acute CT scan was performed for a swimmer with bilateral tibia pain, stress fracture was not possible to detect on the scans. MRI was the only safe way to establish the proper diagnosis. Results: In contrary to data from the literature, we diagnosed stress fractures in very different sports. 2 female handball players, 2 tennis players, and 2 swimmers were affected. 2 of the fractures were bilateral; in one of the cases 2 metatarsal bones were affected. The average recovery time took in average 8 (6–12) weeks; full return to sports was possible after 10 (5–14) weeks. The treatment protocol was partial non weight bearing, and magnethoterapy, orally bone healing supplementing dietary supplements were given. A significant structural risk factor—a valgus alignment of the lower limb—was detected only in our male swimmer patient, who suffered the bilateral tibia stress fracture during the preseassonal conditioning period, while a 5 km jogging session, which was an unusual load for him. In all other five cases no major predisposing factors were detected, overuse was the only risk factor.

S361 Conclusions: All athletes were able to return to their original sports in the same level. Our swimmer patient was even able to win an Olympic Gold Medal in Singapore Youth Olympic Game. To avoid overtraining, overuse, and evaluate the extrinsic and intrinsic risk factors are the main issues to prevent stress fractures in youth elite athletes.

P37-1279 Sports medical experiences with a U12-U15 football academy. Incidence of injuries in a 1 year period A. Ta´llay1, A. Pavlik1 1 National Institute of Sports Medicine, Sports Surgery, Budapest, Hungary Objectives: Football Academies for U17-U16 age groups appeared only a few years ago worldwide. The same for younger players (U12U15) has even less tradition. The aim of our study was to set up a professional sports medical background for our young players, with a full medical service of a preseason screening to sports rehabilitation. The main objective was to register all injuries and their treatment during the study period of 1 year. Methods: All Academy players have a general sports medical screening twice a year with ECG and Urine tests. There is a weekly on site consulting for injured and sick players, and a continuous Hospital background provided. Exercise physiology is also performed once a year. Two sessions of physiotherapy and massage therapy/per week is organized onsite. Proprioceptive prevention training was leaded by the rehab couch. All sporting injuries were registered by the team physician. Results: • Age groups: 95,96,97,98 • Height (cm): 168,155,154,148 • Weight (kg): 54.1,40.6,39.6,36.9 • Previous surgeries: 10 • Number of injured players: 52 • Number of consultations: 125 Out of the 80 players 4 surgeries were needed. The most frequent type of injury was the muscle injures, followed by the tendinopathies. Proximal hamstring injuries (37.5%), and adductor injuries (20.8%) happened most frequently. No major knee ligament injuries occurred during the study period. In 3 cases we diagnosed osteochondritis dissecans of the medial femur condyle. Conclusions: The high number of muscle injuries and tendinopathies occur most probably because of the extremely high number of training sessions; these children have 5 times 1–2 h/week. In our experience, training only on artificial grass is not optimal, may cause high number of overuse injuries. Thanks to the regular proprioceptive prevention training in our Academy, the number of knee ligament and ankle injury is extremely low. In our experience even for this young age group well organized sports medical background is needed.

P37-1388 Arthroscopic treatment of lateral tibial plateau fractures M. Savvidis1, I. Bisbinas1, D. Georgiannos1, I. Theodoroudis1, G. Gouvas1 General Military Hospital, 1st Orthopaedic Department, Thessaloniki, Greece Objectives: Tibial plateau fractures can be a challenging injury to manage for orthopaedic surgeons. Although not all fractures are amenable to arthroscopic reduction and fixation techniques, the arthroscope is still an important instrument for intraarticular diagnostic and treatment purposes. Arthroscopy provides: diagnosis of intraarticular pathology, earlier and better recovery of joint motion, less soft tissue dissection, anatomic restoration of the joint surface.

123

S362 The aim of this study is to present our initial experience and the clinical results in arthroscopic assisted treatment of tibial plateau fractures. Methods: Between 2005 and 2010 55 patients suffering tibial plateau fractures Scharzker I-III, were treated in our institution.18 of them were treated with arthoscopic restoration of the articular surface of lateral plateau, with the use of the ACL- reconstruction aiming devise. In 8 cases calcium sulfate (MIG III) was used to fill the metaphyseal void, in other 10 cases DBM in croutons were used, along with miniosteosynthesis (k-wires or screws). All patients were evaluated with CT scan, in certain cases MRI scan was performed also. All patients kept the epidural catheter postoperatively for 24 h CPM was applied to the affected limp. FWB was allowed 10 weeks postoperatively. Results: The average operative time was 50 min. The average hospitalization was 3 days. No infection, no delayed, non-union, or mal union was noted. All patients returned to their previous activity level with full ROM of the affected limp. Conclusions: Arthroscopy has many applications in the treatment of tibial plateau fractures. It serves as the best method of identifying and treating intraarticular pathology. In addition, arthroscopic reduction and internal fixation of certain types of tibial plateau fractures allows for anatomic reduction of the joint surface and rigid fixation with less morbidity and better visualization. Arthroscopic reduction and internal fixation may be used for types I-III tibial plateau fractures and, given the right clinical scenario, may be superior to open reduction and internal fixation.

P37-1394 The injury incidence in elite female soccer in Iceland A´. A´rnason1, T. Magnu´sson1, T. Sveinsson1 1 University of Iceland, Department of Physiotherapy, Reykjavik, Iceland Objectives: Soccer is the most popular sport within the National Olympic and Sports Association of Iceland, and 30% of registered Icelandic soccer players are females. The purpose of this study was to record the incidence, location, type and duration of injuries, the rate of re-injuries, and identify some risk factors in elite female soccer players in Iceland. Methods: During the 2007 Icelandic soccer season, 192 female soccer players agreed to participate in this prospective study. They were from all of the nine teams that participated in the premier women’s soccer league in Iceland. During the study period injuries were recorded prospectively on a special form by the team physiotherapists or coaches in cooperation with the players. The form included information about the type and location of the injury, former similar injuries, injury mechanism, duration of the injury and diagnosis. The coaches also recorded individual player exposure during training sessions on a special form. Official records were used to monitor the player exposure during matches in the league, cups and in national team matches. A player was defined as injured if she was unable to participate in a match or a training session because of an injury that occurred in a soccer match or during training. The player was defined injured until she was able to play a match or comply fully with all instructions given by the coach. Descriptive statistic was used to describe injury incidence, type and location of injuries. Logistic regression was used to compare uninjured and injured groups of players, treating potential risk factors as continuous variables. Results: Of the 192 players participating in this study 76 (40%) incurred 121 injuries: 93 (77%) acute injuries and 28 (23%) overuse injuries. The total injury incidence (95% CI) was 3.4 (3.1–3.7) injuries per 1,000 h of football. During matches the injury incidence was 15.1 (11.8–19.2) injuries per 1,000 h and during training 1.7 (1.3–2.2)

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 injuries per 1,000 h. Of the total number of injuries, 58 (48%) could be categorized as re-injuries. Of the acute injuries, 59 (63%) occurred during contact with another player. Injuries were most frequently located in the ankle (n = 34, 28%), followed by the knee (n = 18, 15%) and lower leg (n = 17, 14%). The most frequent type of injuries were ligament sprains (n = 37, 31%), followed by muscle strains (n = 24, 20%). The duration of injuries was classified in four categories: 1–3 days (n = 28, 23%), 4–7 days (n = 19, 16%), 8–28 days (n = 39, 32%) and [28 days (n = 35, 29%). Previous injuries in the lower extremities (OR = 2.7, p = 0.009) and the amount of playing time (OR = 1.02, p = 0.023) were the only predictors found for new injuries in the lower extremities. Conclusions: The injury incidence in Icelandic female soccer is similar to other elite divisions, but the amount of re-injuries is higher in Iceland.

P37-1395 Injuries in Downhill-Mountain biking: a high risk sport ? K. Fehske1, L. Eden1, R.H. Meffert1 1 Klinik fu¨r Unfallchirurgie, Universita¨tsklinikum Wu¨rzburg, Wu¨rzburg, Germany Objectives: Down-hill-Mountain-biking (DHMB) is a growing extreme sport. The goal is similar to alpine skiing to descend a parcours made out of natural and artificial obstacles as fast as possible. The riders have especially-designed mountain bikes which enables them to go up to 100 km/h. DHMB is considered to be a high risk sport, eventhough there are just a few investigations published so far. The aim of this study was to analyze DHMB-associated injuries, how they were treated and if they lead to persistent symptoms. Methods: We included all semi-professional riders who took part in the IXS European Cup. The athletes completed a questionnaire. As head of the medical team we documentated every medical consultation occuring during the races. Results: 177 athletes (5.1% female) out of more than 10 nations took part in the survey. The average age was 23.4 ± 6 years, the average size 180.9 ± 7 cm and 75.6 ± 10 kg. 70% reported that they have already suffered from an injury of the upper extremity, 46.4% have had a shoulder trauma (19.8% clavicula fracture, 12% schoulder luxation, 12% injury to the acromio-clavicular-joint). In 63 athletes we found a hand fracture (52 radius, 4 mid-hand, 7 finger). Injuries to the lower extremity have been reported in 43% (9 anterior cruciate ligament ruptures, 27 lateral ligament ruptures of the ankle, 9 ankle fractures, 3 calcaneus fractures, 11 mid-foot fractures, 20 toe fractures, 9 tibia fractures, 1 femur fracture). In 28% we documentated a spine-injury with altogether 13 fractures (7 cervical Spine, 3 thoracical spine, 2 lumbal spine). There was an average lost time due to the injury of 7.95 weeks, 26% of the athletes reported post-traumatic disabilities and 34.5% chronical pain in the injured extremity. Altogether, we have documentated 14.2 severe injuries in 1.000 h (97.5 injuries per 1,000 athletes). Conclusions: Down-hill-mountain-biking is to be classified as a high risk sport. Injuries are more often and more severe as in regular mountain-biking. Injuries could be reduced by better protection gear and athlete’s awareness to classify their riding skills correctly.

P37-1415 Prolotherapy use in medial collateral ligament injuries of the knee: the findings of one English premier league team D. Murray1, N. Jain1, S. Kemp2 1 North West Deanery, Trauma and Orthopaedics, Manchester, United Kingdom, 2Wolverhampton Wanderers Football Club, Physiotherapy Department of Wolverhampton, United Kingdom Objectives: Knee injuries are a common occurrence amongst footballers. Medial collateral ligament (MCL) injuries make up a

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 significant proportion of these in professional players. There is a recent trend of treating such injuries with prolotherapy, the injecting of a sclerosant. This study reports on a consecutive series of MCL injuries. The aim of the study was to determine the grades of injury treated with sclerosant therapy and to report on the outcome of treatment within the professional football club setting. Methods: Data was collected prospectively for all MCL injuries suffered by professional players between 2009 and 2011, spanning two English Premier League (EPL) seasons at one EPL club. Each player’s demographics were recorded along with grade of injury, treatment received and outcome (measured as a return to competitive play). The head physiotherapist assessed all injuries clinically. All players received a magnetic resonance (MR) scan, which was reported on by a consultant radiologist with an interest in musculoskeletal radiology and sports medicine. Results: 12 separate MCL injuries occurred over the course of this study, accounting for 34% of all knee injuries sustained during this time. 50% of the MCL injuries were grade I injuries (sprains) and 50% were grade II injuries (partial rupture). No grade III injuries (complete rupture) were observed. 17% of injuries were a recurrence from previous sprains. All grade I injuries were treated conservatively. Players with grade I injuries had a mean time of 20.5 days to return to play. All grade II injuries were treated with prolotherapy. Players treated with scleroscant injections had a mean time of return to play of 33 days if they did not have any complications. One player with a grade II injury suffered complications and returned to play after 211 days. Conclusions: Our study shows MCL injuries to be the commonest type of knee injury in our study group. The findings suggest that players who received prolotherapy return to play quicker than elite athletes treated without, when comparing our data with previously published papers. We have shown prolotherapy to provide good outcomes and have a low complication rate in the treatment of grade II MCL injuries.

P37-1417 The incidence of acromioclavicular joint injury in national collegiate athletic association American football players H.J. Braun1, A.H.S. Harris1, J.L. Dragoo1 1 Stanford University, Orthopaedic Surgery, Redwood City, United States Objectives: American football has been played competitively at the collegiate level in the United States for over 100 years. It has been estimated that anywhere from 11% to 81% of participants will sustain an injury over the course of their careers. Injuries to the shoulder compose 10–20% of total football injuries, and are the fourth most common, behind the hand, knee, and ankle. In collegiate football, injuries to the acromioclavicular (AC) joint account for up to 41% of all shoulder injuries. The National Collegiate Athletic Association (NCAA) Injury Surveillance System (ISS) provides current and reliable data on injury trends in intercollegiate athletics. The purpose of this investigation was to analyze the NCAA ISS men’s football AC joint injury database in order to (1) compare the incidence of AC joint injury on natural grass and artificial turf playing surfaces and (2) describe the incidence and severity of AC joint injuries in NCAA football players by player position and activity at the time of injury. Methods: During 2004–2005 through 2008–2009 seasons, the NCAA ISS collected data on injury mechanism and severity and exposures. A reportable athlete-exposure (A-E) was defined as 1 student-athlete participating in 1 practice or competition in which there was the possibility for athletic injury. Incident rates were calculated as the ratio of injuries per player exposure. Ninety-five percent confidence intervals for the incident rates were calculated using assumptions of a poisson distribution.

S363 Comparison of number of injuries per exposure based on factors such as surface type (artificial vs. grass) were made with negative binomial regression models adjusted for the number of players and the sampling weights. Results: Between 2004–2005 and 2008–2009, NCAA collegiate football players sustained a total of 748 injuries to the AC joint. Per 10,000 A-Es, there was a significantly greater incidence of AC joint injuries on artificial turf with fill (3.4) and natural grass (3.5) versus artificial turf with no fill (2.0, p \ 0.001). Overall, injuries most frequently occurred in wide receivers (10.6%, 79/748), running backs (7.2%, 54/748), quarterbacks (5.6%, 42/748), special teams players (5.5%, 41/748), and linebackers (4.9%, 37/748). Players were most likely to be injured during offensive passing plays (25.5%, 191/748), offensive running plays (21.9%, 164/748), general play (17.5%, 131/748), and running play defense (14.3%, 107/748). Injuries resulting from contact with another player (72.2%, 540/748) and contact with the playing surface (26.7%, 200/748) were most common. Conclusions: In male NCAA football players, the risk of AC joint injury is significantly greater on artificial turf with fill and natural grass surfaces compared with artificial turf with no fill. Offensive skill players were most frequently injured and the majority of injuries occurred during offensive passing or running plays. The majority of injuries were due to player contact or contact with the playing surface.

P37-1442 Isolated ankle syndesmosis injuries in elite professional footballers: mechanism (using video analysis), investigations and treatment D. Murray1, N. Jain1, S. Kemp2, J. Calder3 1 North West Deanery, Trauma and Orthopaedics, Manchester, United Kingdom, 2Wolverhampton Wanderers Football Club, Physiotherapy Dept, Wolverhampton, United Kingdom, 3The Lister Hospital, Clinic for Foot and Ankle Surgery, London, United Kingdom Objectives: Most ankle sprains involve the lateral ligament complex, with the anterior talo-fibular ligament (ATFL) having the highest incidence of injury. Injuries to the distal tibia fibular syndesmosis, otherwise known as ‘‘high ankle sprains’’, are less common, but are problematic with regards to treatment, and result in an increased time taken to return to play. This study reports on mechanism of injury, investigation findings and outcome of both non-surgical and surgical management in syndesmosis injuries occurring in an English Premier League (EPL) professional football team. Methods: All severe foot and ankle injuries occurring over the course of two EPL seasons in one EPL football club were recorded. A database of all isolated syndesmosis injuries was gathered. Injuries occuring in competitive games were recorded on video for puposes of analysis. Players received X-ray, Ultrasound Scan (USS), or Magnetic Resonance Imaging (MRI) scan before diagnosis. Grades of injury and ligaments involved were recorded. Scans were reviewed by a musculoskeletal radiologist and an orthopaedic surgeon each with a specialist interest in sports injuries. Treatment was either non-operative or surgical with an interosseous suture and endo-button configuration using the TightropeTM fixation technique (Arthrex, Sheffield, United Kingdom). Time taken to return to play was recorded for all patients. Results: 20 severe foot and ankle injuries were recorded, which included 7 (35%) syndesmotic injuries. Of the syndesmosis injuries, 5 (62.5%) of the injuries occurred during a competitive game, and 2 during training. All but one of the injuries were a consequence of a tackle. All injuries except for one in the series occurred as a consequence of forced external rotation of the foot (Fig. 1). 6 players had MRI scan to diagnose the injury, and one had an USS. The strains and partial ruptures (grade 1 and 2) were treated conservatively Of the 5

123

S364 patients treated conservatively 2 received concomitant plasma-rich protein injections (PRP). Players who sustained a complete rupture to AITFL (grade 3) were treated by surgical measures in the form of a syndesmosis repair using the TightropeTM fixation technique performed by the same surgeon. Conclusions: Isolated syndesmosis injuries, are traditionally reported as having a low incidence of occurrence. Our study contradicts this, we suggest that these injuries were under diagnosed in the past, rather than having the low incidence suggested by the existing literature. We have shown specific mechanisms of injury with video analysis, to aid healthcare professionals in future diagnosis. As no randomised controlled trials exist that suggest a best treatment option, it remains to be seen whether we are treating these injuries in the most appropriate manner. However early results are promising with only minor complications and a relatively quick return to play.

P37-1444 Utilization of platelet rich plasma (prp) in orthopeadic practice: clinical and biological analysis of terapeutic effectivenes F. Franceschi1, R. Papalia2, B. Zampogna3, C. Nobile3, M.C. Tirindelli3, V. Denaro2 1 Campus Biomedico University, Orthopaedic and Trauma Surgery, Rome, Italy, 2Campus Biomedico University, Orthopaedic Surgery, Rome, Italy, 3University Campus Biomedico of Rome, Rome, Italy Objectives: PRP is obtained by centrifugation of whole blood with the aim of producing plasma with high concentration of platelets. The literature provides little information about the variability of the clinical efficacy of PRP in relation to the biological characteristics of the product. Objectives of the study in question are: clinical evaluation of effectiveness of PRP ‘‘home made’’ a group of 89 patients, analysis of cytokine profile and concentrations of growth factors in a 10 patients homogeneous group, during treatment with intra-articular and peritendinous injections of PRP; description of possible correlation between changes in cytokine pattern and concentration of growth factors with clinical response to treatment in the same group of patients. Methods: 89 patients were selected and declared fit to receive treatment with PRP ‘‘home made’’. In view of the underlying disease, (Tendinopathy, Epicondylitis, Chondral Lesions, Degenerative lesions of grade I–II–III–IV, Rotator Cuff Lesions) patients underwent to 3–5 cycles of intra-articular or peritendinous injections of PRP, at intervals of 7 or 15 days apart. Scores of clinical evaluation were administered to each patient before treatment, before any infiltration, and 6 months after the last injection. The concentration of growth factors, and inflammatory cytokines present in the PRP, were evaluated in a homogeneous group of 10 patients with degenerative lesions of the knee cartilage of grade III–IV sec. Kellgren-Lawrence century. Finally correlation between growth factors, cytochine and haematological parameters of the patient and PRP with clinical parameters was analyzed. Results: Overall Response Rate (ORR) was noted in 69 of 83 patients (84%). In addition, all the scores used have shown a significant rate (P \ 0.05) in the period immediately after the second injection. Analysis of the growth factors showed a progressive increase in concentration than baseline, after the first application, with a peak level at the second (beginning of the period of clinical improvement statistically significant), and subsequent decrease at the third infiltration. Several more (P = 0.01) or less (P = 0.05) significant correlations were identified, both with clinical scores and haematological index. The pattern of inflammatory cytokines, after a stable trend from time 0, showed a decrease after the second clinical application, at the peak of growth factors and beginning of the patient’s clinical improvement.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Conclusions: The results reported clinical efficacy in 84% of patients treated. In addition, the biological analysis showed at the beginning of the clinical improvement the increase in growth factors and decrease in inflammatory cytokines. More level I studies are necessary to identify the best production methodology and application of the product.

P37-1472 Mri findings in dominant and nondominant shoulders of asymptomatic elite handball players M. Leyes1, C. Flores1, M. Gonzalez1 1 CEMTRO CLINIC, Madrid, Spain Objectives: Quantify and analyse the MRI findings in dominant and non-dominant shoulders of asymptomatic elite handball players without history of previous injury. Methods: An MRI of the dominant and non-dominant shoulder was performed in 8 male and 8 female professional Handball players Exclusion criteria History of previous shoulder symptoms or surgery Goalkeepers Shoulder symptoms on physical exam Participation in other sport or physical activity involving high shoulder strain Inclusion criteria Male players enrolled in ASOBAL league during the 2009/2010 season. Female players enrolled in Honour Division league during the 2009/2010 season. Players with negative glenohumeral and subacromial shoulder provocative shoulder tests. Variables Demographic (age, height, weight, laterality, armspan, position) Physical exam (Strength and range of motion) MRI (rotator cuff injury, subacromial bursitis, capsular injury, acromio-humeral distance, calcification, articular cartilage injury, articular fluid, glenoid labrum injury, LHB injury, greater tuberosity injury and acromioclavicular joint injury). The study was performed with a 1.5 teslas Magnetic Resonance Results: Laterality Right-handed: 13  Left-handed: 3 Position Pivot: 4  Left wing : 4  Right wing: 2  Left back: 2  Middle back: 2 Right back: 2 Rotator cuff MRI fIndings Men (dominant shoulder): 12.5% normal rotator cuff. 62.5% tendinosis. 25% partial tear, 0% full thickness taer of the rotator cuff. Women (dominant shoulder): 50% normal rotator cuff. 50% tendinosis. 0% partial tear, 0% full thickness tear Men (not dominant shoulder): 87.5% normal rotator cuff. 12.5% tendinosis. 0% partial tear. 0% full thickness tear. Women (not dominant shoulder): 75% normal rotator cuff. 25% tendinosis. 0% partial tear, .0% full thickness tear. Multivariate analysis: The probability of finding an abnormal rotator cuff was 5.5 times higher in dominant versus non-dominant shoukders. Subacromial bursitis was present in 25% of male dominant shoulders and 12.5% of female dominant shoulders. Acromio-humeral distance was lower in dominant shoulders. Glenohumeral cartilage injuries were found in 25% of dominant shoulders and 12.5% of non-dominant shoulders. Labral injuries were only found in 12.5% of female non-dominant shoulders. Humeral subchondral cysts were found in 25% of dominant shouldersAC joint arthritis was more frequent in dominant shoulders (62%) compared to

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 non-dominant ones (37.5%). The probability of no pathologic MRI finding was 0.357 times greater in the dominant shouder. Conclusions: (1) Dominant shoulders in asymptomatic elite handball players show a higher incidence of pathologic MRI findings than nondominat shoulders. (2) The higher incidence of rotator cuff MRI findings correlates with a decreased acromiohumeral distance.

P37-1473 Arthroscopic treament of the arthritic ankle of the soccer player at the end of his career: is it worth? L. Osti1, R. Papalia2, F. Barca1, A. Del Buono2, N. Maffulli3, V. Denaro2 1 Hesperia Hospital, Modena, Italy, 2Campus Biomedico Roma, Roma, Italy, 3Queen Mary University of London, Barts and The London School of Medicine and Dentistry, Centre for Sports and Exercise Medicine, London, United Kingdom Objectives: The arthroscopic treatment of arthritic ankle has been debated in the literature with few specific reports. Purpose of this study is to evaluate the outcomes of this treatment in a group of former football players with arthritic changes into the joint and anterior bone impingement. Methods: 17/(seventeen) former active football players all males. age 33–58, amatorial and semi professional including indor soccer with pain and limited ankle function for both sports and daily living activities, who underwent arthroscopic treatment, were evaluated at follow-up of a minimum 2 years (2–9.1 years mean 4.7) In all the cases the orthopaedic surgeon arthroscopically removed the anterior tibial spur and the osteochodral lesions were treated using microfractures. Associated synovial pathologies were treated using radiofrequency (arthrocare device, Naples, Fl). All the patients underwent a similar post-op program with protecetd weight bearing of 6 weeks and return to light sport activitities including jogging no sooner than 3 months and were examined by an indipenden examiner with a minimum 2 year follow-up. AOFAS and SF 36 were administred in the pre-op and at follow-up Results: AOFAS and SF36 scores were both significantly improved compared to the pre-op values with no significant complications related to the procedure. Sf 36 values were significantly emproved compared to the pre-op values (P \ 0.01) with a mean bodily pain of 36 in the pre-op compared with a 57 post op. AOFAS values were statistically different betwenn the pre-op and post post-op values (P \ 0.01) with a mean value of 26 versus 71. Conclusions: Arthroscopic treatment of the footballer ankle at the end of his career can lead to improved function and quality of life.

P37-1513 The high-level gymnasts shoulder; results of surgical treatment P. Boileau1, P. Gendre1, G. Moineau1 1 L’Archet 2 Hospital, University of Nice-Sophia-Antipolis, Orthopaedic Surgery and Sports Traumatology, Nice, France Objectives: The results of surgical treatment for shoulder traumatic lesions in high-level gymnasts have not been described yet. The am of the study was to evaluate the results of surgical treatment of shoulder traumatic lesions in high-level gymnasts. To determine if the arthroscopic Bankart procedure is enough to stabilize the shoulder of a high-level gymnast and if the arthroscopic biceps tenodesis procedure does not preclude the return to the previous level of gymnastics participation. Methods: Twenty-four gymnasts (27 shoulders), operated between 1994 and 2008, were included. There were 22 men and 2 women, for

S365 an average age at surgery of 22 years old. Eight gleno-humeral instabilities and 9 unstable painful shoulders (UPS) were arthroscopically stabilized. Nine lesions of the supra-spinatus tendon were treated by arthroscopic debridement or disinsertion/reinsertion. Ten lesions of the biceps were treated by tenodesis or proximal reinsertion under arthroscopy. One acute acromio-clavicular disjunction was treated by percutaneous pinning. All these cases were reviewed radiologically and clinically by an independent observer at an average follow-up of 5 years (range 2–15 years). Results: Average Constant score was significantly improved from 77 to 94 points. Mean Walch-Duplay and Rowe-Zarins scores were respectively of 90 and 91.25 points for gleno-humeral instabilities, and of 89.5 and 87.5 points for UPS. Twenty-two gymnasts (92%) have resumed the practice of sports, 21 at the same level. The gymnasts subjectively evaluate the function of their own shoulder at 82% of a normal shoulder for the practice of gymnastics. Fifteen patients out of 17 have resumed the practice of gymnastics at the presurgical level after an arthroscopic Bankart procedure. All gymnasts who had an arthroscopic biceps tenodesis resumed their sports practice, but only 2 out of 4 who had a proximal reinsertion could do so. Conclusions: The return to a presurgical level of gymnastics participation is possible after shoulder surgery. The arthroscopicBankart procedure enables the return to gymnastics practice in case of instability or UPS. Biceps tenodesis with an interference screw is an efficient alternative to proximal reinsertion in case of lesion to this tendon in high-level gymnasts.

P37-1514 The high-level gymnasts shoulder; epidemiology and anatomical lesion P. Boileau1, P. Gendre1, G. Moineau1 1 L’Archet 2 Hospital, University of Nice-Sophia-Antipolis, Orthopaedic Surgery and Sports Traumatology, Nice, France Objectives: Pathology of high-level gymnast shoulder is not well studied, while the practice of gymnastics puts the shoulders under very high constraints. The aim of the study was to assess the epidemiology of surgical lesions of the shoulder in high-level gymnasts. Methods: Twenty-eight gymnasts (32 shoulders) were operated from 1994 to 2010 for injuries attributed to the practice of gymnastics. Evaluation of traumatic circumstances (apparatus, gymnastic movements) and the analysis of clinical data, imaging and lesions at arthroscopic evaluation were made by an independent observer, who is a retired high-level gymnast and current orthopaedic surgeon. Results: There were 26 males and 2 females; average age was 22 (16–33). The orthopaedic consultation was motivated by pain in 18 cases. Overuse injuries accounted for 2/3 of cases. There were 9 traumatic shoulder instabilities, 10 unstable and painful shoulders (UPS), 14 lesions of the long head of the biceps, 11 partial thickness supraspinatus tears and one spontaneous acute acromioclavicular joint dislocation. The lesions were multiple in 1/3 of cases. In 85% of cases the traumatic movement was a forced abduction, associated with humeral rotation during hanging stage in the apparatus. There was a predominance of shoulder instability (59%), with Bankart lesions occurring predominantly in the lower quadrant (58%), affecting mostly the support arm (74%) and involving mainly the apparatus bars (58%). Conclusions: The pathologies of the shoulder in the high-level gymnasts involve mostly male sportsmen, with a preponderance of traumatic shoulder instability and UPS. Lesions are however frequently mixed. There are essentially overuse injuries, possibly uncompensated by acute trauma, during hanging stage in the apparatus in forced abduction.

123

S366

Injury prevention

P38-191 The prevention of injuries in American flag football in Israel: a one-year pilot study Y. Kaplan1 1 Hebrew University, Lerner Sports Center, Jerusalem, Israel Objectives: American Flag Football (AFF) is becoming a very popular amateur sport in Israel, with over 85 teams and 1,100 players. Although it is a non-tackle sport, many moderate-severe contact-type injuries have been reported. The author conducted a two-season prospective cohort injury surveillance study (historical cohort) over the 2007–2009 playing seasons. The purpose of this study was to introduce a one-season (2010/2011 winter league), longitudinal, prospective injury pilot prevention program in an attempt to significantly reduce the incidence and the severity of sports-related injuries in American flag football in Israel (AFI). The results were compared to the historical cohort. Methods: A total of 724 amateur male (n = 610) and female (n = 114) players (ave. age 20.49 SD = 3.791) participated in the study (injury prevention cohort). Four intervention methods were introduced. The no pocket rule was enforced, mouth guards, ankle braces and a pre-season information brochure were distributed. All time-loss injuries sustained in game sessions were recorded by the off-the-field medical personnel. This was followed up by a more detailed phone questionnaire by the author, 7–14 days following the injury. Results: There was a highly statistically significant reduction in the number of finger injuries related to fingers being caught in the opposition player’s pockets as well as in the number of ankle sprains (p \ 0.01). There was no statistically significant difference in the severity nor in the incidence of facial injuries. Conclusions: This pilot prevention program demonstrates that the intervention methods introduced, resulted in a highly significant reduction of both finger and ankle injuries. This pilot program will form the basis of a further two-season longitudinal, prospective injury prevention study, the recommendations of which will be sent to the AFI for implementation, as well as, to the various organizations responsible for the game of American flag football world-wide.

P38-551 Effect of knee braces for MCL on dynamic and static balance H. Fujiya1, K. Tateishi1, N. Yui1, K. Yatabe1, T. Kohno1, H. Aoki2 1 St.Marianna University School of Medicine, Department of Sports Medicine, Kawasaki, Japan, 2FIFA Medical Centre of Excellence at Kawasaki Japan, Kawasaki, Japan Objectives: Many knee braces are used for preventing medial collateral ligament (MCL) injuries in American football. We investigated the stabilizing effect, effect on athletic performance and subjective evaluation (fitting condition) among those braces in past studies. The purpose of this study is to estimate the effect of knee braces for MCL on dynamic and static balance. Methods: This study compared following two braces; A: SHORTRUNNER AIRMESH (BREG Inc.) and B: Townsend-Rebel TM5 (Townsend). Brace A is soft type consisted of mesh nyron sleeve, dual-upright with hinges and two rigid straps. Brace B is hard type consisted of hard shell, dual-upright with hinges and five rigid straps.

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Twelve collegiate American football players (Avg. Age:19.8 years) participated in this study. None of the players had any symptoms nor past histories of injuries to either knee joints. Under three conditions (group C: no-brace, group A: brace A on right knee and group B: brace B on right knee), we performed following three tests; a: one leg standing test with eye open for 10 s. (static balance), b: one leg forward drop jump test with eye close for 5 s. (dynamic balance) and c: one leg lateral (right) drop jump test with eye close for 5 s. (dynamic balance) using Zebris PDM-S system (zebris Medical GmbH). Each test was performed three times, and measured Total track length of center of pressure (COP) (mm) and Confidence ellipse area (mm2) to determine the balance index. Statistical significance was examined by using Friedmen test. Differences were considered significant at the 0.05 level of confidence. Results: In Total track length of COP (mm), there were significant differences (p \ 0.05) among group C, A and B (C: 539.2 ± 74.4, A: 552.0 ± 66.5, B: 566.2 ± 54.7) in test c (one leg lateral drop jump test with eye close for 5 s). In Confidence ellipse area (mm2), however, there were no significant differences among three groups in all tests. Conclusions: This study showed that dynamic balance ability of bracing was worth than that of no bracing, furthermore, hard type brace is worth than soft type brace (group C \ A \ B). In this study, significant differences of balance index were not found in forward drop jump test but in lateral drop jump test. It is suggested that contact pressure of dual-upright with hinges to the medial and lateral of knee joint, inhibited the function of joint proprioception after the jump landing. In addition, it is indicated that brace B consisted of hard shell, dual-upright with hinges and five rigid straps is stronger for influence of diffused contact stimulation than brace A consisted of soft stick sleeve to the knee joint wholly. Further study is needed for details of structure and materials of braces and for players who sustained MCL injury with valgus instability.

P38-601 Increased trunk stability modulates sagittal plane lower extremity biomechanics during single-leg landings Y. Shimokochi1, T. Igawa1, A. Oishi1, N. Kato1, E.G. Meyer2, J.P. Ambegaonkar3 1 Osaka University of Health and Sport Sciences, Sennan-gun, Osaka, Japan, 2Lawrence Technological University, Southfield, United States, 3George Mason University, Manassas, United States Objectives: Improving trunk stability has been suggested to be important for knee injury prevention. Still little research has actually examined how increased trunk stability affects lower extremity biomechanics and subsequently knee injury risk. Our objective was to examine how increased trunk stability modulates sagittal plane lower extremity biomechanics during single-leg landings. Methods: Twenty physically active females (20.3 ± 1.4 years, 53.5 ± 7.2 kg, 157.9 ± 6.1 cm) performed single-leg landings from 30 cm box in two landing conditions; (1) self-selected landings (SSL), (2) enhanced Trunk Stability landings (TSL). In TSL, a straight cardboard bar was attached externally on subjects’ sacrum to thorax using an elastic bandage (Fig. 1), and subjects had a mouthpiece with a small hole in their mouth. When performing TSL,

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370

S367 Results: All results are presented in Table 1. Subjects demonstrated 10.5% and 13.2% greater thigh-thorax and thigh-sacrum flexion excursion angles during TSL than SSL. Subjects also had 15.0% larger peak hip extensor moment and 6.3% lesser peak knee extensor moment in TSL than SSL. Further, sagittal plane hip eccentric work was twice in TSL than SSL, but no differences in knee and ankle eccentric work between landing conditions. Conclusions: Our experimental model to increase trunk stability resulted in subjects having greater trunk flexion, hip joint energy absorption, and peak hip extensor moment but less peak knee joint extensor moment than self-selected landing. Our findings indicate that increasing trunk stability may lessen the muscle contraction demands on knee extensors while concurrently increasing hip extensor muscle contributions during landings, both mechanisms that have been previously reported to be protective to the knee.

Fig. 1 TibiaImplant_combined

Table 1 Results of paired sample t tests SSL

TSL

t

p

Thigh-thorax flexion excursion angle ()

40.0 ± 12.4

44.2 ± 10.6

3.09 \0.01

Thigh-sacrum flexion excursion angle ()

25.8 ± 10.3

29.2 ± 8.3

2.65

0.02

-0.160 ± 0.078 -0.184 ± 0.087 2.87

0.01

Peak hip extensor moment (Nm kg-1 m-1) Peak knee extensor moment (Nm kg-1 m-1) Peak ankle plantar flexor moment (Nm kg-1 m-1)

0.160 ± 0.064

0.150 ± 0.057 2.69

0.02

-0.102 ± 0.041 -0.106 ± 0.041 1.24

0.23

Sagittal plane hip eccentric work -0.013 ± 0.012 -0.026 ± 0.017 4.17 \0.01 (J kg-1 m-1) Sagittal plane knee eccentric work (J kg-1 m-1)

-0.081 ± 0.034 -0.077 ± 0.033 0.88

0.39

Sagittal plane ankle eccentric work (J kg-1 m-1)

-0.059 ± 0.027 -0.060 ± 0.026 0.15

0.89

subjects forcefully exhaled and hollow their abdomen while maintaining their spine straight. Using a 3-Dimensional electromagnetic tracking system, we calculated thigh-thorax and thigh-sacrum relative flexion excursion angles and hip, knee, and ankle peak extensor (plantar flexor) moments (Nm kg-1 m-1) and sagittal plane eccentric work (J kg-1 m-1). Paired-sample t tests compared landing conditions.

P38-633 Which screening tools can predict injury to the lower extremities in team sports? A systematic review A. Benjaminse1, J.M. Dallinga1, K.A.P.M. Lemmink1 1 University Medical Center Groningen, University of Groningen, Center for Human Movement Sciences, Groningen, Netherlands Objectives: Injuries to lower extremities are common in ball team sports. Considering the personal grief, disabling consequences and the high costs caused by these injuries, the importance of prevention is evident. From this point of view it is important to know which screening tools can identify athletes at risk. This systematic review describes the predictive values of physical and anthropometric screening tests for injuries to the leg, knee, ACL, hamstring, groin and ankle in ball team sports. Methods: An electronic search was done from January 1966 to September 2011. Inclusion criteria were as follows: 1. Full text; 2. Published in English, German or Dutch; 3. Acute musculoskeletal injuries of lower extremities; 4. Athletes participating in ball team sports; 5. Average age of athletes (C 13 years); 6. Physical screening tests and/or anthropometry; 7. Reliability, validity, sensitivity or specificity described in numbers; and 8. Predictive value described in numbers. Results: From 134 identified studies, 22 met the inclusion criteria. Based on this review there is support that general joint laxity (OR = 5.30, p \ 0.001), an older age (OR = 1.1 per year, p = 0.05), low postural sway (OR = 0.31, p = 0.005), lower H/Q ratio (OR = 0.93, p = 0.02) and may predict the occurrence of leg injuries. The anterior right/left reach distance [4 cm (OR = 2.50, p \ 0.05) and composite reach distance \ 4.0% of limb length in girls during the star excursion balance test may predict leg injuries as well (OR = 6.50, p \ 0.05). Height of the player could significantly predict a knee ligament injury (OR = 3.93, p \ 0.05). Regarding ACL injury, hyperextension of the knee (OR = 4.78, p = 0.02), sideto-side differences in anterior-posterior knee laxity (OR = 4.03, p = 0.002) and side-to-side differences in knee abduction moment (OR = 6.40, p \ 0.001) were significant predictive tests. Furthermore, an age [ 23 (RR = 3.80, p = 0.044) and age (OR = 1.4 [1 year], p \ 0.001) is a predictive risk factor for sustaining a hamstring injury. No agreement could be found in the literature regarding hamstrings flexibility and H/Q strength ratio as a predictive screening tool. Hip adduction-to-abduction strength ratio was the best predictor for adductor strain (RR = 17:1 [based on hip adduction less than 80% of abduction strength], p = 0.003). No consistency could

123

S368 be found on ROM as a predictive measure for groin injury. More strength of the plantar flexors (OR = 1.22, p \ 0.05) as well as decreased dorsiflexion ROM (OR = 0.63, p \ 0.05) may be predictive for sustaining an ankle injury. Finally, four of five studies found postural sway to be a predictive test for ankle injury (OR from 1.22 to 10.20, p \ 0.05). Conclusions: The significant predictive screening tools mentioned in this systematic review are recommended to coaches and medical staff to screen their athletes and provide an individual training program for those athletes who are at risk.

P38-641 Injury profile of competitive alpine skiers: a 5 year cohort study at the Swedish Ski High schools M. Westin1, M. Alricsson2, S. Werner3 1 Karolinska Institutet, Stockholm Sports Trauma Research Center, Stockholm, Sweden, 2Mid Sweden University, Department of Health ¨ stersund, Sweden, 3Karolinska Hospital, Stockholm Sciences, O Sports Trauma Research Center, Stockholm, Sweden Objectives: The aim of this study was to make a survey of injuries in young elite alpine skiers. Methods: During 5 years all students at the Swedish Ski High schools were prospectively followed regarding injuries, time of exposure, injury location, type of injury, injury severity and gender. Four hundred and thirty- one skiers, 215 males and 216 females fulfilled the entire study. Results: One hundred and ninety-three skiers (91 males, 102 females) sustained 312 injuries. The injury incidence for males was 1.62 injuries/1,000 ski hours or 2.97 injuries/100 months at a Ski High school and the corresponding figures for females were 1.77 and 3.25, respectively. One-hundred and ninety-three skiers (91males, 102 females) sustained 312 injuries during the 5 year period that the skiers were followed. Fortyone percent of the injuries were located to the knee. Sixty-nine percent were ligament injuries and 49% were classified as severe injuries. Out of 190 injured skiers 120 injured their left lower extremity as the first time injury (P = 0.0097). The risk of sustaining a re-injury or a new injury increased the sooner the first time injury occurred (P = 0.001). Conclusions: There was a high risk for alpine ski students to get injured during their education at the Swedish Ski High schools. The knee joint was responsible for most of the injuries, especially ligament injuries, in both males and females. Both genders were more likely to sustain injuries to the left than to the right leg. Nearly 50% of the injuries were classified as severe injuries.

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Results: Incidence of injury was 6.48/1,000 h/player (forwards: 6.6, defensemen: 7.2, goalies: 2.5) and was stable over the 4 years of observation. Incidence of injury resulting in time loss was 2.3/ 1,000 h/player (forwards: 2.6, defensemen: 2.0, goalies: 0.68). Most of the lesions were moderate (51%), (minor 36%, major 13%). Collision between players was the most common mechanism of moderate to severe injury. Contusions were the most frequently seen injuries (38%) tracked by strains/sprains (29%). Knee and shoulder were the most frequently injured part of body (both 12%). Concussion had a higher incidence in defensemen (0.64/player/1,000 h) compared to forwards (0.2/player/1,000 h). August (beginning of the ice summer camp) was the month with the highest injury rate with and without time loss (18.7 and 18.6% respectively). Conclusions: In spite of its kinetics, ice hockey injuries rate is comparable to other contact team sports. Risk factors are first month of the season, summer training deficiency, and to be a defenseman for concussion. Measures of prevention might include optimal physical conditioning before and during the season especially for hip muscles, stronger helmet for defenseman, better hand protection. Ice Hockey epidemiology Injuries in professional players is still not well known. Precise data will allow to determine risk factors for severe injuries and to develop prevention program.

P38-968 Injuries among elite Norwegian road cyclists: a prospective study B. Clarsen1, R. Bahr1, G. Myklebust1 1 Oslo Sports Trauma Research Center, Norwegian School of Sports Sciences, Oslo, Norway Objectives: There are no prospective studies of injuries among elite road cyclists. Results of cross-sectional studies suggest that knee and lower back pain may be particular problems in this group. The goal of this study was to register the injuries sustained by a group of elite cyclists over a three-month period, with particular focus on overuse injuries in the knee, lower back, thigh and shoulder. The study was part of a larger project to develop and validate a new method for registration of overuse injuries in sport. Methods: Members of ten elite Norwegian cycling teams (four ‘‘UCIContinental’’ teams, one UCI professional women’s team, and five elite junior teams) were invited to participate in this study (n = 98). All cyclists received a questionnaire by email once a week for 13 weeks that registered overuse injury symptoms in the knee, lower back, thigh and shoulder. In addition, all injuries that led to time-loss from cycling training or racing were registered using standard Table 1 Reduced training volume

P38-934 Professional ice hockey injuries: a 5 years prospective study G. Ornon1, D. Fritschy2, J. Menetrey1, J.-L. Ziltener3 1 University Hospital of Geneve, Department of Orthopaedic Surgery, Geneve, Switzerland, 2University Hospital of Geneve, Department of Surgery, Geneve, Switzerland, 3Hopitaux Universitaires, Hoˆpital beau-Se´jour, Geneve, Switzerland Objectives: To determine on several seasons the epidemiology of professional ice hockey injuries, the mechanism of injury, and risk factors to allow the development and implementation of a prevention program. Methods: All injuries occurring in one of the best Swiss Professional Hockey Team were recorded and monitored during 4 years. Injuries were graded as minor (time loss: 1–7 days), moderate (8–28 days), and major ([28 days). Summer preparation and return to play were controlled throughout the entire study.

123

Mild reduction

Knee

Lower back

Shoulder

Thigh

21

12

3

8

Moderate reduction

7

7

1

3

Severe reduction

6

5

1

4

Unable to participate

6

3

1

3

Table 2 Reduced sporting performance Knee

Lower back

Shoulder

Thigh

Mild reduction

18

19

5

7

Moderate reduction

10

7

2

7

Severe reduction

14

7

3

6

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 Table 3 Pain

Table 1 Group demographics (range, mean (SD)) Knee

Lower back

Shoulder

Thigh

Mild pain

37

33

21

15

Moderate pain

15

9

6

11

Severe pain

S369

5

5

2

1

epidemiological methods. At the end of the study, all cyclists were contacted by telephone to confirm the accuracy of the information collected by the questionnaires. Results: During the 3 months that the study was conducted, 61% of the subjects registered overuse injuries in the knee, 50% in the lower back, 32% in the shoulder and 32% in the thigh. Only 20% of cyclists were completely injury-free during the study period. The average weekly prevalence was 23% (SD 9) for knee problems, 16% (SD 7) for lower back problems, 8% (SD 4) for thigh problems and 7% (SD 2) for shoulder problems. Tables 1, 2, 3 summarize the consequences of the overuse injuries in these areas. Conclusions: The results support the findings of previous investigations that overuse injuries in the knee and lower back are common among elite road cyclists, whereas problems in the shoulder and thigh are less frequent. Knee injuries were the most prevalent and the most severe, and should be a focus for future injury prevention research.

Experimental group (n = 13)

Control group (n = 16)

All (mean (SD))

Age (years)

16–32, 19.1 (2.0)

18–32, 23.8 (4.5)

21.7 (4.2)

Height (cm)

164–176, 170.3 (4.3)

162–182, 171.4 (6.0)

170.9 (5.3)

Weight (kg)

59–71, 64 (4.4)

53–75 (65.7 (6.2)

64.9 (5.4)

BMI (kg/m2)

19.6–24.5, 21.1 (1.6)

18.8–24.8, 22.4 (1.6)

22.2 (1.6)

Results: A 3-way interaction (time by limb by group) was found for frontal plane angles at the knee during DVJ (p = 0.015). Post hoc analyses revealed no change in either limb in the EG, whereas the mean angle on the left side in the CG increased (greater valgus) over time. For the SLL a significant time by group interaction was found (p = 0.016) due to slight changes in the EG and CG towards less and greater valgus angles, respectively. Correlations were found between changes across time at each limb for the two tasks (p \ 0.05). Conclusions: Regular exercises, targeting specific muscles of the hips and trunk, may influence movement patterns during dynamic tasks in female team-handball players. Specifically, they may lessen the likelihood of players developing high-risk movement patterns during the competitive season. Reference 1. Willson JD, Davis IS., JOPST, 2008.

P38-1251 Specific training may influence knee frontal plane projection angles during dynamic tasks in female team-handball players H.M. Sveinsson1, E.O. Thorvardarson1, S.S. Svavarsson1, K. Briem2 1 University of Iceland, Physical Therapy, Reykjavik, Iceland, 2 University of Iceland, Research Centre of Movement Science, Reykjavik, Iceland Objectives: The serious impact of anterior cruciate ligament (ACL) injuries in athletes has been acknowledged by organizations such as the International Olympic Committee, as has the importance of exercise programs to reduce the risk of ACL injuries, in particular for women, whose injury rate is greater than that of men. Noncontact ACL injuries involve the collapse of the knee into a dynamic valgus, typically as players land from a jump or perform a trick movement involving a quick change in direction. As neuromuscular and biomechanical risk factors are modifiable, these have become the focus of research in recent years. The objective of this study was to explore the effects of a simple exercise program on movement patterns during dynamic tasks. Methods: 29 women from 4 premier league handball teams participated in this prospective study (Table 1). Baseline data were collected at the start of the competitive season, after which the teams randomly received either experimental (EG) or control group (CG) assignment. Isometric hip abduction and external rotation strength was evaluated, as well as lateral trunk strength, using a hand-held dynamometer. Frontal plane projection knee angles [1] were evaluated during a drop vertical jump (DVJ) and a single leg landing (SLL). Video data were collected with a high speed camera and angles identified with commercial software via markers placed on participants’ hips, knees and ankles. Members of the EG teams performed 4 specific exercises as part of their warm-up during all regular practices for 8 weeks, after which endpoint data were collected. The exercises targeted hip abductors and external rotators, and core strength. Data were analyzed for each dependent variable using a mixed model ANOVA for repeated measures of each lower limb (within-subjects factors) and groups (betweensubjects factor).

P38-1284 Late effects of team handball in female national team players. High prevalence of osteoarthritis, and functional limitations in knee joints 10 years after silver medal winning at the Sydney olympic gGames A. Ta´llay1, T. Halasi1 1 National Institute of Sports Medicine, Sports Surgery, Budapest, Hungary Objectives: To determine the prevalence of functional limitations and osteoarthritis (OA) in different joints, focusing on knee-related symptoms in Hungarian female national team handball players 10 years after winning an Olympic Silver Medal. Methods: 24 Female National team handball players involved to the Olympic Games in Sydney 2000. All players’ anamnestic data for injuries were collected prospectively before the preparations to the Olympic Games. All players were physically examined, and with standardized weight-bearing knee radiography was taken in athletes with previous knee or ankle surgeries. Ten years after elite sports self administered patient questionnaires were taken (the Knee Injury and Osteoarthritis Outcome Score questionnaire) and all athletes were examined by to senior orthopedic sports surgeons. Results: Of the available cohort of 24 female players the following orthopedic procedures were done: knee surgeries: 23 (ACL-repair: 9), ankle ligaments reconstruction: 10, hand + finger ligament surgeries: 5, elbow: 1, shoulder-stabilization: 1. By the end of the 10 years study period 20 (75%) answered the questionnaires, one player died in car accident, 3 were lost for the study. 50% consented to undergo knee, 25% ankle radiography. The mean age at assessment was 26.25 years, the mean sports age was 13 years by the start of our study. The mean weight was: 67.7 kg, height: 176.8 cm. Our players have 30–60 training days together, daily 2–3 trainings, up to 6 training hours daily. In team handball elite players have 20–25 national level plus 50 club level games/year. All female athletes with any lover limb surgeries in the past had radiographic changes in their index knee, and 18 (75%) fulfilled the

123

S370 criteria for radiographic knee OA. Answering the self administered questionnaires, 18 (75%) reported having severe symptoms affecting their knee-related quality of life. More than 50% of the players had undergone reconstructive surgery or debridement in the study period. Conclusions: A very high prevalence of functional limitations pain, and radiographic knee OA, was observed in this young female population, who’s majority undergone different orthopedic surgeries during their long handball carrier. These findings support the theories of preventive training in handball. Players must be aware of early osteoarthritis of the lower limb after knee and ankle reconstructions in the early period of their sporting carrier.

P38-1511 Reduction of sports injury incidence in a Luxembourgish sport school during a 3-year prospective follow-up D. Theisen1, A. Frisch1, T. Windal1, A. Urhausen2, R. Seil2 1 Public Research Centre for Health, Sports Medicine Research Laboratory, Luxembourg, Luxembourg, 2Centre Hospitalier Luxembourg, Center of the Locomotor System, Sports Medicine and Prevention, Luxembourg, Luxembourg Objectives: Sports injury surveillance is the basis of prevention. It defines critical issues, such as injury characteristics, provides information on risk factors and helps define prevention strategies. The aim of this study was to establish a sports injury surveillance system in a Luxembourgish sport school and to describe injury incidence over a 3 year period. We hypothesized that an increased injury awareness as well as simple measures of prevention might be able to reduce injury incidence in a general sports school setting. Methods: During the school years 2008–2011 an injury surveillance system was implemented at a Luxembourgish sport school, following successively 201, 215 and 269 pupils between 12 and 19 years during the school season, from September until July. Different disciplines were analyzed in racket sports (badminton, tennis,

123

Knee Surg Sports Traumatol Arthrosc (2012) 20 (Suppl 1):S101–S370 table tennis), (other) individual sports (athletics, canoe-kayak, cycling, gymnastics, judo, karate, swimming, triathlon) and team sports (basketball, handball, football, volleyball). A personal sports diary was used to record amount and context of sport practice daily for every athlete. Sports injuries were evaluated with a standardized questionnaire (time-loss injury definition). In the first 2 years all data was collected by members of our team, who also performed weekly checks to ensure compliance and provided regular feedback for enhanced motivation. In the last year an electronic surveillance system (TIPPS, Training and Injury Prevention Platform for Sports) as well as simple measures of preventive fitness programmes were introduced. Results: A total of 1,081 injuries were recorded for a total of 685 athletes, representing a rate of 1.66 injuries/athlete in 2008–2009, 1.89 in 2009–2010 and 1.27 in 2010–2011. In 2008, 73.1% of athletes incurred a sport-related injury during the observation period. In 2009, the rate was 74%, and in 2010 it dropped to 67.3% (v2 = 2.54; p = 0.11). The injury incidence evolved from 3.7 in the first year to 4.5 in the second (c2 = 6.66; p = 0.01) and down to 2.8/1,000 practice hours in 2010 (v2 = 9.02; p = 0.03, compared to 2008). Injury-caused time-loss from regular sports practice was stable over the 3 observation periods (9.0, 9.7 and 9.5% of available time, respectively). Injury gravity, as evaluated by days lost for normal training, was globally stable over the years. Chronic overuse injuries changed from 26.1% in 2008 to 28.8% in 2009 and 22.1% in 2010 (N.S.). The percentage of recurrent injuries (same type of injury at the same anatomical location) changed from 19.5 to 26.5% in the two preceding years to 10.9% in 2010 (v2 = 9.66; p = 0.002 compared to 2008). Conclusions: Injury incidence was significantly decreased in the 3rd observation year. This reduction was associated with a significant regression in recurrent injuries. Injury surveillance and awareness programmes as well as simple measures of prevention might have been responsible for this positive evolution.