Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 DOI 10.1007/s00167-010-1119-z

Posters Shoulder P10-3 Arthroscopic treatment of distal clavicle fractures Pujol N.1, Hardy P.2 1 Hopital Andre Mignot, Orthopaedic Department, Le Chesnay, France, 2 Hopital Ambroise Pare´, West Paris University, Boulogne, France Objectives: To evaluate preliminary results of a new technique for the arthroscopic treatment of distal clavicle fractures using a double button device. Methods: Eight consecutive patients underwent arthroscopic reduction and internal fixation for displaced acute distal clavicle fractures using a double button device, between January 2007 and November 2008. Six were male and two were female (mean age 35, 25 to 45). Mechanism of injury was a fall on the shoulder in all cases. Clinical symptoms and X-ray findings of all patients were evaluated at least at 6 months postoperatively. Results: There were no intraoperative or postoperative complications, such as fixation failure, or infection. At 6 months postoperatively, no patients complained of pain or shoulder discomfort. All patients regained full range of motion. The Constant score averaged 95 (90-100). Two patients had some discomfort on the subcutaneous knot on the superior side of the clavicle. Follow-up radiographs showed bony union in all cases. One case presented a heterotopic bony formation without any subjective complaint. Conclusions: It is possible to treat fractures of the distal part of the clavicle with a minimally invasive arthroscopic assisted procedure. The preliminary results are encouraging, needing a longer follow-up to state indications and limits of this procedure.

P10-57 Influence of joint laxity on the occurrence of rotator interval lesion in throwing shoulder injury Nakagawa S.1, Yoneda M.2, Mizuno N.3 1 Yukioka Hospital, Department of Orthopaedic Sports Medicine, Osaka, Japan, 2Osaka Kosei-nenkin Hospital, Department of Sports Medicine, Osaka, Japan, 3Yukioka Hospital, Orthopaedic Surgery, Osaka, Japan Objectives: Since Burkhart et al reported the influence of internal rotational deficit of the shoulder on the occurrence of type 2 SLAP lesion in throwing shoulder, the importance of posterior shoulder tightness has been noticed. On the other hand, loose shoulders were frequently seen in throwing athletes. However, the significance of shoulder tightness and laxity was not clarified. In this study, the influence of joint tightness and laxity in throwing shoulder on the occurrence of rotator interval lesions was investigated. Methods: Sixty-one shoulders with throwing injury which underwent arthroscopic surgery were investigated. There were 50 tight shoulders and 11 loose shoulders. Frequency of rotator interval injuries was retrospectively investigated by arthroscopic findings, and compared between 2 groups. The investigated factors were long head of biceps muscle (LHB) lesion (injury or instability of LHB), superior and middle gleno-humeral ligament (SGHL, MGHL) injuries, and subscapularis and anterior part of supraspinatus tendon (SBS, SSP-A) injury. Furthermore, clinical features of the shoulders with LHB lesions were investigated. Results: LHB lesion was seen in 8 tight shoulders (16%) and 6 loose shoulders (55%), SGHL injury was seen in 14 (28%) and 8 (73%), and SBS injury was seen in 6 (12%) and 5 (45%), respectively. They showed statistically significant difference. On the contrary, MGHL injury was seen in 4 (8%) and 3 (27%), and SSP-A injury was seen in 19 (38%) and 3 (27%), respectively, and there was no statistical difference. In 14 shoulders with LHB lesion, SGHL injury was seen in almost all shoulders. On the contrary, while SBS lesion was seen in 4 of 6 loose shoulders, it was

never seen in 8 tight shoulders and there was significant difference between 2 groups. Conclusions: In throwing shoulder injury rotator interval lesions, especially LHB lesions, were frequently seen in loose shoulders. They often accompanied SGHL and SBS injuries. Structures around rotator interval may be injured by overstress with repetitive throwing motions, as rotator interval is important coordinate portion of joint laxity.

P10-65 Reversed total shoulder arthroplasty in the treatment of glenohumeral osteoarthritis with massive rotator cuff tears: a critical analysis of 68 Deltaprosthesis with a minimum follow-up of 2 years Sadoghi P.1, Hochreiter J.2, Jansson V.1, Mu¨ller P.E.1, Hausdorf J.1, Pietschmann M.1, Utzschneider S.1, Weber G.2 1 Ludwig-Maximilians-University, Department of Orthopaedics, Munich, Germany, 2Krankenhaus Barmherzige Schwestern Linz, Department of Orthopaedic Surgery, Linz, Austria Objectives: Our aim was a retrospective clinical and radiological evaluation of 68 shoulders operated with the Delta reverse-ball-and-socket total shoulder prosthesis by the senior author with a mean follow-up of 42 months. Methods: We included all patients with a minimum follow-up of two years (range from 24 to 89 months) who had been operated on by the senior author from January 2002 to May 2007. All patients had been evaluated preoperatively and at follow-up in terms of the following clinical scores: Constant score for pain, Constant Shoulder Score, Oxford Shoulder Score, UCLA Shoulder rating scale and DASH Score. Stability after the joint replacement was assessed by use of the Rowe Score for Instability and Oxford Instability Score. Conventional X-ray was performed at follow-up in all cases and graded according to the classification of Nerot et al. All complications were analysed and graded with respect to the paper from Goslings and Gouma. The whole patient population was evaluated before surgery and at a mean clinical follow-up of 42 months. We are able to report more than 98 percent of the follow-up data from the whole patient population. Results: Treatment of massive omarthrosis and rotator cuff tears by use of the inverted total shoulder prosthesis Delta showed a statistically significant improvement in all clinical scores. On the average, the Constant score for pain increased from 4.62 to 11.08 points (p\0.05); the Constant Shoulder Score from 32.65 to 60.31 (p[0.05); the Oxford Shoulder Score increased from 32.65 to 60.31 (p\ 0.05) and the UCLA Shoulder rating scale increased from 15.08 to 27.42 (p\0.05). Furthermore, patients reported a significant benefit in terms of a gain of stability which was quantified by an increase from 49.42 to 80.19 points in the Rowe Score for Instability and from 22.04 to 37.62 in the Oxford Instability score (p\0.05). Radiological analysis showed the following data according to the Nerot classification: Sixty-five percent of patients were graded as ‘‘0’’, 20 percent as ‘‘1’’, 3 percent as ‘‘2’’, 6 percent as ‘‘3’’ and 6 percent as ‘‘4’’. Eight complications occurred in this series in terms of a nerve lesion which was graded as ‘‘1’’ once, loosening of the humeral stem which was graded as ‘‘2’’ three times and loosening or fracture of the glenoid component which was graded as ‘‘2’’ in five times. At mean follow-up of 42 months, one patient of this series had died of decrepitude which was graded as ‘‘4’’ and one patient was lost of follow-up. Conclusions: All patients were satisfied with the treatment and none of them refused further analysis during the investigation. We summarize, that there were significant advantages identified in terms of the Constant score for pain, all clinical scores and the instability scores. Radiological analyses showed 85 percent of patients without or with a small notch only. On the other hand, the rate of complications should be taken into account. We conclude that shoulder arthroplasty with the Delta prosthesis shows significant benefits in terms of less shoulder pain, a higher stability and a gain

123

S124 of range of motion but on the other hand, we emphasize that this treatment remains a salvage procedure in the elderly only.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 differences were found between the two treatments. The type of acromion and severity of symptoms had a greater influence on the clinical outcome than the type of treatment. One patient was lost to follow up. Conclusions: As a result, we believe that the primary subacromial impingement syndrome is largely an intrinsic degenerative condition rather than an extrinsic mechanical disorder.

P10-93 Comparison of interscalene regional block and general anesthesia in arthroscopic Bankart repair Yildirim C.1, Toker T.2, Demirtas M.3 1 Tatvan Military Hospital, Orthopaedics and Traumatology, Bitlis, Turkey, 2Van Military Hospital, Anesthesiology and Reanimation, Van, Turkey, 3Ankara University Faculty of Medicine, Orthopaedics and Traumatology, Ankara, Turkey Objectives: The objective of this study was to demonstrate that interscalene regional block anesthesia has several benefits over general anesthesia for this type of surgery. Methods: We retrospectively reviewed the cases of 50 patients who underwent arthroscopic Bankart repair with suture anchors for posttraumatic recurrent anterior glenohumeral instability. All the patients were male. Patients were divided into two equal groups, based on the type of anesthesia administered. Twenty-five patients received interscalene regional block and 25 a general anesthesia. Interscalene regional block anesthesia was accomplished by use of 0.5% levobupivacaine alone using the multi injection technique. All of the blocks were performed by experienced anesthesiologists. For general anesthesia, endotracheal intubation was performed. All surgery was performed in the beach-chair position. Complete anesthetic, orthopaedic and hospital records were available for all patients. Results: Using the interscalene method, 19 regional blocks were entirely successful. Six patients required conversion to general anesthesia for adequate pain control. Portal entry sites required supplementation with local anesthetic in seven patients. No patient had a seizure, pneumothorax, cardiac event, or other major complications. Fourteen of the patients who had a block had minor complications, which included transient Horner’s syndrome in five patients, hoarseness that resolved in a few days in two patients, anxiety which was controlled with sedation in four patients and nausea in three patients. Conclusions: Interscalene anesthesia provided excellent intraoperative and postoperative analgesia and muscle relaxation with low morbidity. The regional block was found to be safe and effective, with a high degree of patient acceptance. Postoperatively, regional anesthesia resulted in fewer side effects, fewer hospital admissions, and a shorter hospital stay than did general anesthesia. In summary, interscalene anesthesia offers significant advantages for shoulder arthroscopy.

P10-119 The anterosuperior quadrant of the bony glenoid is poorly vascularized DeMers A.1, Arnoczky S.2, Lubowitz J.3 1 Taos Orthopaedic Institute, Taos, United States, 2Laboratory for Comparative Orthopaedic Research, College of Veterinary Medicine, East Lansing, United States, 3Taos Orthopaedic Institute Research Foundation, Taos, United States Objectives: The blood supply to the bony glenoid is relevant to both shoulder capsulolabral reattachment and glenoid osteonecrosis. Our purpose is to describe the microvascular anatomy of the bony glenoid. Our hypothesis is that discrete areas of the bony glenoid are poorly vascularized. Methods: Fourteen fresh-frozen adult cadaveric shoulders were evaluated using an India ink injection model, frozen to -40 degrees Celsius, and sectioned, axially or coronally, using a band saw. Sections were decalcified, cleared with modified Spalteholz technique, digitally imaged, and evaluated by two separate reviewers to characterize glenoid intraosseous microvasculature. Results: The glenoid microvasculature arises from arterial foramina based 11 mm (range 8-13mm) medial to the articular cartilage. Arterioles then arborize laterally toward the glenoid face. In all specimens, a qualitative decrease in the microvascular blood supply to the anterosuperior quadrant was observed. In addition, the blood supply to the bony glenoid does not supply the articular cartilage, and appears separate and distinct from the microvasculature of the capsule and labrum. Conclusions: Our results demonstrate that the anterosuperior quadrant of the bony glenoid is poorly vascularized and represents a watershed region. A limitation is that our methods provide only ex vivo, static, twodimensional analysis. Our results may have clinical implications: during capsulolabral repair, a common practice is to abrade the glenoid rim to encourage soft tissue healing to bone. However, abrasion of the relatively avascular anterosuperior glenoid may not effect the desired result of increasing bleeding to promote healing. Additionally, anchor placement in hypovascular regions may contribute to glenoid osteonecrosis.

P10-117 Bursectomy compared with acromionplasty in the management of subacromial impingement van Arkel E.1, Henkus H.E.2, de Witte P.3, Brand R.4, Nelissen R.3 1 MC Haaglanden, Orthopedic Surgery, Den Haag, Netherlands, 2HAGA Ziekenhuis, Den Haag, Netherlands, 3Leiden University, Orthopedic Surgery, Leiden, Netherlands, 4Leiden University, Statistics, Leiden, Netherlands Objectives: A prospective randomised study to compare the results of debridement of the subacromial space (bursectomy) without acromioplasty with those of acromiaplasty, in patients without a rotator cuff rupture who failed conservative treatment for primary subacromial impingement syndrome. Methods: A total of 57 patients with a mean age of 47 years (31 to 60) suffering from primary subacromial impingement without a rupture of the rotator cuff who had failed previous conservative treatment were entered into the trial. The type of acromion was classified according to Bigliani. Patients were assessed at follow-up using the Constant score, the simple shoulder test and visual analogue scores for pain and functional impairment. Results: At a mean follow-up of 2.5 years (1 to 5) both bursectomy and acromioplasty gave good clinical results. No statistically significant

P10-173 Intraoperative ultrasound localization in arthroscopic treatment of calcifying tendinitis of the shoulder Sabeti-Aschraf M.1, Graf A.1, Ziai P.1, Nemecek E.1 1 General Hospital of Vienna, Vienna, Austria Objectives: Calcifying tendinitis of the shoulder is a common condition with long lasting pain and impairment of the shoulder. Different conservative treatments including extracorporeal shock wave therapy and infiltrations of the shoulder joint are described and recommended to ease pain and improve shoulder function. In cases, refractory to conservative therapy arthroscopic surgery is recommended to curette the calcific deposit. The localization of the lime deposit itself is in some cases difficult and the use of an image intensifier in combination with the needling of the rotator cuff is than needed. This study was planed to evaluate if the intraoperative ultrasound localisation of the calcific deposit is feasible, reduces the number of needle perforations of the rotator cuff and the operation time and increases post operative pain and shoulder function. According to our knowledge there is currently no other randomized controlled study dealing with this topic. Methods: 20 Patients with therapy refractory pain and shoulder impairment are included into this prospective randomized controlled study trail after having passed the inclusion criterion. The study was approved by the

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 ethics committee of the author’s hospital. For clinical assessment immediately before and two and six weeks after surgery the Constant and Murley Score and the Visual Analogue Scale were used. Additionally, radiographs and a standardized ultrasound investigation, performed by a blinded radiologist were carried out. The population was randomized into two groups of equal numbers. Group one was treated with conventional arthroscopy. In group two the localization of the calcific deposit was performed with intraoperative ultrasound investigation. Results: At the time of abstract submission 13 of 20 patients have already been operated. The authors expect to finish the investigation by the end of the year 2009. Results will be presented. Conclusions: The concluding statement will be presented.

P10-174 The infiltration of the AC joint: ultrasound versus palpation a prospective randomized pilot study Sabeti-Aschraf M.1, Ochsner A.1, Schueller-Weidekamm C.1, Schmidt M.1, Graf A.1 1 General Hospital of Vienna, Vienna, Austria Objectives: The Acromio-Clavicular (AC) joint is very susceptible to degenerative processes that result in pain and functional impairment. One common modality of treatment has been local infiltration of the joint space. Although this procedure has produced notable positive results, needle misplacement occurs frequently. The aim of this investigation is to evaluate the effects of an intraarticular infiltration by comparing precise needle placement into the joint space using high-resolution-ultrasound with the conventional palpation technique. Methods: This prospective and randomized pilot study analysed 20 patients who were assigned either to the ‘‘ultrasound’’ or the ‘‘palpation’’ group. Clinical examinations were performed before treatment and at one hour, one week and three weeks after a single infiltration of local anaesthetic and corticoid. Results: In both groups significant improvement in pain and function was obtained up to one week post injection. Function remained significantly improved until the last follow-up and did not differ between the two groups. The agent was administered in all patients into the joint space in the ultrasound group. Conclusions: Ultrasound guided infiltration of the AC joint is an easily achieved procedure without any complications. However, clinical followup did not differ between free-hand and ultrasound-guided AC joint space infiltration.

P10-233 Biomechanical and tendon-bone interface evaluation of three different techniques for rotator cuff repair Baums M.H.1, Spahn G.2, Buchhorn G.H.3, Klinger H.-M.1 1 Georg-August-University, Department of Orthopaedic Surgery, Go¨ttingen, Germany, 2Clinic of Orthopaedic Surgery and Traumatology, Eisenach, Germany, 3University of Go¨ttingen Medical Centre (UMG), Department of Orthopaedic Surgery, Go¨ttingen, Germany Objectives: The repair of the rotator cuff remains a challenge, because structural failure and recurrent tears are frequent postoperative problems. This implies that current techniques fail to establish an adequate environment to facilitate healing of the tendon to the bone. We hypothesized that a modified suture-bridge technique combined with arthroscopic Mason-Allen stitches would have a superior response in cyclic loading and load-to-failure as well as tendon-bone interface pressure compared to a repair using the single-row technique with arthroscopic Mason-Allen stitches and double-row technique using arthroscopic Mason-Allen stitches and mattress stitches. Methods: Thirty fresh-frozen sheep shoulders were randomly assigned to three repair groups: (1) suture anchor single-row repair, (2) suture anchor double-row repair, and (3) modified suture-bridge repair using a titanium screw.

S125 Arthroscopic Mason-Allen stitches were used (single-row and suturebridge) and combined with medial horizontal mattress stitches (doublerow). Shoulders were cyclically loaded from 10 to 180 N. Displacement to gap formation of 5- and 10-mm at the repair site, cycles to failure, and the mode of failure were determined. Additionally, tendon-bone contact pressure was measured by using a pressure-sensitive film system. Results: Contact pressure was lowest in group 1, while group 2 and group 3 presented highest contact pressure. All specimens of group 2 and 3 resisted against 3000 cycles, while only 1 specimen of group 2 did. The ultimate tensile strength was highest in group 3 followed by group 2 and was lowest in group 1. Group 3 had lowest frequency of 5-mm gap formation followed by group 2 and 1. None of the specimens in group 3 reached a 10-mm gap formation. Conclusions: The modified suture-bridge technique in the current study combined with arthroscopic Mason-Allen stitches provides initial strength superior to single- and double-row repair under isometric cyclic loading as well as under ultimate loading conditions. Contact pressure was similar for double-row and suture-bridge techniques while single-row shows lowest results. The presented modified suture bridge technique might optimise the conditions of the healing biology for the reconstructed rotator cuff tendon and therefore reduce failure rates.

P10-246 Traumatic anterior dislocation of the shoulder: arthroscopic versus open stabilization. A 10 to 17 year follow up study Zaffagnini S.1, Giordano G.1, Zarba` V.2, Bondi A.1, Nitri M.1, Delcogliano M.1, Marcacci M.1 1 Rizzoli Orthopaedic Institute, Biomechanics Laboratory, Bologna, Italy, 2 Azienda Ospedaliera Universitaria ‘Vittorio Emanuele’, Orthopedic Department, Catania, Italy Objectives: Traumatic anterior shoulder dislocation and subluxation are common injuries. But few studies have compared arthroscopic and open stabilization of the shoulder at long-term follow up. The purpose of our study is to show whether an arthroscopic approach to repair Bankart lesion can obtain the same results at long follow up as an open procedure. Methods: We analyzed 110 nonrandomized consecutive shoulders in 110 patients who underwent a surgical repair of recurrent anterior shoulder instability between 1990 and 1999. Eighty-two patients were available at long term follow up (74,5% retrieval rate). In particular, 49 patients (59.8%) (group A) were treated with arthroscopic trans-glenoid suture (modified Caspari) between 1990 and 1995 (mean 15,7 year FU), whereas, 33 patients (40.2%) (group B) were treated with open repair between 1995 and 1999 (mean 12,7 year FU). We evaluated the patients in terms of failure rates, Rowe and UCLA scores. Results: The failure cases in the forty-nine patients treated with arthroscopic suture were 13, six dislocations and seven subluxations. The group A had also a Rowe score: function 24.2±8.2, stability 42.4±13.9, range of movement 18.6±3.8, total score 85.0±22.46. The UCLA score was: pain 8.8±1.7, function 8.6±2.1, muscle power 9.2±1.6, total score 26.4±4.8. Of the thirty-three patients treated with open repair, three had at least one post-op dislocations and four felt sometimes subluxations. The Rowe score in group B was: function 23.6±9.7, stability 41.2±14.9, range of movement 18.3±3.9, total score 83.2±24.4. Moreover the UCLA score was: pain 8.8±1.9, function 8.8±1.9, muscle power 9.2±1.2, total score 26.9±4.2. Conclusions: We showed that both techniques were fairly good in treatment of shoulder instability. In our series no significant difference was observed in redislocation rate and in Rowe and UCLA scores between the two groups. The recurrence rate (subluxations and dislocations) was high in both groups: the arthroscopic group had 26.5% and the open one had 21.2%. Our recurrence rate following open repair was higher than in many studies, while the rate after arthroscopic transglenoid procedure was almost equivalent. We hypothesize that one of the reasons for these higher recurrence rates may be the long term follow up. Another cause could be our decision to include subluxation as a failure value, even if there is no agreement about. In fact we believe it to be an important disability factor in sport as in life activities. After surgery, most of the patients returned to

123

S126 their preinjured activities. But at long term follow up almost all patients have stopped high level sport activity. Moreover, at this long term follow up, some patients told us a feeling of muscle weakness in the last years. In conclusion patients had good impressions about their shoulders thanks to surgery, but also because of lower functional demand.

P10-263 Prosthetic overhang most effective way to prevent scapular conflict in reverse total shoulder prosthesis De Wilde L.1, Poncet D.2, Ekelund A.3 1 University Hospital, Department of Orthopaedic Surgery and Traumatology, Gent, Belgium, 2DePuy France Company, Orthopaedics, St. Priest Cedex, France, 3Capio St. Gorans Hospital, Department of Orthopaedics, Stockholm, Sweden Objectives: Despite good clinical results of the reverse total shoulder arthroplasty inferior scapular notching remains a concern. The aim of this study was to evaluate the effect of 6 different parameters on notching. Methods: An average shape A-P view 2-D computer model of scapula was created, using data from 200 scapulae, so that the position of the glenoid and humeral component could be changed, as well as design features such as depth of the polyethylene insert, size of glenosphere and centre of rotation. The model calculates the maximum adduction (notch angle). Results: A change in humeral neck shaft inclination from 155 to 145 resulted in a gain of 10 in notch angle. A change in cup depth from 8mm to 5mm resulted in a maximum gain of 12. With no inferior prosthetic overhang a lateralisation of the centre of rotation from 0 to 5mm resulted in a maximum gain of 15 on notch angle. More lateralization resulted in increased gain in notch angle. With an inferior overhang of only 1 mm no effect of lateralizing the centre of rotation was calculated. Glenoid varus of 0 to 10, without inferior overhang, results in a gain of 10 on notch angle. A change in glenosphere radius from 18 to 21mm resulted in no gain of notch angle without prosthetic overhang. A prosthetic overhang to the bone from 0 to 5mm results in a maximum gain on notch angle of 39. Conclusions: To prevent an inferior scapular conflict in reverse total shoulder arthroplasty the change in neck-shaft angle or depth of the polyethylene insert had a modest gain in notch angle. The effect of lateralization of the centre of rotation and putting the glenosphere in more varus was completely eliminated by adding a small inferior overhang. The main effect of increasing the size of the glenosphere was if it created a prosthetic overhang. Of all 6 tested parameters the prosthetic overhang resulted in the biggest gain in notch angle and this should be considered when designing the reverse arthroplasty and defining optimal surgical technique.

P10-277 Arthroscopically assisted fixation of glenoid fractures: results and functional outcome Marsland D.1, Al-Khateeb H.1, Andrews E.1, Goldie B.1, Ahmed H.1 1 Whipps Cross University Hospital, Orthopaedics, London, United Kingdom Objectives: Intra-articular glenoid fractures are uncommon but potentially debilitating injuries, comprising approximately 10% of scapular fractures. Surgical fixation achieved through open reduction and internal fixation often requires extensive dissection, however, improved arthroscopic techniques have facilitated minimally invasive fixation. We present the results and functional outcome of a series of patients with intra-articular glenoid fractures managed with arthroscopy assisted reduction and percutaneous fixation. Methods: Three male patients (mean age 21 years, range 17-24) presented with glenoid fractures following high energy road traffic accidents. Two were Ideberg type III fractures and one was Ideberg type IV. Pre-operatively, CT scan reconstructions were obtained to define the fracture patterns and all 3 patients had significantly displaced fragments warranting surgical fixation. All patients were first arthroscoped in the beach chair position. After clearing the fracture and washing out the haemarthrosis, percutaneous

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 guide wires were introduced just anterior to the lateral end of the clavicle in a direction perpendicular to the main fracture line. The guide wires were visualised arthroscopically emerging from the superior fragment and passed across the fracture site after provisional reduction using probes. Definitive reduction and fixation was achieved utilising percutaneous cannulated screws over the guide wires with arthroscopic visualisation of the reduction and intra-operative radiological confirmation. Post operative CT scans were obtained to confirm reduction and the position of screws. The Constant Murley Shoulder score was used to assess outcome at six months and at one year follow-up. Results: Radiological union was achieved in all patients at 6 weeks. At three months, all patients were able to elevate to 130 degrees and abduct beyond 90 degrees. All patients returned to their previous level of activity 6 months post-operatively with no associated complications. The mean Constant Murley Shoulder score at one year was 95 out of a total of 100 points (range 94-96). Conclusions: Fractures of the glenoid cavity are rare and may result in chronic instability or degenerative joint disease. Our results show that arthroscopically-assisted reduction and percutaneous fixation of displaced intra-articular glenoid fractures can achieve anatomical reduction and good fixation with minimal soft tissue dissection. This technique allows early rehabilitation with rapid return to near normal function.

P10-289 Single row vs double row arthroscopic rotator cuff repair: clinical and 3T MRI arthrography results. A retrospective study of 24 patients Tudisco C.1, Bisicchia S.1, Savarese E.1 1 University of Rome Tor Vergata, Orthopaedic Surgery, Rome, Italy Objectives: The objective of this study was to evaluate the clinical and radiological results of two groups of patients operated for a rotator cuff lesion with two different techniques. The hypothesis was that there is no clinical difference, but the re-tear rate in the single row group is greater than in the double row group evaluated on 3T MRI arthrography study. Methods: We retrospectively evaluated two groups of patients. In 12 cases, rotator cuff repair was performed with a single row suture anchor technique, in 12 cases rotator cuff repair was performed with a double row double pulley suture anchor technique according to Arrigoni et al. All the patient were evaluated at a minimum of 2 years after surgery with the Constant and Murley Scale and the Simple Shoulder Test (SST), radiological examination was obtained with a 3T MRI arthrography. Results: The mean follow-up in the single row group was 29 months and 26 months in the double row group. The mean Constant score was 70 in the single row group and 68 in the double row group. The mean SST was 9.3 in the single row group and 10 in the double row group. The re-tear rate was 10/12 in the single row group and 8/12 in the double row group. Conclusions: To the best of our knowledge this is the first report on a 3T MRI arthrography for the evaluation of the rotator cuff repair. Single and double row techniques provide comparable clinical outcome at follow-up. The re-tear rate was greater in the single row group. However a statistical analysis could not be performed because of the small number of patient in each group. A randomized trial is needed.

P10-292 Higher fasting plasma glucose levels within the normoglycemic range and rotator cuff tears Longo U.G.1, Spiezia F.1, Maffulli N.2, Denaro V.1 1 Campus Biomedico University, Rome, Italy, 2Keele University, Institute of Science and Technology in Medicine, Stoke on Trent, United Kingdom Objectives: To determine the plasma glucose levels in non diabetic patients with rotator cuff tear. Methods: The study included 194 subjects who were operated at our institution. Group 1 included 97 consecutive patients (36 men and 61 women; mean age: 62.9 years, range 37 to 82) who underwent

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 arthroscopic repair of a rotator cuff tear in 2007 and 2008. Group 2 (control group) included 97 patients (36 men and 61 women; mean age: 61.6 years, range 36 to 80) with no evidence of shoulder pathology. These patients were frequency-matched by age (within 3 years) and gender with patients of Group 1. Results: Patients with rotator cuff tears (Group1) showed statistically significantly higher fasting plasma glucose levels within the normoglycemic range (p=0,007) when compared with patients of the control group (Group 2). Conclusions: The present study suggests that normal, but in the high range of normal, increasing plasma glucose levels may be a risk factor for rotator cuff tear. An enhanced understanding of these factors holds the promise of new approaches to the prevention and management of rotator cuff tears.

P10-325 Results of double row cuff repair with a titanium transfixation screw using a modified, subjective Constant score Geyer M.1 1 St. Vinzenz Klinik Pfronten, Orthopa¨dische Chirurgie, Pfronten, Germany Objectives: Results after double row cuff repair using a titanium transfixation screw with a modified subjective Constant score. Methods: From 2005 to August 2009 1000 patients had an open double cuff repair with a titanium transfixation screw technique (LASA DR, Ko¨nigsee). All patients completed preoperative a modified, subjective Constant score (max. 100 points). 57 of the first operated 67 patients could be reached after 1 year (85%). These 57 patients were further asked with the score every year. After 2 years 40 patients, after 3 years 42 and after 4 years 45 patients were reached. Score values were evaluated in a 5-year follow up for all patients and for tear size according to Bateman. Furthermore from consecutive 524 patients 374 (71,3%) completed the score with a mean follow up of 30 months. Patients with previous or following operations were not excluded. Results: The preoperative mean score value of 50,7 points increased after 1 year to 75,5, after 2 years to 81,9, after 3 years to 82,3 and after 4 years to 83,9 months. For the 30 months follow up the mean score improved from 50,1 points before operation to 78,7 points postoperative. Score values correlated with tear size preoperative and postoperative for Bateman 1-2 from 51,8 to 79,4 for Bateman 3 from 46,5 to 78,2 and for Bateman 4 from 47,8 to 73,3 points. Conclusions: Continuous improvement of results in double row cuff repair is shown by a modified, subjective Constant score in this longitudinal 5-year follow-up-study as well as in the 30 month follow up. After a steep increase in the first postoperative year the curve gradually flattens out. Further investigations have to show if arthroscopic double row cuff repair can provide comparable results.

P10-334 Arthroscopic release for the stiff shoulder Shishido H.1, Konno S.1, Otoshi K.1 1 Fukushima Medical University School of Medicine, Fukushima, Japan Objectives: Clinical courses after arthroscopic release for stiff shoulders were examined over time. Methods: The subjects comprised 24 shoulders in 21 cases that received arthroscopic release for stiff shoulders. Mean age of subjects at the time of surgery was 58 years (range, 43-76 years). Therapeutic results were assessed in terms of the Japanese Orthopaedic Association shoulder scoring system (JOA score), range of motion and visual analog pain scale (VAS) for nocturnal and motion pains. Therapeutic results were assessed before surgery and 1, 3 and 6 months and 1 and 2 years after surgery. Results: Comprehensive evaluation by the total JOA score revealed that compared with the preoperative status there was no statistical difference one month after the operation, but a significant improvement was observed

S127 for the first time three months after the operation. A significant improvement was found until six months after the operation, but not later. Taken together, comprehensive evaluation by the total JOA score showed that compared with the preoperative status, there was an improvement three months after the operation but not at six months after the operation and later. Parameters as to functions by the JOA score, flexion, and abduction improved significantly for the first time one month after the operation. They significantly improved until three months after the operation, but not later. Conclusions: In arthroscopic release, parameters as to functions by the JOA score, VAS, and the range of motion improved as early as one month after the operation compared with the preoperative status. As clinical courses thereafter, this study revealed that improvement became unremarkable in firstly parameters as to functions, flexion, and abduction, secondly comprehensive evaluation by the JOA score, nocturnal pain, and external rotation, and lastly motion pain.

P10-355 Clinical results of arthroscopic double-row repair of rotator cuff tears involving the subscapularis tendon Sano T.1, Matsuoka H.1, Nakayama K.1, Saji T.1, Hamamoto Y.1 1 Shizuoka General Hospital, Department of Orthopaedic Surgery, Shizuoka, Japan Objectives: The purpose of this study was to evaluate the clinical results of arthroscopic double-row repair of rotator cuff tears involving the subscapularis tendon. Methods: Sixteen patients (16 shoulders: 14 males and 2 females) with rotator cuff tears involving the subscapularis tendon between October 2006 and December 2008 were evaluated. The mean age of patients at the time of surgery was 64.4 years (range from 48 to 74 years). Eleven cases (68.8%) were traumatic. The average time interval between diagnosis and surgery was 3.4 months (range from 1 to 7 months). The mean follow-up period was 13.3 months (range from 6 to 36 months). According to the Ide’s classification, the size of subscapularis tendon tear was classified as follows: twelve superior 1/3 tears (superior 1/2 of tendon portion) and four superior 2/3 tears (all tendon portion). We had no cases of all tear (tendon and muscular portion). The retraction of stump of subscapularis tendon tear was classified as follows: eight minimal retractions, seven moderate retractions and one severe retraction. One case had a concomitant supraspinatus tendon tear, and fifteen cases had concomitant supraspinatus and infraspinatus tendon tears. The Cofield’s classification was used to measure the size of supraspinatus and infraspinatus tendon tears. One case was small tear, three cases were medium tears, six cases were large tears, and six cases were massive tears. The Patte’s classification was used to measure the retraction of stump of supraspinatus and infraspinatus tendon tears. One case was stage 1, ten cases were stage 2, and five cases were stage 3. We performed arthroscopic double-row repair for all cases, and used two to four metal suture anchors in order to repair the subscapularis tendon. All patients’ shoulders were immobilized for six weeks with an abduction pillow. Postoperative rehabilitation was initiated from 2 days after the surgery. The clinical results were evaluated with following four scales: the active range of motion of shoulder joint, the manual muscle testing, the Japanese Orthopaedic Association (JOA) score, and postoperative MRI. Results: The mean degree of active flexion had improved from 113 degrees to 159 degrees. The mean degree of active abduction had improved from 119 degrees to 151 degrees. The mean degree of external rotation decreased from 63 degrees to 61 degrees. The mean JOA score improved from 69 to 92, and the pain score improved from 13 to 26 during the follow-up. The repair integrity of subscapularis tendon evaluated with postoperative MRI was classified in accordance with the Sugaya’s classification. Eight cases were Type 1, four cases were Type 2, three cases were Type 3, and one case was Type 5. The Postoperative MRI showed 1 case of re-tear of subscapularis tendon, and 3 cases of re-tear of supraspinatus and infraspinatus tendon. Conclusions: The clinical results of arthroscopic repair of rotator cuff tears involving the subscapularis tendon were satisfactory. The larger size and

123

S128 more severe retraction of the subscapularis tendon tear were negatively correlated with an intact repair of the subscapularis tendon. The larger size and the more severe retraction of the supraspinatus and infraspinatus tendon tear were correlated with the size and retraction of the subscapularis tendon tear, and negatively correlated with an intact repair of the subscapularis tendon.

P10-410 Arthroscopic rotator cuff repair: single row or double row? Fiodorovas M.1 1 Klaip_eda’s University Hospital, Chief of Sports Trauma Department, Klaip_eda, 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). For single-row repair we do simple or Mason-Allen stitch using Arthrex 5,5 mm Corcksrew suture anchors, for double row repair—Smith&Nephew Footprint 5.5 mm implants. 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 ninety percent 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.

P10-438 Candidates for reverse shoulder arthroplasty: incidence in a tertiary orthopaedic center Braman J.1, Harrison A.1 1 University of Minnesota, Orthopaedic Surgery, Minneapolis, United States Objectives: Cuff tear arthropathy (CTA) is a challenging problem with difficult solutions. Reverse shoulder arthroplasty (RSA) implantation rates have increased dramatically in the United States of America since the introduction of RSA in 2005 because of the improved functional outcomes associated with RSA use in Europe. However, no attempts have been made to document the incidence of CTA in the clinical setting. Our purpose was to determine the incidence of RSA at an American tertiarycare University Shoulder service. Methods: Patients seen over a period of 32 months were identified by CPT codes for rotator cuff tear and arthritis, or arthroplasty failure. Chart review confirmed the diagnosis of CTA and documented active shoulder forward flexion, procedure performed if any, and contraindications to RSA. Criteria for RSA were defined as patient with painful CTA age[65, active forward flexion \90, and functioning deltoid. Results: We identified 72 shoulders (59 patients) with CTA. Thirty-six patients met the specified criteria for RSA and 12 underwent this procedure. In the timeframe of the study, there were 1764 unique patients seen in the clinic. Therefore, the incidence of candidates for RSA was 2% (36/ 1764). Furthermore, we found RSA made up 2.7% (12/451) of procedures performed. Patient factors precluding RSA included axillary nerve deficit,

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 age \65, poor glenoid bone stock, patient choice, and associated medical comorbidities. Conclusions: Despite the good clinical outcomes reported for RSA, the population in need of this device remains small. We believe this to be the first study examining the incidence of candidates for the RSA in a subspecialty orthopaedic practice. When this is extrapolated to a general orthopaedic practice, a single surgeon may see only 4 new patients with CTA per year and on average perform only one RSA per year. Given the complication rates and learning curve reported with RSA, the authors encourage thoughtful use of this relatively new implant as it is applied to a wider population.

P10-441 Effect of risk factors on results of arthroscopic stabilization for recurrent anterior shoulder instability Elsallab R.1 1 Mansoura University, Orthopedic Surgery, Mansoura, Egypt Objectives: Retrospective study to evaluate significance of the presence of what is known as risk factors for success of arthroscopic stabilization.many studies reported high incidence of failure because of risk factors such as; large hill-sachs lesions; high numbers of instability episodes and presence of capsular laxity. Methods: This study was performed on 120 shoulders in 120 patients who had recurrent shoulder dislocation; were managed by arthroscopic anchor suture stabilization from january 2004 to june 2008.mean duration of f.u.is 31months.exclusion factors are; patients had surgery for multidirectional instability; glenoid fractures more than one-third; and failed open surgery. All patients were evaluated by Rowe and Zains score. Results: Mean score increased from 44.7points to97.5 points postoperatively with total 108 excellent;12 good and no patients had poor nor fair results. For the patients had preoperative recurrent episodes less than 5; the final mean Rowe–Zains score is98 points; whereas 97 points for those who had more than 5 episodes this difference was insignificant (p=0.522). Correlation for presence of large Hill-Sachs lesions on final results was statistically insignificant(p=0.531). Correlation for presence of marked capsular laxity on final results was statistically insignificant(0.081). Conclusions: We can conclude that risk factors (mentioned in different literatures as a cause of high rate of recurrence after arthroscopic repair) are not absolute risk factors for success.

P10-445 Scapular manipulation technique for traumatic anterior shoulder dislocations Hattori S.1, Hara K.1, Kuroda H.1 1 Kameda Medical Center, Orthopedic Surgery, Kamogawa, Japan Objectives: Anterior shoulder dislocation is one of the most common dislocations in the setting of sporting events. Our institution recently introduced the scapular manipulation technique (SMT) for reduction of traumatic anterior shoulder dislocations. This retrospective study is aimed at reporting our experiences of using SMT. Methods: Between April 2008 and April 2009, SMT was applied to 11 patients who presented with traumatic anterior shoulder dislocations to the Kameda medical center, emergency medicine department, Chiba, Japan. SMT was performed by senior residents (PGY 3-5) of emergency medicine to patients in the prone position. Results: The study population ranged from 16 to 90 years (SD 55). Only one patient had a past history of shoulder dislocations at the same site. 9 patients had a history of falls (83%). We experienced a success rate of 83% over all without any additional fracture or neurovascular injury. The success rate at the first attempt without any local anesthesia or sedation was 54.5%. Conclusions: We report the successful use of SMT in the prone position for the reduction of traumatic anterior shoulder dislocations, without any complication. This technique can be performed by inexperienced

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 physicians who might encounter the patients with traumatic anterior shoulder dislocations in the setting of sporting events.

P10-475 Prospective clinical and structural evaluation of arthroscopic double-row rotator cuff repair: a novel technique; suture-bridge with FibreChain and SwiveLockTM El-Azab H.1, Buchmann S.1, Beitzel K.1, Waldt S.2, Imhoff A.1 1 Technical University Munich, Department of Orthopaedic Sports Medicine, Munich, Germany, 2Technical University Munich, Institute of Radiology, Munich, Germany Objectives: Assessment of repair integrity and clinical outcome after arthroscopic repair of rotator cuff tears with the use of a new knotless suture anchor; SwiveLock-FiberChain systemTM. Methods: Twenty patients with full thickness rotator cuff tear were treated with arthroscopic repair using the SwiveLock-FiberChainTM system. They were evaluated clinically with functional scores (Constant score, ASES index), visual analog scale (VAS), preoperatively, at 6 and 12 months postoperatively. The repair integrity was evaluated with ultrasonography (US) and magnetic resonance imaging (MRI) at an average of 14 months (12-16) postoperatively. Results: A re-tear rate of 20% was detected with both MRI and US. The subjective parameters (VAS and ASES Index) showed no significant differences between the group of re-tear and intact repair group. Whereas, the objective parameters (Constant score, muscle power and active ROM) showed significantly inferior results in the re-tear group compared to the intact repair one. Conclusions: The functional outcome has improved significantly with this knotless fast SwiveLock-FiberChainTM system and was more superior in shoulders with intact repair. Intact repair is mandatory for a favourable clinical outcome. The arthroscopic suture-bridge repair of the rotator cuff using the SwiveLock-FiberChainTM system produces satisfactory clinical and structural results, which are comparable to those of classical double-row repair while providing advantages to the surgeon and patient through knotless character, easy suture management and reduced operation time.

P10-538 All-inside biceps tenodesis with a new implant: the BiceptorÒ. Surgical technique and preliminary results Diaz-Heredia J.1, Ruiz-Iban M.A.1, Gonzalez-Liza´n F.1, Moros S.1, Del Cura Varas M.S.1 1 Hospital Ramon y Cajal, Dept. de Cirugı´a Ortope´dica y Traumatologı´a, Madrid, Spain Objectives: Long head of the biceps tendon pathology is frequently associated with rotator cuff tears and also presents isolated in SLAP lesions. Although there is strong controversy about wether repair to the glenoid bone, tenotomy or tenodesis are the best option there are some cases (young or high demand patients) in witch tenodesis is considered adequate. When indicated the shoulder surgeon has many available options to perform the procedure. The Biceptor system allows for the placement of the tendon inside a bone tunnel at the humerus without the need to retrieve the tendon outside the wound. The purpose of this study is to present the surgical technique and to review the results of 15 consecutive cases. Methods: The Biceptor system allow for the placement of the biceps tendon in a hole in the humeral head and its fixation with a single PEEK interference screw. The systems does not need the extraction of the proximal side of the tendon outside the wound and can be used all arthroscopically. Fifteen consecutive patients (7 males and 8 females, mean age 53±9 years) in which a biceps tenodesis was indicated (for rotator cuff rupture that affected the biceps groove in 11 cases and bicep tendinopathy in four cases) were treated using the Biceptor system. Surgical time devoted to tenodesis was noted. Patients were followed for at least three months, at this time an echographic evaluation of the shoulder was performed to evaluate if the biceps was still in the biceps groove.

S129 Results: No intraoperative complications occured. Mean operative time for the tenodesis was 13±5 minutes. In all cases the biceps tendon was fixed inside the groove at the 3 months echographic evaluation. Conclusions: The Biceptor system allows for a quick and easy biceps tenodesis that can be done all-arthroscopically.

P10-558 Repair of an anteroinferior glenoid defect by the Latarjet procedure: quantitative assessment of the repair by computed tomography Hantes M.1, Venouziou A.1, Giannakos R.1, Karidakis G.1, Bargiotas K.1, Malizos K.1 1 University Hospital of Larisa, Orthopaedics, Larisa, Greece Objectives: An anteroinferior osseous defect of the inferior glenoid, more than 25% requires a bone grafting procedure to increase glenohumeral stability and prevent anterior dislocation. The purpose of this study was to determine whether the Latarjet procedure (coracoid transfer to the glenoid) is efficient to restore a significant defect area of the glenoid Methods: Four-teen cadaveric shoulders were used. An anteroinferior glenoid defect was created and then the coracoid osteotomized to its angle and transferred to the defect. A three dimensional computed tomography was used to calculate the surface a) of the intact glenoid b) of the osteotomized glenoid and c) the reconstructed glenoid. Results: The mean area of the intact inferior glenoid was 618 ± 89 mm2. After creation of the defect the surface of the glenoid was reduced significantly (p=0.011) to 452 ± 55 mm2. The mean defect area was 27 ± 6% of the intact glenoid. After coracoid transfer the mean surface of the reconstructed glenoid was 646 ± 101 mm2 but it was not significantly larger from the intact glenoid. The mean surface of the coracoid which was used to repair the defect was 198 ± 34 mm[sup[2 or 32 ± 8% of the intact glenoid. Conclusions: Bone loss of the anteroinferior glenoid between 25% to 30% can be successfully addressed by the Latarjet procedure. Humeral head containment is restored after the Latarjet procedure since the surface area of the glenoid after reconstruction is close to its initial size.

P10-563 A comparison of acute versus delayed repair of traumatic rotator cuff tears Hantes M.1, Karidakis G.1, Bargiotas K.1, Varitimidis S.1, Dailiana Z.1, Malizos K.1 1 University Hospital of Larisa, Orthopaedics, Larisa, Greece Objectives: Traumatic rotator cuff tears are frequent injuries after trauma of the shoulder girdle and difficult to identify them immediately. As a consequence, the treatment often delays and this leads to difficulties in surgery and possibly fair results. The aim of this study was to compare the results of acute (within 3 weeks) versus delayed repair (after 3 weeks) of traumatic rotator cuff tears tears. Methods: Thirty-two patients with a traumatic rotator cuff tear have been treated surgically in our department during a 3 year period. Acute repair was performed in 12 patients and delayed repair in the rest 20 patients. The time interval between injury and operation was 12 days on average (range 3 to 20) for the acute repair group and 131 days on average (range 45–403) for the delayed repair group. Follow-up time was 20 and 24 months for the acute and the delayed repair group respectively. The two groups were evaluated with the UCLA and Constant scores. Results: The UCLA score in the acute repair group was 32.8 on average, while in the delayed group was 26.5 (p\0.05). The Constant score was 92.6 and 75.07 (p\0.05) for the acute and the delayed repair group respectively. Range of motion (shoulder flexion and abduction) was significantly better in the acute repair group. Conclusions: Acute repair of a traumatic rotator cuff tear provides better results in terms of shoulder function in comparison to a delayed repair. A delayed diagnosis of a traumatic rotator cuff tear leads to difficulties in surgery and possibly fair results. Therefore, it is mandatory for orthopaedic surgeons to diagnose and treat early these injuries.

123

S130 P10-591 Postoperative results and MRI evaluation of arthroscopic rotator cuff repairs for subscapularis tendon ruptures without LHB tenodesis Kobayashi T.1 1 KKR Hokuriku Hospital, Department of Orthopaedic Surgery, Kanazawa, Japan Objectives: The subscapularis tendon (SSC) tears are frequently detected in arthroscopy, and complete ruptures, partial ruptures, hypertrophes, and dislocations of long head of biceps tendons (LHB) are often found with SSC tears simultaneously. The common procedure for LHB lesion except in complete rupture is tenodesis, but we preserved the LHB in arthroscopic rotator cuff repair (ARCR) with precise repairs of upper SSC, anterior supraspinatus tendon (SSP), and biceps pulley which were the important biceps stabilizers. The purpose of this study is to investigate the postoperative clinical result and MRI evaluation of ARCR for SSC tear. Methods: Twenty-nine shoulders with SSC tear which were performed ARCRs and were followed up for at least one year with one year follow-up MRIs. Fifteen shoulders were men, 14 women, 21 right, 8 left, and the average age at operation was 64.9 (range from 23 to 80) years old. The average follow-up period was 21.9 (range from 12 to 40) months. LHB lesions were classified into four groups, N (normal), G1 (mild), G2 (severe), CT (complete tear). Constant score, postoperative cuff integrity on Sugaya’s classification and the changes of fatty degeneration at preoperation and one year post-operation on Goutallier’s classification by means of MRI were evaluated. Results: As for LHB lesions, six shoulders were classified as N, 7 G1, 10 G2, 6 CT. Constant score was improved from 55 to 88 points postoperatively. Postoperative cuff integrities of SSCs were evaluated as grade 1 in 20 shoulders, grade 2 in six, grade 3 in one, grade 5 in one, and those of SSPs were evaluated as grade 1 in 20 shoulders, grade 2 in four, grade 5 in three. The postoperative clinical results were strongly influenced by postoperative cuff integrities. Two ranks improvement of fatty degeneration in SSCs were observed in 2 shoulders, one rank improvement in 1, no change in 24, one rank deterioration in 1 (including one Sugaya’s grade 5), two ranks deterioration in 1. In SSPs two rank improvement were observed in one shoulder, one rank in 3, no change in 21, one rank deterioration in 4 (including three Sugaya’s grade 5). Postoperative Constant score were improved with no influence by LHB lesions, and no residual symptoms related to LHBs were noticed. Conclusions: The precise repair of upper SSC tendon, anterior SSP tendon, and biceps pulley can create the new LHB passage, and well functioned rotator cuff and LHB can be achieved by our LHB preserved repairing method. The clinical results of ARCRs for SSC tendon rupture were directly related to the post-operative cuff integrity, and no adverse effect by preservation of LHB was indicated after ARCRs. Our results suggested that tenodesis was not always necessary on condition that precise reconstruction around the LHB were performed.

P10-597 Arthroscopic acromioclavicular dislocation treatment using flexor carpi radialis autograft Abellan Guillen J.F.1, Gimenez Belmonte D.1, Melendreras Montesinos E.1, Martinez-Martinez, J. 1, Andre´s-Grau, Josefina1, Arroyo, F. F.1 1 Hospital Morales Meseguer, Orthopedic Surgery, Murcia, Spain Objectives: Standard method for treating Acromioclavicular (AC) injuries is still lacking, with a lot of different surgical reconstruction techniques described. The purpose of this study is to describe the details of our surgical technique and to present the obtained clinical results. Methods: We have recorded patients who presented acromioclavicular (AC) joint dislocation and underwent surgical treatment. Seven patients aged 18 to 45 were treated. They presented chronic Rockwood type III AC injury. They underwent arthroscopically assisted stabilization of AC dislocation using a double-strand flexor carpi radialis (FCR) allograft. In 2 of the cases distal clavicle was resected. The graft was secured with the AC GraftRope System (Arthrex, Naples, FL). The metal clavicle washer and coracoid button are joined by a continuous

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 loop of number 5 FiberWire, providing fixation during the healing phase. Results: Range of motion was full and not painful in all the patients 3 months after the surgical procedure. One of the patients presented loss of reduction, but without clinical relevance. No major complications were seen. Two of the patients presented wound dehiscence because intolerance to system suture. They both healed with conventional daily dressing. At final follow-up of 1 year all patients remain asymptomatic and with no limitations to daily activities, including sports. No symptoms or loss of strength related to donor site were seen. Conclusions: Acromioclavicular joint reconstructions with tendon grafts have shown substantial improvement in initial stability and a load-tofailure equivalent to the intact coracoclavicular ligaments. Tendons from semitendinosus, gracilis, tibialis or peroneus brevis muscles, have been proposed as graft donor site. We describe the use of FCR autograft for coracoclavicular ligament reconstruction. The GraftRope system (Arthrex, Naples, FL) allows achieving anatomic reduction and securing fixation. With an arthroscopically assisted approach the graft can be securely fixed. Our results show good results and no morbidity of the donor site using FCR autograft.

P10-600 The sheet-like structure - the key to new etiology of SLAP lesions Arai R.1, Kobayashi M.1, Toda Y.2, Nakamura S.1, Miura T.3, Nakamura T.1 1 Kyoto University, Orthopaedic Surgery, Kyoto, Japan, 2Kyoto University Graduate School of Medicine, Center for Anatomical Studies, Kyoto, Japan, 3Kyoto University Graduate School of Medicine, Anatomy and Developmental Biology, Kyoto, Japan Objectives: The purpose of this present study was to investigate the total constitution of the superior labrum in order to consider the etiology of SLAP (superior labrum anterior and posterior) lesions. Methods: Forty-nine shoulders of 29 cadavers (12 males and 17 females, average age 85.2) were used for anatomical study. The shoulder girdle was cut off and posterior capsule was incised for observation. After recording macroscopic findings by a digital camera, 11 superior-half glenoids were histologically examined. In 8 serially sectioned glenoids, 4 were parallel and the other 4 were vertical cut to the glenoid surface. Furthermore, 3 glenoids were radially sectioned at every hour clock time position between 10:00 and 2:00. Specimens were examined in HE and EVG stain. Results: Anterior to LHB (long head of the biceps tendon) origin, the sheet-like structure which branched off the rotator interval attached to the anterosuperior labrum (Fig. 1).

No LHB fiber penetrated the sheet-like structure and all of them ran posteriorly along the glenoid. The sheet-like structure contained many elastic fibers. The superior labrum consisted of three components; (1) anterior portion which was abundant in vessels, (2) fibers of the sheet-like structure which intermingled with the labrum fibers of anterior portion and both of them got along posteriorly, and (3) LHB fibers which intermingled with them and became a major substance of the posterior labrum (Figs. 2, 3).

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

S131 and 12 months postoperatively. Repaired rotator cuffs and footprints were evaluated on T1-weighted MRI (T1W) and T2-weighted MRI (T2W). And to evaluate the diameter of the anchor holes, three mean points were measured by quartering the anchor holes on T1W images. Results: The subjects were 11 males and 9 females, and their mean age was 64.7 years (range, 51-71 years). T1W images revealed that bone oedema of the footprint was observed in 90% of cases at first examination (1 month) but decreased 3 months later to 20% and 6 months later to 0%. Bone formation had been noted in the anchor holes in 87% of cases 3 months later. The anchor hole was significantly smaller compared to 3 months later (4.87 mm vs. 4.32 mm; p = 0.001) but it had not enlarged at 6 and 12 months (4.41 mm vs. 4.49 mm). T2W images revealed that half of the repaired rotator cuffs remained at high intensity at 1 and 3 months (58% and 33% respectively) but decrease at 6 and 12 months (17% and 7% respectively), and those with low intensity increased at 6 and 12 months (33% and 53% respectively). Conclusions: Bone oedema of the footprint had been observed in 90% of cases at first examination (1 month) but decreased 3 months later to 20%. Bone formation in the anchor holes had been observed in 87% of cases at 3 months. And the anchor hole tended not to enlarge but was smaller 3 months postoperatively, so anchor fixation was completed 3 months later. T2W images revealed that the repaired rotator cuff had remained at high intensity at 3 months in 33% of cases, but decreased at 6 months (17%). Therefore, it takes more than 6 months for integrity of the repaired rotator cuff to return.

Conclusions: There is no strong fiber bundle between the anterosuperior labrum and rotator interval because all LHB fibers run posteriorly. The sheet-like structure, which is rich in elastic fibers, connects them and would make a complex to resist the anterosuperior micro motion of the humeral head. SLAP lesions could be an injury of this flexible support system. The sheet-like structure is considered to play a key role of ‘‘the missing link’’ to understand the new etiology of SLAP lesion.

P10-607 Healing process of rotator cuffs and the anchor holes after arthroscopic suture-bridge rotator cuff repair by serial magnetic resonance imaging Komatsu T.1, Hashimoto K.1, Akita M.1, Iwasaki T.1, Ito J.1 1 Aomori Prefectural Central Hospital, Orthopedic Surgery, Aomori, Japan Objectives: It is important to evaluate the healing period of repaired rotator cuffs when deciding about rehabilitation and return to employment and sports. Magnetic resonance imaging (MRI) is one of the best methods to evaluate the repaired rotator cuff. A problem associated with MRI evaluation of the repaired rotator cuff is halation of the metallic anchors, which interferes with the evaluation of footprints and anchor holes. Since 2007, we have been carrying out rotator cuff repair using an arthroscopic suture-bridge technique, employing Panalok loop RC anchors (Mitek) constructed with poly-L-lactate polymer. We prospectively evaluated the healing process of repaired rotator cuffs and footprints, and measured the anchor holes enlargement by serial MRI after rotator cuff repair. Methods: Twenty patients underwent suture-bridge technique and were evaluated by serial MRI. The suture-bridge technique used 4 Panalok loop RC anchors (Mitek) that were made of polylactic acid. Two anchors were fixed at the medial side, while the other two were fixed at the lateral side of the footprint in the anchor holes. The anchor holes were drilled at the footprint (diameter, 2.9 mm; depth 18 mm). Serial MRI was done 1, 3, 6

P10-618 Clinical and radiological results 7 years after open repair of rotator cuff tears Steimer O.1, Kusma M.1, Kuschnierz M.1, Kohn D.1, Seil R.2 1 Saarland University Medical Center, Department of Orthopaedic Surgery, Homburg/Saar, Germany, 2Centre Hospitalier Luxemburg - Clinique d’Eich, Orthopedic Department, Luxemburg, Luxembourg Objectives: The reconstruction of the torn rotator cuff is a common orthopaedic operation. In our department, until 1999, these reconstructions were done in a standardised, open technique. The aim of our study was to investigate the subjective and objective results 7-9 years after open rotator cuff repair. In future, these findings could be used as a reference in the evaluation of innovative arthroscopic rotator cuff repair techniques. Methods: Between 1996 and 1999, open repairs of full-thickness rotator cuff tears were performed in 69 patients (18 women, 51 men, mean age of 64±11 years). The refixation was performed using transosseous sutures. 55 patients (15 woman and 40 men, mean age of 66±6 years) were available for follow up examination with an average follow up of 7,3±0,12 years. The evaluation consisted of an assessment of the operated shoulder joint with the Constant Score in comparison to the contralateral side, pre- and post-operative assessment using the Upper Limb-DASHScore, visual analogue scale and true AP x-rays. Furthermore the quality of live was assessed using the SF-36 form. Results: The DASH-Score improved significantly from preoperative 61±18 points to postoperative 16±19 points (p\0,01). The patients had a significant reduction in pain intensity (VAS) from preoperative 7,3±1,8 to postoperative 1,5±2,1 points (p\0,01). The Constant Score showed a significant difference between the operated shoulder (80±16 points) and the nonoperative side (90±12 points) (p\0,01). A comparison of followup radiographs with those obtained preoperatively revealed a significant superior migration of the humeral head, however, this had no significant influence on the clinical result in the Constant score. The SF-36 showed a mean value of 73,97±18,04 points. Conclusions: The open repair of rotator cuff tears increases the pain and improves the subjective symptoms of the affected shoulder joint.

P10-679 Predictive anthropometric measurements for humeral head curvature Fening S.D.1, Miniaci A.1 1 Cleveland Clinic, Dept. of Orthopaedic Surgery, Cleveland, United States Objectives: Osteochondral allograft transplantation for the treatment of osseous defects to the humeral head has recently grown in popularity.

123

S132

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

Because only a portion of the articulating surface of the humeral head is replaced, conformity of the allograft to the native surface is imperative to restore the natural geometry of the joint. To achieve proper conformity, it is essential that the curvature of the humeral head of the allograft tissue match that of the native tissue. Curvature determination is also important for shoulder replacement procedures. Curvature of the humeral head is difficult to directly measure in allograft specimens. As a result, predictive measurements, such as the maximum length of the humerus are used to predict this curvature. The purpose of this study was to investigate the value of various anthropometric measurements for predicting humeral head curvature. We hypothesized that the maximum length of the humerus would be the most predictive of humeral curvature. Methods: 60 (28 female, 32 male) cadaveric humeri were obtained from the Hamann-Todd Human Osteological Collection. Specimens ranged from 20 to 35 years of age at the time of death (27.9 ± 4.5, mean ± SD). Specimens from this collection include height and weight as collected at the time of death. All specimens were scanned with a 3-dimensional laser scanner (NextEngine, Santa Monica, California, USA). This scanner has been shown to be accurate to within 0.005 inches. Linear measurements were made according to the recording standards for skeletal remains. The first linear measurement was the maximum length of the humerus. This measurement was the direct distance between the most superior point of the humeral head and the most inferior point of the trochlea. The second measurement recorded was the epicondylar breadth, which was the distance between the most laterally protruding point on the lateral epicondyle and the most medially protruding point on the medial epicondyle. Both linear measurements were made by choosing points on the 3-dimensional scan, rather than the traditional osteometric board. Humeral head curvature was determined by a custom computational code to fit a sphere to the articulating surface of the humerus. Data analysis was performed in Minitab (version 13, State College, PA, USA). A linear regression was performed for each predictive measurement. A stepwise linear regression with forward and backward substitution was performed for the most predictive variables from the initial linear regression. Results: The most predictive factors (R2 [ 0.5) were epicondylar breadth, height, maximum humeral length, and gender. Based on the linear regression coefficients, these four factors (all normalized) were included in a forward and backward stepwise regression (a to enter and remove = 0.15). The resulting equation (shown below) had an R2 values of 0.807. Linear regression results Predictive Factor

R2

Gender*

0.55

Age

0.07

Height*

0.67

Weight

0.16

Maximum humeral length*

0.55

Epicondylar breadth*

0.68

Humeral diameter = 0.894 + 0.048*epicondylar breadth + 0.043*height 0.020*gender Conclusions: Of the predicted measurements evaluated, patient height, epicondylar breadth, and gender were most correlated with humeral head curvature. Including these three factors in a linear regression model increased the R2 value to 0.807. If only a single measurement can be used to size the humeral curvature, patient height will give approximately the same accuracy as epicondylar breadth, and can more easily be obtained. This is in contrast to our initial hypothesis, where we expected maximum humeral length would be the most predictive.

P10-808 Intra-articular repair of an isolated partial articular-surface tear of the subscapularis tendon Mori D.1 1 Koseikai Takeda Hospital, Orthopaedics, Kyoto, Japan

123

Objectives: The purpose of this study was to discuss preoperative assessments and operative techniques for isolated partial articular-surface tear of the subscapularis tendon (SSC), and to assess clinical outcomes of arthroscopic treatment. Methods: From April 2007 through November 2008, we observed biceps pulley in 189 cases of rotator cuff tear treated with arthroscopically, including 12 cases of isolated partial articular-surface tear of the subscapularis tendon. The mean age was 59.7 years old (range; 46-68), and they were three males and four females. The average follow-up was 16.5 months (range 12 - 26 months). In each case of the partial articular-surface tear of the SSC, a suture anchor was introduced through the anterior portal. One of two suture limbs from the suture anchor was passed through the upper edge of the SSc and the other one was passed through the superior glenohumeral- coracohumeral ligament complex. Clinical outcomes were evaluated using the UCLA scores, the Lift-off test, the Belly-press test and the Bear-Hug test. Results: The mean UCLA score improved from 17.4 to 32.8. Preoperatively, the Lift-off test was positive in three shoulder and the Belly-press and the Bear-Hug test were in five shoulders. Postoperatively, the lift-off test was none and the Belly-press test and the Bear-Hug was positive in one shoulder. Conclusions: We conclude that arthroscopic SSC repair and the stabilizing of the LHB is an effective procedure and that instability of the LHB has been considered as one possible cause of anterior shoulder pain.

P10-846 Shoulder plica El-Herraoui A.A.-S.1 1 Egyptian Military Medical Academy, Orthopedic Surgery, Cairo, Egypt Objectives: From April 2008 to July 2009, 102 patients were managed as shoulder instability. 11 patients of them (10.8%) were complaining of no definite history of dislocation. However these 11 patients (8 male and 3 female) of average age 25 years old (22-34 years) were suffering from shoulder instability with feeling of giving way, popping and pain on motion with no history of obvious trauma. All these 11 cases were not responding to the conservative treatment (NSAID and physiotherapy). Methods: On physical examination, there was a positive shoulder instability tests in all the cases. MRI study did not detect a definite lesion. Shoulder arthroscopy (GA and lateral decubitus position) was done for the 11 patients; where there was a fold and thickened band in the anterior capsule extending from the capsular attachment to the glenoid fossa inferior to the biceps tendon and passing laterally to be attached to the capsule. In 5 cases it was adherent to the capsule and in 6 cases it was separated. All the intraarticular structures were normal in all cases except the presence of this band (PLICA). The presence of this plica in the anterior capsule can explain the engagement of the head of humerus on that fold (plica) on abduction and external rotation; giving the feeling of instability (not complete dislocation), giving way, popping and pain, where the patient can rotate the head of humerus to free it from this plica trap, feeling well. The arthroscopic solution for all cases was to excise this plica by evaporating it using thermal technique. In 3 cases there was a minimal Bankart lesion that was repaired by a suture anchor method. The postoperative program was application of arm sling for two weeks with a physiotherapy training in all direction of movements from the first day and for about four weeks. Results: The results were excellent in all the cases with a painless full range of motion ROM without feeling of instability, popping or giving way. Conclusions: The presence of a shoulder plica in some cases of shoulder instability (10.8%) may be one of the causes of the instability and can be solved by excision of that plica.

P10-853 Minimal invasive AC joint reconstruction (MINAR) - clinical results of a prospective study Rosslenbroich S.1, Zantop T.2, Herbort M.3, Raschke M.1, Petersen W.4

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 1 Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 2Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany, 3University of Muenster, Department of Traumatology, Muenster, Germany, 4Martin-Luther-Krankenhaus, Unfallchirurgie, Berlin, Germany Objectives: Treatment of complete acromioclavicular joint disruption remains controversial and ranges from rehabilitation to extensive surgical reconstruction. However, high-grade injuries Rockwood type IV, V, and VI) are typically treated surgically. Aim of the current study was to present clinical results 24 months after a minimally invasive subcoracoid and clavicular fixation (MINAR). Methods: In a prospective study design, the clinical result of 20 subsequent patients with Rockwood type 5 was evaluated at 24 months postoperatively. All patients were treated using a minimal-invasive AC joint reconstruction (MINAR, Karl Storz, Tuttlingen) with 2 FlippTacks and a linkage material (0.7 mm PDS in 8 cases, 1 mm Ethibond suture in 12). Clinical results were quantified using the Constant score and weightbearing stress x-rays were performed. Results: In all cases the coracoidal flip button could be placed safely under the coracoid. Mean operative time was 28,6 minutes. No peri- or postoperative complications such as nerve and vascular injuries, thoracic injuries, infection, thrombosis) did occur. The mean Constant Score was 94,1 point after a mean of 23,3 month. In two cases (1 PDS suture, 1 Ethibond suture) we observed a slight loss of reposition of less than half of the clavicula width without clinical correlation. No difference in clinical outcome between PDS or Ethibond suture was to be found. Conclusions: Our clinical experience shows that the MINAR technique (Karl Storz, Tuttlingen) is easy to perform and has comparable invasiveness to recently presented arthroscopic techniques. This technique has the benefit of being performed by surgeons that are not well aware of arthroscopy all the way to the coracoid process.

P10-868 Reliability of the palm up test and strength in flexion for biceps pathology determination Randelli P.1, Arrigoni P.1, Monteleone M.1, Polli L.1, Cabitza P.1 1 Universita´ degli Studi di Milano, IRCCS Policlinico San Donato, San Donato Milanese, Milano, Italy Objectives: The aim of this epidemiologic study is to evaluate the reliability of the Palm up test and the strength in flexion in the long head of the biceps pathologies. Design of study: Prospective, case-control study; Level of evidence 4. Methods: 44 patients who underwent to shoulder arthroscopy for cuff pathologies were examined with Palm up test before surgery. Strength in Flexion, measured with a digital tensiometer (Kern, series HBC 20K50) and pain during the week previous to surgery expressed as VAS were recorded as well. Intraoperatively, biceps pathology was classified in terms of region in 1) anchor related 2) intrarticular 3) pulley lesions. Surgery was blinded about preoperative measurement. Results: A total of 25 women and 19 men were examined. The age averaged 61 years old. In 24 cases (54%) the dominant shoulder was involved. The average BMI of operated patients was 27,4. 13 out of 44 patients (29%) did not present any biceps problems. Palm up test positiveness significantly correlated with the presence of biceps pathology. (p\0,05, Fisher’s exact test). The observed sensibility and the specificity of the test were respectively 93% and 38%. Strength in Flexion was significantly higher for patients without any bicipital pathology than for patients with the presence of biceps lesions (3,4±2,8 kg versus 1,1±1,7 kg; p-value\0,05, Mann-Whitney test). No difference was observed for pain between the two patient groups (Presence / Absence of Biceps lesions).

S133 One patient presented a lesion of the Biceps anchor. No significant differences were noted between patients with a pulley lesion (N=16) versus patients with an intrarticular tear (N=14) for strength, (0,9 kg versus 1,3 kg). Conclusions: Our study demonstrate an insufficient specificity of the Palm up test for Biceps tears and shows the importance to measure the Strength in Flexion when suspecting a Biceps tear. The results of our cases suggest that more data and prospective studies are necessary to investigate the correlation between strength in flexion and the area of biceps pathology.

P10-871 Clinical results of arthroscopic interval slide repair of massive, contracted, immobile rotator tears Randelli P.1, Arrigoni P.1, Polli L.1, Monteleone M.1, Cabitza P.1 1 Universita´ degli Studi di Milano, IRCCS Policlinico San Donato, San Donato Milanese, Milano, Italy Objectives: Massive, contracted, immobile rotator cuff tears are rare lesions that represent a surgical challenge. The purpose of this study is to report the results of a patient’s cohort undergone arthroscopic repair using an interval slide technique. Type of study: Cases series; Level of evidence 4. Methods: Twenty patients undergone arthroscopic rotator cuff repair were retrospectively evaluated at a mean follow-up of 32 months after surgery. All of these patients had large, severely contracted tears that could not be mobilized without an appropriated interval slide procedure. A double interval slide was performed in 11 patients and a single posterior interval slide was performed in 9 patients. The supraspinatus tendon was involved in all cases (100%), infraspinatus tendon in 19 cases (95%), subscapularis tendon and teres minor respectively in 9 cases (45%) and in 5 cases (25%). A mean number of 2,7 suture anchors needed to achieve the repair of adequately mobilized rotator cuffs. Patients were retrospectively evaluated using: Disabilities of the Arm, Shoulder and Hand questionnaire (DASH), pain score (VAS), Constant score and Strength in external rotation (SER). Strength was assessed with an electronic tensiometer (Kern, series HBC 20K50). Results: The average patient age was 63 years old. The dominant arm was involved in 80% of the cases. At least follow-up, all patients were satisfied with the procedure. In 66% of cases pain was equal to 0. The mean value of DASH score was 39 (range: 30 to 62). Active range of motion was regained after the index procedure (forward flexion: 147; abduction: 152; external rotation: 24). The normalized Constant score attained at least follow-up time was 92 (range: 56 to 110). SER improved from a preoperative mean of 0,6 kg to a postoperatively mean of 1,7 kg (p\0,05, Wilcoxon Signed Ranks Test). No significant complications related to the procedure were reported. Conclusions: The results of this study show a clear reduction in pain and a recovery of active motion, strength and function after repair of massive, contracted, immobile tears treated by an interval slide technique.

P10-877 Necessary and sufficient information for successive nursing from operating room to ward in arthroscopic rotator cuff repair Ueno M.1, Ikeda A.1, Okuno M.1, Hayashi R.1, Inokuchi T.1 1 KKR Hokuriku Hospital, Department of Nurse, Division of Operating Room, Kanazawa, Japan Objectives: Operating room nurses (ORN) write down the preoperative consultation record and the intra-operative record, and anesthesiologist dose the anesthetic record for not only the smooth and safe operation but also the individual postoperative care and rehabilitation. We, ORN, could not catch whether the ward nurses (WN) could effectively use these three records for individual patient care and rehabilitation program or not. In arthroscopic rotator cuff repair procedures (ARCR) which are the most frequent operative procedures in our operation room, ORN investigated whether WN can effectively use these three records, and whether the information from these three records is sufficient by means of questionnaire for WN.

123

S134 Methods: We conducted questionnaires for eighteen WN about preoperative consultation record, intra-operative record, anesthetic record, and postoperative oral explanation about operative course for WN. Eight contents in preoperative record such as past history and the result of blood examination, nineteen contents in intra-operative record such as operative procedure, operating time, number of suture anchors and its arrangement, seven contents in anesthetic records, and nine contents in postoperative oral explanation are assessed as useful or not. Results: ORN assessed the arrangement and the number of the suture anchors, those were related to the location and extent of the rotator cuff rupture, as the most important information for WN because of their direct concern with postoperative fixation periods, the way of the rehabilitation, and prohibited motions. They were evaluated in all of WN as most important information for ARCR. Fourteen out of nineteen contents, e.g. amount and contents of continuous intravenous infusion pump for postoperative analgesia, timing of antibiotics and precise process during operation were evaluated as useful information in all WN. ORN thought that it was not necessary for WN to record the every procedures and process during operation, because they did not have direct influence on postoperative care. However WN used these records to explain the operative process to other patients. Postoperative explanation for WN may not always necessary, if the necessary and sufficient information is recorded in intra-operative record. However the clear and compact summary during operation told by ORN can make WN image the postoperative state of the patient and perform immediate postoperative care. Conclusions: In this study WN almost agreed with ORN in almost all items which are thought to be necessary for ORN during operations. However WN did not always make use of the information of the individual operative and anesthetic procedures for each patient. In the future we cooperate with WN to be performed the individual postoperative care under necessary and sufficient intra-operative records that are recorded about actions by surgeons and co-medical staffs. To achieve these objects we are now preparing intraoperative clinical pathway for individual postoperative care.

P10-929 Is there an association of SLAP- and biceps Pulley lesions and probably a traumatic etiology of Pulley lesions? A prospective clinical study Patzer T.1, Kircher J.2, Lichtenberg S.3, Magosch P.3, Habermeyer P.3 1 University of Marburg, Department of Orthopaedics and Orthopaedic Surgery, Marburg, Germany, 2University Hospital Du¨sseldorf, Department of Orthopaedics, Du¨sseldorf, Germany, 3ATOS Clinic Heidelberg, Shoulder and Elbow Surgery, Heidelberg, Germany Objectives: Aim of this clinical study was to evaluate an association of long head of biceps tendon (LHB) lesions in form of SLAP and biceps Pulley lesions and the analysis of a questionably traumatic etiology. Methods: From 2004-2008 we analyzed prospectively 3395 consecutive shoulder arthroscopies in regard to LHB associated lesions like SLAP and biceps Pulley lesions. The exclusion criteria SLAP type I lesion, total rotator cuff tears and history of shoulder dislocation left 182 cases with SLAP lesions (group I, n=138 male; 46 years ±12.6; 21-72) and 87 patients with Pulley lesions (group II, n=63 male; 49 years ±13.7; 18-81). Results: The overall prevalence of SLAP lesions greater than type I was 5.4%. In group I SLAP lesions type II according to Snyder and Maffet were present in 90 % of patients, type III lesions in 4%, type IV lesions in 4%, type V lesions in 1% and type VII lesions in 1% of patients. 34% of the SLAP lesions corresponded to type IIA, 9% to type IIB and 57% to type IIC according to Morgan. The overall prevalence of Pulley lesions was just under 3%. Type I biceps Pulley lesions classified according to Habermeyer were present in 44%, type II in 34%, type III im 13% und type IV in 9%. Isolated SLAP without Pulley lesions were present in 157 cases (86%), isolated Pulley without SLAP lesions in 62 cases (71%), a coincident presence was evaluated in 25 cases (10%, p=0.003). 65% of isolated biceps Pulley lesions and 34% of isolated SLAP lesions were associated with preoperative shoulder-trauma.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 The typical trauma mechanism for both lesions was the fall on the outstrechted arm in 42% of cases with isolated Pulley lesion and 17% of isolated SLAP lesions. A positive correlation of SLAP lesions and anterior shoulder instability (p=0.001) but a negative correlation of Pulley lesions and shoulder instability (p=0.004) was seen. An association of glenohumeral chondral lesions with SLAP lesions (p\0.01) but not with Pulley lesions was evaluated. Conclusions: We found out a negative correlation of the coincidence of SLAP and biceps Pulley lesions. Thus one can conclude that there is either a presence of SLAP or biceps Pulley lesion concerning the LHB tendon. The subsequence or coincidence of both lesions is statistically unlikely. The incidence of SLAP lesion in our study agrees with the literature, the incidence of Pulley lesions in our study is lower compared to the literature. This could be explained by more strict exclusion criteria excluding all total rotator cuff tears according to the common classifications of biceps Pulley lesion. Pulley lesions are significantly influenced by a shoulder associated trauma. In most of the cases the typical trauma-mechanism is a fall on the outstretched abducted arm. Traumatic biceps Pulley lesions are more frequent than previously thought. An anterosuperior impingement of the undersurface of the Pulley and of the SSC tendon against the anterosuperior glenoid rim is a possible etiologic factor obviously in combination with degenerative changes of the deep fibers of the SSC tendon caused by higher tension. This tension might be caused by repetitive micro-traumata as well as it can occur after an isolated macro-trauma. The humeral LBH chondral print as described in the literature could not be evaluated in the study in association with Pulley lesion but in association with SLAP lesions.

P10-932 Biomechanical evaluation of the LHB pressure on the humeral head with and without SLAP lesion in regard to glehohumeral chondral lesion Patzer T.1, Habermeyer P.2, Bobrowitsch E.3, Hurschler C.3, FuchsWinkelmann S.1, Schofer M.D.1 1 University of Marburg, Department of Orthopaedics and Orthopaedic Surgery, Marburg, Germany, 2ATOS Clinic Heidelberg, Shoulder and Elbow Surgery, Heidelberg, Germany, 3Hannover Medical School, Department of Orthopaedics, Hannover, Germany Objectives: Aim of this arthroscopically assisted biomechanical cadaveric study was to evaluate the stabilizing function of the long head of biceps tendon (LHB) and additionally the LHB load and pressure on the humeral head with and without SLAP lesion in regard to glenohumeral chondral lesions. This study was based on our hypothesis gained from a clinical observational study showing a significant association of SLAP lesions and glenohumeral chondral lesions in a typical localisation. Methods: 21 fresh frozen cadaver shoulders (average age 64 years, 24% male) without any higher grade of degenerative changes or total rotator cuff tears were used for the testings. The scapula, proximal humerus, shoulder girdle muscles and skin were intact. Humerus and scapula were cement imbedded and mounted in neutral position. LHB was connected via springs and pulley wheels with a force measuring sensor with 5 N (passivity-LHB-status) and 25 N preload (activity-LHBstatus). Testings were performed by a 5 axis industrial robot with a force moment sensor and 20 N joint compression and 50 N force in anterior, posterior, anterosuperior and anteroinferior direction. A iatrogenic SLAP lesion type IIC was arthroscopically created and a SLAP repair, a LHB tenotomy and a labrum refixation of SLAP complex were done. Testings were performed in three groups of 7 specimens with 4 different testing situations in each group with measurements in 0, 30 and 60 of abduction.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: A significant increase of anterior, anteroinferior and anterosuperior translation after SLAP lesion was evaluated whereas a significantly higher increase of translation under 5 N versus 25 N LHB-preload was found out showing that LHB acts as an active stabilizer with a maximum in 60 of abduction. The highest translation was found in anterior direction and 0 of abduction and in 60 of abduction in anteroinferior direction. SLAP lesions without LHB tenotomy showed a significantly higher increase of translation than isolated LHB tenotomy with an intact SLAP complex. The LHB load significantly increased more under active LHB preload (25 N) preload in comparison to the passive LHB preload (5 N). The highest increase of LHB load was measured in anterior, followed by anteroinferior and anterosuperior translation. A decrease of the LHB load was found for posterior translation. LHB load and translation could only partially be reduced with SLAP repair. Conclusions: Increased glenohumeral translation and LHB pressure on the humeral head in anterior direction after SLAP lesion leads to increased LBH chondral pressure on the anterosuperior quarter of the humeral head particularly under active LHB preload. Supporting our hypothesis this LHB instability associated chondral print can probably cause a humeral chondral lesion which was observed by us in a published clinical observational study. As known from several studies the increased glenohumeral translation can cause humeral chondral lesions of the anterior glenoid rim as instability osteoarthritis. The LHB chondral print can be avoided by LHB tenotomy for treatment of SLAP lesions but on the other hand glenohumeral stability is altered. Thus LHB tenotomy and refixation of the superior labrum might be the best treatment for SLAP lesions.

P10-980 Treatment of persistent shoulder pain with subacromial sodium hyaluronate injection. Preliminary results of a randomized controlled trial Abellan Guillen J.F.1, Gimenez Belmonte D.1, Melendreras Montesinos E.1, Ruiz Merino G., 3 Moreno Carillo M.A.. 2, Gil, E.1, Pen˜alver, F. 1 1 Hospital Morales Meseguer, Orthopedic Surgery, Murcia, Spain, 2 Hospital Morales Meseguer, Department of Rehabilitation, Murcia, Spain, 3Fundacio´n para la Formacio´n Sanitaria, Murcia, Spain Objectives: Persistent shoulder pain is a highly prevalent problem that is frequently associated with limited range of motion and decreased function. The purpose of this study is to evaluate the effect of subacromial injections of sodium hyaluronate in patients with pain shoulder, clinical subacromial impingement, and no rotator cuff tear. Methods: This was a randomized, double-blind (blinded observer), corticoid controlled study. It was designed to evaluate the efficacy of subacromial sodium hyaluronate injection in patients with persistent shoulder pain. Twenty patients were included in the study. They were randomized into two treatment groups receiving both a single subacromial injection, either corticoid solution (1cc triamcinolone acetonide + 1cc mepivacaine), or sodium hyaluronate (2cc). Patients were evaluated at baseline and at 1, 3 and 6 months. Demographic data were registered before stating the study. Entity of pain was evaluated with the use of the Visual Analogue Scale (VAS). Functional assessment was performed according to the Constant Score. Results: The demographic characteristics were similar in the two treatment groups. At baseline mean EVA was 6.29 (SD 1.54) and mean Constant score was 66.17 (SD 17.9). Both groups showed significant reduction from baseline in the VAS and an increase in his functional score within the first month. No significant changes were seen between first month and neither 3 nor 6 month assessment. Conclusions: These results show that Sodium Hyaluronate subacromial injections are an effective treatment option in patients with persistent shoulder pain and no rotator cuff tear. Compare to corticoid injection have shown same results at medium term follow-up. However, sodium hyaluronate injections can prevent the occurrence of corticoids-related complication, such as local degradation of tissues, tendon tearing, or

S135 arthropathy. Most of the improvement in functional score and pain relief is seen in the first 4 weeks, after that no significant changes are shown.

P10-981 Two-year results of the stemless eclipse prosthesis - a prospective study Heuberer P.1, Kriegleder B.1, Laky B.1, Anderl W.1 1 St. Vincent Hospital, Department of Orthopedics, Vienna, Austria Objectives: Several shoulder systems with different types of prostheses and fixation methods have been developed and evaluated in the past century to improve shoulder arthroplasty. One of the latest developments regarding humeral surface replacement of the shoulder is the stemless Eclipse prosthesis. To date, we are unaware of any prospective studies evaluating Eclipse prostheses. Therefore, our aim was to report two-year results of the stemless Eclipse prosthesis. Methods: Between September 2005 and October 2007, we implanted 63 Eclipse prostheses in a cohort of 60 patients. Patient’s clinical outcome using the Constant score was evaluated before, 12 months, and 24 months after surgery. Additionally, a single question reflecting patient’s satisfaction and radiological assessment was performed twice after the implantation of the Eclipse prosthesis. Results: The mean Constant score increased from 25 points preoperatively to 56 and 62 (p \ 0.001) points at the 1st and 2nd year follow-up, respectively. The range of motion of the flexion changed from 90 to 130 degrees (p \ 0.001), of the external rotation from 20 to 40 degrees (p\0.001), and of the abduction from 70 to 120 degrees (p\0.001) from before the implantation to the last follow-up. Further results showed that 32 (54%) patients were very satisfied, 20 (34%) satisfied, and 8 (12%) unsatisfied. However, over the two-year follow-up period, all prosthesis were radiological centred, and no mismatch or component loosening was observed. Conclusions: The promising two-year results suggest that the Eclipse shoulder prosthesis might be an expedient alternative to the existing resurfacing prosthesis. With the hollow screw fixation method it offers a higher bandwidth of indication than the common cup or stem prosthesis. Another advantage of the Eclipse prosthesis compared to anatomical shoulder prosthesis includes the less complex bone-sparing implantation method and therefore a shorter time of surgery. Keywords: Eclipse prosthesis; Stemless fixation; Shoulder arthroplasty; Glenohumeral arthritis; Bone-sparing implantation; Constant score.

P10-984 Engaging versus non-engaging Hill-Sachs defects within the functional shoulder range of motion Najarian R.1, Fening S.D.2, Jones M.3, Miniaci A.2 1 Ohio State University, Columbus, United States, 2Cleveland Clinic, Dept. of Orthopaedic Surgery, Cleveland, United States, 3 Cleveland Clinic, Cleveland, United States Objectives: The aim of this study was to investigate the effects of the size and orientation of a given Hill-Sachs defect on its ability to contact or ‘‘engage’’ the anterior-inferior glenoid rim at various arm positions within the functional shoulder range of motion. We hypothesized that progressively larger ‘‘engaging’’ Hill-Sachs lesions would engage the anteriorinferior glenoid rim at arm angles of less than 90 abduction and 90 external rotation, and ‘‘non-engaging’’ Hill-Sachs lesions would not engage the anterior-inferior glenoid rim within any functional range of motion, independent of its size. Methods: A 3D image of a normal proximal humerus and a normal glenoid was recreated from the Virtual Human data set. Hill-Sachs defects (‘‘engaging’’ and ‘‘nonengaging’’ orientations) were made by impacting the anterior rim of the virtual glenoid against the humeral head at varying angles. Factors recorded included humeral abduction angle (0 to 90, 5 increments), humeral rotation angle (45 internal rotation to 90 external rotation, 5 increments, humeral flexion angle (0 and 90) and defect size (small, medium large).

123

S136 Results: At a humeral flexion angle 0, large and medium sized ‘‘engaging’’ defects engaged the anterior glenoid rim at angles less than 90 abduction and 90 ER. Small ‘‘engaging’’ defects contacted the anterior glenoid rim only at 90 abduction and 90 ER. At a humeral flexion angle 90, all ‘‘engaging’’ defect sizes engaged the anterior glenoid rim throughout a greater range of humeral angles (Figure 1‘). For ‘‘nonengaging’’ defect orientations at a humeral flexion angle of 0, large and medium lesions engaged only the inferior glenoid rim at ER angles of greater than 70.

Conclusions: The results of this study have clinical implications when evaluating patients with Hill-Sachs defects. Larger ‘‘engaging’’ defects may engage the anterior glenoid rim at angles less than the traditional 90 abduction and 90 ER, causing a sense of instability that may be detected preoperatively. This may compel the surgeon to address the bony humeral head defect as well as the soft tissue injury at the time of surgery.

P10-993 The effect of arthroscopic Bankart reconstruction on quality of life, activity level and bone mineral areal mass Elmlund A.1, Kartus J.2, Ejerhed L.3 1 Danderyds Sjukhus, Department of Orthopaedics, Stockholm, Sweden, 2 NU- Hospital Organization, Trollha¨ttan/Uddevalla, Department of Orthopaedics, Trollha¨ttan, Sweden, 3NU-Hospital Organization, Department of Orthopaedics, Uddevalla, Sweden Objectives: Arthroscopic Bankart reconstruction and its significance for quality of life, activity level and the influence on bone areal mass (BMA) has previously not been thoroughly investigated. Methods: Patients with posttraumatic recurrent anterior shoulder instability scheduled for arthroscopic Bankart reconstruction were prospectively included in the study. Euroqol (5-15), 5 being the best and 15 the worst, was used to estimate quality of life, the activity level was measured using Tegner activity level (0-10) and BMA was measured in both calcanei using the dual energy X-ray absorptiometry (DXA) technique. The patients were assessed before surgery and after six and18 months. Results: 25 patients (9 female and 16 male) have been followed for 18 months. Quality of life according to Euroqol was 7 (5-10) before surgery and after 18 months 6 (5-9) (p=0.006). The Tegner activity level was 5 (1-9) before the injury, 3 (0-9) preoperatively, 3.5 (1-9) at six months (n.s.) and 3.5 (1-7) (n.s.) at 18 months. The BMA in both calcanei had decreased in female and male patients at18 months with 10-12% (p=0.003) and 6-8% respectively (p=0.04). Conclusions: Arthroscopic Bankart reconstruction rendered a better quality of life. The patients did not increase their activity level according to Tegner. Both female and male patients had a decrease in BMA of

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 between 6-12% at 18 months after surgery, which was more than the expected age related decrease. The surgical trauma induced the BMA decrease.

P10-1004 Atrophy and fatty degeneration of the supraspinatus muscle following arthroscopic rotator cuff repair Kusma M.1, Grub T.1, Kohn D.1, Steimer O.1 1 Saarland University Medical Center, Department of Orthopaedic Surgery, Homburg/Saar, Germany Objectives: The objective of this study was to determine the changes in atrophy and fatty degeneration of the muscles of the rotator cuff after arthroscopic repair. Methods: 20 patients with full thickness rotator cuff tears scheduled for arthroscopic rotator cuff repair were included. An MRI was done preoperatively and 6, 12, 26 and 52 weeks postoperatively following a standardized protocol. Tendon retraction (Patte), fatty degeneration (Goutallier), atrophy (Thomazeau) and tendon integrity were analyzed. Additionally, for correlation with the clinical results, preoperatively and 6, 12, 26 and 52 weeks postoperatively, the Constant and Murley Score (CMS) and the Simple Shoulder Test (SST) were evaluated. Results: Out of the 20 included patients, results of at least 6 months postoperatively are available for 15 patients to date. The CMS increased significantly from preoperatively 62.7 ± 25.5 to 94.8 ± 29.6 postoperatively (p=0.001). The SST improved significantly from 4.7 preoperatively to 9.1 postopertively (p\0.001). Patients with a limited retraction of the tendons (Patte 1) showed significantly better results in the CMS compared to patients with a higher degree of retraction (Patte 2-3) (118.8 ± 21.2 vs. 78.6 ± 23.8; p=0.01). A significant correlation between the preoperative diameter of the supraspinatus muscle and the postoperative CMS could be found (r=0.78; p=0.001), showing significantly better results in patients with less preoperative atrophy of the supraspinatus muscle. Between the preoperative MRI and the MRI after 52 weeks, an increase of atrophy was shown which was not statistically significant (p=0.09). In one patient, a retear was found in the MRI after 26 weeks. Conclusions: Rotator cuff repair is a successful procedure with respect to pain relief and functional outcome. However, the tear size and the muscle quality significantly influence the postoperative results. No reversibility of fatty degeneration and atrophy could be found. In contrast, despite good clinical results, an increase of the muscle atrophy was shown.

P10-1005 Diagnostic value of magnetic resonance imaging in post-traumatic anterior shoulder instability van der Veen H.1, Rijk P.1, Collins J.2 1 Medisch Centrum Leeuwarden, Orthopaedic Surgery, Leeuwarden, Netherlands, 2Medisch Centrum Leeuwarden, Radiology, Leeuwarden, Netherlands Objectives: Glenohumeral anterior instability due to a traumatic dislocation of the shoulder joint is a relatively common injury. Magnetic resonance imaging (MRI) in combination with arthrography is frequently performed in patients with traumatic shoulder instability as a diagnostic tool ahead arthroscopy. This prospective study was designed to evaluate the value of MRI after traumatic anterior shoulder instability and especially how far MRI findings correspond to findings at arthroscopy. Methods: Patients included had 2 or more shoulder dislocations, at least the first being traumatic. Exclusion criteria were previous shoulder surgery and skeletal immaturity. After ultrasound guided intraarticular introduction of contrast, MRI was performed. Images were scored and graded for the existence of Hill Sachs lesions, superior labral anterior posterior (SLAP) lesions, cuff tears, glenohumeral ligament (GHL) lesions, joint capsule tears and Bankart lesions. Consequently, shoulder arthroscopy was performed. The surgeon was blinded to the MRI outcome, scoring the

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 same list of anatomical structures. After surgery, MRI findings and observations during arthroscopy were compared. Interobserver agreement for detection of the several lesions was calculated by using Cohen’s Kappa coefficients (K). In cases K could not be defined, overall agreement was calculated. Results: Eighteen patients (13 male, 5 female) with anterior glenohumeral instability due to a traumatic dislocation were included. The average age was 26.1 years (range 15-50). The average time between MRI and arthroscopy was 13.1 weeks (range 5-65). Hill Sachs lesions demonstrated fair agreement (K = 0.33). Eight of 10 SLAP lesions demonstrated on MRI, were also observed at arthroscopy whereas no lesions were missed on MRI (moderate agreement, K = 0.43). On MRI, 4 partial thickness lesions were seen, which could not be demonstrated by arthroscopy. Glenohumeral ligament lesions were described on MRI in 15 patients, whereas only 2 patients turned out to have GHL lesions at arthroscopy. Overall agreement was 0.22. Two tears of the joint capsule were seen on MRI. These lesions could not be stated by arthroscopy. In 2 patients the joint capsule was redundant at arthroscopy, whereas in one of these patients the capsule was described as normal on MRI and the other was scored as being torn on MRI. All 13 Bankart lesions which were described on MRI where also observed at arthroscopy. On the other hand, 2 other arthroscopically diagnosed Bankart lesions which needed surgical treatment were not detected by MRI (moderate agreement, K = 0.47). Conclusions: In patients with posttraumatic anterior glenohumeral instability MRI shows many lesions that can not be confirmed by arthroscopy and therefore do not have therapeutical consequences. On the other hand some labral lesions which do need surgical treatment are not detected on MRI. Therefore, it can be concluded that MRI has no added value in the arthroscopical treatment of posttraumatic shoulder instability.

P10-1014 A prospective study of the interscalene nerve block for elective arthroscopic shoulder surgery: implications for early discharge Marsland D.1, Gooneratne M.2, Lubis N.2, Ahmed H.1, Chitre S.2 1 Whipps Cross University Hospital, Orthopaedics, London, United Kingdom, 2Whipps Cross University Hospital, Anaesthetic Department, London, United Kingdom Objectives: The interscalene nerve block is a widely used technique to provide analgesia for patients undergoing arthroscopic shoulder surgery. The audit aimed to investigate the analgesic effectiveness of the interscalene nerve block for elective arthroscopic shoulder surgery and the potential to reduce the length of hospital stay post operatively. Methods: Forty five patients (28 male and 17 female) with a mean age of 56 years, undergoing elective arthroscopic shoulder surgery were audited. Surgical procedures performed included subacromial decompression (20), rotator cuff repair (22), arthroscopic stabilisation and Superior Labral Anterior-Posterior (SLAP) repair. Interscalene nerve blocks were performed using either a nerve stimulator alone (22 patients) or in combination with ultrasound guidance (22 patients) before the administration of general anaesthesia. Patients were prospectively followed up over a twenty-four hour period and their analgesic requirements, pain score and any complications were recorded. Results: All patients were prescribed paracetamol as regular analgesia in combination with either a mild opiate or a non steroidal anti-inflammatory drug. The interscalene nerve block failed in three (7%) cases, all of which were performed by an inexperienced anaesthetist under nerve stimulator guidance only. The majority of patients (68%) were pain free in theatre recovery. 28 patients developed breakthrough pain on the ward at a mean of 12.5 hours (range 27 minutes to 20 hours, median=14) post operatively, of which 26 (93%) received oral morphine (mean dose 16mg). Local block complications included voice hoarseness (6 patients), profound arm weakness (9 patients) and ptosis (1 patient). All complications resolved within 24 hours. Conclusions: Successful interscalene nerve blocks provide good early analgesia for elective arthroscopic shoulder surgery with an acceptable side effect profile introducing the possibility of early patient discharge.

S137 However, a high proportion of patients in our study developed break through pain within the first twenty hours post surgery and were administered oral morphine. As to whether their pain warranted opiate treatment is unclear. This trend has implications for patients discharged home early, where strong opiates are not accessible.

P10-1054 Is it the age over 65 years a limit for arthroscopic rotator cuff repair in active and motivated patients? Osti L.1, Papalia R.2, Del Buono A.2, Maffulli N.3, Denaro V.2 1 Hesperia Hospital, Orthopedic Surgery, Modena, Italy, 2Campus Biomedico University, Orthopaedic Surgery, Rome, Italy, 3Keele University, Institute of Science and Technology in Medicine, Stoke on Trent, United Kingdom Objectives: We wished to compare the outcomes of arthroscopically repaired rotator cuff tears in 28 patients older than 65 years (the over 65 group) with a control group of 28 patients younger than 30 years (the under 65 group). Methods: All the patients were treated by rotator cuff repair and biceps tenotomy. Pre and post-operatively, each patient was evaluated for range of motion (ROM), shoulder score (UCLA), SF-36 self administered questionnaire. Results: The groups were similar in regard to sex distribution, surgical technique, and postoperative rehabilitation programs, but different in age. In the over 65 group the mean age was 70.1 years, while in the under 65 group the mean age was 57.6 years. Comparing pre vs postoperative status at a minimum 24 months follow-up, forward elevation, internal and external rotation, modified UCLA rating system scores, and SF-36 scores improved significantly in both groups, with no significant difference between the groups. At the last follow-up, the strength improved significantly in both groups, with better result in the under 65 group (P\0.001). The Popeye sign was detected in 13/28 (46.4%) of the patients in the over 65 group and in 11/28 (39.3) in the under 65 group (v=0.29; P=0.59). Conclusions: On the basis of our results in selected active patients older than 65 years, arthroscopic rotator cuff repair associated with biceps tenotomy can yield clinical and related quality of life outcomes similar to those of patients younger than 65 years.

P10-1073 Influence of preoperative acromial shape and size and location of rotator cuff rupture on outcome of surgery Ho¨fling I.1, Va¨a¨ta¨inen U.1, Sipola P.2, Niemitukia L.2, Kro¨ger H.1 1 Kuopio University Hospital, Orthopaedics and Traumatology, Kuopio, Finland, 2Kuopio University Hospital, Radiology, Kuopio, Finland Objectives: The purpose of the present study was to investigate the influence of acromial shape and rotator cuff tear size and location on longterm results of surgery (open rotator cuff repair). Methods: 81 patients (average age 58.3 ± 7.3 years) with signs and symptoms of rotator cuff rupture were studied. Diagnose was confirmed with ultrasound and MRI. Decision for surgery was based on severity of handicap for the patient who had undergone conservative treatment including physiotherapy without sufficient symptom relief. Preoperative parameters included range of motion (ROM), pain duration and intensity and BMI. Acromial shape was determined from x-rays using Bigliani’s classification (type I: flat, type II: curved and type III: hooked). Open rotator cuff repair and acromioplasty were performed and rotator cuff tear size and location recorded. At follow-up (5.8 years after surgery) ROM, pain intensity, BMI and UCLA-scores were evaluated. Results: Preoperative pain duration tended to increase from acromion I to III (9.6 ± 2.9 vs. 20.4 ± 28.5 vs. 24.0 ± 29.8 months) but differences were not statistically significant and no correlations with rupture size were found. Before surgery 61% of patients experienced pain at rest or night pain, 35% had pain at light work and 4% pain only at heavy work. At follow-up at an average of 5.8 ± 1.2 years after surgery 52% of patients

123

S138 were pain free, 44% had pain only at heavy work, 3% pain at light work and 1% pain at rest or night pain. No significant differences were found between the different types of acromion or rupture size or location regarding pre- or postoperative pain intensity. ROM increased significantly after surgery compared to preoperative values for flexion, abduction and internal rotation (average increase 42.7 ± 49.3, 54.1 ± 52.2 and 34.1 ± 26.3 respectively, p=0,000). Improvement in external rotation was less (5.2 ± 26.9, n.s.). Increase in ROM tended to be higher in type II acromions especially for flexion and abduction but these differences were not significant and showed no correlation with size or location of rupture. Pre- or postoperative ROM did not differ significantly between acromial types or location of rupture and there were no correlations with rupture size either. No significant differences between acromial types or location of rupture or correlations with rupture size were recorded regarding BMI or postoperative UCLA-scores. Conclusions: Most patients clearly benefited from surgery in terms of pain reduction and improvement of ROM. Preoperative acromial shape or size or location of rotator cuff tear did not influence pre-operative pain or ROM and did not have any effect on the amount of benefit from surgery regarding pain, ROM, BMI or UCLA-scores.

P10-1077 Effect of hot application on shoulder proprioception in normal individuals Unal A.M.1, Dilek B.2, Akseki D.3, Gu¨lbahar S.4, Akalin E.2, Pinar H.5 1 Ag˘rı Government Hospital, Orthopaedics, Ag˘rı, Turkey, 2Dokuz Eylu¨l University, Physical Therapy and Rehabilitation, Izmir, Turkey, 3Balıkesir University, Orthopaedics, Balıkesir, Turkey, 4Dokuz Eylu¨l University, _ Physical Therapy and Rehabilitation, Izmir, Turkey, 5Dokuz Eylul University, Orthopedics, Izmir, Turkey Objectives: Proprioception has been shown to be affected by several factors such as exercise, fatigue, injury, surgery, rehabilitation and bandage or splint application. Although it is commonly used in sports rehabilitation, little is known about the effect of hot on proprioceptive capability. The purpose of this study was to investigate the effect of hot application on shoulder proprioception in normal individuals. Methods: Eighteen women, 12 men, total 30 healthy volunteers with normal shoulders whose ages were between 22 and 29 (av. 26.2) were included in the study. Proprioceptive level was measured from the dominant extremity before and after hot pack application with the technique of both active and passive joint position sense by a isokinetic dynamometer. Same measurements were repeated one week later by using the same pack, this time without heating it. 00 neutral and 300 external rotation were selected as the starting positions, and the tests were done at 100 internal and 100 external rotation positions from these angles. The effects of hot pack application on active and passive joint position sense were tested. A total of sixteen comparisons were done according to these parameters. Statistical analyses were done by Wilcoxon Signed Ranks and T-tests. Results: Significant differences were noted (p\0.05) in only two of eight parameters when we compared both applications on the shoulder. However, five of the eight comparative parameters showed significantly decreased reproduction errors following hot application (p\0.05). Hot application minimized the reproduction errors in four of the four passive joint position sense testing, but in only one of the four active joint position testing. Conclusions: Results of this study supports that hot application increases the proprioceptive capability of the shoulder. These findings should be taken into consideration for the prevention and treatment of sports injuries.

P10-1078 Effects of using anchors which allow independent tightening of suture limbs in transosseous equivalent rotator cuff repairs on tendonto-bone contact properties compared to conventional dependent tightening after relaxation Haber M.1, Ball K.1, Demontfort D.2, Appleyard R.3

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 1

Southern Orthopaedics, University of Wollongong, Wollongong, Australia, 2Arthrocare Australasia, Sydney, Australia, 3Murray Maxwell Biomechanics Lab, Kolling Institute, St Leonards, Australia Objectives: It has been hypothesized that even distribution of contact at the tendon to bone (TTB) interface over the entire footprint of the tendon may improve the healing response. Previous studies have demonstrated the superior contact properties of transosseous equivalent (TOE) repairs over conventional double row repairs. This has resulted in an increasing popularity of this technique and new purpose designed anchors for TOE repairs. Recent studies have also demonstrated the loss of contract pressures due to tendon relaxation shows different results than at time zero. To our knowledge there have not been any studies which have looked at the effects of independent tightening of the suture limbs in TOE repairs compared to dependent tightening after relaxation of the repair has occurred. The aim of this study was to investigate the TTB contact properties of TOE repairs using the Opus Magnum2 (M2) and new generation Opus Magnum X (MX) arthroscopic instrumentation (Arthrocare Australia Pty Ltd). Methods: Simulated rotator cuff tears were created in 18mth old sheep shoulders by dissecting the infraspinatus tendon away from humerus at the TTB junction. Double-row parallel (DR-P) and double-row cross-over (DR-C) suture reconstructions were then performed using Opus Magnum2 (M2) and Opus Magnum X (MX) arthroscopic instrumentation (n=6 in each group). Real-time TTB contact properties (force, pressure and area) were measured with an electronic pressure transducer (I-Scan 6900, Tekscan, Boston, MA). The TTB contact properties where recorded and compared immediately after suture tightening (Peak) and after a 300 second tissue relaxation period (Relaxed). Data were analyzed using ANOVA (SPSS v13). Results: The MX system generated a greater ‘Relaxed’ contact pressure than the M2 for both DR-P and DR-C while the contact area was the same. The most important differences were noted with DR-C suturing where: • MX demonstrated a significantly greater ‘Peak’ force (p=0.002), ‘Relaxed’ force (p=0.030) and ‘Relaxed’ pressure (p=0.007) than M2. • MX tended towards a greater ‘Peak’ pressure (p=0.074) than M2. With the DR-P suture configuration, MX tended towards a greater ‘Relaxed’ pressure (p=0.052) than M2. Conclusions: The fundamental difference between the Opus Magnum2 (M2) and Opus Magnum X (MX) arthroscopic instrumentation is that the MX allows each suture limb to be tightened individually while with the M2 both suture are tightened together. The MX system therefore allows independent tensioning to allow constant pressure from each suture limb by the surgeon while with the more conventional M2 system there is no possibility to vary the suture tension independently. As was seen in this study, this differential in suture tension can affect DR-P and DR-C contact pressure with MX system demonstrating higher contact pressure than M2 for suture configurations. The greatest differential in contact pressure between M2 and MX was noted in the DR-C group. This would suggest that clinically, when performing TOE repairs, the MX system which allows independent tensioning would produce better TTB contact pressure. This may improve the healing response.

P10-1098 Assessment of footprint properties of rotator cuff repairs during the repair and upon passive arm movement after the repair: the effects of repair relaxation and movement in single-row versus transosseous equivalent repairs Haber M.1, Dolev E.2, Appleyard R.3 1 Southern Orthopaedics, University of Wollongong, Wollongong, Australia, 2Southern Orthopaedics, Even Yehuda, Israel, 3Murray Maxwell Biomechanics Lab, Kolling Institute, St Leonards, Australia Objectives: Re-tears following rotator cuff repairs may in part be due to inadequate tendon-to-bone (TTB) contact pressure during the initial healing phase when mobilization has begun. Awareness of the poor TTB

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 contact properties with single row (SR) repairs has resulted in attempts at improving contact area and pressure with double row repairs including transosseous equivalent repairs (TOE). Nevertheless there have been very few studies looking at these properties in a dynamic fashion, simulating what happens during the repair as relaxation of the repair occurs and in the early postoperative period when movement occurs before any healing have occurred. It was postulated that 1. Significant relaxation occurs after the repair has been performed and retightening may result in improved TTB contact on completion of the repair and 2. With SR repairs, which utilize a tension-band effect, early abduction may result in compromised TTB contact. TOE repairs may provide a more stable repair. The objective of this study was to assess 1. Relaxation of the repair and the effect of retightening and 2. The effects of movement on repairs. Methods: Simulated rotator cuff tears were created in the supraspinatus tendon of six fresh cadaveric human shoulders. A SRR was then performed using the OPUS AutoCuff System, creating 2 horizontal mattress sutures in the standard technique described in their literature. An I-Scan 6900 electronic pressure-sensor (Tekscan, Boston, MA) was placed between the supraspinatus tendon and bone and the sutures were tensioned with the shoulder in 0 degrees abduction. Measurements were made during initial tensioning, after the repair was allowed to relax for 60 seconds and then upon re-tensioning and locking. The arm was then rested for 300secs (final relaxation) before being passively moved twice through a range-ofmotion (0-90 degrees abduction, 0-45 external and 0-45 internal rotation) and finally returned to neutral. The contact force, pressure and area were recorded throughout each movement. The procedure was then repeated using two TOE techniques: with two parallel sutures (TOE-P) and TOE with a cross over suture pattern (TOE-C). Results: DR-P and DR-C demonstrating a higher peak contact pressure than SR after suture tightening However after a 300 second relaxation period there was no significant difference for all three suture configuration. TOE parallel and cross-over repairs demonstrated no significant change in mean TTB contact properties during abduction, external rotation and return to neutral when compared to the 300sec relaxation state. The SRR demonstrated a significant drop in contact force on abduction (-63%), and return to neutral (-43%) and a trend on external rotation (-34%). Conclusions: Consistent tendon-to-bone (TTB) contact during the healing phase is thought to be vital. The electronic pressure sensor is useful assessing the repair dynamically during the repair and when the arm is moved, simulating the postoperative rehabilitation. We confirmed 1. Relaxation occurs over the first 300s after the repair is tightened and retightening at 60s significantly reverses the relaxation and 2. Significant decrease in contact area occurs with SRR during movement compared to the TOE repairs. Checking and retightening repairs where possible improves the resting contact properties and TOE repairs provide a more stable repair when movement occurs.

P10-1113 The technique of a modified Hill-Sachs remplissage in arthroscopic revision shoulder instability surgery Kuhlee U.1 1 Center of Shoulder Surgery, Berlin, Germany Objectives: The treatment after failed instability surgery is a challenging problem . Advancements in arthroscopic techniques over the last years led to an increasing trend towards arthroscopic treatment of revision anterior shoulder instability. There a different factors contributing to failed instability repair. One is an engaging Hill-Sachs lesion. E. Wolf introduced the arthroscopic Hill-Sachs remplissage. We developed a modification of this technique to avoid a marked limitation of movement postoperatively. Methods: We made a prospective study on 16 patients with failed instability surgery.There were different previous operations (1-3). We used our

S139 standardized arthroscopic instability repair with a modified Hill-SachsRemplissage. The technique is comparable to a transtendon pasta repair. The technique is shown in a video. Results: The early follow up of this ongoing study is 12 month. We have one redislocation due to an epileptic seizure. We have no marked limitation of movement. The Rowe score increased from 40 to 90 points. 15 patients are very satisfied. Conclusions: This minimal invasive technique is one possible option to treat revision cases and can be used as an additive in arthroscopic anterior instability repair.

P10-1122 The Pretzel knot; a new simple locking slip - knot Karahan M.1, Akgun U.2, Mendes J.E.3 1 Marmara University, Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 2Acibadem University Faculty of Medicine, Orthopedics and Traumatolgy, Istanbul, Turkey, 3 Minho and Porto University, Sau´de Atlaˆntica Clinic, Orthopedics and Traumatolgy, Porto, Portugal Objectives: The ability to successfully tie a secure arthroscopic knot is among the most essential techniques for the arthroscopic surgeon. The success of an arthroscopic knot depends on multiple factors such as knot configuration, suture type, tissue quality and the surgeon0 s competency. An ideal arthroscopic knot must have minimum requirements such as easy sliding, slack free configuration and satisfactory knot security. In addition, a simple configuration with a low profile is appreciated. Many arthroscopic knots have been described in the literature with good, optimal and suboptimal results regarding to the individual requirements listed above. In this paper, a new locking slip - knot ‘‘ Pretzel ‘‘ that can fulfill these requirements with a simple, reproducible knot configuration and a minimum profile is described. Methods: Once the anchor is placed in the bone and the desired limb(s) is/are passed through the tissues, suture limbs coming out of the cannula are separated into the short and the long limb. With the short limb on the left hand, a right overhand halfhitch is done. Then a right underhand loop is initiated in a conventional fashion As soon as the loop limb’s end revolves under the post limb the end of the suture should go through the loop of the previous overhand halfhitch rather than the loop its own. At this point the slip-knot looks just like a pretzel. By gentle manipulation of the loop limb, the slack between these two half - hitches is removed. The loop limb is then pulled in line with the post limb. Once the initial slip knot consisting of two half-hitches are snugly placed on top of each other, the post limb is pulled until the initial slip knot slides until it is seated on the tissue. One can adjust the pulling force based on the desired tissue tension. With the pull of the loop limb, the internal locking mechanism strangles the post limb in three points thus causing it to twist around the loop limb, which is similar in effect to the switched-post, thereby locking the post limb. After completion of the initial slip knot, a constant gentle pull should be applied to the new post limb and two more reversed half - hitches on alternating posts should be done. Results: An arthroscopic knot with simplicity, easy reproducibility and strong initial security can reduce failure rates. Many arthroscopic nonlocking and locking slip knots have been described to achieve these goals. The ‘‘Pretzel Knot ‘‘ is a locking slip knot with a simple configuration. The configuration consists of an overhand half-hitch followed by an underhand half-hitch. It is almost like a square knot except that the second throw goes through the loop of the first throw which locks the knot by pulling the loop limb. Evidently, existence of two half-hitches only, result in easy sliding of the knot with minimum internal friction in addition to having low profile. Final advantage of the Pretzel knot is; it is easy to describe, teach and learn. Conclusions: Generally simplicity comes alone with efficacy. In this sense and remembering all the criteria for an effective technique in a difficult domain like the shoulder, the simplicity of the ‘‘Pretzel’’ knot may promote a successful outcome by reducing small failures related with the difficult configurations of some knot-tying techniques.

123

S140

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: The mean Constant score was 88,7 (range 81-97) points (healthy side average 92,3 points). The SSV averaged 92,4 (range 70-100)%. The mean coracoclavicular difference between operated and healthy side averaged 3,4 (range 0-7) mm. Clinical sign of posterior instability was present in 9/16 (56,3%) patients. Skin irritation over the clavicle button was present in 3/16 (18,8%) patients. Conclusions: Acromiocalvicular joint reconstruction and stabilization can be performed safely and effectively with arthroscopic technique. Despite the presence of posterior instability and subluxation of the acromioclaviclavicular joint good to excellent clinical results were observed. Patients returned to preinjured activity level. Procedure gives good cosmetic results.

P10-1145 Comparison of general anesthesia and interscalene block in arthroscopic shoulder surgery Park K.J.1, Kim Y.M.1, Kim D.S.1, Choi E.S.1, Shon H.C.1, Cho B.K.1, Seung Hwan B.2 1 Chungbuk National University Hospital, Department of Orthopedic Surgery, Cheong-ju, Korea, Republic of, 2College of Medicine, Chungbuk National University, Department of Orthoprdic Surgery, Cheongju, Republic of Korea Objectives: After Shoulder arthroscopy, patients complained of severe pain mainly during the early period. As for pain control, only Patient controlled analgesia has been performed in most cases; thus, it was experienced that there was no effective pain control. Therefore, the effectiveness of pain control during the early period with IB for shoulder arthroscopy was investigated in comparison with GA. Methods: A prospective randomized controlled test was conducted on 60 patients who received Shoulder arthroscopy from September 2008 to March 2009. The cases were divided into 30 cases that received surgery under IB (Group 1, ages from 24 to 56), and 30 cases that received surgery under GA (Group 2, ages from 30 to 62). Then, the difference in patients0 satisfaction was compared and assessed by analyzing the scale of pain for 3 days before and after surgery. The scale of pain before surgery and at 1 hour, 4 hours, 8 hours, 12 hours, 48 hours, and 72 hours after surgery was measured using VAS. Concerning the scale of pain before surgery, the scale in which pain was the most severely felt during joint exercise was measured. For after surgery, the scale of pain that was felt during rest was measured. Results: There was no significant difference in pain before surgery between the two groups, For Group 1, pain was significantly lower than that of Group 2 immediately after surgery, and in no case did the severity of pain become aggravated during the 3-day observation period. Regarding Group 2, severe pain was felt immediately after surgery, and the severity of pain has decreased over time. With regard to major complications related to anesthesia, no such complications were discovered in Groups 1 and 2. Group 1, which showed significantly lower pain after surgery, were able to walk and eat immediately after surgery; thus, patients0 satisfaction was found to be higher. Conclusions: Concerning shoulder arthroscopy, the severity of pain after IB was significantly lower immediately after surgery and during the 3 days after surgery when compared to shoulder arthroscopy that was given under GA. After IB, shoulder arthroscopy caused less emotional stress on patients from anesthesia than GA, and further, the patients0 satisfaction was also higher because they were less affected in their whole body and were able to walk after surgery. Also, it is considered to be an effective anesthesia method for pain control after surgery.

P10-1156 The infraspinatus tenodesis and posterior capsulodesis (remplissage) in cases with humeral bone loss (hill sachs defects) - preliminary results Mataragas E.1, Tzanakakis N.1, Vassos C.1, Yiannakopoulos C.1, Antonogiannakis E.1 1 IASO General Hospital, Centre for Shoulder Arthroscopy, Holargos Athens, Greece Objectives: This paper aims to evaluate the Remplissage arthroscopic technique as described by Eugene Wolf used in patients with traumatic shoulder instability that present glenoid bone loss and Hill Sachs defects. Methods: In our study 31 patients (5 women and 26 men) with mean age of 30 yrs underwent arthroscopic stabilization of the shoulder by the same surgeon during January 2007- March 2009 period. All patients presented Hill Sachs lesion, 11 of them had medium or large glenoid bone loss, 10 had an ‘‘inverted pear’’ glenoid shape, 4 had been revised for stabilization in the same shoulder and 15 presented joint hypermobility. Mean age for the age of 1st dislocation was 19.6 yrs and our follow up ranged from 7-32 months (Mean=22). The recurrence of instability and the functional outcome were evaluated pre-op and post-op with the Rowe Zarins Score. The post op rehabilitation was performed by a rehabilitation doctor dedicated to shoulder pathology. Results: None of the patients presented recurrent instability. The Rowe Zarins Score raised from a mean pre op score of 24,79 (15-60) to a mean post op score of 97,03 (75-100) (p\0.05). All the patients that were into sports activities before the presentation of shoulder instability began training again and our post op evaluation of the shoulder’s ROM showed a decrease in the external rotation from 0o-15o. Conclusions: The infraspinatus tenodesis and posterior capsulodesis in patients with humeral bone loss seems to offer so far excellent post op results despite the slight decrease in the external rotation of the shoulder.

P10-1146 Results of arthroscopic stabilization of the acute acromioclavicular joint dislocations Kovacic L.1, Senekovic V.1 1 University Medical Centre Ljubljana, Department of Traumatology, Ljubljana, Slovenia Objectives: Despite the common occurence of acromioclavicular joint dislocation no ideal surgical technique is known. Early and late postoperative complications are common including breakage of the implant, migration, loosening, clavicular erosion and postoperative arthritis. The purpose of this study was to evaluate the clinical and radiological results after arthroscopic acromioclavicular joint stabilization. Methods: Sixteen consecutive patients (3 females, 13 males; mean age 38,2 years, range 16-57) with acute acromioclavicular joint dislocation Rockwood V were treated with arthroscopic acromioclavicular joint stabilization using ThightRope technique with the first generation implant. The time interval from trauma to surgery averaged 11,5 (range 4-22) days. Results were evaluated after a minimum follow-up 12 months.

P10-1163 The role of acromioclavicular joint resection in arthroscopic subacromial decompression Vassilev I.1, Jontschew D.1 1 Southwestern German Sporttraumacenter Uhingen, Uhingen, Germany Objectives: Subacromial decompression is the most frequently arthroscopic shoulder operation. With increasing of the number of operations the failure rate without pain release and continuous functional deficit is increasing too. The most common reason for treatment failure is the untreated ac joint pathology and secondary impingement. Since january 2004 we increased the number of combined arthroscopic operation subacromial decompression and distal clavicle resection (ac joint resection), almost in all of the shoulders and observe significant decrease of failed treated cases. Methods: From january 1999 to december 2003 we did an arthroscopic decompression of 326 shoulders (212 male and 114 female) with main age 62,5 years (44-81 y.). In our study were included patients with simple impingement syndrome without other articular pathologies (rotator cuff lesions, labrum pathology, severe omarthrosis, postfracture impingement, ac joint instability). There were two groups of patients - group A (182

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 shoulders) included patients with combined operation - subacromial decompression and resection of the distal clavicle end and group B (144 shoulders) - patients with only subacromial decompression. All the patients had preoperative MRI and such with other shoulder pathology were not included in the study.We performed an arthroscopic examination of the shoulder and the subacromial space. A typical subacromial decompression through a lateral portal was performed in all of the patients. In those from group B we did an arthroscopic resection of the distal clavicle end. For this procedure was used an anterosuperior portal right under the front edge of the acromion. The resection was performed carefully without destroying of the cranial acromioclavicular ligament and the capsule, because of preserving the cranio-caudal stability of the acromioclavicular space. The size of the resection was measured via shaver blade size (two widths of shaver blade 4,5 mm) and was 9-10 mm. After the surgery a standard protocol of physical therapy was used. The patients were able to come back to work and sports between 4-6 weeks. Results: All of the patients were clinical examinated 3 moths after surgery and a radiological examination was performed 12 months postoperatively. In the A group 70% of the patients were painless, in group B 92,5 % have excellent and good results, (Rowe scala. At 5 from the group A patients was performed another arthroscopic operation with distal clavicle resection, in the group B - only by one. Conclusions: Our results encouraged us to do an distal clavicle resection in almost all of the patients with impingement syndrome. Only in young active patient shell be done only a separation of the coracoacromial ligament.

P10-1209 Anterior bilateral shoulder dislocation after convulsive crisis - case report Vic¸oso Sousa Fernandes S.I.1, Barbosa A.T.1, Fraga Ferreira J.1, Cerqueira R.1, Caetano V.1, Loureiro M.1 1 Centro Hospitalar do Alto Ave, Orthopedics and Traumatology, Guimara˜es, Portugal Objectives: Bilateral anterior shoulder dislocation is rare, and his aetiology is via various traumatic insults, atraumatic occurrences, and through extreme muscular contractions like epilepsy. In epileptic seizures is more common to occur posterior bilateral dislocation. Methods: The aim of this work is to describe a rare case of anterior bilateral shoulder dislocation after a convulsive crisis. Results: Case report: It concerns a case of a 35-year-old male, with alcoholism history, who entered the emergency room in 25/05/08 with a generalized tonic-clonic seizure. After, he had bilateral shoulder deformity and swelling. Radiographs demonstrated a bilateral anterior shoulder luxation and bilateral greater tuberosity fracture. The dislocation was reduced and both shoulders were immobilized. 1 month later, radiographs showed bilateral reduction maintenance and bilateral greater tuberosity fracture deviation. The patient had extremely restriction of active and passive ranges of motion in both shoulders: in the left had 58 of active external rotation and 608 of abduction; in the right 08 of active external rotation and 508 of abduction. At this moment surgical procedure was done with bilateral open reduction and osteosynthesis with ‘‘phylus’’ plate and was orientated to physical rehabilitation. At the 2 month follow up, he had significantly improved both shoulders range of motion, and returned to the normal daily activities and 2 months later returned to work. Conclusions: Displaced fractures of the greater tuberosities after shoulder dislocation may result in motion limitation and functional disability. Open reduction and stable fixation allows for early passive motion of the joint and early return to activities of daily living.

P10-1230 Arthroscopic treatment of acromioclavicluar joint dislocation using TightRopeÒ Park K.J.1, Kim Y.M.1, Kim D.S.1, Choi E.S.1, Shon H.C.1, Cho B.K.1, Lee H.C.1

S141 1

Chungbuk National University Hospital, Department of Orthopedic Surgery, Cheong-ju, Korea, Republic of Objectives: Surgical reconstruction of the coracoclavicular(CC) ligament is a fundamental part of management of high-grade acromioclavicular dislocations. Arthroscopic reconstruction with TightRope is increasing nowadays. So we want to describe the methods to prevent complications in arthroscopic treatment of acromioclavicluar(AC) joint dislocation using TightRope. Methods: We reviewed twelve patients with a mean follow-up 8 months (range 6-14 months) underwent the arthroscopic acromioclavicular joint reconstruction using Tightrope between March, 2008, and March, 2009. The indications for surgery included acromioclavicular joint dislocation Rockwood type V. Results: All twelve patients returned to their work without pain within 3 months of the operation. The average Constant score and KSS scores at the last follow-up was 98 and 97. Two patients showed failures of TightRope fixation on the coracoid side and the acromioclavicular joint was redislocated. The reason for failures of fixation was non-centered drilling in the coracoid process, and the thin wall was fractured during follow up. To prevent this complication, we recommend using a C-arm after guide pin fixation in the coracoid. Another method to simplify drilling the center of the coracoid is to reduce the AC joint first and fix it temporally by K-wire. After TightRope fixation, the K-wire was removed. A 70 degree scope is also helpful to see the base of the coracoid. Six patients show the sinking of the superior endobutton into the clavicle. But patients did not complaint of any discomfort, however we worried about clavicle fracture. After confirming the healing of CC ligament by MRI, we removed the tightrope. In the operation we could see the defect in the clavicle. There was no widening of the CC gap after removal. Conclusions: Considering its lower morbidity, less hospitalization, excellent cosmesis and early rehabilitation, this new technique offers an attractive alternative in acromioclavicular joint stabilization. However, surgical complications can deter use of TightRope. We recommend using this technique and removal of the implants to prevent complications.

P10-1260 Latarjet procedure in the treatment of anterior shoulder instability a 16 years follow up study Chillemi C.1, Garro L.1, Godente L.1, Tucciarone A.1 1 Istituto Chirurgico Ortopedico Traumatologico, VII Divisione, Latina, Italy Objectives: A variety of operations have been proposed for the treatment of anterior shoulder instability. The aim of the present study was to evaluate the functional results of the Latarjet procedure and the prevalence of glenohumeral OA after this intervention. Methods: Forty patients (33M, 7F; mean age: 26 yrs) were included in the study with a minimum follow-up of 16 years. All the patients were clinically evaluated according to the system of Rowe, Duplay and Constant and with X-ray (AP with humerus in neutral and internal rotation, and lateral view). Results: At FU none of the patients had recurrent dislocation, two patients reported occasional subluxation, 1 case had a positive apprehension test. 26 patients returned to the same sports, 8 patient changed it. Pre-op radiographs demonstrated a bony lesion in 37 cases (90%) lesion of the glenoid rim (fracture or erosion of the anterior aspect), posterosuperior defect of the humeral head, 1 case a Samilson grade 1 degeneration. Post-op radiographs showed a correct positioning of the coracoid graft in 32 cases, too lateral in 5 and too medial in 3 cases. At the latest FU in 6 cases was detected a non-union and in 2 cases a partial resorption of the coracoid graft. Conclusions: Our data confirm that Latarjet technique is an efficient procedure in chronic anterior shoulder instability. Shoulders stabilised with this procedure function satisfactorily over time, so to allow to more than 80% of our patients to return to sport activities. Glenohumeral OA developed 16 years after surgery in 11 cases.

123

S142 P10-1287 Arthroscopic treatment of acute traumatic AC joint dislocation Heikenfeld R.1, Listringhaus R.1, Godolias G.1 1 St. Anna Hospital, Center for Orthopedics and Traumatology, Herne, Germany Objectives: The purpose of this study was to evaluate the results after arthroscopic treatment of traumatic AC joint dislocation using a Bosworth screw. Methods: 39 Patients with acute AC Joint dislocation type Rockwood 3 were arthroscopically treated with temporary transfixation using a 7.0mm cannulated titanium screw of the clavicle to the coracoid process. The coracoid process is arthroscopically visualized and a drill guide for tibial anterior cruciate ligament positioning is used to exactly place the screw into the coracoid process. The screws were removed after 8 weeks. Patients were followed using a prospective study using the Constant Score after 3, 6, 12 and 24 months. Results: 37 Patients were completely evaluated. One screw slipped out of the coracoid process 3 days after surgery requiring revision surgery. No screw breakage was observed. There were no other operation conditioned complications. Constant score showed a mean of 94,7 at last follow up. At follow up, no patient hat a redislocation without weight bearing. With 10kg weight a mean clavicular elevation of 1,8mm was observed. All remaining patients were satisfied with the functional and cosmetic result. Conclusions: There is some controversy about the surgical treatment of acute traumatic AC joint dislocation type Rockwood 3. Most open surgery techniques have the disadvantage of a poor cosmetic result or a difficult and dangerous hardware removal, because the scar of the AC joint capsule that is supposed to stabilize the clavicle has to be opened. The Bosworth screw technique does not touch the AC joint at all, but the open procedure has poor cosmetic outcome. It is also important to use a large screw to ensure proper hardware stability to avoid hardware failure. Our technique might be an alternative for the operative treatment of acute AC joint instability, because it is safe and all anatomical structures remain intact in case revision surgery with i.e. arthroscopic AC joint resection and ligamentoplasty is necessary.

P10-1296 Postoperative cryotherapy after acromion resections Kise N.1, Ekeland A.2 1 Martina Hansens Hospital, Orthopedic, Baerum, Norway, 2Martina Hansens Hospital, Bærum, Norway Objectives: Celcus (53 b.C. - 7 a.C.) described five cardinal signs of inflammation; calor, rubor, tumor, dolor and functio laesa. Empiri has taught us that decreasing the inflammatory processes increases healing and limits symptoms after trauma. Acute traumas are today treated by the RICE principle; R/rest, I/ice, C/compression, and E/elevation. After surgery, rest, compression, elevation and to some degree cryo therapy are recommended. The aim of this study was to evaluate the effect of cryo therapy after scopic resections of the acromion. Methods: In this prospective study, 60 patients, 20 men and 40 women, aged 54 years (35-71), operated with scopic resections of the acromion, were randomized to postoperative treatment with or without IceBand cryo bandage. Thirty patients had cryo bandage (Cryo group), and 30 patients had conservative treatment (Conservative group). The groups were comparable with respect of the patients’ age, gender, operating time, and additional pathology in the ipsilateral shoulder. IceBand is an adjustable cryo bandage for the shoulder. It is made of synthetic materials and the cryo element contains water and is chilled in the freezer. The cooling effect lasts for 40-60 minutes after application. The patients in this study used the IceBand cryo bandage 4 times a day the first 4 postoperative days. In this period of time we recorded their need for analgesics (paracetamol, codeine) and their subjective level of pain, measured by visual analogue scale (VAS). Quick-DASH was recorded preoperatively and 3 weeks and 3 months postoperatively. We also recorded duration of sick leave from work.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: The patients in the cryo group had significant lower mean pain score measured in VAS the first 4 postoperative days (p = 0.03). The patients in the cryo group had a tendency to lower need for analgesic drugs the first 4 postoperative days (not significant). Preoperatively and after 3 months there was no difference in function outcome measured by QuickDASH between the two groups, but after 3 weeks there was a slight tendency toward better function outcome in the cryo group. There was no difference between the groups in duration of sick leave. There were no complications related to the cryo bandage. Conclusions: Patients who got postoperative cryo bandage treatment, had significant lower pain score the first 4 postoperative days, and they had tendency toward lower need for analgesic drugs. There was a tendency toward better function scores after three weeks in the cryo group.

P10-1297 Arthroscopic reconstruction of isolated tears of M. subscapularis results after 24 months Heikenfeld R.1, Listringhaus R.1, Godolias G.1 1 St. Anna Hospital, Center for Orthopedics and Traumatology, Herne, Germany Objectives: The purpose of this study was to evaluate the results of arthroscopic reconstruction of isolated lesions of M. subscapularis. Methods: 15 shoulders (14 men, 1 woman, mean age 42 y) with an isolated lesion of M. subscapularis were treated with arthroscopic repair. We used 1 to 3 suture anchors in single or double row technique. 13 patients had an adequate trauma. All patients had preOP MRI and ultrasound and at last follow up. Patients were followed using a prospective study at 6, 12 and 24 months using Constant and UCLA score. Results: 14 cases were completely evaluated. 7 shoulders had a type 2 lesion according to Lafosse classification, 5 type 3 and 2 cases showed a complete tear (type 4). Fatty infiltration was not greater than type 2 according to Goutallier. 8 cases with instability of LHB were treated with biceps tenodesis. All patients improved in scores during follow up (Constent Score 40,3 to 80,4; UCLA 15,6 to 31,6). One patient needed an arthroskopic arthrolysis due to decreased ROM after 4 months. 13 patients were satisfied with the postOP result. MRI showed 2 retears of M. subscapularis and 2 failures of biceps tenodesis. Biceps pathology, fatty infiltration, size of the tear and age did not influence the postOP result. Conclusions: Arthroscopic repair of isolated lesions of M. subacapularis is a technically demanding arthroscopic procedure that leads to good results after 24 months. They are comparable to open reconstruction.

P10-1314 A biomechanical analysis of a new arthroscopic locking slip knot ‘Pretzel’ compared with Nicky‘s and SMC knots Akgun U.1, Nuran R.2, Turkoglu A.N.3, Ates F.3, Yucesoy C.3, Karahan M.4, Turkmen M.5 1 Acibadem University Faculty of Medicine, Orthopedics and Traumatolgy, Istanbul, Turkey, 2Acibadem Hospital, Istanbul, Turkey, 3 Biomedical Institute of Biomedical Engineering of Bogazic¸i, Istanbul, Turkey, 4Marmara University, Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 5Acibadem University Faculty of Medicine, Orthopaedic Surgery, Istanbul, Turkey Objectives: ‘Pretzel‘ is a new arthroscopic locking slip knot whic has a simple and easy to learn configuration. The purpose of this study is to evaluate the biomechanical properties of a new arthroscopic locking slip knot, ‘Pretzel‘ knot and compare it with two knots taken as standard: Samsung Medical Center (SMC) and Nicky’s knots. Methods: All 3 knots were tested in 2 different setup. In the first setup, 3 knots were prepared with out any RHAPs, and in the second setup all knots were prepared and then secured with 3 RHAPs. Ethibond (Ethicon, W4843) 2/0 sutures were used in the study. Biomechanical tests were completed with a computer guided tension device. (LF plus, Lloyd, UK) All knots were tied on two metallic hooks hold by the tension device, by

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 the same shoulder surgeon who has no previous experience in any of the 3 knots. Knots have been tested respectively with load to failure test and cyclic loading test. In load to failure test, knots were tensioned until failure with a speed of 1.25 mm/sec . In cyclic loading test, knots were pretensioned by 7 N at a speed of 40mm/min in order to remove any slack. After pretensioning, all knots were loaded in a cyclic manner between 7 and 30 N at a speed of 12mm/sec for 200 times. Parameters considered for both groups were for loop security, maximum elongation, permanent loosening, load at failure and elongation at failure. Knot tying times were also recorded. Results: In the first setup (without any RHAPs), Nickys knot which is a non locking slip knot has been excluded. In the cyclic loading test, there was no significant difference in maximum elongation of the Pretzel and SMC knots. (ANOVA test) For evaluation of the initial loop security, the initial displacements of the SMC and Pretzel knots were recorded until 7N pretension. Pretzel knot showed significantly less initial displacement. (p \0.05) In load to failure tests, there was no significant difference between ultimate load to failures of SMC and Pretzel knots. At the ultimate load, all knots were failed by loosening. (ANOVA test) In the second setup (with 3 RHAPs); in the cyclic loading test, there was no significant difference in maximum elongation of SMC, Nickys and Pretzel knots. (ANOVA) For evaluation of the initial loop security, the initial displacements of the SMC, Nicky‘s and Pretzel knots were recorded until 7N pretension. There was no significant difference in the initial displacements. In the ultimate load to failure tests there was no significant difference between SMC, Nickys and Pretzel knots. (ANOVA) All knots were failed by suture break. There was no significant difference in the knot tying times. Conclusions: The new arthroscopic locking slip knot ‘Pretzel‘ which has a simple and easy to learn configuration has similar biomechanical properties when compared with SMC and Nicky’ s knots.

P10-1318 Long term follow up of patients arthroscopically treated for shoulder instability Mataragas E.1, Tzanakakis N.1, Vassos C.1, Chiotis I.1, Antonogiannakis E.1 1 IASO General Hospital, Centre for Shoulder Arthroscopy, Athens, Greece Objectives: The evaluation of the results obtained after a long term follow up (over 60 months) from patients that were treated arthroscopically for shoulder instability. Methods: In our paper we evaluated 112 patients (104 men and 8 women) with mean age of 24 years, that were treated surgically by the same surgeon from 1999-2004. Mean age of first dislocation was 19 years and mean number of dislocations was 13. Seventy seven (77) of them (66,4%) were into sports activities and during pre op clinical examination 15 patients (12,9%) were diagnosed with joint hypermobility syndrome taking into account the Beighton criteria. Arthroscopic findings showed that 80 of them (68,9%) had some kind of bone loss, either glenoid (7 Large, 23 Medium, 6 Small) or Hill Sachs lesion (28 Large, 30 Medium, 20 Small) and in 8 patients an ‘‘inverted pear’’ glenoid shape was found. Our follow up ranged from 60-117 months (Mean=84) and the recurrence of instability and functional outcome were evaluated post-op using the Rowe Zarins Score. Results: 22 patients were lost to follow up (19.64%). Recurrent instability presented in 7 patients (6.25%). Five (5) of them was due to high energy accidents, one was due to non-compliance and one was due to minor injury. Of these patients 5 presented Hill Sachs lesion, 3 showed glenoid bone loss (2 Large, 1 Small) and in none of them an ‘‘inverted pear’’ glenoid shape was found. All recurrent cases were into some kind of Overhead/Contact sports activity (6 Amateur, 1 Professional). The post op Rowe Zarins Score ranged from 80-100 (Mean=95,53). Conclusions: Arthroscopic treatment of glenohumeral instability is an excellent method that provides comparable results to the open surgical treatment and the results do not deteriorate with longer follow up.

S143 P10-1331 Shoulder arthroscopy: intra-operative events and early complications Tzanakakis N.1, Mataragas E.1, Mouzopoulos G.1, Hiotis I.1, Antonogiannakis E.1 1 IASO General Hospital, Centre for Shoulder Arthroscopy, Athens, Greece Objectives: To evaluate the frequency and types of intra-operative events and early complications during shoulder arthroscopy. Methods: In a prospective study 177 consecutive shoulder arthroscopies were recorded from May 2008 to July 2009 (15 months period). All operations were performed from the same team, in lateral decubitus position. Sixty-nine females and 108 males, mean age 48 years (range:1582) were arthroscopically operated for: Rotator Cuff tear:100 (56.5%), Instability:49 (27.7%), Frozen shoulder:12 (6.8%), Calcifying tendonitis:11 (6.2%), SLAP:3 (1.7%), Early Osteoarthritis:1 (0.6%), Lavage:1 (0.6%). Total number of implanted anchors: 634 (all double loaded), total number of side to side sutures placed: 98. Results: Intra-operative events: Knot Loosing: 3 out of 1342 knots (0.22%), Anchors pulled-out: 4 out of 634 implanted anchors (0.63%), Anchors breakage: 3 out of 634 implanted anchors (0.47%), Suture breakage: 5 out of 1342 sutures (anchors sutures and side to side sutures) (0.37%), Suture slippage from anchor or tissue: 5 out of 1342 sutures (0.37%), Nitinol needle breakage: 4 out of 100 rotator cuff repairs (4.0%), Instrument breakage: 2 out of 177 arthroscopies (1.13%). Early Complications (3 months post-op): totally we recorded 4 early complications: (2.26%). one anterior interoseous nerve paresis that have fully recovered 6 weeks post-op, one motor and sensor ulnar nerve paresis that has not fully recovered 4 months post-op, one sensor ulnar nerve paresis that has fully recovered 7 months post-op and one septic arthritis that was treated with arthroscopic lavage and antibiotics and has no recurrence 14 months post-op. Conclusions: Shoulder arthroscopy is a safe operation with low early complication rate (2.26%). Draping in lateral decubitus position can cause neurological complications. Intra-operative events may occur rarely and extra care should be taken for anchor placement, knot tighting and instrument usage.

P10-1340 Simultaneous strain measurements of rotator cuff tendons at varying arm positions and the effect of supraspinatus tear: a cadaveric study Lie D.1, Sheng J.M.2 1 Singapore General Hospital, Orthopaedic Surgery, Singapore, Singapore, 2 Nanyang Technological University, c/o Orthopaedic Surgery, SGH, Singapore, Singapore Objectives: Previous studies on strain patterns of rotator cuff tendons measured isolated tendons. This study was designed to measure strain of the different rotator cuff tendons simultaneously and demonstrate the effect of supraspinatus tear on other cuff tendons with an intact glenohumeral joint. Methods: Ten fresh-frozen shoulders were tested on a purpose-built rig. With 10 kg loaded at the rotator cuff muscles, displacement variable reluctance transducers (DVRTs) were used to measure the supraspinatus, infraspinatus and subscapularis tendons strains simultaneously at varying elevation angles and planes. Simulated partial thickness tears were cut at the bursal-side of the supraspinatus and sequentially enlarged to full-thickness tear. Results: Results showed that in abduction in the scapular plane, supraspinatus (bursal-side) strains decreased to a maximum of -7.7% while the supraspinatus (articular-side) increased to a maximum of +14.1% at 60 elevation. Infraspinatus and subscapularis reached their peak strain of +3.3% and +2.9% at 45 elevation. In the sagittal plane, supraspinatus (bursal-side) strains decreased to a maximum of -8.8% while supraspinatus (articular-side) and infraspinatus were observed to increased to a maximum of +5.2% and +3.2% respectively at 60deg elevation. There was insignificant change for the subscapularis. In the coronal plane, supraspinatus (bursal-side) strain decreased to a maximum of -13.7% while the supraspinatus (articular-side) strain increased to a maximum of +12.65% at 60 elevation. Subscapularis strain increased to a maximum of +4.5% at

123

S144 60 elevation. No significant change was detected for infraspinatus strain. At most positions, strain values varied between anterior and posterior supraspinatus with a maximum difference of 4.8%. Increasing tear size on supraspinatus further elevated the strain on infraspinatus and subscapularis. Conclusions: The simultaneous strain measurement of all rotator cuff tendons is unprecedented. This cadaver model demonstrated that rotator cuff tendons showed varying strain behaviour at different elevation planes. The strain difference within supraspinatus implied occurrence of shearing which may cause intratendinous tears. Tears in supraspinatus elevate strain on the two adjacent cuff tendons.

P10-1341 Rotator cuff integrity after arthroscopic repair of a fatty degenerated muscle Chillemi C.1, Garro L.1, Tucciarone A.1 1 Istituto Chirurgico Ortopedico Traumatologico, VII Divisione, Latina, Italy Objectives: 10 to 70% of repairs of the rotator cuff (RC) structurally fail. 4 years after surgery, the rate of RC re-tear is dramatically high, and is directly correlated with the number of tendon repaired (10% to 30% after repair of an isolated supraspinatus tear; 40% to 60% after repair of 2 tendons, and 50% to 90% after repair of 3 tendons) and the presurgical fatty degeneration of tendons graded in accordance with the Goutallier scale. The aim of the present study was to compare clinical and structural results of arthroscopic side-to-side (STS) and reinsertion to bone (RTB) RC repair in patients who had stage 3 and 4 fatty degeneration. Methods: 34 patients (24 men and 10 women; mean age of 60.5 years range 35-81 years), affected with RC tear and a fatty degeneration stage 3 and 4 were operated during the period 2002-2003. The arthroscopic repair consisted of a STS suturing technique (19 cases) and a RTB with bioanchor in a single row technique (15 cases). Patients were evaluated using the Constant score, and after 6 months and 4 yrs after surgery with MR. Results: The Constant score improved from 51.6 to 82.9 in the STS group and from 52.6 to 83.3 in the RTB group, with a not-significant deterioration of the results with time. At the end of the follow-up the number of re-tears was 8 (40.2%) in the STS group (i.e. increase of the tear size) and 6 (40%) in the RTB group, detected with MRI. Conclusions: Although RC re-tear is relatively frequent in highly degenerated cuff, the indication for revision surgery is not the documentation of the re-tear but the symptoms associated with it. In consideration of the imaging results, even if the data collected refers only to 34 patients, we can conclude that a STS repair technique could be the better indication (reduced cost -no device, and shorter time of surgery) in the treatment of RC tear with stage 3 and 4 fatty degeneration.

P10-1364 To evaluation of consistency between findings of shoulder physical examination and MRI arthrography results Mahirogullari M.1, Cakmak S.2, Donmez F.3, Aytekin A.3, Mutlu H.3, Kuskucu M.3 1 GATA Haydarpasa Training Hospital, Orthopaedic Surgery, Istanbul, Turkey, 2Aksaz Military Hospital, Mugla, Turkey, 3GATA Haydarpasa Training Hospital, Istanbul, Turkey Objectives: The aim of this study is to compare between the first diagnosis getting by only physical examination and the final diagnosis getting by MRI arthrography, and to emphasize that importance of physical examination in rotator cuff diseases. Methods: 36 patients have shoulder pain were included in this study. All demographic data from patients were recorded. The specific tests (Neer, Hawkins, Bear Hugger, lift-off, Speed etc.) for rotator cuff problems were performed. The results getting by examination tests were noted. We tried to get a diagnosis. The arthro MRI was taken by using 15 ml diluted gadolinum. If the contrast material could pass the subacromial space it is

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 accepted that there is a full thickness rotator cuff tear. We compared the results. Results: The findings from the rotator cuff problems were consistent to arthro MRI results. The correct diagnosis estimation rate was %90 for only clinical examination. Especially, strength tests against force were found very valuable for diagnosis of rotator cuff complete tear. Conclusions: Having diagnosis for shoulder problems, clinical examination is very important and additional MRI arthrography ensure to enhancement the correct diagnosis.

P10-1372 The labrum reconstruction and osseous reaction after arthroscopic Bankart repair in the MRI and score system Stein T.1, Mehling A.P.1, Ulmer M.1, Buckup J.1, Reck C.1, Ja¨ger A.1, Welsch F.1 1 BG Unfallklinik, Dept. of Sports Traumatology - Knee- and ShoulderSurgery, Frankfurt am Main, Germany Objectives: The primary arthroscopic stabilization using Suture Anchors is the current Gold Standard procedure for traumatic anterior shoulder instability. The new anchor generation with absorbable materials and knotless fixation technique provides comparable biomechanical strength. In this prospective study the labral reconstruction and morphology as well as the bony reaction and biodegradation were assessed structurally in the MRI. The influence on the clinical outcome has been measured by a complex score system. Methods: 30 patients after Bankart repair and 31 volunteers (without shoulder instability; [26.7y) were assessed structurally in a standardized MRI (Stir+T1/512 cor. DE sag., PDW+PDW Spir axial; no CM). The ‘‘Bankart repair group’’ ([24.5y at first dislocation; [28.1y at surgery, no. of [preop. dislocation 4.66 ± 7.74) was examined by MRI preoperatively, 14 and 28 months postoperatively (FU 0+1+2) and by a score system (Walch-Duplay, Rowe, Constant-Murley, ASES, DASH) after isolated Bankart repair (2x double armed Bio-Pushlock, Arthrex). The anterior and inferior Labrum Slope (a/iLS), the Labrum-Glenoid Height Index (a/iLGHI) and the labral degeneration (Randelli graduation) were measured (FU 0+1+2). The grade of osseous reaction (0-III) and anchor biodegradation (0-III) were recorded separately for both anchors (FU 1+2). Results: The Bankart repair group using the Bio-Pushlock enabled a labrum reconstruction comparable to the control group. After Bankart surgery the anterior portion (aLS 24.4 ± 2.5/ aLGHI 3.0 ± 0.6) and inferior portion (iLS 24.8± 1.9/ iLGHI 2.3 ± 0.5) were similar to the data of the control group (aLS 24.4 ± 2.5/ aLGHI 3.0 ± 0.6; iLS 25.0 ± 2.2/ iLGHI 2.3 ± 0.3) (p [ 0.5) and significantly improved to the preoperative status (aLS 13.9 ± 3.4/ aLGHI 1.3 ± 0.5/ iLS 15.8± 3.7/ iLGHI 2.3 ± 0.3) (p = 0.009). There were no longitudinal changes (FU1-2). The labrum degeneration was increased in the Bankart group (p = 0.03). The preoperative number of dislocations had no influence on the Bankart repair data (p [ 0.5). The osseous reaction (OR) and the anchor-biodegradation (ABD) were increased for the implants in the lower glenoid portion (OR 1.9/ ABD 2.2) compared to the upper glenoid (OR 1.1/ ABD 1.5) after 28 months. The 14 months data showed a generally increased status for OR and ABD. There were no anchor (sub-)dislocations or cystic configuration. The grade of OR and ABD had neither an influence on the function and instability specific score system (p [ 0.5) nor on the persisting instability (9.7%; p [ 0.5). The score system 28 months after Bankart repair detected excellent results (WD 87.7±6.4; Rowe 89.9±5.1; CM 92.25±6.2; ASES 94.47±4.9; DASH 26.36±3.4) with minimal loss of external rotation in neutral position 3.85 ± 4.05 and 90 abduction 2.11 ± 3.42. Conclusions: The arthroscopic Bankart repair using knotless, bioabsorbable suture anchors revealed favourable outcome data with restored range of motions. These data proved the secure labrum reconstruction compared to healthy cohort without any gap formations. The biodegradation of the anchor and the osseous reaction occurred without any pathologic findings.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P10-1392 Cement augmentation increases the pullout strength of suture anchors in low bone quality while rotator cuff repair Kirchhoff C.1, Braunstein V.2, Ahrens P.3, Imhoff A.3, Hinterwimmer S.4 1 University Hospital Munich, TU, Klinikum rechts der Isar, Abteilung fu¨r Orthopa¨die und Unfallchirurgie, Munich, Germany, 2Ludwig-Maximilians Universitaet, Department of Orthopedic Surgery and Traumatology Campus I, Muenchen, Germany, 3Technical University Munich, Department of Orthopaedic Sports Medicine, Munich, Germany, 4 Klinikum rechts der Isar, Department of Orthopedic Sports Medicine, Muenchen, Germany Objectives: The fixation of suture anchors in low bone quality is a matter of ongoing interest, especially in case of rotator cuff repair. Therefore, the effect of cement augmentation on suture anchor fixation in low bone quality was evaluated. Methods: 28 osteoporotic humeral heads were included in this biomechanical study. The bone density of two typical anchor insertion regions (group 1: posterior-medial anchor position and group 2: anterior-lateral anchor position) was measured at the greater tuberosity of each bone, using a high resolution CT (voxel size: 82lm). In group 1 (posteriormedial anchor position) and group 2 (anterior-lateral anchor position) preferably similar anchor insertion regions were assigned pairwise according to the bone density for this matched pair analysis. Using these pairs, suture anchors (Arthrex Corkscrew FT 1) were inserted randomized conventionally or in a cement augmentation technique (using PMMA cement). All anchors were cyclically ramp-loaded until pullout using a material testing machine (preload: 20N; load progression per cycle: 0.4N). Results: In group 1 (posterior-medial anchor position) the pullout strength for the conventional anchor insertion technique was 226N and 332N for the cement augmentation technique. In group 2 (anterior-lateral anchor position) the pullout strength for the conventional technique was 209N and 304N for the cement augmentation technique. For both groups the differences between the conventional and the cement augmentation technique were significant (p\0.05). Conclusions: Compared to the conventional insertion technique, the cement augmentation technique increases the pullout strength of suture anchors in low bone quality significantly. Therefore cement augmentation could be a helpful tool for improved suture anchor fixation in rotator cuff repair, especially in low bone quality.

P10-1394 Assessment of bone quality within the tuberosities of the osteoporotic humeral head: relevance for anchor positioning in rotator cuff repair Kirchhoff C.1, Braunstein V.2, Ahrens P.3, Milz S.4, Imhoff A.3, Hinterwimmer S.5 1 University Hospital Munich, TU, Klinikum rechts der Isar, Abteilung fu¨r Orthopa¨die und Unfallchirurgie, Munich, Germany, 2Ludwig-Maximilians Universitaet, Department of Orthopedic Surgery and Traumatology Campus I, Muenchen, Germany, 3Technical University Munich, Department of Orthopaedic Sports Medicine, Munich, Germany, 4Institute of Anatomy, Ludwig-Maximilians Universitaet, Muenchen, Germany, 5 Klinikum rechts der Isar, Department of Orthopedic Sports Medicine, Muenchen, Germany Objectives: Tears of the rotator cuff are highly prevalent in patients older than 60 years, thereby presenting a population also suffering from osteopenia or osteoporosis. Suture fixation in the bone depends on the holding strength of the anchoring technique, whether a bone tunnel or suture anchor is selected. Because of osteopenic or osteoporotic bone changes, suture anchors in the elderly patient might pull out resulting in failure of repair. Therefore the aim of our study was to analyze the bone quality within the tuberosities of the osteoporotic humeral head using high-resolution quantitative computed tomography (HR-pQCT). Methods: Thirty-six human cadaveric shoulders were analyzed using HRpQCT (Scanco Medical, Switzerland). The mean bone volume to total volume (BV/TV) as well as trabecular bone-mineral densities (trabBMD) of the greater tuberosity (GT) and the lesser tuberosity (LT) were determined. Within the GT six volumes of interest (VOIs), within the LT two

S145 VOIs and one control volume within the subchondral area beyond the articular surface were set. Results: Comparing BV/TV of the medial to the lateral row significantly higher values were found medially (p \0.001). The highest BV/TV with 0.030±0.027% was found in the posteromedial portion of the GT (p\0.05). Regarding the analysis of the LT no difference was found comparing the superior (BV/TV: 0.024±0.022%) and the inferior (BV/TV: 0.019±0.016%) portion. Analyzing trabBMD equal proportions were found. An inverse correlation with a correlation coefficient of -0.68 was found regarding BV/TV of the posterior portion of the GT and age (p\0.05). Conclusions: Significant regional differences of trabecular microarchitecture were found in our HR-pQCT study. The volume of highest bone quality resulted for the posteromedial aspect of the GT. Moreover, a significant correlation of bone quality within the GT and age was found, while the bone quality within the LT seems to be independent from it. The shape of the RC tear majorly determines the bony site of tendon reattachment. Though the surgeon has distinct options to modify anchor positioning. According to our results placement of suture anchors in a medialized way at the border to the articular surface might guarantee a better structural bone stock.

P10-1400 The comparison of between the results of open and arthroscopic rotator cuff repair Mahirogullari M.1, Ozkan K.2, Akyildiz F.3, Eceviz E.4, Tirmik U.5, Uygur E.4, Kuskucu M.3 1 GATA Haydarpasa Training Hospital, Orthopaedic Surgery, Istanbul, Turkey, 2Goztepe Training Hospital, Orthopaedics, Istanbul, Turkey, 3 GATA Haydarpasa Training Hospital, Istanbul, Turkey, 4Goztepe Training Hospital, Istanbul, Turkey, 5Merzifon Military Hospital, Orthopaedics, Istanbul, Turkey Objectives: The aim of this study was to compare the results of repair of the open and arthroscopic rotator cuff tear. Methods: Between 2006 and 2008, we performed 31 rotator cuff tear surgery for 31 patients. 16 operations were performed open surgery and 15 were by arthroscopy. Mean age was 52,8 and 57,8 for open and arthroscopic group, respectively. Acromioplasty and rotator cuff repair were performed ll patients. In addition, 2 biceps tenodesis, 1 biceps tenotomy and 1 SLAP repair were done in arthroscopy group. Results: Mean follow-up time was 14 months for open repair group, and 12,2 months for arthroscopic group. Mean Constant score was 37,75 in preoperative period and 76,25 for postoperative period in open group; 46,2 and 92,13 for arthroscopic group, respectively. Conclusions: We found that both operation method were effective for rotator cuff repair but arthroscopic group has better results than open group. In addition, more concomitant pathologies besides rotator cuff tear could be figured out by arthroscopic method. We believe that arthroscopic method is better than open group in respect of that patient comfort and rehabilitation period after surgery.

P10-1403 Radiographic analysis for glenoid morphology using modified Bernageau method Takahashi N.1, Sugaya H.1, Hagiwara Y.1, Tonotsuka H.1, Kawai N.1 1 Funabashi Orthopaedic Sports Medicine Center, Funabashi, Japan Objectives: Glenoid bone loss is significant risk factor in surgery for anterior glenohumeral instability. It is generally accepted that 3DCT is most suitable for evaluation of glenoid morphology. However, because of a lack of facility or a health insurance issue, it is available not all countries. Bernageau introduced a valuable procedure which can demonstrate an anterior glenoid rim with using fluoroscopy control. We formulated a new radiographic technique which enables us to observe the glenoid morphology without the fluoroscopy. The purpose of this study is to assess the effectiveness of the new radiographic technique in a series of patients with anterior glenohumeral instability.

123

S146 Methods: Ninety five patients with recurrent anterior glenohumeral instability were included in this study. The new radiographic technique was applied in both shoulders and a 3DCT was utilized in the affected shoulder, preoperatively. All patients experienced an arthroscopic stabilization after the imaging study. Our new technique is as follows: a radiographic film is placed under the axillar area in a lateral decubitus position with the shoulder abducted fully. The X-ray is applied from 15-20 degrees cranially, which allow for parallelizing the incident X-ray along with the long axis of glenoid. The glenoid views were classified into three groups including a fragment type, an attritional type and a normal type using contra-lateral control by three blinded observers. The results were verified by 3DCT. The fragment types and attritional types were classified to three subtypes, including small, medium, large and mild, moderate, severe respectively using en faced glenoid views by 3D CT. The diagnostic rate in each type and subtype were obtained and a sensitivity and specificity were calculated. Results: According to 3DCT, there were 33 fragment types (16 small, 17 medium to large), 55 attritional types (30 mild, 25 moderate to severe) and 7 normal types. The diagnostic rates using the radiographs were 62.6% in fragment type (small 64.6, medium to large 58.8), 76.2% in the attritional type (mild 55.6, moderate to severe 86.7) and 69.7 in the normal type. Sensitivity/specificity in each group was as follows, fragment: 62.6/94.1%, attritional: 62.4/70.8% and normal: 76.1/82.6%. Conclusions: Our new method diagnosed the glenoid morphology with the rate 63 to 76%. Although there was no difference in diagnosis between the small and medium to large fragment, the better accuracy was obtained in the moderate to severe attritional type compared to the mild attritional type. The modest sensitivities were found in each type and the higher specificity was observed in fragment type. We concluded that our new method can be valuable for the screening of glenoid morphology in patients with glenohumeral instability. Specially, the procedure would be helpful for the diagnosis in significant bony lesion including fragment and moderate to severe attrition.

P10-1404 Subacromial impingement syndrome caused by spurs of the bilateral greater tuberosities. A case report Saji T.1, Sano T.1, Suzuki T.2, Matsuoka H.1 1 Shizuoka General Hospital, Department of Orthopaedic Surgery, Shizuoka, Japan, 2Osaka Red Cross Hospital, Osaka, Japan Objectives: Many causes of subacromial impingement syndrome are known, such as rotator cuff insufficiency, thickening of the rotator cuff, a bony spur of the acromion and so on. But a spur of the greater tuberosity is a rare cause of impingement syndrome. We report a case of bilateral subacromial impingement syndrome caused by massive spurs of the bilateral greater tuberosities. Methods: A seventy-year-old man was seen because of pain in both shoulder, especially in the right. Physical examination revealed limited range of motion and positive impingement sign. Anteroposterior radiographs showed massive spurs of the bilateral greater tuberosities. Bony spurs of the bilateral acromions were also found. Magnetic resonance imaging of the right shoulder showed bony spurs and subacromial bursitis, but no apparent complete rotator cuff tear was found. A diagnosis of subacromial impingement syndrome caused mainly by massive bony spur of the greater tuberosity was made. Right shoulder pain was strong and an arthroscopic operation to remove the spurs in the right shoulder was performed. Arthroscopy of the subacromial bursa showed that massive bony spur of the greater tuberosity impinged against the acromion. The spurs of the greater tuberosity and acromion were resected until there was no impingement. Since the left shoulder pain was not so severe as the right shoulder, the left shoulder was treated conservatively. Results: Within a month after surgery, the patient had no pain in the right shoulder. The range of motion was recovered almost fully. Postoperative radiographs showed a small residual spur of the greater tuberosity. At the follow-up visit 18 months after the operation, the patient did not feel pain in the bilateral shoulder, though range of motion of the left shoulder was slightly limited.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Conclusions: The present case suggests that a spur of the greater tuberosity should be considered as a possible cause of subacromial impingement syndrome. Arthroscopic resection of the spur is an optimal treatment if the pain is strong and the range of motion is severely limited.

P10-1416 Radiofrequency microtenotomy in treatment of calcific tendonitis of supraspinatus Havlas V.1, Trc T.2, Challagunda R.3, Chladek P.4 1 Charles University Prague, 2nd Medical School, Prague, Czech Republic, 2 University Hospital Motol, Orthopaedic Clinic, Prague, Czech Republic, 3 Dumfries and Galloway Royal Infirmary, Department of Orthopaedics and Trauma, Dumfries, United Kingdom, 4Charles University in Prague, Department of Orthopaedics and Trauma, Prague, Czech Republic Objectives: Calcific tendonitis is a common disorder of the rotator cuff. It is a cell-mediated process that is often chronic in nature, but it is usually self-limiting with regard to its acute pain states. The pathological process involves metaplastic changes resulting in fibrocartilage differentiation of the tendon with poor self-repair capacity. Non-operative management has been reported as the treatment of choice and is successful in up to 90% of patients. For the patients remaining symptomatic after conservative treatment, surgical excision of the calcium deposits usually offers a generally reliable pain relief; the recent trend is toward arthroscopic management with limited clinical outcomes, mainly due to the nature of the disease being based on metaplastic change of the tendon with limited blood supply and therefore poor healing and selfreparatory capacity. We have designed a new technique of treatment of supraspinatus calcific tendonitis in attempt to improve the healing potential after the arthroscopic treatment. The aim of this prospective study was to compare the standard arthroscopic needle technique with a technique using radiofrequency-induced plasma microtenotomy as enhancement of the arthroscopic treatment of the supraspinatus calcific tendonitis. Previously published technique (Taverna et al.) of subscapularis tendon stimulation by bipolar radiofrequency-based microtenotomy (microdebridement) was used in addition to an arthroscopic calcific deposits needling. Methods: 20 patients suffering clinically symptomatic painful rotator cuff calcific tendonitis, prospectively included in this study, were divided into two equal groups. All patients were preoperatively X-ray and MRI diagnosed with type A or B calcific tendonitis of supraspinatus tendon. All patients had at least 6 months of unsuccessful conservative treatment. The group A was treated by standard arthroscopic technique with needle removal of calcific deposits. Patients from the group B underwent identical surgery with additional stimulation using bipolar radiofrequency-induced plasma microtenotomy (Topaz-XL Arthrowand, Arthrocare) at the end of the surgery in attempt to enhance the reparatory process of the rotator cuff after the arthroscopic treatment of the supraspinatus calcific tendonitis. Subacromial decompression was performed only in cases with peri-operative appearance mechanic impingement between acromion/AC joint and the rotator cuff. All patients were clinically assessed for pain, ROM and Constant score as the main measure of their clinical symptoms prior to surgery and at 3 and 6 months after the operation. Results: We have observed inferior clinical outcome measured by postoperative Constant score in the group A treated by simple arthroscopic needle calcific deposits removal when compared with patients from the group B treated by additional bipolar radiofrequency microtenotomy. The mean preoperative Constant score in both groups was 53,6 (31-69). The postoperative Constant score in Group A increased to mean 68,3 (39-79). The mean postoperative Constant score in Group B was 82,5 (62-100). The above difference of the two patient groups was evaluated as statistically significant. Conclusions: This study has shown that a use of bipolar radiofrequency microtenotomy in addition to the arthroscopic needle removal of calcific deposits from the supraspinatus tendon provides better clinical outcome to patients with a chronic calcific tendinitis irresponsive to conservative treatment measures.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P10-1423 Tensile strength of glenoid bone-cement and polyethylene-cement interface Sanghavi S.1, Hansen U.2, Amis A.A.3 1 Hillingdon Hospital, Uxbridge & Imperial College London, Mechanical Engineering, London, United Kingdom, 2Imperial College London, Mechanical Engineering, London, United Kingdom, 3Imperial College of Science Technology and Medicine, Mechanical Engineering Department, London, United Kingdom Objectives: Symptomatic loosening of the cemented glenoid component includes mechanical failure of the cement layer or of the glenoid bonecement or PE-cement interface or of the adjacent bone. This in-vitro study determined the tensile strength of the glenoid bone-cement and PE-cement interfaces and compared them with the strength of the bulk cement and bone. We analysed the regional variation of the tensile strength of the glenoid bone-cement interface and the effect of PE surface roughness on tensile strength of PE-cement interface. Methods: Seven fresh frozen human scapulae were used for this study. The glenoid was prepared following standard surgical procedures and sectioned into superior, central and inferior segments. PMMA was applied to the bone under pressure and bone-cement test specimens were made. The specimens were subsequently loaded in tension until failure. The loaddisplacement traces were recorded and the strength was estimated based on the maximum load recorded and the cross-sectional area of the specimen. Results: The overall mean strength of the bone-cement interface was 2.5 ± 0.8MPa. The respective tensile strength of the superior, central and inferior regions was 1.7 ± 0.4MPa, 2.2 ± 0.9MPa and 3.4 ± 1.0MPa (ANOVA, p = 0.010). The respective tensile strength of PE with surface roughness of 1.2 ± 0.6 lm, 2.7 ± 0.5lm and 5.5 ± 0.4lm was 0.0MPa, 1.5 ± 0.6MPa and 3.2 ± 0.7MPa respectively. Conclusions: Fixation at the inferior glenoid region was significantly stronger than in the superior and central regions and that the PE-cement interface would be the weakest point if the PE surface roughness was less than 4lm. However, if the PE surface roughness was greater than 5lm, the glenoid bone-cement interface would then become the weakest point of the fixation.

P10-1438 Pattern of chondral and subchondral plate thickness of the human glenoid Sanghavi S.1, Merican A.2, Hansen U.2, Amis A.A.3 1 Hillingdon Hospital, Uxbridge & Imperial College London, Mechanical Engineering, London, United Kingdom, 2Imperial College London, Mechanical Engineering, London, United Kingdom, 3 Imperial College of Science Technology and Medicine, Mechanical Engineering Department, London, United Kingdom Objectives: The detailed anatomy and morphometry of the glenoid were studied to provide information for glenoid implant fixation in total shoulder arthroplasty. A simple technique using high-resolution images of the cross section of the glenoids with different colour spectra of different layers was used. This study describes the glenoid articular cartilage and subchondral plate thickness variations over its entire surface in specimens with a range of degenerative changes. Methods: Eighteen embalmed glenoid specimens were divided into control and diseased groups. The articular surface was stained with India ink and scanned and analyzed independently for wear pattern by two observers. The India ink staining highlighted areas of cartilage degeneration. The glenoid was sectioned along the grid marked on the glenoid articular surface, dividing it into anterior and posterior halves, then into 3 equal height segments. The cancellous part of each glenoid segment was cleaned with a water jet and stained with a blue ink. This staining enabled accurate visualization of the uncalcified articular cartilage, calcified subchondral plate and the cancellous bone of the glenoid. The stained cross sections were scanned and the areas of the uncalcified cartilage and the subchondral plate were measured.

S147 Results: The overall mean thickness and standard deviation of the cartilage in the control and diseased group was 1.9 ± 0.4mm and 1.5 ± 0.7mm respectively. The overall mean thickness and standard deviation of the subchondral plate in the control and diseased group was 0.4 ± 0.2mm and 1.0 ± 1.3mm respectively. The thickest subchondral plate (1.6 ± 2.2mm) was in the posterior-superior zone in the diseased group. Glenoid specimens from the diseased group have significantly lower cartilage thickness (p=0.015) and higher subchondral plate thickness (p\0.010) than the control group. The average difference in the cartilage and subchondral plate thickness between the two groups was -0.45 units and 0.57units respectively. Also for each unit decrease in cartilage thickness, there was a 0.34 unit increase in subchondral plate thickness (p\0.001) in the diseased group suggesting a relationship between the two layers. Conclusions: The glenoid cartilage and subchondral plate thickness differed between the control and diseased group. There was increase in the subchondral plate thickness as the articular cartilage got thinner in the diseased glenoids. The posterior-superior area had the thickest subchondral plate. The variation in the subchondral plate thickness should be taken into account while reaming for the flat-backed glenoid prosthesis.

P10-1446 Clinical and radiographic evaluation of inferior gleno-humeral ligament laxity following arthroscopic Bankart repair Sanghavi S.1, Meyer A.2, Hardy P.3 1 Hopital Ambroise Pare & Hillingdon Hospital, Orthopaedics, Uxbridge, United Kingdom, 2Ambroise Pare´ Hospital, Boulogne Billancourt, France, 3 Hopital Ambroise Pare´, West Paris University, Boulogne, France Objectives: The purpose of the present study was to clinically and radiographically evaluate the gleno-humeral joint along with the inferior glenohumeral ligament (IGHL) laxity following arthroscopic Bankart repair in young patients who presented with recurrent anterior shoulder instability. In addition, the clinical results were correlated with a new radiographic examination. Methods: This retrospective study included twenty three patients, seventeen men and six woman with a mean age of 29.3 years presenting with anterior shoulder instability following a traumatic event. Seventeen patients sustained sports related injury and fourteen had involvement of the dominant extremity. The mean number of preoperative subluxation and dislocation were 6 and 2 respectively. The mean duration from first consultation to arthroscopic Bankart repair was 3.5 months. A minimum of 3 anchors were used for arthroscopic Bankart repair. Subjective outcome measures, objective evaluation with Gagey’s hyperabduction test, Walch-Duplay and Rowe score and radiographic evaluation with a comparative anteroposterior radiograph of both shoulders were obtained at a mean follow-up of 36 months. Radiographs were taken with the patient supine and the shoulder passively held in maximum abduction in neutral rotation against resistance. The angle (hyper-abduction angle) between the humeral shaft axis and a line drawn between the lateral border of the scapular tubercle and the inferior edge of the glenoid fossa was measured. Results: Pre-operative shoulder motion between the affected and the contra lateral normal shoulder were symmetrical. The mean pre-operative Instability severity index score (ISIS) was 1.84. At 6 month post-operative follow-up, all patients demonstrated 40% reduction in external rotation with the arm by the side and in 90 abduction. However, at the final follow-up both external rotation slightly improved but was still reduced by 26%. The mean Walch-Duplay and Rowe score was 71.8 and 90.2 respectively. The mean difference in the radiographic hyper-abduction angle between the affected and the contra lateral normal shoulder was -5.4. Conclusions: This study demonstrates limitation in both external rotations following arthroscopic Bankart repair indicating shoulder stability. The comparative radiographic examination clearly reveals tightening of the IGHL as evident by the difference in the passive hyper-abduction between the operated and the contra lateral normal shoulder. This mean difference in the hyper-abduction appears to be appropriate rather than too small or

123

S148 too large, otherwise it would affect the functional outcome leading to either recurrence of instability or shoulder stiffness. The comparative hyper abduction radiographic examination is reliable and cost effective. It can be effectively used in patients who present with either recurrent instability or stiffness following the Bankart stabilization procedure rather than arthrographic examination.

P10-1450 Prevalence of preventive training in swimners with shoulder pain in 2008 Brazilian south championship Pacheco I.1, Pacheco, More´ A.2, Longaray, Almeida A.3, Da Silva, Lindner M.3, Ceccato J.4 1 Rio Grande do Sul State Soccer Federation, Porto Alegre, Brazil, 2 Federal Universty of Rio Grande do Sul, Laboratory of Exercise, Porto Alegre, Brazil, 3University of Vale dos Sinos, Porto Alegre, Brazil, 4 Gremio Na´utico Unia˜o, Fisiotherapy, Porto Alegre, Brazil Objectives: The objective of this study was to determine if a preventive program was used by swimmers with shoulder pain. Methods: The sample consisted of 173 swimmers of 2008 brazilian south championship of Mirim and Petiz categories. The instrument used was a closed questionnaire, containing questions that identified complains of pain in the shoulder, swimming technique, training load, age of beginning in sport and if a prevention methods was used when these athletes have shoulder pain. Data collection was performed in September 2008 during the south season of braziliam championship by the medical department and physiotherapy of Greˆmio Na´utico Unia˜o. Results: The statistical analysis was chi-square test or Fisher0 s exact test, Student t-test, Mann-Whitney and Multivariate Logistic Regression, all being significant with p B 0.05. The results showed that 55 swimmers had complaints of pain, against 118 reported no shoulder pain. The butterfly style has been the most evident, representing 38.2% of the swimmers with complaints. Athletes who performed all the techniques of swimming were the ones who had less complaints, representing 16.4% of athletes reported pain. Still, it was observed that 44.9% of the competitors whom do not complain of pain practiced all the techniques. The strengthening of rubber stood out was the best method of prevention of pain (p=0,001), where 94.5% of the swimmers who reported pain did this prevention. When controlling confounding factors, the variable that remains associated with the complaint of shoulder pain is the technique of butterfly (OR = 2.66, 95% CI = 1.19 to 5.94, p = 0.017), otherwise strengthening with rubber is a protective factor (OR = 0.08, 95% CI = 0.02 to 0.40) corresponding on average to a reduction of 92% chance of developing shoulder pain. Conclusions: We believe that a guided preventive method will give to swimmers a muscular balance and, consequently, a lower likelihood of future injuries that might interfere with his career.

P10-1458 Chronic unreduced anterior shoulder dislocation Stevanovic V.1 1 Institute for Orthopaedic Surgery ‘Banjiac’, Belgrade, Serbia Objectives: Literature data are with distinct criterion in nomenclature, classification or therapy option for unreduced anterior shoulder dislocation. Shown are empirical experience after open reposition and anatomic surgical reconstruction. Methods: We analyzed 8 patients between 17 and 70 years old with anterior unreduced shoulder dislocation of 4 to 20weeks at the Institute for Orthopaedic surgery ‘‘Banjica’’ Belgrade. Pain and function loss was the main complain. Open reposition with anatomic reconstruction of capsulolabral complex were done in all cases; glenoid and humeral head defects were not repaired. We used Rowe Zarins scale for results evaluation. Results: According to Rowe-Zarins scale 4 patients were with excellent, 2 with very good and two with good result; postoperative score was average 85. Pain score was improved from 15 to 30, functional from 10 to 25 and elevation averaged 135, ER 25 and IR to L1.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Conclusions: Unreduced chronic anterior shoulder dislocation are best treated with open reposition and anatomic surgical reconstruction with glenoid and humeral head preservation for good outcome.

P10-1464 Radial shock-wave therapy for calcifying tendinitis of the rotator’s cuff Ziltener J.-L.1, Leal S.1 1 Hoˆpitaux Universitaires Geneve, Unite´ d’Orthope´die et Traumatologie du Sport, Confignon, Switzerland Objectives: This single-blind, randomized controlled study was performed to evaluate the effectiveness of radial shock wave therapy (RSWT) in the management of calcifying tendinitis of the shoulder among 35 patients. Methods: 18 patients with a symptomatic calcifying tendinitis of the shoulder for more than six months were treated at the area of maximum pain by low-energy RSWT, with application of 2000 impulses/session, 1session/week, five sessions totally, what means a total of 1200 mJ/mm2. In the control group (17 patients), no energy was delivered. Shoulder function (Constant score) and pain (visual analogue scale, VAS) were evaluated before treatment and at 48 hours, 6 and 12-weeks, after the last treatment. Shoulder X-rays were performed before treatment and at the 12weeks follow-up visit. A Dash score (self-report questionnaire designed to measure physical function and symptoms in people with any of several musculoskeletal disorders of the upper limb) was also completed before treatment and at a 12-weeks follow-up. Results: Improvement in Constant score was not substantial in both groups during the first 4 weeks after RSWT, but was significantly higher in the treatment group (p \ 0.05). The Constant score improved significantly (p\0.0001) in the treatment group at 12 weeks follow-up from a mean of 73.5 to a mean of 90.5 points. Regarding the reduction of pain, there was an improvement in the treatment group compared with the control group at the 4-week follow-up (p\0.01), that became very significant at 12-weeks follow-up (p \ 0.0001). In the treatment group, the X-ray examination at 12- weeks’ follow-up showed a total resorption in 6 patients (30%) a partial disintegration in 20%, and no change in 50% of cases. Conclusions: The results suggest that the use of RSWT for the management of calcifying tendinitis of the shoulder is effective, leading to a significant reduction in pain and improvement of shoulder function. The effect is nevertheless delayed, particularly significant around 12 weeks after the last treatment. The amount of total delivered energy seems to be a crucial factor, more than the method of producing or focusing the shock waves, the frequency and timing of delivery, and whether or not local anaesthetics are used.

P10-1489 Maximizing visualization of long head of the biceps tendon during arthroscopy Dragoo J.1, Hart N.1 1 Stanford University, Orthopaedic Surgery, Redwood City, United States Objectives: Pathology of the biceps tendon is a common cause of shoulder pain. Tendinitis and tendinosis of the long head of the biceps tendon can be very debilitating. ‘‘The Hidden Lesion,’’ in the bicipital groove has been previously described by authors frustrated with pathology in this region. Failure to diagnosis biceps pathology can lead to suboptimal or even full failure of arthroscopic treatment . Previous literature has stressed the importance of pulling as much biceps tendon intra-articular for full evaluation. However, to this date no study has fully investigated the completeness of this maneuver nor has the position of the arm to maximize visualization been verified. Methods: 6 pairs of fresh-frozen shoulders were disarticulated at the scapulothoracic joint proximally and transected at the proximal forearm, distal to the biceps tendon insertion. Specimens were excluded if any evidence of previous surgery is identified. The exclusion criteria included those with fractures, contracture, severe osteoarthritis, and other diseases of the shoulder detectable by direct inspection or radiographs. Five pounds

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 of traction was applied to the humerus inferiorly to recreate traction commonly used in the lateral position, and to confirm the integrity of the capsule of the glenohumeral joint. Two Compass hinged external fixation devices were placed, one at the shoulder and one at the elbow. Standard posterior and anterior-inferior and anterior-superior portals were established. A Tension Isometer was placed through the anterior-inferior portal, and the biceps tendon was pulled into the glenohumeral joint at a tension of 2 lbs. While pulling a constant force on the biceps tendon, the arm was placed in standardized positions of shoulder flexion, extension, internal and external rotation, abduction, and adduction . A marked suture was placed arthroscopically into the biceps tendon at the most distal point of visualization for each of the positions. After all marks were placed, an anterior approach to the bicipital groove was performed. The distance from the sutures to the biceps insertion on the superior labrum was recorded. The overall percentage of visualized tendon was calculated. Results: 2 lbs of tension (pulling biceps into joint) increased intra articular excursion about one centimeter. The average increase in visualization with 30/40/0 shoulder position was 1.0cm. The average increase in visualization with elbow flexion was 8.0 mm. The total increased visualization was 2.8cm which is enough to see all the tendon in the groove (average groove length 2.45cm). All other positions had less improvement in visualization. Supination led to no improvement in visualization. Conclusions: The position yielding the most visualization of the biceps tendon is 30 degrees forward elevation, 40 degrees abduction and 90 degrees elbow flexion. At this position the entire tendon within the bicipital groove could be seen.

Elbow/hand/fingers P11-391 Endoscopic versus open carpal tunnel release in carpal tunnel syndrome: a comparative clinical trial Vasiliadis H.1, Mitsionis G.1, Xenakis T.1, Georgoulis A.1 1 University of Ioannina, School of Medicine, Department of Orthopaedics, Ioannina, Greece Objectives: This study compares the endoscopic carpal tunnel release with the conventional open technique with respect to short and long-term improvements of functional and clinical outcomes. Methods: We assessed 72 outpatients diagnosed with carpal tunnel syndrome. Thirty-seven patients underwent the endoscopic method according to Chow and 35 were assigned to the open method. Improvement in symptoms, severity and functionality were evaluated at two days, one week,two weeks and one year postoperatively Changes in clinical outcomes were evaluated one year postoperatively. Complications were also assessed. Results: Both groups showed similar improvement in all but one outcome one year after the release; increase in grip strength was significantly higher for the endoscopic group. However, the endoscopic method showed higher improvement in symptoms and functional status compared with the open method two days, one and two weeks postoperatively. Separate analysis of the questions referring to the pain reveals that the delay of improvement in the open group is due to the persistence of pain for a longer period. Paresthesias and numbness decrease immediately after the operation with comparable rates for both groups. Conclusions: Endoscopic carpal tunnel release appears to be equally safe to open release and provides a faster recovery to the operated patients, at least for the first two weeks, with faster relief of pain and faster improvement of functional abilities. Paresthesia and numbness symptoms subside in an identical manner between the two techniques. At one year postoperatively, both open and endoscopic techniques seem to be equivalently efficient;only the grip strength appears to be slightly increased in the endoscopic treated patients, but with questionable clinical significance.

S149 P11-470 Prospective randomized controlled trial between endoscopic and open carpal tunnel release Ejiri S.1 1 Fukushima Medical University School of Medicine, Orthopaedic Surgery, Fukushima, Japan Objectives: In Japan, Single portal technique with USE system designed by Okutu are most popular method in endoscopic carpal tunnel release. The purpose of this study was to compare outcomes of this single portal endoscopic carpal tunnel release with open carpal tunnel release. Methods: A prospective randomized controlled trial was performed on 101 hands in 79 patients. The endoscopic method was performed in 51 hands in 40 patients, and the open method was performed in 50 hands in 39 patients. All of the patients had clinical symptoms and signs and electrodiagnostic findings consistent with carpal tunnel syndrome. Follow up evaluations with use of measurement of symptom severity on a self-report scale and motor nerve conduction velocity were performed at preoperative and four, twelve weeks postoperative. Results: There was no significant difference in recover of symptom and motor nerve conduction velocity between the two groups. In open group, all patients were recovered at postoperative. On the other hand, in endoscopic group two patients(4%) became worse their symptoms at the four-week examination and one patient(2%) same at the twelve-week, and three patient(6%) became worse their nerve conduction velocity at the four-week. Conclusions: Both method give good result, but the Single portal endoscopic technique with USE system have a risk of transient damage to median nerve function.

P11-665 Adequate conservative treatment head and neck radius fractures at young sportsmen Kuksov V.1 1 Pirogov Clinic City Hospital, Traumatological, Samara, Russian Federation Objectives: Head and neck radius fractures at young sportsmen are not meeting often. They are the typical intra-articular injuries. The medical treatment is depending from the degree of the displacement head radius. The most of the clinicians are given the preference to surgical methods of the treatment, when the displacement is not more than 30 degrees. At that, as a rule, the function of radius is suffering. Rehabilitation and returning to the sport is suspending on 3-4 months. Purpose: To find the adequate treatment at young sportsmen with head and neck radius fractures with the displacement 30 degrees or more and to realize it at early terms after trauma. Methods: During last 10 years we were provided the treatment to 67 young sportsmen with head and neck radius fractures with displacements of the fragments. The age of the patients was from 7 till 12 years old. The boys were 50, the girls - 17. The sport specialization was: gymnastics - 37, wrestling - 15, volleyball -5, handball - 4, basketball and trampoline - on 3 patients. The terms of arrival young sportsmen after trauma were: at first 6 hours - 55, after 12 hours - 7, more than 14 hours - 5. At X-ray examination were founded the following displacements of the head radius (at degrees): 30-40 degrees - 47, 41-61 degrees - 12, 61-70 degrees - 5, 71-80 degrees - 3. At 6 patients was founded traumatic neuritis of the radius nerve. Used own technique - closed manual reposition on an extent. Reposition technique: a surgeon, fixing both hands a hand of the injured extremity of the patient, is carrying out draft on length; an assistant is keeping a hand in the field of an average third of the shoulder. A surgeon, not stopping drafting, is making rotation movements by a forearm from maximal supination up to the maximal pronation. The number of recurrences was 10-15. Stopped reposition by moving forearm in position of the extreme pronation and full unbending in elbow joint. Imposed back plaster. Then was carrying X-ray control. At 62 patients condition of fragments is excellent, displacement was completely eliminated, an axis of radius is correct. Only at 5 patients with fracture-dislocation of head radius, closed reposition was unsuccessful, them was provided operative treatment. At 7-8 days spent was changed plaster, moved forearm in

123

S150 middle-physiological position and under an angle of 90 degrees in elbow joint. The subsequent plaster immobilization was put for 7 days, and then was done the rehabilitation therapy in full volume. Results: Far distant results of the treatment at all 67 patients in terms from 3 till 5 years after trauma are investigated. X-ray-anatomic indices at 62 patients are good and excellent: head and neck radius are at correct configuration, acceleration synostosis. At 5 patients are satisfactory (insignificant bias of the head and decrease in length of the neck). At all 67 patients - full volume of movements at elbow joint (including rotation). Social and sport prognoses are favorable; successfully are going on sport trainings. Conclusions: High effective method of the treatment young sportsmen with head and neck radius fractures, high degree of fragments displacement (more than 30 degree) is the technique offered by us - closed manual reposition on an extend.

P11-675 Synovial folds of the elbow joint. An anatomical study and relationship to osteoarthritis Takahashi Y.1, Ejiri S.2, Otoshi K.3, Konno S.3 1 Fukushima Medical University, Department of Orthopedic Surgery, Fukushima, Japan, 2Fukushima Medical University School of Medicine, Orthopaedic Surgery, Fukushima, Japan, 3Fukushima Medical University School of Medicine, Fukushima, Japan Objectives: There has been several report of radiohumeral synovial fold in the elbow joint, and it might be the cause of lateral elbow pain that is often accompanied by symptom mimicking those of loose bodies of the elbow joint like snapping or catching(Akagi 1988, Antuna 2001, Clarke 1988). However,there are several reports which describe impingement of synovial fold at the posterolateral ulnohumeral joint, or orecranon fossa which caused limitation of full extension and pain. Methods: This report describes the anatomy of synovial fold of the elbow joint noto only radiohumeral fold but also posterior fold and medial fold. Eighty-four elbows (42 male and 42 female) obtained from forty-two donated Japanese cadavers (subjects were age 54 to 98 years, mean years) were used. Synovial fold were divided into 3 parts; radiohumeral fold.Posterior fold, and medial fold, and each part were segmentalized by its location. Macroscopic and microscopic appearances of each synovial fold were observed. Osteoarthritic changes were also evaluated and relationship to each synovial fold was investigated. Results: Posterior fold and medial fold were observed in all specimens. Anterior and posterior part of radiohumeral fold were also observed in all specimens, but lateral part of radiohumeral fold was observed only 43 elbows. Degenerative changes of radiohumeral fold and posterior fold correlate with osteoarthritic change of the elbow joint. Conclusions: In the elbow joint, onto only radiohumeral fold but also posterior fold and medial fold would be the consistent congenital anatomic structure, and both morphological change of synovial fold and osteoarthritic change of elbow joint correlated with the onset of the clinical symptom delivered from synovial fold.

P11-754 Three dimensional anatomy of radius and consequences for radial prosthesis Abrassart S.1, Barea C.1, Hoffmeyer P.2 1 HUG, Orthope´die, Gene`ve, Switzerland, 2 HUG, Orthope´die-Traumatologie, Gene`ve, Switzerland Objectives: Our purpose was to measure the radial geometry and to use threedimensional tool software for surgical practice. We wanted to find a method to reproduce exact measures of one important radial angle for surgery : angle between bicipital tuberosity and radial styloid process. This torsion angle in relation with radial axis will be useful for radial positioning prosthesis. Methods: 12 cadaveric radial bones were harvested from 8 women and 4 men (average age 85). Computed tomographies were performed and allowed a direct view of positioning of the bicipital process and the radial styloid. The bones were set in symmetric and dorsal decubitus position. A Scout view of both radial head and inferior radial epiphysis allows

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 calculating angles. 3D bones reconstructions were computed using Mimics software (Materialize corp.). Axial view was defined by aligning two points on radial articular plan, and the most inferior point of the radial head articular surface. The obtained views were treated using DICO Measure (View Tec corp.). Three points on the radial head articular surface determined the radial head centre. Two points determined the horizontal articular axis: One point was chosen on the radial styloid process and the other one on the bicipital tuberosity. Radial axis (RA) goes through radial styloid process point and the radial head centre. Bicipital axis (BA) passed through bicipital tuberosity point and the radial head centre. Three angles were measured: a between RA and horizontal axis, b between BA and horizontal axis and u between the RA and the BA. Results: The mean value of Angle a is 152 degrees, the mean value of Angle b is 12.5 degrees for the right and 12,41 degrees for the left .The mean value of Angle u is 45 degrees: But these measures seem to be very different between left and right with an increasing of anterior rotation of the bicipital tuberosity on the right side. In our small series, we found big variations between each patient (standard deviation 8.91). Conclusions: These observations have some importance for radial head prosthesis positioning as for synthesis. It is a good indicator of arthrosis deformity.

P11-1013 Injectable synthetic graft augmentation in distal radial fracture management Chambers M.1, Leach W.2, Wilson C.3 1 Western Infimary, Orthopaedics, Glasgow, United Kingdom, 2Gartnavel General Hospital, Orthopaedics, Glasgow, United Kingdom, 3Western Infimary, Glasgow, United Kingdom Objectives: With dorsally displaced fractures of the distal radius being common we wanted to show the effects of augmenting K-wire temporary fixation with a synthetic Injectable graft. There are many different options ways to treat distal radial fractures, and graft is often used. Autologous graft, although the most effective can cause donor site morbidity, We wanted to show the efficacy of a synthetic graft. Methods: We performed a prospective study of 41 patients with unstable dorsally displaced distal radial fractures who were treated surgically and in addition had the dorsal defect grafted with injectable calcium sulphate graft (MIIG). Results: 29 fractures in 28 patients were followed up for 1 year and a further 6 patients had postal questionnaires at 1 year. 7 were lost to follow up. Radiological results showed improvement in angulations and height of the distal radius post operatively which was largely maintained. DASH questionnaires showed a good to excellent outcome in the majority of patients. Two patients experienced erythema but no other complications were attributable to the graft. Conclusions: Patients with significant extravasation of the graft material were observed resorbing this which is a benefit of calcium sulphate graft substitute. Our preliminary observations suggest that is safe and helps maintain fracture reduction and therefore is a useful adjunct to conventional management of these fractures.

P11-1034 Our technique for all inside arthroscopic repair of carpal triangular fibrocartilage Molano Bernardino C.1, Cancelo R.2, Flores F.J.2 1 Ibermutuamur, Orthopaedic Surgery, Shoulder Unit, Sevilla, Spain, 2 Ibermutuamur, Orthopaedic Surgery, Sevilla, Spain Objectives: Report our technique for all inside FTC repair using common simple non specific instrumentation. Methods: FTC tear Palmer I-B in 24 years old woman. Debridement and repair all inside. Instrumentation were regular 19-G 40 mm cannulated needles, Lynvatec-Conmed Spectrum 458 and arthroscopic knot pusher. Results: Repair was completed all inside. Symptoms relieved and return to pre injury activities at 3 months post OP. We present video and pictures of the procedure.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Conclusions: Arthroscopic repair of TFC has been previously described in a percutaneous fashion, knots lying over soft tissues subcutaneously. The technique all inside decreases the risk of ulnar nerve injury secondary to suture entrapment. Our technique can be accomplished with regular instrumentation available in most standard arthroscopic operating rooms.

P11-1295 Minimally-invasive treatment of acute tear of the distal biceps tendon Heikenfeld R.1, Listringhaus R.1, Godolias G.1 1 St. Anna Hospital, Center for Orthopedics and Traumatology, Herne, Germany Objectives: The purpose of this study was to evaluate the results of minimally-invasive treatment of acute distal biceps tendon tears. Methods: 21 patients with acute tear of the distal biceps tendon were surgically treated 7 days (4-15d) after the initial trauma. A surgical technique according to Morrey using two small incisions was used. The distal biceps tendon is found using a 4 cm wide incision proximal of the elbow joint. Fiberwire sutures are used to hold the tendon. A curved hemostat is introduced into the biceps tunnel and a second incision is made right above the Tuberositas radii. The Tuberositas radii is excavated using a high speed drill and the tendon is fixed transosseous to the Tuberositas. To avoid heterotropic ossifications patients were treated with NSAID (Indomethacin) for 14 days. Follow up was done using the score of Ratanen and Orava. After a mean of 28 months with a minimum follow up of 12 months. An isokinetic strength measurement was performed using the Biodex 3 device. Results: 20 patients could be evaluated. 17 patients had a very good and 3 patients a good result according to Ratanen and Orava. Isokinetic strength measurement showed almost equal values of the treated side compared to the contralateral side. 5 cases showed a slight rotational limitation. All patients were satisfied with the functional and cosmetic result. No operation conditioned complications were noted. Conclusions: The surgical treatment of acute tear of the distal biceps tendon leads to good and very good results using the described technique. The advantage is a minimally-invasive approach compared to the standard technique using a single, large incision. We could not find a higher risk for heterotopic ossifications in our group compared to the literature describing a two incision approach.

P11-1460 Functional outcome assessment after open tennis elbow release: what are the predictor parameters? Siddiqui M.A.1, Koh J.1, Kua J.1, Cheung T.2, Chang P.1 1 Singapore General Hospital, Department of Orthopaedic Surgery, Singapore, Singapore, 2Singapore General Hospital, Department of Occupational Therapy, Singapore, Singapore Objectives: To evaluate potential impact of perioperative factors on postoperative outcome after open surgical release for tennis elbow. Methods: This was a retrospective study. We reviewed the case records of all patients who underwent open surgical debridement from period of January 2000-June 2006. Inclusion criteria was all patients who underwent open surgical release for lateral epicondylitis after failing a trial of conservative treatment of analgesia, activity modification and a physiotherapy protocol. Exclusion criteria was patients who had acute trauma to the elbow or a history of elbow subluxation/dislocations. Patients were recalled back for assessment after surgery. Co-morbidities such as diabetes mellitus and associated pathologies of the upper limb were noted. Post-operatively, parameters assessed were pain score (based on the visual analog scale), sagittal range of motion of the elbow, evidence of instability and recurrence. Any other complications of surgery were also noted. Results: There were 37 females and 24 males aged 44.95 years (Range: 22-60; SD 7.34). Five patients had diabetes mellitus (8.1%) and 1 patient (1.6%) had rheumatoid arthritis. The mean duration of symptoms of the patient prior to surgery was 16.8 months (range 2-84, SD 15.3 months). Fifty-five patients (88.2%) had symptoms for 6 months or more prior to

S151 surgery. The average duration of follow-up post-operatively was 23 months (range 0.4-206, SD 34.5 months). Overall, 59 patients (96.7%) said there was improvement of symptom post-surgery. The median grip strengths on the operated and unoperated sides were 25.5KgN (Range 10.5-44, SD 9.02,) and 23.7KgN (Range 9.3341.50, SD 9.10) respectively. Patients with diabetes mellitus had significant higher pain scores. Patients with diabetes had mean pain score of 2.80 while patients without diabetes had mean score of 0.36 (p\0.01, Diff=2.4, 95% CI 1.2-3.7). Grip strength ratios were significantly reduced in patients with diabetes mellitus. Patients with diabetes had a grip strength ratio of 0.91 (ie 91% of the unaffected side) while patients without diabetes had a grip strength ratio of 1.06 (p=0.038, Diff=0.15, 95% CI 0.29-0.01). Moreover, patients with diabetes mellitus had a significantly increased risk of recurrence (p=0.018). Two out of the 5 patients with diabetes mellitus (40% of diabetic patients) had recurrence when compared to 4 out of 52 non-diabetic patients (7.7%). There was a tendency to have increased post-operative pain in patients with ipsilateral shoulder pathologies. Patients with ipsilateral shoulder pathologies had a mean pain score of 1.57 while patients without shoulder pathology had a mean score of 0.43 (p=0.05, Diff=1.2, CI=0.01-2.3). Conclusions: Open release of tennis elbow provides reliable symptomatic relief with no appreciable weakness or instability. Diabetes mellitus appear to have a negative effect on surgical outcome in terms of pain scores and grip strength ratio and this issue should be reiterated during pre-operative counselling. Patients with lateral epicondylitis with ipsilateral shoulder pathologies should be properly worked up prior to surgery.

Pelvis-groin P12-38 The clinical outcome of patients with normal hip arthroscopy DeFranco M.1, Dasilva J.2, Riff A.3, Bush-Joseph C.4 1 Massachusetts General Hospital, Orthopaedic Surgery, Boston, United States, 2SW Washington Medical Center, Orthopaedic Surgery, Vancouver, United States, 3Georgetown University School of Medicine, Washington, DC, United States, 4Rush University Medical Center, Orthopaedic Surgery, Chicago, United States Objectives: The purpose of this study was to report the clinical outcome of a series of patients who had a normal hip arthroscopy despite a preoperative assessment indicating intra-articular pathology. Methods: This study was a retrospective review of all patients treated with hip arthroscopy by the senior author between 2000-2005. The clinical outcome of 14 patients with a minimum of two-year follow-up was defined using the Modified Harris Hip Score, Hip Outcome Score, and SF-12 14. Results: The majority of the patients experienced an acceptable clinical outcome despite not having treatable pathology at the time of surgery. The average MHHS for all patients (14/14) was 90.1 (range 68.2-100). The average HOS for all patients (14/14) was 85.6 (range 40-100) for normal activity and 90 (range 25-100) for sports activity. Based on the HOS, patients rated their hip as normal in 43% (6/14), nearly normal in 36% (5/14), abnormal in 21% (3/14), and severely abnormal in none. According to the SF-12, Twenty-eight percent (4/14) of the patients reported that they were able to do less with their current physical (hip) condition. Forty-three percent (6/14) reported that they were limited at work with their current physical (hip) condition. Ninety-three percent (13/ 14) patients reported that their current physical (hip) condition did not limit their social activity in any way. Conclusions: The cause of debilitating hip pain may be intra- and/or extra-articular pathology. When nonoperative management for either type of hip pathology fails, patients may become candidates for hip arthroscopy. At the time of surgery there may or may not be a definitive pathologic lesion to treat. Based on this case series, patients without definitive intra-operative pathology achieve a satisfactory level of function after surgery.

123

S152 P12-180 Arthroscopic diagnosis and treatment of septic arthritis of the hip Nusem I.1 1 Logan Hospital, Brisbane, Australia Objectives: Arthrotomy is considered the standard treatment for septic arthritis of the hip. This may be complicated by AVN or postoperative hip instability. Arthroscopic treatment of this condition is still not an established technique despite its minimally invasive nature and being associated with low morbidity. Methods: A three portal arthroscopic technique was used for drainage, debridement and irrigation in 13 patients with septic coxarthrosis. Continuous intraarticular irrigation was not performed, nor was decompression drains used. All patients were treated with intravenous antibiotics for three weeks, followed with oral antibiotics for an additional minimum of three weeks. Results: The patients were followed for 1-7 years. Staphylococcus aureus was identified in 10 of the 13 patients. All patients had a rapid postoperative recovery. The mean Harris Hip Score at the last review was 97.5 points. All patients had a full range of motion of the affected hip. No complications occurred with this group of patients. Conclusions: Three directional arthroscopic surgery combined with large volume irrigation is an effective treatment modality in cases of septic arthritis of the hip. It is less invasive than arthrotomy, and offers low post surgical morbidity.

P12-740 Implant survival of metal-on-metal hip resurfacing: a systematic review Hoekstra H.1, van der Weegen W.1, Sybesma T.2, Bos E.1, Poolman R.3 1 St. Anna Hospital, Orthopedic Surgery, Gelrop, Netherlands, 2St. Anna Hospital, Gelrop, Netherlands, 3OLVG and Stichting Medisch Centrum Jan van Goyen, Orthopedic Surgery, Amsterdam, Netherlands Objectives: To evaluate implant survival and functional outcomes of Metal-on-Metal hip resurfacing arthroplasty. Methods: Electronic databases and reference lists were searched from 1988 to March 2009. Identified abstracts were checked for inclusion or exclusion by two independent reviewers. Data were extracted and summarized by one reviewer and verified by a second reviewer. Main study endpoint was implant survival, which we compared with the NICE benchmark. We also evaluated radiological and functional outcomes, failure modes and other adverse events. The quality of evidence was judged using the GRADE system. Results: We identified 268 articles, of which 17 met the inclusion criteria. Data were extracted from 17 articles, totalling 9772 resurfaced hips, providing details on 5 out of 11 resurfacing devices on the market. Maximum follow up was nine years, mean follow up ranged from 0.6 to 7 years. Implant survival ranged from 88.2% to 99.78%. Of the 9772 hips, 283 were revised (2.9%), with fracture of the femoral neck as most frequent failure mode. With implant survival plotted against time, seven studies showed satisfactory implant survival percentages compared to the 3 year NICE entry-benchmark. Conclusions: None of the HRA implants used to date meet the full 10 year NICE benchmark. HRA using the BRH implant is compliant with the NICE ‘entry benchmark’. Longer follow up on revision rates is needed. The quality of evidence is low according to the GRADE classification. Future research has to address the most important failure mode for HRA trying to explain the large variation in the frequency of femoral neck fractures.

P12-1117 Femoroacetabular impingement secondary to slipped capital femoral epiphysis Abellan Guillen J.F.1, Sa´nchez-Alegre M.L.2, Gonzalez-Lopez J.L.3, Lison A.1, Esparza F.4 1 Hospital Morales Meseguer, Orthopedic Surgery, Murcia, Spain, 2 Hospital Gregorio Maran˜on, Radiology Department, Madrid, Spain,

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 3

Hospital Gregorio Maran˜on, Servicio de Ortopedia Infantil, Madrid, Spain, 4Universidad Catolica San Antonio de Murcia, Chair of Sport Traumatology, Murcia, Spain Objectives: Slipped capital femoral epiphysis (SCFE) is a common adolescent hip disorder, with an incidence of 0.2 to 10 per 100,000. This entity usually creates a residual varus deformity in the coronal plane, between the femoral epiphysis and metaphysis. The deformity causes a secondary femoroacetabular impingement (FAI), suggested as the mechanism that leads to early osteoarthritis of the hip in SCFE. Methods: From January 1990 to December 1998, 36 children between 11 and 14 years-old were treated for SCFE. We have reviewed them for clinical and radiological evidence of FAI. Clinical exam (hip range of motion, anterior impingement test of the hip, and Drehmann sign) and radiological study (alpha angle, and anterior offset of the femoral headneck junction measurements), have been performed. In addition, immediate postoperative radiographs were reviewed, evaluating FAI radiological parameters. Results: We have retrospectively reviewed 30 of the 36 patients with minimum follow-up of 10 years. Patients who presented increased alpha angles and low anterior offset in the postoperative x-rays, also do the current study. However, radiological signs of FAI have not been strictly related with clinical criteria. Conclusions: The structural deformities secondary to SCFE produce an abnormal contact between the acetabular rim and the femoral head-neck junction, leading to hip osteoarthritis. Although remodelling of the head-neck junction after a SCFE has been proposed, this study shows that, after mild and severe SCFE, radiological signs of FAI are already present in mid-term follow-up. These signs are not necessarily related to clinical symptoms. Every patient treated for SCFE should be radiologically evaluated for FAI in order to prevent early progression to hip OA.

P12-1431 Hip instability treated with arthroscopic capsuloraphy Schilders E.1, Dimitrakopoulou A.1, Talbot C.1, Bismil Q.1 1 Bradford Royal Infirmary/ Leeds Metropolitan University, Orthopaedic, Bradford, United Kingdom Objectives: Capsuloraphy can be effective in treatment of symptomatic capsular laxity of hip in patients with associated labral or ligamentum teres tears, except with hip dysplasia. Methods: Retrospective case series study reviewing 16 patients with hip instability treated with arthroscopic capsuloraphy. Inclusion criteria for the procedure were: instability, a positive log roll test, demonstration of hip laxity with the image intensifier under general anesthesia. Patients were also assessed for ligamentous hyperlaxity. We made 4 groups based on associated pathology: I. dysplasia (CE\ 20) II. FAI III. Labral tear/ ligamentum Teres Tear IV. Ligamentous Hyperlaxity. Postoperatively patients were assessed for instability, log roll test and 100 point Modified Harris Hip Score. Results: At 2-3 years follow-up, 15/16 had a negative log roll test and no instability. The average preoperative MMHS was 52 points (range 11-74), the average postoperative score at 2-3 years follow-up was 79.1 points (range 20-100). The MMHS scores per group are: Group 1 (n=4) preoperatively 48 points and postoperatively 65; group 2 (n=2) 51 points preoperatively and 93 postoperatively; group 3 (n=7) 63 points preoperatively and 95 postoperatively; group 4 (n=3) 32 points preoperatively and 53 postoperatively. Conclusions: Although arthroscopic capsuloraphy seems to be effective in treating symptomatic capsular laxity, significant differences in outcomes between the groups are observed. Associated labral tears seem to result in better outcomes except in patients with dysplasia. Capsuloraphy may not suffice as a sole procedure to treat instability patients with dyplasia and ligamentous hyperlaxity without other intraarticular pathology.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P12-1433 Hip arthroscopy in the elite athlete Khanduja V.1, Villar R.2 1 Addenbrooke’s - Cambridge University Hospital, Cambridge, United Kingdom, 2The Wellington Hospital, Hip Surgery, London, United Kingdom Objectives: The aim of this study was to assess the role of hip arthroscopy in the management of hip pain in elite athletes. Methods: It is a retrospective study of prospectively collected data. 31 hip arthroscopies were performed on 27 elite athletes. All the patients were assessed pre-operatively with a thorough clinical examination, radiographs and MRI scans. The modified HHS was also recorded for all the patients’ pre and post-operatively. All the operations were performed by the senior author. The patients were assessed at 6 weeks, 3 months, 6 months and a year following the operation. Results: The M:F ratio was 25:2 with an average age of 38 years (Range: 1858) in the study group. 65% of the patients were professional football players followed by cricket and tennis. The patients were referred at an average of 506 days. Two of the 27 patients demonstrated minimal acetabular dysplasia on their pre-operative radiographs. Of the 31 arthroscopies, a primary diagnosis of a labral tear was made in 20, a chondral flap and delamination in 5, early osteoarthritis in 4 and a ligamentum teres tear in 2. Along with the primary diagnosis of a labral tear, twelve of the 20 patients had a secondary diagnosis of chondral injury and or Femoroacetabular Impingement as well. There were no complications reported. Conclusions: Hip Injuries in athletes are reported late and Chondral Flaps and Ligamentum teres tears are under-diagnosed. Most athletes have a good chance to return back to the original level of sport within 4 months following hip arthroscopy. However, about 40% do continue to have some form of pain. Hip arthroscopy is a safe and effective method for diagnosis and treatment of intra-articular disorders in elite athletes.

P12-1441 Adductor-related groin pain in recreational versus high level athletes: comparative study of the therapeutic effect of pubic cleft injections and the role of MRI Schilders E.1, Bismil Q.1, Talbot C.1, Dimitrakopoulou A.1 1 Bradford Royal Infirmary/ Leeds Metropolitan University, Orthopaedic, Bradford, United Kingdom Objectives: We compare a group of recreational athletes and a group of high level athletes with clinical adductor enthesopathy. Pubic cleft injections are a valuable treatment option, but MRI findings and sporting level must be considered. Methods: Retrospective cohort study (n=52) assessing treatment of clinical adductor enthesopathy with pubic cleft injections: in high level versus recreational athletes. The 2 groups were subdivided into athletes with and without MRI enthesopathy. Inclusion criteria: tender enthesis; pain on passive stretching and resisted adduction. Exclusion criteria: osteitis pubis, sports hernia or hip pathology. At 1 year follow-up the effects of the injections were analysed. Results: On the basis of the sporting level and MRI findings the athletes were divided into four subgroups: high level, no MRI enthesopathy (n=7); high level, MRI enthesopathy (n=17); recreational, no MRI enthesopathy (n= 15); recreational, MRI enthesopathy (n=13). At 1 year follow-up 67% of the competitive athletes (n=24) had a recurrence of symptoms compared to 36% of the recreational athletes (n=28) (p\0.001). The recurrence rate amongst the competitive athlete subgroups was 0% in the MRI negative group (n=7) and 94% in the group with MRI enthesopathy (n=17). In the recreational athlete subgroups the recurrence rate was 33% for the MRI negative group (n=15) and 31% in the group with MRI enthesopathy (n=13). Conclusions: Pubic cleft injections are a valuable treatment option in athletes with chronic adductor enthesopathy. The MRI findings determine the efficacy of this procedure in high level but not recreational athletes. A single pubic cleft injection can give one year of relief in symptomatic adductor enthesopathy in high level athletes with a negative MRI scan; but with MRI evidence of adductor enthesopathy only a short term benefit can

S153 be expected. In recreational athletes long term benefit can be expected regardless of the MRI findings.

Knee: ACL

P13-15 An alternative technique for ACL anatomic double-bundle reconstruction: fixation by press-fit to femur and by metallic interference screws to tibia Santos Assis de Melo N.1, Santos Duarte Lana J.F.2, de Oliveira Neri I.D.1, Pidner Neto H.3 1 Santa Casa de Miserico´rdia de Mogi Mirim, Mogi Mirim, Brazil, 2Clı´nica MOR e Hospital Sa˜o Marcos - Uberaba (MG), Mogi Mirim, Brazil, 3 Hospital da Beneficeˆncia Portuguesa de Bele´m (PA), Mogi Mirim, Brazil Objectives: The authors of this study describe a cost-saving technique for anatomic double-bundle ACL reconstruction, using hamstring tendon grafts.The grafts are threaded from femur to tibia, anchored in the femur with Press-fit and attached to the tibia with interference screws. Methods: Stage 1 - Joint Assessment Joint assessment begins after the conventional anterolateral arthroscopy portal (P1) is created. A second portal (P2) is made in the anteromedial region close to the medial border of the patellar tendon. Stage 2 - Graft - harvest and preparation Standard precautions should be taken to avoid severing the tendons because their full length is needed for the graft construct. The ends of each distinct tendon are united with a simple knot, termed the femoral end (FE) of the graft - at the extreme end of the tendon to make optimal use of its length. In order to avoid slippage of the knot, several sutures transfixing the FE are made. Simple stitches are sewn at the tibial ends (TE) of the grafts, rendering them more cylindrical to facilitate their passage through the tunnels. The assistant then measures the cross-sections of both ends of the grafts. TE diameter determines those of the tibial and femoral tunnels. Likewise, FE measurement determines the diameter of the femoral socket wherein the knot will be seated. Stage 3 - Intercondylar notch preparation The next step, with the aid of a shaver and a curette, is to debride the future site that will house the neoligament in the femoral notch. The major anatomic landmark for drilling the femoral tunnels, the so-called ‘‘resident’s ridge’’ (RR) must be preserved. Another anteromedial arthroscopy portal (P3) is created more medially and distally to P2. Stage 4 - Drilling the tibial tunnel The tibial tunnels are drilled under arthroscopic vision through P1, acording standard landmarks. Stage 5 - Drilling the femoral tunnel The femoral tunnels are drilled under arthroscopic vision through P2, RR being the main landmark and should be located posterior to the RR. Furthermore, the femoral posterolateral bundle tunnel (FPL) is always drilled through P3, while the femoral anteromedial tunnel (FAM) may be drilled either through P3 or by using the PLT.The femoral tunnels will be created from the outside to the inside of the joint, initially using drill bits measuring the same diameters as the grafts - cross-sections of the TE having been previously measured. These same tunnels will be re-drilled, this time with drill bits measuring the same diameters as the FE cross-sections of the grafts. On arthroscopic intra-articular viewing, the larger diameter drill can be seen approaching the joint, stopping short at least 1cm distally. Stage 6 - Graft-passage and fixation The PLB and AMB grafts are passed through the tunnels from femur to tibia (in this order).The PLB is secured with a blunt-threaded metallic interference screw with the knee fully extended. The same fixating device is used for the AMB, but with the knee flexed to 30-408. Results: We have performed 36 ACL reconstruction operations using the technique described here. The patient with the longest follow-up is now at 13 months post operation. Conclusions: The authors of this paper do not purport to discuss, in this article, which surgery is best: single-bundle (SB) or double-bundle (DB).

123

S154 The main objective of this article is to present a cost-saving alternative surgical technique for anatomic double-bundle ACL reconstruction while maintaining the quality of graft fixation.

P13-16 The augmentation of independent bundles of ACL reconstruction Maestro A.1, Fdez-Lombardia J.2, Rodriguez L.3, Olay M.J.4, Casero J.4 1 FREMAP, Orthopaedic Surgery, Sports Medicine, Gijon, Spain, 2 FREMAP, Orthopaedic Surgery, Gijon, Spain, 3Hospital Cabuen˜es, Orthopaedic Surgery, Gijon, Spain, 4FREMAP, Orthopaedic Surgery, Oviedo, Spain Objectives: The goal of this paper is to evaluate the clinical outcome after the reconstruction of one or another bundle of the ACL. Methods: We recruited a total of 16 consecutive patients, underwent reconstruction of the AM bundle (8 cases) or PL bundle (6 cases) with the usual ACL reconstruction technique with autologous pes anserinus tendons (in 12 cases the semitendinosus and gracilis tendons were used and 2 cases only the semitendinous was needed), with an age average 27,5±5,8 years, a length of 169.5±6.3 and 72±9.9 Kgrs. weight. The mean followup was 19.07±4.48 months, and the surgery was made by same surgical team. All patients following identical postoperative protocol. The results were evaluated by mean of the objective and subjective IKDC scale. Results: The preoperative evaluation showed 4 cases in group D, 6 in C and 6 in B, and the postoperative 14 included in group A and 2 in B. The mobility were recovery in all cases, with persistence of 1 case of Pivot Shift, functional test and an anterior laxity in group B. There were no differences in complications or time compared to surgical techniques. The subjective IKDC was 81.42 ±4.73 preoperatively and 88.21±4.24 postoperatively. Conclusions: The reconstruction of one or another bundle of the ACL, allows the recovery of stability and functionality of the knee.

P13-20 Dynamometric platforms for evaluating the rotational instability of the knee (PART I) Sanchis-Alfonso V.1, Montesinos-Berry E.1, Baydal-Bertomeu J.-M.2, Castelli A.2, Marin S.2, Garrido D.2 1 Hospital Arnau de Vilanova, Valencia, Spain, 2Instituto de Biomecanica, Valencia, Spain Objectives: Currently, a suitable and reliable, non-invasive method to evaluate rotational stability of the knee in vivo during sport movements, does not exist. We have developed a study using kinetic analysis to evaluate rotational stability under dynamic loading. Methods: 15 recreationally healthy active athletes were recruited for this study. Patients performed jumping with pivoting with internal tibial rotation and external tibial rotation on the dynamometric platform with both limbs. The quantitative results were graphically plotted and we have evaluated the following parameters: loading moment, pivoting moment, moment amplitude, loading slope, pivoting slope, percentage of pivoting with load, loading impulse, pivoting impulse, and maximum body rotation angle. Results: No statistically significant differences were found between both knees during the jumping with pivoting with tibial external rotation test, in the pivoting moment (p=0.627), or in the pivoting impulse (p=0,162). In the rest of the parameters studied we have found statistically significant differences that have been influenced by limb dominance. We haven0 t found significant differences between both knees for the following parameters: pivoting moment (p=0.744), moment amplitude (P=0.245), loading slope (p=0.08), pivoting impulse (p=0.175) and the percentage of pivoting with load (p=0.889) during the jumping with pivoting with internal tibial rotation test. In the rest of the parameters studied there are significant differences influenced by limb dominance. Conclusions: Kinetic analysis is a suitable and reliable, non-invasive method to evaluate rotational stability of the knee in vivo during sport movements.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P13-21 Dynamometric platforms for evaluating the pivot shift in the ACL-deficient and operated knee (PART II) Sanchis-Alfonso V.1, Montesinos-Berry E.1, Baydal-Bertomeu J.-M.2, Castelli A.2, Marin S.2, Garrido D.2 1 Hospital Arnau de Vilanova, Valencia, Spain, 2Instituto de Biomecanica, Valencia, Spain Objectives: Currently, a suitable and reliable, non-invasive method to evaluate rotational stability in vivo in ACL-deficient knees and after ACL reconstruction, particularly during sport movements, does not exist. We speculate that if there is a rotational instability, the patient would avoid reaching a high moment during pivoting activities as a defense mechanism and therefore that the ground reaction moment, as registered by dynamometric platforms, would be reduced. Based on this hypothesis, we have developed a study using kinetic analysis to evaluate rotational stability under dynamic loading. Methods: Twenty-seven recreationally active athletes, one-bundle ACL reconstructed (9), and ACL-deficient (18), were recruited for this study. Patients performed jumping with pivoting with internal tibial rotation and external tibial rotation on the dynamometric platform with both the healthy limb as well as the injured/operated limb. Results: In patients with a chronic ACL tear we have observed a significant decrease (p=0.004) of the pivoting moment during the jumping with pivoting with external tibial rotation test in the ACL-deficient knee compared with the healthy contralateral knee. The kinetic parameters that should be higher due to limb dominance (loading moment, moment amplitude, pivoting slope, loading impulse and maximum body rotation angle) are lower to those of the contralateral healthy knee, when the ACLdeficient knee is the dominant one. However, in the jumping with pivoting with internal tibial rotation test we haven’t found significant differences in any of the parameters that are not influenced by limb dominance. Kinetic analysis using the jumping with pivoting with external tibial rotation test has a sensitivity of 87.5% in evaluating the pivot-shift phenomenon. In patients operated with the technique of a single bundle ACL reconstruction, and with an excellent clinical result, we haven’t observed any significant differences (p=0.16) in the pivoting moment between the operated knee and the healthy contralateral knee. Conclusions: Kinetic analysis has a high sensitivity to evaluate the pivotshift phenomenon with the jumping with pivoting with external tibial rotation test. Our data shows that an anatomical single-bundle reconstruction is sufficient to restore normal knee kinetics during a simulated pivot-shift event.

P13-23 Double bundle ACL reconstruction using autogenous hamstring tendons; outsides in approach and fixation with ligament plateÒ Lee B.1 1 Gachon University, Orthopedic Surgery, Incheon, Republic of Korea Objectives: To evaluate the clinical outcomes at a minimum 1-year following anatomic double bundle ACL reconstruction with using autogenous hamstring tendons fixed with Ligament Plate. Methods: Double Bundle ACL Reconstruction using Autogenous Hamstring Tendons which use outsides in approach for femoral tunnel and fixation with Ligament Plate was done in 50 patients. Semitendinosus tendon was used for the reconstruction of the anteromedial bundle and the gracilis tendon was used for the reconstruction of the posterolateral bundle. For femoral fixation, we used an anteromedial bundle that was suspended in Ligament Plate and a posterolateral bundle linked with Mersilene tape. For tibial fixation, we used double post-tie. The average follow-up period was 16.5 months. Clinical evaluation was done with IKDC score, Lysholm score, stress radiogram, KT 2000 and tunnel enlargement was measured by simple radiogram the Lysholm score was 92.4±6.8 points. For the IKDC score, there were 35 cases of grade A, 14 cases of grade B and 1 case of grade C. Results: The Lachman test was negative for 40 cases, it was grade 1 for 9 cases and it was grade 2 for 1 case and the pivot shift test was negative for 45 cases and it was grade 1 for 5 cases.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 The side-to-side differences with the KT-2000 and the anterior drawer radiogram were 1.3±1.6 mm and 1.3±1.3 mm, respectively. Conclusions: The femoral tunnel enlarged to 1.7±0.6 mm in the anteromedial aspect and 1.6±0.7 mm in the posterolateral aspect, and the tibial tunnel enlarged to 1.2±0.4 mm in the anteromedial aspect and 1.4±0.5 mm in the posterolateral aspect Anatomic ACL reconstruction using autogenous hamstring tendons with outside in femoral approach that are fixed with Ligament Plate showed good clinical results.

P13-24 Comparison of combined ACL reconstruction and high tibial ostetomy and isolated ACL reconstruction in ACL deficient knee with varus alignment Lee B.1 1 Gachon University, Orthopedic Surgery, Incheon, Republic of Korea Objectives: To find a appropriate treatment in ACL defient knee with varus alignment. Methods: We compare the 9 cases of combined HTO and ACL reconstruction with the 13 cases of isolated ACL reconstruction which was done from march 2005 to Feb 2007.Average follow up was 21.8 months in combined group and 20 months in isolated ACL reconstruction group. Clinical evaluation was done with IKDC score, Lysholm score, Tegner score, Cincinatti score, Lachman, pivot shift test, stress radiogram. Results: There was no difference regarding to ROM,IKCD score, Lysholm score,pivot shift, KT2000, stress radiogram. Conclusions: But the Tegner score and Cincinatti score was 4.8,55.6 in combined treatment group and 5.6 and 72.9 in isolated ACL reconstruction group combined treatment using HTO and ACL reconstruction can limit the sport activity than isolated ACL reconstruction in varus alignment knee.

P13-47 Femoral hybrid fixation vs. interference screw fixation alone in ACL-reconstruction - a prospective matched group analysis Wagner M.1, Holtz A.1, Weiler A.2 1 Charite´ - University Medicine Berlin, Arthroscopy and Sports Traumatology, Berlin, Germany, 2Center for Special Joint Surgery, Berlin, Germany Objectives: To clinically compare femoral hybrid fixation vs. interference screw fixation alone in primary hamstring tendon ACL reconstruction. Additionally to find out if there are gender related differences. Methods: Between 1999 and 2001 primary ACL reconstruction was performed in 151 cases using femoral interference screw fixation (group 1). Since 2002 all ACL reconstructions were performed using femoral hybrid fixation with an interference screw and the EndoPearl device (group 2, N = 493). In all cases an identical tibial fixation was used. All cases underwent prospective follow-up with a minimum of two years. Due to exclusion criteria (N = 46) and lost to follow-up (N = 20, FU-rate 87%) 85 cases were included. For comparison 85 patients out of group 2 were selected for a matched group analysis. Matching parameter were: age, gender, ligamentous and intraarticular comorbidity. Additionally, a gender related analysis was performed. Clinical analysis included the IKDC- and Lysholm-score and an instrumented laxity measurement using the KT1000. Statistical analysis was performed using parametric and non-parametric tests (p B 0.05). Results: Before surgery there was no significant difference between the groups. At two years group 2 demonstrated significantly better results for the IKDC-score, the pivot-shift test, and for KT-1000 (group 1: 2.5 ± 1.6 mm; group 2: 1.7 ± 1.4 mm). The Lysholm score was not significantly different. There was no significant gender related difference neither in clinical scoring nor in knee stability. Conclusions: Femoral hybrid fixation of hamstring tendon grafts using an interference screw and the EndoPearl device is superior in terms of instrumented laxity measurements and the overall IKDC-score compared to interference screw fixation alone. Previously reported gender related differences after ACL reconstruction could not be demonstrated in the present study.

S155 P13-66 Evaluation of specific tunnel location in ACL reconstruction: a comparative analysis of bone-patellar tendon-bone single-bundle and semitendinosus-gracilis double-bundle technique in case of optimal and suboptimal clinical results Sadoghi P.1, Kroepfl A.2, Jansson V.1, Mu¨ller P.E.1, Pietschmann M.1, Utzschneider S.1, Fischmeister M.F.2 1 Ludwig-Maximilians-University, Department of Orthopaedics, Munich, Germany, 2AUVA Unfallkrankenhaus Linz, Linz, Austria Objectives: The aim of this retrospective study was to assess and define specific tunnel positions in anterior cruciate ligament (ACL) reconstruction via two tunnel bone-patellar tendon-bone single-bundle (PTB-SB) and four tunnel semitendinosus-gracilis double-bundle (STG-DB) technique by use of 3D-CT scans in case of optimal and suboptimal clinical results. This radiological method takes the femoral rotation of the ACL insertion into account, which can influence the measured place of the tunnel insertion area and thus lead to inaccurate results in standard methods previously published in literature. Methods: The 3-D CT scans of 52 (27 PTB-SB, 25 STG-DB) patients with comparable optimal (PTB-SBo; STG-DBo) and suboptimal (PTB-SBso; STG-DBso) objective clinical results who had been assessed by the objective IKDC Score and KT 1000 arthrometer were analyzed in terms of their tibial and femoral tunnel position by use of 3D-CT scans. All patients with an IKDC \ B, a negative pivot-shift test and a negative Lachmann test met the inclusion criteria for ‘‘optimal’’ clinical results. All patients not fulfilling one of these criteria were excluded and therefore rated as ‘‘suboptimal’’. We manually measured tunnel position in all four groups in a coordinate system on reconstructed 3D-CT scans. Correlation of tunnel position in relation to optimal and suboptimal clinical results in both techniques was evaluated. Results: We could demonstrate different tunnel positions according to our optimal and suboptimal clinical results in all four groups of the PTB-SB and STG-DB technique. In case of optimal clinical results, the mean tibial and femoral tunnel values in the PTB-SB group were T (5.1; 5.5) and F (2.9; 1.6). Tibial tunnel values of the antero-medial bundle in the STB-DB group were Tam (5.2; 6.6) and values of the postero-lateral bundle were Tpl (5.2; 4.4). The femoral tunnel values of the antero-medial bundle were Fam (2.5; 1.6) and the values of the postero-lateral bundle were Fpl (4.7; 2.9). In case of suboptimal clinical results, the mean tibial and femoral tunnel values in the PTB-SB group were T (4.6; 4.9) and F (3.4; 1.7). Tibial tunnel values of the antero-medial bundle in the STB-DB group were Tam (5.1; 6.1) and values of the postero-lateral bundle were Tpl (5.0; 4.5). The femoral tunnel values of the antero-medial bundle were Fam (2.9; 1.5) and the values of the postero-lateral bundle were Fpl (5.2; 1.5).We evaluated the use of our radiological approach with appropriate inter and intraobserver coefficients. Measurements had been performed from two orthopaedic surgeons two times with a break of two weeks in between. We calculated an average value of these four measurements. Conclusions: We conclude, that our approach satisfies the description of the intercondylar notch more precisely than previous X-ray based studies. Our method could demonstrate the correlation of tunnel positions in the PTB-SB and STG-DB technique with optimal (PTB-SBo; STG-DBo) and suboptimal (PTB-SBso; STG-DBso) clinical results. We proved that better clinical outcome was achieved when the graft placement was anterior enough in the tibia and posterior enough in the femur. According to our measurements, this is evident in case of the single bundle in PTB reconstructions and in case of the antero-medial bundle in the STG-DB technique.

P13-67 Comparison of trans-tibial and anteromedial portal approach in femoral tunneling of anterior cruciate ligament reconstruction Kim K.T.1, Wang L.2 1 Dong-Eui Medical Center, Orthopaedic Surgery, Busan, Republic of Korea, 2Dong-A University Hospital, Orthopaedic Surgery, Busan, Republic of Korea Objectives: Recent development and advances in arthroscopic surgical techniques for Anterior Cruciate Ligament (ACL) reconstruction have led

123

S156 to the ideal location for the isometric point from 10 o0 clock (in right knee) and 13:30 (in left knee) to 10:30 (in right knee) and 14 o0 clock (in left knee) in the frontal plane. This study was performed to compare operative methods and the radiologic results of femoral tunnels made through the tibial tunnel(trans-tibial approach) and the anteromedial portal. Methods: From January 2003 to May 2004, one-hundred reconstructions of anterior cruciate ligament were performed. Group I (femoral tunnel through tibial tunnel) was composed of 50 cases and group II (femoral tunnel through anteromedial portal) was consisted of 50 cases. The study was performed to compare the radiographic results of femoral tunnels made through the tibial tunnel and the anteromedial portal and operative methods. Results: In operative methods at Group II, femoral tunnel was made more easily at isometric point than Group I, a good visual field was achieved because 100 flexion of knee, they can be reduced risk of posterior cortical breakage and tunnel-graft mismatching and decreased divergence of femoral interference screw in radiology (P\0.05). The angle between femoral tunnel and longitudinal axis of ACL was increased at Group II. Conclusions: Anteromedial portal technique was more useful in ACL reconstruction for femoral tunnel toward 10 o0 clock to 10:30(in right) or 1:30 to 2 o0 clock(in left).

P13-80 Avoiding tunnel collisions between fibular collateral ligament and ACL posterolateral bundle reconstruction Camarda L.1, D’Arienzo M.1, LaPrade R.2 1 University of Palermo, Department of Orthopaedic and Traumatology, Palermo, Italy, 2University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States Objectives: Double bundle anterior cruciate ligament (DB-ACL) reconstructions are a recognized technical alternative to primary ACL reconstruction which could restore knee kinematics closer to the normal knee. However, with this procedure, the posterolateral (PL) ACL tunnel is placed more horizontal and closer to the primary posterolateral corner (PLC) structures. The purpose of this study was to evaluate the risk of tunnel collisions of the FCL and PL-ACL tunnels during a combined DB-ACL reconstruction and FCL reconstruction. Methods: Thirty-six 4th generation synthetic femurs (Pacific Research Laboratories, Vashon, WA) were utilized. Eighteen femurs were medium size while other eighteen were large size, with a lateral condylar width of 2.8 and 3.6 cm respectively. Each femur was anchored to a custom-made device to ensure reproducibility of the reconstruction tunnel directions. The femoral PL-ACL bundle attachment point was anatomically located on the lateral condyle and the correct insertion site was confirmed through a lateral X-ray view. Two different exit points of the guide-wire on the lateral femoral cortex were chosen on each femur. This was performed to simulate different PL-ACL tunnel trajectories that could be obtained through the accessory anteromedial portal. In this way, an anterior (A) or a posterior (P) PL bundle tunnel was created for each femur with an angulation of 30 and 32, respectively, in the anteroposterior x-ray view and 53 and 30, respectively, in the lateral view. At this point, a 7 mm reamer was passed over the guide-wire and a PL-ACL tunnel was then created breaching the lateral cortex of the femur. A similar technique was used to direct the FCL femoral reconstruction tunnel. The neutral position (0, 0) was considered when the guide-wire was placed parallel to the distal and posterior condylar line. After the neutral position, different guide-wire orientations were created using 20 intervals in both the coronal and axial planes. At this point, a 9 mm tunnel was reamed over the guide-wire at a depth of 25 and 30 mm. Each tunnel was then filled using an epoxy resin augmented with BaSo4 and a CT was performed on each synthetic specimen. Furthermore, 3D images were obtained and the distance between the PL tunnel and the FCL tunnel was calculated. Furthermore, different tunnel collisions were observed and recorded. Results: No collisions were observed when the FCL tunnel was reamed parallel to the condylar line and with an axial deviation of 20 and 40. This was observed for both PL-ACL orientation and for both 25 and 30mm FCL tunnel depth.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 However, when the FCL tunnel was reamed in neutral position with no coronal and axial deviation (0,0), tunnel collision occurred at the femoral notch and close to the PL-ACL tunnel origin. This was observed for both medium and large femurs and for 25 and 30 mm FCL tunnel depths. A collision rate of 92% was observed when the FCL tunnel was directed proximally at 20 and 40 of axial angulations, for both 25 and 30 mm FCL tunnel depths. However, with regard to the two PL-ACL tunnel orientations, the collision rate decreased from 100% using an anterior PL-ACl tunnel to 83% using a posterior PL-ACL tunnel. Conclusions: Our results show that the risk of tunnel collision during a combined DB-ACL and FCL reconstruction could be reduced. This could be obtained directing the FCL tunnel anteriorly with axial angulations of 20 or 40 and limiting proximal angulation of the tunnel.

P13-82 Anatomic double bundle ACL reconstruction using a bone-patellar tendon-bone autograft: a cadaveric study Pujol N.1, Beaufils P.2, Boisrenoult P.3 1 Hopital Andre Mignot, Orthopaedic Department, Le Chesnay, France, 2 Centre Hospitalier de Versailles, Hopital Andre Mignot, Le Chesnay, France, 3Hopital Andre Mignot, Chirurgie Orthope´dique, Versailles, France Objectives: Conventional double bundle anterior cruciate ligament (ACL) reconstruction are using hamstrings autografts. This article describes the feasibility of an original arthroscopic double bundle ACL reconstruction technique using a bone patellar tendon bone (BPTB) autograft. The aim of the study was to state the reproducibility of the BPTB graft harvesting, the relationships of the femoral tunnels, and to settle technical keypoints for its use. It is a descriptive anatomical study. Methods: Dissections were performed and measurements taken on 10 fresh frozen cadaveric knees after arthroscopic ACL double bundle reconstruction procedure. A rectangular patellar bone block (12 mm), with a double strand patellar tendon (5 and 7 mm), and a double tibial bone block was harvested. The femoral anteromedial tunnel was made using an all-inside technique, by the anteromedial portal. The femoral posterolateral tunnel was made using an outside-in technique, with a 30 degrees divergence between both tunnels. A single tibial tunnel was drilled. The graft was passed through the tibial tunnel, and the bundles were separately tensioned and fixed with three bioabsorbable interference screws. The femoral AM bone block was fixed by the anteromedial portal. Secondly, the tibial bone block was fixed in an oblique manner in order to mimic the ACL orientation with the knee at 30 of flexion. The femoral PL bone block was fixed at the end with the knee in full extension. Results: No complications occured while harvesting the graft. The reconstruction was always performed. The divergence between femoral tunnels was between 30 and 35. The cortical bone bridge was always intact, between 1 and 2 mm. Some difficulties occured in 4 cases to put the two blocks into their femoral tunnels. There was a fracture of the PL bone block in one case, during the fixation by the screw. No iatrogenic injury was seen in the lateral side of the femoral condyle. Conclusions: This technique with a BPTB autograft for double bundle ACL reconstruction is feasible in a cadaveric model, but technically demanding. Bone-tendon-bone fixation allows a good primary fixation and tunnel filling. For BPTB graft users this technique is an alternative to double bundle reconstruction with hamstring tendons. Nevertheless, donor site iatrogenic risks and comparative clinical results have to be evaluated in further studies.

P13-98 ACL femoral tunnel length comparing anteromedial portal versus outside-in technique Lubowitz J.1, Konicek J.2 1 Taos Orthopaedic Institute Research Foundation, Taos, United States, 2 Arthrex, Department of Research, Naples, United States Objectives: Resurgent interest in anatomic ACL reconstruction has led to arguments in favor of independent drilling (AM portal or outside-in

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 technique) of the ACL femoral tunnel. However, a reported risk of AM portal technique for ACL femoral tunnel creation is short tunnel length which can result in reduced length of tendon graft within the femoral bone tunnel. The purpose of this study is to measure ACL femoral tunnel lengths comparing AM portal and outside-in techniques. Our hypothesis is that outside-in technique will result in greater ACL femoral tunnel intraosseous distance than AM portal technique. Methods: After sample size estimate, ACL femoral tunnels were drilled into 12 cadaveric knees using anteromedial portal technique and then outside-in technique on each specimen. Pin intraosseous distance was measured in mm using a 3D Microscribe digitizer. Results: AM portal technique mean ACL femoral tunnel distance was 30.5 mm. Outside-in technique mean ACL femoral tunnel distance was 34.1 mm. The difference is significant (p = 0.04). Conclusions: Our results demonstrate that outside-in technique for creating the ACL femoral tunnel results in longer tunnel lengths than AM portal technique for creating the ACL femoral tunnel. This finding may have clinical relevance for surgeons who desire to perform independent, rather than transtibial, drilling of the ACL femoral tunnel and desire adequate length of tendon graft within the femoral bone tunnel.

P13-99 ACL tibial guide pin accuracy and surgical precision: comparing 3.0 mm and 2.4 mm guide pins Lubowitz J.1, Konicek J.2 1 Taos Orthopaedic Institute Research Foundation, Taos, United States, 2 Arthrex, Department of Research, Naples, United States Objectives: Inaccurate ACL tibial guide pin placement is inefficient, requiring intraoperative, corrective steps. The purpose of this study is to evaluate accuracy of a 3.0 mm diameter ACL tibial guide pin versus a standard, 2.4 mm, drill-tipped guide pin. Our hypothesis is that the 3.0 mm guide pin is more accurate. A secondary purpose is to evaluate surgeon precision in identifying the true (anatomic) center of the ACL tibial footprint using arthroscopic visualization. Methods: Five matched pairs of cadaveric knees were disarticulated, leaving a well-defined footprint of the ACL on the tibial plateau. The tibial footprint was digitally recorded by a bioengineer, and the true center of the footprint was calculated. Next, using arthroscopic visualization, a surgeon identified and marked his estimation of the true center of the ACL tibial footprint. This mark was then digitally recorded by the bioengineer, and compared to the calculated center, allowing quantification of surgeon anatomical precision. Finally, under arthroscopic visualization, the surgeon was given one attempt to aim and drill the guide pin to his mark. Pin position was digitally recorded; distance of the drill pin from the mark quantifies drill pin placement accuracy. Results: Mean accuracy for the 3.0 mm guide pin was 2.87 ± 1.19mm versus 6.98 ± 1.29mm for the 2.4 mm pin. The difference is significant (p=0.005). Surgeon anatomical precision was 3.32 ±2.10 mm. Conclusions: Our results demonstrate that a 3 mm ACL tibial guide pin is significantly more accurate than a 2.4 mm diameter pin. The 3 mm pin accuracy is within the range of surgeon precision; the 2.4 mm pin accuracy is not. Pin accuracy and surgeon precision are clinically relevant measures because anatomic tunnel placement is a determinant of ACL reconstruction outcome.

P13-105 Anatomical single-bundle anterior cruciate ligament reconstruction with a transverse femoral fixation can be performed safely: a cadaveric study Gelber P.E.1, Torres R.2, Gonzalez G.1, Vilches F.1, Monllau J.C.1 1 Hospital de la Santa Creu i Sant Pau, Department of Orthopaedic and Traumatology, Barcelona, Spain, 2Hospitals IMAS Barcelona, Department of Orthopaedic and Traumatology, Barcelona, Spain Objectives: To evaluate if a transverse femoral fixation system, originally developed for reconstruction of the anterior cruciate ligament (ACL) at the over the top position, can be safely used when performing an anatomic

S157 ACL reconstruction from the anteromedial portal, comparing two different length of the femoral tunnel. Methods: An ACL was reconstructed arthroscopically with a hamstring graft in twenty-two fresh cadaveric knees. The femoral tunnel was drilled anatomically in all cases. Knee flexion angle was set at 1108. Femoral fixation was performed with a cross-pin system. A 30 mm length femoral tunnel was performed on eleven knees (group A). In the remaining eleven knees, the femoral tunnel was drilled as long as each lateral condyle permitted (group B). The relationships, in both groups, between the crosspin and the lateral collateral ligament (LCL), popliteus tendon (PT), articular cartilage and peroneal nerve were compared. Results: In 5 group A cases, the cross-pin was placed either through the LCL or between the LCL and PT, whereas it was always posterior to the LCL in group B (p = 0.035). The cross-pin was closer to the articular cartilage in group A (7.14 mm vs 16.9 mm; p \ 0.001). The minimal distance to the peroneal nerve in all specimens was 23.89 mm. Conclusions: Hamstring graft fixation with a cross-pin system from the anteromedial portal with a 30 mm femoral tunnel presents a high risk of injury to the LCL. The femoral tunnel should be drilled as long as possible. A long femoral tunnel is required for safe transverse femoral fixation in an anatomical ACL reconstruction.

P13-108 Meniscal tears in anterior cruciate ligament-deficient knees: effects of tibial slope on meniscal tear Choi C.J.1, Choi Y.J.1, Choi C.H.1 1 Gangnam Severance Hospital, Seoul, Republic of Korea Objectives: The purpose of this study is to evaluate the incidence of meniscal tears in patients with chronic ACL-deficient knees, and to determine the effects of posterior tibial slope on meniscal tears in ACLdeficient knees. Methods: We reviewed 174 patients with a mean age of 30.7 years who underwent ACL reconstruction for chronic ACL tears. The presence of meniscal tears and locations was analyzed. In addition, the slope of the tibial plateau was measured from a lateral radiograph. Results: The incidence of a medial meniscus tear was 44% (77/174), and that of a lateral meniscus tear was 35% (61/174). The posterior horn of the medial meniscus (n = 49) was the most common site of the tear followed by the posterior horn of the lateral meniscus (n = 42). The mean tibial slope in patients with medial meniscal tears was 11.4 ± 3.0, whereas that for patients who had no medial meniscal tears was 9.8 ± 2.4. This difference was statistically significant (p = 0.000). Conclusions: In chronic ACL-deficient patients, posterior horn tear of the medial meniscus were the most common type of tear. Posterior tibial slope was larger in patients with medial meniscal tears. An increased tibial slope may be a risk factor for secondary medial meniscal tears in ACL-deficient knees.

P13-110 MRI analysis of associated injury in acute ACL rupture Park H.1, Yoo H.1, Moon S.1 1 Dankook University College of Medicine, Orthopaedic Surgery, Cheonan, Republic of Korea Objectives: To evaluate MRI of associated injury in acute ACL rupture. Methods: Between February, 2003 and October, 2008, 87 MRIs taken within 2 weeks after injury which were found to have complete rupture of ACL through arthroscopic examination were reviewed. Results: Among 87 cases, joint effusion was found in 79 cases(91%), MCL rupture in 30 cases(34%), MM tear in 22 cases(25%), LM tear in 17 cases(20%), LCL rupture in 14 cases(16%), medial retinaculum injury in 6 cases(7%), vastus medialis injury in 6 cases(7%), popliteus injury in 5 cases(6%), lateral retinaculum injury in 4 cases(5%) and vastus lateralis in 2 cases(2%). Bone bruse was present at the lateral femoral condyle in 42 cases(48%), at the lateral tibial plateau in 32 cases(37%), at the medial tibial plateau in 16 cases (18%) and at the medial femoral condyle in 6 cases (7%).

123

S158 Conclusions: Attention must be paid in clinical diagnosis and subsequent treatment decision of acute ACL rupture because there can be many associated injuries.

P13-122 Effects of tibial tunnel position on the outcomes of ACL double bundle reconstruction Tomihara T.1, Yoshida G.1, Hashimoto Y.2, Shimada N.1 1 Shimada Hospital, Orthopaedic Surgery, Habikino, Japan, 2Osaka City University Graduate School of Medicine, Orthopaedic Surgery, Osaka, Japan Objectives: In ACL reconstruction, the position of the bone tunnel affects the clinical outcomes. If the position of the tibial tunnel is too anterior, there is a risk of roof impingement in knee extension. However, a smaller graft in double bundle reconstruction might diminish the risk of roof impingement compared to the more bulky graft in single bundle reconstruction. Therefore, we have changed the tibial tunnel position of both the anteromedial bundle (AMB) and the posterolateral bundle (PLB) anteriorly in ACL double bundle reconstruction since 2006. The tibial guidewire of the AMB was passed near the extension of the Blumensaat’s line (B line) in knee extension without the off-set between the B line and the guidewire of the AMB. The purpose of this study is to investigate the effect of tibial position on ACL double bundle reconstruction, comparing the clinical outcomes which we found in 2006 (Group A) to those which we found with the off-set between B line and the tibial guidewire of AMB in 2005 (Group B). Methods: There were 51 patients in Group A and 48 patients in Group B. An average age at operation was 27.8 (13-56) years in Group A and 27.9 (14-57) years in Group B, and an average follow-up period was 23.7 (1832) months in Group A and 27.3 (18-47) months in Group B. Femoral tunnel was made through the far antero-medial portal for all patients in both groups. According to the method by Buzzi et al., the respective intersections of tibia plateau with B line, and the position of the tibial guidewire in knee hyperextension was measured on intra-operative lateral X-ray. KT-1000 measurement (manual max), IKDC subjective assessment, Lysholm score, limitation of knee extension (over 10 degrees) and re-injury were evaluated. Results: B line and PLB tibial guidewire intersected tibia plateau at 29.0 ± 4.8% of its width from the anterior cortex of the tibia in Group A and 29.1 ± 5.2% in Group B (p=0.83), and at 46.6 ± 5.4% in Group A and 48.6 ± 5.4% in Group B (p=0.07), respectively. However, the position of the AMB tibial guidewire in Group A (33.0 ± 4.9%) was significantly anterior (p=0.007) than that in Group B (35.8 ± 5.2 %). Anterior laxity measured by KT-1000 was 1.0 ± 1.1 mm in Group A and 0.8 ± 1.0 mm in Group B (p=0.54). There was no significant difference between the IKDC subjective assessment (92.5 ± 9.3 points in Group A and 90.0 ± 10.5 points in Group B) or Lysholm score (90.5 ± 6.8 points in Group A and 89.8 ± 8.4 points in Group B). All patients had no limitation of knee extension. Two patients (3.9%) in Group A and 4 patients (8.3%) in Group B suffered re-injury (p=0.43). Conclusions: The advantage of a more anterior position of tibial tunnel is to keep the length of intra-articular graft longer and to decrease the elongation rate of the graft during knee motion. The disadvantage is the roof impingement. A smaller graft in double bundle reconstruction might overcome this disadvantage. These results indicated that the clinical outcomes were similar in both groups, and the incidents of re-injury were lower in Group A.

P13-131 Comparison of outcomes in second-look arthroscopy after ACL reconstruction using auto-hamstring tendon and allo-tibialis anterior tendon Park H.1, Yoo H.1 1 Dankook University College of Medicine, Orthopaedic Surgery, Cheonan, Korea, Republic of Objectives: To evaluate status of reconstructed ACL and changes around graft through second-look arthroscopy after arthroscopic reconstruction of the ACL with autograft tendons.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: Between Jun. 2003 and Feb. 2007, the second look arthroscopy was performed on 22 cases. Second-look arthroscopy was conducted on average 15.1(7*31) months after reconstruction. 15 cases received hamstring tendon autograft, 7 cases received bone-patellar tendon-bone autograft. We measured graft tension using displacement by probing, synovial coverage by visual analysis at second-look arthroscopy. The assay in Lysholm score, Lachman0 s test and KT-2000 arthrometer were evaluated status of reconstructed ACL. Results: The hamstring tendon group showed normal tension in 11 cases and lax tension in 4 cases. The patellar tendon group showed normal tension in 3 cases, lax tension in 2 cases and partial tear in 2 cases. In the hamstring tendon group, synovial coverage was good in 11 cases, half in 3 cases and pale in 1 case, and the bone-patellar tendon-bone group was good in 4 cases and half in 3 cases. There was statistical improvement in Lysholm score, but statistical insignificance difference in KT-2000 arthrometer results. Conclusions: The hamstring tendon group was superior to the bonepatellar tendon-bone group in second-look arthroscopy after ACL reconstruction with autograft tendons results, but long term follow up will be necessary to further evaluate results.

P13-137 Correlation between osteoarthritis and treatment method in isolated total ACL rupture. A 20 year follow-up results of operative versus non-operative management Widuchowski W.1, Widuchowski J.1, Faltus R.1, Lukasik P.1, Kwiatkowski G.1 1 District Hospital of Orthopedics and Trauma Surgery, Depart. of Knee Surgery, Arthroscopy and Sports Trauma, Piekary Slaskie, Poland Objectives: The main consequence of ACL rupture might be development of early osteoarthritis. We already know that ACL reconstruction (ACLR) does not prevent osteoarthritis, furthermore to date no one proved that any form of treatment more effectively prevents the development of OA. Some studies demonstrated that ACLR limits the severity of OA, whereas others suggest that it may be even increased by this surgery. The correlation between conservative treatment and development of OA is also unclear. The aim of this study was to compare the development of OA during the long-term follow-up in operated and non-operated patients. Methods: 127 patients with an acute isolated total ACL rupture were followed during a 20 year period. In 66 cases ACLR was performed and in 61 cases patients were treated non-operatively. All patients were examined clinically, radiologically and evaluated with use of Lysholm and IKDC scores. Radiological assessment was performed according to KelgrenLawrence scale. Results: According to Lysholm score and IKDC objective score operative group (OP) performed better, but the difference was not significant (p[0.05). Patients of non-operative (NON OP) group demonstrated significantly better results according to IKDC subjective score (p\0.05). Osteoarthritic changes were observed in both groups (Grade 1 and more): OP (47%), NON OP (51%). In operative group more patients developed Grade 2 and more of OA: OP (39%); NON OP (35%). The difference was not significant (p[0.05). Regardless of the treatment method degenerative changes were located mostly in the medial compartment. In both groups more severe changes were correlated with further injuries (menisci, re-ACLR). Conclusions: The management of ACL rupture may influence the development of OA. However, our study confirms that none of the methods (operative, non-operative) more effectively prevents the development of OA and limits its severity. Moreover, in the long term, ACL reconstruction may in certain circumstances increase both: appearance and severity of OA.

P13-141 Anatomical double-bundle anterior cruciate ligament reconstruction restores the correlation between closed kinetic chain muscle strength of the leg and jumping performance Tohyama H.1, Ueda M.2, Chiba T.2, Yuri M.2, Ikoma K.3, Yasuda K.1

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 1 Hokkaido University School of Medicine, Dept. of Sports Medicine, Sapporo, Japan, 2Hokkaido University Hospital, Rehabilitation Division, Sapporo, Japan, 3Hokkaido University School of Medicine, Dept. of Rehabilitation Medicine, Sapporo, Japan Objectives: The jumping is a fundamental skill that is required in many sports activities. Jump tests are widely used to measure the outcome of a training program as a functional measure of an athlete0 s readiness to return to sport following a sports-related injury. Several studies showed that decreased jumping performance accompanies the rupture of anterior cruciate ligament (ACL). We previously reported that closed kinetic chain (CKC) strength of the leg with the ACL-deficiency is correlated poorly with jumping performance compared with that in the normal control subjects. We also showed that anatomical double-bundle ACL reconstruction restores rotational stability of the knee as well as the anterior stability. Therefore, we have hypothesized that anatomical double-bundle ACL reconstruction restores the correlation between CKC muscle strength of the leg and jumping performance. The purpose of the present study was to test this hypothesis. Methods: We totally evaluated 54 subjects with 34 males and 20 females and average age of 24 years. These subjects were classified into the following three groups; the control (Control) group (n=16), the ACLdeficient (ACLD) group (n=22), and the ACL-reconstructed (ACLR) group (n=16). In the ACLD group, all subjects underwent ACL reconstruction due to symptomatic instability within two weeks after the present examination. In the ACLR group, subjects were evaluated at 12-18 months after double-bundle anatomical ACL reconstruction using hamstring tendon grafts. We measured the isokinetic strength measurement and the jumping performance in these subjects. Isokinetic peak strength for the knee extensors in OKC was measured using KIN-COM AP (TN, USA). Isokinetic peak strength for the hip, knee, and ankle extensors in CKC was measured using Strengthergo.240 (Mitsubishi Electric Co., Tokyo, Japan). The one-leg vertical jump and the one-leg long jump tests were also assessed. Pearson’s correlation analysis was performed for statistical analysis. Differences were considered statistically significant at P \ .05. Results: The correlations between CKC strength and the vertical and the long jump performances were weaker in the ACLD group (vertical jump: R=.544, long jump: R=.520) than in the Control group (vertical jump: R=.781, long jump: R=.864). CKC muscle strength in the ACLR group is more highly related to jumping performance (vertical jump: R=.880, long jump: R=.921) than that in the ACLD group, while the correlation analysis showed that OKC and CKC strengths in all groups were significantly correlated with vertical jump performance and standing long jump performance. Conclusions: The present study suggested that ACL deficiency weakens the correlation between CKC strength and jumping performance and that the anatomical double-bundle ACL reconstruction restores the correlation between CKC strength and jumping performance.

P13-142 Biomechanical evaluation using one hamstrings tendon for ACL reconstruction: a human cadaveric study Zamarra G.1, Fisher M.2, Woo S.L.-Y.3, Cerulli G.4 1 IOTI Arezzo, Arezzo, Italy, 2Musculoskeletal Research Center Swanson School of Engineer, Pittsburgh, United States, 3University of Pittsburgh, Dept. of Bioengineering, Pittsburgh, United States, 4University of Perugia, Perugia, Italy Objectives: Harvesting both the semitendinosus and gracilis tendons as a quadruple autograft for anterior cruciate ligament (ACL) reconstruction has negative impact on muscle strength as well as knee function and stability. With a new ‘‘All-inside’’ technique, using only one hamstrings tendon (semitendinosus or gracilis) is possible because of a reduction in length requirements. The research question of this in-vitro study was whether the use of only one hamstrings tendon (semitendinosus or gracilis) could restore knee kinematics and in-situ force in the ACL to the level of an intact knee. Hypothesis: A shortening of the length of the tendon graft in the ‘‘Allinside’’ technique should increase the effective stiffness of the graft. Thus, we hypothesize that a single hamstring tendon, used in triple or quadruple

S159 strands, can restore knee kinematics and in-situ force in the ACL to those of the intact knee. Methods: Ten human cadaveric knees were tested in the following conditions: 1) intact, 2) ACL-deficient, and 3) ACL reconstruction with the ‘‘All-inside’’ technique using the a) single semitendinosus tendon graft, or b) single gracilis tendon graft. Using a robotic testing system, externally applied loads, i.e. 4) an anterior tibial load of 134-N and 5) combined rotatory loads of 10-Nm valgus and 5-Nm internal tibial torques, were applied. The multiple degrees of freedom knee kinematics and the in-situ forces in the ACL and ACL grafts were determined. Results: In response to an 134-N anterior tibial load, the use of either the single semitendinosus or single gracilis graft could restore anterior tibial translation to within 1.3 mm of the intact knee. The in-situ forces in the two grafts were not significantly different from those of the intact ACL. Under the combined rotatory loads, both grafts could restore knee kinematics as well as the in-situ force in the grafts to the level of the intact ACL. Conclusions: The ‘‘All-inside’’ technique using either the semitendinosus or gracilis tendon for ACL reconstruction could satisfactorily restore knee kinematics and the in-situ forces in either graft to those for the intact ACL. These data also compared favorably to published results using both hamstrings as a quadrupled tendon autograft. Clinical Relevance: These results in human cadaveric knees support the clinical findings that the ‘‘All-inside’’ procedure with only a single hamstrings tendon graft for ACL reconstruction to be effective in restoring initial knee stability and graft function post-operatively.

P13-144 Immunohistological changes of the tibial insertion site of anterior cruciate ligament in the organogenetic period of the rat Noyama M.1, Soejima T.1, Kanazawa T.1, Noguchi K.1, Tabuchi K.1, Nagata K.1 1 Kurume University, Orthopaedic Surgery, Kurume, Japan Objectives: Following ACL reconstruction, achieving bone-grafted tendon integration with sufficient strength can accelerate rehabilitation and improve clinical outcome. For bone-grafted tendon integration, it is thought that reproduction of four-layer structure involving cartilage tissue would be ideal. However, it is well known that the structure between bone and grafted tendon after ACL reconstruction differs from the normal structure. On the other hand, to evaluate natural ACL development between bone and tendon can provide effective information to reproduce more accurate histological structure, but few reports examined. In this study, we evaluated natural histological development in the native ACL insertion. Methods: Study design: Pregnant Wistar rats were sacrificed and embryos were harvested at 17 and 21-days post conception (dpc). Neonatal, 1-, 2-, 3-, 4-, 5.5- and 6.5-week old rats were sacrificed and bilateral knees were removed, using a minimum of 3-6 rats per each point. Tissue processing: Specimens were embedded in paraffin blocks. Sections were cut in the sagittal oblique plane at 5lm. Histology: Sections were stained with hematoxylin and eosin (HE), safranin-O and toluidine blue. Immunohistochemistry: We used antibodies against collagen types I, II, III and X and S-100 protein. Results: During gestation, spindle-shaped cell aggregations and irregular alignment were seen in the ACL insertion, but fibroblasts and associated extracellular matrix became organized after birth, and some hypertrophic cartilage-like cells were seen in the insertion adjacent to the epiphyseal nucleus. Up to the second week old, the ligament was directly attached to the epiphyseal nucleus, and no 4-layer structures were seen. However, from the third week after birth, as ossification of the epiphyseal nucleus ceased, anchoring of fibers consisting of type I collagen and a 4-layer structure-

123

S160 like tissue involving cartilage, and resembling normal adult ligament became evident. Type II collagen was strongly expressed in the extracellular matrix around hyaline cartilage was noticeable. As enchondral ossification was proceeding, expression became weak, although expression was still strong at the articular cartilage. At the same period, the type X collagen stained and the unstained hypertrophic cells were coexistent at the insertion. In terms of S-100 protein staining, the coexistence was seen similarly. Conclusions: While enchondral ossification is proceeding in the epiphyseal nucleus, only hypertrophic chondrocytes around the insertion site maintain the cell-form itself. Simultaneously at the ligament side, spindle-shaped cells around the insertion site changed into hypertrophic chondrocyte-like ovoid cells. It seems to mature in phase from both the epiphysis nucleus side and the ligament side. At the border site only, two kinds of ovoid cells distinguishable by staining for type X collagen and S-100 protein coexisted. These findings suggest that a biological approach not only from the bone side but also from the grafted tendon side is necessary to reproduce the normal 4-layer structure of this site. In the future, we need to investigate the factors leading to structural change while the biphasic integration process appears to be occurring.

P13-148 Clinical outcome of ACL revision surgery; a single centre experience van der Meijden O.1, Faber C.1, Peters J.2, Tamminga R.2, Zijl J.1, van der Hoeven H.1 1 Sint Antonius Hospital, Nieuwegein, Netherlands, 2Fysiotherapiepraktijk Tamminga/Medicort, Utrecht, Netherlands Objectives: Annually, approximately 7000 Anterior Cruciate Ligament (ACL) reconstructions are performed in the Netherlands, with numbers increasing. With an average failure rate of 10%, the amount of revision surgical procedures is also rising. The purpose of our study was to review experience with revision ACL reconstruction. Methods: A retrospective case study was performed of 41 patients who underwent revision ACL reconstruction between 2003 and 2008, analyzing patient history, mechanism of primary graft failure and surgical difficulties encountered. Additionally, patients were invited for complete subjective and clinical evaluation. The minimum follow up was 6 months. Results: Five different mechanisms of primary graft failure were identified; technical errors (48%), recurrent trauma (20%), recurrent trauma in combination with technical errors (17%), failure of graft incorporation (10%) and finally persisting posterolateral rotatory instability (5%). New tunnel positions were determined and new tunnels drilled in 35 patients (85%). There was no interference with previous tunnels in 29 patients (71%). In 4 (9%) cases spongiosa transplantation and in 2 (6%) cases Allomatrix were required to fill up the pre-existing tunnels during arthroscopy prior to revision. The follow up for complete clinical evaluation included 28 patients (68%) and was on average 28 months (range 3-72 months). The average Lysholm, subjective IKDC and Dutch KOOS score were 83.6 (SD 12.0), 80.4 (SD 16.7) and 85.2 (SD 11.8) respectively. The mean Tegner rating for sports activities was 5.6 (SD 2.3), for working activities the mean score was 2.1 (SD 1.1). Objective IKDC rating showed 20 patients with a good IKDC score (71%, grade A/B) and 6 patients with grade C (21%, fair IKDC score). In 15 patients a side to side difference in Kneelax3 measurement was found of\3mm (54%), in 9 patients 3-5mm (32%) and in 4 patients [5mm (14%). The mean difference found was 2.6mm (SD=2.0mm). The average scores of the functional jump testing of the involved knee, mentioned as deficit of the score of the uninvolved knee, were -9.0% for the vertical hop, -6.0% for the hop for distance and -14.0% for the side hop. Conclusions: In our study the most important factor of failure of primary reconstruction was technical errors. Though results tend to be slightly inferior to those of primary ACL reconstructions, the overall outcome of revision ACL surgery is good. Preoperative planning of the operation is essential and the position of the tibia and femoral tunnel should be addressed with special attention. ACL revision surgery is a technically demanding procedure, but can give good to satisfactory results in the majority of patients.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P13-170 The effect of bone tunnel coalition at tunnel outlet on the tunnel widening after anatomical double bundle ACL reconstruction Inoue M.1, Shimamoto N.1, Kasahara Y.2, Onodera S.3, Yasuda K.4 1 NTT East Japan Sapporo Hospital, Orthopaedic Surgery, Sapporo, Japan, 2 Hokkaido University Graduate School of Medicine, Orthopaedic Surgery, Sapporo, Japan, 3Hokkaido University Graduate School of Medicine, Sports Med., Sapporo, Japan, 4Hokkaido University, Dept. of Sports Medicine and Joint Surgery, Sapporo, Japan Objectives: In anatomic double bundle ACL reconstruction, bone tunnel outlet positions in the femur and tibia are very critical for clinical outcome. Since the area of ACL footprint is small, antero-medial (AM) and postero-lateral (PL) tunnels easily coalesce at their intraarticular outlet and communicate each other. We hypothesized that tunnel coalition at outlet could cause tunnel widening after anatomic double bundle ACL reconstruction, because micro movement of two grafts could occur at their tunnel outlets when they lacked septum. There have been no reports on how bone-tunnel widening is affected by the existence of septum between AM and PL bone tunnels in anatomic double bundle ACL reconstruction. The purpose of this study is to clarify whether or not tunnel coalition at tunnel outlet of AM and PL bone tunnels affects bone tunnel widening after double bundle ACL reconstruction. Methods: Ninety knees which underwent double bundle ACL reconstruction were evaluated. The bone tunnel outlet diameters were measured on the Workstation from bone tunnel images on accurate oblique-coronal and oblique-sagittal views taken by Multi Planar Reconstructed (MPR)CT, which can make detailed assessment of bone tunnel widening from any direction. Differences in post-op bone tunnel diameters between those measured at 1 week and at an average of 1.3 years were compared with enlargement of diameters that were more than 2 mm being defined as bone tunnel widening. Thirteen femoral tunnels and 77 tibial tunnels showed tunnel communication at tunnel entrance at 1 week after surgery. We investigated the correlation between the incidence of bone tunnel widening (%) and anterior knee laxity (mm), which was compared between tunnel coalition group and no coalition group. Results: In the femur, the incidence of tunnel widening in the coalition group (82%) was significantly greater than that in the non-coalition group (24%) (P\0.05). In the tibia, there was no significant difference in the incidence of tunnel widening between the coalition group (25%) and the non-coalition group (20%). There was no significant difference in the extent of anterior knee laxity between the two groups in the femur and also in the tibia. Conclusions: Bone tunnel widening will be a serious problem after ACL reconstruction, especially after revision ACL reconstruction surgery. Femoral tunnel coalition at tunnel outlet significantly induces tunnel widening after double bundle ACL reconstruction using hamstring graft. The existence of septum between AM and PL femoral tunnels is an important factor in the prevention of femoral bone tunnel widening and knee stability. On the other hand, tibial tunnel communication makes no adverse effect.

P13-185 Tibial rotation in ACL single and double reconstruction: a 3D in vivo analysis Neven E.1, D’Hooghe P.2, Molenaers B.1, Callewaert B.3, Desloovere K.3, Bellemans J.1 1 University Hospital of Leuven, Dept. of Orthopaedic, LubbeekPellenberg, Belgium, 2Stedelijk Ziekenhuis Roeselare, Orthopaedic Surgery, Roeselare, Belgium, 3University Hospital of Leuven, Dept. of Rehabilitation Sciences, Lubbeek-Pellenberg, Belgium Objectives: Earlier in vivo studies concerning ACL single bundle reconstructions have shown that during low demand activities such as walking tibial rotation could be restored to normative values but in high demand activities as jumping and pivoting functional dynamic knee function is not restored. The purpose of our study was to determine whether tibial rotation could be restored in ACL double bundle reconstructions.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: Two groups were included in this study. Eight subjects with an ACL single bundle (3 females, 5 males - mean age 35 years) and eight subjects with an ACL double bundle (3 females, 5 males - mean age 32,4 years) reconstructed knee. A clinical evaluation (anterior tibial translation, pivot shift test) and arthrometric (KT-1000) evaluation was performed pre-operative and at 6 months. During this evaluation Lysholm and Tegner scores were obtained as well. After this clinical evaluation a 3 D in vivo evaluation was performed (preoperative and at 6 months). 3 D data were collected during low-demand activity such as walking, and high-demand activities such as pivot (external rotation) and sidestep turn (internal rotation) both after descent and landing from a jump from a 25-cm high platform. For each motor task, the kinematic and kinetic data of three valid trials of the involved and contralateral side were used for further analysis in this study. The dependent variable examined in the present study was the maximum range of motion of tibial rotation. Results: All the subjects regained objective stability with negative Lachman and pivot shift test results. KT-1000, Lyshom and Tegner at 6 months revealed no statistical difference between the ACL SB and DB group. During 3D analysis we found: 1. For walking : significant lower ROM (p=0.05) tibial rotation in ACL deficient versus single bundle reconstruction 2. For descent : no significant values noted 3. For descent + Pivot : no significant values noted 4. For descent + sidestep turn : significant lower ROM (p=0.04) tibial rotation in ACL deficient versus intact contralateral knee (SB). Significant lower ROM tibial rotation in ACL reconstructed (single (p=0.01) and double(p=0.01) bundle) versus intact contralateral knee. 5. For jump + pivot : no significant values noted. 6. For jump + sidestep turn : Significant lower ROM tibial rotation in ACL double bundle reconstructed versus intact contralateral knee.

Tibial rotation Conclusions: These results show that ACL-deficient and ACL-reconstructed knee present with adaptation/protection mechanism, probably caused by muscle co-contraction. The rotational stabilizing effect of hamstrings and quadriceps in the SB and DB reconstructed knee seems to increase in higher demand activities as descent+side step turn and jump + side step turn. In further studies patients should be followed at various time intervals to investigate the change in kinematics over time. Furthermore Cybex dynamometry and electromyography testing should be included in the investigation as these results could quantify the muscular adaptations at the various time intervals.

S161 P13-191 Hamstring tendon autograft is not better than bone-patellar tendonbone autograft in anterior cruciate ligament reconstruction. A 3-year prospective matched-group analysis Widuchowski W.1, Lukasik P.1, Faltus R.1, Kusak W.1, Widuchowski J.1 1 District Hospital of Orthopedics and Trauma Surgery, Depart. of Knee Surgery, Arthroscopy and Sports Trauma, Piekary Slaskie, Poland Objectives: Bone-patella tendon-bone (BPTB) and four-strand hamstring grafts (4SHS) are the most common used grafts for anterior cruciate ligament (ACL) reconstruction. However, current debate on treatment options complicate the choice between these grafts and there are still controversies especially with respect to knee stability and functional outcome. The purpose of this investigation was to analyze the results of ACL reconstructions using either a bone-patella tendon-bone or a four-strand hamstring autograft and to compare them in terms of clinical outcomes, patient satisfaction, activity level and functional status. Methods: Between 2006 and 2007, 82 patients with a torn ACL were enrolled in a prospective controlled study. The inclusion criteria for this study was isolated ACL insufficiency combined with subjective knee instability. Patients who had had a previous injury or surgery on either knee, patients with bilateral ACL insufficiency; a PCL insufficiency; an injury of PLC, LCL, or MCL tear of grade III were excluded. Patients with concurrent osteoarthritis, meniscal lesions, focal grade III or IV cartilage lesions, were not included either. Forty-one underwent reconstruction with a BPTP autograft, and forty-one were treated with a 4SHS autograft. Each patient was assessed preoperatively and was followed for an average of thirty-three months (range, thirty to fortytwo months). Subjective and objective data were collected using the International Knee Documentation Committee (IKDC) knee score, Lysholm score and the instrumented Rolimeter test of knee laxity to anterior translation. Results: ACL reconstruction improved knee stability and both IKDC and Lysholm knee scores significantly. Both treatments produced similar outcomes in terms of patient satisfaction, activity level, and knee function (ability to perform a one-legged hop, bear weight, squat, climb stairs, run in place, and duckwalk). The Lysholm score was 91.2 in the patellar tendon group and 89.89 in the hamstring tendon group (p = .052). The Rolimeter side-to-side difference was 2.1 mm for the patellar tendon group and 2.3 mm for the hamstring tendon group (p = .061). There were less positive pivot-shift test results in the patellar tendon group (p = .042). Overall IKDC scores were better in the hamstring tendon group, but the difference was not statistically significant (p = 0.62). The graft rupture rate was 2.3% in the hamstring group and 4.7% in the bone-patellar tendon-bone group. Hamstring graft reconstruction was associated with lower patellofemoral crepitus (p = 0.032), but with higher thigh atrophy (p = .041). Conclusions: ACL reconstruction is associated with a significantly better IKDC score, Lysholm score and laxity measurement at 3-year follow-up. We were unable however, to demonstrate a significantly better long term outcome in both objective and subjective knee scores or laxity to anterior translation with either a patella-tendon autograft or a semitendinosustendon autograft. These findings may be at least partially contrary to few previous studies and seem to be attributable to the use of both grafts for anterior cruciate ligament reconstruction, also in high-level athletes.

P13-204 Is there a correlation between commonly used knee outcome scores in ACLD and ACLR subjects? Hohmann E.1, Bryant A.2, Tetsworth K.3 1 Clinical Medical School, Department of Orthopaedic Surgery, Rockhampton, Australia, 2Centre for Health, Exercise and Sports Medicine, School of Physiotherapy, Melbourne, Australia, 3Royal Brisbane Hospital, Department of Orthopaedic Surgery, Rockhampton, Australia Objectives: To assess knee function in the ACL-deficient and ACLreconstructed knee a number of validated knee outcome rating scales are used. These scores use a numeric system to rate findings such as pain, swelling, subjective assessment of function and level of activity.

123

S162 However, it is unknown whether there is a correlation between the outcome rating scales and whether they can be used interchangeably. The aim of this study was to investigate the correlation between the four commonly used outcome rating scales (Lysholm, IKDC, Cincinnati and Tegner). Methods: Inclusion criteria included physically active patients between the age of 18 and 35 years with isolated ACL injuries. A power calculation for sample size was performed. Selecting an alpha level of 0.05 and power value of 0.8, 24 ACL-deficient and 24 ACL-reconstructed subjects were needed to achieve adequate statistical power. Statistical analysis included the calculation of means and standard deviations for the dependant variables. Pearson’s product moment correlation coefficients were used to establish the strength of the relationships. Results: 44 ACL-deficient and 24 ACL reconstructed subjects (mean age 27.0, range 16-49) with a minimum of 12 months post surgery completed the tests. Pre-operatively, strong significant correlations (r=0.53-0.74, p=0.00010.001) between IKDC and the other scoring systems (Cincinnati, Lysholm and Tegner) were observed. The Lysholm score was significantly correlated to IKDC (r=0.74, p=0.0001) and Cincinnati (r=0.60, p=0.001) scores. Nonsignificant moderate correlations were observed between Lysholm and Tegner (r=0.38, p=0.17) and Cincinnati and Tegner (r=0.36, p=0.18) scores. Post-operatively all scores were strongly related (r=0.61-0.93). However, only the relationships between Lysholm and IKDC score (p=0.001) and IKDC and Cincinnati score (p=0.01) reached statistical significance. Conclusions: The results of this study indicate that the commonly used rating scales produce interchangeable results in the ACL-deficient patient. In the ACL-reconstructed patient knee scoring systems seem to measure different aspects of physical activity, physical disability and subjective patient satisfaction - all of which are not interchangeable. As such, the classification of results may vary and may explain the findings.

P13-205 Femoral and tibial tunnel placement in ACL reconstruction: the learning curve Hohmann E.1, Bryant A.2, Tetsworth K.3 1 Clinical Medical School, Department of Orthopaedic Surgery, Rockhampton, Australia, 2Centre for Health, Exercise and Sports Medicine, School of Physiotherapy, Melbourne, Australia, 3Royal Brisbane Hospital, Department of Orthopaedic Surgery, Brisbane, Australia Objectives: Extensive previous research has shown that tunnel placement is critical in ACL reconstruction. The ultimate position of both the femoral and tibial tunnel determines knee kinematics and overall function of the knee post surgery. It is believed that anatomic placement as closely as possible will restore kinematics and reduce complications such as graft elongation and impingement. As with all techniques there is a definite learning curve for the arthroscopic technique. However the effect of the learning curve on graft placement has been studied sparsely. The purpose of this project therefore is to investigate the effect of the learning curve on tunnel placement. Methods: Postoperative radiographs of the first 200 anterior cruciate reconstructions with bone-tendon-bone patella tendon of a single orthopaedic surgeon performed during the first 4 years of independent practice were analysed for tunnel placement. Radiographs were digitalized in blocks of 25 and imported into a CAD program. Tunnel placement both femoral and tibial antero-posterior and sagittal was assessed using Sommer’s criteria. A rating scale was developed to assess overall placement. A total of 100 points indicated perfect placement. A maximum of 30 points each were allocated for sagittal femoral and tibial placement and a maximum of 20 points each was allocated for coronal placement. Results: Tunnel placement scores improved from 66 for the first 25 procedures to 87 for the last 25 procedures. Sagittal femoral placement (zone 1-4 with zone 1 preferred zone of placement) improved from an average of 1.44 to 1.08. Sagittal tibial placement (45% from anterior border of tibia) did not change significantly and remained between 42.82 to 44.76%. Coronal femoral placement (between 10:00-11:00 o’clock for the right knee and 1:00-2:00 for the left knee) ranged from 10.45-11.15 and 12:451:15 o’clock respectively. This finding may be related to the transtibial

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 tibial technique used to place the femoral tunnel. Coronal tibial placement (45% from medial tibial border) ranged from 45-46.58%. Conclusions: This study demonstrated a definitive learning curve and steady improvement of tunnel placement. Whilst there was no significant improvement in sagittal placement overall placement improved significantly. Approximately 100 procedures are required to significantly improve tunnel placement in ACL reconstruction.

P13-207 Biomechanical function of anterior cruciate ligament remnants: effects of remnant pattern and length of time from injury to surgery on knee stability evaluated with a navigation system Nakamae A.1, Ochi M.1, Deie M.1, Adachi N.1, Kanaya A.1, Nishimori M.1, Nakasa T.1 1 Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan Objectives: This study aimed to evaluate the biomechanical function of anterior cruciate ligament (ACL) remnants in antero-posterior (AP) and rotational knee stability in patients with a complete ACL injury. We also documented variations in the ACL remnant pattern of patients with either a complete or partial ACL injury. Methods: A prospective study was performed in 100 patients undergoing ACL reconstruction. ACL remnants were classified into 5 morphological patterns (Group 1: bridging between the posterior cruciate ligament (PCL) and tibia; Group 2: bridging between the intercondylar notch and tibia; Group 3: partial rupture of the posterolateral bundle; Group 4: partial rupture of the anteromedial bundle; Group 5: no substantial ACL remnants). Patients in Groups 1 and 2 underwent intraoperative arthrometry with a navigation system before and immediately after resection of the ACL remnant. Effects of chronicity (length of time from injury to surgery) and ACL remnant pattern on changes in knee laxity after debridement of the ACL remnant were investigated. Results: The percentage of patients in each ACL remnant pattern group was 18%, 12%, 14%, 6%, and 50% for Groups 1, 2, 3, 4, and 5, respectively. The 30 patients of ACL scar pattern Groups 1 and 2 underwent evaluation of the biomechanical function of their ACL remnant. At 30 of knee flexion, the AP knee laxity of patients who underwent ACL reconstruction within 1 year of injury was found to have increased by 2.22 mm after resection of the ACL remnant. In contrast, the AP knee laxity of patients who underwent surgery at more than 1 year after injury was found to have increased by only 0.17 mm. There was statistical significant difference (P \ .01). However, chronicity did not influence either AP knee laxity evaluated at 60 of knee flexion or rotational knee instability. We also investigated whether any changes in knee stability were related to ACL scar pattern, although no effects were found. Conclusions: In the present study, we found that ACL remnants continued to contribute to antero-posterior knee stability evaluated at 30 of knee flexion for up to 1 year after injury, beyond which this biomechanical function was lost. We also found that ACL remnants provided no rotational knee stability at any stage after injury. We recommend that surgeons aim to preserve ACL remnants within 1 year of injury when performing arthroscopy without reconstruction in ACL-injured knees, regardless of ACL scar pattern. The only exception to this should be patients in whom there is notch impingement during knee extension.

P13-235 Cyclops syndrome after double bundle anterior cruciate ligament reconstruction: a 14 case series Sonnery-Cottet B.1, Lavoie F.2, Ogassawara R.1, Scussiato R.1, Kasmaoui H.1, Chambat P.1 1 Centre Orthope´dique Santy, Lyon, France, 2Hoˆpital Notre-Dame, Centre Hospitalier Universitaire de Mont, Montre´al, Canada Objectives: The occurrence and the characteristic of cyclops syndrome has been reported for standard single bundle ACL reconstruction, but little is known about cyclop lesion associated with ACL double bundle

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 reconstruction.The aim of this study is to report the clinical and operative characteristics of cyclop lesion after ACL double bundle reconstruction. Methods: 387 patients who underwent an ACL double bundle reconstructions with quadriceps or hamstring tendons were followed at 6 weeks, 3, 6 and 12 months for clinical examination. When an extension deficit was observed, a MRI was performed to eventually diagnose a cyclops syndrome. An arthroscopic removal of the nodule was performed in all patients with cyclops syndrome. These patients were reviewed at a mean follow up of 12 months (min 6, max 20) after nodule debridement for clinical evaluation with the IKDC form. Results: 14 patients had a cyclops syndrome (3.61%). The occurrence rate in the quadriceps tendon group was significantly higher than in the hamstring group (p = 0.0017. Pearson0 s Chi-squared test). In the postoperative period, these patients had an average loss of extension of 6 (range 3 to 15), 71.4% had swelling. At the six weeks visit, 78.6% of the14 patients had an important quadriceps dysfunction associated with an active extension deficit. At arthroscopic debridement, the cyclops lesion was always located from the roof of the intercondylar notch. At the last followup, 12 patients had full range of motion, an extension loss was still present in two patients. 78.5% patients were graded A, 14.3% B and 7.2% C at IKDC objective evaluation. Conclusions: The cyclops syndrome is a complication after a double bundle ACL reconstruction more frequently observed with the quadriceps tendon graft than with the hamstring graft. The overall occurrence rate of cyclops syndrome in our series is comparable to the ones reported for standard single bundle reconstruction. A prevalent localisation of the nodule on the roof of the notch has been noted. 78,6% of our patients with a cyclops syndrome complained of loss of active extension associated to an important quadriceps dysfunction 6 weeks after surgery. This data modified our follow-up with a first control visit at 3 weeks postoperative for evaluation of extension and quadriceps function.Further studies are required to determine if the active extension loss associated with a quadriceps dysfunction observed in the majority of cases is correlated with cyclops syndrome occurrence.

P13-288 A new technique (GX2) for passing BTB graft, in ACL reconstruction with the femoral tunnel through anteromedial portal Karachalios G.-G.1, Chatziargyropoulos T.1, Andrianopoulos N.1, Kaseta M.1, Tsagkarakis G.1, Vlachos I.1 1 Henry Dunant Hospital, A Orthopaedic Dept, Athens, Greece Objectives: To present the new technique which we have developed and allows an easier pulling and passing of the BTB graft, when the two tunnels are not in the same line, i.e. with the femoral tunnel via AM portal. Methods: After the drilling of the two tunnels with the knee in 110 degrees of flexion,we pass a wire with eyelet through the tibia,directed to the roof of the notch,emerging through the anterior cortex of the femur. Using an other wire via the femoral tunnel we finally pass two suture loops of different color.The two loops are out of the tibial cortex, one coming from the roof of the notch and the other from the femoral tunnel.Initially,with the suture loop of the notch we pull the sutures and the graft itself into the joint.When all the bony part of the graft enters the joint,we change the graft0 s sutures to the femoral tunnel with the other suture loop,and finally we pull the graft in the femoral tunnel easily even in cases of press fit positioning. Results: Since 2007 we performed this technique in 47 cases.The femoral tunnel was in the 10 to 11 o0 clock position for right knees (2 to 3 o0 clock position for left knees).In the two of the first five cases,we took off the graft and redid the manoeuvre with the bony graft trimmed in 25 mm of length.Since then,we always harvest the graft in such a way. In all but two cases we placed the graft without oversizing the femoral tunnel.In four cases there was a mismatch of 5 mm concerning the lengths of the tunnel and the bony graft in tibia.We experienced an elongation of the operating time of about 10 min. Conclusions: We believe that this technique is reliable,reproducible, with a short learning curvature and offers solution to a technical problem which we often confuse and leads us to improper graft placement.

S163 P13-315 Hamstring strength recovery following hamstring tendon harvest for anterior cruciate ligament reconstruction: a comparison between graft types Feller J.1, Ardern C.1, Webster K.1 1 La Trobe University Musculoskeletal Research Center, Melbourne, Australia Objectives: To investigate hamstring strength following harvest of one or two hamstring tendons for anterior cruciate ligament reconstruction. Methods: Fifty participants who had returned to regular sporting activity were recruited to participate in a comparative study at an average of 32.5 months after ACL reconstructive surgery (ST-G = 30; ST n = 20). Isokinetic (at 60/s and 180/s with the peak torque and torque produced at 60, 90 and 105 knee flexion recorded) and isometric hamstring strength (measured at 30, 90 and 105 knee flexion) were measured, and the standing knee flexion angle used to evaluate functional hamstring strength recovery. Results: No significant differences between the groups were found in any of the isometric or isokinetic strength measures, and in the standing knee flexion angle. No relationship was found between the standing knee flexion angle and the isometric hamstring strength results obtained at 105 knee flexion (r2 = 0.034). Conclusions: These findings demonstrate that whether one uses semitendinosus alone or semitendinosus and gracilis, the choice of hamstring tendon graft is unlikely to significantly influence post-operative hamstring strength outcomes in athletes returning to sport post-operatively. Both graft choices demonstrated strength deficits of between 3% and 27% compared to the non-operated limb, indicating hamstring strength deficits persist despite successful completion of rehabilitation. The results also demonstrate that the standing knee flexion angle should not be used as a surrogate clinical measure of hamstring strength.

P13-322 Hamstring harvest for ACL reconstruction: is there a disadvantage? Gupta G.1, Engstro¨m B.2 1 Bayview Orthopaedics, Geelong, Australia, 2Capio Artro Clinic & Stockholm Sports Trauma Research Centre, Stockholm, Sweden Objectives: The aim of this study was to evaluate knee flexor strength after hamstring harvest in ACL (Anterior Cruciate Ligament) reconstructed patients. This was done to understand the implications of harvesting the hamstrings on knee flexor function. Methods: An evaluation of prospectively collected data was done for 1304 patients that underwent arthroscopic ACL reconstruction from January 2003 to December 2006. They comprised of 3 groups: PTB (Patellar Tendon graft, 252 cases), ST (Semitendinosus graft, 234 cases), STG (Semitendinosus & Gracilis graft, 818 cases). After a mean 6.4 months of follow up, all 3 groups were evaluated for Peak torque, Angle of peak torque and Total work during concentric flexor contraction at 90/sec. The data collected was analysed with statistical tools. Results: The Peak torque was 88.1% of the healthy knee in ST, 87.3% in STG and 97.2% in PTB group. The difference between ST and STG groups was insignificant. The angle of peak torque had shifted from 33.1 to 26.6 (p\0.0001) in ST, from 31.0 to 25.0 (p\0.0001) in STG and 29.1 to 26.8 (p=0.0056) in PTB group. The total work was 85.9% of the healthy knee in ST, 84.6% in STG and 97.1% in PTB group. The difference between ST and STG group was insignificant. Conclusions: This study shows that the hamstrings recover to near 90% peak torque 6.4 months after their harvest. This recovery is at a shallow angle of flexion, providing a better agonist to protect the reconstructed ACL. This could be an important advantage of the use of hamstrings as graft material. An insignificant difference between the ST and STG groups in peak flexor torque and total work further establishes the low morbidity associated with hamstrings as graft material.

123

S164 P13-395 ACL reconstruction with bone plugs and hamstring - 4 years results Felmet G.1 1 ARTICO Sportklinik & Orthopa¨dische Praxis, Villingen-Schwenningen, Germany Objectives: Hamstring is used for ACL reconstruction and usually fixed with foreign material. We developed a biological method with bone dowels harvested with diamond instruments and fixed press fit near the native insertion. We proved stability of press fit fixation of hamstring under the aspect of early functional rehabilitation and biological integration after 4 years. Methods: In a prospective study we analysed 147 patients with 4,3 years follow-up between 2003 and 2004. All had foreign material free press fit fixation with bone dowels. The bone dowels have been wet grinded with diamond instruments supported with special guiding devices. The graft has been implanted from bottom to top and fixated under tension press fit in 120 degree knee flexion. Tensioning of the graft is achieved by the geometry of the knee while moving in extension. The patients were scored by the IKDC- and Tegner Activity Score. Clinical investigation comprised the Pivot Shift, Lachman Test with the Rolimeter and X-rays were done weight bearing. Results: The 4,3 years postOP follow up showed in 89% the IKDC score A/B, a Lachman test 0-3 mm in 94%, no Pivot-Shift in 90%. The radiological investigation could be performed in 78 cases. Tunnel widening has been observed in the tibia tunnel in 18 cases and in the femoral tunnel in 12 cases without negative influence on stability. Tunnel widening correlated with thick grafts. Conclusions: The fixation with bone dowels harvested with diamond instruments allow ACL reconstruction with a reproducible biological press fit fixation near the original insertion. This is preventing bone defects with ease of revision under low costs.

P13-412 Observations on measuring rotational laxity of the knee in-vitro: comparison of measurements at the level of the tibia, skin and foot Alam M.1, Bull A.2, Thomas R.3, Amis A.A.4 1 Imperial College London, Biomechanics and Bioengineering, London, United Kingdom, 2Imperial College London, South Kensington Campus, London, United Kingdom, 3Charing Cross Hospital, London, United Kingdom, 4Imperial College of Science Technology and Medicine, Mechanical Engineering Department, London, United Kingdom Objectives: To analyse the accuracy of a simple device to measure rotational laxity of the knee in vitro, and to use the conclusions to help the design of a clinical device. Methods: The device consisted of a cutaneous splint over the tibia and a footplate. 6 lower limbs were used in the study with the femur held rigidly and the knee tested at 90 and 30 flexion. Two consecutive external rotation torques of 8Nm were applied to the knee through the footplate and measurements were taken at the level of the footplate, the cutaneous tibial splint and at the level of the tibial skeleton by two different testers. In order to look at the difference in tibial external rotational laxity between flexed and extended postures, measurements at 90 and 30 knee flexion were compared. The effect of soft tissues on measurement of tibial rotation by instruments resting on the skin was investigated by comparing measurements from the tibial splint to tibial skeletal inclinometers. The use of foot rotation in the dial test as an accurate representation of tibial representation was analysed by comparing knee rotation measured at the level of the tibia to inclinometer measurements at the footplate. The measuring device reproducibility was analysed by looking at intra- and inter-tester correlation. Results: Average rotation measured at the skeletal level was 24 and 26 at 90 and 30 knee flexion respectively. This compared to average external rotations measured at the tibial splint of 30 and 35, and average external rotations measured by the footplate inclinometer of 65 and 82 at 90 and 30 flexion. Comparison of measurements at the tibial

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 splint and at the level of the skeleton showed significant correlation, 0.92 at 90 and 0.9 at 30 knee flexion. Bland-Altman analysis revealed tibial splint measurements overestimated rotations by 6 and 9 at 90 and 30 knee flexion respectively. Footplate measurements did not correlate significantly with tibial rotation and consistently overestimated rotations by 103% at 30 knee flexion. Intra-tester intra-class correlation co-efficient (ICC) were C0.89 (p C 0.02) and above for both testers at 90 and 30 knee flexion for measurements taken at the skeletal level. Intra-tester ICC for measurements using the tibial splint was highly significant at C 0.88 for 90 knee flexion and C 0.82 for 30 knee flexion. Intra-tester ICC was not significant for either tester measurements at the footplate. Inter-tester ICC was highly significant for tibial skeletal measurements at both knee flexions at C 0.87. For the tibial splint measurements, intertester ICC was 0.88 and 0.74 (p = 0.01) for 90 and 30 knee flexion respectively. Inter-tester ICC was not significant for the footplate measurements. Conclusions: Measurements using the tibial splint significantly correlated with tibial skeletal rotation but consistently over-read the degree of external rotation. This has important implications for the clinical measuring device, particularly in terms of controlling for femoral rotation. There were systematic problems with measurements at the footplate, probably due to errors in foot and ankle constraint. This preliminary study has shown that measurement of knee rotation needs to occur at the level of the tibio-femoral joint, not at the level of the foot, in order to improve accuracy. Further development has been suggested and evaluation has begun in the clinical setting.

P13-413 Evaluation of the femoral tunnel position on ACL double bundle reconstruction using 3-dimensional CT Hashimoto Y.1, Tomihara T.2, Yoshida G.2, Shimada N.2 1 Osaka City University Graduate School of Medicine, Orthopaedic Surgery, Osaka, Japan, 2Shimada Hospital, Orthopaedic Surgery, Habikino, Japan Objectives: The aim was to measure the femoral tunnel position using 3-dimensional CT and to evaluate the correlation between the femoral tunnel position on ACL double bundle reconstruction and KT measurement. Methods: 59 patients underwent ACL double bundle reconstruction using hamstring tendon autograft. CT scan was performed 6 month after the procedure. CT images of 1mm slice thickness were made by threedimensional (3D) reconstruction software, Virtual Place Advance (Aze Ltd., Japan). Based on the resident ridge in the postoperative 3-dimensional CT images, the location of anterior-medial (AM) and posteriorlateral (PL) bone tunnel was evaluated. The distance to anterior-medial bone tunnel from posterolateral rim (AD), posterior-lateral bone tunnel from posterolateral rim (PD), between anterior-medial tunnel and a parallel line of the long axis of the femoral shaft drawn from cartilage edge of lateral condyle (AD2) was measured on lateral view of 3-dimensional CT image of the knee. The angle (a) between the long axis of the femoral shaft and the line drawn the center of AM and PL was also measured. A-P laxity using KT-1000(manual max) was evaluated 6 months after the procedure. Correlation between the KT and CT measurement (AD, PD, AD2 and a) was analyzed with Stat view-J 5.0. Results: AD,PD and AD2 were 2.0 ± 1.1mm(0-4.5mm),6.4 ± 2.1mm (1.5-13.9mm) and 1.6 ± 1.3mm (-1-4.8mm), respectively. The angle of the femoral tunnel (a) was 134±11 (106.7-167.1) in correlation with the long axis of the femoral shaft. 7 of AM tunnel were located on the ridge and 52 were posterior to the ridge. All of PL tunnel were posterior to the ridge. There is weak negative correlation (r=0.25) between KT measurement and PL tunnel position. Conclusions: Accurate and anatomic tunnel placements are essential to success of the ACL reconstruction. It is reported that the reconstructed

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 graft will be prone to failure if the femoral tunnel is placed anterior to the resident ridge in single bundle reconstruction. In this study, there was no correlation between AD and KT measurement. This fact is considered that our AM position was almost posterior to the ridge. However, there is negative correlation between KT measurement and PL tunnel position. It can be presumed that A-P stability can be acquired when the position of PL was shallower, because of its length pattern. In addition, it is possible that partial rupture of PL bundle occurred.

P13-420 Outcomes after transphyseal anterior cruciate ligament reconstruction in patients with open growth plates Heikes C.1, Larson C.1, Ellingson C.1, Giveans M.R.2 1 Minnesota Orthopedic Sports Medicine Institute, Eden Prairie, United States, 2Minnesota Sports Medicine, Minneapolis, United States Objectives: Concerns about growth disturbance after transphyseal Anterior Cruciate Ligament (ACL) reconstruction in the presence of open growth plates exist. The hypothesis was that transphyseal reconstruction with soft tissue graft for ACL tears in individuals with open growth plates would result in minimal risk for growth disturbance. Methods: Between May 2003 and October 2007, 30 skeletally immature individuals with a mean age of 13.9 years (range 9 - 15 years) underwent ACL reconstruction with soft tissue grafts (22 hamstring autograft, 8 tibialis anterior allograft) using extraphyseal fixation. Twenty-two knees had wide open growth plates and eight knees had partially open growth plates. Outcomes were prospectively evaluated with KT-1000 measurements, IKDC, Cincinnati, and Lysholm scoring. Radiographs were evaluated for asymmetric physeal closure, growth arrest lines, and knee alignment. Patients were also clinically evaluated for limb alignment and leg length inequality at minimum 2 year follow-up. Results: The mean knee scoring was 93.8 points (IKDC), 97.0 points (Cincinnati), and 92.4 points (Lyshom) at 3.2 years mean follow-up (range, 2 - 6 years). The mean KT-1000 side to side difference at most recent follow-up was 0.77 mm. Minimum two-year radiographic and clinic evaluation revealed three Harris growth arrest lines, no asymmetric physeal closure, and a mean side to side difference of 1.4 degrees for radiographic tibiofemoral angle, and 0.3 cm leg length inequality. Three patients sustained a graft re-rupture (10.0%). All three patients suffered a non-contact valgus injury while participating in sports. Conclusions: Transphyseal ACL reconstruction with soft tissue grafts and extraphyseal fixation in patients with open growth plates resulted in good to excellent outcomes in 90% of patients without evidence for clinically significant growth abnormalities.

P13-439 Health-related quality of life after anterior cruciate ligament reconstruction Ma˚nsson O.1, Sernert N.2, Kartus J.1, Ejerhed L.1, Ahlde´n M.3 1 NU-Hospital Organization, Trollha¨ttan/Uddevalla, Orthopedic Department, Uddevalla, Sweden, 2NU-Hospital Organization, Dept. of Development and Research, Trollha¨ttan, Sweden, 3 Sahlgrenska University Hospital/Mo¨lndal, Dept. of Orthopaedics, Mo¨lndal, Sweden Objectives: Patient0 s evaluation is important to the clinical assessments. Health-related quality of life, evaluated, using SF-36, 2-7 years after ACL reconstruction, compared to a matched normal population. Methods: Two to 7 years after ACL reconstruction, SF-36 questionnaires were mailed to 793 patients. Of those, 544 patients answered, 125 were incomplete leaving 419 patients who were analysed and compared to a Swedish healthy control group comprising 2410 persons. 161 patients were female, 258 male, the mean age was 28 years, 331 had undergone

S165 reconstruction using bone-patellar tendon-bone autograft (BPTB) and 78 hamstring autograft (HS). Results: The patient group scored significantly higher in General Health (GH), Social Function (SF), Role Emotional (RE) and Mental Health (MH). The control group scored significantly higher in Physical Function (PF) and Role Physical (RP). There were no significant differences in the BPTB-group compared with controls, however, the HSgroup had significantly lower values in PF compared with controls. Both the total male-group and the BPTB-male group revealed significantly lower scores in PF, RP, but higher in SF, RE and MH compared the controls. Correspondingly, the total female group scored significantly lower in PF. Conclusions: After ACL reconstruction the patients reported a general good health-related quality of life in comparison with a matched sample of the general population. These normative values are important in the evaluation of health status before and after intervention.

P13-454 Prospective randomized comparison of trans-tibial versus transportal techniques for femoral tunnel placement in anatomic doublebundle ACL reconstruction Takeda Y.1, Iwame T.1, Kondo K.1, Takasago T.1, Goto T.1, Fujii K.1, Naruse A.1 1 Tokushima Red Cross Hospital, Orthopaedic Surgery, Komatsushima, Japan Objectives: Proper placement of the femoral tunnel is the integral part of the anatomic double-bundle (DB) anterior cruciate ligament (ACL) reconstruction. Creating the femoral tunnel through the tibial tunnel (transtibial; TT) and through the far anteromeidal (FAM) portal (trans-portal; TP) are the most common techniques in the DB ACL reconstruction. The purpose of this study was to compare the TT and TP technique with respect to the anatomical placement of the femoral tunnel as the randomized clinical trial. Methods: 31 patients with unilateral ACL deficiency were randomized into two groups with TT and TP techniques. Of these patients, 15 (11 male and 4 female patients; mean age, 28.2 years) underwent femoral tunnel creation with trans-tibial drilling and 16 (10 male and 6 female patients; mean age, 26.8 years) underwent femoral tunnel creation with trans-portal drilling. Demographic data were not significantly different between the 2 groups. In the both groups, anatomic DB ACL reconstruction with hamstring tendon grafts fixed with Endobutton CL on the femoral side and Double Spike Plate on the tibial side were performed. In the TP group, a K-wire was inserted through the FAM portal, and placed at 10:30 (right knee) or 1:30 o’clock (left knee) position with respect to the apex of the notch and approximately 6mm anterior from the posterior cortex for the anteromedial bundle (AMB) tunnel. Once marking the inserting point at 90, the knee was flexed at 120, and the tunnel was created. For the posterolateral bundle (PLB) tunnel creation, K-wire was placed at 6 mm anterior (high) from the posterior margin of osteochondral junction at 90, and the tunnel was created at 120. In the TT group, a Kwire was inserted through the AMB and PLB tibial tunnels, and inserted at the same point with that for the trans-portal technique. Both tunnels were created at 90. 3-D CT image of the lateral wall of the intercondylar notch was reconstructed using the volume rendering technique at 7th day from the surgery. Femoral tunnel placement was evaluated using the quadrant method described by Bernard et al. In their method, position of the tunnel center was expressed in percentage to the total sagittal diameter of the lateral condyle along Blumensaat’s line, and maximum intercondylar notch height. Femoral footprint angle (FA) and the shortest distances of the tunnel to the adjacent cartilage margins were also measured. Statistical analyses were performed by use of the unpaired t test. A value of P\0.05 was considered statistically significant. Results: In the TT group, the center of the AMB was 11.7% of the height of the lateral femoral condyle and 25.0% along Blumensaat’s line, and that

123

S166

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

of the PLB was 39.5% and 33.8%, respectively. In the TP group, the center of the AMB was 15.6% (n.s.) of the height and 19.9% (p=0.005) along Blumensaat’s line, and that of the PLB was 45.5% (p=0.019) and 28.1% (p=0.005), respectively. FA in the TT group was 29.2 and that in the TP group was 31.3 (n.s.). The shortest distance of the AMB tunnel to the adjacent cartilage was 3.1 mm in the TT group and 2.2 mm in the TP group (n.s.), and that of the PLB was 3.5mm and 2.2mm respectively (p=0.019). Conclusions: AMB and PLB femoral tunnels in the TP group were created at deeper and lower position compared with those in the TT group. The results showed that the trans-portal technique is more preferable with regard to anatomical placement of the femoral tunnels.

P13-458 The sagittal alignment of the knee and its relationship to noncontact anterior cruciate ligament injuries Terauchi M.1, Hatayama K.1, Yanagisawa S.2, Saito K.1 1 Social Insurance Gunma General Hospital, Department of Orthopaedic Surgery, Maebashi, Japan, 2Gunuma University Faculty of Medicine, Department of Orthopaedic Surgery, Maebashi, Japan Objectives: Many risk factors for noncontact anterior cruciate ligament (ACL) injury have been proposed. Of these factors, knee hyperextension and tibial posterior slope are related to sagittal alignment of the knee. The purpose of this study was to determine if there is a difference in sagittal alignment of the knee between an ACL-deficient group and a negative control group, and to find risk factors for noncontact injures to the ACL. Methods: Our patient group consisted of 73 ACL-deficient subjects who sustained a noncontact injury to the ACL. There were 33 men and 40 women. A negative control group consisted of 28 men and 30 women. MRI of the knee in full extension were acquired. Three lines were used to measure the sagittal alignment of the knee. The first line was tangent to the anterior cortex of the femur (F line). The second line was tangent to the posterior cortex of the tibia (T line). The third line was tangent to the concave profile of the medial tibial plateau (P line). The three lines were superimposed on a computer and the following angles were measured: The femoral-plateau angle (FP) is the angle between the F line and the P line. The extension angle (EX) is the angle between the F line and the T line. The tibial posterior slope angle (PS) is defined as 90 minus the angle made by the intersection of the T line and the P line. Results: In the women, the mean FP was106.3± 2.6 in the ACL-deficient group and 102.6±3.9in the negative control group. The mean PS was 10.9±3.1 in the patient group and 8.2± 2.3in the negative control group. FP and PS were significantly larger in the ACL-deficient group than in the negative control group. The mean EX was 4.9± 3.0 in the ACLdeficient group and 4.5±3.6in the negative control group. These were not statistically significant. In the men, the differences between the 2 groups were not significant. In the women, EX negatively correlated with PS in the ACL-deficient group (R=-0.55, p\0.001), but not in the negative control group. Conclusions: This study has demonstrated that a difference in sagittal alignment of the knee between ACL-deficient and the negative control group existed only in women. The ACL-deficient group showed a larger FP than the negative control group in women. The larger FP represents the large degree of posterior inclination of the tibial plateau to the femoral shaft at full extension. The larger FP results in greater tibiofemoral shear stress and may be harmful to the ACL. This group also showed a negative correlation between EX and PS, suggesting that knees with hyperextension had a small tibial posterior slope, while knees without hyperextension had a large tibial posterior slope. Knees with large tibial posterior slope tend to decrease in terms of extension, while knees with a small posterior tibial slope tend to increase in terms of extension. There were therefore two types of large FP; one has its origin in an increasing tibial posterior slope and the other results from hyperextension of the knee (Figure 1). Both the posterior tibial slope and hyperextension are risk factors for noncontact ACL injury in women.

123

P13-463 Measurement of the anteroposterior and rotational knee laxity by a navigation system during ACL replacement Jenny J.-Y.1 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France Objectives: Navigation systems may help quantifying the anterior and rotational laxity of the knee during ACL replacement. Methods: 30 patients were operated on for ACL reconstruction. The anterior laxity was quantified before surgery with the KT 1000 device at 25 of knee flexion. A non image based navigation system was used to quantify during the procedure the pre- and post-reconstruction maximal anterior translation and internal/external rotation of the tibia at 25 and 90 of knee flexion. Instrumented measurement was repeated after reconstruction. Instrumented and navigated measurements by each patient were compared with a paired Wilcoxon test at a 5% level of significance, a correlation test and the Bland-Altman method. Results: 1) Anterior laxity: We observed a significant improvement of the anterior laxity at 25 and 90 of flexion after reconstruction. There was a significant difference but a strong correlation between instrumented and navigated measurements of the anterior laxity at 25 of flexion. The difference was in most of cases less than 2 mm. 2) Rotational laxity: We observed a significant improvement of the rotational laxity at 25 and 90 of flexion after reconstruction. Conclusions: The navigation system used for this study allowed us measuring in a precise and accurate way the intra-operative anterior and rotational laxity before and after ACL reconstruction. Initial intra-operative information about anterior and rotational laxity may be used to adapt the technique of replacement to the specific operated case. Post-replacement information about remaining anterior and rotational laxity may be used as a quality control.

P13-466 Hybrid fixation of hamstring graft in anterior cruciate ligament reconstruction Milankov M.1, Ninkovic S.1, Harhaji V.1, Janjic N.1, Savic D.1 1 Clinical Center Vojvodina, Medical School, Department of Orthopaedic Surgery and Traumatology, Novi Sad, Serbia Objectives: While the fixation technique for patellar tendon grafts is a pretty much straightforward procedure, no consensus has been found on the fixation of hamstring tendon grafts. Tibial fixation of the ACL graft is commonly considered more problematic than femoral fixation because forces acting on the ACL substitute are parallel to the tibial drill hole, the bone quality of the tibial metaphysis is inferior to that of the femur and the four-tailed end of the hamstring tendon graft that is fixed to the tibia is more difficult to secure. When interference screws are used for tibial graft

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 fixation, graft sometimes looses its tension, so a hybrid fixation (more than one method of fixation) must be applied. A novel hybrid fixation method which compiles two fixation techniques at the tibial site was developed. This method of tibial hamstring tendon graft fixation provides good initial fixation strength together with the biomechanical and biologic advantages of joint line interference fit fixation. Methods: All-inside arthroscopic procedure was used for ACL reconstruction. The hamstring tendon with its periosteal attachments was harvested through a short oblique skin incision. After a precise graft preparation, the graft was passed through the tibia into the femoral socket, properly positioned in the tunnel and fixed with RCI - round cannulated interference screws (Grujic´ & Upyjbh, Novi Sad, Serbia). While performing full range of knee motion, firm traction was applied to the graft in order to pretense the graft and observe if the full extension causes any impingement. The graft was tensioned using 60N force (Karl Storz, Tutlingen, Germany) and fixed with round cannulated interference screws (Grujic´ & Upyjbh, Novi Sad, Serbia). Additionally, a cating head-bicortical 4,5mm diameter screw with a modified head (Grujic´ & Upyjbh, Novi Sad, Serbia) was placed immediately below the tibial tunnel opening. The screw was tighten completely at first and then untighten for ‘ of a circle. Grafat’s sutures were placed through the slots on the head of the screw, three knots were tied and the screw was retighten completely for ‘ of a circle. Results: Our experimental model showed that tightening the screw for  of a circle generates 40N force while ‘ of a circle generates 80N force. This technique was used in about 200 cases, with no intra- or post-operative complications except in one case, where the sharp edge of the slot cut off the suture. From that time on, our slots have rounded edges. Conclusions: Biomechanical testing showed that a interference screw (RCI) and additional fixation to the distal hamstring tendon resulted in higher load at failure and stiffness compared to either interference screw (RCI), cortical screw, double spike plate, spike washer or WasherLoc fixation alone.Secure initial fixation of the graft is essential for a successful early rehabilitation before the graft fully incorporates into the host’s bone. We believe that our hybrid technique of tibial fixation of the hamstring graft allows simple, precise, strong early fixation and by doing that promotes biological incorporation.

P13-487 Bone tunnel enlargement after anatomic ACL reconstruction: mechanical or biological problem? Silva A.1 1 D. Pedro V Military Hospital, Hospital Militar Regional no. 1, Porto, Portugal Objectives: To evaluate prospectively with magnetic resonance (MR) imaging the changes in femoral tunnel diameter following arthroscopic anatomic ACL reconstruction with hamstring tendons. Methods: 40 patients underwent double-bundle arthroscopic ACL reconstruction using autologous hamstring tendons. Femoral fixation was performed by means of 2 EndoButtons CL devices and tibial fixation was done with 2 bioresorbable interference screws. The MR studies were performed between 84 and 117 days after surgery. Bone tunnel enlargement was assessed by measuring the transverse diameter of both femoral tunnels on PDW-FatSat coronal oblique MR sequences. The diameter of each tunnel was measured at the entrance of the femoral tunnel and at the mid-distance between the entrance and the bottom of the tunnel. The measurements were digitally obtained perpendicularly to the long axis of the tunnel. Results: The drill diameter used for the PL femoral tunnel ranged from 5.0 to 6.0mm (mean 5.4 mm; ± 0.3) and for the AM femoral tunnel ranged from 6.0 to 7.5 mm (mean 6.6 mm; ± 0.5). Regarding PL femoral tunnels, the entrance increased 16% in diameter in three months post-op and the middle of the tunnel increased 30% in diameter (0.9 mm vs 1.6 mm; p\0.001). In the AM femoral tunnel, the enlargement was also significantly higher in the mid section of the tunnel than at the entrance (0.9 mm vs 2.3 mm; p\0.001), with the tunnels enlarging 14% at the aperture and 35% in the mid section in three months post-op.

S167 Conclusions: our results don’t support the biomechanical theory of tunnel enlargement, because the tunnels were narrower at the entrance, where motion is supposed to be higher. We think that the most likely explanation for a higher enlargement at the middle of the tunnels is related with biological factors or, at least, they play a more important role in tunnel enlargement than mechanical factors.

P13-489 Correlation between synovial fluid and serum IL-1b levels after ACL surgery - preliminary results Darabos N.1, Hundric-Haspl Z.2, Haspl M.3, Moser C.4, Cuti T.1, Elabjer E.1 1 University Clinic for Traumatology, Sports Trauma Department, Zagreb, Croatia, 2Croatian Institute of Transfusion Medicine, Zagreb, Croatia, 3 Special Hospital for Orthopaedics and Traumatology, Dept. of Orthopaedic Surgery, Krapinske Toplice, Croatia, 4Center for Molecular Orthopaedics, Duesseldorf, Germany Objectives: IL-1b synovial fluid concentration is elevated after ACL surgery. Possibility of controlling harmful intraarticular influence of elevated IL-1b synovial fluid concentration after ACL surgery could be useful. Methods: We have investigated correlation between serum and synovial fluid IL-1b levels following ACL-reconstruction. We measured IL-1b concentration periodically in 3 synovial fluid and 4 serum samples in each of 20 patients receiving either Autologous Conditioned Serum (ACS) containing endogenous anti-inflammatory cytokines including IL-1Ra and several growth factors (Group A) or Placebo (Group B). Results: IL-1b synovial fluid concentration decrease appeared more pronounced in absolute terms in Group A. In 8 patients serum IL-1b were detected on 6th postoperative day. In 4 of them whose synovial fluid levels were over 10 pg/ml on 6th postoperative day, serum IL-1b were detected on 10th postoperative day. Results were different in Group B. Conclusions: Correlation between serum and synovial fluid IL-1b appearance persists in patients after ACL surgery and ACS application. Our preliminary results should be considered as an example of a possible ACS influence on the ACL posttraumatic healing process after an acute rupture and operative reconstruction and a possibility of controlling the ACS influence on the ACL healing process on the basis of the serum IL-1b level.

P13-508 Reconstructed CT evaluation of bone tunnel enlargement after anterior cruciate ligament reconstruction: comparison of hamstring and bone-patellar tendon-bone autograft Matsubara T.1, Takahashi M.2, Koyama H.2, Hanada M.3, Sano M.1 1 Shizuoka Municipal Shizuoka Hospital, Orthopaedic Surgery, Hamamatsu, Japan, 2Hamamatsu University School of Medicine, Orthopaedic Surgery, Hamamatsu, Japan, 3Hamamatsu Red Cross Hospital, Orthopaedics Surgery, Handayama, Japan Objectives: There have been studies of bone tunnel change after anterior cruciate ligament (ACL) reconstruction, however, there are few studies comparing computer tomography (CT) evaluation of bone tunnel change after ACL reconstruction between hamstring (HS) and bone-patellar tendon-bone (BPTB) autograft. ACL reconstruction with BPTB is considered to be superior to HS to bone union. The aim of this study was to evaluate the tibial and femoral bone tunnel change of reconstructed CT after ACL reconstruction with HS versus BPTB. Methods: We examined eighteen patients who underwent ACL reconstruction with follow-up of 1 year (8 receiving HS and 10 BPTB) between 2004 and 2008. The mean age of patients in the HS group was 32 years old and that in the BPTB group was also 32 years old at surgery. In the HS group, bi-socket reconstruction using the transtibial technique was performed. Both grafts were fixed to the femoral side with an endobutton and to the tibial side with a postscrew. CT was taken on the day after surgery and at a mean follow-up of 1 year in all case. We measured the bone tunnel width and areas of both femoral

123

S168 and tibial bone tunnels on the reconstructed oblique coronal and sagittal images of CT. Bone tunnel width was measured at 0.5 cm from the joint line on the femoral side, and 1 cm on the tibial side. All the measurements of widths and areas were performed on PC using ImageJ software. Anterior-posterior laxity of the knee was evaluated using a KT-2000 arthrometer postoperative 1 year. Results: In the HS group, the mean femoral bone tunnel width (oblique coronal) increased significantly from 8.5 mm to 10.1 mm at 1 year in the anteromedial bundle (AMB) and from 8.3 mm to 10.2 mm in the posterolateral bundle (PLB), the mean femoral bone tunnel width (oblique sagittal) increased significantly from 6.9 mm to 9.2 mm in AMB only. In the BPTB group, the mean femoral bone tunnel width (oblique coronal) increased significantly from 6.7 mm to 7.7 mm. There was no significant difference in the rate of change for bone tunnel width between the two groups. In the HS group, the mean tibial bone tunnel areas (oblique sagittal) decreased significantly from 39.3 mm2 postoperatively to 35.5 mm2 at 1 year. There was no significant difference in the rate of change for bone tunnel areas between the two groups. There was also no significant difference on the average side-to-side difference measured by KT-2000 arthrometer between the HS group (1.3±0.9 mm) and BPTB group (1.9±0.5 mm). No relationship was found between the rate of change for bone tunnel width or area and KT-2000 measurements in both groups. Discussion: We evaluated bone tunnel enlargement of reconstructed CT after ACL reconstruction comparing HS and BPTB. We used reconstructed CT because it depicts the real boundaries of the trans-osseous tunnel more precisely. A previous study found that bone tunnel enlargement by X-ray after ACL reconstruction was significantly larger in the HS group than in the BPTB group. In this study, there was no significant difference in the rate of change for bone tunnel width and areas between the two groups. On the femoral side, it might be possible to reduce bone tunnel enlargement in bisocket reconstruction in the HS group because each bone tunnel is smaller in width and the stress on the bone tunnel wall is dispersed. Conclusions: There was no significant difference in the rate of change for bone tunnel width and area using reconstructed CT between HS and BPTB after ACL reconstruction.

P13-509 Arthroscopic single-bundle anterior cruciate ligament reconstruction with periosteum-enveloping hamstring tendon graft, clinical outcome in 2 to 7 years Chen C.-H.1, Chang C.-H.1, Su C.-I.1, Wang K.-C.1, Wang I.-C.1, Liu H.-T.1, Wong C.-B.1, Yu C.-M.1 1 Chang Gung Memorial Hospital-Keelung, Orthopaedic, Keelung, Taiwan, Republic of China Objectives: This is a case series outcome study with surgical technique for single-bundle ACL reconstruction with periosteum-enveloping hamstring tendon graft at minimal 2 years follow-up. Methods: From 2000 to 2005, ACL reconstruction with a periosteumenveloping hamstring tendon graft was performed on 368 patients (372 knees). Of those, 312 patients who complete at least two years of followup were included for analysis. Four-strand periosteum-hamstring tendon grafts were used for single-bundle reconstruction. Clinical assessments included the Lysholm knee scores, International Knee Documentation Committee (IKDC) scores, KT-1000 instrumented testing, thigh muscle assessment, and radiographic evaluation. Radiographs were used to assess femoral and tibial tunnel widening. Results: The 312 study patients were followed for an average of 4.6 (2-7) years. Their median Lysholm knee scores were 56 (40 to 70) and 95 (60 to 100) points before and after surgery, respectively. After reconstruction, 85% of patients could return to moderate or strenuous activity, 5.1% exhibited grade 2 or higher ligament laxity with anterior drawer test, and 6.1% had positive pivot shift. Complete range of motion was achieved in 88% of patients. IKDC assessment rated 93% of patients as normal or nearly normal. Conclusions: Bone-tunnel enlargement of more than 1 mm was identified in 5.4% of femoral tunnels and 6.1% of tibial tunnels. The study shows that satisfactory results can be achieved with the periosteum-enveloping hamstring tendon graft in single-bundle ACL reconstruction with minimal tunnel widening.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P13-514 Study of anterior cruciate ligament injuries in professional sumo wrestlers Shimizu S.1, Tateishi T.1, Nagase T.1, Nakagawa T.1, Tsuchiya M.1 1 The Fraternity Memorial Hospital, Department of Orthopaedic Surgery, Tokyo, Japan Objectives: Sumo wrestling enjoys a long history as Japan’s national sport and is known and loved throughout the country. During 26 years, 4849 injuries in 1380 professional sumo wrestlers were treated in our orthopaedic department. Among them, injuries of lower extremity account for roughly 50%. Anterior cruciate ligament (ACL) injury is one of the common trauma in professional sumo wrestlers. Not only cases treated by ACL reconstruction, conservative treatment is often applicable for them. The purpose of this study is to investigate the course of ACL injury treated with or without ACL reconstruction in sumo wrestlers. Methods: Between December 1982 and December 2008, 207 wrestlers (223 knees) had been visited to our department and were diagnosed with ACL injury. 104 cases (109 knees) were treated without ACL reconstruction (conservative group), including 24 cases (26 knees) treated with arthroscopic meniscectomy. ACL reconstruction was performed for 109 cases (114 knees) (reconstruction group); 55 knees were reconstructed with BTB graft, 43 knees hamstring graft, 2 knees ITT graft, and 14 knees in other hospitals. Mean age was 22.2 (16-32) years in conservative group and 21.1 (16-32) years in reconstruction group. Body weight was 131 (78-233) kg in conservative group and 127 (89-194) kg in reconstruction group. We compared conservative group with reconstruction group in the laterality of injured knee and progression of official ranking list after trauma. Results: Conservative group consisted of 57 left knees and 52 right knees (p=0.63). Reconstruction group consisted of 66 left knees and 48 right knees (p=0.09). Although mean period of default was only 1.0 tournament (2 months) in conservative group, reconstruction group withdrew from 3.4 tournament (6.8 months) because of long rehabilitation exercise. Consequently, reconstruction group had sliding down the rank until 3 to 4 tournaments and came up with rankings of conservative group around 2 years after surgery. In conservative group, wrestlers who had gone up remarkably in the rankings were rare instead of down in the rankings were minimum after trauma. Conclusions: As a reason why left are more than right knees in reconstruction group, one of the possibility is that most sumo wrestlers hold on by left lower extremities when they step back in sumo exercise. Although a number of ACL injuries in sumo wrestlers are candidate for ACL reconstruction, a part of conservative group were able to continue sumo wrestling and show their ability. Therefore, we decide on courses of treatment with due consideration for physical characteristic and ranking in each ACL injured wrestlers.

P13-515 Anterior cruciate ligament reconstruction - what is success? A critical analysis of the knee injury and osteoarthritis outcome score (KOOS) O’Leary S.1, Hill G.2 1 Royal Berkshire Hospital NHS Foundation Trust, Trauma & Orthopaedics, Reading, United Kingdom, 2Specialist Registrar, North West Thames Rotation, London, United Kingdom Objectives: Patient Relevant Outcome Measures (PROMS) are increasingly used to assess the quality of care and clinical effectiveness of surgical intervention but lack clear definitions of success or failure. We sought to evaluate and compare the ability of two outcome scores, the Lysholm and the KOOS, to determine a successful outcome following ACL reconstruction (ACLR) by defining discrete criteria. Methods: 204 patients underwent single bundle ACLR (four strand hamstring graft) from 2003 - 2007 and had minimum 12 months follow up. 118 (58%) patients had complete KOOS and Lysholm scores at 12 months post operation and 65 patients (32%) recorded scores both pre-operatively and at 12 months post operation. A review of the literature identified several different methods of outcome score analysis and data representation that centred on either establishing a cut-off score indicating success, or by observing the change in the pre-operative and post-operative scores. These methods were each applied to our data to assess their ability to represent a good outcome following ACLR.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: The Lysholm scores at 12 months gave an average score of 90 with 82% of scores categorised ‘‘good’’ or ‘‘excellent’’ according to previously published ratings. The KOOS results were firstly subjected to cut-off values determined by comparison with median scores for a standard population (Sport/Recreation subscale [76, Knee related Quality of Life (QoL)[76); and also by the application of a calculated cut-off score (QoL[87.5 and three or more other sub-scales exceeding values between 85-86) for a Symptomatic Knee. The number of results meeting or exceeding the respective cut-off values at 12 months post-operation, were expressed as a percentage representing the success rate following surgical intervention. Secondly, the percentage of cases reporting an improvement in KOOS score by 20 points or more, and also 10 points or more, between pre-operative and 12 month post-operative scores were calculated to represent the rate of success following surgical intervention. Table 1 Analysis of author’s KOOS results for ACL reconstruction preop & 12 month post-op—calculation of ‘‘successful outcome’’ Method of Analysis

n = Criteria for Sports/recreation Knee QoL ‘‘successful’’ sub-scale sub-scale outcome n= % n= % success Success success Success

i) Cut-Off Score Compared with Standard Population

118 Sport/Rec [76, QoL [76

83

70.3%

56

47.5%

ii) Calculated Cut- 118 QoL [87.5 & Off Score 2/4 other scales [ 85-86*

72.9%

iii) Magnitude of Change in Score ?20

65

Change C20

43

66.2%

53

81.5%

iv) Magnitude of Change in Score ?10

65

Change C 10

55

84.6%

59

90.8%

Conclusions: The Lysholm Score and KOOS have to date been used primarily for comparison studies in ACL reconstruction. There has been no consensus regarding specific scores or methods that define ‘successful outcome’ following a surgical intervention. We have shown that observing the change in KOOS score following intervention produces a higher indication of successful surgical outcome than a fixed cut-off score. Use of ‘‘Change in Score’’ represents the ‘‘Value - Added’’ effect of surgical intervention and helps demonstrate clinical effectiveness, as well as permitting comparison of individual surgeon’s performance with the standards being established by the Scandinavian Ligament Registries.

P13-520 Evaluation of ACL grafts reconstructed with anatomical multi-bundle technique using hamstring tendon autograft by oblique coronal MR imaging Shiozaki Y.1, Horibe S.2, Tanaka Y.2, Yonetani Y.2 1 Seifuu Hospital, Orthopeadic Surgery, Sakai, Japan, 2Osaka Rosai Hospital, Orthopaedic Sports Medicine, Sakai, Japan Objectives: Anatomical multi-bundle ACL reconstruction by hamstring tendon autograft is increasingly performed to reproduce the normal anatomy of ACL. Although the idea is theoretically correct, it is unclear if the grafts mimic the course of the normal ACL. In addition, partially or completely damage of the posterolateral (PL) graft at the femoral tunnel aperture has been reported. Recently, MR scanning techniques have been developed to optimize visualization of the ACL bundle anatomy and the graft healing. In this study, using oblique coronal MR imaging, the course

S169 of the transplanted bundle grafts and the graft healing within the bone tunnel were evaluated. Methods: MR images of 18 patients (mean age of 21 years) who underwent anatomical multi-bundle ACL reconstruction by hamstring tendon autograft were reviewed. 10 patients had double-bundle and 8 had triplebundle ACL reconstructions. Each MR image was performed at six months after operation on both the reconstructed and contralateral normal knees. The oblique coronal T2-weighted images were oriented in parallel with the course of the femoral intercondylar roof. At six months post-op., all patients had negative Lachman test, and the mean side to side difference in KT-2000 measurement was 0.2 ± 1.2 mm. The course of transplanted bundle grafts was compared with that of normal ACL bundle and the signal around the femoral and tibial tunnel aperture was assessed. Results: Both anteromedial (AM) and PL bundles of the normal ACL fanned out toward the tibial insertion sites. The course of each bundle of transplanted grafts mimicked the normal ACL bundle. But the intermediate bundle was not restored in double-bundle reconstructed knee. High signal intensity around the femoral tunnel aperture was observed in one of AM bundles (6%) and in 8 of PL bundles (44%). However, high signal intensity around the tibial tunnel aperture was not found in any bundle grafts. Side to side difference in anterior laxity measured by KT-2000 was not significantly correlated with this high signal. Conclusions: Oblique coronal MR imaging revealed that anatomical multi-bundle ACL reconstruction could mimic the normal anatomy of the ACL. Fan shaped tibial insertion of the anterior bundle was well restored in triple-bundle ACL reconstructed knee. High signal intensity around the femoral tunnel aperture of the PL bundle means the poor graft-bone healing and may result in partial or complete rupture of the PL graft.

P13-541 A new guiding technique for precise placement of the double tibial tunnel in anterior cruciate ligament arhtroscopic anatomic reconstruction. Technical note Diaz-Heredia J.1, Ruiz-Iban M.A.1, Del Cura Varas M.S.1, Moros S.1, Gonzalez-Liza´n F.1 1 Hospital Ramon y Cajal, Dept. de Cirugı´a Ortope´dica y Traumatologı´a, Madrid, Spain Objectives: When an anatomic double bundle ACL reconstruction is performed with double tibial and femoral tunnel one technically challenging step is the correct placement of the tibial tunnels in such a way that these are correctly placed and not communicated. The available guides leave much to surgeon0 s expertise to place the guiding wires by hand or condition the placement of the posterolateral tunnel to prior positioning of the anteromedial tunnel. We present a method to place both tunnels correctly with usually available instruments and prior placement of the posterolateral tunnel. Methods: Two Standard Acufex guides are used, the tip guide and the elbow guide. With the tip guide geared to 558 the center of the posterolateral is determined and the wire is passed; a drillbit of appropriate size is passed over the wire and left in place with the tip intraarticular. The elbow guide geared at 458 is introduced in the joint and the tip is placed adyacent to the anterior part of the drillbit, then the wire is passed, the drillbit withdrawn and the anteromedial tunnel drilled. This allows for enough separation of both tunnels. Results: The technique was initially mastered in sawbones and cadavers. Subsequently it has been used in 30 subjects without complications related to tibial tunnel placement. Conclusions: The described technique allows for precise and adequate placement of both tibial tunnels in a very reproducible way.

P13-552 Relationship between the angle and location of the bone tunnel and isometricity during surgery and anterior laxity of the tibia in ACL reconstruction with BPTB Takahashi M.1, Matsubara T.1, Hanada M.1, Koyama H.1, Furuhashi R.1 1 Hamamatsu University School of Medicine, Orthopaedic Surgery, Hamamatsu, Japan

123

S170 Objectives: Several studies have demonstrated that in ACL reconstruction a less perpendicular the angle of the graft in the coronal plane to the joint line resulted in reduced loss of flexion and anterior laxity. As anatomical double bundle ACL reconstruction has been recently performed, anatomic reconstruction is now preferable to isometric reconstruction. However, the relationship between less perpendicular angles for bone tunnel and isometricity is unknown. The aim of the present study is to investigate the relationship between the angle and location of the bone tunnel and the isometricity during surgery and anterior laxity of the tibia in ACL reconstruction with bone - patella tendon - bone (BPTB). Methods: 40 patients who underwent ACL reconstruction with BPTB autograft were included in the study. They consisted of 18 females and 22 males, aged 14 - 50 years old, averaging 30 years old. Reconstruction was performed with single BPTB transtibial technique. Isometricity was measured as the length change of the positioned graft in the bone tunnel between 90 degree knee flex and 0 degree extension during surgery. The angle and localization of bone tunnels were measured by postoperative radiograph; the width of tibial articular cartilage (W), anterior-posterior tibia length (L), coronal angle of tibial bone tunnel (TA), coronal angle of femoral bone tunnel (FA), the percentage of the location of anterior (SA) and posterior (SP) borders of the tibial bone tunnel from the anterior border on the tibia sagittal plane. Anterior-posterior laxity of the knee was evaluated using a KT-2000 arthrometer postoperative 1 year. Statistical analysis was performed using Mann-Whitney U-test and Spearman rank correlation test. The significance level was set at less than 0.05. Results: For isometricity, the graft length was longer in extension than flexion in all subjects. Patients with less than 3 mm length change numbered 24 (group A), and patients with more than 3 mm length change numbered 16 (group B). The average of each measurement on the radiograph is, W, 78.3 mm, L, 57.7 mm, TA, 55.2 degree, FA, 60.0 degree, SA 30.4%, SP, 55.6%. The average side-to-side difference measured by KT2000 arthrometer was 1.5 mm. There was no relationship between the radiographic measurements and KT-2000 measurements. There was no difference in the radiographic measurements and KT-2000 measurements between group A and B. Among radiographic measurements, TA and TB significantly correlated with W and L. It had been expected that when using transtibial technique, the less perpendicular the coronal angle, the more posterior the location of the sagittal tibial bone tunnel, leading to less isometricity, however, it was not so. Conclusions: There was no relationship between the angle and location of the bone tunnel and isometricity during surgery and anterior laxity of the tibia in ACL reconstruction with BPTB in this study. This was due to the coronal bone angle being less perpendicular and having very small variation in the patients of this study. The coronal bone angle was determined by the size of the tibia as they correlated significantly.

P13-559 Evaluation of the bone bridge between the bone tunnels after anatomic double bundle anterior cruciate ligament reconstruction. A multi detector computed tomography study Hantes M.1, Liantsis A.1, Basdekis G.1, Karantanas A.2, Christel P.3, Malizos K.1 1 University Hospital of Larisa, Orthopaedics, Larisa, Greece, 2University Hospital of Heraklion, Radiology, Heraklion, Greece, 3Habib Medical Center Olaya, Riyadh, Sports Medicine Department, Riyadh, Saudi Arabia Objectives: The aim of this study was to evaluate if special aimers would be helpful to avoid intraoperative bone bridge fracture during doublebundle ACL reconstruction. In addition, given that no tunnel communication occurred during tunnel drilling, a second hypothesis of this study was that the bone bridge between the bone tunnels would maintain its structural integrity and no tunnel communication would be observed postoperatively because of tunnel enlargement or screw contact. Methods: Thirty-two patients undergoing double-bundle ACL reconstruction were included in this prospective study. A multi detector computed tomography study with multiplanar reconstructions was performed in all patients at a mean of 17 months postoperatively. The thickness of the bone bridge between the bone tunnels was measured, in

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 both the femoral and tibial side, on an axial and sagittal plane respectively, at three locations: a) at the level of the joint line (apex of the bone bridge) b) at the mid-portion of the bone bridge and c) at the base of the bone bridge. In addition, the bone density of the bone bridge was measured in Hounsfield Units in the same locations. Bone density of the anterior tibial cortex and adjacent cancellous area, and the cortex of the lateral femoral condyle and adjacent cancellous area were measured for comparisons. Results: Tunnel communication occurred intraoperatively in one patient on the tibial side at the level of the joint line. In the rest of the patients, a well defined triangular bone bridge was present between the two tunnels in both the femoral and tibial side. The thickness at the apex of the bone bridge was 2.0 and 2.2 mm for the femur and tibia respectively. In addition, the density of the bone bridge at its apex was similar to that of cortical bone. Conclusions: Our study demonstrated that anatomic aimers are of great help to avoid bone bridge fracture intraoperatively during double bundle ACL reconstruction. The bone bridge remains intact postoperatively although it is thin at the level of the joint line.

P13-560 Influence of knee flexion and femoral cross-pin insertion angle on posterolateral structures of the knee and lateral fixation lengths during ACL reconstruction Lee Y.S.1, Kim J.G.2, Ra H.J.2, Ahn J.H.3, Jung Y.B.4, Min B.-H.5 1 Ajou University Medical Center, Department of Orthopaedic Surgery, Suwon, Korea, Republic of, 2Inje University, Seoul, Republic of Korea, 3 Samsung Medical Center, Department of Orthopedic Surgery, Seoul, Republic of Korea, 4ChungAng University, Department of Orthopedic Surgery, Seoul, Republic of Korea, 5Ajou University, Suwon, Republic of Korea Objectives: The purposes of this study were to 1) evaluate the influence of knee flexion and femoral cross-pin insertion angles on knee PL structures and 2) evaluate the change in lateral fixation length of the cross-pin based on various combinations of knee flexion and femoral cross-pin insertion angles. Methods: Soft tissues of 10 fresh cadaveric knees were dissected until the lateral collateral ligament (LCL) and popliteus tendon (PT) were identified. A tibial tunnel (8 mm diameter) was made, and transtibial femoral tunnels (1:30 or 10:30 o’clock position) were made at three different knee flexion angles (70, 90, and 110). The cross-pin guide was inserted sequentially into three different femoral tunnels, and two crosspin guidewires (superior and inferior pins) were drilled at three different insertion angles (downward 30, 0 [parallel to floor line], and upward 30) for each knee flexion position. The distances from the insertion point of the two cross-pins to the LCL and PT and the distance from the lateral wall of femoral tunnel to the lateral cortex of the femoral condyle were measured. The measurements were taken twice by two orthopedic surgeons to reduce intra- and interobserver bias. Results: The inter- and intrarater reliability ranged from 0.82 to 0.91. No significant differences were observed in the superior and inferior pin depths (p = 0.56 and 0.39). The distances from the superior pin to the LCL and from the inferior pin to the LCL were significantly shorter in all knee flexions with 0 and an upward 30 insertion angle than with 70 and 90 knee flexion with a downward 30 insertion angle, respectively (p\0.05). No significant difference was observed in the distance between the superior pin and inferior pins and the PT (p = 0.25). The distance from the inferior pin to the PT was sufficient with 70o knee-flexion and 30o downward pin-insertion angle and significantly longer than in most other conditions (p \ 0.05). The length was shorter with a 0 insertion angle in all three knee-flexion angles than under the other pin-insertion conditions. Conclusions: The cross-pin was inserted close to the LCL and PT, and a downward 30 angle was the safest insertion angle. Lateral fixation length was sufficient for the cross-pin fixation in the 10:30- or 1:30-positioned femoral tunnel.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P13-573 Quantitative assessment of knee static and dynamic instability by means of a navigation system Zaffagnini S.1, Lopomo N.1, Bignozzi S.1, Marcacci M.1, Visani A.1 1 Istituto Ortopedico Rizzoli, Laboratorio di Biomeccanica, Bologna, Italy Objectives: Lachman, drawer and pivot-shift (PS) tests are extremely important in the assessment of ACL reconstruction. The goal of this work was to analyze the reliability of PS test using a navigation system, identifying a set of new quantitative parameters and evaluating their clinical relevance. Methods: 18 patients that underwent anatomic double-bundle ACL reconstruction were included in our study. To evaluate the joint laxity and kinematics we used an optical navigation system focused in kinematic acquisitions. The new dynamic parameters we focused on were: antero/ posterior (AP) translation of medial, lateral compartment and joint center; internal/external (IE) and varus/valgus (VV) rotation of the joint during PS test. For each parameter we measured the peaks and the areas obtained during the test (Fig. 1).

Scheme of PS parameters Intra-tester repeatability, comparisons of pre-operative and post-operative laxities and correlations between the PS peaks and the corresponding peaks obtained with standard static tests were evaluated. Areas, peaks and static laxity outcomes were compared grouping patients according to the pre-operative IKDC score. Results: Static and dynamic laxities were significantly reduced by the reconstruction (fig. 2, Left). For what concerns PS test we found reduction in coupled peaks of AP translations, in coupled IE and VV rotations (fig 2, Center). The analysis of the area highlighted a huge recovery of the dynamic stability of the joint (fig. 2, Right). Correlation analysis showed good coefficients both for pre-operative and post-operative values. Patients with a IKDC ‘‘D’’ grade had specific larger areas during the PS compared to patients with a IKDC ‘‘C’’ grade.

S171 Conclusions: The navigation allowed a quantitative evaluation of knee global kinematics before and after surgery: all ACL-deficient knees showed a positive PS-test before the surgery with huge value in the subluxation of the tibia; anatomical DB seems to eliminate both static instability and to control PS phenomenon. The presented analysis was therefore helpful to characterize patient-specific laxity and surgical performance, highlighting the clinical relevance of PS test.

P13-576 Depression, anxiety and quality of life acute after ACL-injury Olsson I.1, Kvist J.1 1 Institute of Medicine and Health Sciences, Physiotherapy, Linko¨ping, Sweden Objectives: The purpose of this ongoing study is to investigate the mental health, i.e. depression and anxiety, and quality of life acute after ACLinjury. Methods: All patients between 18 and 45 years with preliminary diagnosis of ACL-injury are asked for inclusion in the present study. Patients are recruited from orthopedic clinics in one region in Sweden. A package of questionnaires is send to the patients as soon as possible after the preliminary diagnosis. Preliminary results of two questionnaires will be presented in this abstract and more results from the ongoing study will be presented in the conference. One questionnaire, the Hospital Anxiety and Depression Scale (HADS) investigates anxiety (7 questions) and depression (7 questions). The score range from 0 indicating no problems to 21 at each dimension indicating problems that may require medical consultation. The other questionnaire, EuroQol-5 Dimensions (EQ-5D) investigates generic health-related quality of life, divided in five dimensions; mobility, hygiene, the principal activities, pain/ disorders and fear/ depression. The results are presented in an index score ranged from 0 to 1, where 1 indicates full health. In addition, the EQ-5D has a VAS-estimate (100 indicates best imaginable health) where the current health status is measured by self-estimation. In order to get reference values from a non-injured athletic population, the HADS and EQ-5D will be answered from approximately 100 soccer players. Results: Preliminary results from 28 patients and 41 soccer players are presented in the table. The 28 ACL-injured players (13 men and 15 women) answered the questionnaires in a mean 40 (SD24, range 8-116) days after the injury. The 41 soccer players (34 men and 7 women) answered the questionnaires during a training session. The results indicates some trends for higher anxiety (p=0.086, Mann-Whitney test) in the patients with ACL-injury. The health related quality of life was significantly decreased in the ACL-injured patients (p\0.001). These

Laxities; PS coupled peaks; PS areas

123

S172

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

results are preliminary and result from more patients will be presented in the conference.

Depression, anxiety and quality of life ACL-injured n=28

Soccer n=41

HADS depression (median and percentiles)

3 (1–5.75)

2 (1–4)

HADS anxiety (median and percentiles)

6 (2.25–7.75)

4 (2–6)

players

EQ-5D score (mean and SD)

58 (20)

81 (14)

EQ-5D VAS-estimate (mean and SD)

0.5975 (0.2)

0.8936 (0.1)

Conclusions: The preliminary results of the present study, indicates that an ACL-injury have an effect on the patients self rated quality of life, and may also affect their mental health. Previous studies have shown that competitive athletes regard sport injuries to be very stressful psychologically and both depression and anger increase after the injury. Impaired mood has a negative impact on rehabilitation and recent studies have shown that patients’ experience of an acute trauma can influence the recovery and the rehabilitation process. The patient’s perspective on his/ her injury is of great importance in order to improve the rehabilitation process and finally have a well functioning individual who is satisfied with his/ her knee and with the care provided to him/ her.

P13-579 Relationship between femoral tunnel placement and isometricity in bi-socket anterior cruciate ligament reconstruction Furuhashi R.1, Takahashi M.1, Koyama H.1, Hanada M.1, Matsubara T.1 1 Hamamatsu University School of Medicine, Orthopaedic Surgery, Hamamatsu, Japan Objectives: Recently, in anterior cruciate ligament (ACL) reconstruction, anatomical double bundle reconstruction that reconstructs the anteromedial bundle (AM) and posterolateral bundle (PL) is emphasized, however, the isometricity is unknown. The purpose of this study was to investigate the relationship between the femoral tunnel placement of AM and PL and isometricity in bi-socket ACL reconstruction. Methods: 23 knees of 23 patients, with a mean age of 29.7 years, were examined. All knees were autogenous hamstring reconstructions using the bi-socket procedure. We examined the location of the femoral tunnels of AM and PL by inserting a pin into the femoral tunnels and taking a lateral radiograph of the knee with an image intensifier during surgery. The femoral tunnel placements were evaluated using the quadrant method described by Bernard et al. The length changes of grafts were measured between 90 knee flexion and 0 extension. The distances from the posterior margin in the deep-shallow direction and from the roof in the high-low direction to center of the femoral tunnel placement of the 2 bundles were measured. Results: 13 of 23 tunnel placements of AM were mapped in C1 of the quadrant, 7 of 23 were mapped in C2, 2 of 23 were mapped in D1, 1 of 23 were mapped in B2. 11of 23 of PL were mapped in C2, 8 of 23 were mapped in D2, 2 of 23 were mapped in C3, 1 of 23 was mapped in D1, 1 of 23 was mapped in D3. The location of the AM tunnel was averaged at 28.8% from the posterior margin in the deep-shallow direction and 20.8% from the roof in the proximal to distal direction, whereas PL was averaged at 23.2% from the deep margin and 40.2% from the roof. The length change of AM and PL was, on average, 3.4mm and 7.6mm. There was no statistically significant relationship between the femoral tunnel placement and the length change. Conclusions: Our previous anatomical study of femoral insertions of the AM and PL showed that the most common location for AM was C2, and C3 for PL. In this study, the femoral tunnels were reconstructed higher and deeper than the anatomical ACL insertions. There was no statistically

123

significant relationship between the femoral tunnel placement and length change, so we examined the relationship between the distance of the femoral tunnel placement and anatomical location and length change. The distance in the deep-shallow direction of PL correlated with the length change. In the deep-shallow direction, the closer the femoral tunnel placements of PL to anatomical insertion, the shorter the length changes were. Regarding reconstructed PL, the anatomical insertion might be most isometric in femoral tunnel placement reconstruction.

P13-588 Mechanical stability of single and double bundle ACL reconstruction femoral fixation methods in an experimental porcine model Bagger Bohn M.1, Dalstra M.2, Søballe K.3, Lind M.1 1 Aarhus University Hospital, Sports Trauma Clinic, Aarhus, Denmark, 2 University of Aarhus, Dept. of Orthodontics, Aarhus, Denmark, 3Aarhus University Hospital, Dept of Orthopedics, Aarhus, Denmark Objectives: ACL reconstruction is traditionally performed using Single Bundle technique with grafts/drill holes in femur of 8-9mm in diameter. Anatomical double-bundle ACL reconstruction is gaining popularity, and this technique uses two grafts/drill holes in femur and tibia of smaller diameters. The biomechanical properties of tendon fixation in smaller diameter/ drill holes is therefore of importance. The aim of this study is to compare biomechanical properties of fixation devices used for ACL reconstruction on the femoral side in three diameters (6, 9 and 2x6mm) using a porcine in vitro model. Methods: ACL reconstructions were performed using two different fixation techniques in porcine femora: Endobutton CL (EB) and Hexalon interference screw (IS). Each technique was performed in diameters 6, 9 and 2x6mm.Ten specimens were tested for every diameter and technique. The specimens were tested for 1000 cycles between 50 and 250N on an MTS 858 Mini Bionix material testing machine and subsequently tested to failure. Following parameters were determined; displacements at 50, 100, 500 and 1000 cycles, the stiffness over the first 0.5mm of the failure test, the force to failure, displacement to failure and the energy to failure. Results: Load to failure: All Single bundle groups displayed a statistically significant different load to failure, which could be ranked in the following order: EB-9mm (947N), IS-9mm (708N), EB-6mm (569N), IS-6mm (433N). Comparing the two double bundle techniques Endobutton fixation displayed a stronger load to failure; EB-2x6mm (1071N) vs IS-2x6mm (806N) (p=0.01). Elongation: The Elongation of the IS-2x6mm (2.08mm) was less than IS9mm (3.27mm) and EB-2x6mm (3.23mm) groups (p=0.03). No statistically significant differences were found between the elongations of the various single bundle groups. Stiffness: The stiffness of the IS-9mm group (353N/mm) was higher than the EB-9mm (253N/mm) (P\0.01) and IS-6mm (274N/mm) groups (p=0.02). For the double bundle groups IS-2x6mm (351N/mm) was stiffer than EB-2x6mm (285N/mm)(p=0.03). Conclusions: A 6mm diameter graft construct as used in double bundle ACL reconstruction has 40% less strength than a single bundle 9mm construct. This could be failure risk in knee flexion angles where only one bundle is loaded. Endobutton fixation is superior to interference screw fixation concerning maximum load to failure. The interference screw fixation results in higher stiffness of the femur/graft complex than the endobutton fixation. Having two bundles reduced the elongation and stiffness for both fixation methods when compared to the single bundle large graft diameter constructs.

P13-614 A comparative study of ACL reconstruction with three different techniques Apostolopoulos I.1, Karagiannis A.2, Tsolos I.1, Staikidou I.3, Giannikouris G.3, Panigirakis N.1, Mantzikopoulos G.3, Andreakos A.1 1 1st IKA Athens Hospital, A’ Orthopaedic Dept., Athens, Greece, 2IASO Hospital, Athens, Greece, 31st IKA Athens Hospital, MRI, Athens, Greece

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Objectives: The purpose of the present study was the assessment of graft stability using three different hamstring ACL fixation devices - (Retrobutton, Cross-Pin, and AperFix) and the comparative evaluation of the results of these techniques with 24-month follow-up. Methods: From October 2005 until May 2009, 89 patients underwent ACL reconstruction with hamstring tendons (i) Femoral extracortical or nonanatomical device (Retrobutton, Arthrex), (ii) Femoral transfixation pin fixation device (CrossPin, Linvatec) and (iii) Aperture fixation (AperFix, Cayenne medical). Hybrid tibial fixation with interference screw and staple, absorbable screws in the first two techniques and non absorbable in the last one were used. The average age of the patients was 29 years (2044) and the mean time interval from injury to reconstruction was 10 months (3/52-31/52). The evaluation methods were clinical examination, knee scores and instrumented laxity measurements at 3, 6, 12 and 24 months. We also evaluated particular parameters in MRI at 6 months, and 1 year postoperatively. Results: Eleven patients were completely lost to follow up and three revisions were done, leaving 75 patients for analysis. Clinical evaluation by Lachman and pivot shift was normal or near normal in 72 patients. No statistically significant differences between the groups were observed both as regards the KT 1000 (1.5 less with AperFix, 1.0 mm less with Cross-Pin,) and as regards the subjective variables (Retrobutton having a slight advantage). Mean postoperative IKDC evaluation was 87,4/85,3/85,9 and Lysholm score was 89/86/85 respectively. Three patients underwent revision after a new traumatic incidence (2 in Cross-Pin group and one in the AperFix). Conclusions: There were no statistically or clinically relevant differences in the results two years postoperatively and all 3 techniques improved patients performance. All patients returned in their previous occupational activity and the vast majority of them also took up their previous social life (overall patients satisfaction 86,2/83,4/83,9).

P13-632 CT Scan evaluation of tunnel position in 38 failed cases of ACL reconstruction Neyret P.1, Lustig S.2, Elguindy A.3, Demey G.3, Servien E.4 1 Hopital Croix-Rousse - Centre Livet, Chirurgie Orthopedique, Lyon, France, 2Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 3Centre Albert Trillat, Lyon-Caluire, France, 4Centre Albert Trillat, Hospices Civils de Lyon, Lyon-Caluire, France Objectives: Tunnel malplacement is the most common technical error causing failure of ACL reconstruction. Tunnel positioning was commonly evaluated radiologically. The aim of our work was to define a reliable method for tunnel evaluation using 2D and 3D CT-Scan, and to compare the position of tunnels measured on the x-rays to that evaluated on CT scan. Methods: We present a retrospective study comprising 38 patients (29 males and 8 females) consulting for failed ACL reconstructions between 2004 and 2009. Radiographic analysis of tunnels was done by the regular method. We describe a new method for tunnel evaluation using CT-Scan. For the femoral tunnel we developed a quadrant applied on the sagittal cut of the lateral condyle, and used the level of the cut on which the tunnel opens in the notch to allow determining the Anteroposterior position of the tunnel in the notch. We used axial cuts to define its position in the frontal plane. The three readings were correlated to give the tunnel position with exactitude. For tibial tunnel we defined an axial cut for tunnel positioning. Eighteen patients had a 3D CT-Scan which allowed us to further describe the central view which allowed tunnel positioning in a single glance. Results: Our results confirmed the findings of other series, the majority of malpositions were femoral in origin: 69% of cases in our series. Meanwhile CT-Scan analysis allowed us to note that the majority of these malpositions (65%) weren’t anterior placement of the femoral tunnel, but its placement in a too central position, we also found that 47% of the tunnels judged radiographically as anterior were otherwise by the CT analysis. As for errors in tibial tunnel positioning, they were 34% of cases. Conclusions: This study permitted us to prove the difficulty of performing radiographic evaluation of the tunnels and its lack of reliability when compared with CT-Scan analysis.

S173 Many authors stated that CT-Scan doesn’t always allow defining the femoral tunnel position; this was limited by the methods they used for measurements. In fact they only relied on sagittal cuts for analysis of tunnel position. With the help of the axial cuts we have shown that if the femoral tunnel opening wasn’t seen on the sagittal cuts of the lateral condyle, this was synonymous with central or vertical tunnel position. The 3D CT-Scan reconstruction appeared to be very useful in obtaining these results with a single look. The central view permits us to evaluate the position of the tibial tunnel in relation to the notch. The use of CT-Scan in tunnel analysis is essential in failed cases of ACL reconstruction for accurate and reliable determination of the possible causes of failure and improving results in management of these increasingly common cases.

P13-668 Hamstrings versus quadriceps in double-bundle ACL reconstruction Hart R.1, Oka´l F.1 1 General Hospital of Znojmo, Dept. of Orthopaedics and Traumatology, Znojmo, Czech Republic Objectives: The aim of this prospective study was to evaluate early functional results and stability of the knee joint after anatomical (4-tunnel reconstruction with grafts from hamstrings) and semi-anatomical ACL reconstruction (3-tunnel reconstruction with tendon-bone quadriceps graft [2 tunnels in the femur and bone block in the tibia]). Methods: 40 patients (26 men and 14 women) in mean age of 27 years (range, 16 - 44 years) underwent isolated ACL reconstruction in 2007 and 2008. The minimum follow-up was 1 year (range, 12 - 27 months). Group I included 20 cases after anatomical reconstruction, Group II 20 cases after semi-anatomical reconstruction. No patient was lost from the study. Lysholm (0 - 100) and IKDC (0 - 10) scores were evaluated at the last follow-up. The ventral stability was measured by means of KT-1000. Pivot-shift test was used to evaluate the rotational stability. Statistical analysis was used to evaluate the obtained results (p \ 0,05). Results: The mean Lysholm score was in Group I 87,9 ± 11 points (range, 62 - 100 points) and in Group II 88,9 ± 12 points (range, 76-100 points), without statistically significant difference. The mean functional IKDC score was identical in both groups (8 ± 1). Knee joints after semi-anatomical reconstruction were a little bit more stable (KT-1000, 89 N) but without statistical significance; average comparison value in Group I was 1,98 mm and in Group II 1,45 mm. The pivot-shift phenomenon wasn0 t present in any case of both groups. The mean operating time was nearly the same in both groups (96 minutes in Group I and 93 minutes in Group II). Fracture of the patella (stabilized by 2 lag screws) occured in two cases during the graft harvest in Group II. No other complication was observed in both groups. Conclusions: In accordance with achieved results it is possible to declare that early clinical results of both techniques are similar, without any statistical difference. It is necessary to wait for long-term follow-up studies to confirm this contention. We prefer the semi-anatomical reconstruction in small knees with unfavourable anatomical conditions (narrow intercondylar notch, short tibial foot-print, and insufficient hamstring tendons).

P13-673 Anterior cruciate ligament reconstruction using remnant preservation and femoral tensioning: clinical and MRI results Ahn J.H.1, Lee S.H.1, Koo K.H.1 1 Samsung Medical Center, Department of Orthopedic Surgery, Seoul, Republic of Korea Objectives: At the time of anterior cruciate ligament (ACL) reconstruction, preservation and augmentation with the remnant original ACL tissue might be preferable in terms of preservation of mechano-receptors and vascularity. However, there are some concerns of cyclops lesion or incorrect tibial tunnel formation. Furthermore, it is technically demanding procedure to preserve the remnant tissue during ACL reconstruction. The purpose of this study is to report the preliminary clinical results of ACLR

123

S174 using preservation and femoral tensioning of remnant ACL described previously. Furthermore, the continuity of graft and the presence of Cyclops like mass lesion in front of the graft in postoperative MRI were evaluated. Methods: From July 2007 to July 2008, Seventy-four consecutive ACL reconstructions were performed. After exclusion of no remnant tissue to preserve, revision surgery, severe arthritic change, and concomitant other ligaments surgery, 63 patients were included in this study. Among them 55 patients were followed and evaluated clinically and using MRI with a mean 15.4 (±12-25) months of follow-up. Clinically, side-to-side difference in KT-2000 arthrometer, pivot-shift and Lachman test, extension limitation, Lysholm knee scale, Hospital for Special Surgery (HSS) score, and International Knee Documentation Committee (IKDC) subjective and objective knee form was evaluated. Follow-up MRI at 6 months after surgery was obtained in 36 patients for evaluation of graft continuity, the presence of Cyclops like mass lesion in front of ACL graft. Tunnel enlargement at last follow-up visit was measured. The concept was that the remnant tissue underwent only adjuvant therapy. Therefore, the standard ACL was reconstructed as a single bundle transtibial technique with a quadruple hamstring tendon autograft and the sutured remnant ACL bundle was pulled out to the same femoral tunnel. Results: There were 45 men and 10 women with a mean age of 31.3 years (17-54). The mean time from injury to surgery was 26.1 months (1-120). The mean Lysholm score, HSS score, and IKDC subjective scores were improved from 57.8 (±18.0), 81.1 (±12.7), and 58.7 (±16.0) to 89.2 (±5.9), 98.5 (±1.9), and 84.5 (±13.4), respectively (p\0.0001). Lachman and pivot shift test showed negative results in 53 of 55 patients and the mean side-to-side difference in KT-2000 arthrometer was 1.8 (±1.6) mm. Among the 45 patients in which postoperative MRI was available, one complete graft failure and one partial tear was observed. Total of 11 cyclops like mass lesion was observed in MRI but none of them showed extension limitation or pain at extension. Tibial bone tunnel at last clinical visit was enlarged by 2.2 (±1.5) mm compared to original tunnel diameter. Conclusions: Arthroscopic ACL reconstruction with the preservation and femoral tensioning of remnant tissue results in good clinical results without the concerns of extension limitation at minimum 12 months after surgery. Our remnant bundle preservation technique can be applied broader indication and expected minimal tunnel enlargement with precise tunnel placement.

P13-686 Anterior cruciate ligament reconstruction: 6-year minimum follow-up of a prospective randomized clinical trial comparing autologous BPTB and DSTG grafts Losco M.1, Giron F.1, Buzzi R.1, Giannini L.1, Mondanelli N.2, Aglietti P.1 1 University of Florence, First Orthopaedic Clinic, Firenze, Italy, 2AOUC Careggi, Firenze, Italy Objectives: The choice of graft for anterior cruciate ligament reconstruction is a matter of debate, with patellar and hamstring tendons being the two most popular autologous graft options. The objective of this study was to determine in a prospective, randomized clinical trial whether two most popular autologous grafts (bone patellar tendon bone, BPTB, or doubled hamstring tendons, DSTG) fixed with modern devices affect the mid-term clinical and radiographic outcomes of an anterior cruciate ligament reconstruction. Methods: One hundred and twenty patients with a chronic unilateral rupture of the anterior cruciate ligament were randomized, in a strictly alternating manner, to undergo an arthroscopically assisted ACL reconstruction with the use of an autologous BPTB or DSTG graft. Both groups were comparable with regard to demographic data, preoperative activity level, mechanism of injury, interval between the injury and the operation, and the amount of knee laxity present preoperatively. The same well-proven surgical technique and aggressive controlled rehabilitation was used. An independent observer, who was blinded with regard to the involved leg and the type of graft, performed the outcome assessment with use of a visual analog scale, the new International Knee Documentation Committee form, the Knee Injury

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 and Osteoarthritis Outcome Score, and an arthrometric evaluation. All patients were evaluated at a minimum 2-year follow-up. We were able to review at a minimum 6-year follow-up 49 patients (82%) in the BPTB group and 51 (85%) in the DSTG group. Results: At last evaluation, no differences were found in terms of the visual analog score, the Knee Injury and Osteoarthritis Outcome Score, the new International Knee Documentation Committee subjective and objective evaluation scores, the KT-1000 side-to-side laxity measurements or return to sports activities. In BPTB group we found more patients with kneeling discomfort (p\0.05) and hypoesthesia (p\0.05). There were 6 failures in DSTG group and 2 in BPTB group (n.s.). Radiographic evaluation showed an increased incidence of femoral tunnel widening in the DSTG group (15% BPTB vs 51% DSTG; p\.01). There were no differences between the two groups in term of degenerative joint changes. Conclusions: Both hamstring and patellar tendon grafts provided good subjective and objective outcomes at a six years minimum follow-up. BPTB group showed higher prevalence of donor site morbility, while the DSTG group showed higher incidence of femoral tunnel widening.

P13-700 The validity of using the clinical exam to quantify surgical outcomes Jacobs C.1, Branch T.2 1 ERMI, Inc., Atlanta, GA, United States, 2University Orthopaedic Clinic, Decatur, GA, United States Objectives: Recent literature comparing different ACL reconstruction techniques has used the Lachman and Pivot Shift tests to quantify surgical outcomes, suggesting that double-bundle techniques may provide more stable reconstructions. The purpose of this study was to evaluate the variability of these tests in a group of experienced surgeons when instrumented with an electromagnetic measurement system. Methods: Twenty-one board-certified orthopaedic surgeons specializing in sports medicine volunteered for the study (mean age = 47.7 y, years experience = 20 y), and data collection was completed over a 2-day period. A single subject with no history of traumatic knee injury was evaluated by all 21 surgeons. The tests were instrumented with an electromagnetic motion analysis system, with sensors being placed over the medial femoral epicondyle and the tibial tuberosity.

Placement of electromagnetic sensors For each test, surgeons performed 1 trial of 3 repetitions in succession, followed by 2 trials each consisting of a single repetition. The order of testing was consistent for each examiner, and testing was solely performed on the left knee. For each test, the amount of displacement and rotation were determined in all six degrees of freedom. The coefficient of variation over 3 trials was calculated for each surgeon, as well as the Minimum Detectable Change (MDC), which has previously been defined as the minimum amount of difference between 2 tests that can be considered clinically significant. Results: For the Lachman test, the mean displacement of the tibia relative to the femur was 6 ± 1.9 mm. These results are not unlike KT1000 testing,

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 which demonstrated 7 mm of displacement. However, while the mean value was similar to an instrumented test, the amount of displacement ranged from 3.7 to 11.7 mm between the 21 surgeons during manual testing.

Range of results from Lachman test Only 1 of 21 surgeons had a coefficient of variation value less than 10%, and the mean coefficient of variation was 25.8% for the group. The MDC was 2.5 mm, and 9 of 21 surgeons demonstrated differences of more than 2.5 mm between their 3 trials. The Pivot Shift movement patterns demonstrated even more variable results between surgeons. For example, 3 surgeons started with the knee fully flexed whereas the others started in full extension, and 1 surgeon externally rotated the knee while going into flexion, which dramatically differed from the others. Conclusions: The results of common tests of joint laxity vary greatly between surgeons. The results of these tests are influenced by hand placement, the amount and direction of force applied, as well as the rate at which the force is applied. Even when using advanced motion analysis systems to quantify the clinical exam, the results of manual testing performed by one surgeon cannot be generalized to the entire orthopaedic community. As such, the validity of these utilizing these tests to quantify surgical outcomes is questioned.

P13-704 Conservative treatment of partial ACL rupture - results after 15 to 19 years Lukasik P.1, Widuchowski W.1, Widuchowski J.1, Faltus R.1, Kwiatkowski G.1 1 District Hospital of Orthopedics and Trauma Surgery, Dept. of Knee Surgery, Arthroscopy and Sports Trauma, Piekary Slaskie, Poland Objectives: Anterior cruciate ligament (ACL) injury is probably the most frequent lesion of knee joint. There are many studies relating to treatment of total ACL rupture; however the problem of partial ACL rupture is not so often raised in long term studies. In our Department we observed that over 65% of ACL injuries relate to partial rupture. That’s why we consider the problem of partial ACL lesions to be very valid. The aim of this study was to present the long-term outcome of non-operative treatment of partial ACL rupture. Methods: 280 patients (156 males, 124 females, the average age: 24.7 years) were arthroscopically diagnosed to have an isolated, partial ACLrupture. 43 patients were excluded from the study due to a second surgery during the follow up (21 p. underwent ACL reconstruction, 16 - partial meniscectomy and 6 diagnostic arthroscopy). With an average follow-up of 16.8 years all patients were examined clinically (using Rolimeter), radiologically and evaluated with internationally accepted scores (Lysholm, IKDC, Tegner). The standing up and lateral x-rays of both knees were taken and the degree of osteoarthritis was classified according to Kellgren and Lawrence scale. Results: The mean value of Lysholm score was 82.5. According to IKDC scale 81.4% normal (A) or nearly normal (B) results were observed. The side to side laxity difference, estimated with Rolimeter, was 2.1 mm.

S175 Significant radiological changes (Grade II or more in the Kellgren & Lawrence scale) were seen in 34% of cases. The sport activity level, evaluated with Tegner score, was 5.2 pre-injury and 4.8 post-injury. Conclusions: Presented results show over 80% good and very good results according to Lysholm and IKDC scale. Significant osteoarthritis was observed only in one third of the patients. We considered that conservative treatment of partial ACL rupture could produce a good outcome in nonprofessional athletes (Tegner score about 5.0) and let them continue sports activity after treatment at the same level.

P13-708 Biomechanical correlation between common clinical examinations to determine ACL integrity Johansen S.1, Westerhaus B.2, Wijdicks C.2, Anderson C.2, Engebretsen L.1, LaPrade R.2 1 Ulleva˚l University Hospital, Orthopaedic Center, Oslo, Norway, 2 University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States Objectives: The heightened interest in anterior cruciate ligament (ACL) reconstruction has propagated a substantial amount of research on its biomechanics. However, discrepancies exist as to how to properly recreate clinically applied forces in the laboratory. It is important to consider the angles at which these forces are applied, as varying degrees of flexion will promote different changes in the biomechanical response of the knee. Our purpose was to compare different biomechanical tests at different angles of knee flexion with an intact and sectioned ACL to determine how well clinically applied loads are represented in the laboratory. Methods: Twelve non-paired, fresh-frozen cadaveric knees, with no evidence of prior injury or disease, were utilized for this study. Each knee underwent the same biomechanical testing protocol with the ACL intact and sectioned. Six-degrees-of-freedom positional data were collected during an 88 N anterior drawer force, a 5 Nm internal rotation torque, a 10 Nm valgus force, and two combined force applications to simulate the pivot shift test - one combining internal rotation and valgus force, and one combining internal rotation and anterior drawer force - at 08, 208, 308, 608, and 908 of knee flexion. Anterior drawer and valgus loads were applied using a 100 N force Model SM S-type Load Cell (Interface, Scottsdale, AZ), and the internal rotation torque was applied using a 15 N•m capacity Model TS12 Shaft Style Reaction Torque Transducer (Interface). Data was collected using a six-degrees-of-freedom electromagnetic tracking system (Polhemus, Colchester, VT), using The MotionMonitor software (Innovative Sports Training, Chicago, IL). Data reduction was accomplished through algorithm-based software, MATLAB (The MathWorks Inc., Natick, MA). A two-way analysis of variance compared the intact and sectioned states, with Tukey’s HSD used for post hoc detection of significant differences between intact and reconstructed states at respective degrees of knee flexion. Results: There was a significant difference in displacement for almost all testing conditions between the intact and the sectioned states. For the 88 N anterior drawer force, the sectioned state total displacements were significantly increased compared to the intact state at all angles of knee flexion. When a 5 Nm internal rotation torque was applied, the sectioned state produced displacements that were significantly increased over the intact state at each angle except at 90. With a 10 Nm valgus force application, sectioned state displacements were significantly increased over the intact state, except at 90. When the combined internal rotation and valgus forces were applied, the sectioned state displacements were significantly increased over the intact at all angles except 90. When the coupled internal rotation and anterior drawer forces were applied, the sectioned state displacements were significantly increased over the intact state at all angles except 30. Conclusions: Comparison of the two tested variations of pivot shifts found few absolute differences between the two methods. However, when internal rotation was combined with anterior drawer to represent the pivot shift, there was no significance at 30; which is a critical angle in recreating the pivot shift in the laboratory. Thus, it can be concluded that internal rotation combined with valgus is the optimal way to biomechanically re-create the pivot shift test for testing of ACL integrity.

123

S176 P13-711 Anatomic posterolateral knee reconstructions require a popliteofibular ligament reconstruction through a tibial tunnel Engebretsen L.1, McCarthy M.2, Camarda L.3, Wijdicks C.2, Johansen S.1, LaPrade R.2 1 Ulleva˚l University Hospital, Orthopaedic Center, Oslo, Norway, 2 University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States, 3University of Palermo, Department of Orthopaedic and Traumatology, Palermo, Italy Objectives: To our knowledge, no study has validated the importance of the popliteofibular ligament in an anatomic reconstruction, as much debate has arisen in the literature regarding the importance of its inclusion in the reconstruction. Some have described the popliteofibular ligament as detrimental to the restoration of an intact knee function, citing over constraint in internal rotation and subsequent limitation in external rotation. Thus, the purpose of our study is to assess two reconstructions, one with and one without the popliteofibular ligament, in comparison to the intact state. Methods: Six paired, fresh-frozen cadaveric knee specimens were used in this study, each without evidence of prior injury. The femur was sectioned 20 cm from the joint line and the tibia 13 cm from the joint line. The specimens were then potted in polymethylmethacrylate (PMMA) to ensure secure fixation. A customized knee testing apparatus, previously described, was used to firmly hold the femur while allowing movement of the tibia and biomechanical testing at various knee flexion states External forces were applied at 0, 20, 30, 60 and 90 of knee flexion. For each test state, applied forces used were 10 Nm varus/valgus load, 5 Nm internal/external rotation torques and 88 N anterior/posterior drawer loads. Load and motion data were recorded in synchrony using the Motion Monitor software (Innovative Sport Training, Chicago, IL). Each knee in a matched pair underwent one of two anatomic posterolateral knee reconstructions. Group 1 knees had all three major posterolateral knee stabilizers reconstructed in a method previously described. The matched knee, in Group 2, was reconstructed in the same way, except here the popliteofibular ligament was left out of the reconstruction. Instead of reconstructing it through a tibial tunnel, as in Group 1, the fibular collateral graft exiting the posteromedial fibula was brought back proximally and sutured onto itself. Each knee was tested in sequential order: intact knee, followed by reconstruction, followed by testing of complete sectioning of these three structures through the tibial tunnel. Results: For an applied external rotation torque, we found significant changes in comparing the sectioned to intact knees at all degrees of knee flexion tested for both groups of knees (p\.05). In analyzing the data for the two reconstruction techniques, we found no significant differences in reconstructing the fibular collateral ligament, popliteus tendon and popliteofibular ligament (group 1 reconstruction) or the modified reconstruction (group 2) compared to the intact state for external rotation. We found significant increases in varus rotation when comparing the sectioned to intact knees at all degrees of knee flexion tested for both groups (p \.05). There were no significant differences between the intact state and group 1 reconstructions at any flexion angle. There were significant increases in varus gapping between the group 2 reconstructions and the intact knee at 0 (p\.05), 20 (p\.05) and 60 (p\.05) degrees of knee flexion. Conclusions: Our results show the importance of the popliteofibular ligament in anatomic posterolateral knee reconstructions, because its inclusion more closely reproduced intact knee biomechanics. Further, our data did not show abnormal restriction of knee motion with the popliteofibular ligament placed through a tibial tunnel.

P13-712 Non-operative vs. operative treatment of anterior cruciate ligament rupture - 15 to 19 year follow-up results Lukasik P.1, Widuchowski J.1, Widuchowski W.1, Faltus R.1, Kwiatkowski G.1 1 District Hospital of Orthopedics and Trauma Surgery, Depart. of Knee Surgery, Arthroscopy and Sports Trauma, Piekary Slaskie, Poland Objectives: Anterior cruciate ligament (ACL) rupture always leads to different intensified instability of knee joint. The aim of medical treatment

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 for ACL-deficient knee is to restore normal function, stability and prevent the joint from developing osteoarthrosis. The majority of surgeons consider that ruptured ACL requires operative management because only surgical reconstruction can restore pre-operative knee function. But there are some who prove that non-operative treatment can also lead to good clinical results. The aim of our study was to evaluate and compare the long term outcome of operative and non-operative management of ACL ruptures. Methods: 158 patients (120 males, 38 females, the average age: 23.37 years) were arthroscopically diagnosed to have an isolated ACL-rupture and divided into two groups. The first group (98 patients) was treated surgically (OP), the ACL was reconstructed using the bone-patellar tendon-bone autologous graft. The second group (60 patients) was treated non operatively (NON-OP). With a mean follow-up of 16.8 years all patients were examined clinically (using a Rolimeter) and radiologically and evaluated with internationally accepted scores (Lysholm, IKDC, Tegner). The standing up and lateral x-rays of both knees were taken and the degree of osteoarthritis was classified according to Kellgren and Lawrence scale. Results: The side to side laxity difference, estimated with Rolimeter, was 1.8 mm. in OP group and 5.3 mm. in NON-OP group (p\0.05). According to Lysholm scale, the average results were 87.3 points (OP group) and 70.6 points (NON-OP group). Table 1 Postoperative knee function, evaluated with Lysholm and IKDC scales, showed statistically significant better results for ACL reconstruction group (p\0.05) IKDC

OP (%) vs. NON-OP (%)

A (normal)

44.4 vs. 18.2

B (nearly normal)

25.0 vs. 31.8

C (abnormal)

27.8 vs. 40.9

D (severely abnormal)

2.8 vs. 9.1

Table 2 Osteoarthritis was seen in evaluated knees in both groups, but more significant radiological changes (Grade II or more in the Kellgren & Lawrence scale) were seen after operative treatment Radiology (Kellgren&Lawrence scale)

OP (%) vs. NON-OP (%)

Grade 0

19.4 vs. 27.2

Grade I

27.8 vs. 27.3

Grade II

27.8 vs. 31.8

Grade III

25.0 vs. 13.6

Grade IV

0 vs. 0

Table 3 Post operative sport activity evaluated with Tegner score showed no statistically significant difference (p[0.05) Tegner score

OP (%) vs. NON-OP (%)

Pre- injury

6.31 vs. 6.28

Post- injury

5.12 vs. 4.90

Conclusions: The activity level after 15 to 19 years after ACL rupture was almost the same in both groups. A significant difference was seen in postoperative function expressed in IKDC and Lysholm scales with higher level after ACL reconstruction. Also the side to side laxity difference, estimated with Rolimeter, was lower in the knees after surgical treatment. Definite osteoarthritis (grade 2 or more) was observed in 52.8% after ACL reconstruction compared to 45.4% with conservative treatment. After ACL reconstruction we can observe better knee function and stability, but the operative treatment does not prevent osteoarthritis in the long term follow-up.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P13-713 Instrumented tests of tibial anterior translation and internal rotation laxity between single- and double-bundle ACL reconstructions Branch T.1, Siebold R.2, Freedberg H.3, Jacobs C.4 1 University Orthopaedic Clinic, Decatur, GA, United States, 2ATOS Praxisklinik Heidelberg, Abt. Knie- u. Fußchirurgie, Heidelberg, Germany, 3Suburban Orthopaedics, Bartlett, IL, United States, 4ERMI, Inc., Atlanta, GA, United States Objectives: Recent cadaveric studies have suggested that double-bundle ACL reconstruction techniques may result in a greater risk of over-constraining the knee. The aims of this clinical study were to compare instrumented in vivo measures of anterior and rotational stability in patients treated with either single- (SB) or double-bundle (DB) ACL reconstructions, as well as how consistently the two techniques reproduced the biomechanics of the uninjured contralateral knee. Methods: Participants included 64 patients that had undergone unilateral SB or DB hamstring ACL reconstruction. The KT1000 was used to bilaterally test anterior tibial translation in response to a 134 N force. A robotic testing system was used to assess rotational stability.

Robotic rotational laxity testing system Both the reconstructed and healthy knees were cycled into internal and external rotation until a torque of 5.65 N m was reached, and 3 preconditioning and 4 test cycles were performed. During the test cycles, the degree of maximum tibial internal rotation was recorded with an electromagnetic measurement system. In addition to comparing the maximum amount of tibial anterior translation and internal rotation, each reconstruction was evaluated to determine how well the reconstruction mimicked the biomechanics of the normal knee. A reconstruction was defined as mimicking the normal knee if side to side differences in anterior tibial translation were B 3 mm and side to side differences in internal rotation were within 3 degrees. Results: Side-to-side differences in anterior translation were significantly higher for the SB group (2.2±1.4 mm) than the DB group (1.1±1.0 mm, p=0.001). Side-to-side differences in internal rotation did not differ between the SB (0.4±4.7 degrees) and DB groups (0.2±2.6 degrees, p=0.82). A significantly greater percentage of DB patients (81%, p=0.008) had both anterior and rotational laxity similar to the normal knee, compared to 38% of the SB patients. Only 1 knee in the DB group was over-constrained in internal rotation compared to 8 knees in the SB group. Table 1 Reconstructions mimicking the normal knee Within Range (B 3 mm and 3)

Underconstrained

Overconstrained

Single-bundle

12

12

8

Double-bundle

26

5

1

Conclusions: Recreating symmetrical biomechanical characteristics between the ACL-injured and normal knee is vital to both the shortand long-term success of ACL reconstruction. While this can be

S177 achieved with either a SB or DB technique, DB ACL reconstruction more consistently reproduced the biomechanical profile of the uninjured limb than did the SB technique without increasing the risk of overconstraining the knee.

P13-714 A comparison between males and females in the outcome after ACL-reconstruction using hamstring tendon autograft Ahlde´n M.1, Ma˚nsson O.2, Karlsson J.1, Ejerhed L.3, Sernert N.4, Kartus J.5 1 Institute of Clinical Sciences, Department of Orthopaedics, Go¨teborg, Sweden, 2NU- Hospital Organization, Trollha¨ttan/Uddevalla, Orthopedic Department, Uddevalla, Sweden, 3Uddevalla sjukhus, Uddevalla, Sweden, 4 ¨ lvsborg County Hospital, Trollha¨ttan, Sweden, 5NU- Hospital Norra A Organization, Trollha¨ttan/Uddevalla, Department of Orthopaedics, Trollha¨ttan, Sweden Objectives: The hypothesis is that there would be no significant differences between males and females regarding knee laxity, clinical outcome and donor-site morbidity after ACL reconstruction using four stranded hamstring tendon autograft and interference screw fixation. Methods: 244 patients (141 males, 103 females) undergoing ACL-reconstruction using four stranded hamstring tendon autograft and interference screw fixation were included. The femoral tunnel was drilled through the anteromedial portal in all patients. All patients attended a clinical examination pre-operatively and at 2 years involving measurements of knee laxity and range of motion, assessment of donor-site morbidity, functional tests and subjective evaluation. Results: Pre-operatively the study groups were comparable in terms of demographics except for age (males 28 years, females 26 years p=0.03). At the 2 year follow-up, there were no significant differences between the two study groups in terms of KT-1000 arthrometer side-to-side laxity, manual Lachman test, donor-site morbidity, Tegner activity scale, Lysholm knee scoring scale and range of motion. However, both study groups improved significantly in their clinical assessments and functional scores compared to their preoperative values. Conclusions: There are no significant differences between males and females in the clinical outcome at 2 years after ACL-reconstruction using hamstring tendon autograft. It therefore appears, that gender is not a factor to be considered in terms of the choice of graft for ACL reconstruction.

P13-718 A new device for tibial graft fixation in double bundle ACL reconstruction: structural properties of the Mini-EndoTack Lenschow S.1, Rosslenbroich S.1, Raschke M.1, Petersen W.2, Zantop T.3 1 Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 2Martin-LutherKrankenhaus, Unfallchirurgie, Berlin, Germany, 3Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany Objectives: Conventional fixation devices may not be able to appreciate the special needs for a two tibial tunnel ACL reconstruction technique. The aim of this study was to analyze the initial stability of a new smaller implant for the fixation of soft tissue tendon grafts and to compare it to a single bundle implant. Our hypothesis was that there is no significant difference in the structural properties between the two implants. Methods: Fresh frozen skeletally mature porcine knees and human hamstringtendons were used for this study. A 7mm tunnel was drilled. In all tests two stranded ACL grafts with a diameter of 7mm were used. A baseball stitch was used to sew the strands to each other using a non biodegradable Ethibond suture (Ethibond 2, Ethicon). In group 1 the grafts were fixed with the Mini-EndoTack (7mm, Karl Storz, Tuttlingen, Germany), in group 2 the fixation was achieved by the

123

S178 conventional EndoTack (9mm, Karl Storz, Tuttlingen, Germany) (Fig.1).

The MiniEndotack The cortex was opened by using a 7mm (group 1) or 9mm (group 2) cannulated drill bit over a K-wire placed in the tibial insertion of the ACL. The specimens were mounted to a material testing machine (LR5K-plus, Lloyd instruments, Great Britain). The proximal end of the graft was clamped in a custom-made freezing clamp. The load was applied in line with the bone tunnel. A preload of 5 N was applied before the grafts were preconditioned between 0 and 20 for 10 cycles. Then the graft-bonecomplex was loaded to failure. Maximum load, yield load and stiffness as well as failure mode were recorded. Statistic analysis was performed using a student T-test with the significance level set at p\ 0.05 (SPSS version 11.0, SPSS, Chicago, IL, USA). Ten tests were performed in each group. Results: There was no significant difference in stiffness between group 1 (21.75 ± 3.36 N/mm) and group 2 (24.65 ± 3.40 N/mm) (p[0.05). The yield load varied in group 1 between 150.31 and 199.77N. In group 2 values between 165.48 and 215.62 N were found. No significant difference was found between group 1 and group 2 (p[0.05). The maximum load showed no significant difference between group 1 (295.63 ± 27.35 N) and group 2 (308.24 ± 33.31) (p[0.05). The most common failure mode was a rupture of the linkage material. Conclusions: The results of this study support our hypothesis that the structural properties of conventional single bundle and the new double bundle implant show no significant differences. The clinical relevance is that by using the smaller implant, the surgeon can use a smaller skin incision and an overlapping of the tibial buttons in anatomical double bundle reconstruction can be prevented. However, few the limitations apply to this study since we tested a worst case scenario with the force of pullout in line to the bone tunnel. Additionally, care should be taken as we can only speculate about in vivo forces an intact ACL or ACL graft has to withstand. The data acquired in the present study are comparable to the fixation with bigger buttons which are already in clinical use. Our results suggest that the Mini-EndoTack is an alternative to larger sized buttons for tibial fixation concerning initial fixation strength.

P13-719 Strategy of the visual proprioceptive control in patients with injury to the anterior crucial ligament of the knee and healthy individuals (soccer players) Piontek T.1, Ciemniewska-Gorzela K.2, Szulc A.2, Dudzinski W.1 1 Rehasport Clinic, Poznan´, Poland, 2University of Medical Sciences, Clinic of Peadiatric Orthopedic Surgery, Poznan, Poland Objectives: Knee joint dysfunction resulting from injury to the anterior crucial ligament (ACL) is associated not only with mechanical joint instability but also with damage of ligamentous receptors responsible for the joint proprioception. It was found that disturbances of signals from the damaged joint produce disorders in movements perception and position of the analogous joint in the normal limb. It plays an important role in balance maintenance during various motor activities.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 This study is aimed at evaluating control strategy in patients with the injury to the anterior crucial ligament. Methods: 84 men, aged between 15 to 55 years (mean age 27 years) were included into this study. Patients were divided into two groups: those with unilateral injury to ACL (33 patients) and control group of healthy volunteers (soccer players; 51 men). Anterior crucial ligament damage was confirmed with arthroscopic knee joint examination in every patient. The way of visual proprioceptive control was assessed with both dynamic (DRT) and static (SRT) Riva tests in monopodalic stance. Tests were performed with the Delos Postural Proprioceptive System (Delos s.r.l., Corso Lecce, Torino, Italy) in the biomechanical evaluation laboratory at Rehasport Clinic in Poznan´. Results: Statistically significant difference for deviations from the averaged axis in SRT (static Riva test) at closed eyes was found between the limb with damaged ACL and normal limb in the group of patients with injury to ACL (p=0.006) and between the limb with damaged ACL and normal limbs in healthy volunteers (p=0.022). Statistically significant difference for deviations from the averaged axis in SRT at closed eyes was also found between dominating and not dominating limb in healthy volunteers (p=0.013). No significant differences in the results of tests at open eyes were noted. Conclusions: Neurological deficits of proprioceptive perception, associated with the injury to ACL and affecting the balance, may be noted only in the results of tests performed with closed eyes. The results of any systems and their contribution to the visual proprioceptive control suggest an important role of visual system in compensation of archeproprioceptive system disorders resulting from the injury to ACL.

P13-724 Navigated anterior cruciate ligament replacement. An experimental validation study Jenny J.-Y.1, Ciobanu E.2 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France, 2Hoˆpitaux Universitaires de Strasbourg, Centre de Chirurgie Orthope´dique et de la Main, Illkirch, France Objectives: We wanted to test the following hypothesis: the location of the tibial and femoral anterior cruciate ligament (ACL) attachments measured by a navigation system differs from their actual anatomic location. Methods: 10 gross specimens were studied. The standard navigated procedure was used. Metallic reference pins were implanted around the tibial and femoral ACL attachments, and palpated with a navigated stylus. Their position was recorded in comparison to the same tibial or femoral reference point. All measurements were repeated with a caliper. Caliper and navigated measurements were compared with a paired Wilcoxon t-test and a Spearman correlation test at a 0.05 level of significance. Concordance between measurements were analyzed with the graphical Bland-Altman method. Results: There was a significant difference between caliper and navigated measurements at the tibia in absolute value (maximal range of 4 mm, mean difference about 2 mm), but the difference ratio was low (about 3%); there was no significant difference between caliper and navigated measurements at the femur. There was a strong correlation and a good agreement between caliper and navigated measurements. Conclusions: The difference observed were small and may have few, if any, clinical relevance. The navigation system used during the present study has the potential to enhance the precision of the intra-operative measurement of the tunnel placement during ACL replacement.

P13-726 Anatomic attachment of the ACL. A multi-modal study Jenny J.-Y.1, Ciobanu E.2 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France, 2Hoˆpitaux Universitaires de Strasbourg, Centre de Chirurgie Orthope´dique et de la Main, Illkirch, France Objectives: We wanted to test the following hypothesis: the location of the tibial and femoral anterior cruciate ligament (ACL) attachments will be

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 different according to the measurement technique: anatomic, radiologic, CT-scan, navigation system. Methods: 10 gross specimens were studied. Metallic reference pins were implanted around the tibial and femoral ACL attachments. Their location was recorded with reference to the bone contours with a caliper on the anatomic preparation, with a navigation system for ACL reconstruction, with standard plain AP and lateral X-rays and with a CT-scan. Results were compared globally with an ANOVA test at a 0.05 level of significance. Multiple post-hoc comparisons were performed with a paired Wilcoxon t-test and a Spearman correlation test at a 0.01 level of significance and with the Bland-Altman technique. Results: There was a significant difference the different measurement techniques in absolute value (maximal range of 4 mm, mean difference about 2 mm), but the difference ratio was low (about 3%). There was no significant difference within any couple of techniques, with a strong correlation and a good agreement within any couple of techniques. Conclusions: The difference observed were small and may have few, if any, clinical relevance. The navigation system used during the present study has the potential to enhance the precision of the intra-operative measurement of the tunnel placement during ACL replacement.

P13-737 Gender differences in patient-reported outcomes after ACL reconstruction - data from the Swedish knee ligament register Ageberg E.1, Forssblad M.2, Herbertsson P.3, Roos E.M.4 1 Lund University, Orthopedics, Clinical Sciences, Lund, Sweden, 2 Stockholm Sport Trauma Research Center, Capio Artro Clinic, Stockholm, Sweden, 3Lund University, Orthopedics, Clinical Sciences Lund, Lund, Sweden, 4University of Southern Denmark, Institute of Sports Science and Clinical Biomechanics, Odense, Denmark Objectives: To study gender differences in patient-reported outcomes before and at 1 and 2 years after ACL reconstruction and to present reference values. Methods: During 2005 to 2008, 10164 patients (mean age 27 years, SD 9.8, 42% women) with primary ACL reconstruction were registered in the Swedish national knee ligament register. 4438 (44%) and 5255 (52%) of these patients (42% women) had completed the knee-specific questionnaire Knee injury and Osteoarthritis Outcome Score (KOOS) and the generic score of health status EQ-5D before surgery and were included in this study. Independent t-tests were used to study gender differences in KOOS and EQ-5D at pre-op, 1 and 2 years post-op, and over time. Results: At pre-op, women reported worse scores than men in 4 KOOS subscales (pain, symptoms, sport/rec, QoL) and EQ-5D, with the largest difference seen in KOOS sport/rec (mean difference 4.7, 95% CI 3.0 to 6.3). At 1 year post-op, women reported worse scores than men in KOOS pain (mean difference 1.4, 95% CI 0.4 to 2.4) and KOOS sport/rec (mean difference 2.7, 95% CI 0.9 to 4.4), and at 2 years post-op in KOOS sport/ rec (mean difference 4.4, 95% CI 2.1 to 6.7) and KOOS QoL (mean difference 2.4, 95% CI 0.4 to 4.4). Women reported less improvement from 1 to 2 years post-op than men in KOOS sport/rec (mean difference 3.2, 95% CI 0.3 to 6.1). In some age groups, women reported a clinically relevant worse KOOS sport/rec score than men (mean difference C8). Conclusions: Women reported statistically significant worse patientreported outcomes than men before and at 1 and 2 years after ACL reconstruction, and in some age groups this difference was also clinically relevant. We suggest that possible gender differences be analyzed in future studies on evaluation after ACL injury/reconstruction.

P13-755 The arthrotic knee - contraindication for acl reconstruction? Long term FU of 19 salvage cases Harmer J.1, Patt T.1, Stellinga D.2, Buijtendijk J.2, van Heerwaarden R.1 1 Sint Maartenskliniek, Orthopaedic Department, Woerden, Netherlands, 2 SPOMED, Capelle a/d IJssel, Netherlands Objectives: The efficacy of acl reconstruction in patients with advanced cartilage damage in whom pain and giving way occur frequently with

S179 activities of daily living has rarely been described with long-term followup. The purpose of this study was to determine if in chronic acl deficient patients with advanced cartilage lesions (ICRS grade 3 and grade 4) the overall quality of life would improve and recreational activity levels would increase by arthroscopic acl reconstruction with autogenous quadruple hamstring graft. Methods: From our database of 874 patients who underwent acl reconstruction in the decade starting September 1995 we selected all patients who had chronic acl reconstruction and had grade 3 to 4 changes in at least 1 compartment at the time of surgery (N=19). 4 Patients (21,1%) had a failed acl reconstruction prior to our surgery. The cartilage condition of the articular surfaces was determined by direct arthroscopic visualization. All patients were operated on by one surgeon using single bundle quadruple hamstrings, endobutton-fixation on the femur and 2 staples on the tibia. A standardized rehabilitation program was used. These patients were prospectively followed and evaluated by clinical and instrumental assessment criteria pre-operativeley, at 6 months, 1 year and latest follow-up with a mean of 9 years after surgery (4-22 y). Clinical assessment was performed using the IKDC form. Subjective and functional parameters were assessed by the Lysholm Score and Tegner activity scale. Instrumental evaluation was done using the KT-1000. Radiographical examination including standard x-rays (ap/lat/Rosenberg view) were obtained. Results: At the latest FU the scores in IKDC differed from pre-operatively 89,5%D, 10,5% C to post-op 26,3% C and 73,7%B. All patients improved. The Lysholm Score was pre-op 43,8 (16-64) and post-op 91,9 (69-100). The Tegner Activity Scale had a pre-op mean of 1,3 (1-2), post-op a mean of 3,7 (2-6). The Manual Maximum Displacement test showed pre-op 5,3% nearly normal, 78,9% abnormal and 15,8% severely abnormal results with the KT-1000. Post-op measurements with the KT-1000 showed after 9 years 89,5% normal (IKDC A, \ 3mm) and 10,5% nearly normal stability. All patients improved subjectively and objectively from the salvage surgery. No recurrent instability was reported, no graft failure was diagnosed. Conclusions: In this study there is a good and stable salvage solution for patients with chronic acl insufficient knees even though pre-operatively arthrotic changes are present. With the salvage acl reconstruction, all patients had decreased episodes of giving way with daily and sports activities without aggravating the preexisting severe arthrosis.

P13-768 Prospective cohort study on functional evaluation of children 12 years or younger after ACL injury Moksnes H.1, Engebretsen L.2, Risberg M.A.1 1 NAR Orthopaedic Centre Oslo University Hospital, Ullevaal, Hjelp24 NIMI and OSTRC, Norwegian School of Sport Sciences, Oslo, Norway, 2 Orthopaedic Center Oslo University Hospital, Ullevaal, OSTRC, Norwegian School of Sport Sciences, Oslo, Norway Objectives: 1. To evaluate knee function in children the first year after acute anterior cruciate ligament (ACL) injury sustained at age 12 years or younger using functional questionnaires, single-leg hop tests and isokinetic strength measurements. 2. To develop and implement a primary active rehabilitation treatment algorithm for pre-adolescent children with open growth plates. Methods: Prospective cohort study started in the spring of 2006 which now include 33 children who have sustained an acute ACL rupture at age 12 years or younger. Inclusion criterion was acute ACL injury at age 12 years or younger. Exclusion criteria are concomitant fractures and bony avulsions of the tibial attachment of the ACL. The treatment algorithm was primary active rehabilitation, in which children were assessed for ACL reconstruction by one experienced orthopaedic surgeon if they experienced repeated give-way episodes, or sustained a secondary meniscus injury eligible for suture. The questionnaire KOOS was used to evaluate the patients’ knee function at the one year follow-up. Knee performance was measured

123

S180 using four single-leg hop tests (single hop (SH), triple-crossover hop (TCH), triple hop (TH) and six meter timed hop (6MTH)), while quadriceps and hamstrings muscle strength was evaluated using peak torque values from isokinetic strength test (Biodex 6000) with 5 repetitions at 60/sec. Results: Twenty-one boys and 12 girls are included. They were on average 11.1 years (range 8.2 - 12.9 years) at time of injury. The injury activities were; alpine skiing (18 knees/55%), soccer (8 knees/24%), cycling (2 knees/6%), play (2 knees/6%), team handball (1 knee/3%), ski-jumping (1 knee/3%) and trampoline (1 knee/3%). Twenty-four (73%) had gone through non-operative treatment and 18 subjects has been followed for 1 year. One arthroscopy has been performed to suture a medial meniscus. The KOOS subscales were on average (range); Pain 93 (83-100), Symptom 89 (71-100), ADL 99 (89-100), Sport/Rec 83 (55-100) and QoL 68 (31-88). Hop test results are given in average percentage (range) of uninjured knee; SH 99.0% (84-129), TCH 94.3% (84-104), TH 93.6% (82-107) and 6MTH 98.2% (82-110). Isokinetic muscle strength tests showed an average percentage peak torque compared to uninjured leg (range) of; quadriceps 92.2% (71110) and hamstrings 90.5% (69-110). Due to repetitive give way episodes or secondary meniscus injuries nine children have undergone ACL reconstruction using transepiphyseal drillholes and soft tissue hamstring graft with fixations proximal and distal to the femoral and tibial physes. They were on average 13.2 years (range 11.9-14.7 years) at time of surgery, which was performed on average 1.4 years (0.2-2.5 years) after injury. In total three medial and three lateral meniscus sutures and one partial lateral meniscus resection had been performed. No secondary surgeries have been performed. Five children have been through one year follow-up after surgery. The KOOS subscales were median (range); Pain 86 (78-97), Symptom 86 (68-89), ADL 99 (99-100), Sport/Rec 80 (60-90) and QoL 75 (56-81). Conclusions: A treatment algorithm consisting of primary active rehabilitation and delayed ACL reconstruction provides good knee function in children 12 years or younger after ACL injury.

P13-770 Single bundle anterior cruciate ligament reconstruction by two anteromedial portals Choi S.W.1, Kim M.-K.2, Bae J.H.2 1 Jeju University, Department of Orthopedic Surgery, Jeju, Korea, Republic of, 2Inha University, Department of Orthopedic Surgery, Incheon, Republic of Korea Objectives: This study is aimed to report the results of single bundle ACL reconstruction by two anteromedial portals and a standard anterolateral portal, which allows surgeons to decrease inclination of femoral tunnel with good surgical view. Methods: Traditional single-bundle transtibial reconstructions have placed grafts in a less anatomic location, more vertical orientation of femoral tunnel, relative to the true ACL insertion site, because the femoral tunnel is predetermined by the position of the tibial tunnel. To resolve the problem of this vertical orientation of femoral tunnel, femoral tunnel through the standard anteromadial portal was developed. But this technique has problems such as poor surgical view and breakage of posterior femoral wall. So we devised a new technique, which drills femoral tunnels through far anteromedial portal with a telescopy through standard anteromedial portal. This technique allows surgeons to decrease femoral tunnel inclination and short surgical time due to good surgical arthroscopic view. So we can make femoral tunnel at 10 or 2 O,Clock position with good visuality. Average follow-up period was 12 months. The control group (Group A) included 30 patients with ruptured ACL who were executed conventional transtibial ACL reconstruction. We evaluated the clinical

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 results of 30 patients with ruptured ACL who were executed ACL reconstruction by two anteromedial portals(Group B). We checked Lachman test, KT-2000 arthrometer and Pivot shift test both preoperatively and at the final follow up, and also compared the clinical results of both groups using IKDC and Lysholm score. We also checked surgical time of this technique compared with traditional transtibial technique. Results: At the final follow up, Group A showed 23 cases of negative Lachman test (77%), anterior translation of 2.85mm in KT-2000 arthrometer, 23 cases (77%) of negative Pivot shift test. In group B the results showed 24 cases (80%) of Lachman test and average of 2.72 mm of anterior translation in KT-2000 arthrometer, 27 cases (90%) of negative Pivot shift test. Group B showed statistically significant improvement of rotatory stability compared to the group A. Lysholm and IKDC showed better improvement with 90.1 point and 90% of nearly normal level in group B compared with the group A(85.6 point and 86% of nearly normal status), but there was no statically significant difference between the two groups. Simple X-ray (tunnel view) showed an average femoral tunnel inclination of 55 in group A whereas 31 in group B. Group B(40 minutes) spent shorter surgical time than group A(60 minutes). Conclusions: We believe that two anteromeidal portals method is an effective technique for single bundle ACL reconstruction, which can prevent residual rotatory stability with excellent clinical results by establishing the best anatomic position, And this technique does not require excessive notchplasty by decreasing the inclination of reconstructed ACL and reducing the surgical time with good surgical visuality.

P13-773 Femoral and bone tunnel widening after ACL reconstruction using EndoButtonTM or EndoButton Continuous LoopTM. A case control study Lind M.1, Webster K.2, Feller J.3 1 Aarhus University Hospital, Sports Trauma Clinic, Aarhus, Denmark, 2 La Trobe University Musculoskeletal Research Center, Faculty of Health Sciences, Bundoora, Australia, 3La Trobe University Musculoskeletal Research Center, Melbourne, Australia Objectives: To investigate the effect on femoral and tibial bone tunnel widening and clinical outcome of EndoButton CLTM used for femoral graft fixation during anterior cruciate ligament (ACL) reconstruction compared to EndoButtonTM with a polyester tape. Methods: A retrospective case-control study design was used. 120 patients with EndoButton CLTM femoral fixation were compared with 120 patients with EndoButtonTM with a doubled 3mm polyester tape. Tunnel widening was measured on anteroposterior (AP) and lateral radiographs at 12 months follow-up. The largest tunnel width was measured for the femoral tunnel and in the tibial tunnel above the interference screw. Clinical outcome was assessed by objective and subjective IKDC scores and KT-1000 knee laxity measurements. Results: Femoral tunnel widening in the EndoButtonTM group was 46.2% and 38.5% on the AP and lateral radiographs respectively and tibial tunnel widening was 24.9% and 33.2%. Femoral tunnel widening in the EndoButton CLTM group were 38.7% and 28.2%, and tibial tunnel widening was 10.9% and 23% for the AP and lateral radiographs respectively. Tunnel widening was lower for both femoral and tibial tunnels with the EndoButton CLTM (p\ 0.01). There were no differences between the groups for any of the clinical scores or KT-1000 knee laxity. Conclusions: Femoral ACL graft fixation with an EndoButtonTM and continuous polyester loop compared to an EndobuttonTM with a doubled 3mm polyester tape reduced the radiographic femoral and tibial tunnel widening at one year. The reduction in tunnel widening was not associated with differences in clinical outcome with respect to IKDC scores or KT1000 knee laxity measurement.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P13-774 Tibial bone tunnel widening is reduced by polylactate/hydroxyapatite interference screws compared to metallic screws after ACL reconstruction with hamstring grafts Lind M.1, Webster K.2, Feller J.3 1 Aarhus University Hospital, Sports Trauma Clinic, Aarhus, Denmark, 2 La Trobe University Musculoskeletal Research Center, Faculty of Health Sciences, Bundoora, Australia, 3La Trobe University Musculoskeletal Research Center, Melbourne, Australia Objectives: Composite interference screws containing calcium phosphate for anterior cruciate ligament graft fixation could improve implant/ bone integration and thereby reduced tunnel widening and graft slippage. The present study investigated the effect of a polylactate/hydroxyapatite interference screw (HA/PLLA) screw used for tibial graft fixation on tunnel widening and clinical outcomes compared with a metallic interference screw. Methods: One hundred patients with HA/PLLA screw tibial fixation were compared to 100 patients with metallic screw tibial fixation. Tibial tunnel widening was measured on AP and lateral radiographs taken at 12 months follow-up. Clinical outcome was assessed by objective and subjective international Knee Documentation Committee (IKDC) scores, Noyes Sports Activity and Occupational Rating scores and KT-1000 knee laxity measurements. Results: Tibial tunnel widening at the level of the metallic screw group was 36% and 38% on AP and lateral radiographs respectively. Tunnel widening was less in the HA/PLLA group with mean tunnel widening of 30% and 32% (p=0.012 and 0.018) on AP and lateral radiographs respectively. No differences were found for any of the clinical scores or for anterior knee laxity. Conclusions: The use of a polylactate/hydroxyapatite interference screw resulted in less tibial tunnel widening than a metallic screw but did not affect clinical outcome or knee laxity.

S181 Results: Mean knee flexion angle was 23 (range, 11 to 30) at initial contact (IC) and had increased by 24 (95% CI, 19 to 29, p\0.001) 40 ms later. Valgus angle was neutral, 0 (range, -2 to 3) at IC, but had increased by 12 (95% CI, 10 to 13, p\0.001) 40 ms later. The knee was externally rotated 5 (range, -5 to 12) at IC, but abruptly rotated internally by 8 (95% CI, 2 to 14, p=0.017) during the first 40 ms. From 40 ms to 300 ms after IC, however, we observed an external rotation of 17 (95% CI, 13 to 22, p\0.001).

Time sequences of knee joint kinematics P13-775 Reconstruction of the mechanism for ACL injury among female basketball and team handball players from videotapes using a model-based image matching technique Koga H.1, Nakamae A.1, Shima Y.1, Iwasa J.1, Engebretsen L.1, Bahr R.1, Krosshaug T.1 1 Norwegian School of Sport Sciences, Oslo Sports Trauma Research Center, Oslo, Norway Objectives: The mechanisms for non-contact ACL injury is a matter of controversy and several theories have been proposed, principally the quadriceps drawer hypothesis (the quadriceps muscle generates anterior shear forces on the tibia due to the patellar tendon angle), internal rotation (internal tibial rotation on a relatively straight leg), valgus in combination with external rotation (knee valgus collapse with external rotation, which might involve impingement of the ACL against the intercondylar notch) and tibiofemoral compression (tibiofemoral compression loading displaces the femur posteriorly relative to the tibia). A precise description of the injury mechanism is critical to be able to target intervention programs to prevent ACL injuries. Video analysis of injury tapes is the only method available to extract biomechanical information on the mechanism. The objective of this study was to describe knee joint kinematics in actual ACL injury situations using a model-based image matching technique (MBIM) we have developed. Methods: Ten video sequences with at least two views of ACL injuries from women0 s handball (n=7) and basketball (n=3) were analyzed using MBIM based on the animation program Poser. Computer models of the background were modeled and matched to the video image. A skeleton model with 57 degrees of freedom was adapted individually to each of the injured players and matched frame by frame, providing an estimate of the time course of knee joint kinematics and ground reaction forces for the injury sequence.

Peak vertical ground reaction force was 3.2 times body weight (95% CI, 2.7 to 3.7), and occurred at 40 ms (range, 0 to 83) after IC. Conclusions: Based on when the sudden changes in joint angular motion and the peak vertical ground reaction force occurred, we assumed that the ACL injury occurred about 40 ms after IC. Knee kinematics was remarkably consistent for the ten injury situations analyzed. An average flexion angle of 23 suggests that a quadriceps drawer mechanism may contribute to ACL injury. However, valgus loading also seems to be an important factor, because all players had immediate valgus motion within 40 ms after IC. Moreover, the knee rotated internally during the first 40 ms, and then an external rotation was observed, which seemed to have occurred after the ACL was torn. These results suggest that valgus loading in combination with internal rotatory motion seems to be the main mechanism of the ACL injury. Prevention program should focus on acquiring a good cutting and landing technique with knee flexion, without valgus of the knee or internal rotation of the foot.

P13-778 Second-look and MRI evaluation of anatomically augmented ACL reconstructions Shibuya H.1, Deie M.1, Adachi N.1, Nishimori M.1, Nakamae A.1, Nakasa T.1, Ochi M.1 1 Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan Objectives: We evaluated the second-look examination, postoperative magnetic resonance imaging (MRI) and clinical results of 24 patients who had undergone our anterior cruciate ligament (ACL) augmentation procedure using an autogenous semitendinosus tendon and EndoButton CL (Smith & Nephew, Andover, MA).

123

S182 Methods: Between April 2002 and March 2008, 36 patients underwent ACL augmentation at our hospital. We followed up 24 patients for more than 1.5 years after their ACL augmentations. Of these, 10 received anteromedial (AM) bundle reconstructions and 12 underwent posterolateral (PL) bundle reconstructions, while 2 underwent a single bundle augmentation procedure. The subjects were assessed by second-look arthroscopic examination at more than 18 months after their ACL augmentations, and were also assessed using the KT-2000 knee arthrometer (MEDmetric, San Diego, CA) at the 30 lb by joint position sense, as well as MRI 3, 6 and 12 months postoperatively. Results: The arthroscopic findings included a poor synovial covering and poor union at the attachment of the femoral bone tunnel in 3 cases with AM bundle reconstructions. On the other hand, the 10 PL bundle reconstructions were confirmed to show integration with the ACL remnant. We found that of the 2 cases receiving PL bundle reconstructions, one each had AM and PL bundles. There were no exposed bone tunnels in PL bundle reconstructions. On postoperative MRI examinations, 6 cases with AM bundle augmented ACL and 7 with PL bundle augmented ACL had appearances resembling those of a single bundle on sagittal planes of the first postoperative MRI. The mean side-to-side difference in anterior displacement measured by the KT-2000 knee arthrometer at 30 of knee flexion preoperatively was 1.52 mm for the AM bundle reconstruction cases, 1.88 mm for the PL bundle reconstruction cases, and 1.40 mm for the single bundle augmentation cases. At first postoperative examination, the mean was 1.15 mm for AM bundle reconstruction and 0.8 mm for single bundle augmentation, but the value improved significantly to a mean of 0.19 mm for PL bundle reconstruction. Conclusions: PL bundle reconstruction showed improved joint stability, postoperatively, as compared to AM bundle reconstruction. This procedure is a treatment option for patients whose ACL remnants are left intact under certain conditions.

P13-782 Anatomy of normal human anterior cruciate ligament attachments evaluated by divided small bundles Hara K.1, Mochizuki T.2, Yamaguchi K.3, Sekiya I.2, Kuroda H.1, Akita K.3, Muneta T.4 1 Kameda Medical Center, Orthopedic Surgery, Kamogawa, Japan, 2Tokyo Medical and Dental University Hospital, Section of Cartilage Regeneration, Graduate School, Tokyo, Japan, 3Tokyo Medical and Dental University Hospital, Section of Clinical Anatomy, Tokyo, Japan, 4 Tokyo Medical and Dental University Hospital, Orthopedic Surgery, Tokyo, Japan Objectives: Detailed knowledge of the footprint of the anterior cruciate ligament (ACL) is essential for anatomical and functional reconstruction. The purpose of this study was to examine the morphology and attachments of ACL on the femur and tibia by separating the ACL into about twenty small bundles. Methods: His study used 20 cadaveric knees. The ACL was divided into anteromedial and posterolateral bundles, then separated into 10 small bundles of 2-mm diameters, with preservation of their attachment sites marked with color markers. The positional relationship between the femoral and tibial attachments of each small bundle was investigated. Results: A layered positional correlation of small bundles was found between the tibial and femoral attachments. Small bundles aligned in the anterior-posterior direction in the tibia corresponded to the bundles aligned in a high-low direction in the femur in flexion. The femoral attachment pattern was relatively similar in each specimen; however, the tibial attachment showed 2 patterns: an oblique type (12 of 20) and a transverse type (8 of 20). The posterior portion of the posterolateral bundle was separately attached to the medial and lateral portions of the tibial attachment. There was no fibrous insertion in the center of the posterior portion of the ACL tibial attachment in any specimen. In this bare area, there was fat tissue and vascular bundles. Conclusions: Small bundles constituting the ACL showed a relatively layered arrangement between 2 attachments. The tibial attachment showed

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 2 patterns of oblique and transverse types, and the vascular bundles were located in the center of the posterolateral bundle.

P13-787 Low-intensity ultrasound enhances tendon graft- bone interface healing in anterior cruciate ligament reconstruction. A biochemical and image analysis in human Papatheodorou L.1, Malizos K.1, Hantes M.1, Grafanaki K.2, Karantanas A.3, Stathopoulos C.2 1 University of Thessaly, Dept. of Orthopaedics, Larissa, Greece, 2 University of Patra, Depts. of Biochemistry, Patra, Greece, 3University Hospital of Heraklion, Radiology, Heraklion, Greece Objectives: The present study investigates the effect of low-intensity pulsed ultrasound (LiUS) during ligamentization process after anterior cruciate ligament reconstruction (ACL) through biochemical and imaging analysis. Methods: Sixty patients with arthroscopically assisted ACL reconstruction using semitendinosus and gracillis tendon autograft were studied. LiUS (200-lsec bursts of 1,5 LHz sine waves with pulse repetition rates of 1 KHz and average intensity of 30 mW/cm2) was applied for 20 days in 30 patients (study group), while 30 patients did not receive LiUS (control group). Blood samples were collected pre-operatively and 1, 2, 3 and 6 weeks post-operatively. Serum levels of TGF-b1, IGF, OPG, sRANKL, procollagen I and NTx from both groups were measured using ELISA. Multiple Direction CT (MDCT) was used to monitor the progress of grafthost bone healing in both groups postoperatively. LDCT with MPR in 3 planes evaluated the direct integration by means of quantitative measurement of HU and qualitative evaluation of the degree of the cross sectional ossification. Results: Analysis of the serum levels of all the markers showed statistically significant alterations in the study group compared to the control group. Interestingly, IGF and OPG levels were found elevated, sRANKL was decreased and TGF-b1 exhibited a bimodal profile in the study group. Imaging analysis revealed a more efficient ligamentization process after ultrasound treatment indicated by an earlier initiation of healing. Conclusions: Our results suggest that LiUS enhances the healing rate of the tendon graft-bone interface in ACL reconstruction, possibly by affecting important biochemical pathways.

P13-796 Does the lateral intercondylar ridge disappear in ACL deficient patients? van Eck C.F.1, Morse K.1, Lesniak B.1, Kropf E.1, Tranovich M.1, van Dijk C.N.2, Fu F.1 1 University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States, 2 Academic Medical Center, Orthopaedic Surgery, Amsterdam, Netherlands Objectives: It is suspected that the lateral intercondylar ridge exists because of osseous remodeling in response to stress from the ligament fibers, in accordance with Wolff ‘s law. This might be why the ridge is more clearly defined in young active people, because they put more strain on their ACL. If this is true, when there is no ACL (i.e. after ACL rupture), the ridge could possibly gradually disappear. It has been suggested by surgeons that in chronic ACL cases the ridge is less visible or not visible at all. However, to our knowledge this has never been proven in literature. The aim of this study was to determine if there is a difference in the presence of the lateral intercondylar ridge and the lateral bifurcate ridge between patients with sub-acute and chronic ACL injuries. We hypothesized that the ridges would be present less often with chronic ACL deficiency. Methods: Twenty-five patients with a chronic ACL injury were matched for age and gender to 25 patients with a sub-acute ACL injury. On the arthroscopic videos of the patients’ ACL reconstruction, the lateral intercondylar ridge and lateral bifurcate ridge were scored as either present, absent, or indeterminate due to insufficient visualization by three

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 blinded observers. The observers were allowed to view the entire arthroscopic procedure. The surgeon who performed the ACL reconstruction procedure did not necessarily aim to visualize the ridges during the procedure. Results: The kappa for the three observers was .61 for the lateral intercondylar ridge and .58 for the lateral bifurcate ridge. The lateral intercondylar ridge was present in 88% of the sub-acute cases and 88% of the chronic cases. The lateral bifurcate ridge was present in 48% of the sub-acute and 48% of the chronic cases. Figure 1 shows an example of a case with both a lateral intercondylar ridge and a lateral bifurcate ridge.

Conclusions: To definitively conclude that there is no difference in the presence of the ridges between sub-acute and chronic ACL injuries, more than 1000 subjects would be needed. We believe that such a small difference would probably not be clinically significant. The authors would suggest looking for the ridges as a landmark of the native ACL insertion site during ACL reconstruction in both acute and chronic ACL injuries.

P13-799 Augmentation technique with hamstring tendons in chronic partial lesions of the ACL: clinical and arthrometric analysis Vannini F.1, Ghermandi R.1, Ruffilli A.1, Cavallo M.1, Parma A.1, Giannini S.1 1 Istituto Ortopedico Rizzoli, Bologna University, VI Department of Orthopaedics and Traumatology, Bologna, Italy Objectives: ACL (anterior cruciate ligament) partial tears include various types of lesions, and an high rate of these lesions evolve into complete tears. Most of the techniques described in literature for the surgical treatment of chronic partial ACL tears, don’t spare the intact portion of the ligament. Aim of this study was to perform a prospective analysis of the results obtained by augmentation surgery using gracilis and semitendinosus tendons to treat partial sub-acute lesions of the ACL. Methods: The study included 89 patients treated consecutively at our Institute from 1993 to 2003 with a mean injury-surgery interval of 21 weeks (12 - 39). All the patients underwent the same surgical procedure: technique requires an ‘‘over the top’’ femoral passage, which enables salvage and strengthening of the intact bundle of ACL. Patients were followed up by clinical and instrumental assessment criteria at 3 months, 1 year and 5 years after surgery. Clinical assessment was performed with the IKDC form. Subjective and functional parameters were assessed by the Tegner activity scale. Instrumental evaluation was done using the KT-2000 instrument: the 30 pound passive test and the manual maximum displacement test were performed. Results: We had no complications, such as DVT or postoperative infection. At the final follow-up no case of re-rupture was observed and we didn’t observed recurrences in ligamentous laxity. We obtained good to excellent results in 96.6% of cases. In particular, following the IKDC score 61 patients were classified as A, 26 as B and 2 as C. According to the Tegner Activity Scale the mean pre-lesional sports activity level was 6.9, whereas it was 6.5 1 year after surgery and 6.2 at the final follow up. The side-to-side difference in anterior tibial displacement with KT-2000 was \ 5 mm in all patients at the final follow up.

S183 All patients regained full joint movement in extension, whereas there was a reduction in flexion of less than 5 in five patients. None of the patients gave up with sports. There is not a statistically significant relationship between the age of patients and the time needed for return to contact sports participation. Conclusions: In conclusion the described technique has the advantage of being little invasive, compatible with the ACL anatomy and provides good results thanks to the maintenance of the intact bundle, with its mechanical support and vessel and nerve supply, representing an ideal solution for the treatment of chronic partial ACL tears in young and active patients. Moreover this technique enables very rapid functional recovery and return to sport.

P13-802 Primary repair combined with bone marrow stimulation in acute ACL lesions: results in athletes, 3 years follow-up Gobbi A.1 1 Orthopaedic Arthroscopic Surgery International, Dept. of Sport and Medicine, Milano, Italy Objectives: The purpose of this study was to evaluate and followup the outcome of ACL primary repair combined with bone marrow stimulation in the treatment of acute ACL lesions in an athletic population with a 3 year follow up\Hypothesis:[ Primary ACL repair combined with bone marrow stimulation could restore stability and function in athletes with acute ACL incomplete tear. Methods: Among a group of 120 patients clinically diagnosed for ACL acute lesion, 35 athletes with arthroscopically confirmed incomplete ACL proximal tear were treated with primary repair combined with bone marrow stimulation of the ACL at the femoral attachment site. Post-operatively, all patients underwent a specific rehabilitation program. All patients were prospectively evaluated and outcome measures were assessed using Marx, Noyes, Tegner, SANE, Lysholm and IKDC scores. Anterior tibial translation was measured using Rolimeter instrument under anesthesia at 6 - 12 months and at final follow-up. Results: All these athletes were followed-up for an average of 36 months. Mean age was 26.6 years. Mean pre-op Tegner was 7.1 (SD=1.1) and final Tegner was 6.5 (SD=1.4); mean pre-op Marx was 11.0 (SD=3.4) and final Marx was 9.6 (SD=3.1), and mean pre-op Noyes was 82.5 (SD=5.8) and final Noyes was 83.3 (SD=7.2). These scores were not statistically different at P values of 0.020, 0.011 and 0.303 respectively. However, final SANE rating was significantly lower than pre-injury SANE rating. This was mainly related to decreased self confidence in high risk sport and fear of a new injury. Final SANE rating and Tegner scores were significantly higher than their respective pre-operative values. Mean Rolimeter side to side difference of anterior knee translation was significantly reduced from 3.5 mm (SD = 0.7) preoperatively to 1.6 mm (SD = 0.8) post-operatively. Two female patients reruptured their ACL after more than 1 year because of a new injury and underwent standard ACL reconstruction. Conclusions: In our athletic population, ACL primary repair in acute incomplete lesion combined Keywords: ACL; partial tear; primary repair; microfracture; bone marrow stimulation

P13-803 Arthroscopic treatment of combined tibial plateau and intercondilar eminence avulsion fractures Vannini F.1, Di Caprio F.1, Ghermandi R.1, Cavaciocchi M.1, Timoncini A.1, Giannini S.1 1 Istituto Ortopedico Rizzoli, Bologna University, VI Department of Orthopaedics and Traumatology, Bologna, Italy Objectives: The purpose of this study is to evaluate the functional and anatomical outcome of combined arthroscopic treatment of Tibial Plateau Fractures (TPF) and Intercondilar Eminence Avulsion Fractures (IEAF).

123

S184 Methods: From January 2000 to September 2004 we treated 26 patients (12 males, 14 females) mean age 39.0 (min 17-max 70), affected by TPF (Schatzker type I-IV) associated to IEAF (Meyers and McKeever type II-III). Under arthroscopic control, surgical treatment was performed by reduction and percutaneous screw fixation for TPF and by reduction and fixation using a Caspari punch with non absorbable suture for IEAF. All patients were evaluated before surgery and followed up by IKDC form and X-rays. In selected cases CT scan was performed preoperatively. At the final follow-up the stability of the knee was evaluated performing KT-1000 side to side evaluation. Results: No complication were observed after surgery. All patients regained full extension. At a minimum 5 years follow-up (min. 60 - max 117 months), the mean functional score with IKDC Subjective Evaluation was 85.2 Performing the IKDC Objective Evaluation patients were classified as follow: 13 as A, 10 as B, 2 as C and 1 as D. None of the patients presented anterior laxity at the final follow up. The KT-1000 side to side difference was 1.06 mm. Conclusions: Combined arthroscopic treatment of TPF and IEAF provides satisfactory results. The presented technique permits to obtain a stable fixation of the fracture and good stability of the knee at the long term follow up and is easy to be performed.

P13-814 Tibial tunnel widening after bioresobable poly-lactide calcium carbonate interference screw usage in ACL reconstruction Foldager C.B.1, Jakobsen B.W.2, Lund B.3, Christiansen S.E.4, Kashi L.5, Mikkelsen L.R.5, Lind M.6 1 ˚ rhus C, Denmark, 2Aarhus Aarhus, Orthopaedic Research Lab, A University Hospital, Division of Sports Trauma, Beder, Denmark, 3 Division of Sports Trauma, Department of Orthopedic Surgery, Aarhus, Denmark, 4University Hospital in Aarhus, Sports Trauma, Orthopedic Dept, Aarhus, Denmark, 5Aarhus University Hospital, Department of ˚ rhus C, Denmark, 6Aarhus University Hospital, Sports Radiology, A Trauma Clinic, Aarhus, Denmark Objectives: Developing bio-absorbable interference screws for anterior cruciate ligament (ACL) reconstruction that is replaced by bone has proven to be a challenging task. Bio-absorbable poly-lactide carbonate (PLC) interference screw (Calaxo Screw, Smith & Nephew Endoscopy, Andover, MA) consists of 65% of the co-polymer poly-D(85%),L(15%)-lactide-coglycolide (PLGA) and 35% calcium carbonate. Previously, the use of this PLC screw in an ovine model resulted in replacement of the screw with bone within one year after soft tissue ACL reconstruction. The aim of this study was to investigate the osteogenetic response of this poly-lactide carbonate (PLC) interference screws in ACL reconstruction in humans. Methods: Ten patients (median age 28 years) underwent arthroscopic ACL reconstruction with semitendinosus-/gracilis tendon graft and a PLC interference screw. The study protocol was approved by the local ethical committee. They received ACL reconstruction with a bio-absorbable PLC interference screw as tibial fixation (Calaxo screw, Smith & Nephew Endoscopy, Andover, MA, USA). All patients followed the same accelerated rehabilitation protocol. The rehabilitation allowed full weightbearing from first postoperative day and focused on early restoration of full extension and quadriceps function. The patients were scanned with a multi-slice CT scanner two weeks and one year postoperatively. The results of the CT scanning were not known to the doctors in the outpatient clinic at any of the consultations in the follow-up period. Tunnel widening and bone formation was assessed on reconstruction CT-images. Results: All patient were available for CT scanning at both follow-up times. One year postoperative the screw was only visible in two out of ten cases. No unexpected granulomas or cysts were observed. Fourteen days postoperative a mean tunnel area widening of 78% [52%;110%] was observed, which was a significant increase (P\0.01). At the one year follow-up the mean tunnel area widening was 128% [84%;180%] which was significantly higher than at baseline (P\0.001) and fourteen days postoperatively (P\0.001). Conclusions: Factors such as accelerated rehabilitation, micro-motions, and early screw degradation might be responsible for this large tunnel widening. The screw was withdrawn from the market by the company due to a report of a sterile tibial cyst and intra-articular granuloma formation after ACL reconstruction using the PLC interference screws after the inclusion of

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 patients in this study had ended. Our results demonstrate the difficulty in translation of preclinical data. This study illustrates the need for extensive preclinical investigation of new materials for clinical purposes.

P13-816 Transtibial ACL reconstruction technique fails to position drill tunnels anatomically Kopf S.1, Forsythe B.1, Wong A.K.1, Martins C.A.Q.1, Tashman S.2, Anderst W.2, Irrgang J.3, Fu F.1 1 University of Pittsburgh, Dept. of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States, 2University of Pittsburgh Medical Center, Orthopaedic Biodynamics Laboratory, Pittsburgh, United States, 3 University of Pittsburgh, School of Medicine, Pittsburgh, United States Objectives: Traditional transtibial tunnel drilling is the most widely used technique for ACL reconstruction among orthopaedic surgeons. However, there is concern that transtibial drilling may result in non-anatomically positioned bone tunnels. Traditional methods for evaluating tunnel position on standard two-dimensional radiographs have limited utility. Thus, the purpose of this study was to determine whether transtibial drilling results in anatomically positioned bone tunnels using 3D CT models. Methods: 58 knee CT scans were obtained from patients who underwent transtibial single-bundle ACL reconstruction performed by 27 different surgeons; thus resulting in more representative sample of ACL tunnel positions than would be obtained from a single surgeon. Additionally, eight cadaveric knee CT scans were obtained after anatomic double-bundle ACL tunnel drilling to establish a control group with anatomically positioned tunnels. 3D CT models were reconstructed and aligned to an anatomical coordinate system (Fig. 1). Femoral tunnel aperture centers were measured in the anatomic posterior-to-anterior and proximal-to-distal directions on the medial wall of the lateral condyle. Tibial tunnel aperture centers were measured in the anterior-to-posterior and medial-to-lateral directions on the tibial plateau.

Drill tunnel position of anatomic DB ACL-R Results: The mean results are plotted on Figure 2. On the femur, the transtibial tunnel centers were 34.4% more anterior (higher) than the anatomic AM centers and 42.2% more anterior than the anatomic PL tunnel centers relative to the posterior-to-anterior notch height (p\0.001 and p\0.001, respectively). Relative to the proximal-to-distal notch depth, the transtibial tunnel centers were 12.9% more distal than the anatomic AM tunnels (p=0.008) and 17.0% more proximal than the anatomic PL tunnels (p\0.001). On the tibial side, the transtibial tunnel centers (48.6±5.4%) were within the anatomic PL tunnel range

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 (46.4±3.7%) relative to the anterior-to-posterior plateau depth (p=1.0); and relative to the medial-to-lateral plateau width, the transtibial tunnels (47.9±3.1%) were within the anatomic AM tunnel range (50.5±4.2%) (p=0.55).

Position of bone tunnel centers Conclusions: The transtibial ACL reconstruction technique has the potential to result in non-anatomically positioned bone tunnels (Figure 3), which might be responsible for abnormal kinematics 1 and thus, for early onset of osteoarthritis.2

Typical transtibial SB tunnel position. References: 1) Tashman S et al.. 2004. Am J Sports Med 32: 975-83. 2) Chaudhari AM, et al.. 2008. Med Sci Sports Exerc 40: 215-22.

P13-820 ‘‘Healing response’’ technique in surgery of ACL lesions Remizov V.1, Erhan N.1, Zlatin P.1 1 ULIM, Medicine, Chisinau, Republic of Moldova

S185 Objectives: The ‘‘healing response’’ technique (HRT) is alternative to formal ACL - reconstruction. HRT was elaborated for acute incomplete (1/ 3) proximal ACL injury, especially in patients over 30 and as procedure added to primary repair - fixation the torn portion to residual bundle. Aim: It is the attemption to use HRT for chronic incomplete (1/3) proximal ACL injury (ACL) accompanied anterior instability (AI) or anteriormedial instability (AMI). Methods: From 1999 to 2007 154 patients (male 35 - 22,4% and 119 - 77,6 % female, of average age 28 - from 15 to 49) with 1/3 proximal ACL injuries. ACL?AI had 98 patients and ACL?AMI had 56 patients. Patients with ACL?AI (group I) were treated by HRT (n=98). Patients with ACL?AMI (group II) were treated as well by HRT only (n=24). Patients with ACL?AMI (group III) were treated by HRT ? active tibial tenodesis -Slocum-Larson procedure (n=32). Postoperatively patients used 4 weeks controlled motion brace (1758-1558) and full weight -bearing. The clinical, radiological, USG, MRI and arthroscopical methods were used. The patients (n=136) were evaluated after 12 month using 2000 IKDC score. Results: Sixy one patients of group 1 had grade A, 17 - B, 2 -C. Six patients of group II had grade A, 14 - B, 4 - D. Twenty four patients of group III had grade A, 6 - B, 2-C. Four patients with grade C (from group I and III) and 4 patients with grade D (from group II) went to formal ACL reconstruction. Conclusions: HRT is useful for incomplete (1/3) proximal ACL injury accompanied AI. Partial (1/3) proximal ACL injury with AMI may be treat by HRT and active tibial tenodesis augmentation.

P13-822 Three-dimensional morphologic measurement of functional axis in distal femur and its relation to ACL injury risk in women: a 3D CT study Hoshino Y.1, Wang J.1, Lorenz S.1, Tashman S.2, Fu F.1 1 University of Pittsburgh, Dept. of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States, 2University of Pittsburgh Medical Center, Orthopaedic Biodynamics Laboratory, Pittsburgh, United States Objectives: A common choice for the femoral mechanical axis is the transcondylar axis, defined as the line connecting centers of both posterior femoral condyles. The anatomical location of this axis relative to the femoral shaft has not been reported, even though it could have a significant influence on knee kinematics. Gender diversity of distal femur morphology has been implicated in ACL injury risk, though typically without considering kinematic implications. Gender differences in the mechanical axis location might contribute to the increased risk of ACL injury in women. The primary goal of this study was to develop a unique parameter that could reliably describe the three-dimensional mechanical axis location in the distal femur. We hypothesized that this parameter would be different between males and females, and between subjects who have had ACL injuries and those who have not. Methods: 3D CT data of both femurs in 30 patients (36 ± 17y.o., 17 men:13 women) with unilateral soft tissue injury (19 ACL, 6 PCL, 5 meniscus tear) were selected from ongoing IRB-approved projects. CT scans were reconstructed into 3D bone models and analyzed using 3D imaging software (Geomagic Studio 10). First, the bony axis was aligned onto the center of femoral shaft. Spheres were fitted to both posterior condyles, and the intercondylar axis was defined as the line connecting their centers. The condyle offset was calculated as the perpendicular distance between the intercondylar axis and bony axis. The condyle offset ratio (COR) was calculated by dividing the condyle offset by the average radius of the two condyle spheres. Measurement reliability was evaluated by three-time and three different tester repeatability using the Intra-class Correlation Coefficient (ICC). Correlations between the COR and body height were also determined. Side-to-side and gender differences of the COR were assessed. COR of uninjured knees was also compared between ACL-injured subjects (N=19) and those with other knee injuries (N=11). Student t-tests were used for statistical analyses (p\0.05). Results: The Condyle Offset Ratio was similar (n.s.; p=0.19) between the affected knees (0.98 ± 0.17) and the contralateral healthy knees (1.03 ±

123

S186 0.15). Inter-tester ICC was 0.82, and intra-tester ICC was 0.96. There was no significant correlation between COR and body height (R2=0.10, p\0.01). A significantly larger COR was found in women, 1.12 ± 0.16, than men, 0.96 ± 0.09 (p\0.01). In women, the ACL injured group had significantly larger COR, 1.19 ± 0.12, than the group without ACL injury, 1.03 ± 0.16 (p\0.05). There was no significant effect of ACL injury status on COR in men (injured 0.94 ± 0.10; non-injured 0.99 ± 0.06; p=0.16). Conclusions: The Condyle Offset Ratio, evaluated from 3D CT, provides a reliable measure of distal femur geometry that is consistent within subjects and independent of body size. A larger offset of the transcondylar axis was observed in women than in men, and in females the ACL-injured group showed larger COR than the group without ACL injury. The transcondylar axis location affects how the ‘‘lever arm’’ of the knee changes with flexion, and would most likely influence the pattern of ACL elongation with knee flexion and rotation. Thus, the COR may be a useful morphological feature which might be related to ACL injury risk. Further study is warranted to investigate how this parameter might affect joint mechanics and ligament loading.

P13-823 A new device for graft fixation in DB and SB ACL reconstruction: primary stability of the shim technique Lenschow S.1, Herbort M.1, Raschke M.1, Strobel M.J.2, Petersen W.3, Zantop T.1 1 Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 2Orthopa¨dische Gemeinschaftspraxis, Straubing, Germany, 3Martin-Luther-Krankenhaus, Unfallchirurgie, Berlin, Germany Objectives: ACL reconstruction using autologous tendon graft is the gold standard for the treatment of ACL injuries in young active patients. Cortical fixation as well as interference screws are associated with various problems. For this reason a new implant was developed blocking the graft in the tunnel with a wedge-shaped implant (‘‘shim technique’’). Our hypothesis was that this fixation provides high fixationstrength comparable to interference screw fixation. Methods: Porcine knees and four stranded human hamstring grafts were used for this study. 6, 7, 8 or 9mm tunnels were drilled. In group 1 the grafts were only fixed with the ‘‘shim technique’’ (Karl Storz, Tuttlingen, Germany). In group 2 the ‘‘Shim technique’’ was combined with a FlipTack (Karl Storz, Tuttlingen, Germany) and a double looped 1mm suture (Ethibond, Ethicon). An interference screw (Megafix, Karl Storz, Tuttlingen) served as a control group (group 3). The specimens were mounted to a testing machine (LR5K-plus, Lloyd instruments, Great Britain). The proximal end of the graft was clamped in a custom-made freezing clamp. The load was applied in line with the bone tunnel. A preload of 5 N was applied before the grafts were preconditioned between 0 and 20 for 10 cycles. Maximum load, yield load and stiffness as well as failure mode were recorded. Statistic analysis was performed using SPSS version 11.0 (SPSS Chicago, IL, USA). Results: The mean maximum load at failure in group 1 was between 253.7 N for the 6 mm shims and 446.0 N for the 9mm shim. The difference between 9 and 6mm as well as 9 and 7mm was significant (p\0.05). Hybrid fixation showed a mean maximum load of 522.5 N and 655.5 N for 6 and 9mm shim. Interference screw fixation showed no differences to the shim fixation in the 6 and 7 mm group (p[0.05). In the 8 and 9mm group ifs fixation was significantly higher (p\0.05). The stiffness varied in the groups where the grafts were fixed with the ‘‘Shim’’ alone between 28.1 (± 8.4) N/mm in the 6mm group and 64.4 (±17.2) N/mm in the 9mm group. This difference was significant (p\0.05). No differences were found between shim and hybrid fixation or shim and interference screw fixation (p[0.05). Three types of failure mode could be observed. In the 6 and 7mm shim the grafts failed by slippage of the graft passed the implant. In 8 and 9mm group, graft failed by pullout of the shim out of the tunnel with the graft.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 In the interference screw fixation group all graft failed by slippage past the screw. Conclusions: The aim of this study was to analyze the structural properties of a new implant in ACL reconstruction and to compare it to the fixation with a biodegradable interference screw. The data support our initial hypothesis that the shim technique fixation provides high initial aperture fixation of the graft comparable to interference screw fixation. The results of the present study show that the maximum load to failure reached by using the shim technique alone as well as the hybrid fixation seems to be sufficient to prevent a failure of the graft under the loads occurring during rehabilitation. In the 6mm and 7mm ‘‘Shim’’ group however hybrid fixation using a cortical button may be useful to prevent slippage of the graft and improve stability. A few limitations apply to this study since we tested a worst case scenario with the force of pullout in line to the bone tunnel. Our results suggest that the shim technique is an alternative to interference screw fixation concerning initial fixation strength.

P13-829 Relationship between distal femur morphology and in vivo kinematics Hoshino Y.1, Wang J.1, Lorenz S.1, Fu F.1, Tashman S.2 1 University of Pittsburgh, Dept. of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States, 2University of Pittsburgh Medical Center, Orthopaedic Biodynamics Laboratory, Pittsburgh, Pennsylvania, United States Objectives: Femoral bony morphology clearly influences knee kinematics. Previous morphological studies have focused on variability of femur size and shape (primarily for TKA implant design), and have not directly evaluated relationships between morphology and knee kinematics. The purposes of this study were: 1) Define a novel morphological parameter of the distal femur; and 2) Investigate relationships between this parameter and in vivo knee kinematics. We hypothesized that longer anteroposterior (AP) dimension of the distal femur should be related to increased tibial anterior translation during a functional activity. Methods: 1) 3D measurement of distal femur morphology. 3D CT data of both femurs in 30 patients (36 ± 17y.o., 17 men:13 women) with unilateral soft tissue injury (19 ACL, 6 PCL, 5 meniscus tear) were reconstructed into 3D bone models using Geomagic Studio 10. First, the bony axis was drawn as the center line of femoral shaft. The intercondylar axis was then defined as the line connecting the centers of the spheres, which were best fit to the posterior condyle. The condyle offset was calculated as the perpendicular distance between the intercondylar axis and bony axis. Finally, the condyle offset ratio (COR) was calculated by dividing the condyle offset by the average radius of the two condyle spheres. Mean, range and side-to-side difference of the COR were assessed, and variation among subjects was compared to side-to-side difference using one-way ANOVA. Measurement repeatability was evaluated using Intraclass Correlation Coefficient (ICC). 2) in vivo kinematics measurement. Eight contralateral healthy knees in soft tissue injured patients (47 ± 18y.o., 3men:5women, 5ACL; 3medial meniscus root tear) were tested. Biplane radiographic images of the knees were obtained at 100 Hz while the patients walked on a treadmill. A model-based tracking technique was used to align 3D CT bone models to the radiographic image pairs, providing 6-DOF knee kinematics. Anterior tibial translation and knee flexion were determined during the period from late swing through early stance (when the knee was near full extension) and used for comparison with the condyle offset ratio. Results: 1) 3D measurement of morphology. The mean condyle offset ratio was 1.00 ± 0.15 (range 0.72-1.26) in left femurs, and 1.01 ± 0.16 (range 0.76-1.29) in right femurs. Variation among subjects was significantly larger than side-to-side difference, mean 0.09 ± 0.08, (p\0.01). Intertester ICC was 0.82, and intra-tester ICC was 0.96. 2) in vivo kinematics measurement. During walking, the tibia translated anteriorly during late swing phase, by 2.8 ± 1.5 (range 1.1-5.3) mm.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Knees with larger condyle offset ratio demonstrated longer translation (R2=0.59, p\0.05). Conclusions: The condyle offset ratio is a morphological parameter that can be measured consistently and varies among subjects. Knees with larger condyle offset ratios tend to have greater anterior tibial translation just before heel strike, when the knee is near full extension. This suggests a possible relationship between condyle offset ratio and ACL injury risk, since greater anterior translation at the instant of foot contact might create larger strains in the ACL. Further studies are warranted to determine the role of this new morphological parameter for affecting knee kinematics and ligament injury risk.

P13-839 Accuracy of femoral and tibial tunnel placement in anatomic double bundle ACL reconstruction using double bundle ACL guides: evaluation using 3-d CT scan Lenschow S.1, Zantop T.2, Herbort M.3, Raschke M.1, Petersen W.4 1 Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 2Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany, 3University of Muenster, Department of Traumatology, Muenster, Germany, 4Martin-Luther-Krankenhaus, Unfallchirurgie, Berlin, Germany Objectives: Tunnel placement in double Bundle (DB) ACL reconstruction can be one of the most important factors on the clinical results. Aim of the current study was to evaluate the femoral and tibial tunnel placement in DB ACL Reconstructions using 3-d CT scans. We hypothesized that using individual femoral and tibial aiming devices an anatomic tunnel placement of the femoral and tibial AM and PL bundle tunnel can be reproducible be performed. Methods: ACL DB reconstructions were performed in 15 patients collected in a prospective study design. All reconstructions were performed by one single surgeon. For femoral and tibial tunnel placement, double bundle ACL aimers were used to accomplish individual differences of sizes of origin and insertion of the ACL. Aimers were set to achieve at least 2 mm bridge between the AM and PL bundle tunnel. On the femoral side a PL bundle aimer (Karl Storz Tuttlingen/Germany) with an off-set of 8, 9, and 10mm was used. On the tibial side, a tibial double bundle aimer (Karl Storz Tuttlingen/Germany)was used that could vary the distance between center of AM and PL individually. Postoperatively, a CT scan with 3-d reconstruction was performed to evaluate accuracy and reproducibility. Results: The femoral aimer was most commonly used with a distance of 9 mm between the centre of AM and PL bundle (9 cases). An off-set of 8 and 100 between the centres of the bundles was used in 3 cases each. At the tibial site, most frequently a 10 mm off set was used. CT 3-d reconstruction showed the femoral AM and PL bundle tunnel placed in its anatomic origin and a bridge width of 2 mm. The tibial AM and PL bundle tunnel was placed at 31 and 44% of the max sagittal tibial diameter, respectively. In no case, a confluence of the tibial tunnel was seen. Conclusions: The results of our study confirm our initial hypothesis and show anatomic femoral and tibial tunnel placement using double bundle guides that account for individual sizes differences. Tunnel placement using the aimers was found to be accurate and reproducible. We conclude that for anatomic tunnel placement, the differences between the centre of AM and PL bundle may vary in patients according to their knee sizes and diameter of the graft. To avoid non-anatomic tunnel placement or tunnel confluence, the anatomy and the size of the insertion has to be respected.

P13-841 The influence of bony morphology on the magnitude of the pivot shift Musahl V.1, Ayeni O.2, Pearle A.3, Wickiewicz T.3 1 University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States, 2 McMaster University, Orthopaedic Surgery, Hamilton, Canada, 3Hospital for Special Surgery, New York, United States

S187 Objectives: ACL injury results in a pathognomonic pivot shift phenomenon. Greater degree of injury generally results in a greater magnitude pivot shift. Little is known about contribution of bony morphology on the magnitude of the pivot shift. The purpose of this study was to correlate pivot shift grading with femoral condyle size as measured on preoperative MRI. It was hypothesized that smaller lateral femoral condyle size would correlate with higher degrees of pivot shift. Methods: Thirty-nine consecutive patients for anterior cruciate ligament (ACL) surgery were examined under anesthesia. The pivot shift was graded according to Galway and MacIntosh by a single observer. Intraoperative findings of injury patterns to the ACL and other soft tissue structures were recorded. The anterior-posterior (ap) and medial-lateral (ml) diameter of femoral condyles and tibial plateaus were measured on preoperative MRI. Patients were grouped into a grade 1 pivot shift group and a grade 2 pivot shift group. Unpaired student’s t-tests were used to compare bony dimensions between the two groups. Significance was set at p \0.05. Results: Twenty-one patients had a grade 1 pivot shift and 18 patients had a grade 2 pivot shift. Associated pathology was present in 9/21 patients (43%) with a grade 1 pivot shift and 18/18 patients (100%) with a grade 2 pivot shift. The ml diameter of the lateral femoral condyle and tibial plateau, respectively, were significantly smaller in patients with a grade 2 pivot shift (27.7±1.7mm and 28.9±2.0mm) compared to patients with a grade 1 pivot shift (29.6±2.5mm and 31.9±3.6mm; p\0.05). No difference was detected for any of the other measurements taken (p[0.05). Conclusions: This study showed that patients with a high grade pivot shift had smaller lateral femoral condyle and tibial plateau diameter compared to patients with lower grade pivot shifts. Associated pathology was found in all patients with a grade 2 pivot shift. Future investigations will include evaluation of pivot shifts following different ACL reconstruction procedures.

P13-842 Recidive instability after ACL reconstruction: preoperative planning using 3-dimensional CT scans Herbort M.1, Zantop T.2, Lenschow S.3, Raschke M.3, Petersen W.4 1 University of Muenster, Department of Traumatology, Muenster, Germany, 2Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany, 3Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 4Martin-Luther-Krankenhaus, Unfallchirurgie, Berlin, Germany Objectives: Tunnel positioning is one of the most common reasons for reinstability after ACL reconstruction resulting in an increased incidence of ACL revisions. When planning ACL revision, a one and two staged procedure can be followed. Aim of the current study was to evaluate the benefit of CT scans using 3-dimensional reconstruction in planning of ACL revision and grading the tunnel placement. Methods: In 30 prospective collected patients suffering reinstability after ACL reconstruction diagnostics were performed using Lachman, pivot shift and Losee tests. Posterior stress x-rays were performed in all patients to evaluate the integrity of the PCL. For evaluation of enlargement of the femoral and tibial tunnels, CT scans were performed and reconstructed in sagittal, frontal and coronal plane. Additionally, a 3-d reconstruction was performed to visualize the femoral and tibial tunnel placement. Results: Of the 30 patients enrolled in the study, 13 had BPTB and 17 had hamstring graft reconstruction at primary ACL reconstruction. CT scans showed significant tunnel enlargement in 12 patients (7 pat tibial, 3 pat femoral, 2 pat femoral and tibial) resulting in a staged surgical procedure with bone grafting. In 18 patients, primary revision were performed. The 3-d reconstruction showed in 14 of these patients a steep femoral tunnel placement in the high noon position outside the anatomic origin of the ACL and the tibial tunnel placement in the posterolateral footprint. In 4 patients, a primary revision using the same tunnel placements as in the first reconstruction could be performed according to the 3-d scans.

123

S188 Conclusions: Analysis of tunnel placement and enlargement are essential factors in preoperative planning of ACL revision. The results of the current study show that a 3-d reconstruction allows exact visualization of the femoral and tibial tunnel placement preoperatively. This facilitates preoperative planning and the patient can be informed reliable about a primary or staged revision procedure.

P13-843 A biomechanical comparison of fan-folded, single-looped fascia lata with other allograft tissues as a suitable substitute for anterior cruciate ligament reconstruction Chan D.1, Temple H.T.1, Latta L.1, Kaplan L.1 1 University of Miami, Department of Orthopaedic Surgery, Miami, United States Objectives: As the number of allograft anterior cruciate ligament reconstructions performed annually continues to increase, demand on tissue banks for allograft tissue is becoming a concern. As such, viable alternatives to traditional allograft tissue are necessary. The purpose of this study was the evaluate the initial biomechanical properties of a single-loop construct of fan-folded fascia lata allograft in comparison to other allograft tissues. Methods: Eighteen fascia lata specimens were harvested from 11 donors and fan-folded using a proprietary process. Bone-patellar tendon-bone (BPTB), tibialis anterior, tibialis posterior, and peroneus longus tendons were harvested from four additional donors. All soft tissue grafts were tested to failure in an MTS machine in a single-looped fashion. BPTB grafts were similarly clamped in freeze grips. The ultimate load to failure and stiffness were calculated for each graft type tested. Results: The mean ultimate load to failure was 3266 N and stiffness was 414 N/mm for the single-loop fascia lata grafts (n=18). There was no significant difference for either ultimate load to failure or stiffness between the fascia lata and tibialis anterior (3012 N, 342 N/mm), tibialis posterior (3666 N, 392 N/mm), and peroneus longus tendons (3050 N, 346 N/mm). The fascia lata grafts performed significantly better (p\0.001) than BPTB (1404 N, 224 N/mm). Conclusions: The use of a fan-folded, single loop construct of fascia lata allograft in this in vitro study suggests that it is comparable, with respect to initial tensile strength and stiffness, to other soft tissue allografts currently used in clinical practice. In this study, the fascia lata allograft performed significantly better than BPTB allograft. In the face of potential allograft tissue shortages and increasing constraints on health care expenditures, the use of fascia lata has the potential of being a readily available graft for anterior cruciate ligament reconstruction that performs as well as other allografts and at a comparable or lower cost.

P13-850 Conventional tran stibial aiming devices fail to hit the center of the human ACL when used via a medial portal approach Zantop T.1, Herbort M.2, Bremer S.2, Lenschow S.3, Petersen W.4 1 Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany, 2University of Muenster, Department of Traumatology, Muenster, Germany, 3Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 4Martin-Luther-Krankenhaus, Unfallchirurgie, Berlin, Germany Objectives: Aim of this study was to investigate the accuracy and reproducibility of transtibial aiming devices using the medial portal technique. We hypothesize that femoral tunnel placement in ACL single bundle reconstruction using transtibial aiming devices fail to hit the center of the anatomic ACL origin. Methods: In twenty fresh frozen human cadaveric knees (range 42-86 years, 10 right, 10 left specimen) insertions sites of the ACL were marked with a surgical marker and the center of the ACL origin determined using digital calipers and the distance of the center to the roof of the intercondylar notch (rc) and to the deep cartilage margin was determined.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Subsequently, a conventional transtibial over the top guide with an offset of 5.5mm was used in a medial portal technique to locate the center of a single bundle ACL reconstruction tunnel with a k-wire. Again, the distances of the center of the SB reconstruction tunnel to the roof of the intercondylar notch and to the deep cartilage margin was determined. The k-wire was then overdrilled using a 9mm drill. The femoral tunnel positions were documented using digital imaging. Radiographic analysis of the femoral tunnel were performed according to the quadrant technique as described by Bernard and Hertel (1996). Results: For the center of the ACL on the femoral origin, the distance to the roof of the notch was 10.3 mm (±2.1). The distance to the deep cartilage margin was 9.5 mm (±1.7). The center of the femoral single bundle ACL reconstruction tunnel using a conventional transtibial over the top guide however was located at 6.7 mm (±1.5) to the roof of the notch and 4.8 mm (±1.3) to the deep cartilage margin. This difference in distance to the roof and to the deep cartilage margin for the ACL center and the SB femoral tunnel was statistically significant (p[0.05). Radiological analysis of the femoral insertion according to the technique described by Bernard and Hertel showed that the femoral center of the ACL tunnel was measured to be distance a 28.0% of distance t and distance b was measured to be 49.5 of distance h. The femoral center of the femoral single bundle ACL tunnel was measured to be distance a 20.5% of distance t and distance b was measured to be 25.3% of distance h. This difference was statistically significant. Imaging analysis after overdrilling the k-wire showed femoral break off in 18 of 20 cases. Only in two cases a small bridge was still left. Conclusions: The specific aim of this study was to investigate the accuracy and reproducibility of transtibial aiming devices using the medial portal technique. The results support our initial hypothesis. Compared to data from the literature, the aiming devices were successful in drilling a tunnel in the center of the AM bundle in double bundle ACL reconstruction. However, for an anatomic single bundle approach transtibial aiming devices fail to aim at the center of the entire ACL origin. The clinical relevance of the current study is twofold. First, care should be taken to avoid femoral blow when using a transtibial guide via a medial portal approach for femoral tunnel placement. Secondly, these guides fail to locate the femoral k-wire in the center of the fibres of the entire ACL origin. Therefore, when aiming for an anatomic single bundle ACL reconstruction, new medial portal aimers should be developed.

P13-851 Morphometric measurements (tibial and femoral footprints, anisometric envelope) in navigation assisted double bundle anterior cruciate ligament reconstruction Plaweski S.1, Petek D.1, Saragaglia D.2 1 University of Grenoble, Orthopaedic Department, Grenoble, France, 2 Grenoble South Teaching Hospital, Department of Orthopaedic Surgery and Sport Traumatology, E´chirolles, France Objectives: The native ACL does not behave as a simple band of fibers with constant tension but a continuum of fibers with differential length change throughout knee flexion. Computer assisted surgery may be used to assess the behavior of two bundles reconstruction grafts. The goal of this study is to show and to confirm that it is possible to routinely obtain an anatomometric transplant position in a given knee with a navigation system. Methods: 40 ACL deficient patients underwent 2-bundle reconstruction assisted by surgical navigation (semitendinosus graft to reconstruct the anteromedial (AM) bundle and a gracilis graft for the posterolateral (PL) bundle). The AM tibial tunnel was placed anterior and medial into the native ACL footprint and without impingement with the roof. The AM femoral tunnel was referenced from the notch, with an offset of 6 mm at 11 o’clock for a right knee and at 1 o’clock for a left knee and drilled through the medial portal. The PL femoral tunnel was referenced off the anterior articular cartilage with a 3 mm osseous bridge between the two guide pins. The PL tibial tunnel was referenced within a triangle created by the posterior root of the lateral meniscus, the PCL

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 insertion site, and the AM bundle tibial footprint, achieving divergent tunnels. An image free, optical navigation system was used to digitize the tibial and femoral attachment of double bundle ACL reconstruction in 30 knees and to measure the anisometry throughout flexion and extension. Results: Anatomic insertion site fill: The tibial attachment area of both AM and PL bundles was mm2 (range to) and the femoral (). Anisometry profiles: The central fiber of AM bundle was relatively isometric during passiv knee flexion (0-110 degrees) with an average of mm (). The PL bundle was always anisometric but with a favorable curve (anisometry decreasing with flexion) with an average of mm (). Laxity tests: The lachman anterior and rotative laxity before reconstruction was respectively 12±4 mm and 32±9. After transplantation, the lachman anterior laxity was 2± 2 mm and rotative laxity 18±9. Sagittal and rotative laxity gains were statistically significants (p\0.001). Conclusions: We found, as in litterature, that AM and PL bundles provide objective in vivo data in a significant improvement in anterior lachman laxity but also a significant improvement in rotational laxity and however a great interindividual variability. Double bundle reconstruction increase graft size and appears to capture more of the anisometry envelope of the native ACL.

P13-857 Surgical outcome of transphyseal ACL reconstruction in skeletally immature patients using quadruple hamstring graft Plaweski S.1, Courvoisier A.2 1 University of Grenoble, Orthopaedic Department, Grenoble, France, 2 Grenoble South Teaching Hospital, Department of Orthopaedic Surgery and Sport Traumatology, Grenoble, France Objectives: Our purpose was to evaluate the clinical and radiological outcome at maturity of an « adult-like » transphyseal anterior cruciate ligament (ACL) reconstruction performed in skeletally immature patients using quadruple hamstring free graft. Methods: The records of all skeletally immature patients who underwent transphyseal ACL reconstruction between 2004 and 2006 at our institution were reviewed. Inclusion criteria were age less than 16 years and radiographic evidence of wide open physes. 38 children and prepubescent were identified. All underwent postoperative clinical evaluation with International Knee Documentation Committee scores and long leg radiographs. Each patient was followed up until skeletal maturity was confirmed. Results: 30 patients were scored A, five B and two C according to IKDC. At last follow-up, there were no radiographic evidence of malalignment in any of the patients. 5 underwent a reoperation, and 3 suffered traumatic graft disruption. Conclusions: The quadruple hamstring free graft seems a relevant procedure for ACL reconstruction in skeletally immature patients. Our findings support early operative treatment of ACL ruptures even in young patients with open physes.

P13-862 Fixation strength in ACL reconstruction after accidental perforation of the lateral femoral cortex Herbort M.1, Heletta S.1, Lenschow S.2, Petersen W.3, Zantop T.4 1 University of Muenster, Department of Traumatology, Muenster, Germany, 2Wilhelms University Muenster, Department of Trauma-, Handand Reconstructive Surgery, Muenster, Germany, 3Martin-LutherKrankenhaus, Unfallchirurgie, Berlin, Germany, 4Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany Objectives: Tunnel placement for ACL reconstruction with cortical flip buttons especially in medial portal technique implies the risk of perforating the lateral cortex with a drill of larger diameter than the intended 4.5 mm. Aim of the study was to evaluate the structural properties after penetration of the lateral femoral cortex. We hypothesized that a

S189 penetration of the second cortex using 7, 8 and 9 mm drill diameters results in significantly lower structural properties compared to a 4.5mm penetration. Furthermore we hypothesize that after perforation of the lateral cortex a hybrid fixation using a size matching interference screw increases the low structural properties significantly. Methods: In 100 porcine femurs ACL reconstruction using a medial portal approach was performed. The constructs were mounted in a uniaxial material testing machine ZWICK/ROELL Z005. In the control group, a penetration of the second cortex was performed with a 4.5 mm diameter. In compared groups a tunnel with a continuous diameter of 5, 5.5, 6, 7, 8 and 9 mm has been drilled. In the first part of the study the grafts have been fixed by a cortical fixation using a 1 mm Ethibond suture and a flip button (FlippTack, Karl Storz, Tuttlingen/ Germany). In the second part of the study, a hybrid fixation with an additional interference screw (MegaFix, KARLSTORZ, GmbH & Co. KG, Tuttlingen, Germany) has been performed. Every construct has been loaded cyclically 1000 times between 50 and 250 N thereafter all constructs were loaded to failure. Identified parameters were number of survived cycles, stiffness, maximum load, yield load, elongation after 1000 cycles and modes of failure. Statistical analyses were performed using the Kolmogorow-Smirnow Test and the Mann-Whitney U Test (p \ 0.05). Results: The control group without perforation of the femoral cortex survived the cycling testing protocol and showed following parameters: mean elongation after 1000 cycles was 4.3 (±0.8) mm, stiffness 114.9 (±27.7) N/mm, yield load 492.1 (±28.7) N and mean maximum load 670.8 (±104.1) N. Groups with a continuous tunnel diameter of 5, 5.5 and 6 mm (cortical fixation) survived the cycling testing procedure and showed no significantly different parameters during load to failure testing (p[0.05 to control group). In the 7, 8 and 9 mm tunnel group the constructs did not survive the cyclic testing. In the hybrid fixation group the reconstruction evaluating a 7, 8 and 9 mm penetration survived the cyclic protocol. During load to failure testing of the hybrid fixation (7, 8 and 9 mm) we found no significant difference in all identified parameters in comparison to the control group (p[0.05). Conclusions: The results of the present study support our hypotheses. In the first part of the study, the series with tunnel diameters from 5 mm up to 6 mm the structural properties showed no significant differences to the conventional technique with a 4.5mm perforation. However, none of the reconstructions with a 7 to 9 mm penetration survived the cyclic loading protocol. Here, the surgeon needs to be aware of the significantly altered lower stability. The data further supports our second hypothesis that in a in the case of a drill through of the lateral cortex with a 7 to 9 mm drill, a hybrid fixation may increase the low properties significantly. After hybrid fixation no significant differences compared to the control group were to be found.

P13-863 The analgesic efficacy of continuous local anaesthetic infiltration for outpatient hamstring anterior cruciate ligament reconstruction: a prospective trial Chambers M.1, Storey N.2, Rooney B.3 1 Western Infimary, Orthopaedics, Glasgow, United Kingdom, 2Western Infimary, Glasgow, United Kingdom, 3Gartnavel General Hospital, Orthopaedics, Glasgow, United Kingdom Objectives: The purpose of this study was to assess the analgesic efficacy of continuous local anaesthetic infiltration (CLAI) at the hamstring donor site following anterior cruciate ligament (ACL) reconstruction. Methods: Over a 12 month period Patients undergoing primary ACL reconstruction with hamstring tendon graft under general anaesthetic were included in the study. 46 patients were included in this study with 30 acting as controls and 16 receiving CLAI for the first four post operative hours. The control patients received an intra-articular injection of Bupivicaine at the end of their operative procedure. The CLAI group also received a bolus of 10mls into the donor site followed by a 40mls of 0.125% Bupivicaine over a 4 hour period via an Elastometric pump into

123

S190

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

the donor hamstring muscle belly by way of a 18G epidural catheter. The catheters were removed after four hours. Visual analogue scale (VAS) pain scores and opiate consumption were recorded for all patients in the recovery room. Results: The average VAS pain scores for both groups at 4 hours were equal at 3 out of 10. The average total opiate use equivalent to mg of subcutaneous Morphine was 5.5 for the controls and 5.0 for the CLAI group. Of the control group 16 (53%) compared to 7 (43%) of the CLAI grouped were admitted for pain control and mobilisation having failed to have their pain adequately controlled to allow mobilisation by specialist physiotherapist staff (p=0.038). Conclusions: Short term CLAI is recommended for use in outpatient ACL reconstruction with hamstring grafts.

Results: 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: Our initial results suggest that the oral contraceptive users have better capability for neuromuscular coordination. 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 injury prevention.

P13-864 Anatomic reconstruction of the anterior cruciate ligament through the double-bundle technique Gali J.1 1 Catholic University of Sa˜o Paulo, Sorocaba, Brazil Objectives: Result analysis of the anatomic reconstruction of the anterior cruciate ligament with the double-bundle, through the International Knee Documentation Committee 2000 protocol. Methods: Fifty eight patients went under anatomic reconstruction of the anterior cruciate ligament using autholog flexor tendon graft through the double-bundle technique were evaluated according to IKDC 2000 protocol. Patient age varied from 17 to 58 years, with a mean of 35.2 years. Follow-up varied from 24 to 37 months (mean of 28.9 months). Results: On the final evaluation 89.65% of the patient knees were graduated as normal, 8.62% nearly normal and 1,72% as abnormal. Conclusions: The used technique was effective in promoting articular stability return, without range of motion compromise.

P13-878 ACL double bundle reconstruction. Results - one year follow-up Fraga Ferreira J.E.1, Ferrero R.2, Cerqueira R.3, Vic¸oso Sousa Fernandes S.I.3, Barbosa T.3, Basto T.3, Lourenc¸o J.3 1 Centro Hospitalar do Alto Ave Guimara˜es, EPE, Orthopaedics and Traumatology, Guimara˜es, Portugal, 2Centro Hospitalar do Alto Ave Guimara˜es, Ortopedia, Guimaraes, Portugal, 3Centro Hospitalar do Alto Ave, Orthopedics and Traumatology, Guimara˜es, Portugal Objectives: Double bundle ACL reconstruction has become very popular over the past years. The normal ACL is composed of two functional bundles: the anteromedial (AM) bundle and the posterolateral (PL) bundle. The PL bundle, which is not traditionally reconstructed, plays a significant role in rotatory stability in the knee. Numerous clinical studies have demonstrated that this technique closely reproduce the normal ACL anatomy, restoring normal knee kinematics and tibial rotation. The authors aim to review the preliminary results of this technique in our institution. Methods: From June 2006 to July 2008, 26 patients underwent double bundle ACL reconstruction with semitendinous and gracilis autografts. 21 patients were tested. The mean follow-up was 15 months (12 to 36). We used four tunnel technique with EndoButton femoral and interference screw tibial fixation. Subjective evaluation comprised the IKDC 2000 subjective score. Objective evaluation followed the IKDC 2000 Knee examination protocol. Antero-posterior instability was measured by Rollimeter. Pivot shift test was used to evaluate rotational instability and one leg hop test was performed to assess the functional results. Results: The mean subjective result was 90,93. The pivot-shift test grouped 18 patients A and and 2 B. The antero-posterior translation was 2,9, with a 1,1 difference to the other knee. The one leg hop scored 18 patients A and 3 B. The final evaluation was that 47,6% of patients scored A and 38% scored B. Conclusions: Double bundle ACL reconstruction is an effective procedure. A long term clinical study with a larger group of patients is needed. Comparative studies and a more objective analysis of rotational stability is needed. Potential complications and revision surgery are concerns. Our results are within those seen in the literature.

P13-872 The effect of oral contraceptives on ACL stiffness and proprioception Ba´nyai T.1, Molna´r G. B.2, Szı´ver E.3 1 County Hospital, Kecskeme´t, Traumatolgy and Hand Surgery, Kecskeme´t, Hungary, 2SZTE University, Department of Obstetrics and Gynecology, Szeged, Hungary, 3SZTE Ege´szse´gtudoma´nyi e´s Szocia´lis Ke´pze´si Kar Fiziotera´pia´, Szeged, Hungary Objectives: Women suffer 4-8 times the anterior cruciate ligament (ACL) rupture rate for the same sport as males. Our study group in Perugia analyzed the hormonal factor of the possible reasons. Between the phases of the menstrual cycle there was no significant difference in ACL stiffness and proprioception. In the second part of our measurements we analyzed pregnant women because they have significantly higher sexual hormone levels. We found decreased proprioceptive capacity and increased knee laxity in pregnant women. In the third part we measured the effect of oral contraceptives on knee joint laxity and proprioception. Methods: We analyzed 19 female volunteers, 8 of them used regularly oral contraceptive (OCP) triphasic pills, 11 never used it. The two groups were comparable. (Normal distribution according to body mass index and age). The women in second group were on the preovulatory phase of their regular menstrual cycle. Measuring ACL stiffness: – KT2000 arthrometer (15,20,30 pounds force and maximal femoraltibial displacement during 30 flexion of the knee). Detecting elongation of the ACL in millimetres. Calculated parameter: Maximal elongation-elongation with 15lbs. Measuring proprioception: – Stabilometry: active and passive balance control with opened and closed eyes, detecting sway path, splitting to anterior-posterior (ACL role) and lateral axes. – Joint position sense: Joint positions in 30, 60, 90 grad, measuring the differences between the sensed and real knee position 3 times Exclusion criteria: Operated knee, knee instability.

123

P13-879 ACL remnant, to save or not? Nakase J.1, Kitaoka K.1, Goshima K.1, Ueshima K.1, Hayashi M.1, Ryohei T.1, Tomita K.1 1 Kanazawa University, Orthopaedics Surgery, Kanazawa, Japan Objectives: In some cases, anterior cruciate ligament (ACL) remnants are found during knee arthroscopy for ACL injury. There are several morphologies of remnants, but their clinical significance is unknown. In this study, we investigated cases with ACL remnants and report their characteristics. Remnant-preserving ACL reconstruction has been reported to be advantageous in preserving proprioceptive function, and effective in early reperfusion and prevention of bone tunnel enlargement. On the other hand, this procedure can result in pain from tendon graft thickening, limited extension of the knee joint, and cyclops

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 syndrome. There are no guidelines on how the remnants should be treated. Since 2006, we have been performing ACL reconstructive surgery, preserving as much remnant as possible. We report here the clinical outcomes. Methods: The subjects were 51 patients in whom the same surgeon performed ACL reconstruction between October 2006 and September 2008. There were 21 males and 30 females, and the mean age was 23 years (1548 years). The presence or absence of remnants and their morphologies were retrospectively examined using intraoperative videos. The waiting period from the time of injury to surgery and the presence or absence of knee giving way during the waiting period were examined. Then their relationships with remnants were also investigated. The differences in preoperative knee instability using KT-1000 arthrometer were compared in two groups: 31 patients with remnants and 20 patients whose remnants disappeared. The following were also compared in 21 patients with remnant-preserving ACL reconstruction and in 30 patients with nonpreserving ACL reconstruction: differences in postoperative knee instability using KT-1000 arthrometer, presence or absence of extension limitation, subjective recovery levels, and presence or absence of enlarged bone tunnels one year postoperatively. Results: Remnants were present in 60% of the cases. When a comparison was made between the group with surgery performed within 3 months of injury and the group with surgery performed 3 months or more, there was no significant difference in the incidence of remnants. In addition, the presence or absence of giving way in the waiting period did not affect the remnant incidence. The group with remnants had a significantly lower preoperative KT-1000 arthrometer value than the group without remnants. It was 21 patients with remnant-preserving ACL reconstruction in 31 patients with remnants. There were no significant differences between the remnant-preserved group and the non-preserved group in the difference in postoperative KT-1000 arthrometer values, subjective recovery levels, and bone tunnels enlargement. Extension limitation occurred in one patient in the remnant-preserved group and in two patients in the non-preserved group. Conclusions: There were 4 patients with the following conditions among patients with remnants: the difference in preoperative KT-1000 arthrometer values was within 2 mm and without anterior instability. The group with remnants had a significantly lower preoperative KT -1000 arthrometer value, suggesting that remnants are stabilizing anterior tibial translation to some degree. Although the postoperative outcomes tended to be good in the remnant-preserved group, there was no significant difference compared to the non-preserved group.

P13-882 In vivo kinematics study of pivot shift test using 2D/3D matching technique Suzuki T.1, Yamazaki T.2, Nakata K.3, Amano H.3, Nakamura N.3, Shino K.4, Yamashita T.1, Sugamoto K.5 1 Sapporo Medical University School of Medicine, Department of Orthopedic Surgery, Sapporo City, Japan, 2Osaka University, The Center for Advanced Medical Engineering and Informatics, Osaka, Japan, 3Osaka University Graduate School, Orthopaedics, Suita, Japan, 4Faculty of Comprehensive Rehabilitation, Osaka Prefecture Un, Habikino, Japan, 5 Osaka University Graduate School of Medicine, Orthopaedics, Suita, Japan Objectives: Pivot shift instability is a complex motion involving multiple degrees of freedom. No method exists that is able to evaluate pivot shift quantitatively. We visualized such a complex pivot shift motion from clear single-plane fluoroscopy images with flat panel detector using bone models created from computed tomography (CT). The purpose of this study was to evaluate in vivo three-dimensional (3D) kinematics of pivot shift test of anterior cruciate ligament (ACL) intact and deficient knees. Methods: 8 patients (3 men and 5 women) with ACL injury were investigated. The contralateral sides (8 normal knees) of the ACL deficient knees were also studied. The average age was 25.9 years, range 14-40 years. The average time from injury to the test was 11.4

S191 month (6-60). Continuous X-ray images of a pivot shift test by a same examiner were taken using a flat panel detector (7.5 fps). CT-derived bone models were used for model registration-based 3D kinematic measurement. Three-dimensional joint kinematics and the distance between ACL (AM: anteromedial and PL: posterolateral) attachment sites were determined. Results: The average pivot shift for the ACL deficient knees occurred over the flexion 22.7±8.8(±standard deviation). The pivot shift consisted of a mean posterior translation 6.8 mm±4.5 mm at the medial compartment and 11.2 mm±9.2 mm followed by an external rotation of 7.8±4.8 during the posterior reduction. In the ACL deficient knee, shortening was 6.7 mm ± 2.5 mm at the distance between AM attachment sites and 8.1 mm ± 3.1 mm at PM during reduction. there was no significant difference between the shortening of the distance of AM and PL(p[0.01). The shortening of AM and PL attachment sites in ACL deficient knees had significantly difference compared with these of normal knees(p \0.01) Conclusions: This method is useful to visualize the real osseous motion and assess the kinematics measurement of pivot shift test. The results show that pivot shift test result in significant posterior tibial reduction and shortening distances of AM and PL attachment sites in ACL deficient knee. There was a statistical significant difference when compared to the intact knee. It is suggested that in vivo kinematics of pivot shift test were rather the posterior tibial reduction that is direction of ACL fiber arrangement than the tibial rotation.

P13-885 Correlation of graft position, knee laxity and clinical outcome: comparison with native anterior cruciate ligament using magnetic resonance imaging study Saowaprut S.1 1 Institute of Orthopaedics, Lerdsin Hospital, Bangkok, Thailand Objectives: The purpose of this study were, first, to compare the sagittal obliquity of anterior cruciate ligament graft with normal native anterior cruciate ligament in contralateral knee;second, to determine the effect of sagittal obliquity and axial femoral tunnel graft placement on stability and functional knee score (Lysholm), third, to measure size of graft after complete ligamentization. Methods: 70 single tunnel quadruple hamstring anterior cruciate reconstructed knee in unilateral ACL injury patients were evaluated at 18 months after surgery. At follow up patients were evaluated including measurement of knee laxity by using side to side different on KT 1000 arthromeres and clinical outcome by completed Lysholm functional knee questionnaires. Sagittal T1 weighted magnetic resonance image with complete dimension of graft from origin to insertion on each side of knee were depicted to compare the obliquity by measuring the intersection angle of the graft line with the tibial plateau plane. The axial femoral tunnel was determined by angle between anteroposterior axis of distal femur and long axis of femoral tunnel. The diameter of graft were also measured. Results: Graft obliquity was average 58o with range between 41o and 69o. In contralateral native ACL obliquity was average 50o with range between 33o and 63o. The difference between two groups was statistically significant (p\0.0001). Average axial femoral tunnel was 36o with range between 10o and 56o. Knee laxity (KT-1000 arthrometer; average pre-op =6, post-op =3) and Lysholm knee scores (average score; pre-op55, postop score 89) were significantly improved after surgery (p \0.01). There was no correlation between degree of sagittal obliquity and axial femoral tunnel with knee laxity and functional score in this series. Graft size was increased in average 8% after 18 months post-operatively. Conclusions: ACL graft in patients with appropriate tibial tunnel placement were more vertical than native ACL. There was no significant effect of degree of sagittal obliquity and axial femoral tunnel to antero-postero stability and knee score. ACL graft size were increased in diameter during post-operative period. Graft-notch distance should be considered during operation.

123

S192 P13-889 Familial predisposition for anterior cruciate ligament injury Goshima K.1, Kitaoka K.1, Nakase J.1, Ueshima K.1, Hayashi M.1, Takahashi R.1, Tomita K.1 1 Kanazawa University, Orthopaedics Surgery, Kanazawa, Japan Objectives: Although the exact cause of anterior cruciate ligament (ACL) injury is unknown, various intrinsic and extrinsic risk factors have been identified. Intrinsic risk factors include generalized ligamentous laxity, female gender, anatomic, hormonal, biomechanical, and neuromuscular factors. However there are few reports about a familial predisposition toward ACL injury. The purpose of this study was to investigate whether there is a familial predisposition for ACL injury. Methods: The subjects were 350 patients in whom the same surgeon performed ACL reconstruction between January 2005 and September 2008. All the patients received a questionnaire by telephone or mail that detailed their family history of ACL injury, family’s sports, and mechanisms of injury. Then Tegner activity score, posterior tibial slope and general joint laxity were compared between the group with a family history of ACL injury and the control group. The control group consisted of 194 patients that had complete data and excluded patients over 35 years old. Results: Complete information was obtained from 335 of 350 patients (95.7%). There were 110 males and 225 females, and the mean age was 21 years (13- 47 years). Thirty-nine patients (11.6%) had a family history of ACL injury. The family relationships were brothers and sisters (n=24, 5 brothers, 19 sisters), parents (n=16, 6 fathers, 10 mothers) and uncle (n=1). Two families had 3 members with ACL injury. The sports the families participated in were basketball 17 case, volleyball 11 cases, handball 5 cases, and 2 cases each soccer, judo, rugby and skiing. Twenty-five of 39 subjects (64.1%) suffered ACL injury in the same sports. Thirty-five cases (89.7%) had a non-contact injury, while 21 cases (53.8%) injured ACL by the same mechanism as their family members. There was no significant difference in Tegner activity score between two groups. The group with a family history of ACL injury had significantly increased joint laxity and posterior tibial slope compared to the control group. Conclusions: Our results indicated that 11.1% of patients with ACL injury had a family history of this injury. As the injury mechanisms were very similar in many cases, it is highly probable that many of the identified risk factors for ACL injury are passed through families. Intrinsic risk factors, including increased joint laxity and tibial slope, may occur in a family and may lead to the predisposition for ACL injury. Prevention programs are therefore important for subjects with a family history of ACL injury and the risk of ACL injury can be decreased in high-risk subjects with neuromuscular training.

P13-893 Septic arthritis after reconstruction of previous crossed ligament: incident, results and algorithm Ruiz-Zafra J.E.1, Valencia-Garcı´a H.1, Gavı´n-Gonza´lez C.1, Santana-Ramirez S.1, Canosa-Sevillano R.2 1 Hospital Universitario Fundacio´n Alcorcon, Traumatology, Alcorcon, Spain, 2Hospital Quiro´n, Traumatology, Pozuelo, Spain Objectives: To evaluate the incident and results of the septic arthritis after reconstruction of anterior crossed ligament (ACL) to establish an algorithm of performance. Methods: 373 arthroscopic ACL reconstruction realized in our center from December, 1998 until December, 2008 (11 years), are checked. We report 5 patients who developed postoperative infection. (1,3% of incidence). There are analyzed age, sex, affected knee, associate injuries, time of delay up to surgery, technique (Autograft bone-patellar tendon-bone reconstruction was performed with double-incision technique and interference screws in all patients) and time of ischemia. Cefonicid 1 g was given preoperatively to all patients for antibiotic prophylaxis. Results: In all five cases diagnosis was clinical and for the results of artrocentesis, identifying the responsible microorganism in 3 cases (60%, Staphylococcus aureus was cultured in 1, and Staphilococcus epidermidis were isolated in 2 patients). All cases of patients underwent immediate

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 arthroscopic debridement with sinovectomy and irrigation with a minimum of 12 L saline 15 days of average after the first surgery (7-23 days) without hardware removal and ACL autografts were left inside in all patients. Intravenous antibiotic was carried out during 7 days and oral antibiotic to complete 35 days according to specific culture antibiogram. In all 5 cases a resolution of the infection was obtained with a functional satisfactory result to the year of the surgery, without cases of arthrofibrosis (range mobility:38-128 8) or alterations in the osseous tunnels. 2 cases were presenting Lachman0 s test ? / ??? without clinical repercussion, with a sports and work level similar to injury before. Conclusions: ACL reconstruction is a frequent procedure (100000/an˜o in USA) however septic arthritis is a rare strange complication (incidence: 0,14-1,7%) but with possible serious consequences, like arthrofibrosis, osteonecrosis and condrolisis. The risk meets increased with previous surgeries or concomitant interventions, but neither the time of isquemia, nor the technics, nor the type of graft seem to influence. Clinical signs may appear in the first 20 days like pain, fever, local increase temperature, swelling or functional limitation. If we have the minimum suspect of arthritis we must get knee aspirate to laboratory study and if septic arthritic is confirmed (normally with leukocytes[50000, 90% polimorfonucleares or positive culture) or there is a lack of response to the antibiotic empirical treatment, we should carry out an arthroscopic debridement with sinovectomy and lavage. diagnosis may not be as clear and be masked by postoperative changes. erythrocyte sedimentation rate (ESR)and C-reactive protein (CRP) are usually high and helping to define strategies at clinical indolent presentation. The most frequent microorganisms are Staphilococcus aureus and coagulase-negative Staphylococcus but others have been described. In our experience, the conservation of the graft did not have final repercussion, though any works suggest that the plastia behaves as a foreign body in our organism until it is completely incorporated and may promote chondrolysis. Early debridement and lavage. (maximum 3 weeks) can avoid this chondral damage. Consensus does not exist on the duration of the antibiotic treatment, but there seems to be clear that wait-and-see approach is not appropriate if infection is suspected.

P13-895 Comparison femoral transcondylar fixation and endobutton fixation of ACL graft in arthroscopic ACL reconstruction Yung A.W.Y.1, Chang H.C.1 1 Changi General Hospital, Orthopaedic Surgery, Singapore, Singapore Objectives: The aim of this study is to compare the surgical outcome between transcondylar graft femoral fixation(Crosspin) and extraarticular graft femoral fixation(Endobutton). Methods: A prospective case control study was conducted from 2005 to 2008 with 32 patients undergoing ACL reconstruction surgery. 32 patients were divided into group 1 (17 patients) for transfemoral graft fixation (crosspin) and group 2 (15 patients) for extraarticular graft fixation(endobutton). All the patients underwent the same fixation method at tibial side. The subjects were matched evenly in age, sex, time from injury to surgical reconstruction and mensical injuries. Exclusion criteria included acute tear of ACL, multiple ligamentous injuries and initial presence of degenerative joint disease. All surgeries were conducted by senior surgeon (CHC) and patients were subsequently assessed during the 24 months follow-up with IKDC, KT-2000 arthrometer, Lysholm-II form, Tegner score and knee radiography. The result was analyzed with independent sample t test and the statistical significance was set at P\ 0.05. Results: 17 patients underwent transfemoral (crosspin) graft fixation and 15 patients underwent extraarticular (endobutton) graft fixation. The mean age is 27 and 25 years old respectively. The mean BMI of both group patient are 24. Group1 patient pre -operation mean Tegner’s score is 4.35, subjective IKDC mean score is 47.4, objective IKDC 65 % C, 35% D, Lysholm mean score is 63, KT 2000 mean value is 4.29. Group1 patient post -operation mean Tegner’s score is 6.88, subjective IKDC mean score is 82.2, objective IKDC 47% A, 53% B, Lysholm mean score is 90, KT 2000 mean value is 0.588.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Group2 patient pre -operation mean Tegner’s score 4.67, subjective IKDC mean score is 49.4, objective IKDC 60% C, 40% D, Lysholm mean score is 64, KT 2000 mean value is 4.33. Group2 patient post -operation mean Tegner’s score is 6.26, subjective IKDC mean score is 81.6, objective IKDC 53% A, 7% C, 40% B, Lysholm mean score is 94.1, KT 2000 mean value is 0.33. T test analysis comparing post op measuring of group 1 and group 2 patients. Tegner score P value is 0.2598 (CI -0.48 to 1.71), subjective IKDC p value is 0.8996 (CI -0.8473 to 9.6), Lysholm score p value 0.2276 (CI -10.05 to 2.49) and KT 2000 p value is 0.324 (CI -0.26 to 0.77). Conclusions: The is no significant clinical outcome difference between patients with transcondylar crosspin graft fixation and extraarticular endobutton graft fixation.

P13-920 Double bundle hamstring ACL reconstruction has no superior results when compared to anatomical single bundle reconstruction over an anteromedial portal- prospective, matched pairs analysis after one year Hoeher J.1, Braas M.1, Offerhaus C.1, Balke M.1 1 Merheim Hospital, Clinic for Sports Traumatology, Cologne, Germany Objectives: It was the objective of this study to evaluate the results of primary ACL reconstruction using single or double bundle quadrupled hamstring graft one year after surgery using a matched-pairs analysis. Methods: Fifty patients were operated from 2006 to 2007 using double bundle (DB) technique by a single surgeon. From a pool of 279 patients who underwent anatomic single bundle reconstruction (SB) over a anteromedial portal during the same time period, fifty patients were selected for a matched pairs analysis to exclude differences in age, gender and concomitant injuries. After one year patients were reevaluated clinically including subjective and objective IKDC and Tegner score. Results: Surgical time was 93min for DB and 82min for SB. No major complications were observed during the study period. There was one rerupture in each group. Results at follow up were: instrumented knee laxity (manual maximum displacement) at Lachman test 1,48±1,50mm for DB and 1,77±2,05mm for SB group. IKDC subjective was 85,1±13,09 for DB and 88,7±11,96 for SB, Tegner was 6±2,21 for DB and 6,6±2,02 for SB. One leg hop was 91,9% (DB) and 96,7% (SB). Pivot shift Test was negative in 78% (DB) and 92% (SB). All results were not significantly different. Conclusions: We conclude that this study failed to demonstrate superior results of DB technique at one year after surgery when compared to an anatomical single bundle technique over the anteromedial portal at the cost of a longer operating time. Operative Techniques for anterior, cruciate ligament reconstruction remain controversial.

P13-922 Evaluation of intercondylar roof impingement in anatomical double bundle anterior cruciate ligament reconstruction using transparent 3D-CT Iriuchishima T.1, Kubomura T.1, Horaguchi T.1, Morimoto Y.1, Tokuhashi Y.1, Goto B.2, Ingham S.2, Fu F.2, Saito A.1 1 Nihon University School of Medicine, Orthopaedic Surgery, Tokyo, Japan, 2University of Pittsburgh, Dept. of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States Objectives: Intercondylar roof impingement is an important concern in anterior cruciate ligament (ACL) reconstruction. Although, anatomical double bundle ACL reconstruction is becoming popular, roof impingement in the anatomical double bundle ACL reconstruction has not been investigated. Recently, Inoue and Yasuda et al reported the usefulness of transparent 3D-CT (computerized tomography) to evaluate post operative ACL graft characteristics. The purpose of this study was

S193 to evaluate intercondylar roof impingement in the anatomical double bundle ACL reconstruction using transparent 3D-CT. Methods: Nineteen subjects submitted to anatomical double bundle ACL reconstruction were included in this study. The antero-medial (AM) and postero-lateral (PL) tunnels were placed at the native ACL stumps and a trans-portal or trans-tibial technique was used for the femoral tunnels. In all cases, hamstrings auto grafts were used. 6 to 8 weeks after ACL reconstruction, and when the subjects achieved full knee extension, transparent 3D-CT was performed with the knees placed at full extension. Roof impingement was evaluated in an axial view. Femoral tunnels placements were also evaluated (Bernald and Hertel’s technique). Tibial tunnel evaluations were performed using intra operative X-ray (Staubli’s technique). Results: No intercondylar roof impingement was observed. In 10 subjects, the ACL graft touched the roof (Group T) but no graft deformation was observed. In 9 subjects no roof-graft contact was observed. (Group F). No significant difference of femoral and tibial tunnels placements were observed between Groups T and F. All subjects had full knee extension. Conclusions: Although group T was characterized by roof-graft contact, we believe this is physiological and has no deleterious effect on the graft. In conclusion, transparent 3D-CT is effective to evaluate impingement after ACL reconstruction. When the grafts are positioned in an anatomical fashion, no impingement will occur.

P13-925 ACL reconstruction in patients above 50 years - results and factors influencing the outcome Joseph C.J.1, Balasubramanian P.1, Rajan D.1 1 Ortho One, Coimbatore, India Objectives: To assess the usefulness of ACL reconstruction in patients above 50 years and factors influencing the outcome. Methods: We present the results of ACL reconstruction in 38 patients (30 males, 8 females) who were above 50 years with an average followup of 22 months. From 2004 to 2009, 71 patients above 50 years were operated in our centre for ACL reconstruction. but unfortunately 38 were available for follow up and were included for the study. The mean age of the study group is 53.1 years (Range 50 - 64 years). The mechanisms of injury were motorized two wheeler accidents (52.1%) Sports (12.7%) and others (35.2%) like jumping landing, twisting injuries and falls. Patients with gross osteoarthritic changes with malalignment were advised High tibial osteotomy and were excluded from the study. It was hypothesized that late presentation for treatment, meniscal injury and cartilage damage can result in poorer outcomes. All patients underwent ACL reconstruction using a quadrupled hamstring graft. Patients with gross osteoarthritic changes with malalignment were advised High tibial osteotomy and were excluded from the study. It was hypothesized that late presentation for treatment, meniscal injury and cartilage damage can result in poorer outcomes. All the patients were evaluated preoperatively and postoperatively with Lysholm score, Oxford knee score and Tegner activity levels. Results: Postoperatively the mean Lysholm scores improved from 49.3 to 90.5 and the mean Oxford score increased from 27.5 to 53.4 and Tegner activity level improved from 1.9 to 4.2. Statistical analysis was done using paired t test, which showed statistically significant improvement in the patients according to both the scores (P \ .0001). According to Lysholm score, excellent results were seen in 68.4%, good results in 13.2%, Fair results in 10.5% and poor results in 7.9%. Cartilage injuries in the medial joint and patellofemoral joint were more prevalent in those with poor and fair results. No patient had instability in the postoperative period. Though ACL deficient knees develop degenerative changes with time, the time of presentation did not play a crucial role in the outcome in our study. Except for one patient, no one had radiological evidence of worsening of osteoarthritis. Conclusions: ACL reconstruction in young and active individuals is widely accepted. There is still reluctance among surgeons for performing ACL reconstruction in elderly because of the concerns of worsening of

123

S194 osteoarthritis, arthrofibrosis and concerns about the usefulness of this procedure in this age group. In India, the patients present very late because of delay in diagnosis and availability of arthroscopic services and many injuries are due to motorbike accidents which is the predominant personal transport in the country. We conclude that ACL reconstruction is a very useful procedure in elderly patients who want to pursue active life provided the knee does not have gross osteoarthritis or malalignment.

P13-926 The functional outcome of ACL reconstruction in professional sportsmen Carmont M.1, Ennis O.1, Rees D.1 1 The Robert Jones & Agnes Hunt Hospital, Oswestry, United Kingdom Objectives: We reviewed professional sportsmen who had undertaken ACL reconstruction and associated surgery to determine their actual and perceived sporting performance, sporting career and long term outcome following surgery. The specific aim was to determine whether the players returned to the same or a better standard of play, following reconstructive ACL surgery. Methods: Professional sportsmen who had had ACL reconstruction surgery were identified on the Sports Injury Surgery database. After Research and Ethics committee approval a questionnaire survey was distributed to 55 players. Results: Twenty four players returned questionnaires (response rate of 43.6%). The dates of surgery ranged from January 1998-February 2006. The mean elapsed time following surgery was 48 months (range 13-120 months). Twelve patients had injured their left knee, 8 their right and 4 had injured both knees. The respondents played rugby league 37% (9), soccer 33.3% (8), rugby union 21% (5) and netball 8.3% (2). Eight respondents were currently playing at Super League standard and 4 within the Premiership. Fifteen were regular first team players, 6 were squad players and 2 did not know. When asked if their knee returned to normal 71% (17) replied yes, 25% (6) replied no and 1 respondent did not know. 62.5% (15) thought they had returned to their previous standard of play, 29% (7) said that they had not and 2 did not know. The mean time taken to return to play was 10 months (5-21 months) and 1 player retired immediately following reconstruction. 41.6% (10) sustained further injuries to their knee. 4 respondents had torn the ACL in either the opposite knee or re-ruptured their reconstruction. The age at which players are injured would appear to be an important factor with players who returned to the same standard being younger (21yrs) compared to those who did not (25yrs) although this was not statistically significant (P=0.108). In this series additional meniscal injuries within the knee did not influence outcome and at 4 years following injuries most players will have no or only slight symptoms with sport or activities of daily living. Conclusions: The rupture of the anterior cruciate ligament is no longer a career ending injury for the professional sportsman. We present our experience in the management and outcome of Anterior Cruciate Ligament injuries in professional sportsmen. The majority (62.5%) of players will return to their pre-injury standard of play following reconstruction. The age at injury and additional meniscal injuries were not shown to be a significant in this series but may shown to be important factors influencing return to the same standard of play.

P13-931 ACL reconstruction with DGST: a long term follow-up study of a comprehensive approach Monaco E.1, Luzon D.2, Giannetti S.3, Caperna L.3, Conteduca F.1, Ferretti A.1 1 St.Andrea Hospital, University of Rome ‘La Sapienza’, Rome, Italy, 2 Ospedale Israelitico, Rome, Italy, 3Sant’Andrea Hospital, Sapienza University of Rome, Rome, Italy Objectives: The anterior cruciate ligament (ACL) is one of the most frequently injured ligaments in the human body. This paper reports the results

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 of our comprehensive approach to ACL tears and knee instability, based on 30 years of experience on ACL reconstruction with hamstrings, that is based on several cornerstones: the use of doubled semitendinosus and gracilis as a free graft, the use of very strong and stiff fixation devices, the careful examination and repair or reconstruction of the lateral compartment in selected cases, and the use of unaggressive rehabilitation. Methods: Between March 2002 and January 2003, 100 consecutive patients who underwent an ACL reconstruction in our hospital using a hamstring tendon autograft, were selected for this prospective study and re-evaluated at a minimum of six years post-operatively. We used an extra-articular tenodesis in case of severe rotatory instability with a pivot shift 3? and in high risk athletes (23 patients). At final follow-up, eight patients (9%) were lost and eighty patients (80%) returned for an evaluation that included a thorough medical history and physical examination, a radiographic examination and a KT-1000 arthrometric test. The follow-up examination also included calculation of subjective functional Tegner and Lysholm scores and subjective and objective IKDC scores (91 patients). Radiological evaluation was done to evaluate degenerative signs and bone tunnel enlargement using standard AP and lateral view. Results: After 6 years, the International Knee Documentation Committee score demonstrated good or excellent results (A and B) in 98% of patients. Instrumental laxity testing showed a mean KT-1000 Manual Maximum side to side (S/S) difference of 2.5 mm (range: 0-8 mm); 58 patients (72.5%) demonstrated an anterior laxity difference of less than 3 mm, 16 patients (20%) showed a difference between 3 and 5 mm and 6 patients (7.5%) had more than 5 mm compared to the uninvolved contralateral knee. The mean Tegner activity score was 5.4, the mean Lysholm score was 95.7 and the mean subjective IKDC score was 93.7. Radiographic evaluation demonstrated early signs of osteoarthritis in 9% of patients and no signs of tunnel enlargement. There were no intra or post operative complications during the inpatient period. In no cases were revision or other surgeries performed in the follow-up period. Conclusions: The proposed approach demonstrates highly satisfactory results. Although we found the worst objective results at IKDC evaluation (p=0.01) and a trend toward worst arthrometric results (p=0.07) in females, when the extra-articular tenodesis was performed, all female patients resulted in group A at the IKDC. Moreover, when considering as a biomechanical failure a pivot-shift 2? or 3? and/or a KT-1000 S/S difference of more than 5 mm, we reported 6/80 failure (rate:7.5%). However, it is important to point out that, despite the lesser preoperative instability, all biomechanical failures occurred in patients operated without the extra-articular tenodesis.

P13-936 Biomechanical evaluation of different anterior cruciate ligament fixation techniques for DGST graft Monaco E.1, Labianca L.1, Camillieri G.2, Agro` A.1, D’arrigo C.1, Speranza A.1, Ferretti A.1 1 St.Andrea Hospital, University of Rome ‘La Sapienza’, Rome, Italy, 2 University of L’Aquila - University of Rome, II Faculty, Roma, Italy Objectives: A number of different anterior cruciate ligament (ACL) fixation techniques are currently in use. Slippage or failure of the graft by excessive loading may results in an unstable knee. Load and slippage of ACL graft varies by fixation technique used. Methods: graft slippage, load to failure and stiffness was evaluated in 6 different soft tissue anterior cruciate ligament fixation techniques and using a bone cement as a fixation device. Group A: Endo Button CL-Bio RCI; Group B: Swing Bridge-Evolgate; Group C: Transfix-Intrafix; Group D: Bone Mulch-Washer Lock; Group E: Rigidfix-Retroscrew; Group F: Rigidfix-Deltascrew; Group G: Kryptonite bone cement. Maximum failure load, stiffness, slippage at 1st and 1000th cycle and mode of failure were evaluated.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: The maximum failure load is significantly higher in group B (1030N), while is significantly lower in group E (483N) than others. The stiffness of the group B (270N/mm) is significantly higher than others. Mode of failure: group C has shown a failure in the femoral side in all test (4 device ruptures and 2 tendon ruptures). All but one failures of the other groups occurred at tibial side. All failure in group G occurred for slippage of the tendons. Conclusions: load to failure and stiffness was significantly different between anterior cruciate ligament fixation techniques. Al but one of fixation techniques showed sufficient properties for an adequate postoperative rehabilitation. Bone cement used as a fixation device in soft tissue graft doesn’t seem to provide adequate initial fixation properties suitable for early rehabilitation in ACL reconstruction.

P13-942 Navigated knee kinematic after tear of the ACL and its secondary restrains Monaco E.1, Labianca L.1, Maestri B.1, Speranza A.1, D’arrigo C.1, Ferretti A.1 1 St.Andrea Hospital, University of Rome ‘La Sapienza’, Rome, Italy Objectives: In this study we evaluated the role of ACL and its secondary restrain in controlling knee stability using a navigation system. It is well known that the pivot shift is related to an ACL tear but it is significantly increased by a tear of the anterolateral tibiofibular ligament (ALTFL). The purpose of this study was to evaluate the kinematics of the knee in different condition of instability. Methods: Anterior tibial translation and rotation were measured with a computer navigation system in 8 fresh-frozen cadaveric knees by use of a manual maximum load. Anterior translation was evaluated at 30,60 and 90 of flexion; rotation (internal, external and total) at 0,15,30,45,60 and 90. Different conditions were tested: ACL intact, after transection of the PL bundle, after transection of the AM bundle and then after lesion of the ALTFL. Results: Cutting the PL bundle does not increased anterior translation and rotation of the knee. Cutting the AM bundle significantly increased the AP translation at 30 and 60(p=0.01), but does not increased rotation of the knee. Cutting the ALTFL increased anterior translation at 60 and rotation at 30, 45 and 60(p=0.03) producing a Pivot-shift clinically detectable. Conclusions: The PL bundle does not affect anterior translation and rotation of the knee. The AM bundle is the primary restrain of the anterior translation but does not affect rotation of the knee. The lateral compartment become the primary restrain of rotation after cutting the ACL. The primary kinematic effect of an ACL injury is an increase in anterior tibial translation, but there is no significant change in maximum internal and external rotation. The lesion of the ALTFL increases tibial rotation and it is strongly correlated with the pivot shift phenomenon.

P13-944 Instrumented measurements of knee laxity: KT1000 vs navigation Monaco E.1, Maestri B.1, Labianca L.1, Speranza A.1, De Carli A.1, Ferretti A.1 1 St.Andrea Hospital, University of Rome ‘La Sapienza’, Rome, Italy Objectives: The KT1000 is widely accepted as a tool for the instrumented measurement of the anteroposterior tibial translation. The aim of this study is to compare the data obtained with the KT1000 in ACL deficient knees with the data obtained using a navigation system during ‘‘in vivo’’ ACL reconstruction procedures and to validate the accuracy of the KT1000. Methods: An ACL reconstruction was performed using computer aided surgical navigation in thirty patients. Anteroposterior laxity

S195 measurements were obtained for all patients using KT1000 arthrometer (in a conscious state and under general anesthesia) and during surgery (under general anesthesia) using the navigation system, always at 30 of knee flexion. Results: The mean AP displacement of the injured knees, measured by the KT1000, was 14±4 mm and 15,6±3,8 mm in a conscious state and under general anesthesia, respectively. The mean AP displacement of the uninjured knees, measured by the KT1000, was 8,5±2,42 mm and 9±2,32 mm in a conscious state and under general anesthesia, respectively. The mean AP displacement, measured by the navigation system under general anesthesia, was 16,1±3,7 mm. The mean S/S difference, calculated with the KT 1000 was 8,5±3,1mm and 8,5±3,25mm in a conscious state and under general anesthesia respectively. Values of AP translation calculated with the KT1000 were significantly higher and different in the injured knee compared to the uninjured knee in both the conscious and unconscious state. Measurements with the KT1000 under general anesthesia were significantly higher and different to those obtained in a conscious state (p=0,02), while they were not different to the measurements obtained ‘‘in vivo’’ with the navigator (p=0,37). No differences were found in the amount of side to side difference using KT1000 in a conscious state and under general anesthesia (p=0.02). Conclusions: In conclusion this study validates the accuracy of the KT1000 to exactly calculate AP translation of the tibia, in comparison with the more accurate measurements obtained using a navigation system. KT-1000 values must always evaluated as a side to side difference, both in conscious state and under general anesthesia. Minimizing the sources of mistake the KT1000 arthrometer has a great potential for clinic use in the diagnosis of ACL tear.

P13-964 The effect of tunnel position on in-situ force vector direction in single bundle and double bundle ACL reconstruction Kramer S.1, Smolinski P.1, Kato Y.1, Linde-Rosen M.1, Lertwanich P.1, Maeyama A.1, Ingham S.2, Fu F.2 1 University of Pittsburgh, Pittsburgh, United States, 2University of Pittsburgh, Dept. of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States Objectives: To fully restore native ACL functionality, the in-situ force vectors of replacement grafts should match those of the native ACL not only in magnitude but also in direction. The purpose of this study is to investigate the direction of the in-situ force vectors in four single bundle ACL reconstruction techniques with different tunnel positions and double bundle ACL reconstruction and compare those to that of the native ACL. Methods: Sixteen fresh frozen cadaveric knees were used. Eight (n = 8) were used for the following techniques: 1. Anteromedial bundle reconstruction (AM-AM), 2. Posterolateral bundle reconstruction (PL-PL), 3. Classical vertical SB reconstruction (PL-HighAM), 4. Double bundle reconstruction (DB). The other eight (n = 8) knees were used for anatomical middle position SB reconstruction (MID-MID). A robotic/universal force-moment sensor testing system was used. An anterior load of 89N was applied (KT) to the intact knee and the anterior tibial translation (ATT) was recorded at 0, 15, 30, 60 and 90 degrees of flexion. An in-situ force vector composed of 3-D force components in the anterior-posterior, medial-lateral, and superiorinferior directions was then determined. A similar protocol was used for each of the five reconstructed techniques. Once the intact and replacement ACL in-situ force vectors were determined, the 3-D angle between the two vectors, termed here the deviation angle, was determined and illustrated in Figure 1.

123

S196

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 and mechanized pivot shift testing. An intraclass correlation coefficient (ICC) was used to determine variability within each knee at each testing condition. Significance was set at a p value of \ 0.05. Results: The anterior tibial translation recorded with manual pivot shift testing (11.7 ± 2.6 mm) was significantly higher than anterior tibial translation recorded with the pivot shifter (7.4 ± 2.5 mm; p\0.05). Total internal/external tibial rotation with the manual pivot shift testing (18.6 ± 5.4) was also significantly higher than with the pivot shifter (11.0 ± 2.3; p\0.05). The intraclass ICC for translations was 0.76 for manual pivot shift and 0.92 for the pivot shifter. The intraclass ICC for rotations was 0.89 for manual pivot shift and 0.82 for the pivot shifter. Conclusions: This study introduced a new device for mechanized pivot shift testing. Although recorded translations and rotations with the pivot shifter were lower than with manual testing, the clinical advantage of mechanized pivot shift testing is a more repeatable measurement of anterior tibial translations when compared to manual pivot shift testing. This new device may increase consistency of clinical grading of the pivot shift test.

in-situ force vector and deviation angle Results: At 0 of knee flexion, the deviation angle of in-situ force vector for DB reconstruction was significantly lower than PL-PL reconstruction (p=0.018). At 15 of knee flexion, the deviation angle for PL-HighAM was significantly smaller than that for AM-AM reconstruction (p=0.018). At 30 knee flexion, the deviation angle for PL-HighAM was significantly lower than PL-PL and AM-AM reconstruction. At 60 knee flexion, deviation angle for PL-HighAM, DB, and MID-MID were all significantly smaller than PL-PL (p=0.018). At 90 knee flexion deviation angle for PLHighAM, DB, and MID-MID were all significantly lower than PL-PL and PL-HighAM was also significantly lower than AM-AM (p=0.018). Conclusions: In the present study, a deviation angle from the intact in-situ force vector was used as a quantitative measure to compare the effectiveness of various ACL reconstruction techniques. It is important to note however, the in-situ force vector has a magnitude and direction, and magnitude was not taken into account. Also, the present study only determined deviation angles for simulated KT. Although, the ACL’s primary role is to resist ATT, recent studies have shown that the ACL also may provide rotationally stability. In the future it may be of interest to examine the in-situ force vectors for simulated pivot shift and determine which tunnel position compares best with the intact ACL.

P13-966 Repeatability of manual vs. mechanized pivot shift testing Musahl V.1, Voos J.2, Kendoff D.2, O’Loughlin P.2, Pearle A.2 1 University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States, 2 Hospital for Special Surgery, Orthopaedic Surgery, New York, United States Objectives: The pivot shift test is a complex clinical test that is highly dependent on the examiner’s skill and expertise. The objective of this study was to design a navigated mechanized pivot shift test (pivot shifter) and evaluate the repeatability of the pivot shifter in the ACL deficient knee. It was hypothesized that translations and rotations measured with the pivot shifter will be more repeatable when compared to those obtained with a manual pivot shift. Methods: Twelve fresh frozen cadaveric hip-to-toe whole lower extremities were used for this study. The ACL was transected and a manual pivot shift test was performed by three different examiners. A navigation system simultaneously recorded anterior tibial translation and external/internal tibial rotation. The mechanized pivot shift test was then recorded utilizing the pivot shifter. The pivot shifter consists of a continuous passive motion (CPM) machine and a custom made foot holder to allow for application of internal rotation moments at the knee. Valgus moments were achieved by a post. The tibia was supported while the femur was free. The pivot shifter ranged the knee from full extension to 90 of knee flexion while the navigation system simultaneously recorded kinematics. The mechanized pivot shift was repeated three times. Repeated measures ANOVA was used to compare manual

123

P13-968 Objective knee laxity measurements after reconstruction of the anterior cruciate ligament Lorbach O.1, Kieb M.2, Brogard P.2, Pape D.2, Maas S.3, Seil R.2 1 Klinikum Osnabru¨ck, Orthopedic and Trauma Surgery, Osnabru¨ck, Germany, 2Centre Hospitalier de Luxembourg - Clinique d’Eich, Orthopedic and Trauma Surgery, Luxembourg, Luxembourg, 3Universite´ du Luxembourg, Faculte´ des Sciences, de la Technologie et de la Communic., Luxembourg, Luxembourg Objectives: The purpose of the present study was to introduce two objectives and non-invasive measurement systems in order to objectively measure anteroposterior and rotational laxity after ACL reconstruction. Methods: Fifty-two consecutive patients were reviewed after reconstruction of the ACL using bone-patellar-tendon-bone autografts. At a mean follow-up of 17 months, ap laxity was tested using a new objective and non-invasive knee measurement system (Genurob) with an applied pressure of 67N, 89N and 134N. The rotational laxity was measured using the Rotameter measurement device with an applied torque of 5,8 and 10Nm. The measured results were compared with the measurements of the healthy contralateral knees of the patients.Tegner activity system, Lysholm score and IKDC score were used in order to correlate the clinical outcome with the measured laxity. Results: Pivot shift test was negative (33) or glide (16) in 49 patients, Lachman tests were negative in 50 cases. Subjective assessment of the IKDC score showed A in 44 cases, 6 cases had a B and 2 patients showed a C. Mean Lysholm score was 94.5±9.5, mean Tegner activity level 6.5±1.4 preoperative and 5.9±1.8 at follow up (p=.003). For the operated knee, the genurob measured a mean of 2.1±.7mm at 67N, 3.1±.9mm at 89N and 4,9±1.2mm at 134N. For the contralateral nonoperated side, 1.5±.7mm at 67N, 2.2±.8mm at 89N and 3.6±1.0mm at 134N were measured. The measured side-to-side differences (.6-1.3mm) were significant (p\.0001), however, they were within the 2mm interval which is considered to be normal in the IKDC. At 5Nm of applied torque, 18.2±9.3 were measured for the internal rotation, 31.5±9.4 for the external rotation and 49.8±11.1 for the entire rotational range in the unaffected knee. In the operated knee, 16.6±8.9 of internal rotation, 33.2±9.5 of external rotation and 50±12.1 of entire rotational range were measured. At 8Nm of torque, 28.4±10.6 were found for the internal rotation, 43.9±9.6 for the external rotation and 72.4±13.7 for the entire rotational range. In the surgically treated knee, 26.4±10.1 for the internal rotation, 45±10.5 for the external rotation and 71.4±13.7 for the rotational range were measured. At highest applied torque of 10Nm, 34.5±11.9 were measured in the nonaffected side for the internal rotation, 50.9±10.2 for the external rotation and 84.7±16.2 for the entire rotational range. In the operated knee, 32.2±10.9 were measured for the internal rotation, 51.9±11.4 for the external rotation and 84±15.9 for the entire rotational range.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 No significant differences were seen in tibiofemoral rotation between the operated and non-operated knee at all applied torques (p[.05). Conclusions: Knee laxity can be objectively examined after ACL surgery using the introduced measurement systems. Furthermore, anteroposterior and rotational knee laxity were adequately restored after single-bundle ACL reconstruction.

P13-974 Transtibial double bundle with ST/G vs transtibial single bundle technique with BPTB: two years of follow up Cervellin M.1, de Girolamo L.2, Bait C.1, Melegati G.3, Denti M.1, Volpi P.1 1 Galeazzi Orthopaedic Institute, Sport Traumatology and Arthroscopic Unit, Milan, Italy, 2Galeazzi Orthopaedic Institute, Orthopaedic Biotechnologies Lab, Milan, Italy, 3Galeazzi Orthopaedic Institute, Centre for Sports Rehabilitation, Milan, Italy Objectives: The single-bundle ACL reconstruction ensures good outcomes and it is a well-established and widespread technique. Nevertheless some patients still present pain and instability. Some recent studies have added that the double-bundle technique restores a better natural ACL-fitting kinematics compared to the single-bundle technique. Long-term clinical studies to compare the two surgical techniques are not frequent and there is no instrument to evaluate functionality and kinematics during the knee rotation in vivo. The ACL lesion is one of the most frequent events in sport injuries among young sportive people. In this randomized prospective study we compare the most common surgical technique practiced in this patient’s category (ACL reconstruction with BPTB) with ACL double bundle reconstruction technique. Double bundle is supposed to allow an antero-posterior stability as the graft of the patellar tendon, besides to enable a better control of rotations. Methods: From 2002 to 2006, 40 patients were selected and treated for the reconstruction of the anterior cruciate ligament and then randomly assigned to two groups: SB group (Single Bundle) - ACL transtibial single-bundle reconstruction with autologous BPTB graft; DB group (Double Bundle) - ACL 4 tunnels transtibial double-bundle reconstruction with autologous ST and GR grafts duplicated pro AM and PL bundles, respectively. Results: Between the two groups no significant differences regarding age, associated lesions, follow up and pre-operative Lysholm and Tegner scoring scales have been detected (all p\0.05). Comparing the two groups, no statistically significant difference regarding the post-operative Lysholm score (p=0,368) the Tegner activity scale (p=0,519) and the arthrometric evaluation with KT-1000 (p=0,74) were observed. On the contrary, the IKDC evaluation showed a statistically significant difference (p=0,004) between the two groups, with better results in the DB group. Conclusions: Our data seems to suggest that the double bundle ST/G ACL reconstruction technique allow to obtain slightly better outcomes than the traditional technique with single bundle BPTB. Therefore, DB ACL reconstruction could be used in selected patients, requiring higher performances.

P13-975 The influence of the anteromedial and posterolateral bundles of the anterior cruciate ligament on external and internal tibiofemoral rotation Lorbach O.1, Pape D.2, Zerbe T.3, Maas S.4, Kohn D.3, Seil R.2 1 Klinikum Osnabru¨ck, Orthopedic and Trauma Surgery, Osnabru¨ck, Germany, 2Centre Hospitalier de Luxembourg - Clinique d’Eich, Orthopedic and Trauma Surgery, Luxembourg, Luxembourg, 3Saarland University Medical Center, Department of Orthopaedic Surgery, Homburg/Saar, Germany, 4Universite´ du Luxembourg, Faculte´ des Sciences, de la Technologie et de la Communic., Luxembourg, Luxembourg Objectives: The influence of the anteromedial and posterolateral bundles of the anterior cruciate ligament (ACL) on tibiofemoral rotation might be

S197 of great value to detect ACL injuries and investigate the postoperative restoration of the rotational stability. A newly developed clinical measurement device (Rotameter) is introduced that allows for objective evaluation of knee rotation. Methods: Tibiofemoral rotation was measured in 20 human cadaveric knees using a non-invasive external measurement device (Rotameter) and a knee navigation system. The measurements of the knees with the intact ACL were compared with the measurements after isolated resection of the posterolateral bundle and after a complete resection of the anterior cruciate ligament at an applied torque of 5,10 and 15 Nm. Statistical analysis was made using analysis of variance and posthoc Scheffe test. Pearson0 s coefficient was used to compare both measurement techniques. Results: In comparison with the measurements of the knees the intact ACL with the knees after isolated resection of the posterolateral bundle showed significant increase of tibiofemoral rotation at almost all applied torques (p\.05). Total resection of the ACL also produced significant increases compared with the intact ACL at torques of 5, 10 and 15 Nm as measured by the Rotameter. (p\0.05). Total resection of the ACL yielded further increases in rotation compared with PL resection alone, but these differences were not significant. The results of the knee navigation system confirmed the measured results of the Rotameter. Comparison of the two measurement methods revealed a high correlation at all applied torques, with Pearson correlation coefficients ranging from .85 to .95. Conclusions: The anterior cruciate ligament and especially the posterolateral bundle of the ACL has a significant impact on isolated tibiofemoral rotation. Therefore, the developed non-invasive device might be of great importance to investigate the status and the postoperative reconstruction of the ACL in the clinical setting.

P13-977 In vivo kinematic evaluation of rotational stability in single bundle and double bundle ACL reconstruction at two years follow-up Quaglia A.1, de Girolamo L.2, Cervellin M.1, Champlon C.3, Tornese D.4, Denti M.1, Volpi P.1 1 Galeazzi Orthopaedic Institute, Sport Traumatology and Arthroscopic Unit, Milan, Italy, 2Galeazzi Orthopaedic Institute, Orthopaedic Biotechnologies Lab, Milan, Italy, 3Galeazzi Orthopaedic Institute, Movement Analysis Lab, Milan, Italy, 4Galeazzi Orthopaedic Institute, Centre for Sports Rehabilitation, Milan, Italy Objectives: Single bundle ACL reconstruction allow to restore the anteroposterior stability of the knee, but it does not re-establish the normal tibial rotation. Several cadaveric studies seem to demonstrate the need of a postero-lateral bundle in order to achieve a good rotational stability. The aim of our study is to evaluate the tibial rotation after double bundle ACL reconstruction and comparing it with the one after single bundle reconstruction and in healthy controls. Methods: The rotational stability was assessed by the EL.I.TE. System (BTS, Milano, Italy) on 2 different groups of patients at 2-years follow up: 10 patients treated with double bundle ACL reconstruction (DB group); 10 patients treated with single bundle ACL reconstruction (SB group). Data were compared with those from 10 healthy patients (control group). Transtibial single bundle reconstruction was performed with autologous ST-GR grafts, whereas double bundle reconstruction with autologous ST and GR grafts duplicated pro AM and PL bundles, respectively. All ACL reconstructions were performed by the same surgeon. All patients were asked to perform a specific movement of ‘‘flexion and change of directions’’ to reproduce pivot shift. Kinematic parameters were acquired by 6 optoelectronic camera, recording movements of 20 reflective markers placed on selected bony landmarks in according to Davis protocol. All patients activities were recorded 3 times. Results: The average age of patients was 33.4±10.9 in group SB (6 male, 4 female), 36.9±12.3 in group DB (6 male, 4 female) and 30.1±4.9 in control group (8 male, 2 female), with a mean BMI of 21.44±8.0, 24.6±2.6 and 22.1±2.1, respectively, without any significant differences among groups. The EL.I.TE. system allowed us to assess internal tibial rotation in relation with flexion degrees during pivoting task. Preliminary

123

S198 results show that both techniques were not able to restore an comparable internal tibial rotation to the one of healthy knees. However, data provided by this kind of analysis showed that DB ACL reconstruction allow lightly better results than SB reconstruction in restoring internal tibial rotation. Conclusions: Although further studies are required to confirm these results, this study seem to suggest DB ACL reconstruction could be used in selected patients, requiring higher performances.

P13-983 The effect of anterior cruciate ligament reconstruction on the quality of life, activity level and bone mineral areal mass Stener S.1, Kartus J.1, Ejerhed L.1 1 NU-Hospital Organization, Department of Orthopaedics, Uddevalla, Sweden Objectives: Anterior cruciate ligament (ACL) reconstruction and its significance for quality of life, activity level and the influence on bone mineral areal mass (BMA) has previously not been thoroughly investigated. Methods: Patients with ACL ruptures scheduled for reconstruction were prospectively included in the study. The quality of life was estimated using Euroqol (5-15), 5 being the best and 15 the worst, activity level was measured using Tegner activity level (0-10) and BMA was measured in the hip, lumbar spine and both calcanei using the dual energy X-ray absorptiometry (DXA) technique. The patients were assessed before surgery and after six, 18 and 36 months. Results: 35 patients (12 female and 23 male) have been followed for three years. The quality of life according to Euroqol was 7 (5-11) before surgery and 5 (5-7) after 36 months. The Tegner activity level before injury was 9 (1-10), 2 (0-9) preoperatively, 4 (0-10) at six months and 5 (2-7) at 36 months. The BMA in the calcanei decreased in both female and male patients. The female patients had lost 12% (p=0.003 v preoperatively) and the male patients 4% (p=0.0005 v preoperatively). The decrease in BMA in lumbar spine and hips were only significant for male patients, but less than 2% after three years. Conclusions: ACL reconstruction rendered a better quality of life. The patients could increase their activity level, but female patients had significant decrease in BMA in both calcanei. The surgical trauma induced the BMA decrease. The decrease in BMA in the hips and the spine corresponded to the expected age related decrease.

P13-989 Differentiation between intra- and postoperative bone tunnel widening and communication in double bundle anterior cruciate ligament reconstruction. A prospective study Siebold R.1, Cafaltzis K.2 1 ATOS Praxisklinik Heidelberg, Abt. Knie- u. Fußchirurgie, Heidelberg, Germany, 2Universita¨tsklinik Mannheim, Orthopa¨die und Unfallchirurgie, Mannheim, Germany Objectives: The purpose of this study was to determine the amount of anteromedial (AM) and posterolateral (PL) bone tunnel widening and communication following anatomic four tunnel double bundle (DB) anterior cruciate ligament (ACL) reconstruction. Methods: 24 consecutive patients undergoing anatomical four tunnel DB ACL reconstruction with hamstrings and extracortical fixation were included in a prospective case series. MRI scans were performed directly post-op and at 7 months follow-up to access intra- and postoperative bone tunnel widening and communication. It was determined in different planes by digitally measuring the diameters of the AM and PL tunnels. Results: Intraoperative bone tunnel communication of AM with PL by drilling was observed in 23.8% on the tibia. 7 months postoperatively significant bone tunnel widening was found in all four bone tunnels. It was an average of 20% for the tibial AM-, 38% for the tibial PL-, 34% for the femoral AM- and 46% for the femoral PL bone tunnel. 19% of patients demonstrated tibial or femoral postoperative bone tunnel communication of AM and PL caused by tunnel widening. Conclusions: We observed significant postoperative bone tunnel widening in anatomic four tunnel DB ACL reconstruction. It was the highest for

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 both PL bone tunnels. Intraoperative bone tunnel communication occurred tibially and was caused by drilling. Postoperative bone tunnel communication was present in 19% of patients on the tibial and femoral side and was caused by bone tunnel widening.

P13-1003 Tibial rotation after ACL reconstruction is restored with a single bundle hamstring graft Barenius B.1, Feller J.2, Webster K.3 1 Karolinska Institutet, Department of Clinical Science and Education, So¨dersjukhuset, Stockholm, Sweden, 2La Trobe University Musculoskeletal Research Center, Melbourne, Australia, 3La Trobe University Musculoskeletal Research Center, Faculty of Health Sciences, Bundoora, Australia Objectives: It has been suggested that excessive tibial rotation during pivoting tasks is not controlled by single bundle anterior cruciate ligament (ACL) reconstruction. This may in part be explained by the angle of the graft in the coronal plane, with a more vertically placed graft being less able to control rotation. The purpose of this study was to assess tibial rotation after ACL reconstruction with a more obliquely placed graft and to assess if there were any differences between reconstructions with semitendinosus alone (ST) or semitendinosus and gracilis (STGR) tendons. Methods: 20 patients were evaluated. All patients had undergone a primary ACL reconstruction using a quadrupled hamstring tendon graft (10 ST and 10 STGR) for an isolated ACL injury within 6 months of injury. There were 2 females and 8 males in each group. All patients were at least 2 years from surgery (ST: mean 35 months, STGR mean 37 months) and all had made a good functional recovery and returned to their pre-injury sporting activities. Evaluation consisted of IKDC 2000, instrumented laxity testing with KT 1000, and 3D motion analysis to record tibial rotation when the subjects descended stairs and pivoted 90 degrees on landing (using a similar protocol to one which has previously been reported in the literature). Results: All patients had made an excellent recovery (IKDC scores [ 90) and there were no significant differences between the ST and STGR subjects for any of the background variables including anterior knee laxity. There were no differences in tibial rotation between the operated (mean: 21, SD 5,3) and non operated limb (mean: 21, SD 7,6), although there was a tendency towards less variance in the operated limb. There was no significant difference between the graft types (ST: 20, STGR: 21). Conclusions: Contrary to previous reports, we found restoration of normal tibial rotation during a pivoting task after a single bundle ACL reconstruction technique. This may reflect a more horizontal graft orientation in the coronal plane for patients in the current study. The lack of difference between the ST and STGR groups suggests that this restoration of normal tibial rotation is due to static rather than dynamic restraints.

P13-1006 I-Space, a reliable 3-dimensional radiological measurement of anterior cruciate ligament reconstruction Meuffels D.1, Potters J.-W.1, Verhaar J.1, Reijman M.1 1 Erasmus MC, Orthopedic Department, Rotterdam, Netherlands Objectives: Accurate and reliable visualization of anterior cruciate ligament (ACL) reconstruction is essential. A three-dimensional viewing and measurement method was developed for a more accurate and better visualization of the ACL reconstruction. This method uses CT data and a 3D viewing chamber (I-Space). The primary objective of this study was to evaluate the reliability and accuracy of this method compared to existing methods. Methods: Fifty consecutive ACL reconstructions were visualized and measured post-operatively by standard radiographs, CT and CT with the ISpace. Tibial and femoral tunnel position were compared, by two observers, for visibility and measured using the method of Amis, Aglietti, Hoser and the I-Space.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: Tunnel visualization was possible in 50-82% of the radiographs versus 100% in the CT and I-Space. Using the intraclass correlation coefficient (ICC) and the Landis and Koch scaling, the inter- and intraobserver agreement was fair to moderate for the femoral and tibial radiographs. The CT showed higher ICCs compared to the radiographs. The femur and tibia CT inter- and intra observer agreement was moderate to substantial, with an ICC range of 0.49-0.76. The I-Space agreement was almost perfect with an ICC of 0.83 for the femur and 0.95 for the tibia. Conclusions: The I-Space reliability scores, according to Landis and Koch scaling, were almost perfect. The properties of I-Space give an extra dimension to measure, and create a perfect environment for performing scientific research and clinically evaluating of ACL reconstructions.

P13-1007 Transtibial ‘‘all inside’’ ACL reconstruction technique: preliminary results at one-year follow up Thiebat G.1, Quaglia A.1, Cervellin M.1, Galli M.1, de Girolamo L.2, Volpi P.1 1 Galeazzi Orthopaedic Institute, Sport Traumatology and Arthroscopic Unit, Milan, Italy, 2Galeazzi Orthopaedic Institute, Orthopaedic Biotechnologies Lab, Milan, Italy Objectives: Due to its mini-invasiveness, the transtibial ‘‘all inside’’ technique allows the anterior cruciate ligament reconstruction saving cortical bone, using just a semi-tendons (doubled, tripled, quadrupled). Moreover, a further advantage is the possibility to associate this method with other surgical interventions, like osteotomies. Thus, the ‘‘all inside’’ technique is particularly indicated in adolescents and in patients requiring other associated surgical procedures. The aim of our retrospective study was to compare the outcome of patients treated with the ACL transtibial single bundle reconstruction technique using ST with the transtibial ‘‘all inside’’ technique, at one year-follow up. Methods: From 2007 to 2008, 20 patients treated with the ACL transtibial single bundle reconstruction technique with ST/GR and 20 patients treated with the ACL transtibial ‘‘all inside’’ technique were examined at one year-follow up using Lysholm score, Tegner activity scale IKDC form and KT-1000 arthrometer. The surgeon, after having determined the graft diameter, prepared the tibial and femoral tunnels directly from inside the joint using specific instruments. The sleeve was then introduced in the intercondyloid notch from the anteromedial portal. The femoral fixation was performed using RetroButtonTM, whereas the distal tibial one was performed with SutureButtonTM (FiberwireTM). Results: No significant differences have been observed in IKDC, Lysholm and Tegner scores between the two groups of patients at one year-follow up. Moreover, in ‘‘all inside’’ group, the recovery time was reduced and patients experienced less pain due to the limited damages to the tissue. Conclusions: Although at an early follow up, our results seem to suggest that, in selected cases, the ‘‘all inside’’ technique could be an useful alternatives for surgeons for ACL reconstructions. Patients will be further monitored at longer follow up, in order to evaluate the clinical trend during time. Up to now very few studies regarding ACL ‘‘all inside’’ technique have been performed, and thus it would be necessary to realize perspective randomized study to validate the efficacy of this technique.

P13-1015 Stiffer fixation of the tibial double tunnel ACL complex versus the single tunnel - a biomechanical study Meuffels D.1, Docter P.1, van Dongen R.1, Kleinrensink G.-J.2, Verhaar J.1, Reijman M.1 1 Erasmus MC, Orthopedic Department, Rotterdam, Netherlands, 2Erasmus MC, Neurosciences, Rotterdam, Netherlands

S199 Objectives: The primary objective of this study was to evaluate the difference in graft pullout forces,stiffness and failure mode of double bundle anterior cruciate ligament (ACL) reconstruction of the tibial insertion using a single tunnel compared to a double tunnel technique with interference screw fixation. Methods: An ACL reconstruction on the tibial site was performed on 40 fresh frozen porcine knees (average bone mineral density of 0.64 measured by dual-energy x-ray absorptiometry scan), randomly assigned to the single or double tunnel group. Interference screw fixation of the soft tissue graft was used for both types of tibial reconstruction. Maximum failure load, stiffness and failure mode were recorded. Results: There was no significant difference in maximum failure load between the single (400? 26 N) and double tunnel group (440? 20 N). Stiffness of the tibial tunnel complex was significantly higher in the double tunnel group (76? 3 N/mm) compared to the single tunnel group (62? 4 N/mm) (P-value of 0.013). All but two (38 of 40) grafts failed by slippage of the tendon past the interference screw. Conclusions: Both groups were similar in maximum failure load and mode of failure on the tibial fixation side. However, there was a significantly stiffer fixation of the tibial double tunnel ACL complex versus the single tunnel.The present study shows a biomechanical advantage with no potential deleterious side-effects to fixate the ACL with a double tunnel technique on the tibial side.

P13-1030 Radiographic landmarks for tunnel positioning in double-bundle ACL reconstructions Johansen S.1, Pietrini S.2, Ziegler C.2, Westerhaus B.2, Wijdicks C.2, Anderson C.2, Engebretsen L.1, LaPrade R.2 1 Ulleva˚l University Hospital, Orthopaedic Center, Oslo, Norway, 2 University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States Objectives: The purpose of this study was to establish quantitative and qualitative radiographic landmarks for locating the femoral and tibial attachment sites of the anteromedial (AM) and posterolateral (PL) bundles of the anterior cruciate ligament (ACL). We hypothesized that this standardized protocol could be used to consistently and reproducibly describe the positions of the ACL bundle attachment sites in relation to osseous and soft-tissue landmarks, as well as reference lines projected onto radiographic images. Methods: Dissections were performed on 12 cadaveric knee specimens. The attachment sites of the AM and PL bundles and relevant landmarks were labeled with radio-opaque markers. The positions of the AM and PL bundle attachment centers relative to surrounding landmarks and superimposed reference lines were quantified on anterior-posterior, lateral, and axial radiographs. Measurements were performed twice by three independent examiners according to both novel and previously described methods. Intraobserver and interobserver reliability was determined using intraclass correlation coefficients (ICCs). Results: On the AP view of the femur, the AM bundle center was located 11.9 ± 2.2% of the way down from the maximum height of the intercondylar notch at 55 of flexion, whereas the position of the PL bundle center was 39.8 ± 6.7% from the top of the notch height. On the lateral view of the femur (Fig. 1A), the AM bundle attachment was located at 21.6 ± 5.6% of the sagittal diameter of the lateral femoral condyle along Blumensaat’s line (B) and 14.6 ± 7.7% of the maximum notch height (H); the center of the PL bundle was 28.9 ± 4.6% of distance B and 43.1 ± 6.0% of distance H. On the AP view of the tibia, the center of the AM bundle was located 55.8 ± 3.4% from the lateral aspect of the tibia along a line spanning the maximum coronal width of the tibia plateau, while the position of the PL bundle center was measured to be 50.0 ± 2.1%. On the lateral view of the tibia, the percentages of the AM and PL bundle attachments from the anterior tibial margin along the maximum sagittal tibial diameter were 36.3 ± 3.8% and 51.0 ± 4.0%, respectively. On the axial view of the tibia (Fig. 1B), the center of the AM bundle was positioned 52.0 ± 3.0% along the maximum coronal diameter (CD) of the

123

S200

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

axial section and 35.5 ± 6.6% along the maximum sagittal diameter (SD), while the PL bundle was located at 48.6 ± 2.7% of CD and 51.8 ± 4.0% of SD. ICCs for overall interobserver and intraobserver reliability were excellent (0.990 and 0.991, respectively).

Lateral knee radiograph of the femur (A); axial knee radiograph of the tibia (B) Conclusions: The proposed radiographic guidelines are reproducible and can be used to facilitate more accurate tunnel placement in double-bundle ACL reconstructions incorporating intraoperative fluoroscopic techniques, as well as to enhance pre-operative planning and post-operative evaluations of these procedures.

P13-1032 Effect of tunnel position on in-situ force vectors in ACL reconstruction Kramer S.1, Kato Y.1, Lertwanich P.1, Maeyama A.1, Linde-Rosen M.1, Ingham S.2, Fu F.2, Smolinski P.3 1 University of Pittsburgh, Pittsburgh, United States, 2University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States, 3University of Pittsburgh, Department of Mechanical Engineering, Pittsburgh, United States Objectives: Studies have compared the magnitude of the in-situ forces in ACL grafts to that of the native ACL. To restore native ACL function the direction of in-situ force is also important. This study investigated the direction of the in-situ force vectors in different ACL reconstructions with that of the native ACL. Methods: Sixteen fresh frozen cadaveric knees were used. Eight (n = 8) were used for ACL reconstructions: 1. Anteromedial bundle reconstruction (AM-AM), 2. Posterolateral bundle reconstruction (PL-PL), 3. Classical vertical SB reconstruction (PL-HighAM), 4. Double bundle reconstruction (DB) and 5. Anatomical middle position SB reconstruction (MID-MID). A robotic/universal force-moment sensor testing system was to measure the forces and displacements in 6 DOF. Under an anterior load of 89N (AT) to the knee, the resultant ACL/graft force was calculated. Results: Figures 1 and 2 show the average force in the tissue for the intact ACL and PL-High AM cases. A major difference that can be seen is that at lower flexion angles the native ACL provides a greater superior force than the graft. Also, at higher flexion angles there is a difference in the medial component.

123

Conclusions: This study considered the load carrying of ACL ligament/ graft in AT knee loading. In ACL reconstruction force direction in addition to magnitude is an important consideration. It is hypothesized that the force direction of the native ACL plays a role in knee function. The aim of reconstruction should be to restore the load carrying function of the tissue to the intact state.

P13-1033 Osteoarthritis 5 years after anterior cruciate ligament reconstructions - IKDC, Tegner and Lysholm Scores predict progression Ong K.L.1, Lee Y.H.D.1, Tan J.1, Chang H.C.2 1 Changi General Hospital, Singapore, Singapore, 2HC Chang Orthopaedic Surgery, Singapore, Singapore Objectives: The purpose of our study was to review the long-term progression of osteoarthritis in of Anterior Cruciate Ligament Reconstruction using Hamstring Grafts.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: Our patients had their surgeries performed in 2000 and 2001 for subacute or chronic tears of their anterior cruciate ligament. They were recalled to our clinic 5 years after their surgery for review. The outcomes of these patients were assessed with Tegner, Lysholm Scores as well as the International Knee Documentation Committee (IKDC) Form. A radiograph of the operated knee was performed and scored according to Kellgren and Lawrence grading (K& L). Results: 64 patients consented to return to for review 5 years after surgery. The average age of our patients was 25 years. The mean Lysholm score was 85.2 and the mean pre injury Tegner score was 7.28 and the post injury score was 5.58. The mean subjective IKDC score was79.49. 79.7 % of our patients had normal or nearly normal knees (IKDC A or B) with remaining 20.3% were IKDC grade C. The mean side-side difference for anterior translation using the KT-1000 arthrometer at maximal manual traction 1mm. Using the Kellgren and Lawrence grading for knee osteoarthritis 8 patients, 12.5% had Kellgren and Lawrence grade 1, 24(37.5%) grade 2, 27(42.2%) grade 3 and 5(7.8%) had Kellgren and Lawrence grade 4 evidence of osteoarthritis on Knee radiograph. We found that using the IKDC Objective scoring system, patients with an objective IKDC grade A had a median K& L score of 2, whereas objective IKDC grade B or C had a median K& L score of 3. The mean K&L score for patients with IKDC grade A was 2.17 B was 2.49 and C was 2.63. We found that the IKDC subjective score had a significant correlation with K&L score, ie the lower the IKDC subjective score the higher the K&L score (Pearsons p=0.029).We also found that the post operative Tegner score also had a significant correlation with K&L score (p=0.018). This was also true of the Lysholm score, which had a significant correlation with the K&L score (p=0.020). The Interobserver variability of the K and L score was analysed it showed good interobserver agreement with a kappa score of 0.700. Conclusions: Within this cohort of patients, we found that five years after surgery that the IKDC, Tegner and Lysholm can be used to predict development of osteoarthritis. There was signficant correlation between the subjective Knee scores (IKDC, Tegner and Lysholm) and radiographic (K&L) evidence of Osteoarthritis. The K&L grading system also demonstrated good interobserver variability.

P13-1062 Does medial collateral ligament reconstruction associated with ACL reconstruction decrease ROM and increase morbidity? Osti L.1, Papalia R.2, Del Buono A.2, Amato C.2, Maffulli N.3, Denaro V.4 1 Hesperia Hospital, Orthopedic Surgery, Modena, Italy, 2Campus Biomedico University, Orthopaedic Surgery, Rome, Italy, 3Keele University, Institute of Science and Technology in Medicine, Stoke on Trent, United Kingdom, 4Campus Bio-Medico University, Orthopaedic Surgery, Rome, Italy Objectives: We report the outcomes of a series of 22 athletes treated by combined anterior cruciate ligament (ACL) and medial collateral ligament (MCL) reconstruction for associated (ACL) and chronic grade 2 (MCL) tears. Methods: Surgery: single bundle ACL reconstruction and MCL repair in a Krackow et al configuration. We evaluated preoperatively and at a minimum 2 years follow-up the knee antero-medial stability (using Lachman test, pivot-shift test result, side to side KT-1000 arthrometer measures, IKDC, valgus stress X-Ray and Clinical valgus stress test) and functional ratings (Lysholm knee scale and return to sporting activities). Results: The athletes improved clinically significantly (P \.001) from baseline to latest follow-up and 20 of the 22 patients (90.9%) had returned to sport at pre-injury level without detection of knee stiffness or consistent ROM restriction. Conclusions: In case of ACL deficient athletes with complaints of associated chronic symptomatic valgus laxity, the simultaneous ACL-MCL reconstruction may be indicated to restore knee stability allowing return to pre-injury activity level without increasing the knee morbidity

S201 P13-1065 Effect of sequential cutting of the anterior cruciate ligament bundles on tibial rotation during the Lachman test Christel P.1, Akgun U.2, Demirel B.3, Karahan M.4, Tatar Y.5, Nuran R.6, Aydin A.T.7 1 Habib Medical Center Olaya, Sports Medicine, Riyadh, Saudi Arabia, 2 Marmara University, School of Athletic Education, Istanbul, Turkey, 3 Akdeniz University, Anatomy, Antalya, Turkey, 4Besiktas Istanbul, Turkey, 5Marmara University Physical Education and Sport Faculty, Health and Sports Department, Istanbul, Turkey, 6Orthopedics and Traumatology, Kadikoy, Istanbul, Turkey, 7Akdeniz University, Orthopaedics, Antalya, Turkey Objectives: It is known that internal rotation occurs during the Lachman test. Little is known about the contribution of each ACL bundle on the rotation which may occur. We hypothesise a progressive internal rotation increase during sequential cutting of the bundles with a larger contribution of the posterolateral bundle (PLB). Methods: Six fresh-frozen cadaveric knees were used. The collateral ligament attachments and the menisci were left intact. The patella, patellar ligament and quadriceps were excised. Eight screws were inserted in the distal femur and the proximal tibia and used as fixed landmarks for further digitization of the bones position. A metal hook was screwed in the centre of the tibial tubercle and later used to apply a 150N load to the tibia, directed from posterior to anterior, aimed at simulating the Lachman test. The knee received manual axial compression with 6.35 kg of force via a dynamometer during data collection. No torque was applied to the tibia. 3D kinematic data were collected with a MicroScribe G2X digitizer in full extension, at 30 of flexion and finally after a 150N load was applied to the tibial. For each specimen, the measurements were repeated 4 times for each test condition: intact ACL, sequential cutting of the ACL bundles, and after full cut of the ligament. The Surfcam Velocity II software allows calculating the 3-D position of the tibia with regard to the femur. The tibial rotation induced from the starting position was recorded and the differential values related to the ACL status were calculated. Results: With intact ACL, 5 degrees internal rotation occurs during Lachman test (p\0.005). After AMB cut, while the internal rotation of the tibia spontaneously increases, with regard to the native ACL when the knee is bent at 30 degrees of flexion, there is no significant increase of internal rotation., during Lachman test. Cutting the PLB alone induces a 13 degrees increase of internal tibial rotation (p\0.002). Full cut of the ACL induces a further 6 degrees internal rotation increase (p=0.02). Conclusions: As already shown in the literature, tibia internally rotates during Lachman test. In accordance with our initial hypothesis, sequential cutting of the ACL bundles shows a more significant effect of the PLB cut compared to the AMB cut.

P13-1074 Comparative study between mono-bundle, double-bundle and monobundle 1 extra-articular plasty ACL reconstructions:a two years F.U. investigation Wasconcelos W.1, Saggin P.1, Dejour D.1 1 COROLYON - Clinique de La Sauvegarde, Orthopaedic Surgery, Lyon, France Objectives: To the present date, no consensus exists regarding the use of double bundle or extra-articular reconstructions compared to mono bundle in anterior cruciate ligament reconstructions. The objective of this study is to compare three different procedures performed by the same surgeon: patellar tendon mono-bundle reconstruction (BTB); hamstrings doublebundle reconstruction (DBH); and patellar tendon mono-bundle combined to extra-articular plasty (Lemaire) (BTB?L). Methods: Seventy five patients (25 in each group) were evaluated by an independent orthopedic surgeon at a mean follow up of 25 months. Pain complains, sensibility deficits and subsequent surgical procedures were

123

S202 recorded. Eight zones of sensibility were tested in each knee, and the mean number of hypoesthesia zones was calculated for the patients who had it. Patients were evaluated in relation to their ability to kneel over their affected knee, to squat, and sports return. IKDC scores were calculated post operatively. Laxity was assessed post operatively with TelosTM stress radiographs (15 Kg) for the internal and external compartments. Pre operative values were available for comparison. The amount of anterior tibial translation (ATT) corrected by the surgery was quantified, and this was compared between groups. Results: The presence of anterior knee pain and subsequent surgeries was not statistically different between groups. Sensibility deficits were present in 11 patients from DBH group, while in 17 in BTB and 19 in BTB ? L (ANOVA, p=0,052). Quantitative analysis of hypoesthesia zones showed a mean of 1 zone in DBH group versus 1,76 in BTB and 1,84 in BTB?L (ANOVA, p= 0,0001, Tukey test). Kneeling was better in DBH group (Chi-square, p=0,0001). In 22 patients it felt normal, while only in 7 in BTB and 8 in BTB?L. Squatting was similar among the three procedures (DBH 18, BTB 17, BTB?L 17). Sports return was possible in 20 patients in DBH, 20 in BTB, and 16 in BTB?L (Chi-square, p=0,324). Mean IKDC values were 89,1, 90,1 and 86 (ANOVA, p=0,224) for DBH, BTB and BTB?L respectively, and their respective overall ‘‘A and B’’ results were 23, 23 and 24. ATT improved in all 3 groups, from 10,4mm (internal) and 15,2mm (external) pre operatively, to 4,7mm (internal) and 9,2mm (external) post operatively (T-test paired samples, p=0,0001). ATT absolute correction for the internal compartment were 5,5mm for DBH group, 5,4 for BTB and 6,4 for BTB?L. This was statistically not significant (ANOVA, p=0,377). For the external compartment correction, however, BTB?L (8,2mm) showed superiority over DBH (5,6mm) and BTB (4,1mm)(ANOVA, p=0,0001, Tukey test). Conclusions: There are intrinsic differences between the three procedures. Sensibility deficits seem to be worse in the patellar tendon groups, as kneeling also seems to. Absolute correction of anterior tibial translation was statistically not different for the internal compartment, but the patellar tendon reconstruction combined to extra-articular procedure achieved the best external compartment ATT correction, although it gives a stiffer knee. Objective and subjective IKDC, and sports return rates, however, were not statistically different.

P13-1088 Knee rotation assessment in single-bundle ACL reconstructions D’Elia C.1, Bitar A.C.2, Castropil W.1, Schor B.2, Garofo A.G.P.1, Orselli M.I.V.2, Duarte M.2 1 Instituto Vita, Ortopedia, Sa˜o Paulo, Brazil, 2Instituto Vita, Sa˜o Paulo, Brazil Objectives: Reconstruction surgery of the anterior cruciate ligament (ACL) has greatly advanced over the last twenty years. Nevertheless, some literature reviews reveal that approximately 15% to 25% of patients submitted to surgery do not present what can be considered excellent results. We believe, like Lubowitz and Phoeling in 2008, that studies in biomechanics laboratories, evaluating in particular, through more functional tasks, rotation of the knee and the capacity of the ACL reconstruction technique to avoid a dynamic range of rotation, is currently the best way to evaluate reconstruction results. In this study we evaluate knee rotation in patients who have undergone single bundle ACL reconstruction. Methods: We analyzed 8 knees from 8 patients that have undergone single bundle ACL reconstruction with the same technique (operated group). We also evaluated a control group of 8 patients who had never undergone any lower limb surgery. The groups were considered parametric in relation to age and sex. The patients performed three tasks in the biomechanics lab such as walking without change of direction, walking with change of direction and landing with change of direction. Both knees were analyzed and then dynamic range of rotation was compared. The knee joint angles were measured by positioning retroreflective markers on the participant’s body, according to the chosen marking protocol and ground reaction force

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 was also measured, enabling the joint torques acting on the knee during the tasks described below to be eliminated. Results: Comparison between mean range from operated group and non operated knees (contralateral group) did not present significant differences in any of the three tasks evaluated (pC0,251). Also, comparisons between mean knee rotation from the operated group and control group did not present statistically significant difference in any of the tasks (pC0,401). Conclusions: This data represent the preliminary results from only one trial and an aspect to be emphasized is the possibility to use this kind of rotational knee analyst to assess patients with multiple ligament injuries, angular and/or torsional deformities in the lower limbs and any other clinical condition that could be associated with changes in the knee rotation control. Biomechanics lab is a useful tool to objectively assess knee kinematics. Preliminary results showed no difference concerning single bundle ACL reconstruction, contralateral side and control group in spite of the short follow-up period and the small sample size.

P13-1089 ACL reconstruction in highly active athletes (Tegner 9 or 10) Van Dijck R.1, Saris D.2, Willems W.J.3, van Ommeren J.W.4, Fievez A.4 1 Franciscus Hospital, Dept. of Orthopeadics and Traumatology, Roosendaal, Netherlands, 2University Medical Centre Utrecht, Department of Orthopaedics, Utrecht, Netherlands, 3Onze Lieve Vrouwe Gasthuis, Dept. Orthopaedics, Amsterdam, Netherlands, 4Medinova Clinic, Orthopedic Surgery, Zestienhoven, Netherlands Objectives: Despite the high incidence of acl reconstruction in high performance athletes (Tegner 9 or 10), there is less known about the results of acl reconstruction and the return to sport activities in this patient group compared to recreational sports. In this study we evaluate the long term results of acl reconstruction in the elite athlete (Tegner score 9 or 10). Methods: Between 1988-1998, 305 patients underwent an acl reconstruction with a bone-patella-bone autograft (n=210) or a semitendinosis/ gracilis autograft (n=95). Of these patients, we included all 134 patients with a Tegner 9 or 10 pre-operatively. 123 patients were available for evaluation with a mean follow-up time of 7.4 years. Evaluation consisted of a physical examination including questionnaires and Lysholm and Tegner scores, IKDC score, X-rays (Ahlback / Fairbank’s classification). Results: At the final follow up 48 (39%) patients had an IKDC score grade A, 54 (44%) an IKDC score grade B, 15 (12%) grade C and 6 (5%) patients a grade D. The mean Lysholm score was 84.6 and the mean Tegner score 6.3. Of the study group 60 patients (49%) returned to their pre-operative sport level. 47 patients (38%) changed sport activities and 16 patients (13%) ceased sport activities. 76% of the patients changed or ceased sport activities due to fear of re-injury. Of the patients with a Tegner 10 score, 29 of the 31 patients returned to their previous sports level. All patients with a Tegner 10 score had a frequent and intense, personalized rehabilitation program. All patients with persistent or recurring instability after acl reconstruction seized or changed sport activities. Between the time of reconstruction and the final follow-up 75 patients (61%) developed degenerative changes. Patients with cartilage damage and/or meniscal lesions present at reconstruction developed significantly (p\0.05) more degenerative changes and had significantly worse results in relation with the return to sport activities. Conclusions: The results of acl reconstruction in highly active athletes with follow up of more than 7 years are good. All but two Tegner 10 athletes returned to full sports at previous level. However of the full population only 49% returned to their previous sport level. Lack of necessity and fear of re-injury may be a reason to stop while for the elite athletes the consequences and an intense and frequent rehabilitation program play an important role in returning to previous sport activities. A large percentage of highly active athletes (61%) developed degenerative changes after acl reconstruction with a higher propensity in those with existing meniscus or cartilage damage at reconstruction.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P13-1090 Biomechanical assessment in humans of doubl- bundle anterior cruciate ligament reconstruction: preliminary results D’Elia C.1, Bitar A.C.2, Castropil W.1, Schor B.2, Garofo A.G.P.1, Orselli M.I.V.2, Duarte M.2 1 Instituto Vita, Ortopedia, Sa˜o Paulo, Brazil, 2Instituto Vita, Sa˜o Paulo, Brazil Objectives: The ACL consists of two well-defined bundles: anteromedial (AM) and posterolateral (PL). Biomechanical studies in cadavers have proven anterior and mainly rotational improvement of the knee, using the double bundle reconstruction technique, secondary theoretically, to posterolateral bundle reconstruction, since only the anteromedial bundle is reconstructed in the usual one bundle technique that has been used for years. Our objective was to present our preliminary results of a comparative study in patients who underwent double-bundle ACL reconstruction surgery concerning knee rotation analysis in a kinematic analysis. Methods: We analyzed 9 knees from 9 patients that have undergone double bundle ACL reconstruction with the same technique (operated group). The groups were considered parametric in relation to age and sex. We also analysed a control group of nine patients, who had never been submitted to any kind of surgery in their lower limbs. The patients performed three tasks in the biomechanics lab such as walking without change of direction, walking with change of direction and landing with change of direction. Both knees were analyzed and then dynamic range of rotation was compared. The knee joint angles were measured by positioning retroreflective markers on the participant’s body, according to the chosen marking protocol and ground reaction force was also measured, enabling the joint torques acting on the knee during the tasks described below to be eliminated. Results: Comparisons between mean range from operated group and nonoperated knees (contralateral group) did not present significant differences in any of the three tasks evaluated (pC0,205). Also, comparisons between mean knee rotation from the operated group and control group did not present statistically significant difference in any of the tasks. Conclusions: An interesting data that called our attention was the tendency of knees of operated patients to present less rotational range of motion. The mean follow-up period of these patients was 1 year. Preliminary results showed no difference concerning double bundle ACL reconstruction technique in relation to the contralateral side and the control group, in spite of the short follow-up period and the small sample size.

P13-1091 Fixation of double bundle ACL reconstruction with hamstrings by a single ‘‘crosspin ’’ in each tunnel: biomechanical study of 48 fixations Potel J.-F.1, Benhima A.1, Molinier F.2 1 Clinique Cours Dillon, Toulouse, France, 2CHU Toulouse, Toulouse, France Objectives: Postoperative rehabilitation and early activities recovery after reconstruction of anterior cruciate ligament (ACL) requires a primary fixation with adequate mechanical properties. The purpose of this study is to determine the properties of a single ‘‘crosspin’’ RigidFi fixation of tendon transplants in small tunnels simulating a double-bundle reconstruction. Methods: On 24 adult fresh frozen cadavers (12 femurs and 12 tibias dissected), 48 bone tunnels were prepared in which were fixed with a single ‘‘Crosspin’’ RigidFix a tendon harvested from the same specimens, plaited, with pre-tension and with adequate diameter to tunnels (5 to 8mm). Two types of tests were performed: – Cyclic test : during 500 cycles, a sinusoidal load between 50N and 150N was applied to the bone-fixation-tendon complex with continuous recording of the constraints and elongation (slippage). – Ultimate load test : after cyclic loading, an increasing load with a controlled speed 5mm/mn was applied to the bone-fixation-tendon complex until failure, with continuous recording of the elongation. Results: The mean elongation after 500 cycles is 1.89 mm ± 0.84 with a progressive stabilization of its increase with the number of cycles. The yield load average is 262 ± 60 N with an elongation of 4.61 mm ± 1.68. The stiffness of the entire bone-fixation-tendon complex is 61N/mm ± 16

S203 and the ultimate load is 295N ± 79. The yield load and ultimate load are more important in the femur (283N and 320N) than in the tibia (239N and 261N). The strengths increase with the diameter of the tunnel (254N for 5mm, 331N for 8 mm) and there were 6 failures out of 18 tests with 5mm diameter with graft sliding (technical error). Discussion: The stress supported by the femoral fixation is less important than the tibia, in view of the application angle when the knee is extended. The yield load result fulfilled the specifications in the femur. At the tibia, because the stress is in the axis of the tunnel, the ultimate strength is not enough to allow intensive rehabilitation. Conclusions: The use of a single ‘‘crosspin’’ in each tunnel in an anatomic reconstruction of anterior cruciate ligament is possible at the femur with diameters of the transplant greater than or equal to 6mm, but not desirable in the tibia. For 5mm diameters at the femur it is imperative to ensure proper centering of ‘‘crosspin’’ guide before putting in place the graft into the tunnel.

P13-1094 Follow up of double bundle ACL operated patients Sanchez F.H.1, Aune A.K.2, Rosenlund E.A.2 1 Volvat Medical Center, Orthopedic, Oslo, Norway, 2Volvat Medical Center, Oslo, Norway Objectives: Anterior cruciate ligament (ACL) is anatomically composed of 2 bundles, anteromedial (AM) and posterolateral (PL). AM prevents mainly anterior-posterior translation (drawer test and Lachman) and PL pivot shift. Over the last few years the double bundle (DB) ACL reconstruction has become an alternative method to the traditional single bundle techniques. With a new ACL reconstruction technique it is necessary to do a patient follow up and do a clinical examination as a quality control. Methods: Material and methods 45 patients with ACL rupture were operated with DB-ACL reconstruction in VMC by the same orthopedic surgeon (AKA) in 2007-2008. 45 was contacted and 35 underwent a clinical examination. Semitendinosus (AM) and gracilis tendons(PL) were harvested from ipsilateral knee and tunnels drilled according to foot-print anatomy. Proximal fixation with Endobutton CL/direct (S.Nephew) and distally with resorbable interference screw. Testing was done with IKDC and Tegner activity scores. Standardized roentgenographs were taken in standing position (Synaflex frame). Results: 35 of 45 patients underwent clinical examination. 10 did not choose to be examined but were telephone contacted and they had all returned to their pre injury activity level. Follow-up time median 13 months (6-24 months). Median age 22 years at the time of the operation. No one was reoperated. IKDC: 1. (EFFUSION) 34 A and 1 B and no C or D. 2. (ROM) 35 A and no B,C and D. 3. (STABILITY) 30 A and 4 B and 1 C, no D. 4. (PAIN) 32 A and 3 B and no C or D. 5. (HARVEST SITE) 27 A and 8 B and no C or D. 6. (X-RAY) 35 A 7. (ONE LEG HOP) 32 A and 2 B and 1 C and no D. Tegner: Average score 7. 28 of 35 had reached the same activity level regarding sports as they had before the ACL rupture. No one had experiences a give away episode and everybody had normal activity of daily living. Conclusions: The method gives good results and has few complications at short time follow-up.

P13-1108 Rotational instability evaluation for assessing different types of anterior cruciate ligament reconstructions with the use of triaxial accelerometry Maeyama A.1, Kato Y.1, Hoshino Y.1, Lertwanich P.1, Ingham S.2, Kramer S.1, Wang J.2, Smolinski P.2, Fu F.2 1 University of Pittsburgh, Pittsburgh, United States, 2University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States Objectives: Residual rotational instability after conventional single bundle (SB) ACL reconstruction remains unsolved, because such instability is

123

S204

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

hard to be quantitatively evaluated due to its dynamism and complex motion. However, some recent reports have successfully detected the instability by measuring acceleration of the instability. Anatomical double bundle (DB) ACL reconstruction has been advocated as a better technique to restore rotational stability, while new SB technique whose graft placement is closer to the anatomical ACL footprint has also been introduced to restore better stability. The purpose of this study was to compare these different types of ACL reconstructions by measuring acceleration of the rotational instability. Methods: Sixteen human cadaver knees (Age, 59±6) were used. The jerk test which can reproduce dislocation phase of pivot shift phenomenon was performed manually. An accelerometer was used to record triaxial acceleration during the jerk test (x-axis: anteroposterior; y-axis: mediolateral and z-axis: superoinferior). A sensor of this device was attached on the tibial tuberosity with a bone screw. Specimens were tested before and after ACL resection, and after three different types of ACL reconstructions. 1) Anatomic SB reconstruction (MIDMID) (n=8); the graft was placed from the mid section between the AM and PL footprints on both the tibia and the femur sides, 2) Conventional SB reconstruction (PL-high AM) (n=8); the graft was placed from the tibial PL footprint to femoral high AM position, 3) Anatomical DB reconstruction (DB) (n=8); grafts were placed in AM-AM and PL-PL positions. The knees used for PL-high AM reconstruction were recycled for DB because tibial tunnel placements were the same and femoral placements were completely different. The overall magnitude of acceleration was calculated for evaluation by use of the formula {|a|=H (ax2?ay2?az2)}. Results: Significantly larger magnitude of acceleration was found in ACL deficient knees (P\0.01) and the both types of SB reconstructed knees (P\0.01 for MID-MID and P\0.01 for PL-high AM) compared to the intact knee. The overall magnitude of acceleration in the DB group was not significantly different from the intact ACL group (P =0.46). (Figure 1, Table 1)

Fig. 1 Oveall magnitude of acceleration in each reconstructions. P value\0.01

Each magnitude of triaxial acceleration. Direction

intact deficient PL-highAM MID-MID DB (mean±SD) (mean±SD) (mean±SD) (mean±SD) (mean±SD)

X-axis (m/s2)

0.61±0.11

0.75±0.14

0.59±0.27

0.51±0.16

0.27±0.10

Y-axis (m/s2)

0.33±0.27

0.76±0.3

0.84±0.38

0.39±0.16

0.41±0.16

Z-axis (m/s2)

0.27±0.11

0.88±0.22

0.85±0.13

0.61±0.16

0.63±0.12

Overall magnitude 0.75±0.13 of Acceleration (m/s2)

1.47±0.13

1.37±0.32

0.96±0.14

0.82±0.10

123

Conclusions: Comparable evaluation of rotational instability is vital for assessing and improving current ACL reconstruction procedures which already solved anterior instability. The acceleration of the instability measured by triaxial accelerometer could provide quantitative parameter to evaluate the magnitude of the instability. The present study demonstrated that DB reconstruction can restore rotational instability better than SB in terms of the rotational instability.

P13-1110 Clinical results of anatomic double-bundle anterior cruciate ligament reconstruction using hamstring tendon grafts: a minimum 5-year follow-up study Kitayama S.1, Kitamura N.1, Kondo E.1, Ogawa M.2, Tohyama H.1, Yasuda K.1 1 Hokkaido University School of Medicine, Sports Medicine and Joint Reconstruction Surgery, Sapporo, Japan, 2Nara Medical University, Department of Orthopaedic Surgery, Kashihara, Japan Objectives: Several studies reported that the anatomic double-bundle procedures are significantly better than the single-bundle procedures concerning the pivot-shift test and/or the anterior laxity. However, no studies have evaluated the long-term survival of graft function and clinical outcome after anatomic double-bundle ACL reconstruction. The purpose of this study was to assess the clinical results after anatomic double-bundle ACL reconstruction over at least five years. Methods: A prospective cohort study was conducted using 69 patients who underwent anatomic double-bundle ACL reconstruction using the hamstring tendon hybrid grafts in the unilateral knee between 2003 and 2004. Forty-four of the 69 patients were clinically and radiographically evaluated both at the one year and at least five years after surgery. Of these patients, two patients had a revision surgery because of graft failure due to trauma, and three patients had an ACL injury in the contralateral knee. After excluding these five patients from the final evaluation, 39 patients were included in the clinical assessment. There were 20 men and 19 women with an average age of 37 years at the time of the operation. The anterior laxity was measured with a KT-2000 arthrometer. Peak isokinetic muscle torque was measured using Cybex II. Anteroposterior (AP) and lateral radiographs of the knee were used to evaluate the potential degenerative changes. As to overall evaluation, the Lysholm score and the IKDC form were used. Statistical comparison was performed using the paired t-test and Spearman’s rank correlation. The significance level was set at p = 0.05. Results: Partial menisectomies were performed in two patients one year after the ACL reconstruction. The side-to-side anterior laxity averaged 1.3 mm and 1.7 mm at the one-year and five-year follow-up, respectively. Quadriceps and hamstrings muscle torques were 87.2% and 94.8% at the one-year follow-up, and 93.0% and 98.5% at the five-year follow-up, respectively. The mean Lysholm score was 97.0 points at the one-year follow-up and 98.6 points at the five-year follow-up. Assessment using the IKDC score revealed that 65.4% were rated as A and 34.6% as B at the one-year follow-up, and 65.4% were rated as A, 30.8% as B, and 3.8% as C at the five-year follow-up. The osteoarthritic changes were assessed using the Kellgren/Lawrence (KL) system, and 58.2% of the patients were grade 0 (normal), 37.5% were grade I, and 4.2% were grade II at the fiveyear follow-up, while 61.5% of the patients were grade 0 and 38.5% were grade I at the one-year follow-up. There were no significant differences in the anterior laxity, the hamstrings muscle torque, the Lysholm score, the IKDC evaluation, and the KL grade between the one-year and five-year follow-ups, while there was a significant difference in the quadriceps muscle torque (p=0.0315). Conclusions: This study demonstrated that good clinical results were maintained in terms of clinical scores and the knee stability overtime more than 5 years after the anatomic double-bundle reconstruction. The risk of the development of osteoarthritis did not increase and only two patients required meniscus surgery during the follow-up period. These findings suggested that this anatomic double-bundle reconstruction is clinically useful and practical in the treatment of the ACL-deficient knee. However, we should recognize that the examination rate in this study

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 was as low as 64%, thus, more thorough investigation and further follow-up are needed.

P13-1116 Do patients select anatomical double bundle or single bundle ACL reconstruction? A questionnaire survey for patients and medical students Iriuchishima T.1, Horaguchi T.1, Motojima S.1, Morimoto Y.1, Kubomura T.1, Tokuhashi Y.1, Saito A.1 1 Nihon University School of Medicine, Orthopaedic Surgery, Tokyo, Japan Objectives: Anatomical double bundle (DB) anterior cruciate ligament (ACL) reconstruction is becoming popular. On the other hand, traditional single bundle (SB) reconstruction has been widely performed and is still the gold standard technique. In most cases, the selection between DB or SB technique is performed by the medical staff and rarely the patient is consulted about this matter. The purpose of this study was to evaluate the patient’s preferred technique when being submitted to an ACL reconstruction. Methods: One hundred and ten subjects (35 patients and 75 medical students) were included in this study. A questionnaire survey was performed (in Japanese) and the following questions were included: 1) How old are you? 2) Have you ever heard of the ACL? 3) Do you have any knowledge about the ACL reconstruction techniques? 4) Do you know the difference between DB and SB? 5) If you had to have an ACL reconstruction, which technique would you prefer? In the same questionnaire, between questions 3 and 4, a description about SB and DB was provided: In the SB technique, the ACL is reconstructed with only one bundle. This technique has been performed for more than 10 years and many of the patients have returned successfully to their sports activities. In spite of this, it is believed that this technique must still be improved and, for this, the DB technique has been developed. In this case, both ACL bundles are reconstructed and several studies have shown an improvement in biomechanical characteristics and short term outcomes. However, there are still no studies showing the long term results of the DB technique. Results: 69% of patients and 100% of medical students had heard about the ACL and 43% of patients and 85% of medical students had previous knowledge about ACL reconstruction. None of the patients and only 29% of the medical students knew the difference between DB and SB. 6% of patients selected SB and 40% of them selected DB. 9% of medical students selected SB and 67% of them selected DB. Of those who had previous knowledge about ACL reconstruction (both patients and students), 14% selected SB, and 64% selected DB. Conclusions: To our best knowledge, this is the first survey that evaluated the knowledge and choice of patients regarding DB and SB techniques. In this study, more than 60% of subjects selected the operation technique by their self. In the future, It might be possible to be selected DB or SB by the patients with careful informed consent.

P13-1124 Clinical outcome of augmentation procedure in non-functional partial rupture of ACL Bae J.-H.1, Lim H.-C.1, Seok C.-W.1, Song S.-H.1, Cho J.-W.1, Lee J.-W.1 1 Korea University Guro Hospital, Department of Orthopedic Surgery, Seoul, Republic of Korea Objectives: The purpose of this study was to investigate the clinical results of 35 patients who had undergone anterior cruciate ligament (ACL) augmentation procedure using an autogenous semitendinosus tendon or tibialis anterior tendon allograft. Methods: We reviewed 35 patients who underwent augmentation for nonfunctional partial rupture of ACL (pivot sift test positive). The average

S205 follow-up period was 33 months (range, 15 to 47). They were assessed using a KT-2000 knee arthrometer (MEDmetric, San Diego, CA), Lysholm knee score, 2000 International Knee Documentation Committee (IKDC) and Tegner activity score. Results: The mean side-to-side difference of anterior displacement measured by the KT-2000 knee arthrometer at 30 degrees of knee flexion preoperatively was 3.3 ± 2.4 mm, significantly improving to a mean of 0.5 ± 2.7 mm 2 years after surgery. The mean Lysholm knee score significantly improved from 74.3 (range, 64 to 85) to 89.7 points (range, 81 to 100) after surgery. The average IKDC Subjective Knee Evaluation Scores were 65.8 (range, 63 to 83) preoperatively, and 88.3 (range, 83 to 100) at last follow up. The median preinjury and preoperative Tegner score was 7 (range, 5 - 9) and 3 (range, 2 to 5), respectively. The median postoperative Tegner score was 6 (range, 3 - 9). Conclusions: Augmentation procedure preserving remnant ACL provided satisfactory clinical results without early complication in non-functional partial rupture of ACL.

P13-1128 Tunnel enlargement after ACLl reconstruction with a quadruple semitendinosus-gracilis autograft. A CT study Chouliaras V.1, Giotis D.1, Boulis s.1, Kappes G.2, Tatsis C.2 1 General Hospital of Arta, Orthopaedic Department, Arta, Greece, 2 General Hospital of Arta, Radiologic Department, Arta, Greece Objectives: The aim of this study is to find out if there is tunnel enlargement after ACL reconstruction with a quadruple semitendinosusgracilis autograft by the aid of CT, and to correlate this to the clinical result Methods: We review the tunnel enlargement 3 months postoperatively by the aid of CT in a series of 25 consecutive patients who have been operated with a quadruple ST autograft. All patients were male, with a mean age 23.6 years old (18-35). In all the patients we used the same femoral fixation system (XObutton) and the same tibial fixation system (bioabsorbable screw) All the patients follow the same rehabilitation program. We measured the tunnel diameter in the tibia in the transverse, coronal and sagittal level, and the femur diameter in the coronal level. Results: One year postoperatively all the patients had stable knees (Lachman test -) and KT-1OOO \ 2mm. The statistical analysis of the results was performed with the SPSS system. We noticed tunnel enlargement in all the levels and this was statistically significant. Tibial transverse 8.378 vs 8.111 (p\0.005), Tibial sagittal 8.428 vs 8.111 (p\0.001), Femur coronal 8.161 vs 7.694 (p\0.001). Conclusions: Tunnel enlargement after ACL reconstruction can be seen the first postoperative months, especially if someone uses suspensory fixation systems. This phenomenon is not correlated with the clinical result.

P13-1132 Resorbable implants design using medical imaging Marinescu R.1, Laptoiu D.1, Antoniac I.2 1 Colentina Clinical Hospital, Orthopedic and Trauma, Bucharest, Romania, 2University Politechnica Bucharest, Bucharest, Romania Objectives: Virtual modeling is more and more used as a pre-clinical evaluation and testing tool in the biomedical field. It is a non-invasive method that was validated to predict bio-mechanical behavior; it can be performed to gain insights before clinical applications, thus increasing the efficiency of the design process. Methods: MRI and CT scans can provide data on the shape and density distribution of a tissue required for a virtual model of anterior cruciate ligament replacement with a bone patellar tendon-bone graft. MIMICS 10 (Materialise, Belgium) was used to obtain our 3-D model. Using the thresholding procedure, ‘‘masks’’ were created which consisted of a group

123

S206 of pixels within the relevant range of HU’s (gray values: Hounsfield Units, HU). Once appropriate thresholding values are chosen, the program groups pixels with densities within the thresholding range (such as bone densities). The 3D models were surface-meshed with triangles using Mimics a volume mesh was created with tetrahedral shaped elements, and density was assigned to each element, based on the HU’s. FEA test were conducted with ANSYS v11 software.The initial strain in the tendon graft was set respectively to the corresponding 0, 20, 40 and 60 N of pretension. These initial strains were included in the model following the methodology described in Gardiner and Weiss (2003). Different implant models (interference screws with different designs - three market models by Stryker, Stortz, Arthrex, three new models with different pitch, proximal half angle and shape), designed in Catia V5 R and exported as *.igs files) were loaded with a force of 200 N directed along the tunnel axis, an approximation of the graft tension at full extension of the knee during gait (Harrington, 1976 cited by previous). Results: Although it was assumed that a 200 N graft force occurs at full extension, this force could vary during gait and affect the stress results. Maximum principal stress appeared in the posterior region of the femoral insertion of the graft, at 30 degrees of flexion, with average values of 12MPa. Pull-out at a rate of 30 mm/min strongly varies related to implant design (narrow pitch, conical shape give better results). Conclusions: The simulated effects of the implant - graft component on the tunnel wall, isolated stress shielded regions from surgically created tunnels and highlighted areas that are at risk for bone resorption that can lead to tunnel widening and implant failure.

P13-1134 Cadaveric study of acoustic emission monitoring during anterior cruciate ligament failure under uniaxial tension Paschos N.1, Barkoula N.-M.2, Gkartzonikas D.1, Aggelis D.2, Paipetis A.2, Matikas T.2, Georgoulis A.1 1 Orthopaedic Sports Medicine Center, University of Ioannina, Department of Orthopaedic Surgery, Ioannina, Greece, 2University of Ioannina, Department of Materials Science and Engineering, Ioannina, Greece Objectives: A ‘‘pop’’ at time of knee injury is one of the common clinical symptoms during anterior cruciate ligament (ACL) tear. The purpose of the current study was to investigate the ACL failure sequence by monitoring acoustic emission, under uniaxial tension along the ligament. A potential correlation between macroscopical response and acoustic emission activity was evaluated. Methods: Seven fresh-frozen human cadaveric knee specimens were fixed in an Instron machine, post removal of all soft tissues except for the ACL. With the usage of a digital goniometer each knee was positioned in 15 degrees of flexion. Load was applied parallel to the ACL axis. A video extensometer was used for accurate measurement of the ACL displacement. The load-elongation curve was monitored up to failure. Two high-resolution, high-speed digital cameras were used to detect the failure mode of the ACL. For the monitoring of the acoustic emission two wide band acoustic emission sensors were attached to either side of the ACL. The sensors were attached to the femur and tibia being able to record the events related to the detachment between the tissue and the bone, as well as the emissions from tissue rupture events that propagate through the tissue into the bone where the sensors are attached. The macroscopical findings during the experiment were used for comparison with the biomechanical results and the results of acoustic emission monitoring. Results: All ACL specimens demonstrated a non-monotonic response. The load-elongation curve demonstrated a plateau or a second peak post the initial drop of load. Macroscopically, in all specimens one group of fibers was undergoing a tear initially, with the intact fibers having the ability to continue to deform. The acoustic emission hit rate (of both transducers combined) was increasing as load was applied, reaching a maximum value at the time of ACL failure. Between the events of fiber tears, the rate was decreasing. Furthermore, the average frequency increases considerably during the tears of the different group of fibers. A direct connection between the failure patterns in the load-elongation

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 curves, the macroscopic sequence of events during ACL failure and the data from the acoustic emission monitoring was detected. The peaks of the load-elongation curve were coincided with the tears of fiber groups. These events were also accompanied by corresponding bursts of incoming acoustic emission signals at the exact time of the load peak or slightly earlier. Conclusions: The ACL ligament acts as a multi-fiber construction. In our setting, rupture follows specific patterns where a complete or partial tear of the fiber bundles were preceding while the remaining intact fiber bundles had a potential load resistance. A direct relation between the tear events and the acoustic emission activity was observed. When a rupture occurs, simultaneous detectable changes in the acoustic emission parameters were observed. In general, it can be said that the acoustic emission behavior can be used as a precursor of final failure, while study of certain parameters can characterize the process of failure.

P13-1143 Bone morphology of the tibial attachment of anterior cruciate ligament Hara Y.1, Hashimoto Y.2, Nishikino S.2, Takigami J.3, Yamazaki S.2, Nakamura H.2 1 Higashisumiyoshi Morimoto Hospital, Osaka, Japan, 2Osaka City University Graduate School of Medicine, Orthopaedic Surgery, Osaka, Japan, 3Yodogawa Christian Hospital, Orthopaedic Surgery, Osaka, Japan Objectives: Accurate and anatomic tunnel placements are essential to success the reconstruction of the anterior cruciate ligament (ACL). It0 s reported variously about the femoral attachment of anterior cruciate ligament in recent years, but there are few reports about that of tibia. Parsons described a ‘‘little knob’’ at the lateral to the medial tibial plateau and ACL tibial attachment locates from this knob to the midline of tibial plateau. Berg named that as ‘‘Parsons0 knob’’ and that was found in 10% of lateral X-ray image. The aligned trabecular structure behind Parsons0 knob which maybe continues from ACL is often seen by the lateral knee X-ray, but no one has reported that structure. On the other hand, Benjamin et al reported that ligament fiber continues into bone at the enthesis structure histologically. So we evaluated and considered the bone morphology of ACL tibial attachment using the cadaveric knee. Methods: 9 cadavers from the anatomical course of medical students at Osaka city university medical school of medicine, Osaka, Japan were used. The cadaver’s knee had slightly cartilage damages but ACL intact. The age range was 70-86 years. Proximal tibia was extract and all tissue except the ACL was removed. Cross-sections with the thickness of 3 mm were made using a bone trimmer with diamond band-saw blade (1-mm thick) in sagittal plane at the ACL attachment. Four slices were prepared and named group1, 2, 3 and 4 from medial to lateral. The each cut specimens were photographed with digital camera and radiographed with soft X- ray to identify the insertion of the ligament and bone. We examined the type of the bone morphology at the insertion, the height of the ridge, the location of insertion site and the relation to aligned trabecular structure. Results: In photograph with digital camera, a bony ridge anterior to the ACL attachment was identified. The ridge existed 89%, 100%, 94% and 22% in group 1, 2, 3 and 4, respectively. The average height of the ridge were 1.2mm, 1.3mm, 1.1mm, and 0.3mm in group 1,2,3 and 4, respectively. In radiograph, the aligned trabecular structure continued from ACL was identified and existed 100%, 100%, 22% and 6% in group 1, 2, 3 and 4, respectively. An angle of aligned trabecular structure and ACL fiber was 27?/-4.9, 25?/-8.2, 27?/-5.7 and 27 in group 1, 2, 3 and 4, respectively. Conclusions: Berg reported on a bony landmark anterior to the ACL tibial attachment as ‘‘Parsons’ knob’’. But, it was not three-dimensional but two-dimensional evaluation by lateral X-ray image. In this study, we showed that the bony ridge exists anterior to the ACL attachment in the width of about 1 cm. The aligned trabecular structure continued from ACL fiber was dominantly located anteromedial in the attachment of ACL. It is considered that trabecula is the response of the continuous stretch load from ACL. This structure confirmed by lateral X-ray may be

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 available for landmark of AM bundle in double bundle ACL reconstruction.

P13-1144 Biomechanical comparison of anatomic anterior cruciate ligament reconstructions between double-bundle and single-bundle Kato Y.1, Maeyama A.1, Ingham S.1, Wang J.1, Kramer S.2, Goto B.1, Smolinski P.2, Fu F.1 1 University of Pittsburgh, Department of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States, 2University of Pittsburgh, Department of Mechanical Engineering, Pittsburgh, Pennsylvania, United States Objectives: Anatomic double-bundle (DB) anterior cruciate ligament (ACL) reconstruction has been advocated because conventional singlebundle (SB) ACL reconstruction cannot better restore normal knee kinematics. There are, however, few reports which reported the comparison between anatomic SB and DB. Many previous studies compared DB ACL reconstruction with conventional SB ACL reconstruction. The purpose of this study is to compare anatomic SB ACL reconstruction with anatomic DB ACL reconstruction on biomechanics. Methods: Sixteen fresh frozen cadaver knees were used. Eight knees (n=8) were used for the anatomic SB ACL reconstruction, which was placed from the center of tibial ACL footprint to the center of femoral ACL footprint. Another eight knees (n=8) were used for anatomic DB ACL reconstruction. A robotic/universal force-moment sensor testing system was used. A load of 89 N was applied for anterior tibial translation (ATT) at 0, 15, 30, 60 and 90 of knee flexion. A combined internal rotation (5Nm) and valgus (7Nm) moment was applied at 0, 15, 30 and 45 of knee flexion. ATT (mm) and In situ forces (N) during the external loads were measured. Results: When compared to the intact ACL, the DB reconstructed knee had significantly lower ATT at 0 of knee flexion angle during KT and pivot moment. In response to the KT loading, both SB and DB grafts had in situ force which was equal to the intact ACL at 0, 15, and 30 of knee flexion angle, however, had significantly lower in situ force at 60 and 90 of knee flexion angle. In response to the pivot moment, there was no significant difference among all groups.

S207 Conclusions: Both anatomic SB and DB ACL reconstructions were almost equal to the AP and rotatory stability. In full extension, the DB reconstructed knee was less translated than the SB reconstructed knee.

P13-1151 Reconstruction using either a BPTB or Hamstrings autograft does not restore gait variability two years post-op Moraiti C.1, Stergiou N.2, Vasiliadis H.1, Xergia S.1, Zampeli F.1, Ristanis S.1, Georgoulis A.1 1 Orthopaedic Sports Medicine Center, University of Ioannina, Department of Orthopaedic Surgery, Ioannina, Greece, 2HPER Biomechanics Laboratory, University of Nebraska at Omaha, Omaha, United States Objectives: The examination of variability in various medical domains has provided with useful information concerning healthy and pathological conditions. Concerning gait, it has been shown that, under healthy conditions, gait variability (variability from one stride to the next) exhibits properties which allow the neuromuscular system to be flexible and adaptable to stresses. Pathology or aging can result in deterioration of these properties. It is of notice that in some cases changes in variability are predictive of subsequent clinical changes. It has been shown that ACL deficient patients walk in a more periodic way (decreased gait variability) compared to healthy individuals. We wanted to investigate gait variability after ACL reconstruction with either a BPTB or a quadrupled ST/G tendon autograft, in both limbs. Methods: Six patients with BPTB reconstruction (2 years post-op), six with ST/G reconstruction (2 years post-op) and six healthy controls walked on a treadmill at their self-selected pace. Two minutes of continuous kinematic data were recorded with a 6-camera optoelectronic system. Gait variability was assessed using the knee-flexion extension time series from 100 continuous strides using the nonlinear measure of the largest Lyapunov Exponent (LyE). We actually examined how knee flexion-extension changes over time. Clinical examination, IKDC, Lysholm and Tegner scores were performed as well. LyE values for the reconstructed and control limbs were submitted to a one way analysis of variance (ANOVA). Post-hoc differences were assessed using independent Student t-tests. In addition a two by two analysis of variance was performed to assess differences of the two

In situ forces and ATT

123

S208 groups and the contralateral intact knee. The level of significance was set at a=0.05. The statistical analysis was performed using SPSS. Results: The reconstructed limbs in both reconstructed groups exhibited significantly larger LyE values than the control. In addition, the intact contralateral limb exhibited significantly larger LyE values than the reconstructed limb in both groups. No significant differences were found between the autografts. Conclusions: Larger LyE values indicate that the knees of both legs of the reconstructed groups exhibit increased gait variability. Thus, ACL reconstruction failed to restore gait variability to normal values. In addition, variability in the intact contra-lateral was also affected. This could be considered as a compensatory mechanism. The increased gait variability may signify impaired neuromuscular performance and increased susceptibility to future pathology. Considering the lack of differences between the two grafts, our results also point towards modification of the present surgical techniques or of the rehabilitation protocols. In addition, the LyE measure could prove to be an important tool for the evaluation of various conditions that affect the neuromusculoskeletal system, just like as it has occurred in other medical domains.

P13-1175 Double-bundle anterior cruciate ligament reconstruction: a comparative cadaver study of the femoral tunnels performed with in-out and out-in techniques Ronga M.1, Punzetto P.1, Callegari L.2, Genovese E.2, Fu F.3, Pederzini L.4 1 University of Insubria, Dep. of Orthopaedics and Traumatology, Varese, Italy, 2University of Insubria, Dep. of Radiology, Varese, Italy, 3 University of Pittsburgh, Dept. of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States, 4Sassuolo Hospital, Dept. of Orthopaedics and Arthroscopic 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 inside-out 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 outin 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

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 in the inside-out than outside-in 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.

P13-1176 Medial compartment contact force should be inferred within the context of the sagittal plane moment: experimental evidence in acute unilateral ACL rupture Manal K.1, Gardinier E.2, Buchanan T.1, Snyder-Mackler L.3 1 University of Delaware, Mechanical Engineering, Newark, United States, 2 University of Delaware, Biomechanics and Movement Science, Newark, United States, 3University of Delaware, Department of Physical Therapy, Newark, United States Objectives: The external knee adduction moment is balanced by internal forces generated by muscles, passive restraints (ie, ligaments, joint capsule, etc.) and bone-on-bone contact forces. It is these latter forces that are of specific interest to those studying joint pathology and mechanisms of disease onset and progression. The prevailing thought with respect to articular loading is that larger knee external adduction moments are associated with greater medial compartment contact forces. This seems to be the case under certain conditions, but not necessarily valid under others. We propose in this study that the external knee adduction moment should be interpreted within the context of the sagittal plane joint moment if used to infer medial compartment joint contact forces. Methods: Standard motion analysis methods (ie, video-based motion capture and force platforms) were used to compute stance phase kinematics and kinetics for natural cadence walking in subjects with ACL deficiency. Data were collected for the involved and uninvolved limbs. Muscle forces for the dominant knee muscles were computed using an EMG-driven musculoskeletal model. In total, EMGs for 10 muscles were computed and used to predict individual muscle forces. The muscle forces were applied to a moment balancing algorithm leaving only the contact forces or soft tissue loads as the unknowns. The contact forces for each trial were time normalized to 100 samples and normalized by bodyweight (BW) for data averaging. Three trials per subject were averaged in this manner. Results: In general, the pattern described herein is applicable to our cohort of 18 ACL deficient subjects but only one subject’s data are presented due to space limitations. The frontal plane moment for the involved knee was larger compared to the uninvolved, even though the medial compartment contact force was smaller (involved = 2.5 BWs, uninvolved = 3.2 BWs). Noteworthy, the sagittal plane moment for uninvolved knee was larger compared to the ACL deficient knee (Figure 1).

Fig. 1 Net joint moment during stance for one subject. The bold solid and dashed lines represent the sagittal plane joint moments for the involved and uninvolved knees respectively. The thin solid and dashed lines depict the frontal plane moments for the involved and uninvolved knees for the same subject. Note that although the frontal moment for the involved knee was greater, the sagittal moment for the involved knee was smaller compared to the uninvolved knee.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Conclusions: When approximating medial compartment joint contact one must consider the magnitude of the sagittal plane joint moment. For the subject presented herein, as well as for other subjects in our cohort, it is evident that medial contact force can not be predicted solely from the external knee adduction moment. The magnitude of the sagittal plane joint moment should be considered when inferring medial compartment joint loading during gait.

P13-1182 Knee adduction moments and medial joint contact forces during gait in subjects early after acute unilateral ACL rupture Manal K.1, Gardinier E.2, Snyder-Mackler L.3, Buchanan T.1 1 University of Delaware, Mechanical Engineering, Newark, United States, 2 University of Delaware, Biomechanics and Movement Science, Newark, United States, 3University of Delaware, Department of Physical Therapy, Newark, United States Objectives: Individuals sustaining ACL injury are at greater risk of developing knee OA. Recently, Butler and colleagues reported that subjects have larger knee adduction moments compared to healthy agematched controls 5 years after ACL reconstruction. While intriguing as a possible factor associated with knee OA, other factors need be considered including time since injury or surgery that the adduction moment has been larger than normal. The implication is the longer the duration the more damaging the cumulative effect might be to the integrity of the joint. Moreover, knowledge of the bone-on-bone force (i.e., articular loading) can not be derived from the knee adduction moment. Ultimately it is the articular surface-to-articular surface forces that are of interest to those investigating joint loading as a possible mechanism for onset and progression of knee OA. To this end we have initiated a longitudinal study investigating gait mechanics and joint contact forces following ACL injury. Our purpose is to report pre-surgical frontal plane joint moments and estimates of medial contact derived using an EMG-driven musculoskeletal model. The clinical motivation for this work was to determine if early-on after injury frontal plane moments and medial contact forces are greater for ACL deficient knee compared to the non-injured contra-lateral leg. Methods: Thirteen subjects patients (8 men and 5 women) with unilateral, ACL rupture were tested an average of 9 weeks after injury. Mean walking speed was 1.54 m/s, weight was 85 kg and height was 1.76 meters. Standard motion analysis methods were used to compute stance phase kinematics and kinetics for natural cadence walking. Muscle forces for the dominant knee muscles were computed using an EMG-driven musculoskeletal model. The muscle forces were applied to a moment balancing algorithm leaving only the contact forces or soft tissue loads as the unknowns. The contact forces for each trial were time normalized to 100 samples and normalized by bodyweight (BW) for data averaging. Three trials per subject were averaged in this manner. Results: The knee adduction moment had a typical double hump pattern with the first peak larger than the second. The 1st peak for the involved knee was 0.242 (SD 0.066) N-m/Kg*Ht compared to 0.283 (SD 0.095) for the uninvolved leg (p = 0.198). A similar pattern was noted for the 2nd adduction peak. The moment was smaller for the involved limb 0.180 (SD 0.080) compared to the uninvolved leg 0.213 (SD 0.111) Nm/Kg*Ht (p = 0.204) The shape of the medial compartment contact force was similar to the pattern for the knee adduction moment. The first peak medial contact force was 2.21 (SD 0.43) BWs for the involved knee and 2.84 (SD 0.66) for the contra-lateral leg (p \ 0.05). Medial contact forces during the second half of stance were 2.09 (SD 0.36) and 2.71 (SD 0.51) BWs for the involved and uninvolved knees respectively (p \ 0.05). Conclusions: The peak knee adduction moment for the ACL deficient knees was not statistically different when compared to the uninjured knees. These data, albeit for a limited number of subjects, combined with our modeling results suggest that prior to surgery, increased medial compartment joint loading is not typical of patients with ACL deficiency. When, if ever, greater than normal medial compartment loads originate remains to be seen and is a focus of our on-going work.

S209 P13-1184 The association between alignment and post-traumatic osteoarthritis after an anterior cruciate ligament injury Swa¨rd P.1, Fride´n T.1, Roos H.1 1 Lund University Hospital, Department of Orthopaedics, Lund, Sweden Objectives: The prevalence of incident osteoarthritis (OA) after an anterior cruciate ligament (ACL) injury is consequently reported at high numbers, also in young individuals. Concomitant injuries to the knee joint, chronic inflammation, instability and additional damage to the knee due to episodes of give way could be important for the post-traumatic OA development. Another factor presented as a possible risk factor in OA is the alignment of the knee. The objective of this study was to investigate the association between alignment and post-traumatic OA after an ACL injury. Methods: In this investigation, 100 patients with an acute complete ACL tear, consecutively recruited, were followed-up after 15 years. An initial conservative treatment algorithm, focused on neuromuscular training, was amended, and the ACL was initially not reconstructed in any case. Large and unstable meniscal lesions were treated surgically with a partial meniscectomy. No patients had any radiographic or arthroscopic signs of OA at baseline. At follow-up, the patients were examined radiographically. The patellofemoral (PF) joint was examined with a skyline view and the knee in 50 flexion, and the tibiofemoral (TF) joint was examined with an anterioposterior view in weigthbearing. Joint space narrowing (JSN) and osteophytes were graded according to the radiographic atlas of the osteoarthritis Research Society International in the ACL injured and the uninjured knee. Patients were considered to have radiographic TF OA or PF OA of the knee if they had JSN C2 in any compartment, a sum of osteophytes C2 in the same compartment, or grade 1 JSN combined with a grade 1 osteophyte in the same compartment. Patients with bilateral ACL injuries, OA or JSN of the uninjured knee were excluded from analysis. The alignment of the knee was measured in the uninjured, contralateral knee at follow-up. A full-limb radiograph was taken, and the mechanic axis of the knee was assessed by means of the hip-knee-ankle (HKA) angle. Patients with a HKA angle below or equal to the median were considered to be valgus aligned, and patients with a HKA angle above the median were considered to be varus aligned. The association between varus/valgus alignment of the uninjured knee and incident OA of the injured knee was tested by calculating the relative risk, by means of a logistics regression analysis. Multivariate analysis was performed using the same logistics regression analysis, adjusted for age, sex, BMI, and meniscal tears treated with partial meniscectomy, sustained in association with the ACL injury. Results: Sixty-nine patients were included in the analysis. Eleven of the patients with varus alignment had OA of the uninjured knee, and 3 patients with valgus alignment had OA of the injured knee. Individuals with varus alignment of their uninjured knee at follow-up had a greater risk of incident OA in their injured knee 15 years after an ACL injury (odds ratio (OR) (95% confidence interval) 5.3 (1.3- 21.4)). After adjusting for age, sex, BMI and meniscus injuries rendering meniscectomy, sustained in association with the ACL injury, individuals with varus alignment of the uninjured knee were still found to be at greater risk of incident OA in the injured knee (OR 5.1 (1,1-22.6)). Conclusions: Our results suggest that individuals with varus alignment of their uninjured knee have a greater risk of having developed OA of the injured knee 15 years after an ACL injury.

P13-1186 Quadriceps and BPTB grafts in ACL reconstruction: rectangular bone tunnel reconstruction can restore intact knee kinematics significantly better than single bundle hamstring reconstruction Tecklenburg K.1, Zantop T.2, Herbort M.2, Hoser C.3, Fink C.3 1 Medical University Innsbruck, Dept. for Trauma Surgery and Sports Medicine, Innsbruck, Austria, 2University of Muenster, 1Department of Trauma, Hand, and Reconstructive Surgery, Muenster, Germany, 3Sports Clinic Austria, Innsbruck, Austria Objectives: Aim of this study was to investigate the knee kinematics after single bundle ACL reconstruction using a rectangular tunnel placement

123

S210 technique with quadriceps and BPTB graft and compare this data to a conventional single bundle (SB) hamstring reconstruction. We hypothesized that under a simulated KT 1000 and a simulated pivot shift test, ACL reconstruction with rectangular bone tunnel placement using quadriceps and BPTB graft will restore the intact knee kinematics better than a single bundle hamstring reconstruction with conventional round tunnel placement. Methods: In nine fresh-frozen human cadaver knees (range 52-76 years) the knee kinematics were examined using a robotic/UFS testing system. After determination of the passive path each knee specimen was tested under different conditions: intact, ACL-deficient, single-bundle ACL reconstructed with rectangular bone tunnels using quadriceps graft as well as BPTB graft, and ACL reconstructed with conventional round tunnels using hamstring graft. All reconstructions were performed using a medial portal approach and fixed using a cortical femoral fixation.The knee kinematics were evaluated under anterior tibial load of 134 N (to simulate KT-1000 test) and a combined rotational load of 10 N-m valgus and 4 N-m internal tibial torque (to simulate a pivot shift test). To exclude any interference, the order of the reconstructions was randomized. The previously drilled tunnels were filled by using bone cement. Results: Under 134 N anterior tibial load, anterior tibial translation (ATT) of the intact knee was a mean 4.7 (± 0.6) mm, 6.6 (± 1.9) mm, 7.1 (± 1.2) mm, 6.7 (± 2.2) mm, and 7.9 (±1.8) mm at 0, 15, 30, 60, and 90 of knee flexion, respectively. After the ACL was sectioned, the translations increased significantly at all flexion angles. ACL reconstruction reduced the increased ATT significantly at all flexion angles. However, between the groups there was a significant difference between the rectangular reconstruction using quadriceps and BPTB graft and the conventional reconstruction using a hamstring graft at 0 and 15 of knee flexion (p\0.05). In response to a simulated pivot shift, the ATT for the intact knee was 2.8 (±0.9) mm, 4.3 (±1.3) mm, 6.8 (±2.1) mm, 5.1 (±1.7) mm, and 6.2 (±2.7) mm for 0, 15, 30, 60, and 90 of knee flexion, respectively. Again, the ATT increased significantly in the ACL deficient knee at all flexion angles. Rectangular ACL reconstructions using a quadriceps or BPTB graft revealed in a significantly lower ATT at 0 and 15 compared to conventional SB reconstruction. Conclusions: The results support our hypothesis that under a simulated KT 1000 and a simulated pivot shift test, an ACL reconstruction technique using rectangular bone tunnels restores the intact knee kinematics better than conventional SB hamstring ACL reconstruction using round tunnels. Our results show the importance of an anatomical insertion site in ACL reconstruction: rectangular tunnel morphology seems to be able to imitate AM and PL bundle fibers of the native ACL better than a conventional SB hamstring reconstruction with round drill-holes.

P13-1193 ACL single bundle reconstruction with quadrupled semitendinosus and Tape Locking Screw technique: preliminary results of a prospective study of 82 knees Robert H.1, De Polignac T.2, Limozin R.3, Cassard X.4, Collette M.5, Lanternier H.6, Lefevre N.7 1 Centre Hospitalier Nord Mayenne, Orthopaedic Department, Mayenne, France, 2Clinique ge´ne´rale, Annecy, France, 3La Tour Raynalde, Rodez, France, 4Clinique des Ce`dres, Blagnac, France, 5Clinique Edith Cavell, Bruxelles, Belgium, 6Polyclinique de l’Europe, Saint Nazaire, France, 7 Centre Me´dico-chirurgical Paris V, Paris, France Objectives: The aim of this study was to evaluate a preliminary series of 82 ACL reconstructions using the quadrupled semitendinosus autograft and TLS (Tape Locking Screw) technique. Methods: The TLS technique is based on 4 principal steps: quadrupled semitendinosus graft in a 50 to 55 mm. loop, pre-tensioning of the graft to 500 N by means of a polyethylene terephthalate tape at each end, insertion of the graft into inside-out retrograde short femoral and tibial tunnels (10 and 15 mm respectively) and fixation of the tapes with titanium interference screws at the cortex of the femur and tibia distant to the tunnels. In the immediate post-operative period, patients are fully weight-bearing without a brace, and full range of motion is allowed. This is a prospective

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 and continuous series of 82 patients operated on in 2007 for chronic, unilateral, total rupture of the ACL. The procedure was performed arthroscopically by 3 experienced surgeons. The patients had a mean age of 29 years (14 to 51 years) and included 58 males and 24 females. All were sportsmen, 56 of them at a high level. 43 knees also had meniscal tears. The patients were evaluated pre and post-operatively (3, 6, 12, 18, 24 months) with the IKDC and Lysholm scores and stability was measured with a Telos or a GNRB. Results: All the patients were examined by an investigator independent of the surgeon after a mean period of 19.5 months (14 to 24 months). The patients0 overall satisfaction rating was 8.3 at 3 months and 9.1 out of 10 points at the final visit. The mean subjective IKDC score went from 66 to 92 points (p\ 0.0001), and the mean Lysholm score from 68 to 93 points (p\0.0001). Objective pre-operative IKDC classification was as follows: A: 2%, B: 6%, C: 71%, D: 21%, and post-operative classification was: A: 33%, B: 50%, C: 13%, D: 4%. Post-operatively the pivot shift was 0 glide0 in 14 cases and abnormal in 1 case. Laxity according to the Telos at 200 N or the GNRB at 250 N decreased from 6.8 mm to 2.2 mm and from 6.7 mm to 1.7 mm, respectively. 74% of patients resumed sport at the same level as before. There were no complications involving infection or thromboembolism. Two patients developed major arthrofibrosis with residual stiffness, one patient had limited flexion (10) and extension (3) and one patient underwent secondary arthroscopy for Cyclops syndrome. Three patients had residual laxity between 6 and 10 mm and there were no cases of re-rupture. Conclusions: The preliminary results of the TLS technique are very satisfactory on both the subjective as well as objective rating scales and offer a rapid functional recovery using a single autograft hamstring tendon. These results are identical to those of 2 recent meta-analyses of ACL reconstruction (Lewis PB et al., AJSM, 2009 and Biau DJ et al., CORR, 2007). It is necessary to undertake further studies to confirm these results and provide longer-term follow-up.

P13-1199 Prediction of the hamstring tendon autograft sizes before anterior crutiate ligament surgery by using a reproducible magnetic resonance measurement technique; a prospective double-blinded clinical study Beyzadeoglu T.1, Tasdelen N.2, Akgun U.3, Karahan M.4 1 Yeditepe University, School of Medicine, Department of Orthopedics and Traumatology, Istanbul, Turkey, 2Yeditepe University, School of Medicine, Radiology, Istanbul, Turkey, 3Acibadem University Faculty of Medicine, Orthopedics and Traumatolgy, Istanbul, Turkey, 4Marmara University, Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey Objectives: Size of the hamstring tendon autografts in the ACL reconstruction surgery is an important issue. We aimed to predict the sizes of the hamstring tendons in preoperative ACL surgery patients by using a reproducible MR technique. Methods: Between 2007-2008 51 (41 male, 7 female) patients, mean age 30.5 (range 18-67) who were planned surgery for ACL reconstruction were included to the study. Preoperatively, AP and lateral diameters and crosssectional areas of the gracilis(Gr) and semitendinosus(St) tendons were measured on the standard fat-suppressed images obtained by the 3T MRI unit. Datas were recorded by radiology department and surgery team were not acknowledged. Perioperatively diameters of Gr and St tendon autografts were measured separately (two strand) and together (4 strand) by using a standard ACL graft preparation instruments. Till the end of the study perioperative datas were not sent to the radiology department. The data obtained were evaluated statistically by pearson correlation and ROC analysis. Results: Considering cross-sectional area measurements for the Gr (2 strand), St (2 strand), and the total (4 strand), there were moderate to significant positive correlation between perioperative sizes and the measurements obtained by MRI. (r=0.293,p\0,05, r=0.384,p\0,01 ve r=0.419,p\0,05 respectively) In the ROC analysis of Gr (2 strand), St (2 strand) and total (4 strand) tendon datas, diameters of 5, 6, and 8mm were considered as the limit values. It is concluded that if the tendons cross-sectional areas in the preoperative MRI measurements are over 6.35, 12 ve 19.1 mm2 respectively, a

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 sufficiently thick autograft would be obtained by a chance of % 79 - % 55, % 79 - % 62, % 80 - % 62 sensitivity and specificity respectively. Conclusions: With a reproducible preoperative MR technique, the thickness of the hamstring tendon autograft sizes can be predicted with high accuracy.

P13-1201 Results and technique of ACL revision reconstruction Gohm A.1, Osti M.1, Benedetto K.-P.1 1 Academic Hospital Feldkirch, Trauma Surgery and Sports Traumatology, Feldkirch, Austria Objectives: ACL revision reconstruction is an increasing procedure over the last years. According to the etiology of recurring instability following ACL procedure, significant trauma has to be separated from primary graft malposition. Indication for ACL revision surgery is present in active patients with at least ?? Lachmann test with missing endpoint, subjective disability and free range of motion. Detailed preoperative evaluation is essential to avoid intraoperative pitfalls. The objective of this retrospective study was to evaluate the results of ACL revision reconstruction. Methods: Between 1995 and 2006 234 patients underwent ACL revision reconstruction. 188 fulfilled the criterion of minimum 2 years follow up. The age of 139 males and 49 females averaged to 33,2 (18 - 54 Jahre) years. 50 patients sustained adequate trauma with traumatic graft rupture, whereas 138 cases presented with incorrect initial tunnel placement resulting in reinstability in combination with recurrent minor trauma. Hamstrings were used in 4, quadriceps tendon in 23 and patellar tendon in 161 cases. Simultaneous bone grafting was necessary in 10 cases femoral and in 17 cases tibial, partial meniscectomy medial in 70 and lateral in 36 cases. Results: Table 1 Complications were intraarticular infection with arthroscopic lavage in 2 cases, superficial wound necrosis in 4 and restrictions in ROM which required re-arthroscopy in 4 cases. IKCD Symptoms

A/34

B/120

C/25

D/9

Subjective

A/24

B/145

C/13

D/6

S211 operatively, after fixation of the PL bundle, and after fixation of both the PL and AM bundles. The examination consisted of the pivot shift, Lachman and anterior drawer tests, internal/external rotation at 30 and 90 of knee flexion, and varus/valgus rotation 0 and 30 of knee flexion. Paired Wilcoxon tests were performed to evaluate the effects of each bundle. The comparisons included from before surgery to after fixation of the PL bundle; from before surgery to after fixation of both the PL and AM bundles and from fixation of the PL bundle to after fixation of both the PL and AM bundles. Significance was determined using a Bonferroni correction set at p\0.017 to account for the multiple comparisons. Results: All kinematic tests showed a significant decrease in laxity from the time before surgery to after fixation of both the AM and PL bundles. Anterior-posterior (AP) and rotational laxity decreased significantly from before surgery to after fixation of the PL bundle, and fixation of the AM bundle further decreased AP and varus-valgus (VV) laxity (Table 1). Results of kinematics tests Before surgery (n=15)

After PL bundle After PL & AM (n=15) bundle (n=15)

5.66 ± 1.59 (3.90 to 9.10)

5.04 ± 1.31 (3.60 to 7.90)

3.54 ± .88** (2.40 to 5.70)

Varus Valgus at 308 6.37 ± 2.78 (2.80 to 12.40)

5.69 ± 2.05 (3.30 to 11.00)

4.85 ± 1.83** (1.90 to 8.70)

Lachman

15.83 ± 5.47 (7.90 to 29.20)

10.71 ± 3.23* (6.60 to 17.00)

7.20 ± 1.71** (4.80 to 10.40)

Anterior Drawer

10.08 ± 3.67 (5.10 to 19.60)

7.07 ± 2.77* (4.20 to 14.80)

4.92 ± 1.48** (2.80 to 8.20)

Tibial Rotation at 308

30.74 ± 2.88 23.98 ± 4.12* 22.85 ± 4.45** (27.00 to 37.10) (18.20 to 31.20) (15.60 to 32.50)

Tibial Rotation at 908

32.48 ± 3.79 26.61 ± 3.45* 25.71 ± 3.14** (24.80 to 37.70) (22.00 to 30.90) (21.30 to 30.40)

Pivot shift anteroposterior area

295.1 ± 396.2 102.8 ± 91.9* (19.6 to 1639.0) (26.9 to 346.1)

Varus Valgus at 08

80.3 ± 62.7** (19.9 to 224.0)

* Before Surgery means were significantly different than After PL Bundle means ** Before Surgery means were significantly different than After PL & AM Bundle means

Stability

A/25

B/126

C/33

D/4

ROM

A/125

B/40

C/20

D/3

Total

A/25

B/122

C/32

D/9

 After PL Bundle means were significantly different than After PL & AM Bundle means

Conclusions: ACL revision reconstruction is indicated in young and active patients with subjective and objective instability and qualifies as option of choice to address restrictions in patient’s daily activities. Tunnel malposition seems to be the main reason for an unacceptable initial outcome. Our data indicate, that high patient satisfaction and good objective results are predictable after revision surgery, although results are inferior to correct primary ACL replacement surgery.

Conclusions: This in vivo study demonstrated the different function of the AM and PL bundles of the ACL. Our results indicate the importance of the AM bundle for stabilizing the knee in the AP direction at 30 and 90 of knee flexion, and the importance of the PL bundle for AP and rotational knee stability at 30 and 90. Interestingly, we found in this study that the AM bundle almost solely controlled VV laxity, at 0 and 30 of flexion. Based upon the results of this in vivo study, we conclude that DB ACL reconstruction may be necessary to re-establish normal knee kinematics.

P13-1204 In vivo kinematic evaluation of anatomic double bundle ACL reconstruction Kopf S.1, Bignozzi S.2, Irrgang J.1, Zaffagnini S.3, Musahl V.1, Fu F.1 1 University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States, 2 Istituto Ortopedico Rizzoli, Laboratorio di Biomeccanica, Bologna, Italy, 3 Rizzoli Orthopaedic Institute, Biomechanics Lab, Bologna, Italy Objectives: The current study was designed to evaluate in vivo function of the anteromedial (AM) and posterolateral (PL) bundles of the ACL during anatomic double-bundle (DB) ACL reconstruction for acute, isolated ACL tears utilizing a computer navigation system to track intra-operative knee kinematics. Methods: 15 patients with acute, isolated ACL tears who underwent anatomic DB ACL reconstruction were included. An image-free computer navigation system with custom-made software recorded the data during kinematic tests. During surgery the knee was examined by a clinician pre-

P13-1205 Arthroscopic treatment of mucoid degeneration of the anterior cruciate ligament: medium term follow-up results about 29 cases Lintz F.1, Dejour D.2, Boisrenoult P.3, Pujol N.3, Beaufils P.3 1 Hoˆpital de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France, 2COROLYON, Orthopaedic Surgery, Lyon, France, 3 Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France Objectives: Mucoid degeneration of the anterior cruciate ligament (MDACL) is a different pathology than anterior cruciate ligament ganglia. Clinical presentation associated knee flexion limitation and posterior knee pain. Treatment needs an anterior cruciate ligament resection, with some questions about harmlessness of this procedure. Our hypothesis was that arthroscopic anterior cruciate ligament is an effective procedure for pain and mobility but creates some anterior knee laxity.

123

S212 Methods: This is a retrospective cohort study including 29 cases of MDACL (19 men, 8 women). Mean age was 49 years (range 28 to 68). Mean follow-up was 6 years. Diagnosis was done associated clinical and MRI criteria and was confirmed during arthroscopy using Mc Intyre’s criteria. A histological analysis was done in 18 cases. Postoperative functional evaluation was done using IKDC and KOOS score. Knee laxity was appreciated using clinical evaluation and radiological evaluation by TELOS measurement. Statistical analysis was done using Student t-test (level of significance: p\0.005). Results: Posterior knee pain was present in 23 knees, and knee flexion limitation in 14 cases. In 10 cases, MDACL was initially misdiagnosed with an inappropriate primary operative treatment. Preoperatively, none of theses patients have an anterior knee laxity. Partial anterior cruciate ligament resection was done in 12 cases and complete resection in 17 cases. Meniscectomy was associated in 11 cases. In cases with histological study, diagnosis was always confirmed. After resection knee was painless in 27 cases, and knee flexion increase was 21.52. A positive Lachman’s test was noted after surgery in all cases, associated with a positive Jerk test in 8 cases. Mean postoperative radiological laxity evaluation using TELOS measurement was statically significant (operated knee vs normal knee: 12.64 /4.33 mm, p\0.001) Two young patients have need secondary an ACL reconstruction. Two old patients have needed secondary knee prosthesis after 2 and 3 years. Mean postoperative IKDC score was 71.19 (range 42.53 to 91.95) and mean postoperative KOOS score was 78.16 (range 26.40 to 99). Statistical analysis have showed better results for patient older than 50 years, after partial resection and for patient without meniscal associated lesions. Conclusions: Mucoid degeneration of the anterior cruciate ligament should not be confused with anterior cruciate ligament ganglia. Accurate diagnosis could be done using clinical, MRI and arthroscopic diagnosis criteria. Arthroscopic treatment of mucoid degeneration of the anterior cruciate ligament is an efficient procedure for knee pain and to restore a better knee flexion. However, this procedure created a significant anterior knee laxity and could lead in some cases to knee instability especially in young patients.

P13-1207 The excursion of native anterior cruciate ligament during knee range of motion Wang J.H.1, Kato Y.2, Ingham S.3, Maeyama A.2, Linde-Rosen M.2, Smolinski P.3, Fu F.3 1 Korea University Ansan Hospital, Orthopaedic Surgery, Ansan-si, Korea, Republic of, 2Univeristy of Pittsburgh, Pittsburgh, United States, 3 University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States Objectives: Understanding the excursion of the native ACL is important for the restoration of physiologic knee motion after ACL reconstruction. The purpose of this study is to evaluate the excursion of the native ACL during knee range of motion. It is hypothesized that ACL bundles would have different tension pattern according to the location of insertion sites. Methods: Six cadaveric knees were used in this study. The knees with registration screws were mounted on a universal force moment sensor (UFS) robotic system. First, the natural path of passive flexion-extension of the intact knee joint was determined and the positional relationship between the femur and the tibia was captured every 15 degrees of knee flexion (from 0 to 90 degrees) using a 3-D laser scanner (Faro, Lake Mary, FL). The surface data and the insertion data of ACL on the femur and tibia were acquired with the 3D laser scanner. The surface data and the insertion site data was superimposed to the positional data using Geomagic Studio 10 software. Finally, the functional length between the two insertion sites and percentages to the maximal resting length (MRL, maximal length during the passive ROM without external force and torque) were calculated. Results: In the extended position, the functional length of AMB was 39.6 ± 2.5mm, that of PLB was 26.9 ± 3.0 mm. The most anterior point (MAP) of ACL had the smallest excursion along five insertion sites (2.7mm ± 0.8mm, range 1.8 - 4.1 mm). The most posterior point (MPP) had the greatest excursion (7.7 mm ± 3.1mm, range 5.1 - 12.9 mm). The excursion of the anteromedial (AM) bundle and posterolateral (PL) were

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 3.7 ± 1.7 mm (range 2.1- 6.3 mm) and 7.6 ± 2.2 mm (range 5.9 - 11.6 mm), respectively. The functional length between the insertion sites became greatest when knee was fully extended (0 degrees flexion) except for the most anterior point. Conclusions: All fibers including the centers of the AM and PL bundles were tightened in the extended position except for the most anterior fibers (MAP). The most anterior fibers of the ACL showed most the consistent pattern of excursion during the knee ROM (0-90 degrees). The anterior fibers (MAP & AMB) of ACL had a smaller excursion when compared to the posterior fibers (PLB & MPP).

P13-1213 Tension pattern change of the anterior cruciate ligament after anterior tibial load and combined loads (valgus and internal rotation) Wang J.H.1, Kato Y.2, Ingham S.3, Maeyama A.2, Linde-Rosen M.2, Smolinski P.3, Fu F.3 1 Korea University Ansan Hospital, Orthopaedic Surgery, Ansan-si, Korea, Republic of, 2University of Pittsburgh, Pittsburgh, United States, 3 University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States Objectives: The anteromedial and posterolateral bundles of the anterior cruciate ligament (ACL) have different tension patterns at different knee flexion angles and also have different roles during the anterior drawer and pivot shift tests. The purpose of this study was to evaluate the tension pattern change (distance change between femoral and tibial insertion sites) of native anterior cruciate ligament with anterior tibial load and combined load (valgus and internal rotation). It was hypothesized that the tension pattern of the AM and PL bundles are different each other during the anterior tibial load and combined load. Methods: Six cadaveric knees were used for this study. Knee joints with registration screws were mounted on a universal force moment sensor (UFS) robotic system without any dissection. Two loads were applied: 89N anterior tibial load at 0, 30, 60, 90 degrees of knee flexion and a combined load (7N-m valgus and 5N-m internal rotation) at 0, 15, 30, 45 degrees of knee flexion. The positional relationship between the femur and the tibia was registered using a 3-D laser scanner. The insertion data and surface data of ACL on femur and tibia were acquired with the 3D laser scanner after dissection. Five points in the femoral and tibial insertion sites of the ACL’s long axis were marked (most anterior point (MAP) and most posterior point (MPP), middle of AMB, PLB and MID) with the Geomagic studio 10 software (Geomagic, Research Triangle Park, NC). The insertion site data was overlaid to the positional data and the functional length and percentages to the maximal resting length (MRL, length at maximal length during the ROM without external force) were calculated at all positions established by the robotic system. Results: During anterior translation, the functional lengths of MAP and middle of the AMB remained consistently long throughout the range of motion (ROM): 43.6-45.7 mm (MAP, 103-108% of MRL), 36.6-38.6mm (AMB, 99-105% of MRL). The functional length of MPP and the middle of the PLB decreased from 23.8 mm (MPP, 109%) and 28.7 mm (PLB, 107%) at 0 degrees to 16.1mm (MPP, 73%) and 22.0mm (PLB, 82%) at 90 degrees (Table 1.) Here 100% length is taken as the length between points under no external load at 0 degrees flexion. During the combined load, the ratio of functional length of MPP (108%) and middle of PLB (108%) of MRL were greater than the ratio of functional length of MAP (104%) and middle of AMB (106%) at 0 degrees of flexion. The ratio sequences were reversed at 15, 30, 45 degrees of knee flexion. Conclusions: The functional lengths of the anterior fibers of the ACL (MAP and AMB) during the anterior translation of tibia were maintained long throughout the ROM. But the functional lengths of posterior fibers of ACL (MPP and PLB) were long only at the extended positions. It can be suggested that anterior drawer test at 90 degrees would be positive only after a AM bundle rupture and Lachman test positive after both bundles are ruptured. Functional length changes of the MPP and PLB were higher than MAP and AMB during the combined force application. It can also be proposed that the PLB might have a more important role during the pivot shift test.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P13-1220 Attachment areas of the 3 ACL bundles Otsubo H.1, Shino K.2, Suzuki D.1, Kamiya T.1, Suzuki T.1, Watanabe K.1, Yamashita T.1 1 Sapporo Medical University School of Medicine, Department of Orthopaedic Surgery, Sapporo, Japan, 2Osaka Prefecture University, Faculty of Comprehensive Rehabilitation, Habikino / Osaka, Japan Objectives: Normal anterior cruciate ligament (ACL) could be divided into three fiber bundles: anteromedial(AM), intermediate (IM), and posterolateral(PL) bundles. In 2005, Shino developed an anatomic triple-bundle method reconstructing three bundles by dividing anteromedial bundles of conventional double-bundle reconstruction into medial and lateral bundles, focusing on the fact that normal ACL has anterolateral distribution of fibers on the tibial side. The purpose of this study was clarification of arrangement of ACL three fiber bundles in the above mentioned identified ACL attachment sites. Methods: Seven unfixed human frozen knees were used. Macroscopical investigation was performed to clarify the attachment sites of three ACL bundles in the area. Fibers were bluntly divided in the middle of ACL at 90 degrees of knee flexion. Each fiber bundle was carefully detached from the femur and tibia; distribution of attachment sites of each fiber bundle was observed. To compare of the distribution pattern and area of each attachment site, we took images of both tibial and femoral attachment site with digital camera. The obtained images were used to measure each attachment area with image analysis software. Results: In all knees, ACL was able to be divided by three fiber bundles. All bundles consisting of AM, IM and PL were running parallel to each other from inside observation at knee extension. On the femoral side, PL was attached on the distoposterior areas; IM was attached distoanterior to AM. In tibial bundles, AM, IM and PL were attached on the anteromedial, anterolateral and posterolateral aspects, respectively. The average attachment area of ACL was 124.6 mm2. The proportions of the three bundles were 29, 28, and 43% in AM, IM and PL, respectively. In tibial bundles, the average attachment area was 119.1mm2 showing 29, 26 and 45% as proportions of AM, IM and PL. Conclusions: We indicated arrangement of three fiber bundles in ACL. Our findings are useful to determine the tunnel placing in the anatomical triple bundle ACL reconstruction.

P13-1221 Computer aided system for knee biomechanical assessments: preliminary results Grimard G.1, Mezghani N.2, Fuentes A.3, Ouakrim Y.2, Hagemeister N.4, Ranger P.5, Lavigne P.6, Baillargeon D.7, de Guise J.A.4 1 Hoˆpital Sainte Justine, Orthope´die, Montreal, Canada, 2E´cole de Technologie Supe´rieure, Montreal, Canada, 3Universiite´ de Montreal, Medicine, Montreal, Canada, 4E´cole de Technologie Supe´rieure, GPA, Montreal, Canada, 5Hoˆpital du Sacre´-Coeur de Montre´al, Orthope´die, Montreal, Canada, 6Hoˆpital Maisonneuve-Rosemont, Orthope´die, Montreal, Canada, 7Hoˆpital de la Cite´-de-la-Sante´, Orthope´die, Montreal, Canada Objectives: Consultations for knee problems are increasing. Integrating a knee functional assessment into the clinical orthopedic assessment could be very beneficial for health professionals. Most current measurement methods to assess knee joint function are however subjective and based on patient self-report and therefore not always useful in the decision making process. A reliable and validated tibio femoral tracking device (KneeKGTM system) showed to accurately and objectively quantify the knee joint function (knee biomechanical patterns). However, associating a knee biomechanical pattern to a specific pathology can be time consuming and often requires the help of an expert. Therefore, there is a need to develop a computer method capable of distinguishing knee biomechanical patterns of patients with different knee pathologies to help health professionals in their assessment process and to provide a more functional objective assessment. Methods: 29 anterior cruciate ligament (ACL) deficient patients, 53 patients with knee osteoarthritis (OA), 18 patients with patellofemoral syndrome and 31 asymptomatic participants, all confirmed by an experience clinician, took part in a functional assessment consisting of collecting

S213 knee biomechanical patterns using the KneeKGTM system during treadmill walking. An automatic computer method based on wavelet decomposition and on data projection was then used to assign biomechanical data to the corresponding class (i.e. ACL deficiency, knee OA, patellofemoral syndrome and asymptomatic). The total number of patient correctly classified in each class evaluated the accuracy of the method. Results: The proposed classification method obtained a global accuracy of 84.1%. The classification accuracy per class reached 74.2%, 82.8%, 84.9% and 94.4% for the asymptomatic participants, ACL deficient patients, knee OA patients and patellofemoral syndrome patients respectively. Conclusions: These preliminary results show that using the KneeKGTM system to assess the knee joint function combined with an automatic objective computer aided method can help clinicians in their decision process. This new clinical tool could have a major impact in managing knee problems.

P13-1231 Modification in transtibial technique enables oblique femoral tunnel and graft in ACL reconstruction Lee M.C.1, Choi W.C.1, Lee S.1, Seong S.C.1 1 Seoul National University College of Medicine, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: A concern over inadequate femoral tunnel (FT) position after transtibial single-bundle ACL reconstruction (SB-ACLR) has been a matter of debate. We evaluated the FT position after transtibial SB-ACLR by arthroscopic and MRI findings. Clinical outcome was also evaluated. Methods: Consecutive 76 patients who underwent SB-ACLR with autologous quadriceps tendon and followed-up for 2 years or more were enrolled in this prospective study. Tibial tunnel was drilled from far medial, just anterior to MCL, and FT was drilled transtibially, rotating the femoral guide externally and aiming horizontally as much as possible with drawing the tibia anteriorly. From arthroscopic view, FT size and shape were evaluated and tunnel position was measured by 2 clock positions; Clock 1, which the 3- to 9-o0 clock axis corresponding to the joint line and Clock 2, which located conventionally within the intercondylar notch. Angles between graft axis and joint line on coronal (coronal graft angle;CGA) and sagittal (sagittal graft angle;SGA) plane were measured and compared to normal ACL obliquity measured from MRIs of matched control group. We also evaluated instrumented and clinical stability and outcome scales. Results: Arthroscopic findings showed ovoid shape FT covering center of ACL footprint in most of the cases. Distance from FT border to articular cartilage was mean 5.3mm and diameters of FT were mean 10.9mm and 16.2mm for high-to-low and deep-to-shallow direction, respectively. FT located on 10:34 position for a right knee measured by Clock 1 and 10:04 by Clock 2. The CGAs were mean 52.2, 54.6 and 66.2 on plain AP, tunnel view and oblique coronal view of MRI, respectively, and the SGA was mean 61.4. Comparison with control group showed no differences between CGA/SGA and normal ACL obliquity (p=0.119, p=0.068). Mean side-to-side difference measured by KT-1000 arthrometer changed from 3.9mm preoperatively to 1.8mm postoperatively (p\0.001). Lysholm (from 67.5 to 91.2) and subjective IKDC score(from 66.8 to 87.7) improved (p\0.001) after ACLR. Conclusions: With our modification in transtibial technique; adjusting tibial tunnel entrance, changing femoral guide direction and tibial position; oblique femoral tunnel and graft orientation similar to normal ACL, and satisfactory clinical result achieved after SB-ACLR.

P13-1239 Clinical correlation of navigation system for intra-operative evaluation of accurate placement of bone tunnels in reconstruction of the anterior cruciate ligament - 4 yrs follow up Bhattacharyya M.1, Gerber B.1 1 University Hospital Lewisham, Orthopaedic Department, London, United Kingdom Objectives: During ACL reconstructive procedure, the exact placement of drilled tunnels influence the outcome of surgery such as range of motion,

123

S214 knee joint stability, reaction of the synovium in the knee, pain, impingement and potential graft failure. 70% of ACL reconstructions are also carried out by orthopedic surgeons, who perform limited number of procedures in a year with or without arthroscopy [1]. The preciseness of the tibial tunnel placement was evaluated, and the advantages of this navigation system for open technique ACL reconstruction are discussed with clinical follow up. Methods: We performed 57 consecutive ACL reconstruction procedures in our hospital. Patients were sequentially assigned a standard single bundle ACL reconstruction with lateral third of the patellar ligament with a lateral incision open technique. In all ACL reconstruction procedures orthopilot navigation system was performed. The patients who had undergone ACL reconstruction using this system were evaluated regarding the positioning of the tibial tunnel against Blumensaat0 s line using XR and the route of the graft by magnetic resonance imaging (MRI). Results: Kinematic navigation enables us to measure anteroposterior and rotational knee stability, isometry, impingement and the angles of bone tunnel placement. At the 2 year follow-up, maximally extended lateral knee xray revealed that the anterior edge of the tibial tunnel and Blumensaat0 s line were almost aligned and that roof impingement was avoided.The ratio of the distance between Blumensaat0 s line and the anterior edge of the tibial tunnel at the level of the tibial plateau to the anteroposterior width in fully extended true lateral radiographs was 2.3% ± 2.4%. Conclusions: The computer-assisted navigation system improves accuracy and decreases dispersion of the tibial tunnel placement against Blumensaat0 s line in single-bundle ACL reconstruction. Computer assisted surgery allows the reconstruction procedure more reliable, eliminating the problem of skeletal variation among patients. Although, during operation, the planning of the insertion points and accurate drilling of the transosseous tunnels is difficult with the help of the jig, the correct placement of the graft, especially the isometry of the tibial and femoral insertion points, is successfully achieved with the orthopilot software for the navigation developed for use of anterior cruciate ligament (ACL) reconstruction. It is a user-friendly navigation system for intraoperative acquisitions of anatomical and kinematic data.

P13-1240 The relationship between graft tunnel position and in vivo knee kinematics after ACL reconstruction Wang J.H.1, Tashman S.2 1 Korea University Ansan Hospital, Orthopaedic Surgery, Ansan-si, Korea, Republic of, 2University of Pittsburgh Medical Center, Orthopaedic Biodynamics Laboratory, Pittsburgh, United States Objectives: Accurate tunnel positioning of anterior cruciate ligament (ACL) reconstruction is important to restore the normal anatomy and normal function of ACL. The purpose of this study was to evaluate the effect of the femoral and tibial ACL graft tunnel positions on knee kinematics during a high-loading, functional task. Methods: 24 patients underwent kinematic tests during downhill treadmill running (2.5m/s; 10 slope) using dynamic radiostereophotogrammetric analysis (D-RSA; 250 frames/s) one year after primary ACL reconstruction. Tibio-femoral kinematics were determined for the period from heel strike to mid-stance. Descriptive variables (min, max, mean, range) were extracted for tibiofemoral translations (anterior-posterior, medial-lateral) and rotations (internal/external, ab/adduction). 3D bone models were generated from CT scans. Centroids were calculated from the boundaries of the tibial and femoral tunnel apertures and expressed in anatomical coordinate systems using Geomagic software. Sagittal femoral tunnel position was determined using the grid system (described by Bernard et al.), with origin at the high-deep corner. Medial-lateral position was defined relative to the anatomic origin (midpoint between the femoral condyles). Tibial tunnel position was expressed relative to a grid defined by the anterior-posterior and medial/lateral dimensions of the tibial plateau, with its origin at the posteromedial corner. Pearson’s correlations were determined for each kinematic variable relative to the three femoral and two tibial tunnel positions. Significance was set to p\0.05.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: Significant correlations were found for the High/low (anterior/ posterior) tunnel positions (Y-axis) relative to internal/external rotation. High (anterior) position of the femoral tunnel was correlated with reduced external rotation (max: p=0.049; mean: p=0.022). Posterior position of the tibial tunnel was correlated with reduced external rotation (max: p=0.009; min: p=0.002; mean: p=0.004). Lateral position of the femoral tunnel was correlated with increased the range of medial/lateral translation (p=0.037). Conclusions: The combination of a more anteriorly placed femoral tunnel and a more posteriorly placed tibial tunnel would result in a sagittal-plane graft angle at the time of heel strike (near full extension) that is considerably more vertical than expected for an anatomically placed ligament or graft. It is possible that a vertically oriented graft positioned closer to the axis of the knee might pull the joint less towards external rotation. We also found that more laterally shifted femoral tunnels were associated with increased medial-lateral instability. Notchplasty was done for the most of cases in this series. These findings suggest that a large notchplasty and/or a wider notch can adversely affect medial/lateral stability after ACL reconstruction. In summary, the internal/external rotation and medial/lateral translation during a high-loading, functional task were affected by the femoral and tibial ACL graft tunnel positions.

P13-1241 Anatomic reconstructions of anterior cruciate ligament with double bundle graft: study of the diameter of AM and PL bundles in two surgical techniques Lefevre N.1, Robert H.2, Cassard X.3, Lanternier H.4, De Polignac T.5, Collette M.6, Herman S.7 1 Institut de l’Appareil Locomoteur Nollet, Paris, France, 2Centre Hospitalier Nord Mayenne, Orthopaedic Department, Mayenne, France, 3 Clinique des Ce`dres, Blagnac, France, 4Polyclinique de l’Europe, Saint Nazaire, France, 5Clinique Ge´ne´rale, Annecy, France, 6Clinique Edith Cavell, Bruxelles, Belgium, 7Centre Me´dico-chirurgical Paris V, Paris, France Objectives: Anatomic reconstructions of ACL with double bundle gracilis and semitendinosus tendons graft, reproducing anteromedial (AM) and posteromedial (PL) bundles, have been introduced to give a better biomechanical outcome. Nevertheless this reconstruction is not always possible due to the quality of the graft, (the length and thickness can be insufficient). In a double bundles reconstructions with 4 strands hamstring tendon, the diameter of the PL bundle is sometimes less than 6 mm and less than 7 mm for the AM bundle. So the ACL transplant can be fragile whereas with CoLS concept (FH OrthopedicsTM), the ACL transplant are thicker and more resistant. Methods: At the beginning of 2008, we operated on 15 patients with a complete rupture of the ACL, using the standard CoLS technique for each bundle. Gracilis and semitendinosus tendon are used to make two short (50-55 mm) closed loop (4 strands each). An out-in arthroscopic guide was designed for tibial and femoral drilling to guarantee a precise position of PL and AM bundles. Femoral and tibial bone sockets are created by retrograde drilling. Textile tapes are passed through both graft loop’s ends, to pull the graft into the femoral and tibial sockets and are fixed to the bone by 4 metallic screws. At first the PL bundle was fixed with manual tension, the knee in extension. Secondly the AM bundle was fixed at about 45  of flexion with the same tension. We measured the diameter of each bundle, and counted the number of strands. We compared the results with 15 similar cases with an in-out surgical technique, using hamstring fixed with EndobuttonCL devices (Smith&NephewTM) and bio-absorbable interference screws. Results: No intraoperative and postoperative complications were experienced in either group. There were no significant differences concerning the time of operation among the 2 groups. The average diameter of the PL was 6,2 mm for Endobutton group and 7,9 mm for CoLS group (p\0,001). The average diameter of AM was 7,4 mm for Endobutton group and 9,2mm for TLS group (p\0,001). Even though, the size of the graft was more important, there was no conflict in the femoral intercondylar notch and the full extension was possible. Conclusions: This technique had already shown excellent results in the reconstruction simple bundle. With the reconstruction double bundle, we

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 obtain a large transplant caliber. Long-term results should confirm the efficacy of this double bundles reconstruction. P13-1246 Femoral interference screw position in transtibial ACL reconstruction: how to optimize alignment with a mathematical model. A cadaveric study Marmotti A.1, Castoldi F.2, Rossi R.3, Biondi A.4, Collo G.5, Tellini A.1, Germano M.1, Bignardi C.6 1 Mauriziano Hospital, University of Torino, Department of Orthopaedics and Traumatology, Torino, Italy, 2University of Turin, Mauriziano Hospital, Orthopaedics and Traumatology, Turin, Italy, 3University of Torino, Mauriziano Umberto I, Torino, Italy, 4University of Padova, Department of Innovation in Mechanics and Management, Padova, Italy, 5 Mauriziano Hospital, University of Torino, Torino, Italy, 6Politecnico Torino, Torino, Italy Objectives: Femoral interference screw divergence is a potential pitfall associated with ACL reconstruction through transtibial tunnel technique, as angles greater than 15 jeopardize graft fixation. Our mathematical model theorizes the proper degrees of knee flexion during femoral screw insertion and the correct screwdriver position to obtain a minimal divergence of the screw in the femoral tunnel. The cadaveric study confirms our method. Methods: mathematical model: using Transformation Matrices, a correlation is demonstrated between the ACL-tibial-guide angle, the knee flexion and the screwdriver position. A theoretical minimal divergence between femoral interference screw and femoral tunnel is obtainable following these assumptions: 1) knee hyperflexion during femoral screw insertion is obtained adding a flexion corresponding to the ACL-tibial-guide angle to the flexion while drilling the femoral tunnel 2) screwdriver position (through the AM portal) is kept parallel to tibial plateau at a rotation of 15 medial to tibial sagittal plane. Cadaveric study: 24 cadaver knees were used. Transtibial tunnel was drilled with a 8 mm drill bit with the help of a ACL tibial guide set at 55. To simulate femoral tunnel direction, a 2,4 mm K. wire was drilled through the femur with a transtibial 7 mm offset femoral drill guide. To simulate femoral screw direction, a second 2,4 mm K. wire was drilled from the femoral entry point of the first wire through the femur, with a cannulated screwdriver. Screwdriver direction and knee flexion during the simulation were obtained following two different methods: In GROUP A (mathematical model group, 12 knees), screwdriver direction and knee flexion were calculated following the mathematical model; in GROUP B (control group, 12 knees), a maximal knee hyperflexion and a generalized screwdriver medialization were manually obtained by a senior surgeon, as in the standard transtibial ACL reconstruction technique. Results: Divergence between femoral interference screw and femoral tunnel was identified as the angle formed by the two wires, measured on the plane formed by the direction of the wires. Mean divergence angles between the K. wires were significantly different (p\0,005) between the groups: GROUP 1 (mathematical model): 7,25 (± 2,2); GROUP 2 (control): 16,8 (± 2,3). Following the mathematical model we have been able to obtain a minimal divergence, lesser than the aforementioned limit of 15. Conclusions: Our study shows that a proper alignment between the femoral tunnel and the screwdriver can be achieved simply following a mathematical rule for (1) correct knee flexion and (2) screwdriver position without any specialized instrumentation. Namely, during femoral screw insertion through anteromedial portal: 1) Correct knee hyperflexion is the sum of the knee flexion angle while drilling the femoral tunnel and the ACL tibial guide angle 2) Correct screwdriver position is parallel to tibial plateau, engaging the femoral tunnel at a rotation of 15 medial to tibial sagittal plane. Furthermore, following the assumptions of this study, a starting knee flexion angle around 70 during femoral tunnel drilling seems preferable for ACL reconstruction with transtibial tunnel technique. As ACL-tibialguide angles ranges commonly from 50 to 60 and in vivo the maximal intraoperative knee flexion attainable is 130, a starting knee flexion around 70 easily allows additional flexion angles up to 60 before reaching the limit value of 130.

S215 P13-1253 There is still life over 50: ACL reconstruction in middle aged patients Osti L.1, Papalia R.2, Del Buono A.2, Leonardi F.2, Maffulli N.3, Denaro V.4 1 Hesperia Hospital, Orthopedic Surgery, Modena, Italy, 2Campus Biomedico University, Orthopaedic Surgery, Rome, Italy, 3Keele University, Institute of Science and Technology in Medicine, Stoke on Trent, United Kingdom, 4Campus Bio-Medico University, Orthopaedic Surgery, Rome, Italy Objectives: The management of middle aged patients with symptomatic anterior cruciate ligament (ACL) deficiency is still controversial. We compared the outcomes of patients older than 50 years (case group) with a control group of subjects younger than 30 years at minimum follow-up of 24 months. Methods: 20 patients (12 men and 8 women) were enrolled in both groups. Lachman test, pivot shift test and KT1000 arthrometer at manual maximum stress were used to measure pre- and post-operative ACL laxity. Clinical functional evaluation was assessed according to IKDC Committee (IKDC) subjective knee form, IKDC ligament standard evaluation, and Lysholm score. Results: At 2 years follow up, all variables significantly improved in both group, compared to pre-operative values (P\.005), with no significant intergroup difference (P[.005). Conclusions: When faced with ACL deficiency, physiological age, condition of the knee at the time of examination, expectancies of life and activity level are probably more important than chronologic age. Age in itself is not a contraindication to surgery.

P13-1256 ACL double bundle reconstruction with semitendinosus-gracilis and tape locking screw technique: preliminary results of a prospective study Lefevre N.1, Cassard X.2, Lanternier H.3, De Polignac T.4, Robert H.5, Collette M.6, Herman S.7 1 Institut de l’Appareil Locomoteur Nollet, Paris, France, 2Clinique des Ce`dres, Blagnac, France, 3Polyclinique de l’Europe, Saint Nazaire, France, 4Clinique Ge´ne´rale, Annecy, France, 5Centre Hospitalier Nord Mayenne, Orthopaedic Department, Mayenne, France, 6Clinique Edith Cavell, Bruxelles, Belgium, 7Centre Me´dico-chirurgical Paris V, Paris, France Objectives: Analyse and provide preliminary results for a continuous prospective series of ACL repairs using FH orthopedicsTM CoLS doublebundle semitendinosus-gracilis reconstruction. Methods: We operated on 47 patients between May and December 2008, 42 males and 5 females. The average age of the series at the time of the operation was 27.8 years (15 to 44 years). Our study only included complete and isolated ACL ruptures with no associated ligamentary lesions. We used the same CoLS double-bundle technique for each patient. The semitendinosus and gracilis tendons were used to perform two short closed loop, 4-strand grafts, 45 to 55 mm in length. A retrograde arthroscopic technique enabled us to drill the 4 tunnels with precision. The graft was secured using fabric tapes, secured using 4 titanium screws, with the knee in extension for the PL and at 45 for the AM. We assessed the clinical results using the Lysholm score, IKDC (International Knee Documentation Committee) score, manual Lachman test and pivot shift test. We took a laxity measurement using the radiological TELOS technique. Results: There were 25 meniscal or bimeniscal lesions out of 47 cases (53% of knees operated on). 17 were internal meniscus lesions (36%), 3 were external meniscus lesions (6%), and 5 were bimeniscal lesions (10%). We carried out 12 menisectomies (25%), 8 conservative treatments (17%) and 5 meniscal sutures (10%). We encountered 2 complications: a sepsis at 3 weeks post-op, treated by lavage and prolonged antibiotic treatment (staphylococus epidermidis) and one algodystrophy of the knee with stiffness at 3 months requiring mobilisation under GA. There were no cases of Cyclops syndrome and no extension deficit. With an average follow-up of 9.6 months (6.3 to 13.4 months), only 19 files were useable,

123

S216 notably with pre- and post-op Telos assessments. The Lysholm score increased from 65.6 to 92.8 post-op. The IKDC score at the latest followup was 66% group A, 24% group B, 8% group C and 2% group D. The manual Lachman-Trillat test was negative in 77% of cases, positive grade 1 in 21% of cases and grade 2 in 2 % of cases. The pivot shift test was negative in 91% of cases, grade 1 in 7 % of cases and grade 2 in 2% of cases. The result of the Telos assessment at 15 kg revealed an average differential of 2.9 mm (0 to 7 mm). We had one traumatic rupture of the graft at 7 months post-op, following a fall down stairs. The patient was revised with a KJ-Lemaire. Conclusions: The CoLS double-bundle reconstruction technique is a reliable and repeatable technique for any patient and any hamstring transplant. Short-term results are good. Long-term results should confirm the effectiveness of this double-bundle technique.

P13-1257 Review of 29 cases of ACL rupture treated arthroscopically in Hospital Nossa Senhora do Rosa´rio, Barreiro, Portugal Canilho B.1, Lanc¸a N.1, Geada N.1, Franco J.1, Martins J.1, Zurbano J.1, Moreno J.1 1 Hospital Nossa Senhora do Rosa´rio, Servic¸o de Ortopedia, Barreiro, Portugal Objectives: Evaluate functional outcome of anterior cruciate ligament bone-tendon-bone and hamstrings ligamentoplasty procedures in Hospital Nossa Senhora do Rosa´rio, using IKDC and Tegner-Lysholm scores. Methods: Parallel to the generalization of sports practice, there is a potentially rise in the incidence in knee injuries and particularly ACL lesions. We present a review study of 29 patients out of 41 with ACL ruptures treated arthroscopically from 2001 to 2009 in our hospital with two techniques: Single band Bone-Tendon-Bone (27 patients) and Hamstring (2 patients) ligamentoplasty. Graft fixation procedure with bioabsorbable interference screw and bioabsorbable pins. Patients were assessed using Tegner-Lysholm and IKDC scores. Results: The average age of the population studied was 27,03 years. (18 51), 25 # - 4 $ . 58,6% of the lesions treated occurred during soccer practice. The average follow-up was of 48 months (6 - 94 months). We found an average fair result in the subjective Tegner-Lysholm score (81,17) and an average IKDC subjective knee evaluation score of 77,76. There is a strong positive correlation between both subjective scores (r=0,83).The average objective IKDC score was grade B (nearly normal). There was poor correlation between the objective and subjective scores (r=-0,07). Conclusions: It was not possible to compare both techniques due to minor numbers of Hamstring procedures. Although the small series, we had satisfactory results with both techniques, and fair functional outcomes.

P13-1273 Correlation between timing of anterior cruciate ligament reconstruction and tibial rotation Zampeli F.1, Ristanis S.1, Xergia S.1, Georgiou S.1, Stergiou N.2, Georgoulis A.1 1 Orthopaedic Sports Medicine Center, University of Ioannina, Department of Orthopaedic Surgery, Ioannina, Greece, 2HPER Biomechanics Laboratory, University of Nebraska at Omaha, Omaha, United States Objectives: There is inadequate evidence about the appropriate time at which the reconstruction of an anterior cruciate ligament ruptured knee should be performed. The purpose of the study was to investigate the correlation between the timing of anterior cruciate ligament reconstruction and the rotational knee kinematics. Methods: Fifteen patients with a unilateral ACL rupture that underwent ACL reconstruction with a BPTB autograft were included in the study. The patients were examined at an average of 18 months postoperatively. An 8-camera optoelectronic motion analysis system was used to collect kinematic data while the participants descended stairs and after foot contact, pivoted on the landing (operated) leg at 90 degrees. The range of

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 motion of tibial rotation was calculated for the pivoting period. The Pearson0 s r correlation coefficient between the time from injury to reconstruction and the tibial rotation was calculated. Results: The results showed a positive correlation of moderate strength (Pearson’s correlation coefficient r = 0.58, p=0.01) between the timing of ACL reconstruction and the postoperative tibial rotation during a high demanding task. Conclusions: These results suggest that there is a relationship between the timing of ACL reconstruction and the tibial rotation of the operated knee. Our results indicate that as the time from injury to operation increases, the patients show higher tibial rotation values postoperatively during a high demanding activity.

P13-1276 Serial dilation reduces tibial graft slippage compared to extraction drilling in anterior cruciate ligament reconstruction: a randomized controlled trial using radiostereometric analysis Sørensen O.G.1, Larsen K.2, Wulff Jakobsen B.3, Kold S.4, Hansen T.B.2, Lund B.1, Christiansen S.E.1, Lind M.1, Søballe K.1 1 Aarhus University Hospital, Department of Orthopaedics, Aarhus, Denmark, 2Hospital Unit West, Department of Orthopaedics, Holstebro, Denmark, 3Science Center Skejby, Eira Private Hospital, Aarhus N, Denmark, 4University Hospital of Aalborg, Department of Orthopaedics, Aalborg, Denmark Objectives: The hamstring tendon graft has become increasingly popular in anterior cruciate ligament (ACL) reconstruction because of low donorsite morbidity. However, the tibial fixation is considered difficult, partly because of low tibial mineral bone density. Therefore, we tested whether preparation of the tibial tunnel with compaction by serial dilation provided a stronger anchorage of the graft-fixation-device complex compared to traditional extraction drilling of the tibial tunnel. Methods: 40 patients (22 males and 18 females) undergoing ACL reconstruction were randomized to either extraction drilling (group ED) or compaction by serial dilation (group SD) of the tibial tunnel. The hamstring graft was anchored with Retrobutton and a supplementary interference screw (Arthrex) in the femur and a Delta interference screw (Arthrex) in the tibia. Tantalum beads were placed in both the proximal part of the tibia and distal part of the femur. Beads were placed in the hamstring graft at the fixation sites as well. Radio stereometric analysis were performed post-operatively and again after 6, 12 and 24 weeks. The ACL-reconstructed knee was stressed with a Telos stress device. Migration of the tantalum markers in the graft was measured in reference to the bone markers in the tibia and femur. Knee laxity was assessed at every follow by measuring the relation of the tibial bone markers to the femoral bone markers in both the anterior and the posterior stress position. Results: Six patients (3 males and 3 females) were excluded during followup, which resulted in 17 patients in group ED (mean age: 32.5 years (range: 20 - 50)) and 17 patients in group SD (mean age: 32.0 years (range: 20 - 49)). The mean migration of the graft at the tibial fixation site after 3 months was 1.3 (sd 0.6) mm, in group ED and 0.8 (sd 0.5) mm in group (P = 0.02). The knee laxity after 3 months was 13.0 (sd 4.0) mm in group ED and 10.9 (sd 3.1) mm in group SD (P=0.09). Conclusions: This study found a significant smaller mean migration of the hamstring graft at the tibial fixation site in the serial dilated group compared to the extraction drilling group. No significant difference in stress radiographic knee laxity was found between the 2 groups.

P13-1279 Double bundle anterior cruciate ligament reconstruction with semitendinosus tendon Lee J.H.1, Park J.H.1, Bae H.K.1, Lim Y.J.2 1 Chonbuk National University School of Medicine, Chonbuk Nati, Orthopedic Surgery, Jeonju, Korea, Republic of, 2Saint Carollo Hospital, Orthopedic Surgery, Sunchon, Republic of Korea Objectives: The purpose of this study was to compare the clinical outcome between double-bundle (DB) and single-bundle (SB) anterior

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 cruciate ligament reconstructions (ACLR) with semitendinosus tendon (ST) alone. Methods: We retrospectively reviewed 30 consecutive patients with a mimimum of 2 year follow up among 51 patients who underwent DBACLR using ST since September 2006 and compared its clinical results to whom SBACLR was performed in same number of patients. Lachman test, pivot shift test and KT-1000 arthrometer (MEDmetric, San Diego, CA) for stability, isokinetic strength of knee extensor and flexor with cybex test (Lumex, Ronkonkoma, NY), and general knee assessment with International Knee Documentation Committee (IKDC) evaluation form and Lysholm score were performed for evaluating the clinical outcome after reconstruction. Statistical analysis was performed with SPSS (Windows version 12.0, Chicago, Illinois) and a p value of less than 0.05 was considered statistically significant. Results: There were no differences between two groups with regard to age, gender, elapsed time to surgery from injury and correlation with sports injury (P[0.05). At the last follow up, no significant differences were also found between the two groups with regard to Lysholm score, IKDC score, cybex testing and Lachman test (P[0.05). But, DB group had better results than the SB group in according to pivot shift test (P=0.007) and KT 1000 measurements (1.7±1.0 mm vs 2.3±0.9 mm) (P=0.024). Conclusions: Either double bundle or single bundle anterior cruciate ligament reconstruction using semitendinous tendon shows satisfactory clinical outcome but double bundle technique is more stable with regard to stability. So, double bundle anterior cruciate ligament reconstruction is thought to be a viable option to improve the stability in ACL-deficient knee.

P13-1283 Comparative imagistic study between transtibial and anatomical single bundle ACL reconstruction using 64 slice ct reconstruction Prejbeanu R.1, Vermesan D.1, Barsasteanu F.1 1 University of Medicine and Pharmacy Timisoara, Timisoara, Romania Objectives: Though considered a successful technique for more than 20 years, transtibial ACL reconstruction is less used nowadays because the position of the tibial and femoral tunnels is not the one of the anatomically, natively positioned ACL and thus cannot fully restore its function. A series of recent studies of the double bundle technique has shed new light on the anatomical landmarks of the two ACL insertions (tibial and femoral). Methods: From this new data we made a imagistic comparative study between two groups of patients: A (20 patients) - transtibial ACL reconstruction and B (20 patients) - anatomical ACL reconstruction. The patients were evaluated using 64 slice CT postoperatively. We compared the accuracy of the ACL graft insertion sites compared to the findings from the literature from recent years (Fu 2007, Ferreti 2008, Kopf 2009). Results: In group A only 8 (40%) patients had tibial insertion overlapping the anatomical position and 4 (20%) the same situation for the femoral insertion. In addition, these 4 patients had the graft insertion in the region of the AM fascicle of the native ACL, which made it not anatomically positioned. In group B, 18 patients (90%) had a correct tibial graft positioning and 17 (85%) femoral insertion in the middle of the anatomical footprint of ACL insertion. Conclusions: In conclusion, we can say that the transtibial technique leads to a nonanatomical positioning of the graft and in the long term, a failure of the ACL reconstruction.

P13-1310 Intraoperative evaluation of anteroposterior and rotational stabilities in anterior cruciate ligament reconstruction: lower femoral tunnel placed single-bundle versus double-bundle reconstruction Kanaya A.1, Deie M.2, Adachi N.2, Nishimori M.2, Nakamae A.2, Ochi M.2 1 Mazda Co. Ltd., Mazda Hospital, Orthopaedic Surgery, Hiroshima, Japan, 2Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan Objectives: The purpose of this study was to evaluate AP and rotational stabilities using our original device and the navigation system for applying

S217 quantitative loads before and after ACL reconstruction, comparing the lower femoral tunnel placed single bundle ACL reconstruction with double-bundle ACL reconstruction intraoperatively. Methods: Twenty-six patients with anteroposterior (AP) laxity of the knee, associated with torn anterior cruciate ligament (ACL), were prospectively randomized for arthroscopic lower femoral tunnel placed single- or double-bundle reconstruction using hamstring tendons. We evaluated AP and rotational stabilities under regular loads (a 100-N anterior load and a 1.5-N m external-internal load) before and after ACL reconstruction, comparing single- and double-bundle reconstruction with our original device for applying quantitative tibial rotation and the navigation system intraoperatively. Results: No significant differences were found between the two groups in AP displacement and total range of tibial rotation at 30 degrees and 60 degrees of knee flexion. We found that a lower femoral tunnel placed single-bundle reconstruction reproduced AP and rotational stability as well as double-bundle reconstruction after reconstruction intraoperatively. Conclusions: This study showed that a lower femoral tunnel placed singlebundle reconstruction reproduced AP and rotational stability as well as double-bundle reconstruction after reconstruction intraoperatively. Although the exact clinical importance of these findings is unknown, our current data suggest that we may not need to persist in double-bundle reconstruction as long as the single-bundle reconstruction is performed with lower femoral tunnel placement.

P13-1313 Role of the healing response technique in the treatment of partial ACL-ruptures Fehske K.1, Ziai P.2, Eichhorn H.J.3 1 University of Wuerzburg, Trauma Surgery, Wuerzburg, Germany, 2 General Hospital of Vienna, Vienna, Austria, 3Sporthopaedicum Straubing, Straubing, Germany Objectives: Since the early 19900 s there have been discussions concerning a minimal-invasive technique to treat partial acl-ruptures. The technique was first published from Steadman et al (Steadman, J. R.; CameronDonaldson, M. L.; Briggs, K. K.; Rodkey, W. G; ‘‘A minimally invasive technique (‘‘healing response’’) to treat proximal ACL injuries in skeletally immature athletes’’, 2006, J Knee Surg, 19-1, Seiten 8-13) and has been developed over the last decade into an alternative to acl-reconstruction. Basically, a chondro-pick is used as a chisel. The proximal ruptured acl is harmed, as a result blood pours into the acl-stump together with tissue growth factors, which supposedly lead to the building of scarf tissue and consequently to a gain of stabilization. Methods: We have treated 102 of our patients between July 2005 and August 2008 with a modified healing response technique. We included patients with an acute trauma, a partial/complete acl-rupture shown in the MRI and a positive Lachman test with uncertain stability. We excluded patients, who already underwent acl-reconstruction or with complex instability (pcl-insufficiency, medial/lateral instability). Furthermore, we assigned the difference in the KT 1000 measurement to the non-injured knee should not be more than 4 mm and the pivot shift sign should not be higher than 1 ? (ideal would be a slight pivot glide). The patients should not suffer from a major instability in everyday life. The arthroscopic findings should not show a complete acl-rupture. In our study we had 61,8% male athletes, the average age was 31,09 ± 13,1 years. Results: The difference in the KT 1000 measurement could be lowered from preoperatively 2,76 ± 1,6 mm to postoperatively 1,02 ± 1,7 mm in the last postoperative control. So far 6 patients (5,9%) needed to undergo revision surgery. Five of them needed an acl-reconstruction. Two patients suffered from a new trauma, which lead to the acl-rupture. Our patients showed in our postoperative controls satisfying stability results und could return to their level of activity. Conclusions: The Healing Response Technique is a sufficient treatment for partial-acl ruptures, if the indication is drawn differentiated.

123

S218 P13-1316 Dynamic investigation of knee rotational stability following double bundle ACL reconstruction Tsarouhas A.1, Kotzamitelos D.1, Spyropoulos I.2, Chrysanthou C.1, Iosifidis M.3, Giakas I.2 1 G.H. Naoussa, Orthopaedic Surgery, Naoussa, Greece, 2University of Thessaly, Department of Physical Education, Thessaloniki, Greece, 3 Papageorgiou G.H., Orthopaedic Surgery, Thessaloniki, Greece Objectives: Tibial rotation has been used in the past to assess rotational knee instability following anterior cruciate ligament (ACL) reconstruction both in vitro and in vivo. However, studies measuring tibial rotation in the double-bundle ACL reconstructed knee under dynamic loading conditions are still limited. Aim: To investigate whether anatomic restoration of the ACL bundles results in significant reduction in transverse plane instability compared to the conventional single-bundle technique during dynamic functional testing. Methods: The study group included 14 patients who underwent doublebundle ACL reconstruction with hamstrings tendon autograft, 14 patients with single-bundle reconstruction, 12 ACL deficient subjects and 12 healthy control individuals. Kinematic and kinetic data were collected using an 8-camera optoelectronic motion analysis system and one force plate. Knee rotational stability was examined during two independent tasks: a combined 60 pivoting turn and immediate stairs ascend and a combined stairs descend and immediate 60 turn. During both maneuvers the supporting knee was held in extension. The two factors evaluated were the maximum range of internal-external tibial rotation and the maximum knee rotational moment of the supporting knee. Results: There were no significant differences in tibial rotation between the four groups in the examined maneuvers (ascend p=0.3, descend p=0.8). Tibial rotation in the single- and the double-bundle groups were even lower than the control group. Rotational moments did not differ significantly between the four groups in any of the examined maneuvers (minimum p=0.29). In general, rotational moments in the affected side of the ACL reconstructed and deficient groups were substantially but not significantly reduced compared to the unaffected side. Conclusions: Double-bundle reconstruction does not reduce knee rotation further compared to the single-bundle technique during dynamic stability testing under varying loading conditions. The injured side of ACL reconstructed or deficient individuals is exposed to substantially lower rotational moments compared to the intact side.

P13-1339 Reconstruction of the anterior cruciate ligament - dynamic strain evaluation of the graft Handl M.1, Kautzner J.1, Drzik M.2, Povysil C.3, Hanus M.1, Trc T.1 1 University Hospital Motol, Orthopaedic Clinic, Prague, Czech Republic, 2 International Laser Center, Bratislava, Slovakia, 3Charles University Prague, Institute of Pathology, Prague, Czech Republic Objectives: Hamstring grafts are more commonly used for ACL reconstruction. These grafts are better tolerated by the patients, their strength is superior to bone-patellar tendon-bone (BPTB) graft. Fixation of hamstring grafts is difficult and demanding. We assume that the best fixation method is a suspension technique that does not interfere with the graft and does not damage its structure. The purpose of our study is to determine the effect of the suspension fixation compared to the graft piercing transfixation. It is a biomechanical study using human cadavers. Methods: 19 fresh frozen human hamstring specimens from xx cadaveric donors were used. A specially constructed testing device was used to test the loading properties of the grafts. One half of the specimens was suspended over 3,3mm pin, while the other half was pierced by 3,3mm pin to simulate graft piercing technique. Single impact testing was performed, failure force, elongation and acceleration/deceleration of each graft was recorded. Results were then evaluated and loading force vs. elongation of the grafts specimens were calculated. Results for

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 suspended and pierced grafts were compared always comparing the grafts from each donor. Results: Femoral fixation methods using suspension of the graft have superior strength over graft piercing transfixation methods. The ultimate strength of two-strand gracilis was 1287±134 N for suspended over pin graft while the strength of pierced graft was 833±111 N and for the twostrand semitendinosus grafts are the strengths 1883±198 N and 997±234 N, respectively. Thus, the failure load for piercing methods is only 64.7% or 52.9% of suspension methods. Conclusions: Outcomes of this study may help to develop new fixation methods. Suspension techniques of fixation should be used for femoral hamstring graft fixation, because of their superior loading properties.

P13-1344 Femoral tunnel position in acl reconstruction - are we anatomical? A critical analysis using 3D CT scans Spalding T.1, Thompson P.2, Clewer G.1, Bird J.1, Dalton H.1 1 University Hospital Coventry, Coventry, United Kingdom, 2University Hospitals Coventry and Warwickshire NHS Trust, Trauma and Orthopaedic Surgery, Coventry, United Kingdom Objectives: Recent advances in understanding of ACL insertional anatomy has led to new concepts of anatomical positioning of tunnels for ACL reconstruction. Femoral tunnel position has been defined in terms of the lateral intercondylar ridge and the bifurcate ridge but these can be difficult to identify at surgery. Measurements of the lateral wall either using C-arm x-ray control or specific arthroscopic rulers have also been advocated. The aim of this study was to assess the femoral tunnel position at ACL reconstruction using 3D CT scans for two approaches and relate the findings to anatomical positioning. Methods: 30 patients undergoing ACL reconstruction before and after introduction of a new anatomical technique of ACL reconstruction were evaluated using 3D CT scan imaging with cut away views of the lateral aspect of the femoral notch and the radiological quadrant grid. In the new technique, with the knee at 90 degrees flexion, the femoral tunnel was centred 50% from deep to shallow as seen from the medial portal (Group A). Group B consisted of patients where the femoral tunnel was drilled through the antero-medial portal and offset from the posterior wall using a 5mm jig aiming for a 10.30/1.30 position on the clock face description. Results: Ridges were identifiable in only 76% of scans. All tunnels in Group A (anatomical technique) were found to be below (posterior to) the lateral intercondylar (residents) ridge and were within 10% of the optimal position as defined by the Grid method on x-ray. No femoral tunnels in Group B meet anatomical criteria and were malpositioned by a mean of 5mm. Conclusions: We believe 3D CT scan imaging with cut away views of the femoral tunnel is a useful and accurate way of describing tunnel position, and that this technique will be valuable in validating new surgical techniques. According to this CT scan analysis the new anatomical technique correctly placed the femoral tunnel. This work forms the basis of a subsequent randomised trial of techniques in relation to clinical outcome.

P13-1346 Revision anterior cruciate ligament reconstruction: etiology of failures and clinical results Sera S.1, Adachi N.1, Deie M.1, Nishimori M.1, Nakamae A.1, Ochi M.1 1 Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan Objectives: Revision anterior cruciate ligament surgery has become a more commonly performed operation. Reasons for this are an increased number of first-time operations, some with technical faults. A retrospective study was performed to determine the etiology of failed primary anterior cruciate ligament (ACL) reconstruction and evaluate the clinical results of revision ACL surgery.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: From April 2002 to December 2008, 36 patients with failed ACL reconstructions underwent revision ACL surgery and were studied retrospectively. The indication of the operation, the management during the procedure and the clinical results were analyzed. Results: The initial ACL reconstruction was performed with a hamstring (n=18), a patellar tendon (n=2) or an artificial ligament (n=16). Differential laxity (KT-knee arthroeter) was 3.5 ± 3.3 mm. Mean time to revision surgery was 8.1years. The etiology of failed ACL reconstruction included 30 surgical technical errors, 3 traumatic reinjuries, 3 related to synthetic grafts. Substitutes of revision ACL surgery were a patellar tendon (n=1), a quadriceps tendon (n=3), a hamstring tendon (n=32) and an allograft (n=1). Outcome was assessed at 12 months follow-up. The mean Lysholm score was 94.1. All patients are negative of Lachmann test and Pivot shift test. Mean differential laxity measured with KT knee arthrometer was -1.2± 3.2 mm. Conclusions: The functional results and objective stability could be obtained through revision ACL reconstruction. However, outcomes are less favorable than those with primary reconstructions, which might be related to the factors including deficiency of previous surgery and impact of revision procedure.

P13-1347 Kinematic and kinetic analysis of knee rotational stability following single and double-bundle acl reconstruction Tsarouhas A.1, Kotzamitelos D.1, Tsatalas T.2, Iosifidis M.I.3, Giakas I.2 1 G.H. Naoussa, Orthopaedic Surgery, Naoussa, Greece, 2University of Thessaly, Department of Physical Education, Trikala, Greece, 3 Papageorgiou G.H., Orthopaedic Surgery, Thessaloniki, Greece Objectives: Knee rotational stability after Anterior Cruciate Ligament (ACL) reconstruction has been recently investigated by in vitro studies. Published data concerning the in vivo kinematics of the double-bundle ACL reconstructed knee are still limited. Aim: To investigate in vivo the differences in knee rotational angles and moments between single- and double-bundle ACL reconstructions, using hamstrings tendon autograft. Methods: The study group included 10 patients who underwent doublebundle ACL reconstruction with hamstrings tendon autograft, 12 patients with single-bundle reconstruction, 10 subjects with chronic ACL deficiency and 10 healthy control individuals. The four groups did not differ significantly in terms of age, body mass index, duration of follow-up and number of meniscectomies performed. Kinematic and kinetic data were collected using an 8-camera optoelectronic motion analysis system and one force plate. The subjects performed a 60o pivoting maneuver with the supporting knee in full extension. Maximum range of internalexternal knee rotation and maximum knee rotational moment were examined. Results: There were no significant differences in knee rotation between the four groups concerning the range of internal-external knee rotation (p=0.13). The highest rotation was recorded in the ACL deficient group (15.1 ± 3.1). The control group showed increased tibial rotation (13.9 ± 4.1) compared to the single- (13.4 ± 4.7) and double-bundle (12.4 ± 3.3) groups. No differences were identified between the affected and the non-affected side in each of these groups (p=0.20). Side-to-side and between groups comparisons established substantial but not significant differences (p=0.34 and p=0.58 respectively) in knee rotational moment. The rotational moment applied to the affected side was constantly lower than the non-affected one in all four groups. No significant correlation was found between the angle and moment of knee rotation (r =0.13). Conclusions: Our study demonstrated no improvement in transverse plane instability when using a double-bundle technique for ACL reconstruction compared to the conventional single-bundle technique. Further in vivo research is needed to evaluate knee kinematics and kinetics under varying simulated load conditions.

S219 P13-1348 Tunnel widening in ACL reconstruction using allograft tibialis tendon vs autograft hamstrings Verdonk P.1, Mahieu P.2, Claes S.3 1 University Hospital Ghent, Dept. Orthopaedic Surgery and Traumatology, Gent, Belgium, 2Department of Orthopaedic Surgery and Traumatology, Gent, Belgium, 3Department of Orthopaedic Surgery and Traumatology, Pellenberg, Belgium Objectives: Tunnel widening has been described after ACL reconstruction and is explained by mechanical and biological phenomena. The aim of this study was to investigate tunnel widening using Computerized Tomography comparing 1 single biological variable: tibialis tendon allograft vs hamstrings autograft . Methods: 13 patients (4 allograft vs 9 autograft) were evaluated at a minimal follow-up of 1 year using 3D reconstructed CT scan for tunnel widening. All patients were operated using the same technique by a single surgeon (Endobutton on the femoral side, BioRCI and staple on the tibial side, Smith and Nephew). Tunnel widening was measured using a best-fit cylinder model in MIMICS software (version 13, Materialize). Results: On the femoral side, tunnels were enlarged by 50% vs 33% for allograft vs autograft, respectively. On the tibial side tunnels enlarged 44% vs 34%. These differences were significant. Conclusions: These results indicate that tunnel widening is significantly increased in tibialis tendon allografts ACL reconstructions vs autograft 1 year after surgery. The biology of graft healing is significantly influenced by the type of tissue. Revision complexity could be influenced by the presence of larger tunnels.

P13-1361 Biomechanical comparison of two anatomical ACL reconstructions: single-bundle versus single-femoral-tunnel double-tibial-tunnels Debandi A.V.1, Lu S.1, Lertwanich P.1, Hume C.1, Martins C.A.Q.2, Linde-Rosen M.1, Smolinski P.2, Fu F.2 1 University of Pittsburgh, Pittsburgh, United States, 2University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States Objectives: There has been a growing trend to perform more anatomical anterior cruciate ligament (ACL) reconstructions designed to reconstruct both bundles, the anteromedial (AM) and posterolateral (PL), of the ACL. Anatomic studies have shown that the tibial insertion of the ACL is the largest portion of the ligament. Therefore, by creating two tibial tunnels this area is better matched than with only one tunnel. The objective of this study was to compare knee kinematics and in situ force in the grafts between two anatomical ACL reconstructions: single-bundle (SB) versus single-femoral-tunnel double-tibial-tunnels (1FT-2TT) in response to externally applied loads. It was hypothesized that anatomic 1FT-2TT ACL reconstruction restores the intact knee biomechanics more closely than anatomic SB ACL reconstruction. Methods: Sixteen fresh-frozen porcine knees were subjected to a 89-N anterior tibial load (simulated KT-1000 arthrometer test) at 30, 60 and 90 and to 4-N•m internal tibial torque and 7-N•m valgus torque (simulated pivot-shift test) at 30and 608 of knee flexion. Resulting anterior tibial translation (ATT) and in situ force in the ACL and grafts were determined by using a robotic/universal force-moment sensor testing system for intact, ACL-deficient and ACL-reconstructed knee. Both anatomical ACL reconstructions were performed arthroscopically drilling a single femoral tunnel in the center of ACL footprint. In SB group, a single tibial tunnel was placed in the center of ACL insertion site, whereas two tibial tunnels for AM and PL footprints were drilled in 1FT-2TT group. Bone-patellar tendon allograft was used positioning the bone plug on femoral site and the tendon on one tibial tunnel (SB group) or two tibial tunnels (1FT-2TT group) splitting the tendon into two bundles. The graft was fixed using screw-washer post fixation on the femur and staple(s) on the tibia. Results: SB group showed higher ATT under anterior tibial load than 1FT2TT group mostly at high flexion angles, but this difference was not statistically significant (P [ .05) (Fig. 1).

123

S220

Fig. 1 Anterior tibial translation under a simulated KT-1000 arthrometer test with an anterior tibial load of 89 N (mean±SD). Asterisk indicates statistically significan differences. Under simulated pivot-shift test 1FT-2TT group showed lower coupled ATT at 608 of flexion than SB group, but it was not statistically significant (P [ .05) (Fig. 2).

Fig. 2 Coupled anterior tibial translation under a simulated pivot-shift test of combined 7 Nm Valgus and 4 Nm internal tibial torque . There were no statiscally significant differences. The in situ force in the intact knee in response to anterior tibial and combined rotatory load was significantly higher than the reconstructed knees (P \ .05). However, there were no significant differences between the groups regarding in situ force (P [ .05). Conclusions: There was a trend towards better restoration of intact knee biomechanics in the Single-Femoral-Tunnel Double-Tibial-Tunnels ACLreconstructed group. However, there were no significant differences between these two anatomical ACL reconstructions.

P13-1376 Anatomic bundle strain pattern in differing rotations compared to non-anatomic mismatch graft strain pattern Karahan M.1, Christel P.2, Akgun U.3, Demirel B.4, Tatar Y.5, Aksu N.6, Nuran R.7, Aydin A.T.8 1 Marmara University, Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 2Habib Medical Center Olaya, Sports Medicine, Riyadh, Saudi Arabia, 3Acibadem University Faculty of Medicine, Orthopedics and Traumatolgy, Istanbul, Turkey, 4Akdeniz University, Anatomy, Antalya, Turkey, 5Marmara University Physical Education and Sport Faculty, Health and Sports Department, Istanbul, Turkey, 6Akdeniz University, Antalya, Turkey, 7Orthopedics and Traumatology, Kadikoy, Istanbul, Turkey, 8Akdeniz University, Orthopaedics, Antalya, Turkey

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Objectives: The study aimed to answer the following questions. What it the length variation pattern of each ACL native bundle? What does it become when AM is connected to PL and PL to AM? What is the influence of tibial rotation on bundle length variation? Methods: Six fresh frozen unembalmed elderly cadaveric lower limb specimens were used. Specimens were screened before harvesting to verify that there was normal knee range of motion and no previous evidence of surgery and to rule out ligamentous knee laxity before study inclusion. Specimens were harvested after hip disarticulation, leg and foot were saved intact. After thawing they were maintained at room temperature for 24 hours before study use. Specimen preparation: After doing the kinematics measurements with intact ACL, one anchor was planted in the center of each femoral bundle insertion site. Using the Director tibial aimer fitted with the’’ tip’’ arm, one 4.5mm tunnels was drilled in each tibial bundle. The AM stitches was passed through the AM tunnel and the PL stitches through the PL tunnel for the AM / AL testing conditions. Once the AM / PL tests are completed, suture of the anchor placed in the femoral AM site was passed through the tibial PL (PPL) and suture of the anchor placed in the femoral PL site was passed through the tibial AM (PAM). Two load transducers was rigidly fixed to the leg above the Steinman pin. They were linked to the stitches. A ball and socket joint allows a self alignment of each transducer with the corresponding tibial tunnel in order to avoid any friction of the stitch at the tunnel aperture. Measurements: The knee was cycled and 6 - 8 N load is applied to each stitch at 60 degree of flexion (TBD). Axial compression loading (6.5 kg) was applied by one assistant with the T handle. Then the knee is moved from 0 extension to 100 by 10-15 step increment. For each degree of flexion the load sustained by each transducer was recorded. AM / AL testing conditions: The measurements were performed in neutral rotation, in external rotation applying a 5N.m torque and in internal rotation applying a 5N.m torque. Each measurement was repeated 5 times for each condition of test. PPL / PAM testing conditions: Then the femoral AM stitch was passed through the PL tibial tunnel (PPL) and the PL stitch through the AM tibial tunnel (PAM). The measurements were performed in neutral rotation, in external rotation applying a 5N.m torque and in internal rotation applying a 5N.m torque. Each measurement was repeated 5 times for each condition of test. Results: ANOVA variance analysis followed by Post Hoc Tukey test was used to compare the bundle strain obtained during the flexion arc in neutral, external and internal rotation. In all specimens knee flexion arc in internal rotation had more apparent strain than knee flexion arc in neutral rotation or external rotation. Wilcoxon test was used to compare the individual bundle strains. Conventional single bundle graft (PPL) had a similar strain pattern as the PL bundle. All of the non-anatomic mismatch grafts failed to exhibit an isometric strain pattern through the tested range. Conclusions: Anatomic bundles are exposed to varying strains depending whether they are in neutral, external or internal rotation during flexion arc. Anatomic reconstruction of the individual bundles is necessary to control anteroposterior displacement and proper rotation.

P13-1406 Prospective randomized clinical comparison between double bundle and single bundle anterior cruciate ligament (ACL) reconstruction: four tunnel hamstring vs. patellar tendon Dobnikar U.1, Pilih I.1, Kozic M.1 1 University Clinical Centre Maribor, Traumatology, Maribor, Slovenia Objectives: Double bundle (DB) ACL reconstruction has gained popularity in the last few years. Biomechanical and intra - operative studies have shown better antero - posterior and rotational stability after DB vs. single bundle (SB) reconstruction. However, only few studies are found in the literature that have evaluated and compared clinical results between SB and DB ACL reconstruction. All reported studies used hamstrings in both DB and SB techniques. We found no studies that used patellar tendon autograft for the SB. The purpose of the present study was therefore to evaluate and compare the clinical results of DB hamstring and SB patellar tendon ACL reconstruction techniques.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: A total of 40 patients (29 male and 11 female, average (SD) age 36 (11)) were randomized during routine surgery into either double bundle (DB) or single bundle (SB) group. Patients with combined (medial, lateral or posterior) ligamentous injuries were not included in the study. ACL reconstruction in DB group was done by four tunnel hamstring technique with EndoButton femoral and cancellous screw tibial fixation. In the SB group, patellar tendon autograft was fixed with titanium interference screws. The same postoperative rehabilitation protocols were used for both groups. Subjective evaluation comprised IKDC 2000 subjective score, Tegner activity and Lysholm scores. Objective evaluation followed the IKDC 2000 knee examination protocol. Antero - posterior stability was measured by KT 2000 and by Rollimeter. One leg hop test was performed to assess the functional results. Isokinetic thigh muscle strength measurements on Biodex isokinetic dynamometer were performed. Results: At the follow up, 90% of all subjects were tested. The average (SD) follow up was 27.8 (5) months in the SB group and 28.9 (5) months in the DB group. There were no significant differences in accompanying meniscal or chondral lesions between both groups. Subjective results showed no difference between both groups. The median (range) IKDC 2000 subjective score was 77.9 (17) in the DB group and 77.5 (15) in the SB group. Median (range) Tegner score was 5.2 (1.8) in DB and 5.9 (1.5) in the SB group and Lysholm score was 84.9 (15) in DB and 87.4 (13) in the SB group. The average (SD) antero - posterior translation was 1 (0.7) mm in DB group and 1.3 (1.1) mm in SB group with no statistical significant differences. Similar results were obtained with Rollimeter (1 (1.2) in DB and 1.4 (1.1) in SB group).No differences were found in one leg hop test between the groups. 75% of DB patients were grouped A, and 25% were grouped B according to the IKDC 2000 examination protocol. Similar results were obtained in the SB group (76.5% A and 23% B) with no statistical significant differences between both groups. Biodex measurements showed no differences in extensor strength, but showed that flexor deficit was significantly greater in DB group (p = 0.001). Conclusions: The results of the present study suggest similar outcome of ACL reconstruction using single bundle patellar tendon versus double bundle hamstring autograft after 2 years follow up according to subjective scores, objective AP translation and IKDC 200 examination tests. However, SB group showed better flexor muscle strength at the follow up.We conclude that double bundle technique for ACL reconstruction is not recommended due to higher costs, longer procedures and loss of flexor muscle strength.

P13-1408 The meaning of quality and graft preparing manner with hamstring tendons in ACL anatomic double bundle reconstruction Sadlik B.1, Solecki A.1, Brzoska R.1 1 Klinika sw. Lukasza, Bielsko-Biala, Poland Objectives: The anatomical double-bundle anterior cruciate ligament reconstruction development is presently observed, a its purpose is to restore proper ACL anatomy and function. The endobutton anchoring system used in Eichhorn technique is very demanding, considering that one of the rules is to create two bundles using autogenous ST tendon only without GR tendon harvesting. In many cases it is not possible to obtain satisfactory autogenous tendon in its length and thickness. To maintain acceptable size and quality of the graft in such situations, it is necessary to modify grafts by preparation technique with additional GR tendon harvesting. Some cases are fined with poor tissue quality despite the measurements are accepted. The purposes of the study are: 1. Results evaluation after anatomical ACL reconstruction of the knee with modified graft preparation technique, using hamstring tendons. 2. The results influences of ST tendon structural changes used as the graft for anatomical ACL reconstruction. 3. Assignment of manner procedure in case of poor ST tendon quality despite its acceptable size. Methods: One hundred and two (102) consecutive patients aged from 16 to 56 years (average 29,9 years) who had a four tunnel anatomical ACL

S221 reconstruction from March 2004 to April 2006 were followed-up in 1th and 2nd year. All patients were divided into three groups depending on ST tendon quality and grafts preparation technique: Group R - The patients with high quality and good dimensions for the ST tendon prepared as a two grafts in the standard way (reference group); Group I - The patients with poor quality but acceptable dimensions for the ST tendon prepared as a two grafts in the standard way; Group II - The patients with inappropriate ST tendon size, which had modified grafts preparing with additional GR tendon application. Results of the anterior knee instability treatment were evaluated in 1th and 2nd year after surgery by clinical examination with Lysholm score, Arthrometer measurements with KT-1000, pivot-shift test, passive motion deficit and effusion presence in the knee. Results: The results of subjective evaluation by Lysholm score and passive motion deficit were comparable among the groups. Anterior stability evaluated with KT-1000 and rotational stability evaluated with pivot-shift test were comparable between groups, meanwhile very good results percentage (A) according to IKDC2000 was significantly less in group I than in group R in both follow-up periods. Concerning effusion presence in the joint, groups I and II had significantly higher percentage of mild effusion cases in first period only. Conclusions: 1. The modified technique for grafts prepared with hamstring tendons in cases their inappropriate measurements allows to reach clinical results comparable with anatomical ACL reconstruction using a good quality and satisfactory size ST tendon. 2. The most frequent complications in patients with less ST tendon quality are less anterior and rotational stability of the knee. 3. In case of poor ST tendon quality despite its well size, it is advised grafts augmentation with GR tendon.

P13-1415 ACL reconstruction or reinsertion in the adolescent: 10-year follow up Canata G.L.1 1 Koelliker Hospital, Center of Sports Traumatology, Torino, Italy Objectives: The conservative treatment of ACL lesions in the adolescent is unfavourable in the long term risking precocious joint deterioration. Nevertheless, literature does not agree on the timing and the best type of surgery in this age group. Results of ACL repairs with the open technique are evaluated negatively. We present a retrospective evaluation of a sample of adolescents surgically treated with ACL reconstruction with patellar tendon or with arthroscopic reinsertion of the ACL in the case of proximal lesions and of good quality ligamentous tissue. Methods: From 1990 to 2006, adolescents under 18 were surgically treated for instability as a consequence of ACL rupture. 75 of these were evaluated retrospectively and divided into two groups: 59 cases, mean age 16 years (13-17), 28 males and 31 females were treated with patellar tendon, Group A. 16 cases, mean age 14 years (11-17), 8 males and 8 females, were treated with arthroscopic repair of the ACL (mean time between lesion and surgery 24 days, 2-200, Group B. Mean follow up 120 months (18-204). The evaluation of the results was carried out using the IKDC 2000 score. In all cases a clinical, radiographic and MRI evaluation was performed. Results: Group A: 50 A, 5 B, 4 C. Group B: 14 A, 1 B, 1 C. Subjective IKDC: Group A mean 97 (80-100). Group B mean 99 (81-100). Conclusions: The high percentage of good or excellent results (93%) confirms the possibility of a complete functional recovery, after the lesion of the ACL in the adolescent, with surgery. The use of the patellar tendon for the repair does not result as having compromised the functionality of the extensor apparatus. Excellent results are possible both with patellar tendon and with ACL reinsertion in the case of proximal lesions and of a good quality ligament. Reinsertion with arthroscopic technique results efficacious in a high percentage of subjects. Above all, it is indicated when the young biological age of the patient could mean postponing reconstructive surgery with the risk of secondary lesions and of successive unsatisfactory results.

123

S222 P13-1417 Radiographic evaluation of bony tunnel positions in single and double bundle ACL reconstruction Giron F.1, Losco M.1, Giannini L.1, Aglietti P.1 1 University of Florence, First Orthopaedic Clinic, Firenze, Italy Objectives: Double bundle (DB) ACL reconstruction has been introduced trying to replicate the functional anatomy of native ACL. There are no clear landmarks to exactly locate the attachments of anterior-medial (AM) and posterior-lateral (PL) bundles of ACL at the time of surgery. Up to date only few studies reported tunnel positions in DB reconstruction. The purpose of this study was to investigate the radiographic positions of the bony tunnels in our SB and DB ACL reconstructions. Methods: Two groups of 35 patients with an ACL lesion who underwent arthroscopic assisted ACL reconstruction with SB or DB technique were included in the study. A conventional double incision surgical technique was adopted to perform the SB reconstruction. To insert tibial and femoral guide wires, a 65 Howell tibial guide and a Shino front-entry femoral guide were employed. In the femur we aimed to an anatomic position in between AM and PL insertion sites. In the DB technique the AM tunnel was drilled using the same tibial and femoral guides. On the tibial side the PL guide wire was inserted using a prototype rod guide inserted in the AM tunnel which allow the wire to exit posterior and lateral at a fixed distance (9mm). On the femoral side, with the knee at 90 of flexion we chose the insertion points starting from the AM which was placed near the posterior cartilage below the OTT. The PL insertion point was automatically defined by the prototype rod guide based on the AM tunnel. The distance to the first pin was fixed (9mm). With the knee at 90 of flexion the exits of the two tunnel were almost parallel to the tibial plateau. Radiographic evaluation of the tunnel positions were performed on a standard lateral view with condyle superimposition at 2-year follow-up. The position of the center of the tibial tunnels was measured as percentage of the tibial plateau along the plateau line, whereas the position of the center of the femoral tunnels was measured according to the quadrant technique as described by Bernard and Hertel and was expressed as percentage of the width and the height of the femoral condyle. Results: In SB group the center of the tibial tunnel was located on average at 44% of the plateau width, while in DB group the center of AM and PL tunnels was on average at 38% and 50%, respectively. The femoral tunnel in SB group was locate on average at 25% of the condyle width measured from the back and at 33% of the height of the condyle. In DB group the AM and PL tunnel positions were on average 22% and 35% of the width and 23% and 47% for the height, respectively. In both groups no correlations were found between tunnel positions and subjective and objective results. Conclusions: In DB technique the positions of the AM and PL tunnels are critical. To recreate the native bundle attachments the PL tunnel has to be located in the tibia posterior to AM tunnel and in the femur shallow and inferior. Our data confirmed those positions and moreover we showed that in SB technique the positions of the femoral and tibial tunnels were in between to those of the AM and PL tunnels.

P13-1418 Injury pattern and management of ACL injuries in a young active population - single, double, or ,,selective’’- bundle anterior cruciate ligament reconstruction? A case series of our management in 100 consecutive patients Ta´llay A.1, Morris H.2 1 National Institute for Sports Medicine, Sports Surgery Dept., Budapest, Hungary, 2The Park Clinic, East Melbourne, Australia Objectives: The purpose of our study was to study the distribution of the injury pattern in ACL injury and present our indications for the choice of reconstructive technique (single bundle, double bundle, or selective bundle) based on clinical and operative findings. Methods: Between May and October 2007, a total of 100 consecutive patients for primary ACL reconstructions using hamstring tendon or BTB grafts by the senior author (HM) were recruited for the study. Knee

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 stability was assessed with the KT-2000 arthrometer and evaluation under anesthesia. The intraoperative findings of the extent of ACL injury and the operative techniques were documented. Results: In the series, a complete ACL rupture was found in 56 patients and a partial rupture in 44 patients. A total of 35 single bundle (4-HT and BTB), 44 selective anteromedial or posterolateral bundle and 21 double bundle anatomical reconstructions were performed. Conclusions: Complete ACL rupture was seen in 56% of our early reconstruction cases. Our choice of surgical technique is based on clinical and operative findings.

P13-1419 The ability of three different approaches to restore the anatomic AM bundle femoral insertion site during anatomic ACL reconstruction Kopf S.1, Pombo M.2, Wong A.K.1, Shen W.3, Irrgang J.4, Fu F.3 1 University of Pittsburgh, Department of Orthopaedic Surgery, Pittsburgh, United States, 2University of Pittsburgh Medical Center, Center for Sports Medicine, Pittsburgh, United States, 3University of Pittsburgh, Orthopedic Surgery, Pittsburgh, United States, 4University of Pittsburgh, School of Medicine, Pittsburgh, United States Objectives: There is debate regarding the ability of transtibial drilling methods to create anatomically located femoral tunnels, and no large prospective studies in patients undergoing anterior cruciate ligament reconstruction have addressed this issue. We hypothesized that drilling the femoral tunnel for the anteromedial graft via the accessory medial portal, as opposed to drilling the tunnel transtibially, will lead to more frequent location of the anteromedial femoral tunnel within the anatomic anteromedial bundle insertion. Methods: Primary anatomic double-bundle reconstruction was performed on 113 patients. Intra-operatively, a guide pin was placed through the anteromedial and posterolateral tibial tunnels and accessory medial portal, attempting to reach the center of the native femoral anteromedial bundle insertion. For each approach, the position of the guide pin was classified as: (1) within the center of; (2) off-center within; or (3) outside of the femoral anteromedial insertion (Fig. 1).

Classification describing the guide pin location Results: There were significant differences in the ability of each approach to reach the center of the femoral anteromedial bundle insertion. Through the tibial anteromedial tunnel, the femoral anteromedial insertion site center was reached in 4.4% of cases, while it was off-center within and outside of the femoral anteromedial insertion in 23.0% and 72.6% of cases, respectively. Through the tibial posterolateral tunnel, the femoral anteromedial insertion site center was reached in 60.2% of cases, while it was off-center within and outside of the femoral anteromedial insertion in 23.9% and 15.9% of cases, respectively. When approached from the accessory medial portal, the center of the femoral anteromedial insertion was reached in 100% of the cases. Ultimately, the femoral anteromedial tunnel was drilled via the tibial anteromedial tunnel in 0.9%, via the posterolateral tunnel in 62.8%, and via the accessory medial portal 36.3% of cases.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Conclusions: Drilling the femoral anteromedial tunnel transtibially during anterior cruciate ligament surgery can only reproduce an anatomic femoral anteromedial insertion in select cases.

P13-1425 A review of 6 years experience of double bundle ACL reconstruction technique Solecki A.1, Sadlik B.1, Brzoska R.1 1 Klinika sw. Lukasza, Bielsko-Biala, Poland Objectives: Single bundle anterior cruciate ligament (ACL) reconstruction has been the gold standard for several years. It has provided good to excellent results, with many athletes being able to return to sports. However, anatomically, the ACL consists of two major functional bundles, the anteromedial and the posterolateral bundle. Biomechanically, both bundles contribute significantly to the anterior and the rotational stability of the knee. Therefore, anatomical double-bundle ACL reconstruction techniques may further improve the outcomes in ACL surgery, in theory. Methods: In our Clinic during 6 years the reconstruction methods have been changing from single tibial-single femoral tunnel (2T) through single tibial-double-femoral tunnel (3T) to anatomic double tibial-double femoral tunnel technique (4T). Until now 940 ACL reconstructions were done. Evaluation methods like patient’s subjective complaints and Lysholm score, arthrometer KT-1000, pivot-shift test, pain, range of motion, stress radiographs of the knee (TELOS device) were used. The results of 117 surgeries performed in selected period of time were analyzed among groups of 2T, 3T and 4T. Results: The results of subjective evaluation by Lysholm score, passive motion deficit and the presence of pain were comparable among the groups. Anterior stability evaluated with KT-1000 and stress X-ray and rotational stability evaluated with pivot-shift test were significantly better in the 4T group, meanwhile the worst results were observed in group 3T. Conclusions: 1. The best 6 years results of ACL reconstruction were achieved in double-bundle anatomic technique (4T group). 2. The rotational stability of the knee was best restored in double bundle anatomic ACL reconstruction (4T group). 3. The most unsatisfactory outcomes were found in the group of patients treated with non anatomic double bundle ACL reconstruction technique (3T group).

P13-1427 Anatomical study of three bundle footprints of the anterior cruciate ligament in the porcine knee Lu S.1, Debandi A.V.1, Martins C.A.Q.2, Liu M.1, Kato Y.1, Linde-Rosen M.1, Smolinski P.2, Fu F.2 1 University of Pittsburgh, Pittsburgh, United States, 2University of Pittsburgh, Orthopaedic Surgery, Pittsburgh, United States Objectives: The Porcine knee is widely used as an animal model to study the anterior cruciate ligament (ACL). Several anatomic studies have shown that the ACL of the porcine knee can be separated into three distinct bundles: anteromedial (AM), intermediate (IM) and posterolateral (PL). The purpose of this study is to measure the femoral and the tibial insertion site footprints of AM, IM and PL in porcine knee. Methods: Ten mature porcine knees were dissected and the insertion sites of AM, IM and PL bundles exposed, visually outlined with a marker and digitally photographed. On the anterior surface of the tibial plateau is tilted anteriorly, the camera was set parallel to this plane. Measurements were done using the software for image analysis, Image J, developed at the NIH, using a ruler and a round shape metallic plate as reference. The lengths and the areas of the ACL bundles were analyzed. The longest line from anterior to posterior border of the each bundle was used as length. A line drawn at a 90 angle with the length was used to represent the width. However, for the areas, a direct visualization of bundles’ drawings was performed.

S223 Results: In tibial insertion site the anterior lateral meniscus root is between AM and IM bundles. The tibial AM insertion area from 70mm2 to 100mm2 and average is 86mm, simultaneously in femoral side is from 65mm2 to 97mm2 and average is 76 mm2.The PL insertion area in tibial from 22 mm2 to 38 mm2, and the average is 27 mm2, in femoral from 31 mm2 to 52 mm2 and the average is 42 mm2. The IM insertion area in tibial is from 22 mm2 to 40 mm2 and the average is 29 mm2, meanwhile in femoral side from 17 mm2 to 42 mm2 and the average is 29 mm2.The average of the AM length in tibial and femoral are 20mm and 16mm, respectively. The average of the AM Width in tibial and femoral are 9mm and 8mm, respectively. The average of the PL length in tibial and femoral are 3mm and 11mm, respectively. The average of the PL width in tibial and femoral are 8mm and 5mm, respectively. The average of the IM length in tibial and femoral are 5mm and 4mm, respectively. The average of the IM width in tibial and femoral are 6mm and 7mm, respectively. The average of the distance from the center of AM to the center of PL in tibial and femoral are 14mm and 5.5mm, respectively. Table 1 The percentage of area in each bundle AM/ACL (%)

PL/ACL (%)

IM/ACL (%)

TIBIAL SITE

60.3

19.3

20.4

FEMORAL SITE

51.9

28.1

20

ACL tibial insertion

ACL femoral origin Conclusions: The ACL tibial insertion is the broadest portion of the ligament and this area is mostly occupied by AM bundle, which is also the largest bundle on the femoral site. The other two bundles have similar area on the tibial footprint, but PL bundle is larger than IM bundle on the femoral origin. This study provides information about the anatomy of the porcine ACL femoral origin and tibial insertion that may facilitate future studies using this knee as a model.

123

S224 P13-1454 Tibial spine fractues in pediatric patients - treatment and outcome Eberl R.1, Castellani C.1, Weinberg A.1 1 Medical University of Graz, Department of Pediatric and Adolescent Surgery, Graz, Austria Objectives: Fractures of the tibial spine are rare injuries in the pediatric population and are mostly caused by bicycle or skiing accidents and athletic activities. Either hyperextension with inward rotation of the extended knee joint or hyperflexion with inward rotated knee cause the injury. In adults this mechanism of accident usually results in a rupture of the ACL without bony fragments. The optimal treatment for these injuries has been discussed controversially in the literature. The aim of our study was to evaluate the outcome of a consecutive series of pediatric patients treated either operatively or conservatively. Methods: All patients younger than 17 years with an avulsion fracture of the tibial spine treated between 1995 and 2007 were included. The injury was diagnosed by means of conventional X-rays in two planes and was classified according to Meyers and McKeever. In the operation room an initial aspiration of the joint was performed and the fracture reduced with hyperextension of 5 to 10. In case of irreducibility arthroscopic fracture reduction and fixation with sutures was performed. Independent of treatment the knee joint was immobilized in a cast for 6 weeks. The outcome measurement included the Cincinnati Knees Score protocol and the clinical examination according to the International Knee Documentation Protocol Commitee 2000. X-rays were made in two plains. Results: 52 patients were included. The mean time follow-up was 3.2 years (range 1 to 7.5 yr).We found 29 female and 23 male patients. The majority (49 patients) were treated conservatively and 3 were treated operatively. The mean age was 11.7 years (range 6 to 16 yr).There were 16 Type I, 18 Type II and 18 Type III injuries. All operated patients sustained Type III injuries. None of the patients reported limitation of daily life activities. The clinical examination showed free range of motion in 48 patients. The anterior drawer test was performed clinically and differences to the contralateral side of 3 to 5 mm were found in 10 cases. Radiological abnormalities were found in 33 patients but were without correlation to the clinical findings or subjective complaints. Conclusions: The overall outcome in the presented series of avulsion injuries of the ACL was good although conservative treatment in Type II and III injuries is discussed controversially in the literature. We consider our regimen as justified in case of correct fracture alignment. A temporary positive anterior drawer test will be compensated by further growth. Arthroscopic operative intervention was realized in case of futile closed reduction in Type III injuries only. In these cases soft tissue interponation like the medial or lateral meniscus or the anterior transverse knee ligament should be removed for correct fracture reduction. Further studies are mandatory to confirm our findings.

P13-1471 ACL reconstruction with STG grafts - hybrid tibial fixation biomechanical study Roman M.1, Fleaca R.1, Oleksik V.2, Pascu A.2, Deac C.2, Cofaru N.2, Baier I.1 1 University of Sibiu, Orthopedic Surgery and Trauma, Sibiu, Romania, 2 University of Sibiu, Faculty of Engineering, Sibiu, Romania Objectives: The initial graft fixation after ACL reconstruction is mechanical until the biologic integration takes place. Tibial fixation is considered to be the weak link. For the BTB graft there is a consensus that its fixation is strong and safe because. For the STG graft such a consensus, regarding the ideal method of fixation, does not exist. Study hypothesis: Hybrid fixation (direct interference screw?indirect extra cortical fixation over a boney bridge) is more efficient then simple fixation with interference screw alone. The present study compared the biomechanical properties of three methods for tibial fixation of a quadruple hamstring (STG) graft. Methods: The biomechanical study was performed in the mechanical laboratory of the Engineering Faculty (ULB Sibiu). The testing protocol consists of 9 knees of bovine origin, human semitendinos and gracilis

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 tendons prepared in a four bundle fashion. Tibial tunnel drilling was performed (exit point - the ACL footprint). Three different fixation systems were created: 1. direct fixation with interference screw 2. Hybrid fixation (direct fixation with interference screw ? indirect, extra cortical fixation over a bony bridge) 3. Hybrid fixation (direct fixation with interference screw ? indirect, extra cortical fixation over a cortical screw). All the experiments were performed on the universal tensile, compression and buckling testing machine Instron 5587. Progressive traction of each graft was performed until the system failed. Results: The test showed a different behavior for each of the 3 fixation systems. The maximum load failure (in average) was: 430,498 N for the first system, 941,786 N for the second and 564,338 N for the third. There was no graft failure. In all tests the system failed at the level of tibial fixation. Limitations of the study: The use of bovine specimens modifies the fixation properties (maximum load failure values greater as those in human specimens). Conclusions: The experiment confirms the initial hypothesis and validates the concept of hybrid tibial fixation (direct fixation with interference screw ? indirect, extra cortical fixation over a bony bridge). This paper is supported by a national investigation grant (CNCSIS nr. 698 19. 01. 2009).

P13-1472 Partial lesions of the anterior cruciate ligament Roman M.1, Fleaca R.1, Cofaru N.2, Oleksik V.2, Pascu A.2, Deac C.2, Baier I.1, Roman C.3 1 University of Sibiu, Orthopedic Surgery and Trauma, Sibiu, Romania, 2 University of Sibiu, Faculty of Engineering, Sibiu, Romania, 3Spitalul Clinic Judetean Sibiu, Sibiu, Romania Objectives: The ACL is made out of two major functional fiber bundles (AM and PM). Preoperative diagnosis of partial ACL lesions is difficult to do both by clinical exam and MRI. Intraoperative evaluation of the remainig bundle is challenging. Aim: Evaluation of the role of single bundle procedure in partial ACL lesions. Methods: 285 patients with ACL lesions were performed (2004-2009). Only 12 cases had partial lesions of the ACL. There were only 2 cases with PL lesion and 10 cases with AM bundle lesion. The patients with PL lesion had an functionally insufficient AM bundle. Reconstruction of only the ruptured AM bundle was made in 4 cases, the rest required replacement of both. STG graft in three strands was used in all cases. The fixation was performed with bioabsorbable interference screws on both sides. On the tibial side we used the hybrid fixation (direct, anatomic fixation with interference screw ? indirect, transcortical augmentation. Results: Diagnosis and management of these particular situations is very challenging. Appropriate clinical evaluation can provide some valuable information. MRI should always be performed in such situations although it has its limitations. Arthroscopy may offer the correct diagnosis but evaluation of the functional capacity of the nonruptured bundle is challenging even for experienced surgeons. Postoperative clinical evaluation (IKDC, Lysholm, functional tests) showed good results regarding both AP and rotational stability. Conclusions: Evaluation of the intact bundle in partial ACL lesions is challenging. Reconstruction of a partial ACL tear should be addressed to the torn bundle (single bundle procedure) instead of sacrificing both bundles.

P13-1487 Biomechanical evaluation of a one-stage ACL revision technique using a structural bone void filler Dragoo J.1, Vaughn Z.1, Lindsey D.1 1 Stanford University, Orthopaedic Surgery, Redwood City, United States Objectives: CL graft failure commonly occurs due to error in tunnel placement or tunnel widening. ACL revision often necessitates a staged approach with bone grafting to avoid graft pullout. Structural bone void

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 fillers may provide the required structural support to allow for a single stage revision with equivalent biomechanical properties. Methods: 11 matched pairs (n=22) of fresh, frozen cadaveric knees were tested. Controls were treated with autologous bone-tendon-bone primary ACL reconstruction fixed with bioabsorbable interference screws with a 1-mm back wall. The contralateral knee of each pair had tunnel placement in an anterior position to simulate graft failure conditions (6-mm back wall). This tunnel was filled with calcium phosphate cement bone void filler arthroscopically. After solidification in 10 minutes, the tunnel was re-drilled through the bone void filler with a 1mm back wall fixed with bioabsorbable interference screws. Specimens were then tested in an MTS machine for load to failure according to a standard protocol. Results: Failure loads for the control group averaged 312N (±127N), while the calcium phosphate cement bone void filler revision group averaged 302N (±95N), We were unable to demonstrate a significant difference between these conditions, average difference 10.76N (±136.527), p = 0.799. Failure occurred at the tendon bone interface in both groups, but without screw pullout. Conclusions: We were unable to show a significant difference between the average pull out strength of the control group compared to the Callos Impact revision group. This technique may be useful as a one-staged revision ACL reconstruction procedure.

P13-1488 The effect of relaxin on ACL integrity Dragoo J.1, Lindsey D.1 1 Stanford University, Orthopaedic Surgery, Redwood City, United States Objectives: The hormone relaxin, found in pregnant and non-pregnant females, has been shown to have a collagenolytic effect on ligamentous tissue. Relaxin receptors have recently been identified on human female anterior cruciate ligaments (ACL)1. This study evaluated whether the administration of recombinant relaxin and estrogen or relaxin alone will lead to a significant increase in ACL laxity in the guinea pig model. Methods: Guinea pigs were administered 20 ug/hr of recombinant relaxin ±5 ug/hr of estradiol using separately implanted osmotic pumps. ACL laxity was tested by implanting radio-opaque markers in the femur and tibia of each leg. After applying a standard force (22N) anteriorly translating the tibia, the distance between markers was measured radiographically at day 0 and day 21 compared to controls. The animals were then sacrificed and the ACL’s were analyzed for load-to-failure using a material testing machine. Results: Animals treated with relaxin and estrogen (n=4) showed a significant (p=0.02) increase in ACL laxity under an applied force compared to controls (n=4). Animals only treated with relaxin (n=4) also showed a significant (p=0.04) increase in ACL laxity under an applied force compared to controls (n=4). Load-to-failure testing showed hormone treated ACL’s (Relaxin ? Estrogen l=32.7 N) (Relaxin only l= 40.4 N) were significantly weaker than controls (l=64.1 N) (p=0.000). Conclusions: This data suggests that relaxin and estrogen significantly alter the mechanical properties of the ACL in an animal model. The effects of relaxin may contribute to the etiology of female non-contact ACL injuries.

Knee: PCL P14-81 Posterior cruciate ligament reconstruction Alonso R.H.1, Taddeo H.2 1 ARST, Wilde, Argentina, 2ARST, Banfield, Argentina Objectives: The aim is to evaluate our results in 50 arthroscopic Posterior Cruciate Ligament reconstructions in workers, with different techniques, and to describe our personal technique.

S225 Methods: From 1996 to 2007, we performed 50 PCL reconstructions, changing the techniques because of non-satisfactory initial stability in terms of residual posterior tibial displacement. We have made a retrospective study, with data obtained at the moment that they returned to work Level of significance was set at P\0.05. The statistical analysis was performed through analysis of variance (ANOVA) and the Student t test. Current Surgical Technique: Knee placed in 908. Hamstring tendons are removed with stripper. While the surgeon begins the arthroscopy, the assistant surgeon prepares the graft, passing the tendons by the stirrup of a SAR-Bone. The scope is placed through the anterior lateral approach. We can clean the femoral wall and the posterior tibial slope by only using the anterior medial approach. Then we use a special guide to PCL tibial tunnel with a safety device that avoids crossing the guidewire further than the top of the guide. The tunnel is overdrilled outside-inside according to the diameter of the graft. The external hole of the tunnel is drilled 10 mm. in diameter to place the SAR-Bone later. With the same guide we make the femoral tunnel outside-inside. Then with a wire turned over itself we hang the threads of the graft, and pass it from tibia to femur, hitting the SARBone on the tibia. Now, the graft is firmly on the tibia, then while the surgeon with one hand on the ankle and the other forearm behind the proximal leg does strong anterior drawer and more than twenty flexionextension, like an old water pump, the assistant surgeon pulls the graft and fixes it with a screw over a pin, with knee in 908. Results: In the first patients we used Patellar Tendon Bone (PTB) autograft, single bundle fixed by two interference screws, the X-Ray Posterior Drawer Test at the end of treatment was 7.83 mm ± 0.94 (range, 7 to 10 mm). In the following group of patients we used Hamstring (HT) autograft fixed by screws, the X-Ray Posterior Drawer Test at the end of treatment was 7.85 mm ± 1.63 (range 3 to 9 mm). The 2 men with Quadriceps Tendon autograft had similar results to the previous series. In the last 23 patients, with a new technique the X-Ray Posterior Drawer Test was 3.95 mm ± 1.18 (range 2 to 6 mm). When we compared the X-Ray Posterior Drawer expressed in mm. PTB vs. HT SAR-Bone techniques, a statistically significant difference was found (P \ 0.0001), same as with HT vs. HT SAR-Bone (P \ 0.0001). Conclusions: We used then a device designed to fix the HT in femur to replace the ACL, that solved our problems with instant and strong fixation of the quadrupled graft against the anterior cortical of the tibia. We obtained better results than in the past, with maintenance of our patients in Grade I. We believe that our results are better due to: 1) firm fixation on tibia that avoids slippage and allows doing stronger anterior drawer, 2) we do not force the tibia backwards when inserting a screw or a staple as it is usually done when the fixation finishes in tibia; 3) Usually the assistant can pull directly from the graft. Probably the better results with Inlay Reconstruction are because of the stronger tibial fixation and possible slippage in only one tunnel, but we obtain similar results with our all arthroscopic technique, more simply, and avoiding posterior approach.

P14-336 Clinical results of the single-bundle, transtibial posterior cruciate ligament reconstruction: a systematic review Kim Y.-M.1 1 Chungnam National University Hospital, Orthopedic Surgery, Daejeon, Republic of Korea Objectives: The arthroscopic single-bundle, transtibial reconstruction technique has traditionally been the most common method for posterior cruciate ligament (PCL) reconstruction. Unfortunately, clinical studies evaluating this procedure for isolated PCL tears are rare, and there are no pooled clinical analyses evaluating the effectiveness of this procedure for isolated PCL tear. Methods: Ten articles containing the clinical results of this procedure for isolated PCL tears were systematically reviewed with an emphasis on

123

S226 residual posterior laxity, subjective and objective functional outcome, activity level, patient satisfaction, incidence of osteoarthritis, and postoperative complications. Results: Instrumented posterior knee laxity was 1.96-5.9 mm (results of 9 articles), which was comparable with preoperative 8.38-12.3 mm posterior laxity (results of 4 articles). The mean values of the Lysholm knee scores were 81-100 points (results of 9 articles). Overall qualification of the IKDC standard form were categorized 75% as normal or nearly normal (results of 6 articles), and the mean score of Tegner activity scale was 4.76.3 points (results of 8 articles). Data included in this review showed that degenerative changes developed following this procedure. There were few complications reported. Conclusions: The single-bundle, transtibial PCL reconstruction technique for isolated PCL tears can improve posterior knee laxity by one grade, though this procedure cannot reliably restore normal knee stability. Return to recreational and athletic activity was predictable with more than 80 points of subjective knee score and 75% normal or nearly normal objective outcome, though degenerative osteoarthritis did not seem to be prevented by this procedure.

P14-399 Serial assessment of knee joint moments in posterior cruciate ligament and posterolateral corner reconstructed patients during a turn running task Lee Y.S.1, Lim B.-O.2, Kim J.G.3, Lee K.-K.4, Ahn J.H.5, Jung Y.B.6, Min B.-H.7 1 Ajou University Medical Center, Department of Orthopaedic Surgery, Suwon, Republic of Korea, 2Sports Science Institute, Seoul, Republic of Korea, 3Inje University, Seoul, Republic of Korea 4Kookmin University, Seoul, Republic of Korea, 5Samsung Medical Center, Department of Orthopedic Surgery, Seoul, Republic of Korea, 6Chung-Ang University, Department of Orthopedic Surgery, Seoul, Republic of Korea, 7Ajou University, Suwon, Republic of Korea Objectives: The purposes of this study were to understand 1) how these negative aspects of PCL and PLCS reconstructions influence the knee joint moments during turn running tasks, and 2) how these biomechanical properties change with serial assessment. Our hypothesis was that PCL- and PLCS-reconstructed patients would show a decreased extension moment and an increased external rotational moment in the operative knee. Methods: For the experimental group, PCL- and PLCS-reconstruction were performed simultaneously, and all reconstructed patients had stable knees and showed no complications. A motion analysis system was used to measure and calculate kinematic and kinetic data for seven patients after PCL and PLCS reconstruction (experimental group) and seven normal subjects (control group) during a turn running task. The study was conducted on two groups at both three months (return to daily activity) and six months (return to light sports) post-operation. At six months after surgery, the dial test was also performed in order to observe the static rotational stability. Results: Compared with the control group, the experimental group showed a decreased extension moment (-1.15 ± 0.46 Nm/kg versus -3.51 ± 0.69 Nm/kg, p=.000), a decreased valgus moment (-1.36 ± 0.72 Nm/kg versus 2.15 ± 0.54 Nm/kg, p=.041), and a decreased external rotational moment (-0.15 ± 0.11 Nm/kg versus -0.37 ± 0.10 Nm/kg, p=.002) at three months post-operative. However, these results approximated to the normal control, and the experimental group showed an improved extension moment (-2.95 ± 0.67 Nm/kg, p=.188), valgus moment (-1.73 ± 0.58 Nm/kg, p=.359) and external rotational moment (-0.30 ± 0.09 Nm/kg, p=.325) at six months post-operative. A static rotational stability revealed a similar or over-constrained state compared with the contra-lateral knee, and no patient showed rotational instability. Conclusions: PCL- and PLCS-reconstructed patients showed decreased extension, valgus, and external rotational moments during turn running tasks. These results approximated to the normal control with time, but they showed a decreased external rotational moment although static rotational stability was fine.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P14-725 Outcome of surgical treatment of arthrofibrosis after ligaments reconstruction Said S.1, Christiansen S.E.1, Faunoe P.1, Lund B.2, Lind M.3 1 Aarhus University Hospital, Sportstrauma Division, Orthopedic Dept, Aarhus, Denmark, 2Division of Sports Trauma, Department of Orthopedic Surgery, Aarhus, Denmark, 3Aarhus University Hospital, Sports Trauma Clinic, Aarhus, Denmark Objectives: In a retrospective case-series we evaluated the clinical outcome after surgical release and intensive physiotherapy in patients having an arthrofibrotic complication to ligament reconstruction. Methods: 31 Patients operated from January 2003 to December 2007 for arthrofibrosis as post-operative complication after knee ligament reconstruction was included in the study. The included patients underwent brissement force or surgical arthrolysis combined with intensive physiotherapy as a treatment for reduced range of motion. Objective examination, Pain, KOOS and Tegner scores were used to evaluate the clinical outcome at a project follow-up in 2009. Results: 14 patients were treated with brissement only and the remaining 13 patients were treated with arthroscopic arthrolysis. The patients (8 of them were evaluated by phone) had average follow-up of 47 months (1873). There were 12 males and 15 females. Average age was 36. The average time between the primary surgery and the surgical release was 9 months. Average Range of Motion (R.O.M) was increased from 5 to 1 (range -2-9) in extension and from 94 to 126 (range 90-145) in flexion. At follow-up average Pain VAS score (0-10) at rest was 1,5 and in daily activity was 3. Average KOOS subscores were for symptoms 37, pain 37, ADL 29, sport 63, QOL 61. Average Tegner score was 5. Conclusions: Arthrofibrosis is an uncommon but severe post-operative complication after ligament reconstruction. Surgical arthrolysis combined with intensive physiotherapy improved range of motion to nearly normal values. But subjective outcome scores revealed relatively poor outcome levels compared to uncomplicated ligament reconstructed knees.

P14-865 Anatomical and computed tomographic analysis of the femoral attachment of the posterior cruciate ligament Nishikino S.1, Hashimoto Y.1, Hara Y.2, Takigami J.3, Yamazaki S.1, Nakamura H.1 1 Osaka City University Graduate School of Medicine, Orthopaedic Surgery, Osaka, Japan, 2Higashisumiyoshi Morimoto Hospital, Osaka, Japan, 3Yodogawa Christian Hospital, Orthopaedic Surgery, Osaka, Japan Objectives: Accurate and anatomic tunnel placements are important aspect to success the reconstruction of the posterior cruciate ligament (PCL). Although many studies have provided information about the femoral footprint of the PCL, there are few studies detailed anatomical evaluations of the bony landmarks for PCL reconstruction1). The objective of this study was to define the bone morphology of the femoral attachment areas of PCL in cadaveric knees. To define the bony landmarks, we examined the Computed Tomography (CT) scan and CT number was measured that we previously reported about the femoral attachment of anterior cruciate ligament1).and prepare the slices 3-mm thick vertically to PCL insertion and analyzed by radiograph. Methods: 18 human cadaveric knees from the anatomical course of medical students at Osaka city university medical school of medicine, Osaka, Japan were used. The cadaver’s knee had slightly cartilage damages but PCL intact. The age range was 70-86 years. Distal femur was extract and all tissue except the PCL was removed. CT examination of the all distal femur was performed with a helical CT machine and reconstructed using three-dimensional (3D) reconstruction software (AzeLtd.,Japan). Cross-sections 3-mm thick was made using a bone trimmer vertical to the PCL attachment. Five slices were prepared and named group A, B, C, D and E from intercondyle to medial condyle. The each cut specimens were radiographed with soft X-ray to identify the insertion of the ligament and bone. We observed the bone morphology at the insertion and classified the shape of the bony ridge. The height of the ridge and the distance from articular cartilage edge to center of osseous

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 prominence and attachment of PCL were measured. At CT analysis, six zones around the PCL insertion were defined and 10 point of CT value in each area were measured and analyzed. Results: In radiograph, three types of the shape at the medial wall of PCL attachment were identified, that is one ridge type, two ridges type and flat type. One ridge type were found 66.7%, 38.9 %, 22.2%, 33.3% and 50% in group A, B, C, D and E. Two ridges were found 22.2%, 61.1%, 77.8%, 61.1%, and 44.4% and flat were 11.1%, 0%, 0%, 5.6% and 5.6% in group A, B, C, D and E. The average height of the ridge were 0.60, 0.66, 0.76, 0.75 and 0.45mm in group A, B, C, D and E. The average distance from articular cartilage edge to the center of one ridge, distal prominence of the two ridges, proximal prominence of the two ridges and proximal edge of PCL insertion were 9.2, 8.5, 12.6 and 13.8mm, respectively. The average CT number was 604.6, 978.2, 643.5, 520.0, 942.1 and 615.6 at zone 1, 2, 3, 4, 5 and 6, respectively. The CT numbers of zone2 and 5 that footprint of PCL were higher than that of other zones. Conclusions: It was important for surgeons to know about relation between attachment of PCL and bony landmark. In this study, we demonstrated that femoral insertion of the PCL has two ridges frequently, which is include the distal prominence of the two ridges located slight proximally footprint of the PCL and the proximal prominence of the two ridge located proximal edge of footprint of the PCL. The CT number of the bony ridge was higher than that of the other zone in this study. This is indicated that PCL attachment was easily visualized using 3D reconstruction software and the higher bone quality in PCL attachment may be help the surgeons where make bone tunnel.

P14-979 Isolated arthroscopic PCL reconstruction with quadriceps tendon: a series of 17 patients with mean 30 months follow-up Sonnery-Cottet B.1, Zayni R.1, Chambat P.1 1 Centre Orthope´dique Santy, Lyon, France Objectives: The objective of this retrospective study is to report the clinical results of a series of 17 patients treated with isolated arthroscopic reconstruction with a minimum one year follow-up. Methods: The population included 15 patients for which conservative treatment had failed and 2 elite athletes, with major differential laxity (18 mm) and for which early recovery was sought-after. Mean age at surgery was 29 (17- 46). The mean time interval from injury to surgery was 23 months (2 weeks - 95 months). The operative technique used was an isolated arthroscopically assisted reconstruction of the anterolateral bundle of the PCL, using ipsilateral quadriceps tendon autograft with double tibial and femoral fixation. Surgeries were performed jointly by both senior surgeons. Pre- and postoperative workup include full radiological workup, Telos differential laxity measurement with radiological reference points in the middle of the knee to overcome tibial plateau rotation, IKDC objective and subjective scores, along with Lysholm and Tegner scores. The 17 patients were re-examined by the same observer after a mean 30month period (min 12, max 60). Statistical analysis including correlation studies was performed. Results: Mean operative time was 72 minutes (min 52, max 110). An associated lesion of the lateral meniscus was observed in two patients, requiring meniscectomy. An early-stage chondropathy was observed on the medial tibial plateau in four patients. The level of activity of the 17 patients at last follow-up was comparable to their level of activity prior to injury (p = 1 using Fischer’s exact test). 11 patients out of 17 (64.7%) stated that their knee was almost normal or normal at last follow-up. 6 patients out of 17 (35.2%) reported anterior pain on effort and 4 patients (23.5%) complained about moderate effusion during intense efforts. Mean subjective IKDC score (37.7 prior to surgery, 74.5 on last follow-up), Lysholm (40.47 prior surgery, 82.35 at last follow-up) and Tegner (3 prior surgery, 7 at last follow-up) were significantly improved by surgery (p \ 0.05 using the Wilcoxon test). Objective IKDC scores for patients in the series revealed 11,8% of class A or class B pre-surgery, and 88.2% after surgery (no significant p = 0,12 Fisher’s exact test). Mean differential posterior laxity was 11,9 mm (min 8, max 18) prior to surgery and 3,8 mm (min 1, max 7) at the last control assessment

S227 (p = 0.01 using the Wilcoxon test). On last follow-up assessment, 2 patients out of 17 (11.7%) presented a slight flexion deficiency between 6 and 15. 4 patients (23.6%) presented discreet pain at harvesting site level. 3 patients presented moderate radiological pinching of the medial line space. The time interval from injury to surgery was not correlated with residual differential laxity, objective and subjective IKDC scores (p = 0.829, 0.112, 0.251 respectively, Kruskal-Wallis test). Subjective and objective scores on last follow-up assessment were not correlated (p = 0.335 using the Kruskal-Wallis test). However, a correlation was established between preoperative activity level and subjective IKDC score on last follow-up assessment (p \ 0.01, Wilcoxon test). No specific complications related to this reconstruction technique were observed. Conclusions: The results obtained on posterior laxity control and subjective results are encouraging. Longer-term studies with bigger sample sizes are required in order to evaluate the clinical benefit of such reconstruction technique.

P14-1011 Results of the triple and double bundle posterior cruciate ligament reconstruction in the multiple-ligament injured knee: a preliminary report Soejima T.1, Inoue T.1, Kanazawa T.2, Noguchi K.2, Tabuchi K.3, Noyama M.3, Nagata K.2 1 Kurume University School of Medicine, Orthopaedics Surgery, Kurume Fukuoka, Japan, 2Kurume University School of Medicine, Orthopaedic Surgery, Kurume, Japan, 3Kurume University, Orthopaedic Surgery, Kurume, Japan Objectives: In the treatment of the multiple-ligament injured knee, the treatment results of each ligament influence it complicatedly, but it is thought that the result of the posterior cruciate ligament (PCL) reconstruction is the above all most important. Therefore we report it in this study because we investigated results of the PCL reconstruction in the multiple-ligament injured knee. Methods: Fifteen multiple-ligament injured knees of 15 patients which underwent the arthroscopic PCL reconstruction in our hospital were evaluated. There were 13 males and 2 females. An average age at the operation was 31 years old (ranged from 17 to 58 years old). Details of concomitant ligament injuries and its treatment were as follows; only medial collateral ligament (MCL) injury (conservative: 3, reconstruction: 1) : 4, only postero-lateral structure (PLS) injury (repair: 2, reconstruction: 1): 3, only anterior cruciate ligament (ACL) injury (reconstruction: 2): 2, both ACL and MCL injury (ACL conservative ? MCL conservative: 1, ACL reconstruction ? MCL reconstruction: 1, ACL reconstruction ? MCL repair: 1): 3, and both ACL and PLS injury (ACL reconstruction ? PLS repair: 2): 2. In PCL reconstruction, tripled or quadrupled semitendinosus tendon was used for the anterolateral (AL) graft, and doubled or tripled graciles tendon was used for the posteromedial (PM) graft. Nine of 15 patients underwent the conventional double bundle PCL reconstruction which created 2 femoral sockets and 2 tibial tunnels. Remain 6 patients underwent the our original triple bundle PCL reconstruction which created 3 femoral sockets and 2 tibial tunnels. Namely, AL graft was divided to two distinct bundles and grafted into 2 femoral sockets and a tibial tunnel as the bi-socket reconstruction, and PM graft was grafted same as the double bundle PCL reconstruction. For these 15 case, we evaluated objective knee stability using the gravity sag X-ray view and the stress Xrays at an average of 31 month (ranged from 1 to 7 years) after PCL reconstruction. Results: In all cases, the side to side difference of the tibial posterior displacement was 4.0±1.9 mm (ranged from 0.5 to 7 mm) at the time of the final evaluation, whereas it was 13.2±3.1 mm (ranged from 10 to 20 mm) before the operation. 4 patients were categorized in poor results because of more than 5 mm posterior instability, and all of them were in the double bundle PCL reconstruction group. However, the triple bundle PCL reconstruction revealed comparatively good results with 2.9±1.6 (ranged from 0.5 to 5 mm) than that of the double bundle PCL reconstruction.

123

S228 Conclusions: The patients of poor results of the PCL reconstruction simultaneously showed valgus and/or varus instability. Namely, our results suggested the close relations of treatment results of each ligament. Although it was an ideal to reconstruct all ligaments adequately, it was thought that it was most necessary to make efforts in the improvement of results of PCL reconstruction. Thus, we newly developed the triple bundle PCL reconstruction. There was still little number of the cases, but the triple bundle PCL reconstruction had comparatively good results in the multiple-ligaments injured knee.

P14-1155 Reconstruction of acute posterior cruciate ligament tears using a synthetic ligament: 45 cases at five years mean follow-up Boisrenoult P.1, Pujol N.1, Beaufils P.1 1 Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France Objectives: Treatment of recent posterior cruciate ligament tears remains a discussed problem, especially in multiple ligament injured knee. Since 1996, we have choose to perform acute PCL reconstruction with adjunction of a synthetic ligament (LARS), which acts like a tutor for healing of the torn ligament. The aim of this retrospective study was to present our clinical results and complications. Methods: 45 patients (41 men and 4 women) operated after multiple ligament knee injuries have been studied. Mean operative day was 18.5 (7 to 37). The evaluation concerned the safety of use looking for post-operative complications, and the functional result using the IKDC evaluation form. Results: Mean follow-up was 5 years (Range, 6 month to 13 years). Complications were 7 knee stiffness, 2 post-operative infections (after concomitant osteosynthesis), and 3 complex regional pain syndrome. No specific complications linked to the synthetic ligament were found. At final follow-up, the overall IKDC score was A: 6, B: 18, C: 17, D: 4. the posterior drawer test at 70 was absent in 11 cases, quoted at 1? in 24 cases and 2? in 10 cases. At final follow-up, no patient has been operated for degenerative changes. Conclusions: Our results are in compliance with the literature. Results depended on associated lesions particularly posterolateral corner involvement, with however an improvement for the patients operated since 2002 because we have changed our posterolateral corner reconstruction technique. The gain in posterior laxity was substantial and remains stable. The role of the healing tutor of PCL devolved to the synthetic ligament appeared to be achieved, without specific complications. This technique is in our hands, reliable, fast and safe. This technique is easy to done and spares tendon stock in multiple ligament injured knee.

P14-1187 Clinical outcome after double bundle PCL reconstruction. 2- year results of a prospective study Lenschow S.1, Zantop T.1, Herbort M.1, Raschke M.1, Petersen W.2 1 Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 2Martin-LutherKrankenhaus, Unfallchirurgie, Berlin, Germany Objectives: Posterior instability is an often unrecognized injury of the knee. Our hypothesis was, that PCL reconstruction using autologous hamstring tendons can reduce posterior instability significantly. Methods: Patients with posterior instability were included in a prospective study design. Patients were questioned for subjective feeling of giving way and posterior rotational drawer was tested in clinical examination. For the quantification of posterior instability lateral stress- x-rays were taken using the Telos-system. PCL reconstruction was performed using autologous hamstring tendons and a femoral and tibial hybrid fixation. Postoperative management included 6 weeks in a stiff PTS brace in full extension and another 6 weeks of rehabilitation wearing a PCL brace without range of motion limitation. Clinical evaluation took place 24 months after surgery according to the criteria of the IKDC and by stress- x-rays (Telos system). 50 Patients were included in the study. The indication for surgery was a posterior instability of 12 mm or more.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: In 80% of the patients a III posterior instability in combination with an injury of other stabilizing structures of the knee was found. While 29 patients had an additional posterolateral instability, in 11 patients a posteromedial instability was found. In only 11 cases an isolated PCL tear was the reason for instability. The posterior instability under 15 N of posterior tibial stress was reduced from preoperatively 14.3 ± 5.2 mm to 6.9 ± 4.1mm after 24 months. Conclusions: PCL reconstruction using autologous hamstring tendons and a femoral and tibial hybrid fixation can produce satisfying results after 24 months. Associated injuries have to be taken into account. Posteromedial instability seems to play an important role as an associated injury.

P14-1189 Preliminary results of arthroscopic PCL reconstruction in inlay technique using the retro-drill system Osti M.1, Gohm A.1, Benedetto K.-P.1 1 Academic Hospital Feldkirch, Trauma Surgery and Sports Traumatology, Feldkirch, Austria Objectives: The tibial tunnel technique is reported to be the arthroscopic standard procedure for PCL reconstruction. Suboptimal tunnel positioning results in killer-turn effect and evident biomechanical disadvantage however. A more favourable graft position is obtained using the tibial inlay technique, but the open posterior intermuscular approach yields a higher soft tissue morbidity. The objective of this study was to evaluate a new reconstruction technique, which combines a tibial inlay technique with an arthroscopic approach. Methods: The arthroscopic inlay technique is a new development based on a specially designed drill system. Under arthroscopic and fluoroscopic control a posterior tibial socket is created using a drill guide inserted over the anteromedial portal. The femoral tunnel is drilled either retrograd or antegrad. Quadriceps tendon with a 11mm patellar bone block is used as autograft. The graft is inserted, interlocked in the tibial socket and fixed to the anterior tibial cortex by a suture button. The free tail of the graft is guided into the femoral tunnel and fixed with an interference screw. Between 1.8.2007 and 31.12.2008 a total of 17 patients underwent surgery using the presented technique. 12 males and 5 females with an average age of 31,4 (19 - 41) years were followed using clinical, radiographic and computertomographic criteria. The period between trauma and operative treatment was 2 - 12 months. Results: Total IKDC score result: A 5 / 17, B 12 / 17, C 0 / 17, D 0 / 17. One postoperative haematoma was reported as complication but did not qualify for revision. No neurovascular deterioration was documented. Breakage of the drill guide as instrument-related complication occurred in one case. Postoperative CT scans for evaluation of graft position and tunnel placement resulted in malposition of the tibial socket in one case. This patient presented with a loosening and dislocation of the tibial bone block, which required open refixation. Conclusions: The arthroscopic inlay technique using quadriceps tendon autograft seems to be a reproducible option for PCL reconstruction with a low complication rate. Preliminary results are promising.

P14-1304 In vivo tunnel positioning analysis after PCL reconstruction Servien E.1, Gancel E.2, Lustig S.3, Demey G.4, Neyret P.5 1 Centre Albert Trillat, Hospices Civils de Lyon, Lyon-Caluire, France, 2 Centre Albert Trillat, Department of Orthopaedic Surgery, Lyon-Caluire, France, 3Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 4Centre Albert Trillat, Lyon-Caluire, France, 5Hopital CroixRousse - Centre Livet, Chirurgie Orthopedique, Lyon, France Objectives: The aim of this study was to analyze the positioning of the femoral and tibial tunnels after single bundle PCL reconstruction under arthroscopy. Methods: Between 2006 to 2008, we performed twenty-four single bundle PCL reconstruction for isolated PCL lesion or multiple ligaments injury. We were able to analyze 19 patients with entire data including x-rays and

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 a CT-scan. On stress x-rays (Bartlett view), we measured the residual laxity. A Ct-scan was performed by the same radiologist using one protocol in order to analyze the tunnels positioning in the frontal, sagittal and axial planes. Results: The tibial tunnel was located at 32 mm (SD 4.6; range 22.2- 39) from the medial tibial plateau in the frontal plane. The center of the tibial tunnel was emerging at 15 mm above the joint line. One the sagittal plane, the angle of the tibial tunnel averaged 54 (SD 4.5; range 46- 64). Using a ratio to analyse the anteroposterior positioning on the femur, the femur tunnel was located at mid distance on the sagittal plane (ratio : 0.49). The height of the femoral tunnel averaged 12 mm (SD 3.5; range 518) regarding to the distal condyles. In the axial plane, the center of the femoral tunnel was located at 1o0 clock for a right knee and at 11ho0 clock for a left knee. Conclusions: Our CT-scan study attempts to determine some anatomical criteria for PCL reconstruction. Recently some anatomical and radiological studies improved our knowledge on the tibial and femoral PCL insertion. Furthermore, the femoral tunnel positioning is still controversial in the literature. Our work emphasize that the one bundle reconstruction is close to the anatomy of the anterolateral bundle on the femur.

P14-1475 Posterior cruciate ligament reconstruction with transtibial tunnel technique arthroscopically assisted Wajsfisz A.1, El Masri F.1, Djian P.1 1 Cabinet Goethe, Paris, France Objectives: Reconstruction of the posterior cruciate ligament (PCL) using transtibial tunnel technique arthroscopically assisted has been reported as an alternative to the tibial inlay fixation. The transtibial tunnel technique is criticized for its difficulty, even if this arthroscopy procedure have less morbidity at short term. The purpose of this study was to evaluate the results of PCL reconstruction using bone-patellar tendon-bone or quadricipital tendon graft and transtibial tunnel technique. Methods: Prospectively studied 35 patients (36 PCL reconstruction) having isolated or combined PCL reconstruction using the transtibial tunnel technique. The study period was from 2001 to 2008. The follow-up (FU) was 100% and averaged 20.4 months (2 to 84). Twenty-nine patients were male; average age was 30 years. There were 33 primary and 3 revision reconstructions. Surgery was performed in the chronic setting in 80% (29/ 36). Combined reconstructions involving the posterolateral corner, anterior cruciate ligament, or medial collateral ligament were done in 78% (28/ 36). All PCL reconstructions were performed with bone-patellar tendonbone or quadricipital tendon graft, with the bone part of the graft always destined at tibial fixation. The rehabilitation protocol depends on the combined procedures. All patients were evaluated with preoperative and postoperative examination, KT-1000 arthrometer and x-rays. Final followup International Knee Documentation Committee (IKDC), and final follow-up stress radiography (Hamstring contraction and Telos) were performed in all patients. Results: Postoperative posterior drawer (PD) examination demonstrated the following: grade 0 (normal) in 22 patients, grade I in 13 patients, grade II in 1 patients, and none in grade III, as compared to preoperative PD grade II or greater in 58% in this report. Final FU Telos stress radiography with 25 kg posterior load applied at 80 degrees to 90 degrees of flexion demonstrated average side-to-side difference of 4.78 mm; for final FU Hamstring contraction this difference was 5.44 mm. Average flexion gain was 5 degrees (0-17 degrees). No loss of extension. Preoperative IKDC objective evaluation rated 35 knees as severely abnormal based on instability. Final FU objective IKDC evaluation distribution was as follows: A, 5 knees; B, 28 knees; C, 2 knees; and D, 1 knee, as compared to all 36 D preoperatively. Average final FU IKDC subjective score was 85 (50-98). 89% patients evaluated their knee as improved or greatly improved, 53% patients returns at High level activity. Compared to the pre-op status, there was a statistically significant improvement for activity level, symptoms, subjective evaluation, final score and PD. Conclusions: Reconstruction of the PCL-deficient knee with severe posterior laxity is a challenging surgical problem, as combined instability

S229 patterns frequently coexist (78% in this study). When appropriate combined reconstructions or primary repair is used, PCL reconstruction with bone-patellar tendon-bone or quadricipital tendon graft using transtibial tunnel technique was shown to be a successful technique at short to long term. Based on their initial experience with this technique, we continue to use the PCL arthroscopically assisted reconstruction with transtibial tunnel technique for isolated or combined reconstruction of the PCL.

Knee: meniscus P15-8 Long-term outcome of meniscus repair in stable and ACL-reconstructed knees: does it prevent osteoarthritis? Steenbrugge F.1, Verdonk R.2, Verstraete K.3 1 ASZ Campus Aalst, Orthopaedic Surgery and Traumatology, Kerksken, Belgium, 2Gent University Hospital, Dept. of Orthopaedic Surgery and Traumatology, Gent, Belgium, 3Ghent University Hospital, Department of Radiology, Gent, Belgium Objectives: A retrospective study was set up to evaluate the long-term resutls of meniscus repair in stable knees and ACL-reconstructed knees with regard to the development of osteoarthritis. Methods: The study included 137 patients. Group 1 (ACL-intact) included 75 patients, group 2 (ACL-reconstructed) 62 patients. Mean follow-up was 10 years 7 months for both groups and mean age at follow-up was 37 years 10 months for both groups. Evaluations were done using the Modified Hospital for Special Surgery Knee Score. MRI was used for radiographical evaluation. Results: Group 1 showed 83% good-to-excellent results and group 2 showed 75% good-to-excellent results on long-term follow-up. Failure rate of meniscal repair was two times higher in the ACL-repaired group compared to the stable knee group. Radiographically detectable degenerative changes were also two times higher in the ACL-repaired group compared to the stable knee group. Conclusions: The status of the ACL determines the success-rate of meniscal repair. The status of the meniscus determines the success-rate of ACL reconstruction with regard to the development of osteoarthritis.

P15-14 Arthroscopic meniscal repair by outside-in technique using a peripheral endovenous catheter needle and polyglactin sutures Santos Duarte Lana J.F.1, Santos Assis de Melo N.2, de Oliveira Neri I.D.2, Pidner Neto H.3 1 Clinica MOR e Hospital Sa˜o Marcos - Uberaba (MG), Mogi Mirim, Brazil, 2Santa Casa de Miserico´rdia de Mogi Mirim, Mogi Mirim, Brazil, 3 Hospital da Beneficeˆncia Portuguesa de Bele´m (PA), Mogi Mirim, Brazil Objectives: Aiming to render arthroscopic meniscal repair more feasible, using simple, cost-effective materials, readily available in a large number of hospitals, the authors have developed a modified arthroscopic transmeniscal suture technique for the outside-in method described by Cooper, Arnoczky and Warren, using a peripheral endovenous catheter (14G or 16G) and polyglactin sutures. Methods: In the period between January 1999 and October 2008, for suturing meniscal injuries the authors used the outside-inside technique, with a 14 or 16G peripheral endovenous catheter needle and 1 or 2-0 polyglactin strand. 189 repairs were performed, with at least 6-month follow-up post suture. Once the meniscal injury is visualized in the red-red or white-red areas, the shaver is used to roughen its edges, with the aim of unclogging the vascular canaliculi. A 14G or 16G peripheral endovenous catheter and polyglactin strand (1 or 2-0) are the instruments used (Figure 1). The needle and strand are introduced intradermally into the joint, using the transilluminated area as a reference point (Figure 2). Under direct intra-articular arthroscopic vision, the edges of the injury are transfixed (Figure 3). A slight retreating of the needle creates a loop in the

123

S230 strand (Figure 4). This loop allows the surgeon to pull the end of the strand outwards from the inside of the joint through the portal, while an assistant secures the other end at the entry point in the skin (Figure 5). Both ends of the strand are repaired. The same sequence is then repeated, as described before, in a way that, inside the joint, the distance between the strands varies between 3 and 6 mm (Figure 6). The two strands emerging from the portals are tied together in a simple knot (Figure 7) and placed at the tip of a hemostat or grasper (Figure 8). While the assistant applies traction to the ends of the strand in the skin, the surgeon guides the knot, aided by the pincers, insidewards, avoiding soft tissue interposition (Figure 9). The same sequence is repeated if the injury needs further stabilization. In cases where there is associated ligament injury, it is first reconstructed before finalizing the suture. At the site in the skin where the strands were inserted, an incision is made and dissection to the capsule is performed. The ends of the strands are tied together in a triple knot (Figure 10). The stability of the suture is tested with a hook. In order to tension the medial meniscal suture, the knee is flexed to 708 to make a ‘‘4’’. For the lateral meniscus, the knee hangs freely at 908 flexion. Results: 189 repairs were performed;156 with at least 6-month follow-up post suture; of these, 134 were medial menisci and 22 lateral menisci. In 112 cases, concurrent ACL reconstruction was performed and in 2 cases so was the PCL. 42 were isolated injuries. 91.7% good and excellent outcomes were obtained and the complication rate was low. Conclusions: The authors presented an operationally economical, modified outside-in arthroscopic transmeniscal suture, described by Cooper, Arnoczky and Warren (20), with the aid of readily available instruments, easily-performed, with 91.7% good and excellent results and low complication rate. The limitation this technique presents consists of the difficulty or impossibility of suturing more posterior injuries. The technique presented by the authors encourages surgeons to preserve the menisci, whenever possible.

P15-107 The relationship between medial meniscus posterior root tear and medial meniscal extrusion Choi C.J.1, Choi Y.J.2, Choi C.H.2 1 Kangnam Severance Hospital, Seoul, Republic of Korea, 2 Gangnam Severance Hospital, Seoul, Republic of Korea Objectives: The purpose of this study was to evaluate the relationship between medial meniscal extrusion (MME) and medial meniscus posterior root tear. Methods: From January 2008 to December 2009, 189 consecutive cases of medial meniscal tears were treated using arthroscopic surgery. Of these, 121 cases with MR imaging were reviewed. Arthroscopic findings were reviewed for the type of tear and medial compartment articular cartilage lesion. Tears that involved the ‘‘root’’ at the tibial spine were recorded. And MR imaging of the knee were evaluated for the presence and degree of MME. MME C3 mm was considered pathologic. Arthroscopic findings (posterior root tear, type of meniscal tear, and articular cartilage lesion) were compared with respect to the degree of MME. Results: The results showed 61 cases (50.4%) of the medial meniscus to have MME C3 mm. We found an association between C3 mm MME and medial meniscus posterior root tear (p=0.001), radial tear (p=0.016) and medial compartment articular cartilage lesion (p=0.002) when compared with the knees without these findings. Longitudinal, horizontal, oblique and complex tears were not associated with MME. Conclusions: Substantial MME is associated with tears involving the meniscal root. In addition, the presence of significant (C3 mm) MME should prompt the clinician to specifically exclude pathology at the meniscal root attachment site.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P15-112 Meniscal repair using the FasT-Fix device in patients with chronic meniscal lesions Popescu D.1, Sastre S.2, Garcia E.1, Lozano L.1, Segur J.M.3, Macule´ F.3 1 Hospital Clinic de Barcelona, Knee Unit, Barcelona, Spain, 2Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain, 3Hospital Clı´nic i Universitari de Barcelona, Knee Unit. Orthopaedic Surgery Department (ICEMEQ), Barcelona, Spain Objectives: The aim of this prospective study is to evaluate meniscal suturing when using the FasT-Fix device for chronic meniscal lesions. Methods: This procedure was carried out on 25 patients between January 2006 and November 2007. Average patient age was 31.1 years old (14-47) and the gender ratio was M/F 19/6. The mean waiting time until surgery was 27.7 months (6-80). Average follow-up was 19.6 months (14-29). 11 patients (44%) required reconstruction of an Anterior Cruciate Ligament (ACL) associated lesion. 20 patients (80%) showed medial meniscus lesions. All lesions were located in the red zone or red-white zone. Results: According to the Barett criteria, meniscal lesion healing was achieved in 21 patients (84%). 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. There were no neurovascular complications. Revision surgery was necessary in one patient, in whom a partial meniscectomy was performed. Conclusions: The results obtained suggest that chronic meniscal lesions in the zones described can be cured. The FasT-Fix meniscal repair suture system assures a reliable and long-lasting suture and has proved to be a safe and effective method.

P15-134 Mid-term outcomes after meniscal allograft transplantation: comparison of cases with extrusion versus without extrusion Bin S.I.1 1 Asan Medical Center, Department of Orthopedic Surgery, Seoul, Republic of Korea Objectives: Although graft extrusion is of concern after meniscus allograft transplantation (MAT), the correlation between extrusion and clinicoradiologic outcomes remains unclear. We hypothesized that patients with graft extrusion after MAT would have worse clinical outcomes and further arthritic change compared to 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 MRI findings. The mean patient age at the time of surgery was 33.5 years (range, 17-43 years), and the mean follow-up period was 5.1 years (range, 3.5-8.3 years). The Lysholm score was used to evaluate knee function. In addition, pre- 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: MRI 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 two 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 pre- and postoperative JSW across all patients, or in either group. Conclusions: JSW was slightly narrower following MAT. Extrusion had no effect on 5-year clinico-radiologic outcomes.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P15-135 Evaluation of meniscus allograft transplantation with serial magnetic resonance imaging during the first postoperative year: focus on graft extrusion Bin S.I.1 1 Asan Medical Center, Department of Orthopedic Surgery, Seoul, Korea, Republic of Objectives: The purpose of this study was to evaluate change in meniscal allograft extrusion during the first year after meniscus allograft transplantation (MAT) by use of serial magnetic resonance imaging (MRI). Methods: Twenty-three consecutive patients who underwent MAT between June 2002 and March 2006 with total or subtotal meniscectomy of the knee were enrolled in the study. The medial meniscus was affected in 3 cases (15%) and the lateral meniscus in 18 cases (85%). Conventional MRI of the affected knee was performed in all 21 patients at 6 weeks and at 3, 6, and 12 months after MAT. We recorded subluxation at the 4 different time points by measuring the absolute value and relative percentage of extrusion (RPE), defined as the percentage of the width of extruded menisci compared with the width of the whole meniscus, using coronal images. We considered subluxation of greater than 3 mm to indicate extrusion. Results: The mean amount of graft subluxation on serial MRI was 2.87, 2.95, 3.03, and 2.96 mm at 6 weeks, 3 months, 6 months, and 12 months after MAT, respectively. The mean RPEs were 29.2%, 29.4%, 32.4%, and 31.9%, respectively. These values were not considered statistically significant among the 4 different time periods (P[.05). Seven extruded cases (33.3%) at 6 weeks postoperatively remained extruded until the final follow-up at 1 year postoperatively. With 1 exception, the other 14 cases without extrusion at 6 weeks postoperatively did not undergo extrusion until 1 year postoperatively. Conclusions: A meniscus that extrudes early remains extruded and does not progressively worsen, whereas one that does not extrude early is unlikely to extrude within the first postoperative year.

P15-145 Long-term clinical outcome of open meniscal allograft transplantation van der Wal R.1, Thomassen B.1, van Arkel E.1 1 MC Haaglanden, Orthopedic Surgery, Den Haag, Netherlands Objectives: Meniscal allograft transplantation is an accepted treatment for the symptomatic postmeniscectomized knee in younger patients; however, long-term data are scarce on clinical outcome. Hypothesis: Cryopreserved meniscal allograft transplantations can, in the long term, be a good alternative for the symptomatic postmeniscectomized knee in younger patients. Methods: Sixty-three meniscal allografts (40 lateral and 23 medial) were transplanted with an open procedure in 57 patients. Clinical outcome and failure rate were evaluated at 13.8±2.8 years. Mean age at time of transplantation was 39.4±6.9 years. Results: Eight medial and 10 lateral allografts failed (overall failure rate, 29%). Overall Lysholm scores significantly improved from 36±18 preoperatively to 61±20 at long-term follow-up. Scores were not significantly different in the following subgroups: medial allografts, female patients, and left treated knees. All subgroups had poor scores at mean follow-up of 13.8 years, except the male patients group, which had a fair score. Short-term Lysholm scores were 79±19 at 3.1±1.5 years. All subgroups demonstrated a significant difference between short- and long-term Lysholm scores. There were no significant differences in Lysholm scores between posttransplanted survivors and posttransplanted nonsurvivors who received a total knee arthroplasty. For 2 other measures-the Knee injury and Osteoarthritis Outcome Score and the International Knee Documentation Committee scoring system-significant decline was present between men and women only. Conclusions: Long-term follow-up results show that meniscal allograft transplantation is a beneficial procedure. Good improvements in clinical function and pain relief have been shown at short-term follow-up in this

S231 population. Despite the deterioration in function scores over time, there is still improvement in level of function at long-term follow-up but not at a high level. As such, meniscal allograft transplantation is a good salvage option for the treatment of degenerative arthritis of the symptomatic postmeniscectomized knee. Meniscal allograft transplantation can be used to postpone total knee arthroplasty in younger patients.

P15-216 Teaching arthroscopy: analysis of verbal communication in the operating room Tuijthof G.1, Thomas R.2, Sierevelt I.2, Schafroth M.2, van Dijk C.N.2, Kerkhoffs G.2 1 Academic Medical Centre / Delft University of Technology, Orthopedic Surgery / Biomechanical Engineering, Amsterdam, Netherlands, 2 Academic Medical Centre, Orthopedic Surgery, Amsterdam, Netherlands Objectives: It is acknowledged that existing methods of training surgeons in arthroscopy have limitations that can lead to long learning curves. Objective evaluation of training performance remains a challenge in assessing the effectiveness of new training methods in the operating room. Verbal communication reflects the interaction between the surgeon and the resident. Recently, an objective classification method was introduced based on the type and the content of the verbal communication during surgical training. The aims of this study were to adopt the classification method for an arthroscopic setting, to analyze typical communication patterns and to verify its potential use in objective assessment of training performance. Methods: Within a period of 3 months, 17 arthroscopic procedures were recorded with a special capturing system consisting of two video cameras and a tie-clip microphone that was mounted on the supervising surgeon. The operations were performed by three residents who were supervised by one of the two participating surgeons. The four communication types, previously introduced, were adopted: explaining, questioning, commanding and miscellaneous. Categories for communication content were discussed and refined by analyzing one operation until mutual agreement was achieved by two observers. This resulted in seven domains: operation method (e.g. what step, which sequence), anatomy and pathology, instrument handling (e.g. open punch, reposition instrument, stress joint), tissue interaction (e.g. increase portal size, push meniscus backwards), visualization (e.g. move scope, irrigation, focus), miscellaneous (general or private), and indefinable. Each digitized video of an operation was analyzed with Utilius software, where one observer marked all communication events and assigned them into proper categories. The frequency of categories was assessed as well as the presence of significant correlations between the frequency of communication events and experience of the resident, operation time, satisfaction of the surgeon (scale 1-5) and resident (scale 1-5) on the performance of the operation. Results: On average 4.7 (SD 1.2) communication events took place every minute. Explaining (45%) and commanding (38%) show considerable frequency compared to questioning (7%) and miscellaneous (11%). The explaining events were primarily on anatomy and pathology (42%) followed by operation method (25%). The commanding events were primarily on instrument use (30%) and visualization (25%). A significant correlation was found between number of communication events and the experience of the resident (Pearson 0.64, p\0.05). Conclusions: The classification method for verbal communication was adopted for an arthroscopic training setting and shows how surgeons teach residents verbally. The results highlight the high frequency of explaining and commanding events as opposed to questioning events. The latter could stimulate the learning experience of residents. Additionally, the high frequency of explaining anatomy and pathology and operation method could be reduced as these items are suitable for training outside the operating room. Further research is required to determine the value of this method in objectifying training performance.

123

S232 P15-218 The needle scope: old news or a useful new tool? Comparison of a needle scope vs. conventional arthroscopy in diagnostic procedures in the knee joint Diercks R.1, Stevens M.2 1 UMCG Groningen, Dep. of Orthopaedics and Sports Medicine, Groningen, Netherlands, 2University Medical Center Groningen, Orthopedics, Groningen, Niger Objectives: To compare a needle scope with conventional arthroscopy in diagnosis of intra-articular pathology of the knee joint. Methods: In 31 patients, a standard arthroscopic procedure of the knee joint was preceded by an arthroscopy performed with a needle scope. The visualisation of cartilage, syovium, cruciate ligaments and meniscus by the needle scope was compared with regular arthroscopy. Results: Th sensitivity with respect to chondral lesions for the patello femoral and medial compartment was respectively 92.8% and 87.5% and for specificity 94.1% and 93.3%, which can be considered high. For the lateral compartment these values were considerably lower with a sensitivity of 71.4% and a specificity of 75.0%. With respect to meniscus lesions a same pattern was seen. For the medial compartment sensitivity (100%) and specificity (94.4%) can be considered high and for the lateral compartment moderate with respectively a sensitivity of 87.5% and a specificity 56.5%. Finally the accuracy both with respect to the chondraland meniscus lesions could be considered high, once again with exception of the lateral compartment. With respect to chondral lesions sensitivity and specificity for the patello femoral and medial compartment can be considered high and for the lateral compartment moderate. With respect to meniscus lesions sensitivity and specificity for the medial compartment can be considered high and for the lateral compartment moderate. Conclusions: Needle arthroscopy is not an alternative for conventional arthroscopy. Especially with respect to the lateral compartment sensitivity, specificity and accuracy can be considered insufficient. This is also confirmed by means of the moderate associations between the needle- and conventional arthroscopy diagnoses. As a result needle arthroscopy will not lead to any sustainable benefit for the patient, treating physician or care provider.

P15-251 Lysholm and WOMET - knee scores for patients with degenerative meniscal tear: how much is the minimal clinically important improvement? Ja¨rvinen T.L.1, Sihvonen R.2, Ja¨rvela¨ T.3 1 University of Tampere, Faculty of Medicine, Tampere, Finland, 2 Hatanpa¨a¨ Hospital, Department of Orthopaedics and Traumatology, Tampere, Finland, 3Hatanpa¨a¨ Hospital, Sports Clinic and Hospital Mahila¨inen, Tampere, Tampere, Finland Objectives: Lysholm knee score was originally developed for the assessment of patients with knee instability, but has recently been validated also for patients with meniscal tears. Although knee pain is the primary symptom of a meniscal tear, patients often complain of other symptoms, such as impaired knee function and even emotional distress. To more comprehensively elucidate the wide range of various symptoms related to meniscal tears, a Health-related Quality of Life (HrQoL) -instrument called Western Ontario Meniscal Evaluation Tool (WOMET; range 0-100, 0 for the best possible score) was recently introduced and validated for patients with meniscal pathology. Unfortunately, a valid measurement tool does not suffice for the execution of a clinical trial, but one also needs an adequate sample size (sufficient statistical power). Traditionally, the assessment of the efficacy of treatment given has been based on the statistical significance of the intervention-induced change. However, the problem with such a statistical approach is that the change observed might not necessarily be clinically important. Therefore, the aim of this study was to determine the minimal clinically important improvement for Lysholm and WOMET knee scores in patients with arthroscopically-verified and -treated meniscal tear.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: 434 consecutive patients with degenerative meniscus tear treated with arthroscopic partial meniscectomy were assessed prospectively (preoperatively and at 6-month postoperative follow-up) using the Lysholm knee score, the WOMET knee score, and the patients0 global impression of change at 5-point Likert -scale (much better, better, the same, worse, much worse). Of these patients, 355 (82%; 268 medial, 55 lateral, and 32 bilateral meniscus tears) returned all questionnaires at 6month follow-up. The smallest clinically important improvement was estimated for both Lysholm and WOMET scores using a so called ‘‘anchor based’’ -method with the patients0 global impression of the change used as the anchor. Those reporting ‘‘better or much better’’ -improvement were considered responders, whereas those reporting ‘‘the same, worse or much worse’’ were considered non-responders. Responsiveness was analysed using the area under the receiver operating characteristic (ROC) -curve, which is a statistical tool used to define the cut point for a clinical test. The cut point was defined as the point of maximum sensitivity and specificity. Results: Preoperatively, the medians of the scores were 51 (range 11-95) and 55.0 (range 3-94) for the Lysholm and WOMET scores, respectively. At the 6-mo follow-up, the corresponding values were 74 (range 11-100) and 22.8 (range 0-100), respectively. Accordingly, the medians of the changes were 19 (range -32 to 64) in the Lysholm score and 22.3 (range 31 to 76) in the WOMET, respectively. Area under the ROC curve was 0.870 for the Lysholm score, and 0.907 for the WOMET score, both of which depict a high accuracy. The smallest clinically important improvements were 12 for the Lysholm score and 10 for the WOMET score, respectively. Conclusions: According to our study, the minimal clinically important improvement by the patients with arthroscopically verified degenerative meniscal tear was 12 and 10 points in the Lysholm and WOMET scores, respectively. These values can be utilized in the future for the determination of the sample sizes required to detect treatment effects with sufficient probability in arthroscopic treatment of meniscal tears.

P15-259 Comparison of arthroscopic medial meniscal suture repair technique: inside-out versus all-inside repair Choi N.-H.1, Kim T.-H.2, Jung Y.-H.3 1 Eulji College of Medicine, Orthopaedic Surgery, Seoul, Republic of Korea, 2ChungJu St. Mary’s Hospital, Orthopaedic Surgery, Chungju, Korea, Republic of, 3Eulji College of Medicine, Seoul, Republic of Korea Objectives: The purpose of this prospective study was to compare healing rate between meniscal repairs with inside-out and all-inside suture repair in anterior cruciate ligament reconstruction patients. Methods: 48 consecutive patients underwent meniscal repairs of longitudinal tear of the posterior horn of the medial meniscus combined with hamstring ACL reconstructions. 14 patients had all-inside meniscal repairs and 34 patients had inside-out meniscal repairs with absorbable sutures. Identical postoperative rehabilitation protocols were used for both groups. Postoperative evaluations included Lysholm knee scoring scale, Tegner activity levels, Lachman and pivot-shift tests, and KT-1000 arthrometer. Assessment of meniscal status was performed using joint line tenderness, McMurray test, and range of motion. Follow-up MRI scans were obtained on all patients. Results: The average follow-up was 35.7 months. No patient had joint line tenderness, or complained of pain or click on McMurray test in either group. There was no significant difference in range of motion between the two groups. Follow-up MRI scans demonstrated that 10 (71.4%) menisci were healed and 4 (28.6%) were partially healed in the all-inside group. 24 (70.6%) menisci were healed and 10 (29.4%) were partially healed in the inside-out group. There was no significant difference in meniscal healing between the two groups. There were no differences in Lachman test, KT-1000 side-to-side differences measurements, Lysholm scores, and Tegner activity scales. However, there was a significant difference in pivot-shift test between the two groups (p=0.023).

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Conclusions: There was no significant difference in meniscal healing between inside-out and all-inside repair techniques in combination with hamstring ACL reconstructions.

P15-260 Meniscal repair for radial tear of the mid-body of the lateral meniscus Choi N.-H.1, Jung Y.-H.2, Kim T.-H.3 1 Eulji College of Medicine, Orthopaedic Surgery, Seoul, Republic of Korea, 2Eulji College of Medicine, Seoul, Republic of Korea, 3ChungJu St. Mary’s Hospital, Orthopaedic Surgery, Chungju, Republic of Korea Objectives: Radial tears of the meniscus have been treated by partial meniscectomy, although they are more biomechanically detrimental than longitudinal tears. Clinical results after meniscal repair for radial tear of the midbody of the lateral meniscus have been rarely reported. Methods: 14 consecutive patients who had radial tear of the midbody of the lateral meniscus underwent arthroscopic meniscal repair. Inclusion criterion was radial tear involving the red-red or red-white zone. All patients underwent all-inside meniscal repair using suture materials. Postoperative evaluation was performed using joint line tenderness, McMurray test, range of motion, and follow-up magnetic resonance imaging (MRI) scan at 6 months postoperatively. Lysholm knee score and Tegner activity level were evaluated at last follow-up. In 3 patients, 2nd look arthroscopies were performed. Results: The average follow-up was 36.3 months. No patient had joint line tenderness. 3 patients complained of pain or click on McMurray test. The mean follow-up range of motion was 138.6 degree. Follow-up MRI scans demonstrated that 5 (35.7%) menisci were healed, 8 (57.1%) were partially healed, and 1 (7.1%) meniscus was not healed. The follow-up Lysholm score was 94.7 (81-100, SD=6.4) and Tegner score was 5.7 (3-7, SD=1.4). The 2nd look arthroscopies showed partially healing of meniscal tear in 3 patients. Conclusions: Meniscal repair for radial tear of the midbody of the lateral meniscus may be an alternative treatment option to partial meniscectomy.

P15-290 The relationship between osteochondritis dissecans accompanied with discoid lateral meniscus and shape of lateral femoral condyle Kamei G.1, Adachi N.1, Deie M.1, Nishimori M.1, Nakamae A.1, Nakasa T.1, Ochi M.1 1 Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan Objectives: Discoid lateral meniscus is a congenital meniscal anomaly that can produce knee symptoms such as pain, limping and snapping. Occasionally, discoid lateral meniscus cause osteochondritis dissecans (OCD) in lateral femoral condyle (LFC). Various indirect signs with discoid lateral meniscus such as lateral joint space widening, hypoplasia of LFC, hypoplasia of the lateral tibial spine, high fibular head, and cupping of the lateral tibial plateau have been reported. Although there have been several reports on relationship between discoid lateral meniscus and OCD, there are a few reports that pay attention to the relationship between OCD accompanied with discoid lateral meniscus and shape of LFC. The purpose of this study is to evaluate the relationship between OCD accompanied with discoid lateral meniscus and shape of the LFC. Methods: From January 2003 to December 2008, we examined 65 knees in 52 patients who consulted our institution with symptomatic discoid lateral meniscus. There were 26 males (32 knees) and 27 females (33 knees) with the average age of 17.7 years old (range 7 to 66). In these patients, there were 18 knees in 15 patients who was accompanied with OCD (9 males; 11 knees, 6 females; 7 knees) with the average age of 14.9 (range 7 to 41) years old (OCD group).There were 47 knees in 37 patients who was not accompanied with OCD (17 males; 21 knees, 21 females; 26 knees), with the average age of 17.6 (range 7 to 66) years old (non-OCD group) All patients had standardized Rosenberg view radiographs of the knee. We evaluate the shape of LFC using Rosenberg view because we thought it would provide the proper image around the condyle. We measured the condylar prominence of the medial and lateral condyles adjacent to the intercondylar notch. Line A (A’) were drawn through the outermost points

S233 of the femoral condyle medially (A) and laterally (A’). Line B was drawn through the lowest points of femoral condyles. Point a and a’ intersect lines A (A’) and line B. Point c is the highest point in the intercondylar notch. Line C (C’) are the lines through point c and point a (a’). The prominence P (P’) were measured as the longest distance between the prominence of condyles and lines C (C’). The prominence ratio was defined as P’/P. Results: The OCD group had a significantly larger prominence ratio than non-OCD group (0.87±0.14 in OCD group, 0.73±0.09 in non-OCD group: P\0.0001). A prominence ratio of male was significantly larger than that of female (0.81±0.02 in male, 0.74±0.02 in female : P=0.0273). Conclusions: In this study, we clearly demonstrated that a prominence ratio in OCD group is significantly larger than that in non-OCD group, indicating that prominence of LFC can be a cause of OCD in LFC. One possible explanation is that discoid lateral meniscus can cause excessive stress to prominent LFC to produce OCD lesion. Prominence ratio of male is significantly larger than that of female, and OCD accompanied with discoid lateral meniscus tend to be more frequently in male in this study. These findings are consistent to previous reports. Our results that prominence ratio in male was higher than that in female can explain the higher occurrence rate of OCD in LFC in male patients.

P15-294 The DUPRA-trial - a randomized controlled trial for the effectiveness of intra-articular Ropivacain and Bupivacain on pain after knee arthroscopy Campo M.1, Kerkhoffs G.2, Sierevelt I.2, van der Vis H.1, Albers R.1 1 Tergooiziekenhuizen, lokatie Hilversum, Orthopaedics, Hilversum, Netherlands, 2AMC, Orthopaedics, Amsterdam, Netherlands Objectives: Effective pain relief is important after diagnostic and therapeutic arthroscopic knee surgery to permit early discharge and improve comfort, mobility and revalidation at home. Current literature on this topic remains inconclusive. The aim of this Randomized Controlled Trial was to assess the efficacy of Bupivacaine, Ropivacaine, or normal saline as placebo for postoperative pain relief after arthroscopic knee surgery. Methods: After power analysis, three groups of 90 healthy patients undergoing knee arthroscopy under general anaesthesia were randomized to receive one of the following substances intra-articular postoperatively. Group A: 10 mL of Bupivacaine (0.5%); Group B: 10 mL of Ropivacaine (0.75%); Group C: 10 mL normal saline (0.9%). Oral Paracetamol 1g and Diclofenac 50 mg were used as rescue drugs. Pre-operatively pain was assessed at rest and in flexion and postoperatively at 1 and 4 hours as well as at 5-7 days; patient satisfaction and side effects were also recorded. Results: Preliminary results showed that all groups experienced pain relief at 1 and 4 hours as well as at 5-7 days postoperatively. However, at 1 and 4 hours the Bupivacain an Ropivacain groups both had lower pain-scores compared to the saline group. There was no significant difference between the three groups on the 5-7 days assessment, not for pain in rest, nor in flexion, nor for the amount of rescue medication used. All patients were satisfied and there were no significant demographic differences between the three groups. No significant side effects were noted. Conclusions: From the preliminary results can be concluded that at 1 to 4 hours postoperative Ropivacain and Bupivacain both are more effective for post-operative pain relief than Saline. There is no significant difference between Ropivacain and Bupivacain 1 to 4 hours postoperative. Therefore, either Bupicacain 0.5% or Ropivacain 0.75% should be injected intraarticular after knee arthroscopy to optimize post-operative analgesia.

P15-443 Radial displacement of lateral meniscus; relationship with popliteomeniscal fascicle Ichiba A.1, Makuya K.2 1 Takatsuki Red Cross Hospital, Department of Orthopaedic Surgery, Osaka, Japan, 2Veritus Hospital, Department of Orthopaedic surgery, Kawanisi, Japan

123

S234 Objectives: Radial displacement (RD) of the meniscus suggests meniscus dysfunction. There have been a number of reports on RD of the medial meniscus, but reports on the lateral meniscus are rare. Lateral meniscus injury is frequently combined with anterior cruciate ligament (ACL) injury. The popliteomeniscal fascicle (PMF) is an important structure that stabilizes the lateral meniscus, and disruption of PMF has been demonstrated in knees with ACL injury. The purpose of this study was to evaluate how PMF disruption affected RD of lateral meniscus in knees with ACL injury retrospectively. Methods: This study included 96 patients (96 knees). In the ACL group (n = 63, mean age 29 years), arthroscopic isolated ACL reconstruction was performed. Meniscus tear was confirmed on arthroscopy, and meniscectomy or repair was performed if necessary. In the ACL group, knees with lateral meniscus tear were classified into the LMACL group. As a control, patients who had isolated lateral meniscus tear confirmed on arthroscopy with meniscectomy or repair, excluding ligament injury, discoid meniscus and osteoarthrosis, were classified into the LM group (n = 33, mean age 33 years). RD was evaluated by a coronal view on MRI. The length extruded from the lateral tibial plateau was measured and corrected for scale. Discontinuity of PMF was evaluated on sagittal view of MRI in both the anteroinferior fascicle (AIPMF) and posterosuperior fascicle (PSPMF). Results: Mean RD was 1.6 mm in the ACL group and 1.5 mm in the LM group (p = 0.41). In the ACL group, mean RD was 2.1 mm in the LMACL group (n = 27), and 1.0 mm in knees without LM tear (n = 36) (p = 0.032). In terms of the type of lateral meniscus tear, mean RD was 1.5 mm in the knees with longitudinal tears (n = 16), 2.9 mm in the knees with radial or complex tears (n = 11) in the ACL group (p \ 0.001), and the corresponding figures in the LM group were 0.7 mm (n = 9) and 1.7 mm (n = 24), respectively (p = 0.003). In the LMACL group, the mean RD was 2.7 mm in the knees that needed treatments (n = 14), and 1.5 mm in the knees without treatment (n = 13) (p \ 0.001). The ratio of discontinuity of the PMF was 81% in the ACL group, 88% in the LMACL group and 39% in the LM group. Discontinuity of PMF was expressed as (AIPMF/PSPMF); and the ratio of discontinuity was as follows: ACL group; (-/-) 19%, (-/?) 25%, (?/-) 13%, (?/?) 43%, LMACL group; (-/-) 11%, (-/?) 18%, (?/-) 15%, (?/?) 60%, LM group; (-/-) 61%, (-/?) 12%, (?/-) 21%, (?/?) 6%. The ratio of discontinuity of both AIPMF and PSPMF was higher in the knees with ACL tears. Mean RD was as follows: ACL group; (-/-) 0.9 mm, (-/?) 0.9 mm, (?/-) 1.3 mm, (?/?) 2.0 mm (p = 0.043), LMACL group; (-/-) 0.7 mm, (-/?) 1.8 mm, (?/-) 1.1 mm, (?/?) 2.5 mm (p = 0.22), LM group; (-/-) 1.8 mm, (-/?) 0.7 mm, (?/-) 1.3 mm, (?/?) 0.3 mm (p = 0.39). In the knees with ACL tears, RD was greater in the knees with discontinuity of both PMFs. Conclusions: Although the amount of RD was equal in the ACL and LM groups, the ratio of discontinuity of PMF was quite different. It was speculated that differences in mechanism of injury between the ACL group and the LM group resulted in the difference of PMF discontinuity. Because RD was significantly greater in knees with radial or complex tears and in knees with treatments, disruption of circumferential fibers of lateral meniscus might be associated with RD.

P15-472 Role of the anterior intermeniscal ligament in tibiofemoral contact pressure distribution on axial joint loading Poh S.-Y.1, Yew K.-S.A.1, Wong P.-L.K.1, Koh J.1, Chia S.-L.1, Howe T.S.1 1 Singapore General Hospital, Orthopaedic Surgery, Singapore, Singapore Objectives: The anterior intermeniscal ligament (AIML) is an anatomically distinct structure that connects the anterior horns of the medial and lateral menisci. We hypothesized that both menisci work together as a unit in converting axial joint loading into circumferential hoop stresses, due to intermeniscal attachments. Therefore, loss of the AIML could lead to increased tibiofemoral contact stress and predispose to arthritic change. In this cadaveric study, we compared tibiofemoral contact pressures on axial loading, before and after sectioning of the AIML. Methods: Five fresh frozen human cadaveric knees were mounted on a linear x-y motion table and loaded in extension under axial compression of 1800 N (about 2.5 times body weight for a 70 kg individual), using a

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 materials testing machine. Tibiofemoral contact pressures before and after sectioning of the AIML were measured using resistive pressure sensors. Results: Sectioning of the AIML produced no significant increase in the total force, mean contact pressure, peak contact pressure or change in contact area, in either the medial or lateral compartments of the knees. Conclusions: The menisci appear to work independently to convert axial loads into circumferential hoop stresses. This is probably due to their individual root attachments to the tibia. Inadvertent sectioning of the AIML during knee surgery, e.g, anterograde tibial nailing, anterior cruciate ligament reconstruction, meniscus transplantation and unicondylar knee replacement, may thus be functionally insignificant.

P15-537 The difference between the subjective and objective status of patients after operative treatment of the nontraumatic lateral meniscus tear with parameniscal cyst - one year follow up Pedak K.1, Rahu M.2 1 Arthron OY, Rehabilitation, Tallinn, Estonia, 2North Estonian Regional Hospital, Orthopaedics and Traumatology, Tallinn, Estonia Objectives: Most of lateral meniscus horizontal cleavage or flap tears on the antero-middle part of menisci are caused by minor or no trauma and at times those tears are accompanied by meniscal cysts. In those cases most patients, specially in lower sports activity level are visited doctor due to cysts formation and discomfort, not because of pain and limitation of sports activities. The operative treatment consist of carefull resection of the damaged menisci and debridement of cysts. Purpose of study to compare subjective and objective status of patient with nontraumatic lateral parameniscal cyst and antero-middle meniscus tear a year after operation. Methods: 13 patient (12 male, 1 female), mean age 32,7 (21 - 58) with lateral parameniscal cyst with antero-middle menical tears without trauma mechanism and significant sports activity limitations before operation were selected. Patients were operated by same surgeon using same technique : partial resection of the antero-medial horizontal cleavage or flap tears and debridement of meniscal cysts with in - out - in technique. Postoperative rehabilitation was done by following recommendations in paper. Patients had symptoms of discomfort and parameniscal cyst with mean time 13,6 months(2 - 60) before surgery. The evaluation follow up was performed one year after surgery using Tegner, Lysholm, KOOS, Sarimo scores and objectively differences of legs in One Leg Hope test and with Isometric Dynamometry Tests were compared. Results: All patients were one year after operation in previous sport activity level: Tegner score 5,6 (4 - 8). Results of subjective scores one year after surgery were: Lysholm 91,1 (75 -100), KOOS 93 (71,4 - 100), Sarimo cyst score 8 excellent, 5 good . The difference in One Leg Hope test was in operated leg non significantly decreased (6%) but in Isometric Maximal Voluntary Contraction (MVC) was in operated leg significantly 14,3% (p\ 0,05) decreased and rate of Force development was at 25% (G25) - 11,4% and at 50% (G50) - 20% decreased. Conclusions: According the facts, that all patients did not have before operation significant limitation of their sports activities (probably because their activity level was low) his period without sports activities in pre- and postoperative period was very short and recovery seems to be more faster as in traumatic cases. However despite good results one year after operation in subjective scores and one leg hope tests, were found significant differences in muscle isometric dynamometric tests, which may lead to the reinjury, specially in more higher ,,pivoting’’ sports activity level and must have under special attention in rehabilitation period.

P15-585 Relationship of the neural structures to arthroscopic posterior portals in different knee positions: an anatomic study Koo K.H.1, Ahn J.H.1, Lee S.H.2 1 Samsung Medical Center, Department of Orthopedic Surgery, Seoul, Korea, Republic of, 2Samsung Medical Center, Orthopaedic Surgery, Seoul, Republic of Korea

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Objectives: Given the proximity of the neural structures, special care is warranted in the placement of posterior arthroscopic portals on knee joints. But there have been limited studies assessing the relative safety of the posterior portals used in knee arthroscopy in terms of their distance from neural structures. So the aim of this study was 1) to investigate the relationships between nearby neural structures and the standard posterior portals in their proximity in different knee positions, 2) to provide specific surgical landmark to establish the posterolateral portal. The hypothesis was that the neural structures are at increased risk for injury by the formation of posterior portal as the knee is brought into increased extension. Methods: Ten fresh cadaveric knees were used to evaluate the proximity of neurovascular structures to the posteromedial and posterolateral portals of the knee at different positions. Each knee was secured with the proximal part of the femur by a specimen holder and performed diagnostic arthroscopy in supine position. The standard posteromedial and posterolateral portals were established using an outside-in technique with transillumination technique and marked using an 18-gauge spinal needle. Each specimen was carefully dissected after procedures, and the distance from each portal site to the adjacent relevant neurovascular structures (Infrapatellar branches of sapenous nerve (IPSN) and sartorial branch of saphenous nerve (SSN) from posteromedial portal, common peroneal nerve (PN) from posterolateral portal) was measured using a precision caliper. Distances were recorded with the knees in extension, 45 of flexion and 90 of flexion to examine the effect of dynamic knee motion on nerve position. The distances between bony prominence of fibular head and posterolateral portal with the knees in 90 of flexion, were also measured. Results: The mean distance from the posteromedial portal to SSN at 90 of flexion was significantly greater as 26.1 mm (range, 18.1-48.4 mm), compared with 18.4 mm (range, 10.5-38.3 mm) at 45 of flexion and 14.0 mm (range, 1.5-35.5 mm) at full extension (p\0.0001). The mean distance from the posterolateral portal to PN at 90 of flexion was also significantly greater as 25.4 mm (range, 16.0-32.6 mm), compared with 22.5 mm (range, 15.0-28.7 mm) at 45 of flexion and 20.1 mm (range, 11.0-28.5 mm) at full extension (p\0.0001). The mean distance from the posteromedial portal to IPSN was no significantly difference with different knee position. The posterolateral portal was consistently located to proximal and posterior direction from the fibular head in all cases. The mean distance from the posterolateral portal to the bony prominence of fibular head was 37.8±2.8mm proximally and 12.1±4.0 mm posteriorly in 90 of knee flexion. Conclusions: In our cadaveric study we found that the lower knee flexion angles might have the potential risks of damage to the neural structure, especially SSN and PN. Therefore, the establishment of posterior portals at knee flexion angle of 90 is recommended to ensure safety of the neural structures. The bony prominence of fibular head was useful landmark to make the posterolateral portal if it is not possible to use the transillumination technique.

P15-587 Bilaterality in symtomatic discoid lateral meniscus knees: prospective magnetic resonance imaging study Lee S.H.1, Ahn J.H.1, Chang M.J.1 1 Samsung Medical Center, Orthopaedic Surgery, Seoul, Republic of Korea Objectives: The incidence of bilaterality in discoid lateral meniscus (DLM) is poorly defined but has been reported to be 10-20%. These studies could not determine the true prevalence of bilaterality in DLM because most of the previous studies included only the patients who underwent an operation. The same shape menisci in both knee was found in 253 (90.7%) of 279 cadavers of a Japanese cadaveric study. The hypothesis was that the possibility of the other knee having a DLM would high, when a DLM is clinically found unilaterally. Therefore, we think that the possibility of an unaffected knee having a DLM is high when a unilateral DLM is clinically found. The aim of this study was to analyze the contralateral DLM status of 33 patients who underwent

S235 an operation for symptomatic DLM by performing magnetic resonance imaging (MRI). We formed the hypothesis that 1) the patients with a symptomatic DLM may have a significantly higher prevalence of bilaterality than has been previously reported, and 2) the knee of the unaffected side may be associated with tears of the DLM or chondral lesion. Methods: Thirty five patients underwent arthroscopic surgery for a symptomatic DLM at our institution between Oct 2006 and Nov 2008. We prospectively collected a diagnostic study of these patients, who all agreed to undergo clinical and radiographic evaluations that included MRI of the opposite knee. This diagnostic study included 33 consecutive patients who underwent arthroscopic surgery for a unilaterally symptomatic DLM. All the patients’ opposite knees were checked with simple X-rays and MRI scans. They were also evaluated by physical examinations and clinical scoring. Mean age was 22.4 years (range: 19 to 35 years) and 22 of the 33 were left knees. The MRI findings of the lateral meniscus were simply categorized into three types: the normal type (NM), the incomplete discoid type (ICDM) and the complete discoid type (CDM). The combined meniscal tears and chondral lesion were evaluated. Results: On the simple X-ray, 23 (69.7%) of 33 knees were seen characteristic findings of DLM. On the MRI scans, there were 26 (78.8%) CDM knees, 6 (18.2%) ICDM and 1 (3.0%) NM. Subsequently, 32 (97.0%) of 33 knees were diagnosed complete or incomplete DLM. According to exam the paired knees of the lateral meniscus shape 29 (87.9%) had the same shape menisci in both knees. A DLM exhibiting tears was found in 11 (33.3%) cases and chondral lesions in 4 (12.1%) cases of the only CDM cases. Conclusions: Our study indicated that DLM commonly occurs bilaterally and especially in patients with symptomatic DLM that requires an operation. We recommend that orthopedic surgeons should mention these patients about the possibility of DLM as well as combined tear or degenerative change in the opposite knee, even if the opposite knee is clinically negligible.

P15-603 A prospective outcomes study of meniscal allograft transplantation LaPrade R.1, Wills N.1, Spiridonov S.1, Perkinson S.1 1 University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States Objectives: The menisci have been reported to perform vital roles in the knee joint, including load bearing, load transmission, shock absorption, joint stability, joint lubrication, and joint congruity. It is well established that premature, progressive chondromalacia of the ipsilateral compartment occurs after meniscectomy. Therefore, increased emphasis has been placed on the preservation of meniscal tissue whenever possible. In these patients, meniscal transplantation is an option that can potentially result in pain relief, decreased effusions, and improved function. The purpose of this prospective study was to report the clinical outcomes of meniscal transplantations in a consecutive series of patients who had undergone a previous total or subtotal meniscectomy and who had pain and activity related effusions. We hypothesized that meniscal transplants would significantly reduce pain and improve patient function, demonstrated by a significant improvement in patient outcome scores. Methods: We prospectively followed the outcomes of meniscal transplants in forty consecutive patients between July 2003 and December 2006. The presenting author performed all of the operations. The meniscal transplantations were performed with an arthroscopically-assisted technique. Baseline outcome scores were obtained before surgery. Subjective functional data was collected through the International Knee Documentation Committee (IKDC) Subjective Evaluation form and the modified Cincinnati Knee Rating survey. Objective outcome scores were obtained through the IKDC scoring system. The same outcome scores were obtained at routine clinical follow-up in all patients. Statistical comparison between the preoperative and follow-up outcome scores was performed using a Student’s t-test.

123

S236 Results: Baseline overall Cincinnati Knee Rating scores on the entire population averaged 55.2. At an average of 2.5 years (range, 1.8 to 4.0 years), 34 patients were available for follow-up. Their overall Cincinnati knee rating scores improved to an average of 75.3 (p\0.01). The IKDCsubjective form demonstrated a baseline score of 54.5 for all patients. Thirty four patients had complete IKDC-subjective forms available at follow-up. These patients demonstrated a significant improvement (72.0) in overall function (p\0.01).The twenty one patients who underwent lateral meniscal transplants had baseline Cincinnati scores of 57.8. Fifteen of these patients had follow-up scores averaging 77.9 (p\0.01). Nineteen patients who underwent medial meniscal transplants had baseline Cincinnati scores of 52.3 which improved eleven points to 73.2 (p\0.01). On the IKDC subjective questionnaire, patients who underwent lateral meniscal transplants (n=21) improved from baseline scores on average of 57.6 to 76.6 (p\0.01). For patients undergoing medial meniscal transplants baseline scores (n=19) averaged 51.2 and improved to 68.2 (p\0.01). The preoperative IKDC objective scores measuring effusion improved significantly from 6-A (normal), 29-B (nearly normal), 5-C (abnormal), when compared to the postoperative scores of 33-A (normal) and 1-B (nearly normal). Conclusions: In conclusion, the results of this study demonstrate that meniscal transplantation reduces pain, decreases activity related effusions, and improves function in patients experiencing the signs and symptoms of post-meniscectomy arthritis, when ligamentous instability or axial malalignment was also corrected, at an average follow-up of 2.5 years.

P15-651 Meniscus preservation by suture repair for the tear in discoid lateral meniscus Yamada Y.1, Nakata K.2, Nakamura N.3, Mae T.4, Kanamoto T.5, Shino K.6 1 Moriguchi Keijinkai Hospital, Orthopaedic Sports Medicine, Moriguchi, Japan, 2Osaka University Graduate School, Orthopaedics, Suita, Japan, 3 Osaka University Graduate School, Orthopaedic Sports Medicine, Suita, Japan, 4Osaka University Graduate School of Medicine, Department of Orthopaedics, Suita, Japan, 5Osaka University Graduate School of Medicine, Suita, Japan, 6Faculty of Comprehensive Rehabilitation, Osaka Prefecture Unit, Suita, Japan Objectives: The purpose of this study was to evaluate the results of meniscus preservation by suture repair for the symptomatic discoid meniscus tear without partial resection and to elucidate the effectiveness of this procedure. Methods: A symptomatic tear in discoid lateral meniscus was examined by MRI before surgery and was accessed by arthroscope during surgery. If the tear was limited in the peripheral region of meniscus without the degenerative change in the body, the longitudinal tear was repaired by inside-out stacked-suture or outside-in technique with the use of autologous fibrin clot in order to preserve meniscus. We made three portals to evaluate the thick discoid meniscus.

Shema of suture repair for discoid lateral meniscal According to this indication, eleven menisci in 9 patients out of 91 surgically treated patients with complete discoid lateral meniscal tear has been repaired since 2002. Mean age was 14-year-old, ranged from 6 to 22. The average follow-up period after surgery was 29 months (range, 14 to 49).

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: Preoperatively, eleven had a knee pain, snapping, tenderness of the lateral side of the knee, and positive McMurray test, and four had a lack of extension. Postoperatively, all knees decreased knee pain and snapping except one knee which had mild pain on sport activity. Weight bearing X-ray examination showed that narrowing of the lateral joint space was found in 2 knees. No degenerative changes were seen by radiography. Preoperative and postoperative MRI demonstrated that the high intensity area of the periphery of the discoid lateral meniscus diminished in size in two knees and was no change in six.

Changing of MRI of the tear region. Nine knees were graded as excellent, one as good and one as poor according to the Ikeuchi’s scale. Conclusions: Meniscus preservation by suture repair for peripheral longitudinal tears in discoid lateral meniscus without the degenerative change of the meniscus body relieved pain and mechanical symptoms at the average of 29 months. In the case of this condition, preservation by suture repair could be one of the surgical options for discoid lateral meniscal injury in the young patients.

P15-657 Arthroscopic pullout suture repair of posterior root tear of the medial meniscus: radiographic and clinical results with a 2-year follow-up Lee J.H.1, Bae H.K.1 1 Chonbuk National University School of Medicine, Chonbuk Nati, Orthopedic Surgery, Jeonju, Korea, Republic of Objectives: This study was undertaken to document the short-term clinical efficacy of arthroscopic pullout suture repair in treating posterior root tears of the medial meniscus. Methods: From March 2004 to August 2006, 26 patients (27 knees) with posterior root tears of the medial meniscus were treated with arthroscopic pullout suture repair surgery by the senior author. Of these, 20 consecutive patients (21 knees) with a minimum of 2 years’ follow-up treated by arthroscopic pullout suture repair were analyzed. Clinical results by use of the Lysholm knee and Hospital for Special Surgery scores and radiographic grade were evaluated, both preoperatively and at final follow-up. In addition, the second-look arthroscopic findings for 10 knees were analyzed. Results: A radiographic evaluation using the criteria of Kellgren and Lawrence at final follow-up showed an increase in radiographic grade by 1 grade in only 1 knee. On the second-look arthroscopies performed in 10 knees (47.6%), all repaired menisci had healed completely without additional chondral lesions in the knee. The mean Hospital for Special Surgery scores improved from 61.1 preoperatively to 93.8 at final follow-up (P _ .0001), and the mean preoperative Lysholm knee scores improved from 57.0 to 93.1 at final follow-up (P _ .0001). Retear was found in 1 knee at the 6-month follow-up, and reoperation was performed with the same procedure used for the index surgery. Conclusions: Arthroscopic pullout suture repair is an effective treatment for alleviating meniscal symptoms in patients with a symptomatic posterior root tear of the medial meniscus with degenerated articular cartilage of less than grade III. In addition, no discernable degenerative arthritic changes were found in terms of radiographic features with our limited short-term follow-up.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P15-667 Long-term results after medial meniscectomy Hart R.1 1 General Hospital of Znojmo, Dept. of Orthopaedics and Traumatology, Znojmo, Czech Republic Objectives: The menisci play an important role in load transmission and shock absorption of the knee. The purpose of this retrospective study was to determine radiological and clinical long-term results after medial meniscectomy. Methods: Between 1985 and 1989, 66 patients underwent isolated total or subtotal medial meniscectomy. Cases with other injury (ACL, cartilage, lateral meniscus etc.) of both knees before, at the time of surgery or afterwards were excluded from the study. 51 patients visited the last follow-up control. The minimum follow-up was 20 years (range, 24 - 20 years). The age ranged at the time of surgery between 15 to 57 years (mean, 35 years). Long both limbs weight bearing radiographs were performed for all patients. The tibiofemoral axis difference between operated and normal knees and grade of osteoarthritic changes were evaluated. The clinical condition was evaluated according to Lysholm. Results: Considerable osteoarthritic changes (grade II or more) in the medial compartment of the operated knee were evident in 12 % of cases (6 patients), 2 patients already underwent high tibial osteotomy (HTO), and in 1 knee, the medial unicompartmental arthroplasty (UCA) was already implanted. In all but one case, there was no marked degeneration of the medial compartment of the contralateral knee (grade II or more). The significant tibiofemoral axis difference (varus deformity 2 or more) between operated and normal knees was apparent in 45 % of cases (23 patients; 3 cases after HTO or UCA were not included in this group). The results after the surgery according to Lysholm was excellent in 24, good in 14, moderate in 5 and bad in 5 cases; 3 patients who already underwent HTO or UCA were not evaluated. Conclusions: Osteoarthritis after the medial meniscectomy developed ‘‘only’’ in approximately 1/6 of cases but the tibiofemoral axis changed in 1/2 of evaluated knees; in all cases with moderate and bad results according to Lysholm the varus deformity was present. Surprisingly despite of considerable meniscectomy, 3/4 of patients at minimum 20 years after the surgery don’t feel marked subjective troubles; in these cases, there was no or only a little difference between operated and normal knees in Lysholm score.

P15-671 Results of medial meniscus bucket handle tears which were repaired with inside-out horizontal sutures Eren O.T.1, Gunaydin B.1, Armagan R.1, Sever C.1, Kuzgun U.1 1 Sisli Etfal Research and Training Hospital, Orthopaedics and Traumatology 1st Clinic, Istanbul, Turkey Objectives: To evaluate the clinical and MRI results of patients who had medial meniscus bucket handle tears which were repaired with only inside-out horizontal sutures. Methods: Thirty-two patients underwent medial meniscal repair with only inside-out horizontal sutures. Repairs for lateral meniscus, and repairs with other techniques were not included to the study.Arthroscopy knee surgery was performed in all patients by senor author (OTE). Their average age at operation was 25.5 years (range 17-35 years). The age of the tears varied from 1 month to 4 years (average 20 months). Only five patients had an isolated tear of the meniscus, the other 27 having meniscal tears in combination with ACL injuries which were reconstructed at the same operation. All patients were clinically evaluated at the average 22-month follow-up period. In 24 patients MRI were available and healing was evaluated. Results: Clinically, only 3 patients had minimal joint line tenderness. On MRI scans 17 repairs were completely healed (all were clinically asymptomatic), 7 cases were incompletely healed (5 were clinically asymptomatic 2 patients had joint line tenderness). There was no patients with recurrent locking or catching except one who had a second tear after the initial repair in which rerupture was at the sutured area without any

S237 trauma and it was rerepaired which was healed at the final follow up. In three ACL reconstructed patients had saphenous nerve lesion. Conclusions: The inside-out horizontal sutures for medial meniscal bucket-handle tears are reliable method for meniscal repairs.

P15-732 Does arthroscopic partial meniscectomy result in knee osteoarthritis? A systematic review with a minimum of eight year follow-up Petty C.1, Lubowitz J.2 1 Taos Orthopaedic Institute, Taos, United States, 2Taos Orthopaedic Institute Research Foundation, Taos, United States Objectives: Our purpose is to test the hypothesis that arthroscopic partial meniscectomy results in knee osteoarthritis at long term follow-up. Methods: We systematically reviewed PubMed search terms ‘‘meniscus AND arthritis AND knee’’ and ‘‘meniscectomy AND arthritis AND knee’’, and included English language, Level I-IV evidence studies reporting either radiographic and clinical osteoarthritis outcome measures with a minimum of 8 year follow-up after partial arthroscopic meniscectomy. Results: Five studies met the inclusion criteria. All reported both radiographic and clinical measures. All studies compared the normal, contralateral knee as a radiographic control, but none included a clinical control group. Follow-up ranged from 8 to 16 years. In all studies, operative knees demonstrated a statistically significant incidence of radiographic signs of osteoarthritis as compared to control knees. However, clinical symptoms of osteoarthritis were not observed. Furthermore, clinical outcomes did not correlate with radiographic findings. Conclusions: Our results demonstrate that radiographic signs of osteoarthritis are significant at 8-16 year follow-up after knee arthroscopic partial meniscectomy, but clinical symptoms of knee arthritis were not observed. Limitations include absence of clinical control groups, and heterogeneity of reported outcome measures. Future research of higher levels of evidence, and longer term follow-up is required to determine if the radiographic signs ultimately foreshadow clinical symptoms in patients post-arthroscopic partial meniscectomy. Radiographic signs of osteoarthritis are significant at 8-16 year follow-up after knee arthroscopic partial meniscectomy, but clinical symptoms of knee arthritis are not significant.

P15-766 Partial meniscectomy - results after a 6 to 9 year follow up Mauricio H.1, Pessoa P.1, Silva T.1, Silva C.1, Mendes D.1, Carvalho M.1, Virgolino M.1 1 H. O. S. Outa˜o, Setu´bal, Portugal Objectives: The objectives of our work are to evaluate the clinical and radiological results of partial meniscectomies in a mid-long term followup. Methods: We searched our database for partial meniscectomies performed between the year 2000 and 2003. We reviewed the clinical files, registering the cause, local and type of meniscal lesion and the initial complaints. The associated lesions and the complications were also registered. The patients were clinically evaluated during July and August of 2009. We registered the actual clinical status and the functional status using the KSS Scale. We compared the initial and the current radiological status of the joint using the Ahlback classification. The results were compared between the degenerative and traumatic meniscal injuries, the lateral or medial meniscus and with or without LCAassociated injury. Results: We found 348 partial meniscectomies in our database, and were able to clinically review 61 patients, with an average follow-up of 7yr 10months. 37 Cases were degenerative and 28 were traumatic. The average ages were, respectively, 55 and 34yrs. The major initial complaints were pain (84%) and joint effusion (55%), with locking in 57% of traumatic injuries. The number of lateral and medial injuries was similar in both groups. Most of the degenerative lesions were of the posterior horn, and the traumatic were mostly bucket-handle. 37% patients with degenerative lesions had documented chondral lesions (2,8 degree in average).

123

S238 9 patients of the traumatic group had an associated ACL tear that was repaired within the next year of the first surgery. There was a radiological degradation of the joint in both groups. The degenerative group was 0.86 degrees worse (0,7 initial to 1.39 final) and the traumatic was 0.96 worse (0.04 initial to 1.0 final.). 7 (19%) patients of the degenerative group already had a joint replacement (6 TKA and 1 UKA). Clinically, 40% of the degenerative group still had moderate pain (VAS 3-7) against 25 of the traumatic group). Both groups had a similar KSS orthopaedic (KSS-O) score (87), but the traumatic group had a better KSS functional (KSS-F) score (94 vs 87). The lateral meniscectomy had lower KSS-O and KSS-F scores, specially in the traumatic group (10 and 7 points lower, respectively) when compared with the medial meniscectomy. The lateral meniscectomy was also associated with a higher degree of pain (2,3 vs 1,7) and more patients complained of moderate pain (30% vs 8%). The patients with associated lateral meniscus and ACL injury had also lower KSS-O and F scores (19,7 and 14,4 points lower) when compared with isolated latera meniscus. We found no difference when comparing the medial meniscus with or without associated ACL injury. Conclusions: This is a retrospective descriptive study. Our results are similar to the results already published in the international literature. The long-term results are worse with associated chondral changes, lateral meniscal injuries and with associated ACL and lateral meniscus. The partial meniscectomy isn’t a consequence-free procedure, because it accelerates the joint degradation. The surgeon must be aware of the longterm effects of this procedure and try to save as much meniscus as possible, when the repair isn’t possible.

P15-826 Design of a polycarbonate-urethane meniscal implant: finite element approach Linder-Ganz E.1, Elsner J.1, Portnoy S.1, Zur G.1, Condello V.2, Zorzi C.2, Guilak F.3, Shterling A.1 1 Active Implants Corporation, R&D Center, Netanya, Israel, 2Sacro Cuore Don Calabria Hospital, Negrar, Italy, 3Duke University Medical Center, Orthopaedic Research Laboratories, Department of Surgery, Durham, United States Objectives: Meniscus dysfunction due to tear is a common knee injury which leads to degenerative arthritis, attributed primarily to the changes in knee load distribution. Clearly, there is a substantial need to protect the articular cartilage by either repairing or replacing the menisci. A ‘‘floating’’ Polycarbonate-Urethane (PCU) meniscal implant is proposed as a solution for restoring the function of the missing meniscus and for the reduction of pain, through improved tibial and femoral pressure distribution. The implant is composed of PCU embedded with polyethylene reinforcement fibers and its design is based on the geometry of the femur and tibia. Our goal was to develop an optimal meniscal implant, whose contact pressure with the tibial plateau (TP) would be similar to that of the natural meniscus. We hereby present one aspect of the implant bench-tests, finite element (FE) analyses of the medial knee under loading. Methods: The FE model geometry was based on MR-scans of a left knee from a cadaveric specimen. The implant was built based on the actual implant design and the fibers were taken as ‘‘ring-shape’’ parts with a fiber cross-sectional area of 0.0314mm^2. Finally, the 3D model was transferred to an FE solver (ABAQUS) for non-linear strain/stress analyses under load-bearing. Lower nodes of the tibia were fixed for all degrees of freedom and a vertical load of 1200N was applied to the femur. Flexionextension angle was selected to be 0, while the implant was located freely between femur and tibia. Several compositions of the implant were tested, e.g., utilization of 3 to 5 fiber conduits and 2 to 13 fibers per conduit. For each configuration, peak and average TP contact pressures were calculated. In addition, peak and average von Mises and tensile stresses were calculated for PCU and fibers, respectively. The model was validated by comparing pressures measured in cadaveric knees to calculated TP contact pressures. Results: Pressures calculated from the FE analysis were similar to measured TP contact pressures. In general, peak and average TP contact pressures

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 across all simulation cases, were in the same order of magnitude, e.g., *5.5 and *1 MPa, respectively. Similarly, PCU peak and average von Mises stresses were *4.5 and *0.8 MPa, respectively. Focusing on the fibers, the highest of all peak tensile stresses (565 MPa), was attained in the case of 4 and 5 conduits, with 2 to 4 fibers placed in the superior tunnel. Peak values were predicted in the superior conduit, where fewer fibers could be located due to practical molding reasons. However, in the case of only 3 conduits containing 9 fibers each, the peak tensile stress (444 MPa) was reduced by *20%. The average tensile stress in the fibers was 122-152 MPa. Peak and average tensile strains were0.38-0.58% and 0.12-0.15%, respectively. Conclusions: In this study, we presented strains/stresses results of FE analyses of a medial meniscal implant. These analyses allowed us to optimize the design of a composite PCU implant embedded with fibers by altering the geometry and composition such that contact pressures developed on the TP cartilage were comparable to those measured under an intact natural meniscus (5-9 MPa). Safety-wise, strain/stress values in both the PCU and the fibers remain within the allowed limits. We believe that the current device configuration will lead to the most uniform and physiological TP pressure distribution, thus decrease pressure on the cartilage and presumably reduce pain.

P15-833 A polycarbonate-urethane meniscal implant: size increments and population coverage Elsner J.1, Zur G.1, Condello V.2, Zorzi C.2, Guilak F.3, Shterling A.1, Linder-Ganz E.1 1 Active Implants Corporation, R&D Center, Netanya, Israel, 2Sacro Cuore Don Calabria Hospital, Negrar, Italy, 3Duke University Medical Center, Orthopaedic Research Laboratories, Department of Surgery, Durham, United States Objectives: Geometrical similarity dictates the degree of conformity between the menisci and the condyles. It has been shown that even very small changes in allograft meniscal implant sizes demonstrate a dramatic change in stress levels. Specifically, an accurate match between implant size and candidate knee is critical for the success of an artificial meniscal implant, as a means for restoring the contact pressure distribution to normal. We hypothesized that an artificial polycarbonate-urethane (PCU) meniscal implant could improve joint congruency and cartilage protection through size availability and material pliability. Our aims were (i) to determine the necessary increments of implant size providing optimal pressure distribution on the cartilage surfaces, and (ii) to predict the number of required sizes that will cover the candidate population, based on experimental and computational studies. Methods: A set of 12 geometrical parameters, representing typical lengths, widths and areas within the knee were measured in 130 MRI scans. Mean values were calculated to define the most prevalent implant size in the general population. A meniscal implant was then created (NUsurface) by means of an MRI-based reconstruction. The next step was to test experimentally and computationally the effect of implant sizes on the implant biomechanical functionality. Various PCU implant sizes were created by increasing/decreasing the reference implant size by 1-2% increments. These sizes were evaluated in the joint setting in terms of implant location, pressure distributions and resistance to static loading. First, an experimental evaluation was conducted by indicating whether the implant is located properly within the joint. Second, we measured the pressure distributions on the tibial plateau. For each test, a pressure score was calculated (0 to 100% grade, Linder-Ganz et al. 2009), taking into account peak pressure and contact area coverage with respect to the natural meniscus. An analogous finite element (FE) model of the knee was constructed based on MRI-scans. FE models of the various implant sizes were placed in the joint space and loaded under similar conditions. Tibial cartilage pressure and internal stresses/strains within the implants were calculated in each case. Similarly, a pressure score was calculated for each simulation. Results: The experimental evaluation narrowed our focus to the range of ±10% due to PCL interference at larger sizes and extrusion at smaller sizes. A qualitative examination of pressure maps followed by a quantitative evaluation showed that a ±2.5% deviation in the implant size still yielded an

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 optimal pressure distribution like the reference implant. Similar results were attained for FE analyses. In addition, internal stresses within the implant components remained within the safety limits allowed by the manufacturers. Conclusions: Experimental and computational methods have been employed to assess the effect of correct matching between an implant size and a candidate knee. Both methods confirmed that the PCU meniscus implant performs equally well in distributing joint compressive loads on the tibial plateau in a 5% range around the ‘true’ joint space. The meaning of this being that a relatively lenient safety-range exists for the choice of implant by the surgeon. This finding, together with statistical analysis of the natural distribution of knee sizes imply that an array of 9 implant sizes would adequately accommodate the needs of 95% of the population.

P15-1119 Comparison of healing rate after meniscal suture repair: red-red versus red-white zone Choi N.-H.1, Kim T.-H.2, Jung Y.-H.3 1 Eulji College of Medicine, Orthopaedic Surgery, Seoul, Korea, Republic of, 2ChungJu St. Mary’s Hospital, Department of Orthopaedic Surgery, ChungJu, Korea, Republic of, 3Eulji College of Medicine, Seoul, Republic of Korea Objectives: There are controversies of differences in healing potential according to meniscus tear site. The purpose of this prospective study was to compare healing rate between red-red and red-white zone using modern arthroscopic suture techniques in anterior cruciate ligament reconstruction patients. Methods: 48 consecutive patients who had longitudinal tear of the posterior horn of the medial meniscus underwent arthroscopic suture repair with concomitant hamstring ACL reconstructions. 34 patients had meniscal tears in red-red zone and 14 in red-white zone. In the red-red group, 9 patients underwent all-inside suture repair and 25 underwent inside-out repair. In the red-white group, 5 patients underwent all-inside suture repair and 9 underwent inside-out repair. All patients had MRI examination to evaluate meniscus healing 6 months after surgery. Clinical evaluation included Joint line tenderness, McMurray test, and range of motion for assessing meniscal status. Lachman test, pivot-shift test, and KT-1000 arthrometer were done to assess knee stability. Lysholm knee score and Tegner activity level were evaluated. Results: The average follow-up was 40.7 months. No patient had joint line tenderness, or complained of pain or click on McMurray test in either group. There was no significant difference in range of motion between the two groups. Follow-up MRI scans demonstrated that 23 (67.6%) menisci were healed and 11 (32.4%) were partially healed in the red-red group. 11 (78.6%) menisci were healed and 3 (21.4%) were partially healed in the red-white group. There was no significant difference in meniscal healing between the two groups. There were no differences in Lachman test, pivotshift test, Lysholm scores, and Tegner activity scales. However, there was a significant difference in KT-1000 side-to-side differences measurements between two groups (p=0.041). Conclusions: There was no significant difference in meniscal healing between red-red zone and red-white zone following arthroscopic suture repair in concurrent hamstring ACL reconstructions.

P15-1138 Outside-in meniscus suture technique: ten years follow-up Marinescu R.1, Laptoiu D.1, Iozefina B.1 1 Colentina Clinical Hospital, Orthopedic and Trauma, Bucharest, Romania Objectives: The outside-in technique of arthroscopic repair remains effective for repair of most meniscal tears. The overall indications are similar to those for the commonly used inside-out technique. Previous studies have demonstrated that the location of the tear and the condition of the anterior cruciate ligament are important factors in determining the success of meniscal repair.

S239 Methods: A vertical suture orientation is preferred in order to evenly coapt the meniscus to the capsule. Knot-end sutures (so-called Mulberry knots) were used, 2 vertically stacked sutures, with 1 on each surface of the meniscus. If a mattress suture is used, a vertical orientation can be easily achieved with the outside-in technique. Use of an exogenous fibrin clot is suggested for isolated tears. The clot can be secured to the site of repair by a suture that has been placed through a spinal needle with the outside-in method. Between 1997 and 2001 we enrolled 68 patients in this study, from our 1045 knee arthroscopy cases. The initial prospective group consisted of 57 men and 11 women; average age was 27.6 years (range 1745 years). The time of the surgery was related to the first knee injury in 37 cases; the other 31 patients suffered repetitive injuries of the knee. The types of tears suitable for suture consisted of acute vertical tears, solitary or in association with a radial tear in middle third. The technique was always outside-in using resorbable sutures (2-0 PDS, 2-0 Maxon). Followup evaluation included clinical examination, radiographic examinations and second-look arthroscopies; 12 cases were lost at follow-up (median 9 years). Results: There were four initial failures in this study, consisting in rerupture of the meniscus. Healing was demonstrated in 5 of 11 patients with an unrepaired tear of the anterior cruciate ligament. Clinical results at ten years were good in 46 cases (82%) with complete recovery and return to physical and sportive activity, good in 3% with mild symptoms, and failure in 6% consisting of meniscus re-rupture. A radial tear was found in all re-ruptures. Fairbank0 s changes were found in three cases. Conclusions: The overall results with use of the outside-in technique are comparable with those reported with use of the inside-out method. Such implants are not yet reliable, easy to use, and share biomechanical properties equivalent to the vertical-mattress suture technique. Patients with concomitant tears of the medial meniscus and the anterior cruciate ligament should have combined meniscal repair and reconstruction of the anterior cruciate ligament. In this setting, it may be advisable to use multiple permanent sutures, and the patient must be counseled regarding the higher rate of failure with this approach. Repairs of the lateral meniscus have a higher rate of success, and repair of the lateral meniscus should be considered even in the presence of injury of the anterior cruciate ligament. Loss of meniscus results in abnormal load transmission across the knee and may lead to degenerative joint disease. Preservation of meniscal tissue is therefore important. The most successful repairs occur in younger patients who have an acute, vertical tear in the vascular portion of the meniscus. Currently, arthroscopic meniscal repair procedures include the inside-out, the outside-in and the all-inside technique. Vertical suture techniques are superior to horizontally placed sutures. From a biomechanical point of view, 2-0 sutures can be recommended for suture repair.

P15-1173 Good outcome after meniscal repair using an all inside suturing system in combination with high-frequency bio-stimulation Leal-Blanquet J.1, Monllau J.C.2, Pelfort X.3, Puig-Verdie L.1, Tey M.3, Voss C.1, Pavlovich I.4 1 IMAS Hospitals (Mar-Esperanc¸a), Orthopaedic Surgery, Barcelona, Spain, 2Hospital de la Santa Creu i Sant Pau, Department of Orthopaedic and Traumatology, Barcelona, Spain, 3Instituto Universitario Dexeus, Arthroscopic Unit, Barcelona, Spain, 4Orthopedics & Sports Medicine Institute, Hermosillo, Mexico Objectives: The latest generation of all-inside devices combines the use of bio-absorbable materials and stitches in such a way that its final pullout strength is closer to the conventional suture. As the meniscus heals poorly, stimulation of revascularization at the lesion site is one of the most important factors in order to obtain a good meniscal repair outcome. The present study evaluates the clinical and functional results of an arthroscopic meniscal suture done with the FasT-Fix system in combination with healing stimulation of the lesion with radio-frequency (RF). The hypothesis was that this combination would be able to arrive at a healing ratio similar to standard suturing systems and stimulation procedures.

123

S240 Methods: A prospective study was carried out on a series 43 patients with an injury to the meniscus that was operated on over a 3-year period. The series consisted of 34 males (72.7%) and 9 females (27.3%) with an average age of 29.8 years. The inclusion criterion was a total rupture of the meniscal tissue greater than 10mm in length in either the red-white or white-white zone. All the patients had had a meniscal repair using the FasT-Fix suturing system with RF applied at the margins of the lesion. All patients were evaluated with Lysholm and IKDC scores. To analyze the extent of improvement of the Lysholm score as well as the IKDC, the obtained data were tested using the Student T-test for paired data. The level of significance was set at p\ .05. MRI and X-ray were performed in patients after surgery in order to evaluate the rupture healing and the evolution to osteoarthritis. Results: As two of the patients were lost in the follow up, 41 were finally evaluated. Those forty-one patients were analysed along an average of 26.1 months (from 18 to 47 months). The subjective IKDC and the Lysholm scores improved. IKDC averaged 59 preoperatively and 92 postoperatively (p\.009) and the Lysholm score averaged was 91.96 points postop while it was 60.9 preoperatively (p\ .008). Four repairs (9.8%) were considered to be failures at the time of evaluation; 2 in the ACL repair group and 2 in the rest of the series were meniscal repair alone was done (p\ .832). No changes were seen in the Rosenberg view at the last follow-up according to the Ahlback criteria when compared to the preoperative radiological exam. With respect to the MRI’s, the repaired site can always be recognized by an altered signal. Conclusions: Most of the current literature proposes combining the meniscal suture with a mechanical synovial abrasion. The results of the current work suggest that the combination of FasT-fix all-inside suture and RF stimulation leads to the same healing rate as it does in combination with mechanical abrasion. Furthermore, mid-term results of this type of meniscal repairs are at least as satisfactory as current methods. There are few complications with it and its ease of use makes it a worthwhile method even though long-term and comparative studies are still needed.

P15-1179 Asymmetric bilateral discoid medial menisci with posterior cyst formation in one knee Lee J.H.1, Lim Y.J.2, Bae H.K.1 1 Chonbuk National University School of Medicine, Chonbuk Nati, Orthopedic Surgery, Jeonju, Korea, Republic of, 2Saint Carollo Hospital, Orthopedic Surgery, Sunchon, Republic of Korea Objectives: Discoid medial meniscus is very rare condition of the knee. Even less frequent is the presence of bilateral medial discoid menisci, in fact only 18 cases are described in literature. We report one patient who presented a bilateral medial discoid. Methods: In march, 2009, a 22-year-old male addressed us complaining he had been suffering with medial knee pain of the right knee for 1months. The symptoms began with playing basket ball that was light pain associated with swelling on affected knee and limiting the patient’s sports and general daily activities. On physical examination, there was some discomfort on medial aspect of the knee with extension and 10 degree limitation of extension was checked without any mechanical abnormality. Medial joint line tenderness was existed. McMurray and squatting test results were positive. No abnormality was detected on radiographs. There were no ligamentous instability, patellofemoral abnormality on his affected knee. MRI of the right knee showed a discoid medial meniscus with a horizontal cleavage tear in mid-body to posterior horn and lateral meniscus was normal. (Fig 1A,1B) The patient underwent arthroscopy of his right knee, revealing complete medial discoid meniscus with horizontal cleavage tear in mid-body and posterior horn (Fig 2A,2B) and a pathologic medial plica syndrome. We had performed saucerization with partial meniscectomy (Fig 3). Upon finding a pathologic thick medial patella plica that were impinging between patella and medial femoral condyle during extension, we proceeded to remove it. Even if he had no complaints on his left knee and physical examination didn’t demonstrate abnormal signs, we evaluated MRI of the opposite knee due to possibility of

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 combined anomalies on the left knee. MRI showed an incomplete medial discoid meniscus with posterior cyst formation of the medial meniscus with high signal intensity on a T2 weighted image. Results: We report one patient who presented a bilateral medial discoid. One knee had an complete type of medial discoid with horizontal cleavage tear confirmed by both magnetic resonance imaging. Conclusions: The use of MRI was helpful for the diagnosis of the unaffected knee as well as the affected knee. If there are subclinical incomplete discoid medial meniscus with pathologies in the unaffected knee, the patients may be recommended about activity modification to prevent possible injury.

P15-1211 A new arthroscopically assisted all-inside technique for lateral meniscal allograft transplantation: a cadaveric study Wajsfisz A.1, Meyer A.1, Hardy P.1 1 Ambroise Pare´ Hospital, West Paris University, Boulogne, France Objectives: The efficacy of meniscal transplantation was demonstrated at intermediate-term studies with high levels of patient satisfaction. The meniscus can be implanted with the use of either an open or an arthroscopically assisted technique. Although results of both methods have been reported to be similar, arthroscopic techniques are routinely used. Whatever the technique, the meniscus should be fixed in an anatomical position. Bone anchoring of the anterior and posterior horns is essential regarding meniscal function. Although it is technically easier to secure the graft to soft tissue alone, research has indicated that load transmission is superior when the graft is secured to the bone. The most common methods use either bone plugs or bone bridge. These techniques are technically highly demanding. Besides this difficulty, bone tunnel have their specific morbidity. We performed a cadaveric study on feasibility of a new arthroscopic lateral meniscal allograft technique with bone fixation and without bone tunnel. Methods: Twelve cadaveric specimens were used for this study. For each specimen, one knee was dedicated to receive the graft, the other one was used to assess meniscal positioning and as a donor site for another specimen. The technique was performed with cadaver associated 2 by 2, with similar sizing in order to have the best adapted lateral meniscal allograft. The donor knee was disjointed to take a picture of the lateral meniscus positioning before it was removed and implanted on the other cadaver. Second, an arthroscopic lateral meniscectomy was performed on every knees elected to receive a graft.. The procedure was an all-inside arthroscopically assisted technique using three approaches: antero-medial, antero-lateral and an accessory lateral portal. The meniscal horns were fixed on the tibial plateau by an Arthrex Swivel-LockTM (Naples Florida) suture anchor. The meniscal wall was fixed to the articular capsule by all-inside (Meniscal Cinch TM Arthrex Florida) and out-in techniques. The knee which received the meniscal allograft was disjoint too. A picture was taken of the lateral meniscal allograft positioning. A statistical analysis was performed to compare the meniscal allograft to the native meniscus. Results: Eleven implantations were performed. One failed because of a too short meniscal allograft. On a standard tibial plateau (75mm x 40 mm), the meniscal allograft posterior horn was measured at a mean distance of 4,6 mm to the native posterior horn. This difference was not significant (p[0, 05) Conclusions: This original technique doesn’t use any bone plug nor bone tunnels. In case of simultaneous ligament reconstruction, this technique doesn’t interfere with ligament fixation. This procedure is easier than the others, but needs a learning curve. This original technique of meniscal allograft implantation is feasible, with poor morbidity, reproducible and allows a satisfactory meniscal posterior horn positioning.

P15-1229 Meniscus suture - 14 years overview Pasa L.1 1 Trauma Hospital Brno, Dpt. of Traumatol. of Medical Fac Masaryk University Brno, Brno, Czech Republic

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Objectives: Authors present their experiences and results of meniscus preserve operations in overview of 14 years. Methods: Chondromalacy and pain begins usually in some years after subtotal menisectomy in injured compartment. Suture of injured meniscus preserve preinjured conditions in injured compartment and could prevent overloading of cartilage, which comes after meniscectomy. From January 1994 to December 2007 authors sutured 563 injured meniscus in time 2-23 days after injury, 425 medial and 138 lateral. Contemporary lesions: ACL rupture in 241 patients, PCL rpt in 5 patients, second meniscus injury 36 patients, cartilage damage in 258 patients. FOR Meniscus suture were used inside-out,out-side in and all-inside technique, usually PDS stitches No.0-1 and only for posterior horn were used T-fix or Ultra Fast fix. In 62 patients were contemporary ACL reconstruction, in 154 patients were ACL reconstruction in second operation, after regress of acute inflammations of synovial tissue, 25 patients remain without ACL reconstruction because no clinical instability of the knee. Results: All patients were controlled clinically after 1 year after operation. Criteria for clinical success included absence of joint-line tenderness, locking, swelling, and a negative McMurray test. Clinical evaluation also included the Tegner and Lysholm knee scores, Only 12 patients were symptoms for meniscus pathology. In these patients were provided control arthroscopy, in 7 patients menscectomy and in 5 patients resuture with good clinical results. Only 398 patients were possible to controll until June 2009. Lysholm score: 286 excellent, 102 good, 10 satisfied, IKDC score: all in A and B group, Tegner score: 7,6. From these group 24 patients had new meniscus injury in time 2- 8 years after meniscus reinsertion and arthroscopy for acute haemathros. In 21 patients were cartilage without damage and 3 patients with chondromalacy until gr. II Outerbridge. Conclusions: Suture of acute meniscus tears is a method by which is possible to preserve preinjured conditions in weight bearing compartment. Synovial proliferation after acute injury improve healing of the injured and sutured meniscus. Cartilage changes are minimal after meniscus preservation in compare to patients after meniscetomy. Level of Evidence: Level IV, Therapeutic Study, Cases Series (no, or historical, control group).

P15-1277 Phenotypic gradient from the avascular to the vascular zone of the meniscus. New insights in the use of the term ‘‘fibrochondrocytes’’ referred to meniscus cells Deponti D.1, De Palma C.2, Pozzi A.3, Ballis R.4, Fraschini G.4, Peretti G.1 1 University of Milan, Faculty of Exercise Sciences, Milan, Italy, 2 San Raffaele Scientific Institute, Milan, Italy, 3University of Milan, Residency Program in Orthopaedic and Traumatology, Milan, Italy, 4 San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy Objectives: The meniscus plays an important role in the biomechanics of the knee joint: it has a function in load bearing and transmission, shock absorption, joint stability and lubrication. This tissue has a poor healing potential, partly due to the absence of vasculature: blood vessels are present only in the outer 10-30% of the meniscal body. Normal meniscus is composed of extracellular matrix (ECM), water (72%) and cells; the ECM consists of collagen, mainly type I (90%), and glycoaminoglycan. Due to the large amount of collagen type I, the meniscus has properties of fibrous and cartilaginous tissue. The term fibrochondrocytes has been introduced to identify the typical characteristics of the meniscus cells, but a clear description of the phenotype of these cells is still missing. This work was aimed to study the characteristics of the meniscus cells by focusing on three areas of the meniscus, the inner avascular zone, the intermediate, and the external vascular zone. Methods: Meniscus cells proliferation and differentiation were compared in these three different areas by FACS analysis; moreover, the expression

S241 of cartilage specific genes was confirmed by real time PCR; additionally, cell morphology was analyzed both in the native meniscus by histology and by microscopy analysis after enzymatic isolation from the tissue. Meniscus cells were compared to isolated articular chondrocytes and tendon fibroblasts. Results: The data show that the inner region of the meniscus is composed of cells that are similar to articular chondrocytes in terms of morphology (round shape) and proliferation rate, but they express lower amount of cartilage specific genes, such as collagen type II and aggrecan; the intermediate and external areas are composed of cells that showed a gradient in morphology and cartilage specific gene expression compared to the cells of the inner area; however, interestingly, they show a lower proliferation rate. Conclusions: These results suggest the presence of an ‘‘intermediate phenotype’’ located in the middle and external meniscal areas having morphological and functional characteristics intermediate between the fibrochondrocytes of the inner region and the tendon fibroblasts; these evidences could lead to new options in the choice of the optimal cell source for meniscus tissue engineering.

P15-1303 A basic arthroscopy course based on motor skill training Unalan P.1, Akan K.2, Orhun H.3, Akgun U.4, Poyanli O.2, Baykan A.5, Yavuz Y.1, Beyzadeoglu T.6, Nuran R.5, Kocaoglu B.7, Topsakal N.8, Akman M.1, Karahan M.9 1 Marmara University Faculty of Medicine, Istanbul, Turkey, 2Ministry of Health Go¨ztepe Education and Research Hospital, Istanbul, Turkey, 3 Ministry of Health Kartal Education and Research Hospital, Istanbul, Turkey, 4Acibadem University Faculty of Medicine, Orthopedics and Traumatolgy, Istanbul, Turkey, 5Acibadem Hospital, Istanbul, Turkey, 6 Yeditepe University, School of Medicine, Department of Orthopedics and Traumatology, Istanbul, Turkey, 7Acibadem University Faculty of Medicine, Orthopaedic Surgery, Istanbul, Turkey, 8Marmara University Physical Education and Sport Faculty, Istanbul, Turkey, 9Marmara University, Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey Objectives: To introduce, evaluate and suggest a basic arthroscopy training course with interactive methods as using bovine knees, joint models but mainly focusing on practices with motor skill learning devices. Methods: This study presents the level one results (participants’ feedbacks) and a level two (a summative exam of the arthroscopic procedure) of a skill training program which is based on ‘‘motor skill training’’ theory and using ‘‘bovine knees’’ to simulate a specific task as arthroscopy both in education and evaluation. During this evaluation of the task, a four dimensional checklist which was prepared previously by the consensus of 3 experts was used and a global rating of each rater was added at the end of the checklist. In this 2 - day course, small group teaching sessions are integrated with active and experiential learning methods, coupled by short presentations of the teachers, demonstrations using joint models, audio - visual teaching material and mainly continuous feed back of the teachers. Participants’ expectations and post - course feedbacks were obtained using a brief questionnaire mostly with open ended questions and a self-rating of ‘‘own basic diagnostic arthroscopy skill’’ between 0-10 was completed by them both in pre and postcourse intervals to establish a student centered learning environment.. Overall rating of the course was asked on a 1 5 (poor - excellent) Likert scale. Results: Feedbacks of a total of 64 participants who took the course in the last 5 cycles, constitutes a 100% return. The overall rating of the course was 4. 36 ± 0. 47on the 1 - 5 scale. Most of the participants (90%) mentioned that the course was on the target according to their expectations. The most beneficial activity was accepted to be; arthroscopic procedure applications on bovine knee, to repeat the manipulations under the control of the trainers, to get feed back immediately and to be able to

123

S242 correct simultaneously (35/64). Pre and post-course self ratings of the participants about their ability to perform arthroscopy alone was significantly improved (4.2 vs. 7.7 p = 0.000, paired samples T-test). All participants were found to be competent in the evaluation of their diagnostic arthroscopy skills on bovine knees. Conclusions: To plan a skill teaching course on the needs of the trainees, focused on basic motor skill training exercises and using bovine knee as a simulator is a safe, inexpensive, humanistic and replicable method that constitutes basic arthroscopic skills learning prior to patient encounter.

P15-1328 Clinical efficacy of a novel, synthetic meniscus scaffold for treating irreparable, partial meniscal tissue loss Verdonk R.1, Verdonk P.1, Cugat R.2, Laprell H.1, Bellemans J.3, Djian P.4, Paessler H.H.5, Beaufils P.6, Colombet P.7, Neyret P.8 1 University Hospital Ghent, Dept. Orthopaedic Surgery and Traumatology, Gent, Belgium, 2Clinica del Pilar, Barcelona, Spain, 3KU Leuven, Leuven, Belgium, 4Paris, France, 5ATOS Clinic, Center of Knee and Foot Surgery,Sports Trauma, Heidelberg, Germany, 6Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France, 7Centre de Chirurgie Orthope´dique et Sportive de, Bordeaux-Me´rignac, Bordeaux Me´rignac, France, 8Hopital Croix-Rousse Centre Livet, Chirurgie Orthopedique, Lyon, France Objectives: A novel, synthetic, slowly degradable, polyurethane scaffold developed for irreparable partial meniscal tear treatment was evaluated in a prospective multicenter, non randomised study. The device is intended as a scaffold for meniscal tissue ingrowth in subjects with an irreparable partial medial or lateral meniscal tear or meniscal tissue loss. Efficacy of the device was assessed by changes over time at 3, 6, 12 and 24 months, in subjects’ perceived pain, functionality and quality of life as compared to baseline. Methods: Subjects with an irreparable medial or lateral meniscal tear, intact rim, presence of both horns and stable knees with normal alignment received the scaffold during standard a arthroscopic procedure following partial meniscectomy. Subsequent to surgery subjects were required to undergo a specific rehabilitation program designed to provide optimum conditions for healing. Efficacy was assessed using the Visual Analog Scale (VAS), Knee and Osteoarthritis Outcome Score (KOOS), International Knee Documentation Committee (IKDC) and Lysholm scores performed at baseline, 3, 6, 12 and 24 months. Results: The 52 subjects entered in the study were predominantly Caucasian (98%), male (75%) and had a mean age of 30.8 ±9.4 years. Surgery involved 34 medial and 18 lateral menisci, and the mean length of the defect was 47.1 ± 10.0 mm. The majority of subjects had undergone either one (65%) or two (23%) prior surgeries on the involved meniscus. Of these subjects 47 completed 12 months follow-up continuing to the 24months study endpoint, 3 were lost to follow-up and 2 discontinued due to a serious adverse event unrelated to the investigational device. Statistically significant improvements in IKDC scores, Lysholm scores and all KOOS subscales were reported at 12 months post implantation (p\0.0001). In addition, clinically relevant reductions in knee pain on VAS were reported at 3, 6 and 12 months post implantation compared to baseline. Improvements in all KOOS subscale scores were reported at 12 months compared to baseline, with the greatest magnitude of change reported for sports/recreation and Quality of Life. Preliminary data obtained at 24 months (n=9) post implantation corroborate the 12 months data. The 24 months data for all subjects will be presented. No serious device related adverse events have been reported to date, in one case no integration of the scaffold to the rim was observed. The causality is unknown. Conclusions: Following implantation of the novel, synthetic scaffold, statistically and clinically significant improvements for all clinical outcome scores including pain relief were achieved at 12 months post implantation. It is anticipated that data obtained at 24 months follow-up will further corroborate these findings.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P15-1343 Anatomical study on lateral meniscal transplantation: an original arthroscopic technique developed on 12 cadaver specimens Panarella L.1, Be´lot N.2, Boisrenoult P.3, Beaufils P.3 1 University of Rome Tor Vergata, Department of Orthopaedic Surgery, Rome, Italy, 2University of Rennes, Department of Orthopaedic Surgery, Rome, Italy, 3Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France Objectives: Meniscal transplantation is still discussed as a valuable technique in the treatment of degenerative osteoarthritis of the knee. However it seems to be effective in the post-meniscectomy syndrome, especially in patients with a normal knee alignment. The objective of this anatomical study was to verify the feasibility of an all-arthroscopic lateral meniscal transplantation. Methods: An original arthroscopic technique was developed on 12 cadaver specimens and was performed through the two standard antero-medial and antero-lateral approaches. The fixation was secured through two bone plugs into two orthogonal tunnels. Six fresh right knees and six fresh left knees were tested. We carefully preserved meniscal insertions on the anterior and posterior horns during the preparation of the transplant from the meniscal bank. Through the arthroscopic portals, the graft was secured on the two tibial tunnels and fixed all over the menisco-synovial junction. At the end of the procedure, all the specimens were tested for anatomical positioning and mechanical resistance of the graft after dissection of the knee joint. Results: Progressing with the technique, the overall position of the lateral meniscus on the knee joint was found to be more and more adequate. The mechanical resistance to pull-out improved according to bone plugs positioning, as well. Conclusions: Bone plugs quality, correct meniscal sizing and careful dissection of the menisco-synovial junction are mandatory to ensure a good preparation of the meniscal transplant. Despite a large number of papers, meniscal transplantation can not be defined a standard treatment. We believe that this technique could be further evaluated to perform lateral meniscal allograft with a less invasive procedure.

P15-1396 Comparison of the accuracy of clinical examination and arthroscopic findings in medial meniscus injuries Trc T.1, Kos P.1, Chudacek V.2, Nemejc M.1, Havlas V.1, Kuzilek J.2 1 University Hospital Motol, Department of Orthopaedics and Traumatology, Prague, Czech Republic, 2Czech Technical University in Prague, Department of Cybernetics, Prague, Czech Republic Objectives: The menisci plays an important role in the biomechanics of the knee joint. The meniscus acts in load bearing and transmission, shock absobtion, joint stability and congruity. Meniscal impairment can lead to early degenerative arthritis. The fast developing field of diagnostic technology may sometimes obscure the importance of clinical examination. Our study was designed to assess the value of clinical examination in the diagnosis of meniscal injuries in comparison with arthroscopic findings. Methods: 138 patients with clinical suspicion of medial meniscus injury were included in our study (mean age 45 years (18-73); 48.3% men, 51.7% women; 50.5% right, 49.5% left knees). 12 clinical tests and symptoms were evaluated by 7 experienced orthopaedic surgeons (medial joint pain, swelling, hyperextension, Mc Murray, Steinmann I., Steinmann II., Apley, Childress, Payer, Boehler-Kroemer, Bragard, Cabot test). The subsequent arthroscopic surgery of the knee joint was performed by 6 experienced surgeons within 3 days in all cases - there was no new injury of the knee observed in the meantime. Correlation coefficient was assessed using the Spearman0 s rank correlation coefficient computation, the specificity and sensitivity for single clinical tests were determined. Additionally testing of sensitivity of multiple tests combinations had been undertaken. All possible permutation of all different clinical tests combination was computed. Results: We sorted clinical tests and symptoms by sensitivity - Childress 88.5%, medial joint pain 78.2%, hyperextension 75%, Payer 71.6%, Mc Murray 61.6%, Apley 57.7%, Steinmann I. 53.6%, Boehler-Kroemer 52.9%, Steinmann II. 49.3%, Bragard 44.3%, swelling 32.8%, Cabot test

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 30.9%. Significant specificity was observed in following tests - Cabot 78.5%, swelling 74.2%, Bragard 62.3%, Steinmann I. 58.9%, Steinmann II. 58%, Boehler-Kroemer 54.9% and Apley test 51.6%. There is no significant correlation between the tests and postoperative diagnosis when focusing on presence of medial meniscus injury. While testing sensitivity of all multiple tests combinations, all combinations of more than 3 tests yielded sensitivity of less then 40% - which was considered unsatisfactory and therefore only combination of 2 and 3 tests was investigated further. Adding medial joint pain symptom to Childress test resulted in slight decline of sensitivity but great increase of specificity 74.4% and 34.4% respectively. Combination of Childress and hyperextension yielded sensitivity of 66.7% and specificity of 36.6%. When combining three parametres namely medial joint pain, Boehler-Kroemer and Childress test improved specificity further to 75.6% but lowered sensitivity to 51.8%. Conclusions: According to our results the importance of skilled physical examination continues to be essential. We confirmed high reliability of Childress test, hyperextension test, Payer and McMurray tests. Medial joint pain is highly sensitive but with little specificity for medial meniscus injury. We determined that suitable combination of more above stated tests and symptoms could be a standard of clinical knee examination with suspicion of medial meniscus injury. Using more then three tests at once does not bring any improvement in the decision making, nevertheless combinations of two tests usually lead to significantly better specificity while keeping sensitivity relatively high.

P15-1409 A polycarbonate-urethane meniscal transplantation for medial meniscus insufficiency: preliminary clinical results Zorzi C.1, Condello V.1, Linder-Ganz E.2, Shterling A.3, Zur G.4, Shani J.3 1 Sacro Cuore Don Calabria Hospital, Orthopaedic Department, Negrar, Italy, 2R&D Center, Active Implants Corporation, Netanya, Israel, 3Active Implants Corporation, Netanya, Israel, 4Active Implants Corporation, R&D Center, Netanya, Israel Objectives: The goal of this study is to demonstrate that a ‘‘floating’’ artificial meniscal substitution made of polycarbonate-urethane (PCU) embedded with polyethylene reinforcement fibers (‘‘Dyneema’’, DSM), designed on the geometry of the articulating surfaces of the femur and tibia, could improve joint congruency and cartilage protection. The medial meniscus plays an important role in the knee joint. Meniscus dysfunction due to tear is a common knee injury which leads to degenerative arthritis, attributed primarily to the changes in knee load distribution. Clearly, there is a substantial need to protect the articular cartilage by either repairing or replacing the menisci. A PolycarbonateUrethane (PCU) meniscal implant is proposed as a solution for restoring the function of the missing meniscus and for the reduction of pain, through improved tibial and femoral pressure distribution. Previous biomechanical studies in a sheep model have shown that PCU meniscus transplantation may protect the articular cartilage from further degenerative changes. Methods: Since May 2008 we implanted 8 meniscal artificial implants (Nusurface) in 8 patients suffering for medial meniscal insufficiency. We had very restricted inclusion criteria: no axial deviation or ligament insufficiency, no previous surgeries on the affected limb, no cartilage degenerative changes more than grade 3 according to Outerbridge classification. The surgical technique includes an arthroscopic total medial meniscectomy, joint debridement when necessary and a mini arthrotomy for meniscal substitution insertion. Since the device does not have any attachment to the intra-articular tissues, the instrumentation is very simple: an inserter and an extractor are the only two tools needed for surgery. Preoperatively an MRI with a specific protocol is performed in order to measure the size of the implants. Previous MRI studies on cadavers have shown that five sizes of the implant are able to cover 95% of the population. Each patient was evaluated with KOOS, TEGNER and Lysholm score pre operatively, at 6, 12 and 18 months. Results: Five patients had an improvement of clinical results according to the KOOS schedule. Two of them were able to return to pre-operative activity level The first follow up at six months showed no improvement of

S243 clinical results but at one year follow-up subjective and objective scores were better than before surgery. Three out of 8 patients needed revision surgery for persisting pain: they never improved and had an extension deficit since the first week with antero-medial impingement of the implant. One of the three had a posterior luxation of the device after a knee sprain. Conclusions: This study shows only preliminary results about a new floating device that substitutes meniscal function in patient candidate for unikee replacements. At this stage we can’t give definitive conclusions on its application but further studies are going on in order to develop an effective device that is able to release overload from a meniscectomized knee and give patients more choice before knee replacement using a simple non aggressive solution that can be easily removed without any joint or peripheral tissue damage.

Knee: cartilage P16-138 Arthroscopic surgery for osteoarthritis of the knee: clinical outcome at 2 and 6 years follow-up Widuchowski W.1, Lukasik P.1, Koczy B.1, Kwiatkowski G.1, Szczesniak M.1, Widuchowski J.1 1 District Hospital of Orthopedics and Trauma Surgery, Depart. of Knee Surgery, Arthroscopy and Sports Trauma, Piekary Slaskie, Poland Objectives: The role of arthroscopic surgery in arthritic knee remains controversial and unclear. The purpose of this study was to evaluate clinical results after the use of different arthroscopic methods in patients with osteoarthritic knee. Methods: We performed an analysis of 131 patients who underwent arthroscopic surgery for the treatment of the Kellgren-Lawrence grade 1 and 2 (K-L1, K-L2) knee osteoarthritis. Average patient age was 51 (range: 37-75 years). The outcomes were reported at 2 and 6 years followup using the WOMAC score and the Lysholm score. Results: Arthroscopic debridement, lavage and loose body removal carried out in patients with K-L1 at 2 years follow-up resulted in not significant (p[0.05) improvement in WOMAC and Lysholm scores. At 6 years follow-up no improvement was reported. In K-L2 group at 2 years follow-up no improvement was noted and at 6 years follow-up we recorded significant deterioration in both scores (p\0.05). There was significant (p\0.05) deterioration according to WOMAC and Lysholm scores in both groups regardless of follow-up period in cases where microfracture was performed. Forty patients (30 of K-L2 group and 10 of K-L1 group) had further surgeries; 15 had total knee replacement, seven had high tibial osteotomy and five had unicondylar knee arthroplasty. Conclusions: Our study confirms that inappropriate selection of patients and the inappropriate use of certain techniques might increase the number of bad and not-satisfactory results in treatment of the osteoarthritic knees. However, this study supports also the contention that there is a group of patients with mild knee osteoarthritis that might benefit from adequate arthroscopic surgery.

P16-186 The operative treatment for osteonecrosis of the knee - the indication and clinical results of total knee arthroplasty, uncompartmental knee arthroplasty, high tibial osteotomy, and autologous osteochondral transplantation Kubo M.1, Matsusue Y.1 1 Shiga University of Medical Science, Department of Orthopaedic Surgery, Otsu, Japan Objectives: The indication of operative treatment for osteonecrosis of the knee (ON) is controvertial. The purpose is to report our indication of operative treatment for ON and its clinical results. Methods: Fifty patients (fifty-three knees) were operatively treated with diagnosis of ON by TKA, UKA, HTO, HTO, autologous osteochondral transplantation (OAT), or HTO ? OAT by our indication. Average age

123

S244 was 62.5 years (23-82). They were evaluated radiologically by Koshino’s classification (stage), Lotke’s scaling (size) and femoro-tibial angle (FTA). Thirty-four knees followed for more than one year were evaluated clinically by Lysholm knee scoring scale and range of motion (ROM). The average follow-up was 35.8 months (14-87). Results: There were one knee in stage I, nine knees in stage II, twenty-two knees in stage III, and twenty-one knees in stage IV. The size of lesion was 38.9% (20.5-54.5). FTA was changed from 178.4 to 170.4. At their final follow-up, average Lysholm knee score was increased from 54.1 to 94.2. ROM was improved from -4.9-135.6 to -0.9-139.7. Conclusions: We selected UKA for senile patients with low lebel of activity, however we performed TKA for the patients who were not compatible for indication of UKA (mono-compartmental lesion, ligament stability, good alignment and ROM). We selected HTO for young patients and senile patients with high lebel of activity, moreover we added OAT when the necrotic site was localized and unstable. For the patients whose lesion was not on the weight bearing-line, we performed only OAT. Clinical results of the patients operatively treated by these our indication was good and these indication is thought to be recommended.

P16-230 Weight-bearing axis changes after osteochondral autogenous transfer for the knee joint Kobayashi M.1, Nakamura S.1, Arai R.1, Nishitani K.1, Shirai T.1, Satake T.1, Kuroki H.2, Nakagawa Y.1, Nakamura T.1 1 Kyoto University, Orthopaedic Surgery, Kyoto, Japan, 2 Kyoto University, Human Health Sciences, Graduate School of Medicine, Kyoto, Japan Objectives: There remain some debates on osteochondral autogenous transfer (OAT; mosaicplasty) for an early stage of osteoarthritis in knee joints (OA). The objectives of this study are to investigate effectiveness and validation of OAT for the early OA. Methods: Consecutive series of eighteen cases (eighteen knees) were retrospectively reviewed. Six males, and 12 females, and average age at the surgery was 57.8±12.7 years old. Inclusion criteria was those patients who had isolated chondral lesion (ICRS grade-3 and -4) at medial femoral condyle, whose standing lateral femoro-tibial angle (FTA) was less than 180 degrees, and whose radiographic OA grading was less than 3 (Kellgren-Lawrence (K/L)). Several numbers of osteochondral plugs were transferred from lateral and medial edges of femoral groove. IKDC subjective score, FTA, and K/L grading were assessed pre-operatively and at the final follow-up (18.5±9.9 months in average). Results: Post-operative IKDC score improved comparing to pre-operative one. FTA significantly increased from 178±2.2 degrees pre-operatively to 179.5±3.0 degrees post-operatively (P=0.0030). In some cases where the pre-operative FTA was close to 180 degrees, the post-operative FTA increased being over 180 degrees. And there was moderate negative correlation between pre-operative IKDC score and the increase in FTA. Regarding K/L grading, nine cases did not change, while 9 cases deteriorated. Conclusions: The present study showed favorable short-term clinical outcomes after OAT for the early knee OA where FTA was less than 180 degrees, although OAT cannot improve lower limb alignment. Further long-term clinical study is essential to evaluate the influence of increased FTA on the clinical outcomes.

P16-254 Viscosupplementation injections augmented with corticosteroid for knee osteoarthritis: patient expectations and clinical outcomes Briggs K.1, Steadman J.R.2, Matheny L.1, Rodkey W.3 1 Steadman Research Foundation, Clinical Research, Vail, United States, 2 Steadman Sports Medicine Clinic, Knee Surgery, Vail, United States, 3 Steadman Research Foundation, Chief Scientific Officer, Vail, United States

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Objectives: The use of intraarticular viscosupplementation has gained wide acceptance; however, clinical outcomes reports have been variable. The purpose of this prospective cohort study was to document patient expectations and outcomes following treatment with Hylan G-F 20 injection in a series of three injections in which corticosteroid was used with the initial injection. Methods: Patients with forty-seven knees (42-80 years of age) with a diagnosis of knee osteoarthritis completed a self-administered questionnaire before the injection series, and at 3, 6, 12 weeks and 6 months after the final (third) injection. The questionnaire included WOMAC score, Lysholm score, SF-12, and a patient satisfaction with outcomes question (10-point scale; 0=unsatisfied, 10=highly satisfied). Prior to the injections, patients also completed a validated expectation questionnaire and activity level questions. Patients were excluded if they had undergone surgery within the past 6 months. Results: The most important expectations for patients in this group were to have confidence in their knee, avoid future degeneration of their knee, and improve ability to maintain general health. There were no adverse events related to the viscosupplementation injections. Patients’ WOMAC pain subscale improved from pre-injection to 6 months post injection (p=0.003) as did the overall WOMAC score (p=0.038). Function, measured by Lysholm score, significantly increased from pre-injection to post-injection (40 to 62, p=0.001). SF-12 physical component score also was significantly improved. Conclusions: Our results showed significant improvement in pain relief and function following the viscosupplementation injection series. These improvements were maintained at 6 months. A combination of Hylan G-F 20 and corticosteroid was a safe and effective treatment option in this population of older patients suffering from osteoarthritis who desired to return to an active lifestyle while avoiding surgical treatment.

P16-291 Combined overexpression of human IGF-I and human FGF-2 enhances proteoglycan and DNA contents in an osteochondral defect model in vivo Orth P.1, Kaul G.1, Menger M.D.2, Kohn D.-M.1, Cucchiarini M.1, Madry H.1 1 Universita¨tsklinikum des Saarlandes, Klinik fu¨r Orthopa¨die und Orthopa¨dische Chirurgie, Homburg/Saar, Germany, 2Universita¨tsklinikum des Saarlandes, Institut fu¨r Klinisch-Experimentelle Chirurgie, Homburg/ Saar, Germany Objectives: Separate overexpression of human insulin-like growth factor-I (IGF-I) or fibroblast growth factor-2 (FGF-2) in transfected chondrocytes was shown to improve cartilage repair following implantation in osteochondral defects in vivo. Here, we tested the hypothesis that administration of encapsulated fibroblasts genetically modified to overexpress a combination IGF-I/FGF-2 in such lesions enhances the proteoglycan and DNA contents and improves the histological parameters of the repair tissue in vivo. Methods: NIH 3T3 fibroblasts were transfected with plasmid vectors carrying the E. coli lacZ gene, a human IGF-I cDNA or a combination of human IGF-I and human FGF-2 cDNAs using FuGENE 6 and next encapsulated in alginate. In vivo, two cylindrical osteochondral defects were created in each patellar groove of 12 female rabbits (n = 48 defects). The resulting lacZ, IGF-I and IGF-I/FGF-2 spheres were press-fit into the defects. The right and left knees alternatively received lacZ or IGF-I spheres (n = 12) or lacZ or IGF-I/FGF-2 spheres (n = 12). At 3 weeks post operation, the proteoglycan and DNA contents in the retrieved repair tissue (n = 24) were detected by DMMB and DNA assays. Cartilage repair was assessed on safranin-O-stained sections using a histological grading system. This system ranges from 0 points (normal articular cartilage) to 31 points (no repair tissue) and comprises individual scores for filling of defect, integration, matrix staining, cell morphology, architecture of defect and surface and formation of subchondral bone and tidemark. Data are given as mean ± SD or mean with 95% confidence intervals. Results: At 3 weeks after in vivo transplantation, there were no macroscopic signs of adverse reactions of the grafts in any group. The

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 proteoglycan content of the repair tissue was of 4.9 ± 1.8 lg/mg dry weight in the IGF-I/FGF-2 defects, compared with those receiving the IGF-I spheres (3.2 ± 1.1 lg/mg dry weight; P\0.05) or lacZ spheres (3.0 ± 1.4 lg/mg dry weight; P [ 0.05). The DNA content was also significantly higher in defects receiving the IGF-I/FGF-2 spheres (2,860 ± 96 ng/mg dry weight) compared with the IGF-I and lacZ control groups (2,244 ± 102 and 2,762 ± 542 ng/mg dry weight, respectively; both P \ 0.05). In good agreement with the biochemical findings, combined IGF-I/FGF-2 gene transfer significantly improved average individual histological scores for integration, cell morphology, subchondral bone and tidemark formation. The average total score after 3 weeks in vivo was significantly improved for defects treated with IGF-I/FGF-2 (11.2, 8.3013.8; n = 6) compared with defects receiving lacZ spheres and IGF-I spheres alone (P \ 0.05). In contrast, gene transfer of IGF-I alone (15.5, 12.7-18.2; n = 6) improved the average total score without reaching statistical significance when compared with defects receiving lacZ spheres (17.9, 15.1-20.7; P [ 0.05; n = 12). Conclusions: The data indicate that combined overexpression of IGF-I and FGF-2 in genetically modified NIH 3T3 cells significantly enhances the proteoglycan and DNA contents in the repair tissue of osteochondral defects in vivo. Furthermore, the transplantation of IGF-I/FGF-2-transfected fibroblasts significantly improves the corresponding histological score in a magnitude that is larger than with IGF-I alone. Future studies need to evaluate the long-term properties of the repair tissue.

P16-317 Operative outcome of osteochondral grafts for knee joints diseases Nakagawa Y.1, Matsusue Y.2, Mukai S.1, Kobayashi M.3, Nakamura S.3, Tsubouchi N.1, Nakamura T.3 1 Kyoto Medical Center, Orthopaedic Surgery, Kyoto, Japan, 2Shiga University of Medical Science, Department of Orthopaedic Surgery, Otsu, Japan, 3Kyoto University, Orthopaedic Surgery, Kyoto, Japan Objectives: Since Matsusue first reported autologous osteochondral grafts of the knee in 1993, more than 10 years passed. There were many reports of good short-term outcome for the osteochondral grafts of the knee, but there were little reports of medium or long-term outcome of it. Therefore, we report the medium-term (more than 5 years) outcome of the osteochondral grafts for the knee joint diseases. Methods: Between July 1997 and March 2003, there were 30 cases 36 knees who had osteochondral grafts for their knee joints in our hospital, and had more than 5 years follow-up, comprising 13 men and 17 women, with the right knee being affected in 14 cases and the left knee in 22. The follow-up ratio was 71%, and the mean follow-up period was 80.1 months (range from 60 to 127 months). The mean age of the subjects at the time of the operation was 39.3-years (range from 13 to 76-years). We measured the operative age, basic diseases (osteonecrosis or cartilage injury), the area of the lesion, the follow-up period and the lesion site (femoral condyle or patellofemoral joint). As the clinical outcome, IKDC subjective score and IKDC objective score were used. The statistical analysis was performed, and we defined it as significant if P value was less than 0.05. Results: The mean area of the lesion was 424.8 mm2. There were 21 knees in osteonecrosis group and 15 knees in cartilage injury group. There were 25 knees in femoral condyle group and 11 knees in patellofemoral joint group. In IKDC objective score, all cases were abnormal or severely abnormal at preoperative period, and 33 knees (92%) were normal or nearly normal at follow-up period. In IKDC subjective score, the mean score was 46.2 points at preoperative period and 83.0 points at follow-up period. There was significant difference. In preoperative IKDC subjective score, osteonecrosis, femoral condyle and the larger areas of the lesion were the risk factors. In follow-up IKDC subjective and objective score, the longer follow-up period was the risk factor. Ten cases (33%) were able to sit up straight in Japanese style at preoperative period, but 23 cases (77%) were able to do it at follow-up period. All cases had no donor symptoms. There were some reports that the outcome of patellofemoral group was worse than the outcome of femoral condyle group, but in our study there

S245 were no significant difference in 2 groups. The operative age was not risk factor in our study, and the older patient is also able to expect the good clinical outcome. Conclusions: The medium-term (more than 5 years follow-up) clinical outcome of osteochondral grafts for knee joints diseases was good, and 92% of the patients were normal or nearly normal in IKDC objective score.

P16-352 New one-step arthroscopic technique for the treatment of osteochondral lesions of the knee with bone marrow derived cells Vannini F.1, Cavallo M.1, Ruffilli A.1, Grigolo B.2, Buda R.1, Giannini S.1 1 Istituto Ortopedico Rizzoli, Bologna University, VI Department of Orthopaedics and Traumatology, Bologna, Italy, 2Istituto Ortopedico Rizzoli, Laboratory of Immunology and Genetics, Bologna, Italy Objectives: Articular cartilage demonstrated to have limited potential to spontaneously heal. Various surgical options have been proposed in the past to achieve long-lasting hyaline cartilage, among all arthroscopic Autologous Chondrocyte Implantation (ACI) proved to be able to achieve good results even in the long term. Limits of ACI are the necessity of two surgeries and high costs. Aim of this study is to present a one-step arthroscopic technique using concentrate autologous bone marrow-derived cells, hyaluronic acid scaffold and Platelet Rich Fibrin for the treatment of osteochondral lesions of the knee (OLK). Methods: 25 patients (age 32.6±12.9 years) were treated between March 2006 and August 2007. 17 cases involved the medial femoral condyle, 6 cases the lateral and 2 cases affected both the medial and the lateral condyle. Patient evaluation included clinical IKDC score, X-Rays and MRI preoperatively and at different established follow-ups. The bone marrow-derived cells were harvested from posterior iliac crest and concentrated directly in the operatory room, loaded on a hyaluronic acid scaffold with Platelet Rich Fibrin and arthroscopically implanted in the same surgical session. One patient underwent a second arthroscopy with biopsy at 1 year followup. Results: Mean IKDC score before surgery was 33.7±14.6 and 91.2±9.8 (p\0.0005) at mean 31±5.1 months. Control MRI evidenced a regeneration of cartilaginous layer and subchondral bone. Histology performed on the bioptic sample of the regenerated tissue highlighted the expression of collagen type-II and proteoglycans. Conclusions: The presented technique demonstrated to be able to provide good clinical results and a regenerated tissue which closely approximates the characteristics of the original hyaline cartilage at 2 years follow-up. The use of bone marrow-derived cells permitted a one-step treatment of OLK, with marked reduction in procedure morbidity, duration and costs, relevant drawbacks of previous repair techniques.

P16-365 Safety and efficacy of a single injection of lightly cross-linked hyaluronic acid formulation for knee osteoarthritis Schmitz-Elvenich G.1, Hardung A.2 1 Eduardus-Krankenhaus, Rheumaorthopedics, Cologne, Germany, 2 Eduardus-Krankenhaus, Cologne, Germany Objectives: To determine the safety and efficacy of a single injection of lightly cross-linked hyaluronic acid formulation for treatment of osteoarthritis of the knee. Methods: Eighty patients received a single intra-articular injection of lightly cross-linked hyaluronic acid formulation (MONOVISC, 4mL) in the knee. Baseline pain score was assessed using the summed WOMAC Index in VAS pain scale format. Patients with a baseline pain score between 200 and 400 and Kellgren-Lawrence Grade of I, II or III were included in the study. Treatment efficacy was assessed to baseline pain score using the summed WOMAC Index, Patient Global Assessment and

123

S246 Investigator Global Assessment in VAS pain scale at 4, 8, 13, and 26 weeks after the treatment. Adverse events were recorded to assess treatment safety. Results: All patients (N=80) completed study visits and were included in the Intent-to-Treat (ITT) population for safety analysis. Seventy three patients (n=73) were included in Per-Protocol (PP) population for efficacy analysis, seven patients were excluded from the efficacy analysis due to protocol violations (missing visits). Demographics of Per-Protocol (PP) population: mean age - 62.3 years, mean BMI - 24.8 (SD = 2.6) kg/m2, 62% females, 38% males. On average, the significant improvement in the summed WOMAC pain score was observed at 4 weeks after the treatment and maintained through the end of the followup period (26 weeks). The average decrease in summed WOMAC pain score of 27mm (44%) from the baseline assessment sustained at 26 weeks after the treatment. At the end of the observation period (26 weeks) ninety-nine percent of the PP population demonstrated the improvement from a baseline WOMAC pain score The average Patient Global Assessment and Investigator Global Assessment scores were also improved by 37% and 33%, respectively. Two minor adverse events manifested by pain and minor swelling at the injection site for less than 48 hours were observed; no serious adverse events were reported. Conclusions: This study demonstrated the safety, efficacy and sustainable results of MONOVISC treatment for symptomatic reduction of pain associated with osteoarthritis of the knee. A single injection of MONOVISC resulted in improvement of summed WOMAC pain score to baseline in 99% of the PP population for at least 6 months.

P16-385 Malalignment and cartilage lesions in the patellofemoral joint treated with autologous chondrocyte implantation-long term results Vasiliadis H.1, Brittberg M.2, Lindahl A.1, Peterson L.3 1 Sahlgrenska Academy / Gothenburg University, Molecular Cell Biology and Regenerative Medicine, Gothenburg, Sweden, 2University Go¨teborg, Cartilage Research Unit/Dept. Orthopaed., Kungsbacka, Sweden, 3 University of Go¨teborg, Department of Orthopaedics, Go¨teborg, Sweden Objectives: The aim of our current study is to present the long term follow up of patients with cartilage lesions of the patellofemoral joint, treated with Autologous Chondrocyte Implantation (ACI) with the use of periosteum. Methods: 92 patients having patella or trochlea lesion participated in our study. Lysholm and Tegner-Wallgren questionnaires were filled 12.6 years in average after the surgery. They patients were asked whether they feel better, worse or unchanged compared to previous years and whether they would do the operation again. Complications or subsequent surgeries were also assessed. Results: Tegner-Wallgren score was 7.1, improved by 0.95 compared to preoperative values (p=0.01). Lysholm score was 68.1, improved by 9 points in average (p=0.3). Seventy two % of the patients were better or unchanged while 93% would do the operation again. Patients with no kissing lesions appeared to have a better prognosis. Patients with malalignment or instability that had a realignment procedure had comparable outcomes to the cases which did not need any additional surgery. Realignment procedures increased the incidence of serious complications but they were associated with decreased incidence of periosteal hypertrophy. No association was found between the age at the time of the ACI or the size per lesion and any of the clinical outcomes. Conclusions: It seems that correcting the coexisting background factors with realignment, stabilizing or unloading procedures, along with the treatment of cartilage lesions, is improving the results over time. ACI provides a satisfactory result even for the difficult cases with concomitant patellar instability. Our long term follow up study reveals preservation of good results and of high level of patients’ activities, even 10 to 20 years after the implantation, in both isolated trochlea and patella lesions and also in multiple and in kissing lesions where an intervention could be considered as a salvage procedure.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P16-397 Effect of intraarticular pH change on the structure and metabolism of chondrocytes Kocaoglu B.1, Ergu¨n S.2, Akgun U.3, Basci O.4, Karahan M.5 1 Acibadem University Faculty of Medicine, Orthopaedic Surgery, Istanbul, Turkey, 2Marmara University Faculty of Medicine, Istanbul, Turkey, 3Marmara University, School of Athletic Education, Istanbul, Turkey, 4Marmara University Faculty of Medicine, Orthopaedic Surgery, Istanbul, Turkey, 5Marmara University, Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey Objectives: Decrease in pH of synovial fluid in the presence of osteoarthritis and trauma has been reported in previous investigations. We hereby aimed to show the negative effects of acidic synovial fluid on cartilage metabolism by Real Time Polymerase Chain Reaction method. Methods: Cartilage tissue obtained from bovine femoral condyles were incubated in 3 different medias; acidic (pH 7.2), normal (pH 7.4) and basic (pH 7.6), for four days. At the end of the incubation, mRNA isolation and quantitative gene expression analysis were done. Type II collagen, aggrecan, hypoxia inducible factor I alpha (Hif 1 alpha) and beta-actin (housekeeping gene) were the genes investigated. Normalizing was done by 2-DDCt method. According to results, expression levels of all genes in normal pH media was higher compared to acidic and basic media. Results: Expression level ratio of gene type II collagen was 0.739 in acidic and 0.755 in basic media when compared to normal pH media expression level. This ratio was 0.615 in acidic and 0.681 in basic media of gene aggrecan and 0.695 in acidic and 0.652 in basic of gene Hif 1 alpha. Conclusions: Conclusion, decrease in synovial fluid pH as a consequence of osteoarthritis, trauma or intraarticular hyaluronic acid injection, negatively effects chondrocyte metabolism and production of extracellular matrix components.

P16-405 Articular inlay resurfacing for the treatment of focal full thickness femoral condyle cartilage defects Bollars P.1, Bosquet M.2, Bellemans J.1 1 KU Leuven, Orthopaedic Surgery, Leuven Pellenberg, Belgium, 2Europa Hospital Brussels, Orthopaedic Surgery, Brussels, Belgium Objectives: Localized full thickness defects of the femoral condyle can be highly symptomatic. Treatment options for these lesions are plentiful in young patients, however they become increasingly challenging in middle aged and older patients, in particular when repeat and biological revision procedures have been exhausted. In order to delay traditional joint replacement procedures and to provide a soft tissue and bone sparing alternative, a contoured focal inlay resurfacing procedure was introduced to the market in 2004. Methods: 17 patients (18 implants) were treated and re-assessed at an average follow-up of 37 months (range 20-57). 16 were female, 1 was male. The average age was 50 years (range 43-62). 11 were diagnosed with early focal osteoarthritis, 5 had a localized traumatic full thickness defect, and 1 osteonecrosis. Implant diameters were 15mm (N=12) and 20mm (N=6). 15 patients were treated on the medial femoral condyle, 2 on the lateral condyle, and 1 patient on both condyles. 10 were implanted in left knees, 8 on the right. The mean mechanical axis deviation was 2 degrees into the affected compartment (range 0-7). 50% of patients had a decreased meniscal function at index procedure. All knees were ligamentously stable. 2 patients had grade 1 opposing cartilage changes, all others were normal. Results were assessed using the KOOS scores, as well as physical examination and radiographs. Results: On physical examination, 13/17 had a normal physical examination with full extension, flexion to a minimum of 125 degrees, no swelling, ligamentous stability and full quadriceps function. 4/17 had variable deductions from a normal exam. Average KOOS domains at last follow-up were 90/100 (Pain, range 75100), 87/100 (Symptoms, range 57-100), 93/100 (ADL, range 83-100), 73/ 100 (Sport, range 50-100), 76/100 (QOL, range 63-100). The mean

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 duration in postoperative hospitalization was 4 days (range 3-7). No implant loosening, subsidence, cystic formation, or progressive loss of joint space was observed in any patient during the radiographic follow-up. No postoperative complications were observed. No secondary knee procedures were performed in this patient population. Conclusions: Limited femoral condyle resurfacing has demonstrated excellent pain relief and function improvement in middle aged 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, remain critical patient selection criteria for successful outcomes.

P16-446 A new avenue for cartilage repair - xenograft chondrocytes source for hyaline cartilage Nierenberg G.1, Maor G.2 1 Rambam Health Care Campus, Division of Orthopaedics, Haifa, Israel, 2 Technion, Israel Institute of Technology, Cell Biology, Haifa, Israel Objectives: Damage to joint cartilage, caused by trauma or osteoarthritis, results in pain, and altered quality of life. Since hyaline cartilage does not heal spontaneously, the desired biological long term solutions are based on tissue replacement. Currently tissue engineering, is focused on autologous transplantation (ACI) which universally results in mixed fibrous / hyalinelike tissue in various proportions. The key element for spontaneous generation of hyaline cartilage is proper stable cell differentiation. Chondrocytes, originating from neonatal porcine mandibular condyle have demonstrated capacity to produce genuine hyaline cartilage. Differentiated cells from a unique epiphyseal structure with high capacity to maintain phenotypic expression are targeted and explored for possible tissue implantation. Methods: Mandibular condyle (MC)-derived cells (MCDC), harvested from neonatal SPF porcine, cultured in vitro, resulted in spontaneously differentiating hyaline cartilage-producing cells. The cultured cells form a continuous cartilage film (CartimoveTM, offering the possibility for simple mechanical handling, as well as re-culturing while preserving its proliferating and differentiating activities. Pre-clinical studies consisted of controlled lesion in medial femoral condyle of 8 goats, in which Cartimove was transplanted. Three, 6 and 13 months implants were analyzed for hyaline morphology and biomechanics. Results: Within 10 days in culture, the expression of type II collagen is amplified by about 50 folds (RT-PCR), at the same time, expression of type I collagen is gradually reducing. In co-culturing studies (with humanderived cells), no transmittable pathogen were found. In the preclinical studies, parameters as tissue filling, environmental integration and hyaline characteristics were demonstrated. Local as well as lymphatic organ analysis indicated no detectable immune response. Conclusions: It is demonstrated that cultured porcine-derived MCDC cells, are capable to replenish cartilage lesions in a goat model with genuine hyaline cartilage.

P16-502 Apoptotic cell death induction of synocvial cells by an array of RGD peptides Sasho T.1, Sugioka K.1, Nakagawa K.1, Matsuki K.1, Ikegawa N.1, Saito M.1, Akagi R.1 1 Graduate School of Medicine, Chiba University, Department of Orthopaedic Surgery, Chiba, Japan Objectives: Small peptides including Arg-Gly-Asp (RGD) motif (RGD peptide) are known to induce apoptotic cell death in variety of cells. Two different mechanisms have been proposed for this cell-death induction. One mechanism is due to inhibitory character of RGD peptides against cell binding to extra-cellular matrix (ECM). As main ligand for integrins on cell surface is RGD motif included in fibronectin and other major ECM members, occupation of integrins by RGD peptides disturbs cells from

S247 attaching to ECM. Cell attachment is crucial for cell survival, therefore adhesion-blocking agents including RGD peptides cause caspase-dependent apoptotic cell death known as anoikis. The other mechanism is direct activation of caspase-3 by RGD peptides where incorporated RGD peptides into cytoplasm convert procaspase-3 into active form. With the latter mechanism, RGD peptides induce cell death even in un-adherent cells such as lymphocytes. With simple structure and easy productivity, therapeutic usage of RGD peptides is expected in several fields such as cancer therapy. Administrating RGD peptides into the joint might be useful in synovial proliferative disease such as rheumatoid arthritis (RA) or pigmented villo-nodular synovitis (PVS). For this purpose in mind, we studied the effects of RGD peptides on cells in two major joint component; chondrocytes and synovial cells. Methods: Chondrocytes and synovial cells were enzymically isolated and cultured from the knee joints of New Zealand White rabbits (n=4). After one passage, cells were incubated in serum-free medium with 5mM of RGD peptides (RGD-X, X represents single amino acid) for 4hours, and the rate of apoptotic cells were evaluated. As a negative control RGES (non RGD small peptide) was used. Using double staining of Annexin-V (Annexin-V: early marker for apoptotic cell) and propidium iodide (PI), the rates of live cells and apoptotic cells were measured by flow cytometry. Statistic analysis was performed using Mann-Whitney’s U and p\0.01 was considered as significant. Results: Six RGD peptides induced significant cell death in chondrocytes comparing to RGES treated cells and 10 peptides in synovial cells. Among them RGDC significantly induced cell death of synovial cells but not chondrocytes. Percent of live/apoptotic cells treated with RGD p Chondrocytes

Synocvial Cells

Live

Live

Apoptotic

Apoptotic

No peptide

96.5

0.57

94.5

3.9

RGE-S

95.6

0.52

93.4

5.4

RGD-A

87.6

7.1

79.4

15.6*

RGD-R

85.3

6.1

76.1

15.2

RGD-N

86.9

6.3

77.5

18.1*

RGD-D

18.8

42*

38.3

50.9*

RGD-C

88.2

3.7

15.4

76.9*

RGD-Q

81.8

8.5

61.5

31.2*

RGD-E

86.9

8.3

52.9

36.2*

RGD-G

82.6

9.5

84.2

11.5

RGD-H

81.6

0.3

77.2

18.6*

RGD-I

91.1

2.8

86.1

15.5*

RGD-L

91.1

2.8

86.1

11.9

RGD-K

84.8

6.6

83.9

12.1

RGD-M

85.8

6.8

83.6

11.9

RGD-F

74.7

18.9*

87

10.6

RGD-P

82.5

9.3

83.9

13

RGD-S

87.5

6.1

83.1

12.9*

RGD-T

84.5

8.9*

99.9

0.02

RGD-W

95.4

0.6

94.6

3.8

RGD-Y

88.1

9.6*

81.9

13.7*

RGD-V

83.5

8.5*

88.8

8.7

Average percent of live cell and apoptotic cell was shown (n=4). *: Significant difference comparing to RGE-S. #: Significant difference between chondrocytes and synovial cells. Conclusions: Among 20 RGD peptides, RGDC is strongest cell death inducer in synovial cells but not in chondrocytes.

123

S248 P16-523 Decreased degradation of cartilage by hyperbaric oxygen and hyaluroniate, an animal study in rats0 intervertebral disc model Chang C.-H.1, Chen C.-H.1, Whu S.-W.1, Wang I.-C.1, Sun H.-J.1, Wu M.-Y.1, Liu H.-W.2 1 Chang Gung Memorial Hospital-Keelung, Orthopaedic, Keelung, Taiwan, Republic of China, 2Fu Jen Catholic University, Life Science, Xinzhuang, Taiwan, Republic of China Objectives: The process of disc degeneration is believed to have a biochemical basis, with inhibition of nuclear proteoglycan synthesis and enhanced matrix degradation cause by chemical mediator that may include interleukin-1 (IL-1), interleukin-6 (IL-6), nitric oxide (NO), prostaglandin E2 (PGE-2) and matrix metalloproteinaes (MMPs). Hyperbaric oxygen (HBO) treatment stimulated bone healing and endothelial cell, fibroblast proliferation but inhibited cancer division. Furthermore, HBO treatment resulted in a decreased synthesis of NO but an increased synthesis of hyaluronic acid, proteoglycan, and collagen. Growth factors induce stimulation of type II collagen synthesis in chondrocytes. In this study, effect of HBO and hyaluroniate (HA) on degradation of cartilage in intervertebral disc was evaluated. Methods: Sprague-Dawley (SD) rats had the proximal two intervertebral discs in the tail injected with 0.10ml (0.25unit/ml) chondroitinase ABC (ABC) as degenerated model. After the injection surgery, thirty six SD rats were divided into six groups: Group 1: normal animal without any treatment. Group 2: exposed to HBO and injected 0.10ml PBS without ABC treatment as control. Group 3: injected ABC only. Group 4: exposed to HBO after injected ABC. Group 5: injected 0.10ml (10mg/ml) HA after ABC treatment. Group 6: exposed to HBO and injected HA after ABC treatment. Secretion of NO and PGE-2, histology and biochemistry assay were examined after ABC injection for 14 days. Results: After 14 days, lower concentration of iNOs in HBO groups (2, 4 and 6 groups) than those in other groups were found. Lower concentration of PEG2 in 4 and 6 groups than in 3 and 5 groups were observed. Thickness of cartilage in group 4, 5 and 6 were similar as those in group 2, but were wilder than in group 3. Glycosaminoglycan (GAG) and total collagen in 2*6 were much lower than in group 1. The highest GAG and collagen content in group 6 among group 2*6 was determined. The similar effect of HBO and hyaluroniate on cartilage degradation was concluded. The degradation of damaged cartilage of intervertebral disc treated with both of HBO and hyaluroniate was decreased. Conclusions: By treating HBO and hyaluroniate, cartilage degradation will be decreased by reducing iNO and PEG2.

P16-575 Clinical outcome and radiographic prognostic factor of juvenile osteochondritis dissecans of medial femoral condyle treated with transartucular drilling Yonetani Y.1, Shiozaki Y.2, Tanaka Y.1, Horibe S.1 1 Osaka Rousai Hospital, Sakai, Japan, 2Seifu Hospital, Orthopeadic Surgery, Sakai, Japan Objectives: The purpose of this study was to evaluate the clinical outcome and radiographic prognostic factor of the transarticular drilling of patients with juvenile osteochondritis dissecans of medial femoral condyle after 6 months of unsuccessful non-operative treatment. Methods: A total of 15 osteochondritis dissecans lesions of medial femoral codyle in 15 skeletally immature patients were treated with transarticular drilling. There were 13 boys and 2 girls with a mean age of 12 years old. The functional outcomes were evaluated by use of Lysholm score at a mean follow-up of 17 months after drilling, and preoperative condition and healing of the lesions was evaluated by use of computed tomography and magnetic resonance imaging. Results: The Lysholm score in all cases significantly improved postoperatively (a mean, from 74.9 to 99.3). All lesions except 3cases healed after transarticular drilling. Healing was achieved at a mean of 4.8 months on computed tomography and 7.2 months on magnetic resonance imaging. Radiographic non-healing lesions in 3 cases were tend to have the chondro-osseous findings on the computed tomography, opposite to the rapid

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 healing lesions in 6 cases having the findings of subchondral defect on the computed tomography. Conclusions: Transarticular drilling for patients with stable juvenile osteochondritis dissecans of the knee is one of good treatment choice. Chondro-osseous radiographic finding might be poor prognostic factor in transarticular drilling.

P16-608 The effect of autologous mesenchymal stem cells on the articular cartilage regeneration (experimental examination) Kostub A.1, Zasadnyuk I.1 1 Kyiv, Sport Trauma, Kyiv, Ukraine Objectives: The thesis is devoted to the study of effect of autologous mesenchymal cells of bone marrow with various degree of chondrogenic differentiation on the course of reparative chondrogenesis of the articular cartilage with its traumatic injury. Methods: According to the light microscopy, the positive effect of intraarticular infusion of the culture of autologous mesenchymal cells of bone marrow on the reparative chondrogenesis has been proven. The analysis of the obtained data for the alternative OS scale has revealed the reliable best result with the use of undifferentiated culture of autologous mesenchymal cells (8,6 ± 0,24) of the model of traumatic injury of the articular cartilage as compared with the culture of autologous mesenchymal stem cells with former directed chondrogenic differentiation (7,6 ± 0,24), (p \ 0,05). It has been proven by the method of tracing of autologous mesenchymal cells of bone marrow with the aid of fluorescent probes PKH-26 that the given cells take part directly in the processes of chondroreparation with their intra-articular infusion into the articular cavity. Results: It has been proven on the basis of performed biochemical examinations that the intra-articular infusion of autologous mesenchymal cells of bone marrow at the earlier stages of pathologic process stabilizes the metabolic processes in cartilaginous tissue, and farther on normalizes them reaching the physiologic norms that are characteristic for intact animals. Conclusions: The practical value of the examination results consists in argumentation of high efficiency of the articular cartilage treatment with the aid of autologous mesenchymal cells of bone marrow in experimental animals and the experimental substantiation of the use of autologous mesenchymal stem cells of bone marrow in the treatment of orthopedic and traumatologic patients with traumatic injuries of the articular cartilage with the aim of renewal of function of the injured joint and prevention of the posttraumatic osteoporosis progression.

P16-625 Biomechanical analysis of the relationship between the plug and the deep of the receiving tunnel in autologous osteochondral transplantation Fleaca R.1, Roman M.1, Oleksik V.2, Pascu A.2, Deac C.2, Baier I.1 1 University of Sibiu, Orthopedic and Trauma Surgery, Sibiu, Romania, 2 University of Sibiu, Faculty of Engineering, Sibiu, Romania Objectives: The purpose of this study is to analyze the biomechanical behaviour a a distal femur with autologous osteochondral transplantation depending on the relation between the long af the plug and the receiving tunnel. Methods: For the analysis of the mechanical properties of the osteochondral transplant depending of the plug long and the deep of the receiving tunnel we used a fresh bovine femur and we have done a sagittal section of the internal femoral condyle. We have realized 2 receiving tunnels deep 15 mm and 18 mm with 1/3 of their perimeter outside the section, then we have harvested 2 osteochondral cylinders with diameter 8 and long 15 mm and transplanted at the side of the tunnels. Thus, one was at 3 mm from the base of the tunnel, and the other one at the level of the base. The surface of the section was then covered with white mated paint and sprayed with graphite, which was necessary for image acquisition with the Optic System Aramis 2M. The prepared femur was positioned at the

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Machine for testing Instron 5587. After the calibration of the Aramis 2M System we have loaded the femur along the axis of each of the 2 cylinders sequential up to 5.000 N, with concomitant acquisition of the optic data in rhythm of 1 pictures/sec. The data acquired by the system allowed the analysis at different pressure levels of the Major Strain, Minor Strain, Equivalent Von Misses Strain, as well as Displacement on X, Y axis, and Total Displacement E. We have realised with the soft of the Aramis System sections at the level of the 2 analysed cylinders, which allowed us to analyse the behaviour of all the points of the sections. Results: The results achieved shows that at 5.000 N the major strain is 4,52% when the plug doesn’t reach the base of the tunnel and 7,92% when it reach the base, the minor strain is 4,94% vs. 6,8%, the von Misses strain is 5,557% vs. 8,94%. These values obtained in both loads are small, but we have noticed higher levels in the second case. The data analysis show that in the first case the behaviour on axial load of the plug is determined by the processes from the plug/tunnel interface and in the second case mainly by the strains and displacements at the level of the base. Conclusions: When the plug doesn’t reach the base of the tunnel the two bodies behaviour is different in compression, more different with the increasing of the loading force, and when the plug reach the base the two bodies the behaviour is unitary increasingly with the pressure. Acknowledgments: This work was carried out within the framework of the research grant named ‘‘Theoretical and experimental analysis of the static and dynamic behavior of the grafts in the autologous osteochondral transplantation’’ support by the Romania Ministry of Education and Research and National University Research Council.

P16-636 Effects of HYLAN G-F 20 [Synvisc Ò] supplementation on cartilage preservation in osteoarthritis of the knee assessed by MRI: a two-year, single-blind clinical trial Hall S.1, Wang Y.2, Cicuttini F.2, Hanna F.2, Wluka A.2, Feletar M.1, Grant G.3 1 Monash University Cabrini Health, Medicine, Malvern, Australia, 2 Monash University, Alfred Hospital, Epidemiology, Melbourne, Australia, 3Monash University Cabrini Health, Malvern, Australia Objectives: To assess the effect of viscosupplementation with HYLAN G-F 20 on the progression of cartilage loss over two years in patients with knee osteoarthritis (OA). Methods: A single-blind, parallel control group pilot clinical trial was performed in 78 eligible patients with symptomatic knee OA (KellgrenLawrence grade II,III). Patients were assigned to either an intervention group (n=39, receiving four courses (3 x 2.0cc) of intra-articular HYLANG-F 20injections six monthly or a control group (n=39, without injections but receiving usual care for OA). Magnetic resonance imaging of the target knee was performed at baseline,6,12,and 24 months. Images were analysed blinded both to patient group and sequence. Tibial cartilage volume, tibiofemoral cartilage defects and bone marrow lesions were assessed at baseline and follow up. Results: 55 subjects (71%) completed 2 year follow up. There was no significant difference in age, gender, BMI, baseline cartilage volume or bone area in those who completed and those who did not (all P[0.13). Analysis of completers demonstrated a significantly reduced annual percentage rate of medial, lateral and total tibial cartilage loss in the intervention group (mean±SD, -0.3±2.7%, -1.4±4.3% and -0.5±2.3%) compared with the control group (2.3±2.6%, 1.4±2.6%, 1.6±1.8%, P=0.001, 0.005 and 0.001 for difference, respectively). The intervention group also showed a significant reduction in the increase of cartilage defect score in the medial and total tibiofemoral compartments (0.1±1.3 and 0.5±2.0) compared with the control group (0.8±1.5 and 1.6±2.0, all P=0.05). There was no significant difference of change in bone marrow lesions between the intervention and control groups.

S249 Conclusions: 6 monthly intra-articular injections of HYLAN-G-F 20 administered without regard to symptoms have a beneficial effect on knee cartilage preservation as measured by both cartilage volume and cartilage defect score: Over two years, the control group continues to lose cartilage while there is no significant loss of cartilage in the HYLAN G-F treated group. HYLAN-G-F 20 could be further evaluated in larger trials as a possible disease-modifying agent in patients with knee OA.

P16-677 Long term results of hyaluronic acid in mild to moderate osteoarthritis of the knee; a prospective case series study in 304 patients Hoekstra H.1, van der Weegen W.1, van Drumpt R.1, Bootsma H.-P.2, Vellema J.1, Bos E.1, Sybesma T.3 1 St. Anna Hospital, Orthopedic Surgery, Gelrop, Netherlands, 2St. Anna Hospital, Clinical Pharmacology, Gelrop, Netherlands, 3St. Anna Hospital, Gelrop, Netherlands Objectives: To evaluate the long term changes in knee pain and functioning after treatment with hyaluronic acid. Secondary outcome measures were safety and patient satisfaction. Methods: Three intra-articular injections with Fermathron (Hyaltech Ltd, Edingburgh, Scotland) at weekly intervals, were given to 304 consecutive patients with mild unilateral or bilateral osteoarthritis of the knee (Kellgren&Lawrence grade 1-3). Clinical outcomes, adverse events and patient satisfaction were measured three months and 12-18 months after the final injection. A two-tailed paired t-test for parametric data or a Wilcoxon signed rank test for non-parametric data were used to test for any statistical differences. Results: Mean pain significantly reduced from 5.49 to 4.12 (p\0.001), 48% were classified as responders and 45.5% was satisfied with treatment results. In 5% of all cases, minor adverse events were observed. Conclusions: After treatment with hyaluronic acid, pain is significantly reduced at long term follow up (C 12 months). No serious adverse events were recorded. Fermathron can be regarded as a safe brand of hyaluronic acid for patients with mild osteoarthritis of the knee.

P16-688 Cartilage impactor shape alters pore pressure: a finite element analysis Fening S.D.1, Miniaci A.1 1 Cleveland Clinic, Dept. of Orthopaedic Surgery, Cleveland, United States Objectives: Impact injury to cartilage is thought to lead to degenerative osteoarthritis. Development of a reliable model for the early stages of cartilage injury and osteoarthritis would enable the testing of novel treatments that could slow or even reverse the progression of this debilitating disease. Previous research utilizing direct impact of rabbit knee articular cartilage has reported the most consistent results. However, cartilage damage from direct impact does not appear to be uniform, despite a curvature matching impactor. We hypothesize that stress concentrations arise from the rigid impactor, and that changes to impactor geometry will alter the pore pressure distribution in the cartilage. Methods: Three axisymmetric model geometries were constructed in ABAQUS (Fig. 1). Bone was modeled as a linear elastic material, cartilage as a biphasic poroelastic material, and the steel impactor as a rigid body. Sensitivity analyses were performed on the material properties (bulk modulus, Poisson’s ratio, permeability) and thickness of the cartilage. Three levels of each parameter were chosen for the sensitivity analyses based on previously published data. This resulted in a total of 243 simulations for all impactor geometries. Outcome measures were defined as the peak pore pressure and the position of the peak pore pressure. ANOVA was used to evaluate the effect of impactor shape on these outcome measures.

123

S250

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

Concave

Flat

Convex

Fig. 1 Impactor geometries Results: Peak pore pressure was significantly dependent upon the impactor geometry (Fig. 2). The concave impactor head had the lowest peak pressure, and the convex had the greatest. The position of peak pressure was also significantly dependent upon the impactor geometry.

Max Pore Pressure (MPa)

0.8

Effect of Impactor Shape on Max Pore Pressure (p < 0.001)

0.6 Convex

0.4 0.2

Flat Concave

0 0.247

0.416 Thickness (mm)

0.670

Fig. 2 Effect on pore pressure Conclusions: Stress concentrations were observed in all cases, even when the impactor geometry matched the cartilage geometry. Changes to geometry significantly altered the pore pressure distribution within the cartilage. The concave impactor shape had the most even pressure distribution of the three geometries evaluated. Still, the pressure distribution was not even, and was typically concentrated at the cartilage-impactor interface and at the radial edge of the impactor.

P16-696 Biological reactivity and rheological characteristics of two crosslinked HA-based viscosupplements: JonexaTM and DurolaneÒ Skrabut E.1, Yu L.-P.2, Dethlefsen S.3, Voschin E.2, Moran N.2, Corazzini R.2, O’Brien C.4 1 Genzyme Corporation, Waltham, Massachusetts, United States, 2 Genzyme Corporation, Framingham, Massachusetts, United States, 3 Genzyme Biosurgery, Cambridge, Massachusetts, United States, 4 Genzyme Corporation, Cambridge, Massachusetts, United States Objectives: Studies were conducted to compare the physical properties and biological reactivity of two commercially available crosslinked hyaluronan (HA)-based viscosupplements, JonexaTM (hylastan SGL-80) and Durolane (NASHA). Methods: The rheological properties of the materials were assessed by a control stress rheometer. The dilution tolerance of the viscosupplements was determined by evaluating the rheological properties at various dilutions. The percent dilution at which the phase angle increased to 50% of its original value was defined as the dilution durability. The biological reactivity of each test material was evaluated in guinea pigs by injecting into the knee joint fat pads. Animals were sacrificed at 7 or 30-days following injection and the fat pads were collected and processed for histological evaluation. Additionally, the biological reactivity of each material was evaluated in a rabbit muscle implant model at 1, 4, and 12weeks after implantation.

123

Results: Rheology results suggest that hylastan SGL-80 behaves as a soft and cohesive gel that maintains its physical properties against dilution. In contrast, NASHA behaves as a suspension of hard particles in buffer. Hylastan SGL-80 and NASHA were determined to have dilution durabilities of 550% and 25%, respectively. This suggests that hylastan SGL80’s viscoelastic properties were maintained at a high level of dilution whereas NASHA’s viscoelastic properties were not maintained, even at a low level of dilution. On day-7, residual HA-based material was observed microscopically in fat pads taken from both hylastan SGL-80 and NASHA-treated animals. Test material was noted as a basophilic substance that formed discreet clusters within the adipose tissue. The cellular response to these two materials was characterized by the presence of a thin layer of macrophages at the host / test-material interface. Macrophages appeared to easily infiltrate and degrade the test material in the hylastan SGL-80-treated animals but not in the NASHA-treated animals. At Day 30, residual hylastan SGL-80 was not observed in any of the fat pad tissues; however, residual NASHA was observed as discrete aggregates of basophilic material associated with a single layer of macrophages at the host material interface. Several layers of fibrous connective tissue completely circumscribed and encapsulated the test material. In the rabbit muscle implantation study, the materials showed similar tissue reactivity at 1-week post implantation. At 4 and 12-weeks, hylastan SGL-80 showed lower reactivity scores than NASHA. Microscopically, NASHA was still observed at the injection site and was associated with a cellular infiltrate of lymphocytes and macrophages. Conclusions: Hylastan SGL-80 is a much softer, smoother gel with superior dilution tolerance and less tissue reactivity than NASHA. As a viscosupplement, hylastan SGL-80 is expected to provide better lubrication, shock absorption (under moderate dilution) and therefore, better physical protection to cartilage than NASHA. In the long-term muscle implantation studies, hylastan SGL-80 was less reactive than NASHA. The hylastan SGL-80 injected fat pads demonstrated a typical biomaterial response with macrophages digesting the material at an early timepoint and disappearing at a later timepoint once the material was resorbed. The NASHA injected tissue demonstrated a similar early timepoint response, but at the later timepoints showed a cellular and fibrous encapsulation of the injected material.

P16-697 Safety and efficacy of repeat treatment with hylan G-F 20 in osteoarthritis of the knee Raman R.1, Johnson G.2, Dutta A.3, Shaw C.3, Sharma H.3 1 Hull Royal Infirmary, Trauma and Orthopaedics, Normanton, United Kingdom, 2Academic Dept of Orthopaedics, Normanton, United Kingdom, 3Hull Royal Infirmary, HULL, United Kingdom Objectives: Efficacy of viscosupplementation is variable ranging from 3-12 months. This warrants the need of a repeat course of treatment. There is a paucity of clinical evidence addressing the efficacy and safety of repeat courses of this treatment. The aim of this study was to assess the safety and efficacy of repeat intra-articular injections of hylan GF-20 (Synvisc) in the treatment of osteoarthritis of the knee. Methods: In this independent study, patients who had previous viscosupplementation treatment to the knee were offered repeat treatment after a minimum of 6 months after the initial course. The inclusion criteria were pain score of [6 on a VAS (0-10) in the target knee and one course of previous viscosupplementation treatment. As an extension of our previous study, we identified 375 consecutive patients, who were prospectively reviewed by blinded independent assessors at pre injection, 1 week, 6 weeks, 3, 6, 12 months after repeat treatment with hylan G-F 20. The primary outcome variable was knee pain on VAS at 6 months. Functional outcome was assessed using WOMAC, Oxford knee score and EuroQol EQ-5D. All adverse events (AE), including injection and treatment related AE in the target knee were recorded at 1week and 6 weeks after treatment. Mean follow up was 12 months. Results: The mean time from primary course of treatment was 35.2 months (30.2 -sodium hyaluronate group, 47.3 - hylan G-F 20 group). 187 patients received sodium hyaluronate at initial treatment and the rest had hylan G-F

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 20. Knee pain on VAS improved from 6.9 to 3.9 at 6 months (p=0.02) and to 4.8, p=0.04 at 12 months. Significant improvements from the baseline in the WOMAC pain and function subscales and Oxford knee scores at 3, 6 and 12 months were recorded. Subgroup analysis revealed greater improvement in pain in patients with previous sodium hyaluronate treatment at 3, 6 months although no statistically significant difference could be demonstrated when compared to the patients with previous hylan G-F 20 treatment. Overall AE compared to the patients with previous hylan G-F 20 treatment. Overall AE were recorded in 23% (19% in initial course). Treatment and/or injection related minor AE in the target knee were observed in 4.7% (4.1% in the sodium hyaluronate hylan group and 5.1% in the hylan - hylan group). Previous treatment regime did not influence the incidence of AE in the patients. None of the patients had serious AE. Conclusions: There is a significant reduction of pain and improvement in function following treatment of symptomatic OA of the knee with repeat injections of hylan G-F 20. The magnitude of pain relief and the longevity of symptom control were comparable with the results from the first course of treatment. Repeat injections of hylan G-F 20 are well tolerated as shown by the low rate of AE. From this study, it appears that repeat injections of hylan G-F 20 can be both safely and effectively administered in patients with symptomatic OA of the knee.

P16-722 Reduced chondrocyte viability is associated with marker pen ink Getgood A.1, Mcnamara I.1, Kili S.2, Bhullar T.3 1 University of Cambridge, Orthopaedic Research Unit, Cambridge, United Kingdom, 2Genzyme Biosurgery, Oxford, United Kingdom, 3Edith Cavell Hospital, Peterborough, United Kingdom Objectives: Sterile surgical marker pens are commonly used in cartilage repair surgery to aid in the placement of periosteal patches or collagen membranes in autologous chondrocyte transplantation. The aim of this study was to investigate the effect that methylene blue dye would have on human chondrocytes when cultured on a collagen membrane in-vitro. Methods: Bilayered collagen membranes were seeded in duplicate with 12 million human chondrocytes per ml and cultured for 24 hours under standard conditions. Group A consisted of a membrane marked with methylene blue ink on its smooth side, group B marked on its porous side, and group C acting as an unmarked control. At the end of the culture period the membranes were qualitatively analysed for cell survival by live/ dead fluorescent staining under confocal microscopy. Results: The control membranes of group C showed normal ‘live’ staining of cells. A marked reduction in chondrocyte viability was seen in groups A and B, with a significant reduction in viability seen opposite to the ink mark on the smooth side in A, and practically no cell viability in the same position on the porous side in group B. Conclusions: Methylene blue dye in surgical marker pen ink has been shown to be cytotoxic to chondrocytes cultured on bilayered collagen membranes. Surgeons need to be aware of the potential harmful effects of methylene blue dye when using surgical marker pens in cell based regenerative therapies.

P16-790 Infiltrative treatment with autologous platelet rich plasma in early osteoarthritis: results at 6 months Gobbi A.1, Boldrini L.2 1 Orthopaedic Arthroscopic Surgery International, Dept. of Sport and Medicine, Milano, Italy, 2OASI Bioresearch Foundation, Research, Milano, Italy Objectives: Cartilage lesions represent a significant clinical problem because of their limited ability to regenerate. The rationale for topical use of platelet-enriched plasma (P.R.P.) preparations is to stimulate the natural healing cascade and tissue regeneration by a ‘‘supra-physiological’’ release of platelet-derived factors directly in the site of treatment. Recent studies have documented the effectiveness of growth factors in chondrogenesis and preventing degeneration of the joints.

S251 Methods: We followed prospectively a group of 34 patients affected by Gr.III or Gr. IV knee chondral lesions, diagnosed at MRI or previous arthroscopic evaluation, treated with 2 intra-articular injections (1 month between each injection) of autologous P.R.P. (RegenLab, Switzerland) by a supra-patellar approach. VAS and KOOS scores were collected in all patients before the treatment and after 3 to 6 months. Results: 34 patients with a mean age of 50.3 (SD 6.4) have been treated in our institution in the last year with this technique for grade 3 and 4 cartilage knee lesions. At a mean follow-up of 5.9 months (SD 2.3), patient showed improvements in all scores. Mean pre-treatment values were: KOOS Scores: P=71.5/S=69.6/ADL=77.0/SP39.8/QOL= 42.5; VAS scale 4.0 (SD 2.9). At final follow-up mean scores were: KOOS P=80.5/S=77.0/ ADL=87.3/SP=50.4/QOL=54.9; VAS 2.3 (SD 2.3). Similar improvement were reported in patients with and without previous surgeries. No adverse reaction or post-treatment complications were noted in all patients. Conclusions: This report shows that in early osteoarthritis knee patients the treatment with injections of platelet-enriched preparation can lead to an improvement of perceived pain and functional capacities at 6 months.

P16-791 Mesenchymal stem cells implantation for full-thickness cartilage lesions treatment: preliminary report Gobbi A.1, Boldrini L.2 1 Orthopaedic Arthroscopic Surgery International, Dept. Sport and Medicine, Milano, Italy, 2OASI Bioresearch Foundation, Research, Milano, Italy Objectives: Cartilage lesions represent a significant clinical problem because of their limited ability to regenerate. Recent advances in our understanding of the functions of mesenchymal stem cells (MSC) have shown its chondrogenic potential. The use of autologous MSC represent an improvement on the currently available techniques for cartilage transplantation avoiding the first surgery for cartilage biopsy and chondrocyte cultivation. Methods: A group of 25 patients have been treated with one step cartilage Transplantation with M.S.C. in the last 3 yrs. at our institution and prospectively followed-up 12 patients reached a minimum of 2 yrs of followup and have been included into this preliminary study. Inclusion criteria for this study were: patients with Gr.III or Gr. IV chondral lesions treated with standard arthroscopic approach and implanted with concentrated MSC from the iliac crest covered by a second generation collagenic scaffold. All patients followed the same specific rehabilitation program after MSC implantation. IKDC, KOOS, Lysholm and Tegner were collected at pre-op and at 6 months, 1 and 2 years postoperatively. Results: 12 patients (7 males and 5 females) with a mean age of 45.3 yrs. treated with this technique reached the 24 months follow-up. All these patient showed improvement in all scores: Tegner Pre. op 2 Post. 5, IKDC subjective Pre. 39.5 Post. 70.1, KOOS Pain Pre. 72.1 Post. 96 KOOS Symptoms Pre. 72 Post. 87.6 KOOS ADL Pre. 71.6 Post. 96.7 KOOS Sport Pre. 41 Post. 63.3 Quality of Life Pre. 38.2 Post. 63.7. No adverse reaction or post-op complication were noted in all patients. Conclusions: This preliminary report shows that Mesenchymal Stem Cell Implantation can be a viable technique in the treatment of Gr. III and Gr. IV chondral lesions.

P16-819 Effects of combined 3D- and hypoxic culturing on cartilage-specific gene expression in human chondrocytes Foldager C.B.1, Munir S.2, Ulrich-Vinther M.2, Bunger C.2, Lind M.3 1 ˚ rhus C, Denmark, 2Aarhus Aarhus, Orthopaedic Research Lab, A ˚ rhus C, Denmark, University Hospital, Orthopaedic Research Lab, A 3 Aarhus University Hospital, Sports Trauma Clinic, Aarhus, Denmark

123

S252 Objectives: Articular cartilage defects are known to have a very limited potential of self-repair. In cartilage, native chondrocytes reside in a 3-dimensional (3D) network and are exposed to low oxygen levels. In vitro expansion of autologous chondrocytes is an essential part of clinically used treatments such as autologous chondrocyte implantation (ACI) and matrix-assisted chondrocyte implantation (MACI). The aim of this study is to investigate conventional monolayer culturing compared to an in vivo-mimicking environment using a combination of 3D scaffold and hypoxia. Methods: Cartilage biopsies were collected from the intercondylar groove in the distal femur from 12 healthy patients undergoing anterior cruciate ligament reconstruction. The protocol was approved by the local ethical committee under the Danish National Committee on Research Ethics. Isolated chondrocytes were cultured separately for each patient in DMEM/ F12 medium with antibiotics in normoxia (21% oxygen tension) until they reached confluence. Following trypsination cells were divided to monolayer or scaffold seeding. For the monolayer culture cells were seeded in 24-well plate with a density of 20,000 cells/cm2. MPEG-PLGA scaffolds (ASEED, Coloplast A/S, Denmark) soaked in culture medium (4 mm in diameter) were placed in agarose coated 24-well plates to prevent cells from adhering to the wells. 125.000 cells in 10 lL media were added on top of the wet scaffold (seeding concentration 5x106 cells/mL). 24 hours after seeding, the cells for baseline measurement (t=0) were harvested. The remainders were divided in three groups for incubation in either normoxia (21% oxygen tension), hypoxia (5% oxygen tension) or severe hypoxia (1% oxygen tension) in a designated hypoxia workstation (Xvivo System, BioSpherix, NY). RNA extractions from these subcultures were performed after 1, 2 and 6 days after the baseline time-point respectively (passage 2). Oxygen and carbon dioxide tensions as well as temperature were measured throughout the experiment. Real-time quantitative polymerase chain reaction (qRT-PCR) was performed on a 7500 Fast Real-Time PCR system (Applied Biosystems) using commercially available TaqMan Gene Expression Assays Quantitative RT-PCR was performed using assays for collagen type 1 alpha 1 (COL1A1), collagen type 2 alpha 1 (COL2A1), aggrecan (AGC), SOX9, and ankyrin repeat domain 37 (ANKRD37). Expressions are normalized to the two reference genes ribosomal protein L13a (RPL13A) and b2microglobulin (B2M) that have proven stable in hypoxic culturing of chondrocytes, and to the baseline expression. Data was analyzed by three-way ANOVA analyses. P-values lower than 0.05 were considered significant. Results: The hypoxic challenge of the cells were observed by a significant increase in ANKRD37 expression with lovering of oxygen tension. The cartilage-specific transcription SOX9 was significantly induced with lowering of oxygen, with no difference between the two culturing surfaces. COL2A1 and AGC expression increased with lowering of oxygen tension. The expression of the genes for these extracellular matrix proteins were higher in 3D culture after 6 days at all oxygen levels. However, this was only significant for COL2A1. Conclusions: These new results suggest combined positive effects of 3Dand hypoxic culturing on cartilage-specific gene expression in human chondrocytes. Moreover, we found a positive effect of seeding the cells onto the scaffold at least 6 days prior to implantation at all oxygen levels.

P16-827 Six years follow up of cartilage defects in the knee: patients improve their knee function score, but fail to recover completely Løken S.1, Heir S.2, Holme I.3, Engebretsen L.1, A˚røen A.1 1 Ullevaal University Hospital, Orthopaedic Centre and Oslo Sports Trauma Research Centre, Oslo, Norway, 2Martina Hansens Hospital, and Oslo Sports Trauma Research Centre, Bærum, Norway, 3Oslo Sports Trauma Research Center, Oslo, Norway Objectives: The natural history of focal cartilage injuries is largely unknown. We investigated the The functional outcome of patients with a known focal cartilage injury in their knee over a 6-7 years observation period.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: In a previous report of 993 knee arthroscopies 98 patients were less than 50 years of age and had an International Cartilage Repair Society grade 3-4 focal cartilage injury. Sixty-four patients had no cartilage surgery of the defect performed at baseline, except debridement, while 34 patients had cartilage surgery performed at baseline (microfracture n=21, autologous chondrocyte implantation n=7, osteochondral cylinder transfer n=2, fixation of osteochondral fragment n=4). During the follow-up period additional nine patients had cartilage surgery performed (microfracture n=3, autologous chondrocyte implantation n=4, osteochondral cylinder transfer n=2). Eighty-four of the 98 patients completed a follow up at average 6.2 years after baseline. The patients filled in the International Cartilage Repair Society Knee Evaluation form at baseline and at followup. In addition Lysholm score, and other functional knee questionnaires were filled in at follow-up. Weight-bearing radiographs were obtained from 68 patients and classified according to Kellgren and Lawrence. Results: The average International Cartilage Repair Society functional level, Visual Analog Scale pain score, and the patients’ functional rating of the knee compared to the contra-lateral knee all improved from baseline (p \ 0.001), while International Cartilage Repair Society activity level decreased (p\0.001) from baseline. At follow up the average Lysholm score was 75 (SD 20). The other functional scores were in the same level. In a linear regression analysis there were no association between any of the outcome scores and the following variables: Age, sex, BMI, area of the cartilage lesion, localization of the cartilage lesion, additional ligament/ meniscal surgery or whether cartilage surgery had been performed or not. The degree of radiographic osteoarthritis in the knees with cartilage defects was significantly higher compared to the contra-lateral knee (p\0.001). Conclusions: Patients with a known cartilage injury in their knee, with or without various treatments performed, can expect a stable or improved knee function over the next 6-7 years. Focal cartilage injuries seem to be associated with later development of radiographic osteoarthritis. Further studies are needed to compare the outcome of non-surgical and surgical treatment of cartilage lesions.

P16-838 Co-cultures of juvenile and adult human cartilage fragments: in vitro study for one-stage cartilage repair procedure Bonasia D.E.1, Martin J.2, Marmotti A.1, Rossi R.1, Buckwalter J.2, Amendola N.2 1 University of Torino, Mauriziano Umberto I, Torino, Italy, 2University of Iowa, Orthopaedic Surgery, Iowa City, United States Objectives: Treatment of articular cartilage lesions is a common issue in orthopaedics. One stage repair is a promising approach: repair process is promoted directly through a viable cell source in a resorbable scaffold without ex vivo cell expansion and without patients’ exposure to multiple procedures. The use of cartilage fragments has shown the matrix breakdown, the release of cells and good clinical results in one-stage cartilage repair. Juvenile chondrocytes have shown superior capabilities of producing cartilage extracellular matrix. This in vitro study is a proof of concept of using allogenic juvenile and autologous cartilage fragments as a viable cell source for a one-stage cartilage repair procedure. Our starting hypothesis was that adult and juvenile cartilage fragments in co-culture would produce a matrix biochemically and histologically superior to isolated adult cartilage cultures and inferior to juvenile ones. Methods: Adult cartilage was harvested from intraoperative pieces of three different patients: a) 62 year old female, affected by severe knee arthritis that underwent total knee replacement (WE); b) 69 year old female, affected by primary knee arthritis that underwent unicompartmental knee replacement (DD); c) 18 year old male, affected by hip osteochondral defect that underwent hip arthroscopy (GC). Juvenile cartilage source was ISTO technologies company (St. Louis, MO). All cartilage fragments were manually minced in order to obtain pieces smaller than 1 mm. The cartilage of every patient was cultured alone and with juvenile fragments. Juvenile cartilage as well was cultured alone. Each culture was made both with agarose and with a Hyaluronic-

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Acid Scaffold (Hyaff 11, Fidia Advanced Biopolymers, Italy). Two cultures per each type were made (total 28) in order to analyze them at 2 and 6 weeks. The cultures underwent biochemical (proteoglycan/DNA ratio) and histological (both Safranin O and collagen type II immunofluorescence staining) evaluation at 2 and 6 weeks. The DMMB method was used for glycosaminoglycan content. At 4 weeks a Confocal microscopic examination was conducted, without impairing the cultures. Results: At 2 weeks a chondrocytes migration out of the fragments was detectable in every culture and no statistically significant difference was noted biochemically and histologically between the different groups. At 4 weeks Confocal evaluation showed the creation of bridges between the cartilage fragments in every culture. At 6 weeks, co-cultures and juvenile cultures showed significantly better biochemical and histological properties, compared to isolated adult cultures. Even if not significant, co-cultures showed properties superior to isolated juvenile cultures. No significant differences were noted between agarose and Hyaluronic Acid scaffold. Conclusions: The null hypothesis was that in the co-cultures (1:1) the value of PG/DNA ratio and Saf-O positive cells would have been comparable to the mean value of isolated adult and juvenile cartilage fragments cultures. The hypothesis was refused and, surprisingly, in the co-cultures the values were significantly higher than expected. This indicates a stimulation of adult cells by juvenile ones. These results are promising for the development of one-stage cartilage repair with autologous adult and allogenic juvenile cartilage fragments. These findings support further investigation in vivo in an animal or human model.

P16-848 Arthroscopic fixation of matrix associated autologous chondrocyte implantation - joint compression forces following biodegradable pin fixation Herbort M.1, Zelle S.1, Rosslenbroich S.2, Petersen W.3, Zantop T.4 1 University of Muenster, Department of Traumatology, Muenster, Germany, 2Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 3Martin-LutherKrankenhaus, Unfallchirurgie, Berlin, Germany, 4Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany Objectives: Recently arthroscopic techniques for m-ACI fixation have been described. A fixation technique using biodegradable pin fixation can be used. However, arthroscopic pin insertion is technically demanding and a not perpendicular inserted pin may cause increased joint compression forces thus resulting in cartilage damage at the tibial plateau. We hypothesize that perpendicular insertion of a biodegradable pin for the fixation will not result in altered pressure distribution of joint compression forces when compared to the intact knee. Furthermore, it was hypothesized that an insertion 30 tilted will result in significantly altered increased joint compression forces. Methods: In 15 porcine knee joints the lateral femoral condyle was resected leaving the meniscus attachment intact and mounted in 30 knee flexion in a material testing machine (Zwick/Roell, Ulm/Germany). Joint compression forces were recorded using a digital pressure sensor (Novel, Munich/ Germany) underneath the medial meniscus and an axial compression of 100 N. The forces were recorded for the intact femoral condyle, standardized cartilage defect of 25 x 20 mm, after matrix associated ACI (BioSeed CTM, Biotissue, Freiburg/Germany) fixed in transosseous suture technique according to Erggelett and fixed in pin fixation using a biodegradable pin (SmartNail, Conmed Linvatec,16mm) perpendicular and 30 tilted to the matrix surface. Statistical analyses were performed using a two-factor repeated-measures analysis of variance (ANOVA). (p\0.05). Results: After creation of the standardized cartilage defect, the joint compression forces (824 kPa) were significantly increased when compared to the intact knee joint (564 kPa). After MACI implantation using the transosseous and perpendicular pin fixation, the joint compression forces of the cartilage defect was significantly decreased. No significant differences were found in maximal contact pressure between the intact, the transosseous fixed and perpendicular pin fixed matrices (p\0.05). Contact pressure was 564 kPa for the intact, 581,3 kPa for the transosseous fixed

S253 and 630,7 kPa for the perpendicular pin fixed matrices. Cartilage defect resulted in significantly increased joint contact pressure compared to the intact and reconstructed knees using transosseous and perpendicular pin fixation technique. Joint contact pressure forces after pin fixation using a 30 tilted insertion was a mean of 1740 kPa and significantly increased compared to all other groups. Conclusions: The results support our initial hypothesis and show that a perpendicular inserted pin for the fixation of m-ACIs does not result in significantly increased joint compression forces. Insertion of the matrix and fixation using a transosseous suture technique or perpendicular pin fixation reduced the joint compression forces to the values of the intact knee. A malplacement of the pin however, was shown to significantly increase the joint contact forces. To avoid a tilted insertion, the surgeon needs to keep the same angle of the aiming device during drilling and insertion of the pin. Additionally, the knee joint flexion needs to be identical during drilling and insertion.

P16-900 Predifferentiation does not foster quality of human MSC in osteochondral hybrid constructs stimulated by mechanoperfusion techniques Haasper C.1, Budde S.2, Richter B.2, Krettek C.1, Hurschler C.2, Jagodzinski M.1 1 Hannover Medical School, Trauma Department, Hannover, Germany, 2 Hannover Medical School, Orthopaedic Department, Hannover, Germany Objectives: Osteochondral lesions represent a common problem in orthopaedic surgery. A promising therapeutical approach is provided by the concepts of tissue engineering. The purpose was to examine whether a chondrogenic medium and/or mechanical stress is necessary to differentiate human bone marrow stromal cells (hBMSC) into chondrocyte-like cells in a biological osteochondral matrix stimulated in a bioreactor system. Methods: HBMSC were harvested, density centrifugation was performed prior to resuspension and three cell passages. For 7 days a predifferentiation culture was initated by replacing FGF-2 with 100 ng/ml IGF-1 and 5 ng/ml TGF-b2. Afterwards cells were seeded into the biologic hybrid scaffold out of CaReS (rat collagen I, Arthrokinetics) and Tutobone (bovine spongiosa, Tutogen Medical) with a concentration of 1x106 cells/ ml. The constructs were exposed to a cyclic compression protocol (10% compression, 0.5 Hz) under continuous perfusion in a mechano-bioreactor for 14, 21 and 28 days. Effects were evaluated using light microscopy and collagen 2, 3, and 10 staining. GAG and DNA were quantified. Biomechanical characterization was conducted using a confined compression quasi-static loading setup. Results: GAG/DNA quantification showed no differences between mechanical and static stimulation after three and four weeks (p\0.05). Mechanical tests showed no difference over four weeks, but the mechanical groups were stiffer compared to the static control (p\0.05). The quality of tissue was not improved by IGF/TGF in our system focusing histology, biochemistry and mechanical properties of these tissues. Conclusions: Media supplements to foster the quality of the tissue showed no progress in our system although it is well known that those are important to induce a chondrogenic phenotype. With regard to later clinical applicability and financial concerns, it could be postulated that stimulation time should be kept short and growth factors should be left out in such a system whereas cyclic compression enhanced matrix stiffness.

P16-943 Fresh osteochondral allograft of the knee Giannini S.1, Buda R.1, Fornasari P.M.2, Bevoni R.3, Ruffilli A.1, Cavallo M.1, Grigolo B.4, Vannini F.1 1 Istituto Ortopedico Rizzoli, Bologna University, VI Department of Orthopaedics and Traumatology, Bologna, Italy, 2Rizzoli Orthopaedic Institute, Muscoloskeletal and Tissue Bank, Bologna, Italy, 3Rizzoli Orthopaedic Institute, Bologna, Italy, 4Istituto Ortopedico Rizzoli, Laboratory of Immunology and Genetics, Bologna, Italy

123

S254 Objectives: Post-traumatic mono-compartmental and tri-compartmental arthritis of the knee in the young active patient represents a reconstructive challenge. Fresh osteochondral allografts are an increasingly popular option in the reconstruction of various amounts of cartilage defects, and although the role of partial monopolar allograft is well established in literature, concerns for both bipolar and total osteochondral allografts are reported. Purpose of this study is to describe the results obtained in a series of patients operated with partial and total allograft of the knee. Methods: 7 patients affected by mono-compartmental post traumatic knee osteoarthritis received partial (4 monopolar and 3 bipolar) allograft of the knee, while 8 patients affected by tri-compartmental post-traumatic arthritis underwent total osteochondral allograft. The mean age of the patients was 38.5 ± 10.2 years and the mean follow-up was 28 months (range 18-36). The ideal patient to allograft match was permitted through CT scan. Pre-operative and post-op results at 4, 6, 12 and at maximum follow-up were evaluated clinically and radiographically by X-Rays, MRI and CT scan. Two steps surgery, one for allograft preparing and one for the recipient site, were performed by using specifically designed jigs. Results: No intraoperative complications occurred. Satisfactory allograft consolidation and ROM were evident at 4-6 months. Partial allografts were allowed weight-bearing at 4 months, while total at 6-8 months in all the cases. Satisfactory results with normal ROM of the knee and regular gait with no pain and no need of support were resumed in all the partial allograft even bipolar and in 2 of the total, while a severe joint laxity developed at 10-16 months in 6 total allografts requiring arthroplasty. Histological evaluation of the samples showed cartilage degenerative changes with hypocellularity. Fibrocartilagineous aspect and positivity to extracellular matrix degrading enzymes were evident in the soft tissues. Conclusions: Both monopolar and bipolar partial allografts of the knee, were shown to be a reliable method of treatment for monocompartmental knee arthritis. Total knee allografts demonstrated high rate of failure. Although specifically designed jigs helped reproducible cuts and nice allograft fit, and good consolidation rates with no pain were obtained, the development of a total joint laxity and effusion produced unsatisfactory results. Immunological reaction to the large amount of transplanted tissue, may play a major role in the failure and should be deeply investigated.

P16-945 A novel route in bone tissue engineering: magnetic biomimetic scaffolds Russo A.1, Tampieri A.2, Bock N.3, Goranov V.4, Sandri M.2, Dionigi C.3, Riminucci A.3, Panseri S.1, Casino D.1, Shelyakova T.1, Dediu V.3, Marcacci M.1 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, 4 Belarussian State Medical University, Minsk, Belarus Objectives: Tissue engineering approaches in osteochondral and long bone defects employ three-dimensional scaffolds to provide the necessary support for the defect site and for the cells to proliferate and maintain their differentiated function. Nowadays severe defects remain unsolved problems, and the complete histomorphological and biological maturation of tissues is achieved only if angiogenesis is permanently stimulated by angiogenic factors. The proposed innovative approaches involve the use of nanoparticles in magnetically controlled delivery of selected angiogenic growth factors. Methods: Magnetic scaffolds are prepared following two different methods, both based on biologically inspired synthesis involving biomineralization in vitro of collagen. Apatite/collagen porous scaffolds are prepared by nucleating biomimetic apatite on self-assembling collagen fibres and then, infiltrating these bio-hybrid composites with ferrofluids (aqueous dispersions of magnetite nanoparticles coated with biopolymers). The magnetic nanoparticles are therefore entrapped in the construct leading to the magnetization of the scaffold. Magnetite functionalized with

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 biocompatible coatings are chosen because of their superparamagnetic behaviour at room temperature. The second approach is based on the direct nucleation of biomimetic apatite on self-assembling collagen fibrils in presence of magnetite nanoparticles, thus realizing the magnetization of the scaffold material in situ. Results: The stability of the magnetic scaffolds is assessed in contact with simulated body fluids. The analysis revealed an extremely low percent of released magnetite after 8 days in simulated body fluid under constant flow. The scaffolds became magnetic (17 emu/g) maintaining their specific porosity and shape. Magnetic scaffolds were tested in vitro with mesenchymal stem cells from human bone marrow. The preliminary results showed no toxicity of the magnetic scaffolds. They retained the ability to support cell adhesion, growth and proliferation. Throughout the whole cultured period (40 days), there were no apparent differences in the dynamics of cell growth between the non-magnetized scaffolds and the magnetic scaffold. Conclusions: The proposed scaffolds work like a magnetic local field amplificator: their relatively strong internal magnetization can be aligned in the same direction by relatively weak external field. This approach will activate a continuous interaction and material exchange of the scaffold with the external supplier under the effect of magnetically guided nanoparticles. In this way the scaffold will be manipulated in vivo over a long period of time thus controlling and directing where and when appropriate optimal tissue regeneration. The process of reloading of the scaffold will be based on the use of variously biofunctionalized magnetic nanoparticles with selected growth factors and stem cells and to act as shuttles that can transport these growth factors towards the static scaffold, leave them in/on/at the scaffold and, if required, being removed from it. At the same time this approach provide the fascinating possibility to achieve efficient scaffold fixation via magnetic forces providing a very elegant and simple solution to the problems of fixation that many scaffolds meet.

P16-953 Second-generation autologous chondrocyte implantation in the treatment of the knee cartilage defects. Findings at five years Podsˇkubka A.1, Vaculı´k J.1, Povy´sˇil C.2, Masˇek M.3, Sˇprindrich J.4 1 Bulovka University Hospital, Orthopaedic Clinic 1st Medical School Charles University, Praha, Czech Republic, 21st Medical School Charles University, Institute of Pathologic Anatomy, Praha, Czech Republic, 3 Bulovka University Hospital, Clinic for Radio-diagnostics, Praha, Czech Republic, 43rd Medical School, Clinic for Radio-diagnostics, Praha, Czech Republic Objectives: From November 2003 to January 2005 we implemented clinical trial in the treatment of deep chondral defects of the knee using ACI with 3D hyaluronan-based scaffold (Hyalograft C). In the prospective study we followed and evaluated functional, macroscopic, histological and MRI outcomes five years after operation. Methods: 11 patients were treated. We evaluated results in 9 patients, eight men and one woman, with an average age 31 years. Defects of an average size 3.9 cm2 (2-6) were localized on femoral condyles. We evaluated functional outcomes according to IKDC, KOOS, Lysholm, Tegner score and performed MRI examination prior to and after ACI (9-12, 30-36 and 60 months). The newly formed tissue was controlled visually (ICRS visual score) at second-look arthroscopy 9 to 12 months after ACI. Specimens for histological analysis were harvested from the site of implanted chondrocytes. Results: The average IKDC subjective score improved from 46 points before surgery to 74 points, 77 points and 85 points at the interval between 9 and 12 months, 30 and 36 months and 60 months after surgery respectively. The Lysholm score was 61 points preoperatively, 83 points at the interval between 9 and 12 months after operation, 86 points at the interval between 30 and 36 months after operation and 91 points 60 months after operation. Similar improvement of the KOOS was also noted. The average Tegner score before ACI was 3 (2-4). 5 years after ACI the average score was 6 (4-7).

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 At second-look arthroscopy the newly formed cartilage was macroscopically evaluated as nearly normal in 7 cases and as abnormal in 2 cases. The average ICRS visual score was 9.4 points. The newly formed tissue had histological characteristics of hyaline-like cartilage in 1 patient and characteristics of mixed (hyaline/fibro) cartilage in 8 patients. In the interval between 30 and 36 months after ACI control assessment with standardized cartilage sensitive MRI evaluating the integrity, morphologic features and signal of the articular surfaces was performed. Changes in the subchondral bone were also evaluated . The surface of the newly formed cartilage was smooth in 6 patients. In the remaining patients it was slightly irregular. In 5 patients the new cartilage protruded slightly above the level of the surrounding cartilage. In the majority of patients residual irregularity of subchondral bone plate was found. 60 months after ACI in all 9 patients a bony substrate (osteophyte) of various size at the base of the original defect could be found. The bony prominence was covered by a thin layer of cartilage (1-3 mm) that was at the same level with the surrounding cartilage. In one case the cartilage was absent above a segment of the osteophyte. Small bone oedema was noted in one patient. Conclusions: In the interval from 9 to 12 months Hyalograft C leads to good filling of defects with newly formed tissue. Mainly mixed hyaline/ fibro tissue adhered well and was formed in less than a year. The scaffold resorbed without problems. Significant functional improvement could be seen 1, 3 and 5 years postoperatively. MRI findings 5 years after operation were unexpected. The bony outgrowth at the base of the original cartilage defect found in all patients is not an incidental finding and its cause is unclear. Long term results remain uncertain.

P16-976 A correlation of clinical outcome scores and MRI/CT scan findings up to 24 months following insertion of a PLA/PLG scaffold implant for osteochondral defects of the knee Melton J.1, Wilson A.1 1 North Hampshire Hospital, Trauma and Orthopaedics, Basingstoke, United Kingdom Objectives: This study aims to characterize the early imaging characteristics and appearances of TruFit CB (Smith and Nephew) osteochondral scaffold implants on plain radiographs, magnetic resonance imaging and computed tomography (including CT arthrogram). Methods: We analysed the early imaging of 16 implants in nine patients who had undergone implantation of TruFit CB plugs in the knee. The median age of the patients was 45 years at the time of implantation (range 39-72) and mean length of follow up at the time of imaging was 12 months (8- 24 months). Plain radiographs and MRI scans were obtained for all patients up to two years after implantation and the scan appearances are presented. The imaging characteristics are correlated to clinical outcome scores (OKS and KOOS). Results: The T2 weighted images consistently show high signal from the implanted scaffold up to two years. The T1 weighted images are presented which are said to be the best sequences for evaluation of incorporation of the implant. Computed Tomography scans are presented which show that in some cases there appears to be no tissue apparent on CT scan despite the appearance of tissue ingrowth on T1 weighted MRI scan. At two years one patient appeared to have no tissue ingrowth on CT scan and this correlated with poor clinical outcome scores. Conclusions: Scaffolds are designed to allow the regeneration of chondral tissue when implanted and eventual incorporation. This is new technology and the gold standard for radiological evaluation of these implants is not yet established. We suggest that CT scan (with or without arthrogram) is a useful adjunct to MRI scanning for follow up of these implants and may be better for assessing the chondral surfaces and incorporation of the implant. We also suggest that radiological studies with longer follow-up are conducted to ensure that these implants are eventually incorporated.

S255 P16-986 Knee arthroscopy outcome for patients over 70 years of age Malagelada F.1, Ballester M.2, Jimenez Obach A.1, Badia J.M.1, Lopez de Vega J.3 1 Hospital de Mataro´, Mataro´, Spain, 2Hospital de Sant Rafael, Orthopaedic, Barcelona, Spain, 3Hospital de Mataro, Orthopedic Surgery, Barcelona, Spain Objectives: To evaluate the future outcome of patients over 70 years old who had undergone a knee arthroscopy with different findings and their results and/or need for extra surgery after follow up. Methods: We present 50 patients over 70 years old who had undergone a knee arthroscopy, with a minimum of 10 year follow-up. We collected the data retrospectively from their clinical notes and intra-operative findings. We analyzed their previous clinical condition and radiographic findings, evaluating the osteoarthritis grade by Kellgren-Ahlba¨ck classification and the cartilage damage during the arthroscopie using the Outerbridge scale. After arthroscopic treatment, we measured patient’s postoperative satisfaction, and time needed for another knee procedure (i.e. another arthroscopie, total knee replacement, osteotomy). Results: Most of the patients experienced substantial pain relief at the first control one month after the surgery, and most of them were able to return to their daily activities at that time. The most frequent findings in arthroscopy were cartilage damage and meniscal degeneration. Other less frequent results were condrocalcinosis, free bodies and sinovial plica. The commonest procedure performed after arthroscopie was total knee replacement. Conclusions: Our study shows that this procedure is successful in selected cases. Results depend mainly on the grade of articular degeneration. In the short term we experienced good results, but some clinical worsen has been seen at long time follow up.

P16-1016 Human bone marrow and adipose tissue derived mesenchymal stem cells chondrogenic differentiation Havlas V.1, Kos P.1, Jendelova P.2, Trc T.1, Lesny P.3, Sykova E.4, Danisevic L.5 1 Charles University Prague, Department of Orthopaedics and Trauma, Prague, Czech Republic, 2Center for Cell Therapy and Tissue Repair, Charles University Prague, Prague, Czech Republic, 3Institute of Experimental Medicine, Prague, Czech Republic, 4Department of Neuroscience, Prague, Czech Republic, 5Center for Cell Therapy and Tissue Repair, Prague, Czech Republic Objectives: Treatment of patients with cartilage damage or degeneration is difficult because mature cartilage has a limited regenerative potential. Small cartilage defects are regenerated by the migration of chondrocytes, while full-thickness damage heals by the formation of inferior fibrocartilage; in many cases osteoarthritis develops. Cell therapy using chondrocytes has some disadvantages: chondrocytes undergo a dedifferentiation process when expanded in vitro, gradually change their morphology and the production of type II collagen is replaced by the production of collagen type I. Recently, stem cell therapy has offered new possibilities for solving this problem. The sources of adult mesenchymal stem cells (MSCs) include bone marrow stromal cells (BMSCs) and mesenchymal adipose tissue-derived cells (MADCs). MSCs are characterized by self-renewal ability and are capable of differentiating into different cell lineages under proper in vitro conditions. The aim of our study was to verify the in-vitro chondrogenic differentiation potential of human BMSCs and MADCs in the presence or absence of transforming growth factor beta (TGF-beta 1). Methods: Human BMSCs and MADCs were collected from healthy donors during orthopaedic surgeries, expanded in vitro and sub-cultured for three passages to obtain a sufficient quantity of cells. For chondrogenic differentiation, a pellet culture system was used. Pellets were formed by centrifugation of 29106 cells in 15 ml polypropylene tubes. Chondrogenic medium consisted of DMEM-F12, 10% FBS, 100 U/ml penicillin, 100 lg/ml streptomycin and TGF-beta 1. Control pellets were cultured in the same

123

S256 culture medium but without the addition of TGF-beta 1. After three weeks, pellets were fixed, embedded in paraffin and cut into 5 micrometer thick sections for histological analysis and immunohistochemistry. Cartilagespecific gene expression was examined by real-time PCR. Results: We observed spontaneous chondrogenic differentiation in all of the pellets. The condensation of pellets cultured in chondrogenic medium and in medium without TGF-beta 1 into single aggregates was observed after 24 hours. After 21 days of culture, aggregates cultured in chondrogenic medium were larger than aggregates cultured in control medium; the content of glycosaminoglycans and collagen was higher in the pellets cultured in chondrogenic medium. Real-time PCR confirmed the production of collagen type II and aggrecan in all tested groups. Conclusions: The results from this study demonstrate that both the human bone marrow and the adipose tissue-derived MSCs have chondrogenic potential in vitro and therefore both types are likely to play an important role in the future cartilage repair engineering.

P16-1017 Characterization of Tet-regulated, lentivirally mediated BMP-2 expression in chondrocytes: a new gene therapy approach in cartilage healing Vogt S.1, Wu¨bbenhorst D.2, Dumler K.2, Wexel G.1, Imhoff A.1, Gansbacher B.2, Anton M.2 1 TU Munich, Orthopaedic Sport Medicine, Mu¨nchen, Germany, 2 TU Munich, Experimental Oncology, Mu¨nchen, Germany Objectives: Therapy of cartilage defects is still challenging due to poor selfhealing capacity. A gene and cell therapeutic approach for in situ production of growth factor BMP-2 by implanted autologous chondrocytes might be advantageous. To allow for regulation of gene expression the Tet on System was chosen, delivered by VSV-G pseudotyped lentiviral vectors. Methods: Experiments were carried out on primary rabbit chondrocytes. Cells were coinfected with two vectors, expressing the reverse Transactivator (rtTA) and eGFP or BMP-2 under control of the Tet-responsive element (TRE), respectively. Transgene expression was induced by doxycycline. Results were obtained by FACS analysis for eGFPexpressing samples or BMP-2 ELISA for BMP-2 expressing samples. Synthesis of proteoglycans was determined by alcian blue staining. Results: Induction of eGFP expression in chondrocytes was possible. The highest induction rate and gene expression was achieved when using the spleen-focus-forming-virus LTR promoter to drive the reverse transactivator expression after addition of doxycycline. The induction and expression levels for BMP-2 in chondrocytes were similar comparing oneand two-vector system infected cells. In addition BMP-2 expression was repetitively inducible. An increase in proteoglycan production demonstrated functionality of the expressed BMP2. Conclusions: The lentivirally mediated tet-on system demonstrated efficient, repeatedly inducible expression of BMP-2 in primary rabbit chondrocytes. Therefore, this system is promising for further in vivo experiments concerning the treatment of cartilage defects.

P16-1029 PRP intra-articular injection and viscosupplementation as therapeutic treatments for early osteoarthritis: multicentre retrospective cohort study in 150 patients at 6 months follow up Kon E.1, Buda R.2, Mandelbaum B.3, Filardo G.1, Delcogliano M.1, Di Martino A.1, Timoncini A.2, Giannini S.2, Marcacci M.1 1 Rizzoli Orthopaedic Institute, Biomechanics Lab., Bologna, Italy, 2 Istituto Ortopedico Rizzoli, Bologna University, VI Department of Orthopaedics and Traumatology, Bologna, Italy, 3Santa Monica Orthopaedic and Sports Med Research Foundation, Santa Monica, United States Objectives: The influence of the growth factors on cartilage repair is not yet widely studied and its application in clinics is still experimental. Platelet Rich Plasma (PRP), a blood derived rich in growth factors, is a

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 promising method for treatment of cartilage defects. Aim of this study is to evaluate and compare the efficacy of PRP and Viscosupplementation (HA) i.a. injections for treatment of severe chondropathies of the knee. Methods: The study involved 150 patients affected by chondropathy, early Osteoarthritis and severe Osteoarthritis. 50 symptomatic patients were treated with 3 autologous PRP intra-articular injections and evaluated prospectively. For PRP production 150 ml of venous blood were collected from each patient and underwent a double centrifugation, increasing platelet concentration. All patients were clinically evaluated at the enrolment, at the end of the treatment, at 6 months follow up. The results obtained were also compared with two homogeneous group of patients treated in two different centers with HA injections, one group was treated with injections of High Molecular Weight Hyaluronan and the other group was treated with Low Molecular Weight Hyaluronan. IKDC and EQ-VAS scores were used for clinical evaluation and patient satisfaction and functional status were also recorded. Results: The comparison between the outcomes of the three groups was statistically significative (p\0,0005), reporting a superiority of PRP group at any times of F-up. Conclusions: Autologous PRP injections demonstrated more and longer efficacy than HA injections in reducing pain, symptoms and recovering articular function in patients affected by severe chondropathies of the knee.

P16-1039 Novel nano-composite biomaterial for osteochondral tissue engineering: pilot clinical study at 2 years follow up Kon E.1, Delcogliano M.1, Di Martino A.1, Filardo G.1, Zaffagnini S.1, Marcacci M.1 1 Rizzoli Orthopaedic Institute, Biomechanics Lab., Bologna, Italy Objectives: Current surgical techniques to repair osteochondral defects lead to poor subchondral bone regeneration and fibrocartilage formation, often associated with joint pain and stiffness. From a surgical and commercial standpoint, an ideal graft for osteochondral defect repair would be an off-the-shelf product; thus, some new biomaterials were recently proposed to induce ‘‘in situ’’ cartilage regeneration after direct transplantation onto the defect site. We performed this clinical pilot study where a newly developed biomimetic scaffold was used for the treatment of chondral and osteochondral lesions of the knee joint in order to evaluate the safety and the reproducibility of the surgical procedure and to test the intrinsic potential of the device without any other cells culture. Methods: A gradient composite O.C. scaffold, based on type-I collagenHA, was obtained by nucleating collagen fibrils with hydroxyapatite nanoparticles at physiological conditions30 cases (9F, 21M, mean age 29,3 years) with knee osteochondral lesions (8 medial femoral condyle, 5 lateral condyle, 12 patella, 8 femoral troclea) were treated with scaffold implantation from January 2007 to July 2007. The lesions size went from 2 cm2 to 6 cm2. The clinical outcome of all patients was analyzed prospectively, at 6 moths at 1 year and 2 years using the Cartilage standard Evaluation Form as proposed by ICRS and an high resolution MRI . Results: Statistical analysis demonstrated a significant improvement (Non Parametric paired Wilcoxon test, p\0.0005) from pre-operative to 12 and 24 months follow up. IKDC objective score showed preoperatively 46.1% of normal or nearly normal knees and 79.3% of normal or nearly normal knees at 24 months. Statistical analysis showed a significant improvement in the IKDC subjective score from pre-operative (37,5±14,6) to 24 months follow up (82,4±11,9) (Non Parametric paired Wilcoxon test, p\0.0005). MRI evaluation in at 12 months was analyzed according to the MOCART scoring scale. The complete filling of cartilage defect was noted in 86.2% of the patients and the congruency of the articular surface was seen in same patients.1 case failed and were reoperated. Conclusions: This open one-step surgery was used for the treatment of chondral and osteochondral defects. The results of this technique at short follow-up are very encouraging and show satisfactory results even in big ostechondral defects.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P16-1048 Gene and microRNA expression in uncultured articular chondrocytes, proliferating and hyaline matrix producing articular chondrocytes and in de-differentiated and proliferating articular chondrocytes Karlsen T.A.1, Shahdadfar A.1, Brinchmann J.1 1 University of Oslo/Rikshospitalet-Radiumhospitalet, Institute of Immunology, Oslo, Norway Objectives: Human articular chondrocytes (AC) can be obtained for analysis from three different phases: 1) uncultured AC (day 0), 2) proliferating and hyaline matrix producing AC cultured in their own ECM (day 7-14), and 3) de-differentiated and proliferating AC in monolayer culture (day 28). The objectives of this study were to assess the gene and microRNA (miRNA) expressions in these different phases. Methods: Articular cartilage was cut to pieces and digested with collagenase for 90-120 min. The partly degraded cartilage structures were established in culture. Real-time PCR analysis was performed to determine the mRNA levels of a large number of relevant molecules. Global arrays were performed to assess the miRNA changes between the different phases and flow cytometry was used to analyze expression of cell surface molecules and intracellular levels of ECM molecules. Cartilage from three donors was included in the study. Results: Uncultured cells were characterized by moderate levels of mRNAs for COL1, 2, 9, 10 and 11 and matrix molecules aggrecan, COMP and decorin. Transcription factor SOX9 mRNA was high, while SOX5 and 6 mRNA was much lower. Moderate levels of matrix degrading enzyme MMP3 was also found. In phase 2 the mRNA levels of COL1, 2, 9 and 11, versican and HAS2 all increased, while COL10 was unchanged or decreased. Aggrecan, COMP and decorin mRNA were unchanged. SOX9 mRNA was reduced, while SOX5 and 6 were greatly increased. mRNAs for MMP1, 3 and 13 all increased considerably. At dedifferentiation (phase 3) mRNA for COL2,9 and 10 dropped dramatically, COL11 and aggrecan remained unchanged while COL1 and versican mRNAs increased. mRNA for SOX5.6 and 9 all dropped, as did MMP1,3 and 13.mRNA for alkaline phosphatase and RUNX2 were increased. miRNA arrays revealed 4 different miRNA expression profiles (P \ 0.01): upregulated in uncultured cells only, upregulated in COL2 producing cells (phase 1 and 2), upregulated in proliferating cells (phase 2 and 3) and upregulated in dedifferentiated cells only (phase 3). Validation of one miRNA from each profile was performed using real-time RT PCR. Expression pattern was very similar to that obtained from miRNA arrays. Flow cytometry analysis showed different expression of integrins between day 0 and 28. Intracellular staining of ECM molecules confirmed the mRNA levels. Conclusions: Expression of many genes and miRNA changed during in vitro culture. However, AC can maintain aggrecan, COL2, 9 and 11 production for 14 days if cultured in their own ECM. Ongoing studies are in progress to see if these AC produce hyaline cartilage better than that seen with dedifferentiated AC. The importance of the miRNA changes is unknown but functional studies are in progress.

P16-1055 Correlation of dynamic blunt impact testing, histopathology and visual macroscopic assessment in human osteoarthritic cartilage Kos P.1, Varga F.2, Handl M.1, Kautzner J.1, Chudacek V.3, Drzik M.4, Povysil C.5, Trc T.1, Amler E.2 1 Charles University in Prague, Department of Orthopaedics and Traumatology, Prague, Czech Republic, 2Charles University in Prague, Department of Biophysics, Prague, Czech Republic, 3Czech Technical University in Prague, Department of Cybernetics, Prague, Czech Republic, 4 International Laser Center, Bratislava, Slovakia, 5Charles University in Prague, Institute of Pathological Anatomy, Prague, Czech Republic Objectives: Articular cartilage has unique mechanical properties due to its composition that optimize transmission of load to subchondral bone. Degradation of cartilage resulting from osteoarthritis influences its functional properties. Improved staging of cartilage degeneration is required for prognosis and early treatment, particularly in low grades when minimal surface

S257 damage is visible. We examined correlation of dynamic mechanical properties of cartilage with histological findings and macroscopic visual score. Methods: 14 patients were included in our study (mean age 71.2 years (5684); 5 men, 9 women; 4 right, 10 left knees). Osteochondral samples (6 mm diameter, cylindrical) were obtained during total knee joint replacement from lateral and medial condyle of femur and tibia. 2 samples were harvested from the same place for histological and biomechanical testing. Prior to mechanical testing, samples were macroscopically classified following International Cartilage Repair Society (ICRS). Dynamic blunt impact testing was used, response to a single impact evaluated. Parameters of loading resembled physiological cartilage loading. Sample deformation was read simultaneously by a sensitive piezoelectric accelerometer and laser Doppler vibrometer. Relation of acting force vs. deformation was expressed by loading diagrams. Normal stress, dissipated energy, tangent modulus and stiffness at 1 MPa stress were evaluated. Histological examination was performed on formalin-fixed parafin-embedded decalcinated samples. Slides were stained using hematoxylin-eosin, classified following ICRS visual histological scale. Results: Regarding individual histological ICRS features, significant differences were found in dissipated energy values as well as related specific damping capacity for features 1, 2 and 3 (cartilage surface, extracellular matrix, cell distribution). No other mechanical characteristics correlated significantly with the histological score (T-test, p[0.05). Significant correlation was found between the macroscopic ICRS score, specific damping capacity and dissipated energy (ANOVA, p\0.01). Neither stiffness nor elasticity modulus values followed the macroscopic classification. Mean values of specific damping capacity increased between macroscopic grades 0-1 and 1-2 by 11.3% and 18.6% respectively. It implies that this material characteristic could serve for distinguishing between early stages of cartilage deterioration. Though no correlation between cartilage sample thickness and total histological score was proved, strong relation was found between energy dispersed in unit volume - relative dissipated energy and thickness (p\0.001, R2=0,69). Conclusions: Among mechanical quantities characterizing cartilage, those dealing with part of compressive energy lost within the tissue were so far neglected. The current standard for staging cartilage degeneration is histology requiring invasive sample harvesting and therefore is not suitable in vivo. Clinical cartilage status examination is based on arthroscopic visual evaluation of the articular surface that gives no information about the tissue morphology. In cartilage with low grade of macroscopic degradation (ICRS 0-1) histopathologic findings varies a lot. Targeted treatment esp. of low grade arthritis demands on early diagnosis and requires improved tools for less invasive and more exact classification of cartilage status. According to our results, mechanical testing evaluating energy dissipation appears to be a promising approach.

P16-1060 Use of an unloader brace for medial or lateral compartment osteoarthritis of the knee Briggs K.1, Matheny L.2, Herzog M.3, Watts C.3, Steadman J.4 1 Steadman Hawkins Research Foundation, Clinical Research, Vail, United States, 2Steadman Research Foundation, Clinical Research, Vail, United States, 3Steadman Hawkins Research Foundation, Vail, United States, 4 Steadman Hawkins Clinic, Knee Surgery, Vail, United States Objectives: The purpose was to determine if there was a difference in factors affecting outcomes when comparing medial versus lateral unloader braces. Our hypothesis was that patients with unloader braces will have improved disability(WOMAC score) over 6-months. Methods: Forty-six patients(17 females,29 males), with mean age of 61 years(range:48-87), had standard AP and long-standing radiographs taken prior to brace fitting. Patients completed WOMAC score prior to brace use, 6-weeks, 6-months. Fourteen patients were fitted with a lateral unloader and 32 with medial unloader. Results: There was no difference in average age between medial and lateral braced patients(both=61;p=0.9). Mean medial joint space for medial group was 2.4mm(range:0-4.2) and mean lateral joint space was 4.2mm(range:2.08.0) for lateral braces. Percent deviation from neutral alignment was 26%(medial) and 19%(lateral). Pre-brace pain was greater in medial group

123

S258 compared to lateral(8vs.5;p=0.008). Medial group also had higher pre-brace total WOMAC(34vs.23). Both groups showed significant improvement in total WOMAC from pre-brace to 6-weeks and 6 months. At 6-weeks and 6 months there was no difference in total WOMAC between groups. This showed more improvement for medial(12points) in WOMAC compared to lateral group(9points). For lateral group, 6-month total WOMAC was strongly correlated with deviation from neutral alignment(r=0.847). For medial group, pre-brace total WOMAC correlated with medial joint space(r=-0.561), as did 6-week total(r=-0.61) and 6-month total WOMAC(r=-0.496). WOMAC function subscale showed the strongest correlation(r=-0.621). Conclusions: Few differences were seen between medial and lateral unloader bracing. Both groups significantly improved. Medial group had more disability prior to brace use and more improvement.The smaller the medial joint space, the more disability in medial group. Lateral group showed more disability in patients with more varus alignment.

P16-1070 Evaluation of hylastan SGL-80 in symptomatic osteoarthritis (OA) of the knee Housman L.1, Arden N.2, Birbara C.A.3, Conrozier T.4, Waddell D.5, Wei N.6, Yates J.7 1 Tucson Orthopaedic Institute, Tucson, United States, 2Southhampton General Hospital, Southhampton, United Kingdom, 3Clinical Pharmacology Study Group, Worcester, United States, 4Centre Hospitalier Lyon-Sud, Pierre Benite Cedex, France, 5Orthopedics Specialists of Louisiana, Shreveport, United States, 6Arthritis and Osteoporosis Center of Maryland, Fredrick, United States, 7Coastal Orthopedics Associates, Conway, United States Objectives: A single injection of hylastan SGL-80 (JonexaTM, Genzyme Biosurgery, Cambridge, MA), provides long lasting OA knee pain relief and is a safe, convenient alternative to multiple-injection viscosupplements. Methods: Currrently available viscosupplements provide safe and effective therapy to treat knee OA, but are typically administered via multiple injections. This study assesses 2 intra-articular injection regimens of hylastan SGL-80 (H-80). 391 patients participated in this multicenter, randomized, double-blind study; 130 received 1 injection of H-80 and 129 received 2 injections, 2 weeks apart. 132 received methylprednisolone as an active comparator. Percent improvement in pain (WOMAC A) from baseline at 26 weeks was determined. At Week 26, qualifying patients were randomized again and re-treated with H-80. Safety was assessed by the incidence of treatmentrelated adverse events (AEs). Results: Demographic variables were balanced between study groups at randomization. Baseline Kellgren-Lawrence (KLG) OA severity was reported: 29.2% patients had KLG II and 65.9% had KLG III. After initial treatment, (n=391) the onset of pain relief was similar and percent improvement from baseline at 26 weeks was statistically significant (p\0.0001) in all groups. Improvements of 34.5% with 1 injection H-80, 35.6% with 2 injection H-80 and 34.6% with steroid were observed. After re-treatment, (n=201) statistically significant improvements (p\0.0001) in pain from baseline to re-treat Week 26 were observed for both H-80 arms. AEs were comparable among all treatment groups. Most were of mild or moderate severity. The incidence/severity of AEs did not increase after retreatment. There were no study treatment-related target knee serious AEs. Conclusions: Both H-80 treatments were well tolerated and comparable to steroid. A single injection of H-80 provided long-lasting pain relief and a safe, convenient alternative to multiple injections.

P16-1084 ACL deficiency, meniscal and cartilage tears. A minimum 5 years follow-up in athletes Osti L.1, Papalia R.2, Del Buono A.2, Maffulli N.3, Denaro V.2 1 Hesperia Hospital, Orthopedic Surgery, Modena, Italy, 2Campus Biomedico University, Orthopaedic Surgery, Rome, Italy, 3Keele

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 University, Institute of Science and Technology in Medicine, Stoke on Trent, United Kingdom Objectives: To compare the postoperative (minimum 5 years) clinical and radiological outcomes of 50 patients, similar for ACL rupture and meniscal tears, but different for grade cartilage lesions. Methods: Group 1 included 25 patients undergoing microfracture management of grade III-IV cartilage lesions, Group 2 included 25 patients with grade I-II cartilage lesions, managed by radiofrequency. Preoperatively, at 2 years and minimum 5 years follow-up, Lachman test, pivot shift test, and KT-1000 side to side difference were used to assess ACL stability; IKDC, Lysholm score and Tegner activity scale were used to rate functional status. At the latest follow-up, symptoms of knee osteoarthritis were assessed using the WOMAC OA index. Results: Comparing pre- and post-operative status, Lachman test, pivot shift values, and KT 1000 side to side difference measurements improved significantly (\.001) in both groups, with no statistically significant difference between the two groups ([.05) at the intermediate and latest assessments. At both postoperative appointments, in both groups, the average Lysholm score and IKDC ranking rates improved significantly (\.001) compared to preoperative values, but slight worsening was observed in Group 1 patients at the latest review. Concerning the WOMAC index score and sport activity level rating, Group 1 patients had significantly lower scores than Group 2 patients (p \.05). Conclusions: In patients with symptomatic ACL instability combined to a single deep cartilage lesion less than 2 cm2 in diameter, microfracture is a safe, with excellent short term clinical and functional improvement, providing excellent rates in return to pre-injury sport activity level. A slow rehabilitation protocol it is mandatory to reach good results.

P16-1087 Comparative analysis by finite element method on a theoretical model between 6 mm and 8 mm diameter plug osteochondral autologous tansplantation Fleaca R.1, Roman M.1, Oleksik V.2, Pascu A.2, Deac C.2, Baier I.1, Stanciu T.3 1 University of Sibiu, Orthopedic and Trauma Surgery, Sibiu, Romania, 2 University of Sibiu, Faculty of Engineering, Sibiu, Romania, 3University Hospital Sibiu, Orthopedic and Trauma Surgery, Sibiu, Romania Objectives: The purpose of this study is to analyze the biomechanical behavior on axial load of a distal femur with 6 mm and 8 mm plug osteochondral autologous transplantation on a theoretical model using the Finit Element Method. Methods: For this evaluation we have harvested a distal femur and a proximal tibia from a fresh cadaver. After all remaining soft tissue was dissected free and the distal femur was degreased the bone surface was covered with a white mate powder and marked with several red points. Then we have scanned the femur with the 3D the NextEngine Scan and medium meshed (discretisated) its body. The same method was utilised for the proximal tibia. The next step was to computed assemble the two scanned bones and than to import the whole femur-tibia assemble in the Ansys soft. Then we have simulated several osteochondral autologous transplantations on the medial femoral condyle with plugs 20 mm long and 6 and 8 mm diameter harvested from the femoral trochlea. We have created four different simulations: one model for the equivalent of a chondral defect of approximate 2 cm2 with 6 mm, another one with 8 mm diameter plugs, and other two models for 4 cm2 chondral defect also with 6 and 8 mm plugs. We have then applied for each model separately a force of 1.200 N along of the mechanical axis of the femur cranio-caudal, considering fix point the section level of the tibia diaphysis. Then we have performed the analysis by Finite Element Method, focusing on the behavior of the transplant area. The data obtained allowed the analysis of the Maximum Principal Stress, Equivalent Von Misses Elastic Stress, Equivalent Elastic Strain and Total Deformation. Results: The comparative analysis between the models with plug diameter of 6 and 8 mm witch covers equivalent chondral defect areas of 2 cm2 and 4 cm2 shows that the Equivalent Elastic Strain of the distal femur was obtained in all models at the level of the chondral surface of the transplant and has values with approximate 0,002 mm/mm higher in 6 mm than in

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 8 mm plugs diameter. The Maximum Total Deformation has maximum values around 2,2 mm obtained at the level of the femoral diaphysis and was minimal at the level of the articular surface of the condyles. Maximum Principal Stress was also obtained at the surface of the osteochondral transplant for all simulations and has values of approximate 68 Mpa for the 8 mm plugs and more than 111 Mpa for the 6 mm diameter plugs. The Equivalent Von Misses Elastic Stress was maximum at the level of the femoral condyles and with 50 Mpa higher in area corresponding simulations with 6 mm plugs than in the ones with 8 mm diameter plugs. Conclusions: The results of this analysis shows that the maximum principal stress and the total deformation have similar behavior in all models, with values slightly increased in the models with 6 mm comparative with 8 mm plugs. A significant difference was obtained for the equivalent elastic stress and the elastic strain between the 6 and 8 mm plug diameter. These results demonstrate the importance of the primary plug stability and also the importance of the anatomical reconstruction of the articular surface. Acknowledgments: This work was carried out within the framework of the research grant named ‘‘Theoretical and experimental analysis of the static and dynamic behavior of the grafts in the autologous osteochondral transplantation’’ support by the Romania Ministry of Education and Research and National University Research Council.

P16-1153 Clinical outcome after femoral condyle transplantation using large frozen osteochondral allografts (Mega OATS transplants) Said S.1, Christiansen S.E.1, Lund B.2, Faunoe P.1, Lind M.3 1 Aarhus University Hospital, Sportstrauma Division, Dept of Orthopedics, Aarhus, Denmark, 2Division of Sports Trauma, Department of Orthopedic Surgery, Aarhus, Denmark, 3Aarhus University Hospital, Sports Trauma Clinic, Aarhus, Denmark Objectives: Large osteochondral lesions in the knee joint are a challenging clinical problem. Fresh osteochondral allograft transplants are used to replace tissue in large symptomatic osteochondral defects in young, active patients in United States and Canada. In Europe there is mainly access to frozen osteochondral grafts. The purpose of this study was to examine the clinical outcomes of eight patients treated with femoral condyle transplantation using large frozen osteochondral allografts. Methods: Eight patients were operated from September 2006 to January 2009 with femoral condylar transplantation using 25-30 mm cylindrical frozen osteoarticular allograft for large osteochondral defect of the femoral condyles (Mega OATS technique). Two of the patients had osteochondritis dissecans, five had traumatic osteochondral lesions. And one patient had cartilage lesion due to loose interference screw after an ACL reconstruction. In all patients cartilage lesions were preoperative evaluated by MRI. At follow-up VAS pain score, Tegner, and KOOS scores were determined. Results: There were 5 males and 3 females. Average follow up was 28 months (7-37) average age at the time of the follow up was 36 years (23-44). Average KOOS sub scores were for symptoms 35, pain 70, ADL 54, sport 87, QOL 81. Average VAS pain at rest was 4 and in activity were 7. Average Tegner score was 4. Three of the patients had arthroscopic reoperations due to increasing pain. In all patients degenerative changes in transplanted cartilage was found. One patient is waiting for total knee arthroplasty. Conclusions: Studies using fresh and living osteochondral allograft have demonstrated good functional and clinical outcome. In this study the patients were treated with frozen osteoarticular allograft. Subjective outcome scores and reoperations revealed relatively poor clinical and biological outcome indicating that using frozen osteoarticular allograft transplantation may not achieve satisfactory clinical results.

P16-1169 Mid-to-long term longitudinal outcome of autologous chondrocyte implantation in the knee joint: a multilevel analysis Richardson J.1, Bhosale A.2, Kuiper J.H.2, Johnson E.2, Harrison P.2 1 Institute of Orthopaedics, Oswestry, United Kingdom, 2Robert Jones & Agnes Hunt Orthopaedic Hospital, Orthopaedics, Oswestry, United Kingdom

S259 Objectives: Autologous Chondrocyte Implantation (ACI) is a cell therapeutic approach for the treatment of chondral and osteochondral defects in the knee joint. Previously we have reported on the histological and radiological outcome of ACI in the short-to-mid term, which gives mixed results. Here, we report on the clinical outcome of ACI for the knee in the mid-to-long term. Methods: We analysed 80 patients who had undergone ACI of the knee with mid-to-long-term follow-up. The mean patient age was 34.6 years (SD 9.1 yrs), with 63 males and 17 females. Seventy-one patients presented with a focal chondral defect, with a median defect area of 4.1 cm2 and a maximum defect area of 20 cm2. The modified Lysholm score was used as a self-reporting clinical outcome measure to determine: 1) What is the typical pattern over time of clinical outcome following ACI? and 2) Which patient-related predictors for the clinical outcome pattern can be used to improve patient selection for ACI? Results: The Average follow-up time was five years, with a range of 2.7to-9.3 years. Improvement in clinical outcome was found in 65 patients (81%), while 15 patients (19%) showed a decline in outcome. The median pre-operative Lysholm score of 54 increased to a median of 78 points. The most rapid improvement in Lysholm score was over the 15 month period post-operation, after which the Lysholm score remained constant for up to nine years. We were unable to identify any patient-specific factors (i.e. age, gender, defect size, defect location, number of previous operations, pre-operative Lysholm score) that could predict the change in clinical outcome in the first 15 months. Conclusions: ACI seems to provide a durable clinical outcome in those patients demonstrating success at 15 months post-operation. Comparisons between other outcome measures of ACI should be focussed on the clinical status at 15 months after the surgery. The patient-reported clinical outcome at 15 months is a major predictor of the mid-to-long term success of ACI.

P16-1171 The choice of the arthroscopic procedure for chondral defects determines the intraarticular synthesis of growth factors and liberation of MSCs Varoga D.1, Pries F.2, Lippross S.1, Pufe T.3, Seekamp A.1, Hartz C.1 1 University Hospital of Kiel, Department of Trauma, Kiel, Germany, 2 Mare-Hospital, Department of Arthroscopic Surgery and Sportstraumatology, Kiel, Germany, 3RWTH Aachen, Department of Anatomy and Cell Biology, Aachen, Germany Objectives: Efficacious treatment of chondral defects of weight bearing articular surfaces is a daily challenge in orthopaedic practice. Techniques used for arthroscopic treatment of OA of the knee include shaving, abrasion chondroplasty (ACP) and bone marrow stimulation techniques such as abrasion arthroplasty (AAP) and microfracturing (MF). Subsequently regeneratory cells, growth factors and cytokines are released into the defective area and the joint cavity. Fibroblast growth factor (FGF), insulin-like growth factor-1 (IGF-1), transforming growth factor beta-1 (TGF-b1), vascular endothelial growth factor (VEGF) and katabolic cytokines such as tumor necrosis factor alpha (TNF-a), interleukin-1 (IL-1) and interleukin-6 (IL-6) play a pivotal role in the regeneration process of chondral defects and in the regulation and differentiation of mesenchymal stem cells (MSC) to chondrocytes. The aim of the current study was to monitor the intraarticular (i.a) release of growth factors and cytokines dependent on the choice of arthroscopic treatment. Methods: Postoperative haemarthrosis from 119 Patients after different arthroscopic knee procedures were collected (70 men, 49 women with a mean age of 48 years). In 95 patients 2.5 mg of the steroid Dexamethasone was administered into the knee joint after the procedure. Samples from the drainage bottle were taken 5, 22 or 44 hours after surgery and growth factors were measured using immunohistochemistry (IHC), Enzyme-linked immunosorbent assay (ELISA), Western-blot and Luminex assay. Total mononuclear cells were isolated from haemarthrosis and cultured.Collagen type I and II proteins were detected using

123

S260 IHC und RT-PCR in the adherent MSCs. For further testing the cell pellet was stored at liquid nitrogen. Group differences were considered significant if p\0.05. Results: AAP or MF of the knee joint results in significant higher levels of i.a TGF-b1 when compared to expression levels of patients after meniscectomy (22 ng/ml vs. 11.4 ng/ml) measuring 22 h after treatment. Similar findings were demonstrated for i.a VEGF expression. 22 hours after arthroscopic treatment the experiments revealed no differences in the expression of other growth factors between the different treatment groups. The assays showed an increase of i.a. IGF-1 expression 44 hours after AAP when compared to levels measured in haemarthrosis from the same patient 22 hours after treatment. Steroid injection after AAP led to increased VEGF levels (5.6 ng/ml) compared to control samples (2.3 ng/ml). In contrast, IL-6 levels decrease after corticosteroid administration in case of AAP/ACP. Bone marrow stimulating techniques release more MSC into the joint cavity compared to ACP and meniscectomy. Even solely chondral procedures release a significant number of MSC, which were much higher when compared to meniscectomy. The i.a steroid injection in AAP results in higher numbers of MSC. Conclusions: Despite the high frequency of arthroscopic procedures in case of chondral defects of the knee joint, studies about the expression of i.a growth factors or the release of MSCs dependent on the choice of surgery were nearly abundant. Taken together, our examination demonstrates for the first time the regulation of growth hormones and mesenchymal progenitor cells in postoperative haemarthrosis dependent on the choice of the arthroscopic procedure. Our results provide evidence for a therapeutic effect of subchondral penetrating procedures such as MF or AAP and the additional steroid injection.

P16-1200 ‘‘Fixation Plus’’: evaluation of a novel fixation technique in adult osteochondritis dissecans of the knee Lintz F.1, Pandeirada C.1, Boisrenoult P.2, Pujol N.2, Beaufils P.2 1 Hoˆpital de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France, 2Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France Objectives: Conservative treatment of osteochondritis dissecans (OCD) of the adult knee requires correct bone and cartilage integration of the OCD fragment. Mechanical fixation using screws offers good primary stability but could be insufficient to ensure secondary consolidation. So, we have developed the ‘‘Fixation Plus’’ technique, using a double fixation by screws and additional biological fixation using mosaicplasty-type osteochondral grafts to improve OCD fragment integration. The aim of this retrospective study was to evaluate the short-term results of this technique. Methods: Eight patients (mean age 17-years, range 13-24) who underwent surgery for OCD of the medial femoral condyle, Bedouelle IIB or III, between 2003 and 2008 were included. The technique involved curetting of the condyle defect followed by mechanical fixation of the OCD fragment using metal screws. Mosaicplasty grafts were then press-fitted through the fixed OCD fragment. Subchondral bone loss was overcome by intercalated spongious autograft. Patients were followed-up clinically and radiologically (Hugston OCD score). The screws were removed arthroscopically at 3 months and the cartilage was graded by ICRS-OCD score. OCD fragment integration was evaluated in 5 patients by arthro-magnetic resonance imaging (MRI). Results: Mean follow-up was 17.4 months (range, 3-36). Mean clinical Hugston score changed from 1.6 (range, 0-3) preoperatively to 3.4 (range, 2-4) postoperatively. Mean radiological Hugston score changed from 2.5 (range, 2-4) to 3.2 (range, 3-4). Arthroscopic ‘‘second look’’ showed that the OCD fragments were well integrated in each case. ICRS-OCD score was I in 2 patients, II in 5, and III in one. Postoperative MRI confirmed that cartilage and bone integration was complete. Conclusions: Mechanical fixation of OCD fragments using screws frequently fails due to mal-union. Mosaicplasty is an alternative approach but does not retain normal condylar anatomy and curve. The ‘‘Fixation Plus’’ technique adds a biological ‘‘booster’’ to OCD fragment fixation, with

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 excellent clinical and radiological results. Further studies are required to confirm the long-term results with this technique.

P16-1202 The effect of hyperbaric oxygen therapy on cartilage healing Akgun U.1, Kocaoglu B.2, Basci O.3, Zeren S.4, Saglican Y.5, Cimsit M.4, Basdemir G.6, Karahan M.3 1 Acibadem University Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 2Acibadem University Faculty of Medicine, Orthopaedic Surgery, Istanbul, Turkey, 3Marmara University, Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 4Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey, 5Acibadem University Faculty of Medicine, Istanbul, Turkey, 6 Acibadem Hospital, Istanbul, Turkey Objectives: We aimed to evaluate the effect of hyperbaric oxygen therapy (HBOT) alone or combined with drilling on full thickness cartilage defects. Methods: 32 adult rat (Wistar, [350gr) were divided into 4 groups. (A 8, B - 8, C - 8, D - 8,) Under general anesthesia, after medial parapatellar arthrotomy full thickness cartilage defect has been made by a burr on femoral sulcus of both knees of the animals in group A-B-C-D. Right knees (control knees) were sutured without further surgery. In left knees full thickness lesions were drilled by K wires and then sutured. Group A - C had daily HBO therapy in the postoperative period. Group B - D had no HBO therapy. At the end of 2 weeks, animals in groups A and B were sacrificed and all knees were harvested and evaluated by pathology department. At the end of 4 weeks group C and D were sacrificed and all knees were harvested and evaluated by pathology department. Specimens were stained with hematoxylin and eosin for osteochondral evaluation. Results: Cartilage lesions were evaluated by using semi-quantitative measures. (osteoblastic and osteoclastic activity, cartilage proliferation, remodelling of bone). At 2nd week specimens, the maturity of the healing tissue was significantly better in the HBOT group (p \ 0,05). At the 4th week specimens, complete healing of the lesion was significantly high in the HBOT group (p\0,05). In all specimens, we could not find significant difference in the control knees which had cartilage lesions without drilling. Conclusions: This experimental study shows that, a combined therapy of drilling and HBO is an effective way of treatment which accelerates the healing process of full thickness cartilage lesions.

P16-1206 One stage osteochondral repair with cartilage fragments in a hyaluronic acid/fibrin glue/platelet rich plasma scaffold: in vitro human and in vivo rabbit and goat animal model Marmotti A.1, Castoldi F.1, Rossi R.2, Bruzzone M.3, Bonasia D.E.4, Maiello A.4, Peirone B.5, Mauthe Von Degerfeld M.5 1 University of Turin, Mauriziano Hospital, Orthopaedics and Traumatology, Turin, Italy, 2University of Torino, Mauriziano Umberto I, Torino, Italy, 3University of Torino - Mauriziano ‘Umberto I’ Hospital, Department of Orthopaedics and Traumatology, Torino, Italy, 4University of Torino, Torino, Italy, 5University of Torino, Department of Animal Pathology, Section Surgery, Torino, Italy Objectives: Treatment of articular cartilage lesions is a common issue in orthopaedics. ‘‘One stage’’ repair is a promising approach: repair process is promoted directly through a viable cell source in a resorbable scaffold without ex vivo cell expansion and without patient’s exposure to multiple procedures. Our study shows a new ‘‘one-stage’’ surgical procedure for cartilage repair combining ‘‘in situ’’ minced autologous cartilage fragments, as source of viable cells, with a resorbable scaffold composed of hyaluronic acid derivative (Hyaff-11), fibrin glue and platelet rich plasma. Methods: In vitro (explant culture): articular cartilage from human, rabbit and goat knees was minced and loaded onto the scaffold; constructs were cultured for 1 and 2 months and successively evaluated with histological,

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 immunohistochemical analysis. In human explant cultures, immunofluorescence characterization of cells migrated into the scaffold has also been performed. In vivo (unilateral trochlear model): osteochondral defects were created in 30 adult rabbit knees (4,5 mm diam., 4 mm depth) and 16 goat knees (7 mm diam; 3,5 mm depth); defects were either treated with cartilage fragments embedded in the scaffold (Group A) or loaded with scaffold alone (Group B) or left untreated (Group C). Rabbit were sacrificed at 1, 3, 6 months, goats at 3, 6, 12 months. Repair process was evaluated with morphological, histological and immunohistochemical analysis. In goat, treated defects have also been evaluated with nanoindentation and results compared with normal trochelar goat cartilage. Results: In vitro, chondrocytes from cartilage fragments migrate and proliferated into the scaffold both in human and in animal (rabbit and goat) explant cultures. In human explant cultures, cell migration and outgrowth was age dependent and exhibited an increase under stimulation with TGF-beta or G-CSF (10 ng/ml) in culture media. Cells migrated into the scaffold showed a chondrogenic phenotype (sox9?, CD151?) under immunofluorescence. In vivo, repair tissue of Group A (cartilage fragments embedded in the scaffold) was better than control groups (Group B and C), showing some typical features of articular cartilage, as the presence of collagen type II in extracellular matrix. This result was more noticeable in goat model. Nanoindentation of samples of neoformed tissue from group A showed mechanical properties similar with normal trochelar goat cartilage. Conclusions: This study suggests that minced autologous cartilage fragments embedded in a HA/PRP/fibrin scaffold provide a viable source of cells and chondrogenic stimuli for a simple one-stage non-culture-based repair of osteochondral defects in a rabbit and goat model. Although in vitro results from human explant culture are promising, further careful clinical studies in human are required to allow this approach to become a therapeutic alternative in the treatment of articular cartilage lesions.

P16-1216 Gene expression and biochemical study of an engineered cartilage tissue: in vitro and in vivo maturation Mangiavini L.1, Deponti D.2, Scotti C.3, Pozzi A.3, Sosio C.4, Fraschini G.4, Peretti G.2 1 University of Milano-Bicocca, Residency Program in Orthopaedic and Traumatology, Milan, Italy, 2University of Milan, Faculty of Exercise Sciences, Milan, Italy, 3University of Milan, Residency Program in Orthopaedic and Traumatology, Milan, Italy, 4San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy Objectives: The purpose of this work was to evaluate the maturation in vitro and in vivo of an engineered cartilaginous tissue obtained by isolated swine articular chondrocytes embedded in fibrin glue at different experimental times. Methods: Isolated swine articular chondrocytes were embedded in fibrin glue. Sample groups were divided as follow: Some samples were left in standard culture conditions for one, five and nine weeks. Some others were implanted in nude mice for the same time points. The remaining samples were cultured in vitro for one or five weeks and then implanted in subcutaneous pouches of nude mice for four additional weeks. Gross evaluation, biochemical analysis (DNA and GAGs content) and gene expression (collagen type I and type II, aggrecan and Sox 9) were performed. Results: Generally, samples retrieved from nude mice experienced shrinkage and mass reduction; they also resulted in a higher content of DNA and GAG. Collagen type II was higher in the 5- and 9-week samples with respect to those cultured in vitro. Opposite finding was recorded for aggrecan expression, while Sox9 was not significantly different from in vitro and in vivo. Conclusions: The results of this study demonstrate that in vivo implantation of engineered cartilage composite results in increasing the cell proliferation and matrix formation. Pre-culturing the samples before implantation does not seem to interfere with the capacity of cell

S261 proliferation and synthesis, but, on the other hand, does not appear to ameliorate the quality of the engineered samples in this model.

P16-1226 The effect of hyperbaric oxygen therapy on synovial repair mechanism in cartilage lesions - experimental study Akgun U.1, Kocaoglu B.2, Basci O.3, Zeren S.4, Saglican Y.5, Cimsit M.4, Basdemir G.6, Ergun S.7, Karahan M.8, Turkmen M.2 1 Acibadem University Faculty of Medicine, Orthopedics and Traumatolgy, Istanbul, Turkey, 2Acibadem University Faculty of Medicine, Orthopaedic Surgery, Istanbul, Turkey, 3Marmara University Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 4 Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey, 5 Acibadem University Faculty of Medicine, Istanbul, Turkey, 6Acibadem Hospital, Istanbul, Turkey, 7Marmara University Faculty of Medicine, Istanbul, Turkey, 8Marmara University, Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey Objectives: We aim to evaluate the effect of hyperbaric oxygen therapy on synovial repair mechanism in full thickness cartilage lesions in animals. Methods: 44 adult rat (Wistar,[350gr) were divided into 6 groups. (A - 8, B - 8, C - 8, D - 8, E - 6, F - 6) Under general anesthesia, after medial parapatellar arthrotomy full thickness cartilage defect has been made by a burr on femoral sulcus of both knees of the animals in group A-B-C-D. Right knees were sutured without further surgery. In left knees full thickness lesions were drilled by K wires and then sutured. In groups E and F, right knees had no surgery in order to evaluate the effect of HBO in healthy knee, and left knees were only drilled without any cartilage lesion. Group A - C - E had daily HBO therapy in the postoperative period. Group B - D - F had no HBO therapy. At the end of 2 weeks, animals in groups A, B and E, F were sacrificed and all knees were harvested and evaluated by pathology department. At the end of 4 weeks group C and D were sacrificed and all knees were harvested and evaluated by pathology department. Specimens were stained with hematoxylin and eosin for synovial and osteochondral evaluation. Results: Synovial healing mechanism were evaluated by using semi quantitative measures. (synovial migration, synovial thickness, synovial vascularisation and cartilage defect cover) In healthy knees of control groups E and F, there was no change in the synovial tissues with or without HBO. However in all the knees with cartilage lesions there was synovial migration. At 2nd week specimens, the migration of the synovial tissue to the cartilage lesion area was significantly better in the HBOT group. (p\0,05) At the 4th week specimens, synovial vascularisation and thickness in the cartilage lesion area was significantly high in the HBOT group. (p \ 0,05) Conclusions: This experimental study showed that, HBO therapy accelerates the response of synovial healing mechanism in full thickness cartilage lesions in rats.

P16-1238 Chondrocytes in vitro density and cellular phenotype Deponti D.1, Mangiavini L.2, Pozzi A.2, Ballis R.3, Fraschini G.3, Peretti G.1 1 University of Milan, Faculty of Exercise Sciences, Milan, Italy, 2 University of Milan, Residency Program in Orthopaedic and Traumatology, Milan, Italy, 3San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy Objectives: Articular chondrocytes acquire a fibroblastic-like phenotype when cultured in monolayer, characterized by different cell morphology and by expression of non-cartilage specific genes such as type I collagen. We focused on some of the events which could affect the chondrocytes phenotype in vitro: the cell density and the time of culture. Methods: Swine articular chondrocytes were isolated and seeded at different cell densities (from 12,000 cell/cm2 to 100,000 cells/cm2). Samples were cultured for 6 and 8 days. Phenotype was evaluated in terms of

123

S262 aggrecan and collagen type I and II expression, while proliferation was assessed by cell counting. Results: Chondrocytes proliferated more at lower cell densities and did not show any proliferation at highest cell density; moreover, no further proliferation was observed when chondrocytes were kept in culture for 8 days. Chondrocytes seeded at 50,000 and 100,000 cells/cm2 were able to maintain better level of differentiation, characterized by higher expression of type II collagen and aggrecan and lower expression of type I collagen, but in all conditions a significant loss of chondrocyte phenotype from 6 to 8 days of culture was observed. Conclusions: Cell density influences chondrocyte capability of maintaining a differentiated phenotype in vitro; moreover, the proliferation is allowed only at low cell densities. Some other important modulators of proliferation and differentiation, such as growth factors and growth substrates, will be valuated in future studies in order to better modulate proliferation and the loss of phenotype in vitro.

P16-1250 Long-term effectiveness of extracorporeal shockwave therapy in the treatment of early spontaneous osteonecrosis of the knee: a 2 years follow-up Vitali M.1, Sosio C.2, Peretti G.3, Mangiavini L.4, Fraschini G.2 1 University of Milan, Residency Program in Orthopaedics and Traumatology I, Milan, Italy, 2San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy, 3University of Milan, Faculty of Exercise Sciences, Milan, Italy, 4University of Milano-Bicocca, Residency Program in Orthopaedic and Traumatology, Milan, Italy Objectives: This work reported of 15 cases of early spontaneous osteonecrosis of the knee successfully treated with a novel extracorporeal shockwave treatment (ESWT). Traumatic and vascular theories have been proposed as a causative factor of the spontaneous osteonecrosis of the knee, but the precise etiology still remains speculative. The lack of blood in some critical areas, such as the subchondral bone of femoral condyles or the tibial plateaus, has been considered the underlying condition of this pathology. The ESWT, thanks to its neo-angiogenetic effect, can be suggested as an effective conservative treatment for spontaneous osteonecrosis of the knee. Methods: 15 patients with medial femoral condyle osteonecrosis of the knee (one bilateral) were evaluated in this work. Their history was negative for steroid therapy, systemic lupus erythematosus, alcoholism, Caisson decompression sickness, Gaucher0 s disease and hemoglobinopathies. Exclusion criterium was the evidence of a structural collapse of the subchondral bone. Two patients had received a femoro-popliteal bypass within the last year, while others five presented a deficit of the vascular axis of the homolateral lower limb documented by an ecocolordoppler. A clinical evaluation was taken at the diagnosis using KSS, McGill Pain Questionnaire (PPI, NRS, VAS). Plain radiographs, radioisotope bone scan and MRI confirmed the diagnosis of osteonecrosis. The patients were treated with a cycle of three ESWT performed with 2000 pulses of 0,28 mJ/mm2 with Wolf Piezoson 300 with 6,5 MHz ultrasounds for three times in a month. A clinical evaluation was performed at one and three months and at 2 years follow-up, an RMI evaluation was also performed. Results: The clinical evaluation showed a significant improvement of the symptoms (p\0,001) and the articular functionality (p\0,001). MRI in all cases revealed the continuity of the cartilage with a reduction in bone marrow edema and no collapse of the lesion. Conclusions: The most effective treatment for osteonecrosis lesions without evidence of structural collapse is conservative, with rest and protected weightbearing. In our study, a single cycle of ESWT produced an improvement of the clinical and MRI aspects in eleven cases of spontaneous medial femoral condyle osteonecrosis of the knee. The neo-angiogenetic effect of the ESWT appears to accelerate the time for the symptom remission. The ESWT might have the potential to curtail the progression of the disease and to avoid the need for surgical treatment.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P16-1251 Evaluation of the articular cartilage of the knee in delayed gadolinium-enhanced MRI (dGEMRIC) at 3.0 Tesla Okuhara A.1, Deie M.1, Nakamae A.1, Nishimori M.1, Nakasa T.1, Adachi N.1, Ochi M.1 1 Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan Objectives: The purpose of this study was to evaluate the articular cartilage of the knee and to detect the prognoses the osteoarthritis of the knee using magnetic resonance imaging with use of Gd(DTPA)2- (Gadolinium diethylenetriamine-pentaacetic acid) for measuring glycosaminoglycan (GAG) concentration in delayed phase (delayed gadolinium-enhanced MRI (dGEMRIC))at 3.0 Tesla. (hypothesis) The 3.0 Tesla MR imagines with Gd (DTPA)2- could be useful to evaluate early articular cartilage damage of the knee joint. Methods: Informed consent was obtained from all patients. Twenty six patients’ twenty six knees (five male, twenty one female) who had the pain in their knee joints were included in this study. These patients were divided into four grades according to Kellgren-Lawrence classification, grade I two knees, II12 knees, III 5 knees and IV 6knees. At the time of evaluation with dGEMRIC, the patients’ mean age was 66.1years (ranged from 42 to 76) . Mean femoro-tibial angle (FTA) was 179.7 degrees (ranged from 162 to 185) and range of motion of affected knee joint was 6.9 (ranged from -30 to 0) / 129.8 (ranged from 90 to 145) degrees. These patients were imaged twice at first and second year. We followed the protocol issues reported on by Burnstein et al. Double dose Gd(DTPA) 2- (0.2mmol per kilogram of body weight) was injected intravenously, and the exercise after intravenous contrast administration was needed for effective penetration into the articular cartilage by walking up for 10 minutes. MRI imaging (T2, proton density image and T1 mapping) was performed 2 hours after an intra-venous injection and the evaluation was done at the coronal plane for femoro-tibial compartment by dividing articular surface of the medial knee joint into six zones. Region of interest (ROI) was set up and measured manually. P\.05 indicated statistical significance. Results: All six zones showed that as the grade of Kellgren-Lawrence classification progresses, the T1 value became lower, and T1 value was significantly lower at the femur side compared to the tibial side (P\.05). In cases of grade 2 and 3, there were no changes in normal 3.0 Tesla MR imaging, however, in 3.0 Tesla MR imaging with dGEMRIC the values of ROI at second year were decreased lower than ones at first year. Conclusions: This results indicate that dGEMRIC at 3.0 Tesla could be used for detecting the early stage of the articular cartilage damage and could have a great value in diagnostics of the early osteoarthritis of the knee. And it suggests that it is important to be conscious of the cartilage damage progression at the tibial surface.

P16-1292 Staging the failing medial compartment of the knee Bottomley N.1, McNally E.2, Gill H.S.1, Dodd C.1, Murray D.1, Beard D.1, Price A.1 1 University of Oxford, NDORMS, Oxford, United Kingdom, 2 Nuffield Orthopaedic Centre, Oxford, United Kingdom Objectives: There is a sizeable population of patients, up to 65% of all osteoarthritis presentations, who have disabling early stage osteoarthritis with only partial thickness cartilage loss (PT-OA) in the knee. In contrast to the very end stage ‘bone-on-bone’ disease, which is well treated by arthroplasty, PT-OA is treated using a variety of surgical interventions with varying success. We describe the MRI findings of partial thickness osteoarthritis. Methods: 50 patients with radiographic PT-OA underwent dedicated coronal, sagittal and axial 3 Tesla MRI sequences. Ligament integrity, osteophytes, meniscal anatomy, subchondral high signal and cartilage damage severity and location were assessed and scored. Scoring was by a Consultant Radiologist using a validated system (Interobserver - Kappa 0.84) based on WORMS (Whole organ MRI score).

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: Different patterns of disease exist; including subchondral bone marrow oedema with no cartilage change, isolated partial thickness changes on the femoral condyle, partial thickness on both femur/tibia and full-thickness damage on one surface with partial on the other. Meniscal damage correlated with joint space narrowing but not focal cartilage damage. All intra-articular ligaments were intact. Conclusions: We believe PT-OA is a continuous spectrum of increasing irreversible damage, beginning in either the cartilage or bone and progressing to partial femoral condylar damage, then combined femoral and tibial partial thickness changes and ending with the development of focal, then confluent full thickness cartilage loss. We do not know how to best treat this failing knee joint, with all surgical techniques showing variable outcomes. We believe one of the main reasons for variability in outcome is a lack of characterization and precise staging of partial thickness disease. Different stages of disease will have different outcomes. Treatment should be based on the stage of disease and this work is the first step in providing a scoring system to stage partial thickness osteoarthritis.

P16-1299 What proportion of knee surgery practice is osteoarthritis and are we treating it correctly? Bottomley N.1, Kendrick B.1, Ferguson J.2, Al-Ali S.2, Dodd C.1, Murray D.1, Beard D.1, Price A.1 1 University of Oxford, NDORMS, Oxford, United Kingdom, 2 Nuffield Orthopaedic Centre, Oxford, United Kingdom Objectives: Total knee arthroplasty (TKA) accounts for 84% of all knee replacement surgery in the UK (NJR 2009) despite published epidemiological data showing that single compartment disease is most prevalent. We investigated this incompatibility further by describing the compartmental pattern and stage of cartilage loss of all patients with osteoarthritis (OA) presenting to a specialist knee clinic over one year. Methods: All new primary referrals in a calendar year by local General Practitioners to knee clinic at a United Kingdom Hospital were assessed. Tertiary referrals and second opinions were excluded. The final diagnosis after all imaging was recorded and tabulated. The standing AP, lateral and skyline radiographs of all cases of arthritis were scored to assess the pattern of disease. Results: 1029 new patients attended clinic. OA accounts for 52% of all attendances, meniscal pathology 17%, other diagnoses less than 10% each. 533 patients presented with osteoarthritis, mean age was 65 years (SD 11.6). Only 23% showed a tricompartmental pattern and another 3% medial and lateral compartment. 62% were medial and 8% lateral, approximately half with patellofemoral involvement, and 4% pure patellofemoral (Fig.2). 65% had symptomatic cartilage degeneration without full thickness ‘bone-on-bone’ loss.

Number of Patients

600 500 400 300 200 100 0 Meniscal

PFJ - Not Ligaments OA

Summary of clinic attendance

OA

Soft Tissue

Other

S263

70% PFJ Involvement

60%

No PFJ Involvement

50% 40% 30% 20% 10% 0% Medial

Lateral

PFJ

Bi / Tri

Pattern of arthritis presenting over a calendar Ye Conclusions: Osteoarthritis still presents the majority workload for the modern knee surgeon. Tricompartmental disease accounted for only 23% of presentations although joint registry data shows that the majority of patients undergo TKA. A controversial interpretation is that current practice involves the unnecessary removal of healthy joint surfaces. Further research is required. Interestingly, the majority of these patients did not have full-thickness cartilage loss. Best treatment of end stage, ‘boneon-bone’ OA is well known to be arthroplasty but the treatment of choice for the commonest presentation, the failing joint with early cartilage loss, is presently unknown.

P16-1329 Retrograde arthroscopic osteochondral autograft transplantation for chondral lesion of the tibial plateau of the knee: about 4 cases Wajsfisz A.1, Djian P.1 1 Cabinet Goethe, Paris, France Objectives: Arthroscopic osteochondral autograft transplantation is often chosen to treat osteochondral lesions on femoral condyles. Osteochondral lesions on tibial plateaux are less well known, not only because these lesions are rare, but also because there is no gold standard surgical procedure. Recently, a new arthroscopic procedure has been described to treat these lesions, but only case reports have been published. The aim of this study is to describe this surgical technique, its pitfalls and the pre-operatory planning. We report the clinical and radiological results about 4 cases. Methods: A prospective study was performed. Four patients were included age range was 16 to 40 years. All were treated for tibial plateau osteochondral lesion (3 lateral, 1 medial plateau). The diagnosis was achieved on standard X-rays and arthro-CT-scan or RMI. None of these patients presented a laxity. The shortest follow-up was 12 months (14 to 22months). The procedure was based on a mosaicoplasty autograft technique, but only one osteochondral plug could be implanted. First, arthroscopy revealed a deep chondral lesion, involving subchondral bone. The injured cartilage was debrided using a curette and an abrader until normal healthy cartilage bordered the defect. Osteochondral plugs were 8 to 10mm in diameter and 20mm long and were harvested from the most peripheral, proximal and medial part of the trochlea. They were harvested so that their chondral surface was orientated 20 to 30 degrees obliquely. A tunnel was drill through the tibia to the center of the defect, using a core reamer. The osteochondral plug was inserted by retrograde approach, from the tibial window through the tunnel under arthroscopic control. To enhance the stability of the osteochondral graft, a bioabsorbable screw was used to fill the space below the plug when primary stability was insufficient. Weight bearing was relieved for 6 weeks after surgery. An arthro-CT-scan or MRI was performed post-operatively to control plug implantation. All patients had pre-operative and latest follow-up functional evaluation using IKDC and KOOS documentation.

123

S264 Results: All four patients were satisfied with the surgery. In one case, a secondary intervention was need to remove the screw. Arthroscopic control during this procedure showed a perfectly integrated osteochondral plug, level with the tibial plateau cartilage. All patients had A IKDC scores before the surgery, and at the last follow-up. The subject if IKDC score improved from 40 to 65 at latest follow-up. The KOOS evaluation showed an improvement of 20 points by item. Every patient agreed to undergo a control MRI or arthro-CT-scan. In each case, this showed perfect integration of the osteochondral plug. Conclusions: We found in the literature only one study concerning 3 patients and one case report which have similar results as us. The procedure is very difficult and needs a good knowledge of knee arthroscopy and an experience with mosaicplasty. The surgical treatment of isolated lesions of tibial plateaux by retrograde osteochondral autograft transplantation under arthroscopic controlshow good results at mid term.

P16-1350 Biomechanical properties of tissues used in the autologous chondrocyte implantation Handl M.1, Drzik M.2, Adler J.3, Povysil C.4, Kos P.1, Trc T.1 1 University Hospital Motol, Orthopaedic Clinic, Prague, Czech Republic, 2 International Laser Center, Bratislava, Slovakia, 3Clinic Sao Vicente de Paulo, Lisbon, Portugal, 4Charles University Prague, Institute of Pathology, Prague, Czech Republic Objectives: Autologous cultured chondrocytes as a hyaline-like cartilage in the form of a solid chondrograft (fixed by fibrin glue) are now one of several possible procedures used for reconstructing full-thickness chondral lesions of joints. This study was focused on evaluating biomechanical properties of human hyaline cartilage and cultured autologous chondrocytes used in autologous chondrocyte implantation (ACI), with an emphasis on dynamic compressive testing. Methods: Hyaline cartilage samples were obtained from operated knee joints and served as the source of graft material. Special harvesting cutting tubes were employed to provide the same shape and size samples for the measuring test as well as for histology.The experimental set up of the pendulum-type testing device was constructed. The dynamic movements of the impactor’s mass were measured by a Laser Doppler Vibrometer (LDV), which was correlated in parallel to a piezoelectric transducer fixed on the impactor. A Polytec OFV-302 LDV and a Bruel & Kjaer Type 4375 piezoelectric accelerometer were used for this purpose. The cartilage samples were fixed, clamped and set up planparallel for the measurement, so that it was possible to adjust the load in a direction normal to the original cartilage surface. The pendulum impactor deceleration/acceleration signals from both the sensors were preconditioned and digitized by a Tektronics TDS 220 oscilloscope. The information was processed by evaluation software developed in a LabView environment. The study was focused on measuring the material properties such as the ultimate strength, stiffness and compression dynamic loading curve at loads similar to a human0 s average weight. Results: The mean values of compressive strengths were determined to be 11.32 ± 2.32 MPa for the non-weight-bearing zone, 10.37 ± 4.11 MPa for the defect zone and 7.44 ± 3.98 MPa and 8.73 ± 4.46 MPa for both cultured tissue engineered materials. The corresponding strains were identified as e = 0.35 ± 0.09 (non-bearing zone), e = 0.33 ± 0.10 (defect zone) and e = 0.76 ± 0.09 and e = 0.69 ± 0.14 for the artificial materials. Cartilages were tested as circumferentially unconstrained where sidewall bulging was possible under load. Conclusions: The tissue-engineered cartilage used nowadays as hyaline-like cartilage for ACI must be anatomically suitable and biocompatible, corresponding to the mechanical properties of human cartilage. The suitability of substitute materials depends on the requirements of the original cartilage type and on its standard mechanical value. Due to its reliability, precision and contactless nature, this method can be used in a variety of biomechanical tests for measuring dynamic load-induced deformations in natural tissues. In order to characterize hyaline cartilage biomechanics in dynamic loading conditions the Laser Doppler Vibrometer was used as a basic sensor for detecting time history of samples’ deformation during the impact. The results obtained have shown the feasibility, reproducibility and overall

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 reliability of the dynamic tests. The introduced method facilitates the application of high strain rate loadings and the measurement of deformation response even on small-sized samples.

P16-1366 TruFit CB plugs for articular cartilage repair in the knee: clinical and radiological results at 3 years Spalding T.1, Clewer G.1, Bird J.1, Thompson P.2 1 University Hospital Coventry, Coventry, United Kingdom, 2University Hospitals Coventry and Warwickshire NHS Trust, Trauma and Orthopaedic Surgery, Coventry, United Kingdom Objectives: To report on the clinical and radiological results of articular cartilage repair in the knee using the TruFit CB plug - a synthetic scaffold designed to support tissue differentiation, potentially allowing rapid mobilisation. Methods: 24 patients undergoing repair of chondral or osteochondral defects with the TruFit CB plugs (Smith and Nephew, Andover, US) were prospectively analysed using Lysholm, IKDC subjective, Tegner, KOOS, and SF-36 outcome assessments in addition to MRI imaging, at 6, 9, 12, 18, 24 and 36 months post surgery. Results: The 20 male and 4 female patients had a mean age of 34 years (range 19-50). 20 chondral and 4 osteochondral defects were repaired on the medial femoral condyle (n=14), lateral femoral condyle (n=4), and lateral or central trochlea (n=6). Mean defect size was 1.8 cm2 and 1 - 4 plugs were used (mode 2). There were 13 primary procedures and 11 revision procedures, following microfracture (7), osteochondral grafting (1) MACI grafting (1) and failed fixation OCD (2). Statistically significant improvements was seen in all outcome scores at 12 months; Lysholm improved 25 points (52.5 to 77.7), IKDC Subjective improved 24 points (42.6 to 67.1), SF36 physical improved from 39 to 50 and all components of KOOS improved: Pain by 21 points (62 to 83), Symptoms by 16 points (57 to 73), ADL 18 points (69 to 87), Sport & Recreation 42 points (25 to 67) and Quality of Life 28 points (33 to 61). After patients were allowed to return to sport at 12 months the Tegner activity score improved from 2.9 to 5.6. Five patients underwent repeat arthroscopy for persistent symptoms and this demonstrated that the plugs stay soft for 9 months or more. MRI evaluation, including T2 mapping, has demonstrated regeneration of the articular cartilage, reformation of the subchondral bone lamina and scaffold resorption with subchondral bone reformation. Conclusions: TruFit CB plugs offer a potential solution for small focal chondral defects. However integration is slow and the surface remains soft for up to 9 - 12 months. These observations have important implications for the usage of these plugs, and our recommendation is that the plugs are not used for large lesions unless well supported by a bone bridge between plugs or by surrounding articular cartilage.

P16-1470 Focal cartilage defects in the knee impair quality of life as much as severe osteoarthritis. A comparison of KOOS in four patient categories scheduled for knee surgery Heir S.1, Nerhus K.2, Røtterud J.H.3, Løken S.4, Ekeland A.2, Engebretsen L.5, A˚røen A.6 1 Oslo Sports Trauma Research Center, Norwegian School of Sports Sciences, Oslo, Norway, 2Martina Hansens Hospital, Bærum, Norway, 3 Akershus University Hospital, Lørenskog, Norway, 4Ulleva˚l University Hospital, Orthopaedic Centre, Oslo, Norway, 5Medical Faculty, University of Oslo, Oslo, Norway, 6Oslo Sports Trauma Research Center, Orthopedic Center, Oslo, Norway Objectives: To evaluate the extent to which cartilage defects in the knee affect patients’ quality of life, compared to three other knee disorders using KOOS as the outcome measure. Methods: Previously registered KOOS baseline data on patients enrolled in different knee treatment studies were included in the present study, the patients being 18-67 years (working population) at data registration. The different patient categories were

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 A) Patients with knee osteoarthritis enrolled for knee arthroplasty, B) Patients with knee osteoarthritis enrolled for osteotomies around the knee, C) Patients with focal cartilage lesions enrolled for cartilage repair, and D) Patients with ACL deficient knees enrolled for ACL reconstruction. The KOOS subscale Quality of Life was the main parameter for comparison of complaints. Results: At preoperative baseline, patients with focal cartilage defects in the knee scored 27.5 on the KOOS subscale QoL, not significantly different from the 28.8 and 27.2 in the patients with osteoarthritis enrolled for knee osteotomies and arthroplasties respectively. For all the subscales of KOOS, the cartilage patients scored significantly lower than the ACL deficient patients did. Conclusions: Patients with focal cartilage lesions have major problems with pain and functional impairment. Their complaints are worse than those of ACL deficient patients, and quality of life is affected to the same extent as in patients scheduled for knee replacement.

P16-1474 Cartilage repair: mosaic plasty resulted in higher degree of tissue filling but impaired subchondral bone more than Microfracture technique. A randomized, controlled, blinded, long term follow-up trial in 100 rabbit knees Heir S.1, A˚røen A.2, Løken S.3, Holme I.4, Engebretsen L.5, Reinholt F.6 1 Oslo Sports Trauma Research Center, Norwegian School of Sports Sciences, Oslo, Norway, 2Oslo Sports Trauma Research Center, Orthopedic Center, Oslo, Norway, 3Ulleva˚l University Hospital, Orthopaedic Centre, Oslo, Norway, 4Oslo Sports Trauma Research Center, Oslo, Norway, 5Medical Faculty, University of Oslo, Oslo, Norway, 6 University of Oslo/Rikshospitalet-Radiumhospitalet, Institute of Pathology/Pathology Clinic, Oslo, Norway Objectives: To investigate the results regarding tissue filling and subchondral bone changes with time, comparing Mosaic plasty and Microfracture technique for cartilage repair in the rabbit knee. Methods: In New Zealand rabbits age 22 weeks, a 4 mm pure chondral (ICRS grade IIIb) defect was created in the medial femoral condyle in both knees. A stereomicroscope was used to optimize the preparation of the defects. In one knee (randomized) the defect was treated with microfracture technique, whereas in the other with mosaic plasty. The animals were sacrificed 12, 24 and 36 weeks post surgery. Defect filling and the density of subchondral mineralized tissue was estimated using Analysis Pro software on micrographed histological sections. Animals with identical but untreated defects served as controls. Results: Tissue filling following mosaic plasty increased with time resulting in a significantly 34.0% higher filling than microfracture technique at 36 weeks, standard deviation of mean difference being 33.9%. Mosaic plasty impaired subchondral mineralized tissue density significantly whereas microfracture technique did not. The differences between the two techniques in both outcomes were apparent only at the 36 weeks follow up. Conclusions: Mosaic plasty resulted in higher degree of tissue filling than microfracture technique, however large standard deviations imply unpredictable results even in a standardized animal model. Mosaic plasty impaired subchondral bone more than microfracture technique, the implication being unclear. This study demonstrated the necessity of long term follow ups after cartilage repair also in animal models.

Knee: patellofemoral P17-37 A two year follow-up of patients receiving a southwick-fulkerson osteotomy with operative guidance of alignment via femoral nerve stimulation Albright J.1, Mellecker C.1, Ebinger T.1 1 University of Iowa, Iowa City, United States

S265 Objectives: The objective of this study is to evaluate the outcomes of patients who recieved a southwick-fulkerson osteotomy under the guidance of femoral nerve stimulation. It is believed that the stimulation and contraction of the quadriceps gives the surgeon a more accurate picture of patellar tracking, resulting in improved placement of the tubercle and MPFL graft. Because these are placed based off dynamic movement, it should theoretically improve outcomes in the awake patient. Methods: 32 patients for a total of 37 knees were contacted. This included all patients who received this procedure and were at least two years postoperative. 26/31 patients (31 knees) returned for follow-up. Patients were evaluated using KOOS and IKDC scores, and physical exam features of apprehension, dynamic tracking and crepitus. Results: 29 knees reported they were happy with the procedure and 29 reported they would do it again. 1 knee (3%) reportedly redislocation, but did not return for exam for verification. 30/31 had relatively normal tracking. 1 knee displayed a residual J sign. 4/16 knees with MPFL repair and 0/15 with reconstruction exhibited apprehension. Increased age and apprehension were correlated with lower outcome scores. Demonstration of chondromalacia did not affect functional outcome scores. Conclusions: Intraoperative femoral nerve stimulation is an effective way of evaluating patellar tracking intraoperatively that leads to 97% successful postoperative results. The Southwick modification of the Fulkerson Osteotomy decreased recovery times. MPFL reconstruction eliminates the persistence of the apprehension sign.

P17-68 The role of Denk0 s operation in the treatment of recurrent patellar dislocations Ba´nyai T.1, Gera L.2, Csomor L.1, Nagy E.1 1 County Hospital, Kecskeme´t, Traumatology and Hand Surgery, Kecskeme´t, Hungary, 2SZTE Teaching County Hospital, Department of Traumatology and Hand Surgery, Kecskeme´t, Hungary Objectives: The aim of our study was to evaluate distal and proximal surgical procedures to find the best guideline for the treatment of recurrent patellar dislocations in our daily practice. This is a second phase of our study with analysed x-ray results. Methods: Between 2002 and 2008 in the Department of Traumatology of County Hospital in Kecskeme´t we treated 358 patients with patellar dislocations. We asked the selected (operated) patients to come back for control examination and to fill a questionnaire. We described the outcomes with Lysholm score. The patellofemoral congruence and alignment detected by radiographs. We used medical documents to evaluate the surgical procedures. The state of the patellofemoral arthrosis was described with the help of the arthroscopical findings (Outerbridge). The follow up periods were 1-6 years. Results: During the time-period we treated 358 patellar dislocations(41% male, 59% female, average age: 25,5 years). There were 181 monoluxations and 177 recurrent dislocations. In 283 cases were spontan reposition, 75 times reposition by the examiner. Haemarthros was in 21%, joint aspiration 50 times. Gyps cast in 78 cases. X-ray results: Patellar morphology: Wiberg 1:10%, 2:72%, 3:18%. Trochlear morphology: Dejour et al. A:63%, B:10%, C:27%. Sulcus angle 1358[41%, 1358-1458: 25%, 1458\34%. The main Caton-Dechamps index: 1,27. In 73% we performed conservative treatment (physiotherapy, brace), in 27% operation. Within the operated group: Arthroscopy 77%, tibial tubercle medialisation 26%, lateral retinaculum release 40%, medial capsular shift 18%, in single case of extreme genu valgum we performed femoral osteotomy. In 17 cases we found osteochondral fragment of the patella, lesion on the femoral condyle in 2 cases. In one case was the refixation possible (KFI screws). According to the trochlear morphology in our cases we didn’t perform thorchlear plasty. The follow-up of the operations possible in 65 cases. Main Lysholm score before operation: 42,1 points, after operation mainly 82,6 points, specially after Denks operation:87,25points. The screws were removed mainly 8,5 months after the surgery. In 20% we found rare patella subluxations after operation.

123

S266 Conclusions: According to the literature the surgical treatment of recurrent patellar dislocations can give good results. Important is the correct diagnosis, preoperative planing, and the choice of adequate technic. By skeletally immature patients in case of recurrent dislocation we prefer physiotherapy and if it not helps watchfully capsular release and shift. After maturity when the problem remains we perform tibial tubercle medialisation. The x-ray radiographs and MR images show us the possible indication for bony interventions. By adults with symptomatic tight lateral retinaculum we perform lateral release with our without arthroscope. When the patellar tilt is increased and the patella lateralised we often use tibial tubercle medialisation (Op.sec Denks), and if the intraoperative congruence-evaluation demands, carefully lateral release. In our opinion the reason of the good success rate after Denks operation depends on the adequate patient selection, possible avoiding the operations inside the knee joint and on the early mobilisation.

P17-128 Conservative versus surgical treatment for repair of the medial patellofemoral ligament in acute dislocations of the patella Bitar A.C.1, Camanho G.L.2, Viegas A.d.C.2, Demange M.K.2, Hernandez A.J.2 1 Instituto Vita, Sa˜o Paulo, Brazil, 2IOT - HCFMUSP, Sa˜o Paulo, Brazil Objectives: The objective of this study was to analyze and compare the results obtained after 2 types of treatment, surgical and conservative, for acute patellar dislocations. Methods: We divided 33 patients with acute patellar dislocations into 2 groups. One group with 16 patients underwent conservative treatment (immobilization and subsequent physiotherapy), and the other group with 17 patients underwent surgical treatment. A radiographic examination was performed in the evaluation of the patients to verify predisposing factors for patellofemoral instability, and the Kujala questionnaire was applied with the intention of analyzing the improvement of pain and quality of life. The x2 test, t test, and Fisher test were used in the statistical evaluation. A significance level of P \ .05 was adopted. Results: The groups were considered parametric in relation to age and sex. The conservative treatment group exhibited a higher number of recurrent dislocations (8 patients) than the surgical treatment group, which did not have any relapses. In addition, the surgical treatment group obtained a better mean score on the Kujala test (92) than the conservative treatment group (69). Conclusions: We conclude that surgical treatment afforded better results. There were no recurrences in the surgical treatment group, but there were 8 recurrences in the conservative treatment group. The mean Kujala score was 92 in the surgical treatment group and 69 in the conservative treatment group.

P17-188 Superficial slip of quadriceps tendon for reconstruction of medial patello-femoral ligament in persistent dislocation of patella Goyal D.1 1 Saumya: Center for Advanced Surgeries of the Knee Joint, Ahmedabad, India Objectives: Acute Patella dislocation, not responding to conservative treatment is a frequent problem. Untreated cases can cause severe disability at young age, with high potential of gradual development of Patellofermoal Arthritis. In recent years, the importance of the medial patellofemoral ligament (MPFL) as the primary soft-tissue restraint to lateral displacement of the patella has been recognized. Hence, reconstruction of the MPFL is recommended for persistent dislocation cases, when they do not respond to conservative means. We use superficial slip of Quadriceps tendon for MPFL Reconstruction. We hypothesized that superficial slip of Quadriceps tendon will give results that are comparable to the other graft materials used in International literature. We can spare hamstrings and remaining thickness of Quadriceps tendon that can be used for future major reconstructions of the Knee.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: 32 cases of recurrent dislocation of Patella were studied in a prospective study. There were 22 females and 10 males between age group 16-30. The longest follow up was 4 years, and shortest was 11 months. Cause of dislocation in all cases was traumatic dislocation of Patella. All cases had conservative treatment as the first line of treatment. Duration of surgery from time of 1st Injury varied from 26 days to 2 years. Inclusion Criteria included cases with persistent instability even after 3 weeks of conservative trial and when lateral push of Patella causes it to move out of Trochlea. Presence of bony causes of Patellar instability on affected side with normal bony architecture on opposite side; was exclusion criteria. We also excluded all cases that also had other ligaments injured. All patients underwent MPFL Reconstruction using superficial Slip of Q tendon. Superficial slip is broad, band like structure & can mimic native MPFL when reconstructed in proper fashion. We aimed to develop a band made up of superficial slip of Quadricpes tendon in all cases. Breath of the band was between 8-10 mm in all cases. We predetermined selection of Patellar and femoral attachment sites in all cases. The graft is secured to the femoral attachment site with an interference screw to have intraosseous fixation. Patellar fixation is not required as distal attachment of Quadriceps tendon is left intact and is folded medially so as to mimic natural patellar attachment. Results: Clinical & Radiological Assessment, Pre-op MRI, Lysholm Knee Score, VAS scale and Kujala Patellofermoral Score were done in all cases. The study was also compared with different international publications using different graft choices. There was not a single case of Re-dislocation. Lysholm score, Kujala Score and VAS, all improved to near pre-operative level. Two cases had slightly short graft length, but still it could be fixed intra-osseous. This was the only complication we had. Conclusions: MPFL Reconstruction using superficial slip of Quadriceps tendon for persistent dislocation of Patella is a safe and reliable method. It spares important hamstring and Q tendons for future reconstructions. It also spares any bony procedure done on Patella. Results are comparable when done with other graft choices. The technique is easy, reproducible with very few complications.

P17-210 Treatment for recurrent dislocation of the patella - reconstruction versus repair of the medial patellofemoral ligament Nishimori M.1, Deie M.1, Adachi N.1, Nakamae A.1, Kanaya A.2, Ochi M.1 1 Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan, 2Mazda Co. Ltd., Mazda Hospital, Orthopaedic Surgery, Hiroshima, Japan Objectives: The purpose of this study was to evaluate the surgical repair of the medial patellofemoral ligament (MPFL) compared to the reconstruction of MPFL in patients with recurrent dislocation of the patella. Methods: Surgical treatment was performed on 26 knees in 26 patients with recurrent patella dislocation (aged 13 to 38 years; with a mean age of 21.4 years) from July 2004 to June 2007. Eighteen cases underwent MPFL reconstruction and 8 cases underwent MPFL repair. Before surgical treatment, we evaluated the pre-operative X-rays and CT scans. The type of surgical procedure varied, depending on the preoperative evaluation: 1) we performed MPFL repair if we could confirm that a bony fragment was detached from the patella, 2) we performed MPFL reconstruction if we could not confirm that a bony fragment was detached from the patella. We evaluated the results of 2 MPFL procedures. The clinical outcomes were evaluated by both the pre-operative and follow-up Kujala scores. Clinical data included the incidence of recurrent dislocation, lateral patellar apprehension, range of motion and radiography postoperatively. For the radiographs, the patients were examined from the anterior-posterior view, lateral view, and Merchant’s view, as well as with a 2 kg stress patellar X-ray from an axial view at 45 of knee flexion. We used the Merchant’s view to measure the congruence angle (CA), the tilting angle (TA), sulcus angle (SA), and the lateral shift ratio (LS). In the stress X-ray,

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 the lateral stress shift ratio (LSS) and the medial stress shift ratio (MSS) were measured. The stress X-ray was taken with the 2 kg stress applied from medial to lateral and from lateral to medial using a pushing apparatus. Results: There was no recurrence of dislocation after surgery. Both groups tested negative to the apprehension test. All knees had full range of motion. The mean Kujala score at follow-up was significantly improved in both groups. Post-operative radiological assessment showed that the mean values of CA, TA and LS were demonstrated to be within the normal range in both groups. On stress X-ray evaluations, the mean values for the MSS ratio and LSS ratio were demonstrated to be within the normal range, but there were significant differences between the post-operative mean value of MSS in the repair group and that of reconstruction group. Conclusions: The vast majority of this study’s results for both MPFL repair and MPFL reconstruction were satisfactory. Therefore, we recommend performing MPFL repair for recurrent patella dislocation cases with a bony fragment detached from the patella.

P17-261 Outcome of surgery for recurrent patellar dislocation based on of the tibial tuberosity trochlear groove distance Tecklenburg K.1, Whitehead T.2, Webster K.3, Elzarka A.2, Feller J.3 1 Medizinische Universita¨t Innsbruck, Unfallchirurgie und Sporttraumatologie, Innsbruck, Austria, 2La Trobe University, Medical Centre, Bundoora, Australia, 3La Trobe University Musculoskeletal Research Center, Faculty of Health Sciences, Bundoora, Australia Objectives: This study evaluated the outcomes for a cohort of patients who underwent either medial tibial tuberosity transfer or an isolated lateral release as primary surgery for recurrent patellar dislocation. The decision to use one or other procedure was based on the preoperative tibial tuberosity trochlear groove distance. Methods: 46 of 63 knees in 35 of 48 patients were evaluated at a mean of 38 months postoperatively. Medial tibial tuberosity transfer was performed in 33 knees and isolated lateral release in the remaining 13 knees. Evaluation included the Kujala, IKDC (subjective and objective) and SF36 scores. 23 knees in the tibial tuberosity transfer sub-group also underwent repeat radiological examination. Results: There were further episodes of patellar dislocation in 6/46 knees (13%). Only one knee had more than one further episode of instability. The rate of further instability was 15% in the tibial tuberosity transfer subgroup and 8% in the lateral release sub-group. The mean subjective IKDC score was 80, the mean Kujala score was 88 and the mean objective IKDC scores were 79% A and 21% B. The mean postoperative tibial tuberosity trochlear groove distance in the tibial tuberosity transfer sub-group was 8.9mm compared to the mean preoperative value of 16.8mm. Conclusions: The protocol resulted in satisfactory outcomes for most patients, but a further reduction in recurrence rates may be able to be achieved by selectively addressing patellar height, the medial patellofemoral ligament and trochlear dysplasia in addition to the tibial tuberosity trochlear groove distance. The reliability of CT measurement of the tibial tuberosity trochlear groove distance remains to be established.

P17-324 Intraoperative femoral nerve stimulation in evaluation of patellar tracking: the effects of tourniquet use and catheter placement Albright J.1, Cox E.2, Cohen E.2, Mellecker C.2, Raw R.3, Fraser A.3, Williams G.4 1 University of Iowa, Orthopaedics and Rehabilitation, Iowa City, United States, 2University of Iowa, Iowa City, United States, 3University of Iowa, Anesthesia, Iowa City, United States, 4University of Iowa, Physical Therapy and Rehabilitation Science, Iowa City, United States Objectives: The purpose of this study was to evaluate the effects of using a pneumatic tourniquet and the importance of catheter placement for femoral nerve stimulation of the extensor mechanism during intraoperative assessment of patellar tracking. We hypothesize that the pneumatic tourniquet would decrease the sensitivity of the femoral stimulation of the quadriceps

S267 over time and require greater stimulus amplitudes for generating muscle contraction. A second hypothesis is that the proximity of the catheter’s placement to the femoral nerve is important for stimulation at lower amplitudes. Methods: Electromyographic (EMG) analysis of muscle activity was performed by insertion of percutaneous EMG needles into the quadriceps and sartorius muscles to observe changes in amplitude threshold (mA) required to generate a palpable muscle contraction from femoral nerve stimulation. Eleven patients used ultrasound for catheter placement and ten were manually placed based upon body landmarks. Repeated analysis of muscle activity was performed prior to and after inflation of the tourniquet at 30 minute intervals. Results: Tourniquet application time correlated positively with the change in amplitude threshold required to generate muscle contraction. On average, patients had a four-fold magnitude increase in required amplitude of stimulus from the baseline threshold (pre-tourniquet inflation) to the final threshold (tourniquet inflated) with a two-hour tourniquet inflation time. The use of ultrasound for catheter placement significantly decreased the baseline amplitude required in comparison with catheters placed without, p = 0.0330. Conclusions: When performing intraoperative femoral nerve stimulation for assessment of patellar tracking, the use of a pneumatic tourniquet will require an increase in stimulus amplitude over time to generate the same extent of muscle contraction. Ultrasound guidance can provide improved positioning of the stimulating catheter in proximity of the femoral nerve to achieve contraction of the extensor mechanism at lower stimulus amplitudes.

P17-542 UKA in combination with PFR at average 12 years follow-up Heyse T.1, Khefacha A.2, Cartier P.2 1 University Hospital Marburg, Department of Orthopedics and Rheumatology, Marburg, Germany, 2Clinique Hartmann, Institut de Genou, Neuilly sur Seine, Paris, France Objectives: Safety and efficacy of unicompartmental knee arthroplasty (UKA) has been shown in large patient series. Patellofemoral replacement (PFR) is known to be a viable solution to end-stage patellofemoral arthritis. Bicompartmental osteoarthritis (OA) affecting the medial tibiofemoral and the patello-femoral compartment (medio-patellofemoral OA) is often treated with total knee arthroplasty (TKA). It was hypothesized that medio-patellofemoral OA can successfully be treated with bicompartmental arthroplasty. Methods: In a retrospective approach 9 patients who had received UKA in combination with PFR were included into the study. Intact ACL and lateral compartment were condition for the indication. Patients were clinically examined including clinical scores (KSS and WOMAC) and radiographies were evaluated. Satisfaction of patients was recorded in four categories. Results: Average follow-up after bicompartmental arthroplasty was 11.8 ± 5.4 years (4 - 17 years). Among the 9 patients there were 8 female and 1 male at an average age at operation of 64 ± 5 years. No surgical revisions were required following bicompartmental arthroplasty. The KSS score increased from a preoperative 68.8 ± 26.2 to 175.5 ± 22.9 at latest followup (p = 0.002). WOMAC was 18.3 ± 8.6 at latest follow-up. All patients included were satisfied (n = 3) or very satisfied (n = 6) with the outcome of this surgical procedure. Conclusions: This small case series shows that a bicompartmental arthroplasty can be a successful approach to prevent or postpone TKA. However, this intervention is technically demanding and requires experience in both UKA and PFR.

P17-546 Patellofemoral axial radiographic technique: consistency across clinic sites Arendt E.1, Nord A.2, Agel J.1 1 University of Minnesota, Orthopaedic Surgery, Minneapolis, United States, 2University of Minnesota, Medical School, Minneapolis, United States

123

S268 Objectives: Diagnosis and treatment decisions for patella dislocations are made on a variety of factors including standard radiographs. Standard radiographs include a true lateral, and patella axial view(s), from which one can determine the amount of trochlear dysplasia (femoral groove shallowness) and the position of patella in relationship to the femur (lateral patella translation and tilt). The axial view demonstrates whether the patella is relocated in the femoral groove after a lateral PF dislocation. Axial radiographs can be used to define patellofemoral (PF) alignment including excessive lateral patella and/or translation. It is our clinical impression that ‘‘standard’’ radiographs are taken with the patient are taken in a variety of positions. This variability in radiographic techniques causes the radiographic measurements on which diagnostic and treatment decisions are made to be variable and therefore potentially inaccurate. The goal of this study is to determine if axial images vary based on clinic site, and if this variability exists across hospital systems and/or within hospital systems. Secondarily, is there variability between the radiographs taken with different radiographic techniques that yield different clinical assumptions? Methods: Seven sites agreed to take a demonstration axial x-ray on a phantom limb used for training radiology technicians. The patella on the phantom limb was placed in an excessive lateralized position for the radiographic image. This position was consistent across all sites. In addition a human subject was used to document the normal patient set-up for taking an axial view. Radiographic techniques were recorded by a photo taken of the phantom set-up and the model set-up. Results: Radiographic technique: Beam orientation to the femoral shaft ranged from 26-70 degrees for the phantom limb and from 29-75 degrees for the patient. Knee position ranged from 27-124 degrees of flexion for the phantom and 45-144 degrees for the model. The accuracy of obtaining a requested degree of knee flexion was greater when the knee flexion angle was 45 degrees or greater. (Phantom limb averaged 3.5 degree, range 0-10; human model averaged 3.25 degrees, ranged 0-5 degrees). When the requested knee flexion was less than 30 degrees, the accuracy of obtaining the requested degree of knee flexion was less. (Phantom limb error averaged 6.25 degree, range 4-7, human model error averaged 15 degrees, ranged 11-25 degrees). Radiographic Images: Ten x-rays taken at 7 sites measured sulcus angle ranging between 139-150 degrees, with those views taken in lower degrees of knee flexion recording a higher sulcus angle. Of the 3 sites utilizing a ‘‘merchants’’ 30 axial X-ray, the congruence angle varied from ?12 to ?29, being normal at one site (?12) and abnormal ([?16) at 2 sites. Patella tilt, as recorded as converging lines along the long axis of the patella and along a line across the medial and lateral trochlear prominences, ranged from 10 - 38 degrees, with the lower knee flexion angles recording more patella tilt. Conclusions: Axial imaging of the patellofemoral joint utilizing digital radiographs is inconsistent in regards to imaging protocols and degree of knee flexion. Standardization of specific test requests would yield more useful images for the clinician. The clinician should be wary of interpreting radiographic findings in the absence of documentation of specific knee flexion angle and knowledge of the radiographic technique utilized.

P17-569 The influence of riskfactors of patella instability on the result of anatomic MPFL-reconstruction using a gracilis-tendon transplant Pfalzer F.1, Wagner D.1, Hingelbaum S.1, Huth J.1, Mauch F.1, Bauer G.1 1 Sportklinik Stuttgart, Stuttgart, Germany Objectives: The MPFL is the most important passive medial patella restraining force. Once destroyed by an occurred patellaluxation and followed by inadequate healing, it becomes insufficient which results in chronic patella instability. So the reconstruction of the MPFL is most important to recover patella stability. The question, which has to be answered, is how anatomic predispositions of patella instability (trochleadysplasia, TTTG-distance, patella alta, lateral trochlea slope) influence the outcome of the surgical procedure of MPFL-reconstruction.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: Since 2007 over 200 patients with chronic patellar instability have been treated in our department with MPFL-reconstruction with gracilis tendon. By now we have performed a prospective study in 50 cases with a minimum follow-up of 12 months. The clinical results have been measured by objective knee scores, according to Kujala and Lysholm. In addition, the sports activity level was gathered by Tegner- and Valderrabano-score. Pre-existing anatomic risk factors of patella instability have been detected by conventional x-rays and MRI. Results: 80 % of the patients showed a good to excellent result. The average Kujala score increased significantly about 20 points. One reluxation of the patella could be observed. In two cases a surgical revision because of woundhealing problems in the femoral fixation area was necessary. By follow-up, 80% of the patients returned to a similar respectively higher sport activity level in comparison to the time before the injury. A pre-existing femuropatellar cartilage lesion could be seen intraoperatively in 48% of the cases. In 90% a trochleadysplasia could be found, divided into 50% type I, 40% type II and 10% type III, according to the Dejour classification. The severity of trochleadysplasia had a significant influence on the outcome of the scores mentioned above.The follow-up MRI showed a good integration of the reconstructed MPFL by gracilis tendon with concomitant positive effect on the patellatilt. Conclusions: The MPFL-plastic is an appropriate therapeutic procedure to recover patella stability even when patella instability comes along with low or moderate trochleadysplasia. It is associated with a low risk of intraand postoperative complications and highly-satisfied patients. To what extend less invasive procedures are suitable for such cases has to be shown in further studies. In cases of high-grade trochleadysplasia the MPFLplastic as a single procedure has to be seen critically. Here, additional procedures like trochleaplastic may be necessary to provide patella stability permanently.

P17-639 The effect of medial patellofemoral ligament reconstruction on patellar tracking Kita K.1, Horibe S.2, Toritsuka Y.3, Nakamura N.1, Shino K.4 1 Osaka University Graduate School of Medicine, Orthopaedic Surgery, Suita, Japan, 2Osaka Rosai Hospital, Orthopaedic Surgery, Sakai, Japan, 3 Kansai Rosai Hospital, Orthopaedic Surgery, Amagasaki, Japan, 4Osaka Prefectural University, Faculty of Comprehensive Rehabilitation, Habikino, Japan Objectives: The medial patellofemoral ligament (MPFL) reconstruction has been performed to treat recurrent patellar dislocation, however, its effect on patello-femoral (PF) articular status and patellar tracking has not been clear. In this study, alteration of arthroscopic chondral status and of patellar tracking pattern was evaluated after MPFL reconstruction. Methods: MPFL reconstruction using a double-looped semitendinosus tendon was performed for 24 knees with recurrent patellar dislocation (16 females and 8 males). Neither tibial tuberosity transfer nor lateral release was performed in any patients. The age at operation ranged from 14 to 33 years(mean 22). The average duration between MPFL reconstruction and second-look arthroscopy was 14 months. At reconstruction and secondlook arthroscopy, chondral status of PF joint and patellar tracking pattern was evaluated arthroscopically during passive knee motion. Chondral injury was graded by Outerbridge classification. Results: No patients had patellar dislocation postoperatively. Before MPFL reconstruction, patella was persistently laterally tilted and shifted throughout the range of knee motion, while the patella was centralized in the femoral groove in all knees immediately after MFPL reconstruction. At second-look arthroscopy, two different patellar tracking patterns were observed. In 10 knees, the patella was centralized throughout the range of motion, while in 14 knees, the patella was tilted and shifted laterally at knee extension and became centralized with the increase in knee flexion. Deterioration of chondral surface was observed in no case. Conclusions: MPFL reconstruction did not aggravate chondral status in PF joint but did not restore normal patellar tracking pattern in all cases.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P17-680 Depression trochleoplasty for patellar instability: report of 20 cases at 34 months follow up Bessie`re C.1, Beaufils P.1, Thaunat M.2, Boisrenoult P.1, Pujol N.1 1 Centre Hospitalier de Versailles, Trauma and Orthopaedic Surgery Department, Le Chesnay, France, 2Hopital Pitie´ Salpe´trie`re, Paris, France Objectives: Trochlear dysplasia is one of the bony abnormalities involved in patellofemoral disorders. It is characterised by a flatness (crossing sign) and a trochlear projection. Thus, it acts as a springboard for the patella (which is abnormal too) and is a supplementary factor of recurrent patellar dislocation. At the first degrees of flexion the patella can’t engage in the trochlea correctly. Associated with other procedures (‘‘a la carte surgery’’), trochleoplasty may enhance the engagement of the patella in the trochlear groove. Standard trochleoplasty is a complex procedure. We chose to use the so called depression trochleoplasty, as described by Goutallier, whose goal is to reduce the trochlear bump. A retro-trochlear wedge, with a proximal basis, is resected. It doesn’t modify the trochlear flatness and the patellofemoral congruence. Hypothesis: In case of severe trochlear dysplasia, the depression trochleoplasty has a low morbidity and is efficient on patellar stability. Methods: This retrospective study reports the results of a consecutive case series of 18 patients (20 knees). 19 knees presented with recurrent patellar instability, 1 with persistent retropatellar pain. Main inclusion criterion was a major trochlear dysplasia with a bump exceeding 5 mm. There were two groups: primary surgery (12 knees), iterative surgery (8 knees). In association with the trochleoplasty, all patients but one underwent a tibial tubercle transfer and 8 underwent a medial patellofemoral ligament reconstruction. Patients were reviewed for clinical (satisfaction index, IKDC, Kujala score) and radiological assessment. Results: Mean follow up was 34 months (3-71). 1 patient was lost for follow up. Apart hardware removal (ATT), there were one knee stiffness requiring an arthroscopic arthrolysis and one sus trochlear ossification requiring an arthroscopic resection, both with good final result. 94% of the patients were satisfied or very satisfied and would have underwent the procedure again. 2 patients (iterative surgery group) had post operative instability (one with a unique traumatic dislocation, one with recurrent subluxations). The mean IKDC score was 67.3% (standard deviation 17.2; range 21 to 92), the mean Kujala score was 80 (17; 33 to 98). 7 knees out of 8 which were painless pre-operatively (all belonging to the primary surgery group) had a slight climatic pain post-operatively. On the contrary, pain present in 11 knees preoperatively improved in 10, didn’t change in 1. No trochlear non union nor fracture has been observed. The mean trochlear projection and the trochlear bump decreased from 4.8mm to -0.8mm and 9.2mm to 3.4mm, respectively. The patellar tilt decrease from a mean of 14 to 6.3. MPFL reconstruction did not influence the postoperative patellar tilt (from 14.3 to 6 without, and from 14 to 5.1 with MPFL). As expected, trochlear angle did not change. Conclusions: Depression trochleoplasty is a reproducible procedure with low morbidity, which can be associated with ATT transfer. At a midterm follow-up, it is efficient on stability, and decreases pre operative pain. It reduces the bump without modifying the trochlear flatness. It can be proposed in cases of failure of a previous surgery or as a primary procedure, when there is a trochlear dysplasia with a bump exceeding 5 mm. In our hands, adding a medial patellar ligament reconstruction is not necessary.

P17-694 Clinical outcome and morphological MRI changes after trochleoplasty for patellar instability due trochlear dysplasia Koch P.P.1, Zingg P.O.1, Meyer D.C.1, Schmitt J.1, Pfirrmann C.W.A.2, Fucentese S.F.1 1 University of Zurich, Department of Orthopaedic Surgery, Zurich, Switzerland, 2University of Zurich, Department of Radiology, Zurich, Switzerland

S269 Objectives: Patellar dislocation can occur in patients with trochlear dysplasia. Trochleoplasty is a surgical procedure which strives to deepen the trochlear groove due patella can engage properly with good clinical results reported in the literature. However, some failures are reported. Aim of this study was to evaluate the clinical outcome and MRI changes after trochleoplasty according to the technique by Bereiter and to find predictive factors for successful results. Methods: The study group consisted of 38 patients (44 knees) with a mean follow-up of 4.2 years (range, 2 to 7.8). Clinical assessment consisted of structured interview and standardized physical examination including the Kujala score. Additionally, the visual analogue scales (VAS) was assessed postoperatively. Special focus was set to the postoperative pain, stability, and satisfaction. The imaging assessment consisted of pre- and postoperative X-rays and MRI. Results: The apprehension test at follow-up was still positive in 11 knees, but only six patients of these 11 reported an instability sensation. One recurrence with dislocation occurred. Pain was unchanged in 27, decreased in 14, and increased in 3 knees. The Kujala score in-creased with a mean from 65 to 85 points. 27 knees were ranked as excellent, 10 as good, two as fair, and five as poor. An increase of cartilage changes in the MRI were found for trochlea. Tibial tuberosity to trochlear groove distance (TTTG), patellar tilt, and translation normalized. No chondrolysis or necrosis of cartilage were remarked. Age had not any correlation with the clinical outcome or MRI-findings, while follow-up time had a correlation with cartilage changes of lateral trochlea. Conclusions: Trochleoplasty is a valuable and relatively safe surgical technique for patients suffering patellofemoral instability with underlying trochlear dysplasia. The clinical results are encouraging even with a relevant number of instability sensations. MRI-findings have only concerning follow-up a correlation. A preoperative predictor had not been detected.

P17-761 Rare case of vertical patellar dislocation Vic¸oso Sousa Fernandes S.I.1, Barbosa A.T.1, Fraga Ferreira J.1, Cerqueira R.1, Basto T.1, Vasconcelos P.1, Lourenc¸o J.1 1 Centro Hospitalar do Alto Ave, Orthopedics and Traumatology, Guimara˜es, Portugal Objectives: Patellar acute dislocation is a relatively common pathology and occurs in the grater majority of the times laterally in the coronal plane. Despite this fact, more rarely other types of dislocations that involve the rotation of the patella can happen. Methods: In this work we describe a rare case of vertical patellar dislocation. Results: Case report: It concerns a 37-years-old without previous diseases, that in 26/01/09 was victim of a direct blow to the lateral side of the right knee and that entered the emergency room with vertical patellar dislocation. Radiographs and clinical examination showed that the patella was submitted to a rotation of 908 in the vertical plane and it was impacted against the femoral lateral condyle without fracture. The initial attempt of reduction of the dislocation in the emergency room with sedation was fruitless. The closed reduction was obtained with general anaesthesia. The patient was immobilized with a cast during 3 weeks and was posteriorly orientated to physical rehabilitation. At the 2 months follow-up the patient was asymptomatic, with normal gait or dismetria; also without alterations of the morphotype of the leg, knee with normal range of movement and without deviation in the frontal or sagittal plane and the patella was centred. The knee had no ligamentar laxity and the patella had normal mobility without slippage or excessive external tilt. At the 3 months follow-up the patient returned to work has a bartender. Conclusions: This case is a rare case of vertical patella dislocation; only 5 cases are reported in the literature. A similar type of vertical dislocation was found also, but in this case the patella was impacted on the intercondylar notch. In this case of rare patellar dislocations, the surgeon should be prepared to realise an open reduction because many times the attempts of reduction even with general anaesthesia are not well succeeded, but that didn’t occur in this case.

123

S270 P17-762 Radiological examination of position of the patella after medial patellofemoral ligament reconstruction Shimizu S.1, Tateishi T.1, Nagase T.1, Nakagawa T.1, Tsuchiya M.1 1 The Fraternity Memorial Hospital, Department of Orthopaedic Surgery, Tokyo, Japan Objectives: Malalignment and abnormalities of patellofemoral joint are often seen in the cases of recurrent dislocation of the patella. Patella alta is thought to be one of the mechanical factor of recurrent dislocation of the patella. We evaluated position of the patella after medial patellofemoral ligament (MPFL) reconstruction radiologically. Methods: From 1998 to 2007, we treated 39 patients of recurrent or traumatic dislocation of the patella ranging in age from 13 to 38 years with MPFL reconstruction. In these cases, we intended for 36 study patients (17 men, 19 women), which radiological examination was available. Position of the patella was measured T/P ratio in lateral view of X-ray by method of Insall-Salvati. According to preoperative T/P ratio, we compared patella alta group (T/P ratio[1.2) with normal group (T/P ratio\1.2). Results: Alteration in T/P ratio from pre-operation to post-operation was 1.38 to 1.32 for patella alta group and 1.10 to 1.15 for normal group. No significant differences were detected between preoperative and postoperative T/P ratio in both groups. Additionally, among 10 patients with bilateral disease, 8 patients were classified with patella alta group. Conclusions: However the cases of dislocation of the patella which indicate patella alta preoperatively tend to improve position of the patella after MPFL reconstruction, there was no significant differences in preoperative and postoperative T/P ratio. Whereas patella alta is one of the mechanical factor of recurrent dislocation of the patella, our study indicated that MPFL reconstruction has little influence on longitudinal position of the patella.

P17-870 Influence of the femoral torsion on patella maltracking Seitlinger G.1, Scheurecker G.2, Fuhrmann G.3, Schmied P.3, Hofmann S.4 1 LKH Stolzalpe, Orthopa¨die, Stolzalpe, Austria, 2Institut fu¨r CT & MRT Diagnostik am Schillerpark, Linz, Austria, 3LKH Stolzalpe, Stolzalpe, Austria, 4Allgemeines u. Orthopa¨disches LKH Stolzalpe, Endoprothetik, Stolzalpe, Austria Objectives: The torsion of the femur has an influence on the patellofemural joint. Nevertheless there is no clear consensus which impact on stability it has. An analyses of the femoral torsion in different sections in patients with patella maltracking and healthy volunteers could lead to a better understanding of this disease. Methods: MR-imaging was performed with a 1.0 Tesla MRI using a body coil for the pelvis and a mobile knee coil for the knee and ankle with the patients fixed on the table. Imaging protocol included T1-FFE sequences, axial views, slice thickness 5mm with an average examination time of 8 min. 30 patients (average age 25 [13-44]) with episodic patella dislocations and 30 healthy volunteers (average age 24 [18 - 48] were included. The femur torsion was measured as described by Berger. Additionally the proximal torsion (angle between column and trochanter minor) and the distal torsion (angle between facies poplitea and trochlea line) were measured. Statistical analysis with paired-samples T-tests were performed additionally. Results: In all patients the relevant bony landmarks could be identified without problems. The patients showed an overall torsion of Ø12 (12retro torsion - 36ante torsion) and the volunteers of Ø11 (5retro torsion - 29ante torsion). The proximal torsion was Ø30 (11 - 55) within the patient group and Ø29 (13 - 46) within the volunteers and a distal torsion of Ø13 (7 - 20) for the patients and Ø17 (9 - 30) for the volunteers were measured. Conclusions: In our patient group the distal femur torsion proofed to be an important factor for patella instability. While the overall torsion and also the proximal torsion seems to have only a poor correlation to patella maltracking.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P17-886 Modified elmslie-trillat technique for treating patello-femoral dysfunction Russu O.1, Nagy O.1, Gergely I.1, Pop T.S.1 1 University of Medicine and Pharmacy Tg. Mures, Orthopedics, Tg. Mures, Romania Objectives: The purpose of our study was to evaluate retrospectively the functional results obtained after a modified Elmslie-Trillat procedure, in a group of patients with patellofemoral dysfunction, regarding pain in the anterior aspect of the knee, recurrent patella dislocation or subluxation and correction of patella alta. This procedure was designed to obtain simultaneously joint decompression and realignment of extensor mechanism, by isolating the tibial tubercle as a pedicle and its antero-medial transfer, with lateral release. Methods: Between 2005-2008, 29 knees in 24 patients with a history of patellofemoral pain for recurrent patella subluxation or dislocation were operated. The mean age was 25 years (range 19-27 years) and the average follow-up period was 24 months (range 7-38 months). Preoperatively the average duration of symptoms was 32 months (range 18-45 months). The clinical evaluation was based on history and physical examination. Standard anteroposterior and lateral radiographs with the knee in 30 of flexion and standard tangential radiograph of the patella with the knee in 45o of flexion were performed. We determined the angle of congruence (index of subluxation) and the patellofemoral angle (index of tilt). The BlackburnePeel index was between 0,75-1,31 (average 1,04) and 18 cases were diagnosed as patella alta. We created a pedicle with a 10-12 mm width and we performed a medial transfer of 8-10 mm. The mean degree of elevation of the tibial tubercle was 11 mm (range 9-13 mm). The evaluation of pain was subjective and the results were considered excellent if the pain was absent during normal physic activities, including weekend sports, good if the pain was present occasional with other than day-to-day activities and failure if the pain remained unchanged or it became worse. Results: 21 knees (72%) had excellent (asymptomatic knee) or good results (occasional pain) and 8 knees (28%) were considered insatisfactory (persistent or worse pain). The congruence angle improved in 26 cases (93%), the sulcus angle had a mean value of 136o and the Blackburne-Peel index was 0,89. In the 18 cases with patella alta, the distal transfer of the tibial tubercle was between 6-10 mm (mean 8 mm) and even though the recurrent patellar dislocation reappeared in 2 knees (6.8%). The radiographic examination confirmed the residual patella alta. All of the patients had a full range of motion of the knee preoperatively and all regained a full range of motion of the knee within an average of 2,4 months after surgery. All osteotomies showed radiographic union without displacement or fractures. In 1 case we had a delayed wound healing that needed a secondary suture. Kneeling has not been a major problem in neither of the case. Conclusions: This modified Elmslie-Trillat procedure improved the symptoms for patients with patellar dislocation. Continuous pain in the anterior aspect of the knee seems to be related to the duration of symptoms rather than the age of onset. The persistence of pain is correlated with the presence of irreversible cartilage lesions of femuropatellar articular surface. Correction of patella alta appears to be an important factor in the prevention of recurrent patellar dislocation.

P17-959 Pathology of the medial patellofemoral ligament - a classification scheme based on magnetic resonance imaging of 25 patients with chronic objective patella instability and 25 healthy volunteers Scheurecker G.1, Seitlinger G.2, Schmidt P.3, Hofmann S.2, Kramer J.1 1 Institut fu¨r CT & MRT Diagnostik am Schillerpark, Linz, Austria, 2 LKH Stolzalpe, Orthopa¨die, Stolzalpe, Austria, 3MRT-Institut Stolzalpe, Stolzalpe, Austria Objectives: To establish a classification scheme for chronic pathology of the medial patellofemoral ligament (MPFL) as in chronic-recurrent lateral patella displacement (LPD). Methods: We analyzed retrospectively the MRI scans of knee joints of 25 patients (mean age 21.2y, range 15 - 36y; 21 females, 4 males) with at

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 least one episode of LPD. This was proven and/or confirmed by several physicians, orthopedic surgeons and prior imaging studies. We enrolled 25 age and gender matched healthy volunteers (HV) (mean age 20.7y, range 17 - 26y; 20 females, 5 males) without any history or signs and symptoms of prior knee injury for MRI scans of one of their knee joints as a means of comparison. The MRI scans were all performed on a 1Tesla open MR unit. An axial T1-weighted sequence with a slice thickness of 3mm was performed and evaluated. The following grading system was invented to describe the MPFL: grade 0 (continuous, straight, no thickened or wavy appearance), grade 1 (continuous, but thickened or wavy), grade 2 (discontinuous at either of its attachments or in its course). The performance of all grades was assessed with calculations of sensitivity, specificity, accuracy and likelihood ratio. Results: Grade 0 was found in 20 HV and in no (0) LPD. Grade 1 was found in 5 HV and in 18 LPD. Grade 2 was found in 1 HV and in 21 LPD. Sensitivity (%) / specificity (%) / accuracy (%) / likelihood ratio are 100/ 80/90/5 for grade 0; 72/80/76/3.6 for grade 1; 84/96/90/21 for grade 2. All 25 LPD are either grade 1 or 2, but also 5 HV are grade 1 or 2, thereof one (1) HV with grade 2. Conclusions: Pathology of the MPFL as in chronic-recurrent LPD can be feasibly classified according to our scheme with the use of MRI. Grade 0 rules out pathology, and grade 2 almost rules in pathology. Grade 1 serves as intermediate with likely pathology.

P17-995 Patella tendinosis: the short term outcomes of an arthroscopic patellar tendon treatment technique Campbell J.1, Murray P.1 1 The Galway Clinic, Galway, Ireland Objectives: To review the short term out-come of an arthroscopic treatment for patellar tendinosis, with regards to pain and function. Methods: The nature and severity of pain was preoperatively and post operatively assessed in 87 patients using the short form McGill pain questionnaire (SF-MPQ) and their pre and post operative function using the Knee and Osteoarthritis Outcome Score (KOOS). 66.7% (58) were males and 33.3% (29) were females, with a mean age of 29.9(14-58) years. Following arthroscopy patients were reviewed at 2-3 weeks and the same questionnaires completed. Results: Results were analysed for difference using Wilcoxon signed rank testing (SPSS, Chicago, Illonois). All sections of the KOOS showed a mean improvement (p\0.05) post op apart from ‘symptoms’ sections (p=0.095). In the SF-MPQ the subjective and visual analogue scale (VAS) sections showed a mean improvement also. However the affective section did not have a significant improvement (P=0.072). Conclusions: We have shown a favourable outcome with arthroscopic treatment which also allows for exclusion and treatment of other conditions which can cause patello-femoral joint syndrome. In our cases we found an improvement at 2-3 weeks post-op in the nature and severity of symptoms using the SF-MPQ. Although ‘symptoms’ had not improved significantly in the KOOS, the questions in this section are not specific to tendon pathology and is therefore not sensitive enough to changes in this condition.

P17-1045 In vivo positioning analysis of Medial patellofemoral ligament reconstruction Servien E.1, Debarge R.2, Lustig S.3, Demey G.2, Neyret P.4 1 Centre Albert Trillat, Hospices Civils de Lyon, Lyon-Caluire, France, 2 Centre Albert Trillat, Lyon-Caluire, France, 3Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 4Hopital Croix-Rousse Centre Livet, Chirurgie Orthopedique, Lyon, France Objectives: The purpose of the study was to analyze the femoral graft positioning in MPFL reconstruction.This work provides an in vivo description for femoral tunnel placement in medial patellofemoral ligament reconstruction.

S271 Methods: Several techniques have been described for reconstruction of the medial patellofemoral ligament. The anatomical insertion of the medial patellofemoral ligament has been defined. However, the femoral positioning of post-operative graft fixation has not been studied. We reported a prospective series of 29 MPFL reconstruction with a minimum follow-up of 12 months. The tunnel positioning analysis was performed using plain radiographs and MRI at one year follow-up. Results: We analysed 29 femoral tunnels. 20 femoral tunnel (69%) were considered to be in good position on plain radiographs. On MRI we found 19 femoral tunnels (65%) in a proper location, 5 (17.5%) in a high position and 5 in anterior and/or high position. Conclusions: The study highlights that reproducible medial patellofemoral ligament reconstruction is difficult. The surgical procedure continues to be improved and finding a reliable technique to anatomically place the graft remains challenging.

P17-1066 Anterior knee pain treated with hip osteotomy - 9.7 years average follow-up on 66 osteotmies in patients with miserable malalignment and excess femoral anteversion Teitge R.1, Latteier M.1, Torga Spak R.2 1 Wayne State University School of Medicine, Orthopaedic Surgery, Grosse Pointe Shores, MI, United States, 2Instituto de Traumatologia y Rehabilitacion, Buenos Aires, Argentina Objectives: The Objective was to measure the clinical outcome of a cohort of patients seen with anterior knee pain and an associated increase of internal femoral torsion who were treated by intertrochanteric external rotational femoral osteotomy. Methods: All patients with anterior knee pain were subjected to a standardized clinical examination, plain radiographs, and when indicated a standardized CT rotational study to determine femoral torsion, TT-TG, and tibial torsion. 53 patients elected to undergo femoral osteotomy (19 patients underwent osteotomy on both femurs). The patients underwent regular examination and during the study period (2006) 92% were available for follow-up and collection of data to determine Kujala, Lysholm and Tegner scores. Additionally complications were tabulated and patients were questioned regarding any improvement or worsning of their condition, the percentage of estimated improvement, and whether they would undergo surgery again. The average time from surgery to follow-up was 9.7 years with a range of 2 to 17 years. Results: The average Kujala score increased from 53 pre-operatively to 86 at follow-up, the Lysholm score increased from 49 to 89, the Tegner score increased from 2.2 to 4.0. The average percentage of improvement defined by the patients was 91%. Fifty-five percent of patients described 100 percent relief of anterior knee pain. 95 percent of patients stated they would undergo the surgery again. Three patients would not, primarily because the degree of improvement was not considered to be enough. No patient described their condition as being worse although those with a more severe PF arthrosis, those older and those with multiple prior surgeries tended to have less improvement than younger patients and particularly those without prior surgery. Conclusions: Malalignment of the femur in the horizontal plane (torsional malalignment) is one factor of many which can contribute to the syndrome of anterior knee pain. In this cohort of patients in whom torsional malalignment was diagnosed and considered a more significant factor than patellar instability, trochlear dysplasia, abnormal TT-TG, varus-valgus malalignment, patella alta or articular cartilage pathology the improvement in symptoms following corrective osteotomy was significant. Long bone maltorsion should be evaluated and addressed in patients with anterior knee pain.

P17-1093 Medial femoropatelar ligament reconstruction for acute and chronic femoropatelar instability - a minimum 2-year follow-up Bitar A.C.1, Camanho G.L.2, D’Elia C.O.3, Demange M.K.4, Viegas A.d.C.2 1 Instituto Vita, Sa˜o Paulo, Brazil, 2IOT - HCFMUSP, Sa˜o Paulo, Brazil, 3 Instituto Vita, Ortopedia, Sa˜o Paulo, Brazil, 4IOT - HCFMUSP, Sao Paulo, Brazil

123

S272 Objectives: Our objective is to present a case series with preliminary results from using a technique for reconstructing the medial femoropatellar ligament (MFPL) with a medial strip from the patellar ligament, regarding recurrences and functional scores. Methods: Between March 2004 and June 2007, 46 patients underwent surgery using the MPFL reconstruction technique with a medial strip of patellar ligament, in accordance with the technique described by Camanho et al (2007) (SURGICAL TECHNIQUE). Only 34 of the patients presented a minimum of two years of follow-up and therefore these were the patients analyzed in this case series. Out of these 34, 22 patients underwent operations in the acute phase, less than four weeks after their injuries. The remainders were chronic cases (more than four weeks after their injuries). Results: Our patients presented a mean Kujala score of 89,23 (minimum of 62 and maximum of 100). Only one patient presented episodes of subluxation and none of the patients presented recurrences of femoropatellar instability. MPFL injuries due to acute dislocation of the patella are well documented, the MPFL is considered the principal restraint to lateral displacement of the patella. There is still some controversy in the literature regarding the treatment for acute dislocation. The management methods of immediate reconstruction and/or repair versus conservative treatment are topics for debate. In cases of chronic femoropatellar instability, the insufficiency of the medial containment structures for lateral luxation has also been well demonstrated. Various techniques have been described for treating this dislocation when conservative treatment fails. Among them are distal, proximal and combined realignment techniques and lateral release. Among the patients with chronic femoropatellar instability with more than two episodes of dislocation and tomographic measurements of the AT-TG (anterior tuberosity-troclear groove) that were less than 20 mm, we used MPFL reconstruction with a medial strip of patellar ligament. This technique could also be performed in acute luxation cases. Conclusions: Our case series using the MPFL reconstruction technique with a medial strip of patellar ligament presented excellent functional results from the Kujala score. The technique was also effective in improving symptoms from femoropatellar instability.

P17-1154 Non-operative management of patellofemoral dislocation: results of a survey of members of the ESSKA Arthroscopy Committee Lynch A.1, Arendt L.2, Dejour D.3, Zaffagnini S.4, Snyder-Mackler L.5 1 University of Delaware, Newark, United States, 2University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States, 3COROLYON, Orthopaedic Surgery, Lyon, France, 4Rizzoli Orthopaedic Institute, Biomechanics Lab, Bologna, Italy, 5University of Delaware, Department of Physical Therapy, Newark, United States Objectives: Patients experiencing a primary, isolated lateral dislocation of the patellofemoral joint may be treated non-operatively. There is no evidence based protocol for managing these cases. Members of the ESSKA Arthroscopy Committee were surveyed for their recommendations on the best practices to prevent further complications and give the patient the best opportunity to succeed. A consensus opinion was sought from this survey to guide clinical practice. Methods: An on-line survey was created and edited to reflect important aspects of non-operative care for patellofemoral dislocations. Twenty-six members of the ESSKA Arthroscopy Committee were invited to respond to the survey. Surgeons were questioned about immobilization, bracing, weight bearing, range of motion and quadriceps strengthening. Surgeons were asked to comment on timing, duration and methods of treatment in a forced choice paradigm and were also given an opportunity to provide comments. Results: Seventeen of 26 (65%) surgeons completed the survey. A majority (73%) of respondents recommended waiting at least 2 weeks to begin rehabilitation activities, most favoring constant immobilization. A straight leg immobilizer was most frequently recommended, with others opting for a hinged immobilizer or a brace with patellar stabilization. Weight bearing was not typically limited. A variety of restrictions on the acceptable range of motion for both open and closed kinetic chain exercises were recommended although there was a tendency to allow unrestricted range of motion when exercise was allowed. Open and closed

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 kinetic chain exercises were initiated variably throughout rehabilitation, from as early as immediately after injury to 8 weeks post-injury. Immobilization time frames were divided evenly among three recommendations: no immobilization, 2 weeks of immobilization or more than 4 weeks of immobilization. Equivocal responses were received for limiting both active and passive range of motion early on after dislocation, however, a majority opposed continuous passive motion. Conclusions: Immobilization is generally recommended with a long leg brace for at least 2 weeks. Full weight bearing is allowed early after injury rehabilitation. Range of motion and quadriceps exercise are more controversial rehabilitation topics, with broad ranges of recommendations. From these results, rehabilitation professionals can be confident in the need for nearly constant immobilization early after injury with a gradual resumption of range of motion and quadriceps exercises, however, no single ‘‘safe’’ knee range of motion for exercises was identified.

P17-1160 Arthroscopic patellar denervation for patellofemoral pain and normal patellar alignment. Results with a minimum 5 year follow-up Vega J.1, Marimo´n J.2, Golano´ P.3, Perez-Carro L.4, Sanchez S.1, Aguilera J.M.1 1 Hospital Asepeyo Sant Cugat, Orthopaedics, Sant Cugat del Valles, Spain, 2Fundacio´ Hospital Salut Alt Emporda`, Orthopaedics, Figueres, Spain, 3University of Barcelona, Laboratory of Arthroscopic and Surgical Anatomy, L’Hospitalet de Llobregat, Spain, 4Hospital Universitario Marque´s de Valdecilla, Orthopaedics, Santander, Spain Objectives: Nociceptive receptors are richly distributed in the peripatellar soft tissue. A lesion in this region would lead to theoretical desensitization of the patella, in what we term ‘‘arthroscopic patellar denervation’’. This study presents our experience and the results with a minimum fiveyear follow-up of arthroscopic patellar denervation in patients with intractable patellofemoral pain and normal alignment. Methods: Between 2002 and 2004, fifteen patients with patellofemoral pain were considered for the patellar denervation procedure (10 women, 5 men; mean age 32.6 years, range 20-49). The clinical outcome was evaluated using the Grana and the Kujala scores, at 2 and 5 year follow-up. Statistical analysis was performed using the Wilcoxon and Kendall’s Tau tests. The level of significance for these studies were set at p\0.05. Results: At five year follow-up, the clinical status had improved considerably in all patients and all but two had returned to their usual daily activity. None of the patients reported that their knee pain was worse on follow-up than before arthroscopic patellar denervation. Preoperative assessment with the Grana score placed all patients in categories D (restriction in extreme activity) and E (restriction in daily activity). At follow-up, the Grana score placed eight patients in group A, four in group B and three patients in categories C and D, indicating satisfied results in twelve patients. According to the Kujala score, the preoperative mean score was 70 (ranged from 52 to 83). At follow-up, all patients reported an improvement in the Kujala score. The postoperative mean score at five year follow-up was 91 (ranged from 69 to 100). Conclusions: These results suggest that arthroscopic patellar denervation can be effective in cases of intractable patellofemoral pain and normal alignment or chondromalacia patellae, with a minimal morbidity. The clinical relevance of this study is that arthroscopic patellar denervation decreases pain sensitivity in the anterior region of the knee and is effective in this patient population.

P17-1306 Comparative study of quantitative stress radiography of the patella under non-anesthesia and anesthesia Niimoto T.1, Deie M.1, Adachi N.1, Nishimori M.1, Nakamae A.1, Nakasa T.1, Ochi M.1 1 Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan Objectives: When deciding the treatment policy for patellar instability, since 1988, we have performed quantitative stress radiography of the patella and

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 assessed the inward and outward lability in order to decide application of lateral release. In this study, we radiologically measured and comparatively studied different types of quantitative stress radiography of the patella. Methods: The subjects comprised 21 knees of 19 patients (7 male subjects, 12 female subjects) with an average age of 20 years old (11 to 49 years old) who agreed to undergo perioperative stress radiography during patellar stabilization for patellar instability between March 2007 and September 2009. Before surgery, 20N of stress was inwardly and outwardly applied to the patella at the outpatient department in order to obtain axial images of the patella with a knee flexion of 45. Under perioperative anesthesia, a soft wire was inserted in the center of the patella from the outside to the inside and 20N of stress was similarly applied in order to obtain axial images. Radiological assessments were performed using the Medial Stress Shift Ratio (MSSR) and Lateral Stress Shift Ratio (LSSR). Results: The average MSSR was 7.82±13.7% under non-anesthesia and 4.62±15.78% under anesthesia, which showed a significant decrease, and the inward lability increased. The average LSSR was 33.39±18.05% under non-anesthesia and 45.24±33.57% under anesthesia, which showed a significant increase, and the outward lability increased. After all of the subjects were examined, MPFL reconstruction was performed. However, lateral release was not performed for one subject in whom the inward instability under anesthesia had significantly increased, though lateral release surgery was concomitantly performed for the other subjects. Conclusions: In the present study, in addition to conventional quantitative stress radiography of the patella, similar stress radiography was performed under anesthesia and the inward and outward balance was assessed, and it was revealed that there were differences in the assessments depending on whether anesthesia was used. In the subjects who did not undergo lateral release, the differences in the measurements depending on the use or nonuse of anesthesia were large, and it was discovered that medial instability that was not revealed under non-anesthesia was present. Therefore, it was believed that it was necessary to make determinations based on both of these indices and thereafter consider the surgical indications when assessing patellar instability.

P17-1322 Clinical outcome after MPFL reconstruction in primary patellar dislocation vs. salvage surgery Verdonk P.1, Mulliez A.1, Thibaut E.1 1 University Hospital Ghent, Dept. Orthopaedic Surgery and Traumatology, Gent, 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).Prospective cohort study with minimal follow-up of 1 year. Methods: 47 patients have been treated between April 2007 and September 2008. Thirty-two patients received and MPFL reconstruction of which 11 were in combination with a transposition of the tibial tubercle (group 1). Fifteen of the 47 patients already had a patellar stabilizing operation in the history of which 2 Insall plasty, 6 trochleoplasty and 11 TTT (group 2). The clinical follow up was evaluated using KOOS and KUJULA scores preoperatively and at 1 year. The minimal follow-up was 1 year. Results: For group 1, the KOOS and KUJALA score increase significantly, while only KOOS subscores pain, function, ADL and KUJALA increase in group 2. The net gain for the KOOS subscales and KUJALA is higher in group 1. Overall, the outcome is significantly better for group 1 vs group 2. No patients experienced patellar dislocations in the follow-up. 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. The clinical outcome is nevertheless significantly higher in primary cases.

S273 P17-1327 Prospective 1 year results after anatomic reconstruction of the medial patellofemoral ligament in double bundle technique Beitzel K.1, Mu¨nch M.1, Imhoff A.1, Scho¨ttle P.B.1 1 Technical University Munich, Klinikum rechts der Isar, Department of Orthopaedic Sports Medicine, Munich, Germany Objectives: Since proof of the biomechanical importance of the medial patellofemoral ligament (MPFL) for patellofemeral instability (PFI) the reconstruction of the MPFL gained more importance. Aim of this study was to evaluate the clinical and radiological results of the anatomical reconstruction with a autologous gracilis tendon after one year postoperatively. A special focus has been given to initial and revision surgery. Methods: 02/07 till 05/08 50 patients (33$ /17#, mean-age: 23 a) underwent reconstruction of the MPFL (gracilis tendon & suture anchor fixation). 2 of these had to be excluded (1x traumatic dislocation, 1x additional surgery necessary). Dysplastic Trochlea type A & B has been found in 37, type C & D in 11 patients. 25 patients underwent revisionsurgery because of persisting PFI after lat. release and med. reefing (22) or transfer of the tuberosity. (3). in 20 cases additional surgery as cartilage therapy, femoral osteotomy, closure of the lateral release had to be done. Tegner, Kujala and IKDC had been evaluated preoperatively and one year postoperatively. Results: Table 1 Detailed results Tegner-score Pre-OP

Altern. Kujala-score

12m post OP Pre-OP

IKDC-score

12m post OP Pre-OP

12m post OP

MPFL isolated / initial-OP

4,0

7,3 (±2,6) (±2,0)*

64,5 90,1 (±21,8) (±12,3)*

63,8 90,4 (±18,8) (±12,4)*

MPFL isolated / revision-OP

2,1

5,5 (±0,7) (±2,9)*

51,0 76,5 (±17,1) (±25,3)*

50,3 75,9 (±12,5) (±27,9)*

MPFL combined/ initial-OP

2,3

7,3 (±3,2) (±1,7)*

42,3 92,3 (±33,5) (±8,0)*

48,9 90,5 (±32,0) (±4,3)*

MPFL combinied/ revision-OP

2,3

3,2 (±1,6) (±1,1)*

40,5 56,0 (±15,1) (±14,1)*

42,1 59,4 (±17,8) (±9,6)*

Detailed results are shown in table 1 (*=p\0,05) Conclusions: The minimal invasive reconstruction of the MPFL with an autologeous gracilis graft offers very good clinical and radiological results. It has been shown that patellofemoral stability can be restored and postoperative pain can be impeded. The significantly better outcome of patients with initial surgery in comparison with those who underwent revision surgery let conclude that the MPFL reconstruction and additional surgery if necessary should be chosen as an initial intervention in case of patellofemoral instability to achieve an optimal result.

P17-1379 How does the patella track following knee replacement: a novel use of ultrasound and motion capture Monk A.1, Simpson D.2, Cooke W.1, Chen M.1, Mellon S.1, Gill H.3, Price A.4, Dodd C.5, Murray D.6 1 University of Oxford, Oxford, United Kingdom, 2Nuffield Orthopaedic Centre, Oxford, United Kingdom, 3Nuffield Orthopaedic Centre, Orthopaedics, Oxford, United Kingdom, 4Oxford University, Nuffield Department of Orthopaedic Surgery, Oxford, United Kingdom, 5Nuffield Orthopaedic Centre, Department of Orthopaedic Surgery, Oxford, United Kingdom, 6University of Oxford, Department of Orthopaedic Surgery, Oxford, United Kingdom Objectives: Total knee replacement is the standard treatment for end-stage osteoarthritis when conservative measures have failed. Registry data suggests that the patella is re-surfaced in around 50% of cases. Previous

123

S274 attempts to study the coronal plane patellofemoral kinematics have suffered from the patella being obscured by the components and/or metal artefact. The aim of this study was to assess whether there was any significant difference in patellofemoral kinematics between normal and replaced knees. Methods: 30 patients were recruited into three groups; normal healthy volunteers (Healthy), TKR patients with the patella resurfaced (TKR?P) and TKR without the patella resurfaced (TKR-P). A Motion Capture System was used to image multiple 4D and 2D ultrasound probes providing coordinates for the patella and bony landmarks on tibial and femoral segments, during a squat exercise. At each flexion angle, patellofemoral kinematics were described relative to both the femur and tibia in 6 degrees of freedom, including medial-lateral shift, PTA, PFA, patella spin and tilt. Results: The accuracy of the motion capture system is ±0.1mm. The accuracy of registering the ultrasound images within the motion capture system is ±2.2mm. Preliminary data shows medio-lateral movement of the patella between 0 and 90 degrees relative to the tibia was 18.6mm laterally for the normal group, and 11.69mm medially for the TKR group (TKR?P 17.65mm, TKR-P 5.69mm). Conclusions: We present a new, in-vivo technique for measuring patellofemoral kinematics. Preliminary data suggest that many aspects of patellofemoral kinematics are absent following TKR whether or not the patella is resurfaced.

P17-1412 Double bundle reconstruction of the medial patellofemoral ligament a prospective evaluation of isolated initial- and revision surgery with a minimum follow up of one year Beitzel K.1, Mu¨nch M.1, Beermann I.1, Imhoff A.1, Scho¨ttle P.B.1 1 Technical University Munich, Klinikum rechts der Isar, Department of Orthopaedic Sports Medicine, Munich, Germany Objectives: Due to biomechanical and retrospective studies, the reconstruction of the medial patellofemoral ligament (MPFL) in a double bundle technique is nowadays an established surgical procedure for treatment of chronic patellofemoral instability (PFI). However until today there is still a lack of prospective studies evaluating these procedures. Therefore it was the goal of this prospective study to evaluate the clinical and radiological outcome of isolated reconstruction of the MPFL in double bundle technique with a one year follow up. Additionally a special focus was given on differentiation in between initialand revision surgery. Methods: Inclusion criteria for this study was an isolated anatomic double bundle MPFL-reconstruction using an autologous gracilis tendon for treatment of a PFI with a positive apprehension sign form 0 to 30 of flexion with a minimum follow up of 12 months. Patients with additional surgical procedures due to femoral malrotation, valgus malalignement, advanced patellofemoral osteoarthritis or a convex trochlear dysplasia have been excluded. From 06/07 to 09/08, a total of 29 patients (20 female; 8 male) with a mean age of 21,0 years (±6,5) could have been included in this study. Eighteen of these (mean age 20,6 years ±6,7; 14 female; 4 male) underwent initial surgery, while previous surgery with persisting PFI have been performed in 10 patients (mean age: 21,7 years ±6,4; 6 female, 4 male). These were a medial reefing in six, an insufficient medial anchor-refixation in one and a lateral release in four cases. IKDC and Kujala score have been recorded preoperatively as well as 3, 6 and 12 months postoperatively. The patella height has been evaluated on lateral radiographs, using the Caton Dechamps index (CDI), Patella shift, -tilt and cartilage degeneration according to ICRS have been recorded on axial MRI pre- as well as 12 months postoperatively. Results: No patient reported a redislocation at the 12 month follow up. The results of the evaluated scores are shown in table 1. Changes regarding cartilage degeneration in the patellofemoral joint have not been observed at the 12 month follow up. Detailed results are shown in Table 1.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Table 1 Detailed results (*=p\0,05) MPFL initial Pre OP IKDC Patella shift Patella tilt Patella height (CDI)

MPFL revision 12 m post OP Pre OP

12 m post OP

61,1 (±17,2) 89,5 (±12,3)* 52,8 (±14,4) 81,9 (±21,9)* 6,6 (±4,3)

3,9 (±2,1)*

6,2 (±3,3)

3,2(±2,3)*

20,5 (±5,4)

14,5 (±6,8)*

20,5 (±10,5)

9,7 (±6,1)*

1,3 (±0,2)

1,0 (±0,1)*

1,2 (±0,2)

1,1 (±50,1)*

Conclusions: This prospective study could show good and excellent results of the presented surgical technique at a one year follow up in initial as well as in revision cases. All groups showed significant decrease of patella tilt, shift and CDI without increase of cartilage degeneration. This suggests normalization of patella positioning by the presented anatomic technique. Anatomic reconstruction of the MPFL generates no cartilage degeneration at the 12 month follow up, if the graft is tensioned and positioned properly. Patients who underwent unsuccessful non-anatomical surgery before showed initially lower results regarding the subjective scores. Gaining an equal benefit postoperatively, the results are still slightly lower than in the initial group 12 months postoperatively. Therefore the presented anatomic procedure should be used as initial surgery for stabilization in PFI.

P17-1483 Restoration of knee volume using selected arthroscopic releases Dragoo J.1, Miller M.1, Vaughn Z.1, Schmidt J.1, Handley E.1 1 Stanford University, Orthopaedic Surgery, Redwood City, United States Objectives: Inflammation and subsequent fibrosis, adhesions or plica may limit normal capsular compliance and decrease volume capacity of the knee. Hypothesis: Patients with fibrosis, anterior interval scarring, adhesions or symptomatic plica will have decreased knee volumes when compared to controls, and selective arthroscopic releases will restore volume to normal levels. Methods: Part I: Knee volume and pressure were recorded in 25 fresh-frozen human cadaveric knees, and the maximum volume capacity was identified before capsular disruption. Part II: 51 patients undergoing arthroscopy were divided into two groups based on intra-operative volume assessment at 50mm H20 pressure: a control group (n=13) with normal volume ([ 1 SD of mean established in Part I) and an experimental group (n=38) knees with deficient volume (\1 SD below mean). The experimental group underwent volume-changing procedures such as lysis of adhesions, anterior interval release, and plica resections, while the control group underwent volume neutral procedures including meniscal or chondral surgery. The knee volume was then reassessed after arthroscopy. Results: The average volume capacity of the knees in the cadaveric study was 87.5ml (± 21.7ml) with an overall range of 50-120ml’s. There was no statistical difference between the pre-surgical (97.9 ± 24.2ml) and postsurgical volumes (97.7 ± 23.0ml) in the control group; p=.92. The presurgical volume in the experimental group (47.5 ± 12ml) was significantly lower than the control group (97.8ml); p\0.001. The experimental group volume increased to (83.0 ± 24.0ml) after surgery; p\0.001, with an average change in volume of 40.7% (35.4ml). The change in volume postsurgery was significantly greater in the experimental group (36.7ml) versus the control group (3.4ml); p\ 0.001. Clinical outcome data with an average 6 month follow-up showed similar reduction in VAS pain scores in experimental group (from 4.30 to 1.50, p\0.001) and control group (from 3.80 to 1.30, p\0.02) after surgical

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 intervention. Lysholm knee score improved (from 45.77 to 68.82, p\0.001) in the experimental group, and (from 52.78 to 65.89, p\0.01) in the control group. Conclusions: The average volume of the human knee is between 65 and 110 ml. Patients with chronic knee pain may have a diminished knee volume and selected arthroscopic releases can restore knee volumes to near normal levels. Pre-arthroscopic injection of 60cc of saline can be used to help identify patients with volume limitation.

Knee: osteotomy P18-28 Early complications of medial opening wedge high tibial osteotomy using autologous tricortical iliac bone graft and T-plate fixation. How to avoid loss of correction and lateral condyle fracture Nha K.1, Chae D.2, Choi H.3, Han S.4, Shetty G.5 1 Inje University, Ilsanpaik Hospital, Orthopedic Surgery, Goyangsi, Korea, Republic of, 2Bonplus Hospital, Orthopadic Surgery, Seoul, Korea, Republic of, 3Inje University, Ilsanpaik Hospital, Goyangsi, Korea, Republic of, 4Korea University, Anam Hospital, Orthopaedic Surgery, Seoul, Korea, Republic of, 5Breach Candy Hospital, Orthopaedic Surgery, Mumbai, India Objectives: Despite several advantages of medial opening wedge high tibial osteotomy(HTO), this procedure has been noted to have a high rate of complications. The purpose of this study was to evaluate the early complications of medial opening wedge HTO using autologous tricortical iliac bone graft and lower profile T-plate fixation. Methods: We prospectively evaluated the early complications of 138 arthroscopic examination and medial opening wedge HTO done using autologous tricortical iliac bone graft and lower profile T-plate fixation in 128 patients. The preoperative diagnosis was primary osteoarthritis in 134 knees and medial femoral condyle osteonecrosis in 4 knees. At a mean follow-up of 36.8 months (13 to 78), 26 knees (18.8%) developed complications. Results: Complications varied from osteotomy site infection(3), loss of correction(6), broken screws(6) and lateral tibial plateau fracture(3) to joint fluid leakage(3), pseudoaneurysm(1) and iliac bone fractures(3). There was no nonunion. Conclusions: Most complications associated with this procedure were a result of technical error in initial 50 cases and could be minimized with proper planning and taking adequate intra-operative precautions. Even though there were 5 lateral cortical fractures in safe zone(lateral capsule area), there was no nonunion and loss of correction. Using the ‘‘safe zone’’ technique and penetrating the lateral cortex with Steinmann pins may help to avoid complications such as loss of correction and lateral tibial plateau fractures.

P18-168 Surgical and clinical experience with a novel technique and implant system for medial opening wedge osteotomy of the tibia Engebretsen L.1, Litchfield R.2, Dimmen S.3, Slynarski K.4, Parker D.5, MacDonald P.6, Ho C.7 1 Ullevaal University Hospital, Orthopaedic Center and Oslo Sportstrauma Research Centre, Oslo, Norway, 2Fowler Kennedy Sport Medicine Clinic, Surgery, London, Canada, 3Ullevaal University Hospital, Orthopaedic Center, Oslo, Norway, 4Sports Medicine Centre, Complex Medical System, Warszawa, Poland, 5Royal North Shore Hospital, Sydney Orthopaedic Research Institute, Sydney, Australia, 6Pan Am Clinic, University of Manitoba, Surgery, Winnipeg, Canada, 7Vail Imaging Center, Vail, United States Objectives: To determine the safety and efficacy of the iBalance AKRFX Surgical Instrument System and the iBalance iFX PEEK implant. Methods: Thirty-two patients underwent medial opening wedge osteotomies using the iBalance AKRFX Surgical Instrument System and the iBalance iFX PEEK implant. Each of the 32 study patients was case-matched to a historical control patient who had undergone MOW-HTO using a

S275 commercially available plate and screw system. The primary outcome of the study was time to unassisted ambulation, secondary outcomes included disease specific quality of life (KOOS), SF 36, adverse events and radiographic assessments of union and maintenance of correction. Clinical union was primarily measured by the patient’s ability to completely withstand full weight bearing and to walk without the use of crutches. Adverse events were studied and compared in terms of severity and frequency. Results: At six months, more patients in the study group (81%) met the clinical criteria for osteotomy healing than did those in the control group (55%), but the difference did not reach statistical significance (p \ 0.06). Fewer patients in the study group exhibited radiolucency around the implant at 6 and 12 months (p=.0005, p=.001 respectively) or bone resorption within the osteotomy at 12 months (p=0.03). With regard to safety, there were no new types of adverse events in the study group and the rates and severity of adverse events in the two groups were similar. One type of adverse event, tibial plateau fracture, was found to be highly technique dependent. An intraoperative check prior to the osteotomy cut reduced the rate of this complication from 17% in the first 12 patients to 0% in the final 20 patients. Both groups showed similar and statistically significant improvement in symptoms and in function. In the study group, all five subscales of the KOOS score showed statistically significant (P \ 0.001) increases from baseline to 12 months. The largest increase (35 points) was seen in the pain subscale. Both groups showed an approximately 10-point improvement in the SF-36 physical component score at 12 months, which was also statistically significant (P \ 0.001). Conclusions: The AKRFX tibial osteotomy implant system achieved excellent rates of union, maintenance of correction and clinical improvement equal to or superior to current standards of care.

P18-215 Comparison of the change of femorotibial angle and tibial posterior slope between closed wedge high tibial osteotomy and opening wedge high tibial ostotomy Hanada M.1, Takahashi M.2, Koyama H.2, Matsubara T.2, Furuhashi R.2 1 Hamamatsu Red Cross Hospital, Orthopaedics Surgery, Handayama,Hamamatsu City,Shizuoka, Japan, 2Hamamatsu University School of Medicine, Orthopaedic Surgery, Hamamatsu, Japan Objectives: High tibial osteotomy (HTO) is commonly performed for unicompartmental osteoarthritis of the knee. In the past, it was considered that HTO corrects the femorotibial angle (FTA) in the coronal plane of the knee, but changes in the tibial posterior slope (TPS) in the sagittal plane were not considered when planning HTO. Several studies have examined increased TPS after opening wedge HTO, and a few have examined closed wedge HTO. To our knowledge, only one study has been published investigating the difference of changes in TPS between closed wedge HTO and ope wedge HTO. The purpose of this study was to compare the difference of changes of TPS between closed wedge HTO and open wedge HTO and investigate any correlation of the degree of correction of FTA in the coronal plane and changes to TPS in the sagittal plane. Methods: From January 2005 to March 2009, patients with medial unicompartmental osteoarthritis who underwent closed wedge or open wedge HTO were included in this study. There were 24 patients (30 knees), 5 males (6 knees), 19 females (24 knees), with an average age of 65.7 years (range: 52-78 years). 19 patients (23 knees), with an average age of 65.0 years (range: 52-75 years) underwent the closed method and 5 patients (7 knees), with an average age of 67.9 years (range: 56-78 years) the open method. Antero-posterior and lateral long leg weight bearing radiographs of the knee were obtained both preoperatively and postoperatively, and FTA and TPS were measured. TPS was measured based on the perpendicular line to the posterior tibial cortex. Results: FTA on preoperative radiographs of patients with closed wedge HTO averaged 185.3 degrees, and FTA on preoperative radiographs of patients with open wedge HTO averaged 182.6 degrees. FTA on postoperative radiographs of patients with closed wedge HTO averaged 168.3 degrees, and FTA on postoperative radiographs of patients with open wedge HTO averaged 172.9 degrees. There was no significant difference between FTA with closed group and open group both pre- and

123

S276 postoperative. TPS on preoperative radiographs of patients with closed wedge HTO and open wedge HTO averaged 7.2 degrees and 6.7 degrees, respectively. There was no significant difference between preoperative TPS with closed group and open group. TPS on postoperative radiographs of patients with closed wedge HTO averaged 2.9 degrees, a decrease of 4.3 degrees from preoperative TPS. TPS on postoperative radiographs of patients with open wedge HTO averaged 11.7 degrees, an increase of 5.0 degrees from preoperative TPS. Conclusions: There was no significant difference between postoperative FTA of the closed and opening groups. There was a significant difference between pre- and postoperative TPS of both closed group and open group. There was a significant difference between postoperative TPS of the closed and open groups. Several publications suggested that the change of TPS after closed and open wedge HTO were mainly caused by surgical technique. The change of degree of FTA between pre- and postoperative radiographs was not correlated to the change of TPS.

P18-422 The precision of intra-operative fluoroscopy in open-wedge valgus high tibial osteotomies van de Pol G.1, van Kampen A.2 1 Radboud University Nijmegen Medical Center, Nijmegen, Netherlands, 2 Radboud Unversity Nijmegen Medical Center, Department of Orthopaedics, Nijmegen, Netherlands Objectives: The aim of this study was to evaluate the results of open wedge high tibial osteotomies (HTO) using a specially designed rod, replacing the more frequently used electrocautery cord for assessing intraoperative coronal alignment. We hypothesised that there is a strong relationship between intra-operative alignment and the final alignment on standing full-length radiographs (FLR). Also we hypothesised that the actual used wedge size does not correspond to the preoperative planned correction angle. Methods: The results of 27 HTO’s were evaluated using pre- and postoperative FLR’s and intra-operative fluoroscopy images. The mechanical axis (MA), expressed as a percentage of the tibial width (0% is medial tibial edge, 100% is lateral edge) and the hip-knee-ankle angle (HKA) were calculated and analysed. Results: The preoperative standing MA was 26.6%, or about halfway the medial tibial plateau. The MA was corrected to 56% intra-operatively, meaning a slight valgus. Postoperative, on the standing radiograph, the mechanical axis had shifted significantly to 62% (95% CI=-9.2 to -2.7; p\0.01) compared to the intra-operative axis. The HKA angle changed 8.2 from 5.1 varus preoperative to 3.1 valgus postoperative, corresponding to the desired angle of 3. The used wedge size also closely corresponded to the planned correction angle. Conclusions: The mechanical axis can be assessed intra-operative by using a rigid rod, which proves to be an accurate and easy to use tool. When aiming at the lateral tibial eminence, the final postoperative standing mechanical axis showed a slight valgus shift compared to the intra-operative images. However, the final outcome of 3 valgus was satisfactory. Preoperatively calculating the wedge size remains important for surgical planning, as it closely corresponds to the actual used wedge size.

P18-452 High tibial osteotomy modulates the MRI intensity of the osteochondral graft plug in the early postoperative period Mukai S.1, Nakagawa Y.2, Matsusue Y.3, Kobayashi M.4, Nishitani K.4, Tsubouchi N.1, Nakamura T.4 1 Kyoto Medical Center, Department of Orthopaedic Surgery, Kyoto, Japan, 2 Kyoto Medical Center, Orthopedic Surgery, Kyoto, Japan, 3 Shiga University of Medical Science, Department of Orthopaedic Surgery, Otsu, Japan, 4Kyoto University, Orthopaedic Surgery, Kyoto, Japan Objectives: The postoperative evaluation of the osteochondral graft is uncommon and there are few reports of the fate of osteochondral plug. In

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 this report, we retrospectively research whether the combination of high tibial osteotomy(HTO) modulate the MRI evaluation of the osteochondral plugs. Methods: Between March 1998 and October 2008, there were 26 persons 26 knees (6 males and 20 females) of osteonecrosis of the medial femoral condyle, and they were treated by osteochondral graft with or without HTO. The mean age of the subjects at the time of the operation was 64.0 years old (range from 44 to 74) and the mean follow-up period was 40.2 months (range from 13 to 101months). The total numbers of osteochondral graft plugs were 78 (average 3 plugs per a case). MRI was taken at 6 and 12 months after the operation, the sagittal and coronal T1 images were viewed and the images of osteochondral plugs were evaluated in two points, whether the intensity of the subchondral bone including the plug turned to low, and that whether the plugs were distinguished from the surrounding subchondral bone. The statistical analysis was performed, and the significance was determined if P value was less than 0.05. Results: HTO was performed in 15 knees (Group A) whose mean age was 65.9, and there were 11 knees of osteochondral graft without HTO (Group B), whose mean age was 63.0. There were no cases of non-union at the osteotomy sites nor were no cases of free body or non-union of osteochondral plugs in the plain roentgenograms. Evaluating the intensity of the medial femoral condyle, 15 knees showed lowered change in the subchondral area, including 6 from Group A(40%)and 10 from Group B(91%). On the rest 11 knees, the intensity of was almost the same against the surrounding bone, which included 9 from Group A (60%) and 1 from Group B(9%). Incidence of low intensity change was significantly higher in Group B than in Group A (P=0.01). Next, we evaluated whether the plugs were identified in the suchondral area. The plugs were not distinguished from surrounding bone on 6 in 15 (60%) of Group A, and on 2 in 11 (18%) of Group B. This difference was not significant (P=0.44) between Group A and B. Finally, the relation between low intensity change and plug identification was reviewed. There were only 2 cases in which plugs were not distinguished among 16 showing lower change. On the other hand, there were 6 cases in which plugs were not identified among 10 showing no intensity changes. This difference was significant (P=0.03). MRI often demonstrates fracture line even if the fracture united on the plain roentgenograms, because MRI can show subtle changes in the bone marrow. Shortly after the transfer, osteochondral graft plugs are easily distinguished on MRI, but they are becoming difficult to detect while the union proceeds. MRI also demonstrates the low intensity change in the subchondral bone, which is supposed to be a bone marrow edema. This hypothesis is supported by the result that HTO modulate the incidence of lower change in subchondral bone. It is natural to suppose that the bone marrow edema is reduced by changing mechanical axis. Our data also indicates that the change in the mechanical axis may accelerate the bone union of the osteochondral plugs. Conclusions: High tibial osteotomy modulated the MRI intensity of subchondral bone of the medial femoral condyle in the early postoperative period, which consolidates the bone union of the plugs on MRI.

P18-478 Limb alignment after open wedge high tibial osteotomy: effects on sagittal plane and clinical outcome El-Azab H.1, Morgenstern M.1, Ahrens P.1, Schuster T.2, Lorenz S.1, Imhoff A.1 1 Technical University Munich, Department of Orthopaedic Sports Medicine, Munich, Germany, 2Technical University Munich, Department of Medical Statistics, Munich, Germany Objectives: High tibial osteotomy in open wedge technique (OW-HTO) is a reasonable treatment option for isolated medial varus osteoarthritis of the knee. The aim of our study is to evaluate the accuracy of alignment after high tibial osteotomy and its relation to the clinical outcome. Methods: A prospective case series of 50 patients underwent OW-HTO and fixation with Tomofix plate fixatorTM. The correction angle was determined radiologically preoperatively and at 6 months postoperatively. The patients were examined clinically preoperatively, at 3, 6 and 36 months postoperatively.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: 43 patients showed an acceptable correction between 50%-70% of the tibial plateau width (group I). Undercorrection (\50%, group II) and overcorrection ([70%, group III) was found in 4 and 3 patients, respectively. The mean postoperative Lysholm scores at 36 months in group I, II and III were 88±13, 65±15 and 86±6, respectively (test of group heterogeneity (ANCOVA): p=0.017), whereas solely the difference between group I and II met a statistical significance (p=0.004). Conclusions: OW-HTO results in improvement of symptoms and functions in all correction groups in early and midterm results. Undercorrection is associated with significant lower clinical outcome than accurate correction and overcorrection. Ligamentous laxity or soft tissue slackness of the knee can influence the overall correction after HTO and has to be considered in the preoperative planning.

P18-482 Results of the high tibial osteotomy for medial compartment arthrosis of the knee Zazirnyi I.1 1 Hospital ‘Feofania’, Centre of Orthopaedia, Trauma and Sport Medicine, Kiev, Ukraine Objectives: The aim of our study was to evaluate the results of treatment of patients with high tibial osteotomy for medial compartment arthrosis of the knee. Methods: We have treated 48 patients with high tibial osteotomy for medial compartment arthrosis of the knee. Clinical and radiographic data were evaluated with Lisholm knee score and IKDC score. Results: Excellent and good results were achieved in 67,5% patients. 86% patients reported clinical improvement at 24 month compared to preoperative status. 67,5% of patients returned to their predisease sports activity level at 24 month after the operation. In one case we had intraoperation fracture of the lateral part of tibia plato. In one case we had early infection. Conclusions: High tibial osteotomy with an open-wedge technique using the plates like Puddu seems to be a safe and efficient procedure. Our data show that majority of patients reached the good results of operations from 12 till 24 month. This fact might be related to a removal of implant.

P18-521 Enhancing bone regeneration with an biomimic bone graft containing bone marrow stem cells and platelet rich plasma Chen C.-H.1, Whu S.-W.1, Lin B.-N.1, Chang C.-H.1 1 Chang Gung Memorial Hospital-Keelung, Orthopaedic, Keelung, Taiwan, Republic of China Objectives: Bone tissue engineering has been heralded as the alternative strategy to regenerate bone. Up to now, several biocompatible materials such as bioceramics, polymers and composites have been proposed. PRP contains many bone formation growth factors and can improve bone reconstruction. But PRP gel will degrade so fast that its structure collapse in short time. By mixing bone-like beads, beads can serve as scaffold of PRP gel. Then platelet rich plasma (PRP), thrombin and bone-like beads containing rabbit bone mesenchymal stem cells (rbBMSCs) were mixed to form a bone graft, and evaluate the performance of bone regeneration of bone graft. Methods: According to the principle of complex coacervation, type I collagen beads can be formed by extruding collagen solution into chondroitin sulfate A (CSA) solution, and then the biomimic growth of nanohydroxyapatite on the surface of fibrous collagen soaked in simulated body fluid solution (SBF) to fabricate bone-like beads. The effect of biomimic bone graft will be put into biochemistry, gene expression, and histological study. Results: Actin expression of rbBMSCs increased with time was found. rbBMSCs also secreted collagen type I and alkaline phosphate (ALP) in biomimic bone graft. rbBMSCs in biomimic bone grafts still maintained collagen I and osteoprotein (OPN) expression. According to above results, the biomimic bone graft can be utilized as a new material for bone regeneration. Conclusions: Biomimic bone graft containing bone marrow stem cells and platelet rich plasma could accelerate bone regeneration.

S277 P18-612 Histological evaluation of a synthetic bone graft in a high tibial osteotomy at 2 years post implantation Roberts S.N.J.1, Mangham D.C.1 1 Robert Jones & Agnes Hunt Orthopaedic Hospital, Oswestry, United Kingdom Objectives: The case study presented is a high tibial opening wedge osteotomy using a standard titanium plate and screw fixation used in conjunction with a synthetic bone graft material in place of traditional autologous bone graft. Methods: The 30 year old patient presented with a varus deformity of the right tibia with a Lysholm Score of 38. Following the osteotomy, a resorbable synthetic bone graft (Genex, Biocomposites Ltd) was prepared in accordance with the manufacturers guidelines, and 10cc was injected into the space of the opening wedge. The fixation was achieved using a titanium plate (Puddu plate, Arthrex) and screws. The procedure resulted in a 12.5mm correction. At 25 months post-operation, the patient’s progress was satisfactory with a Lysholm Score of 85, however all of the titanium screws had broken. During routine removal of the metalwork two biopsy samples were taken with patient consent. The samples were approximately 20x8mm, and were harvested from two sites immediately behind the plate; at the site of synthetic graft placement (larger fragment); an adjacent site representing a transition from host bone to the site of graft placement. Both samples were processed for decalcified histology (H&E stain). Results: Histological analysis of the largest fragment showed mostly bone present with an architecture of mainly compact bone with no clear cortical/medullary distinction. This is likely to be due to: a) re-distribution of osteoinductive forces around the screws and plate and b) possible osteoinduction in response to the graft material. Remodelling activity was present in the bone. Little graft material remained in the bone tissue indicating complete resorption; however there were a number of regions of woven bone with small slit-like/ovoid features, encased by lamellar bone which may indicate areas of trace residual graft material. In the region of fibrous tissue that was adjacent to the metal plate, trace particles of the graft were evident, in addition to metal fragments. The smaller biopsy sample had a composition of predominantly bone, with some fibrous tissue and a small area of fibrocartilage, which although adjacent to bone, had not developed into bone tissue, possibly as a result of poor local blood supply and a resulting low oxygen tension. The fibrous tissue was appositional to the titanium screws, and plate. Evidence of early and recent haemorrhage was also present in the sample. Again, no residual graft material was evident in the bone tissue. There was evidence of trace amounts of residual graft material in the fibrocartilage. Metal fragments were again evident. The lesser degree of bone formation in this smaller biopsy sample may be attributed to the proximity to a broken screw giving rise to local movement in the surrounding tissue. Conclusions: Histological analysis of both biopsy samples confirms almost complete resorption of graft material and replacement with viable bone. This synthetic graft warrants further investigation as an alternative to autologous bone in this indication.

P18-794 Femoral extension osteotomy to treat sequela of femoral epiphysiodesis. A case report Vila` G.1, Matamalas A.1, Puig L.1, Leal J.1, Montserrat F.1, Hinarejos P.1 1 IMAS, Barcelona, Spain Objectives: Some angular deformities, leg length discrepancy or a combination of both can occur as a consequence of physeal lesions in lower limbs. It is important an early diagnosis to avoid the development of severe deformities. Methods: We present a case of thirteen year old girl who had suffered an undiagnosed distal femoral epiphysiodesis secondary to a car accident in China when she was six year old.

123

S278

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

She complained of right hip pain, mild left knee pain and limp deambulation. Physical exam showed up a 408 flexion contracture of the left knee and reducible equinus contracture of the ipsilateral ankle and a limb length discrepancy of 10 cm, which was not diagnosed or treated before. At x-ray examination we could see right hip subluxation secondary to pelvic disbalance and a flexion deformity of the distal femur as a result of a distal femoral epiphysiodesis. Length leg x-ray showed a left varus deformity of 158. A posterior addition distal femoral osteotomy was performed supported by a lateral plate.No tendon lengthening was necessary to correct the flexion contracture. Results: Six months postoperative the patient is able to walk with no pain in the treated knee neither the contralateral hip. Control x-ray shows a consolidation of the osteotomy. Correction of varus deformity was achieved and limb length discrepancy has been corrected to 4 cm treated with shoe supplement. Conclusions: It is important to control children after a physeal fracture until the end of growth in order to detect angular deformities or leg length discrepancies and in case it occur we must treat them as soon as possible to prevent severe deformities. In cases of combined angular and length deformities the correction of the first one can improve the second one.

P18-818 Tibial slope - which measurements are reliable? A comparative study in 3D-CT reconstructions Lorenz S.1, Lu¨tzner J.2, Fu F.1, Tashman S.3 1 University of Pittsburgh, Dept. of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States, 2Technische Universita¨t Dresden, Department of Orthopedic Surgery, Dresden, Germany, 3University of Pittsburgh Medical Center, Orthopaedic Biodynamics Laboratory, Pittsburgh, United States Objectives: Reconstruction of the tibial slope in knee arthroplasty and high tibial osteotomy is known as important factor for favourable clinical outcomes in terms of ROM, pressure distribution and stability in the knee joint. Several methods to determine the slope on plain radiographs have been described, though the reliability of the measurement is low. We hypothesize that the tibial slope can be more accurately determined on 3D CT scans than on any previously described 2D radiographic method. Methods: 5 CT scans of healthy volunteers were identified from the control group of an IRB-approved ongoing study. The CT scans consist of images in axial orientation of the knee at least 10cm proximal to 20cm distal from the joint line. One additional slice was obtained through the ankle at the level of the maleoli. The mechanical tibial axis was defined from the midpoint between the two intercondylar eminences to the ankle center. The tangent to the lateral and medial plateau was defined as a line drawn on a sagittal image that is exactly between the central point of the longitudinal axis and the lateral and medial border of the plateau, respectively. Digitally reconstructed radiographic images were obtained by projection through the CT data set, using AMIRA. True sagittal plane views were generated and verified by superposition of the femoral condyles. The following lines were evaluated (Fig.1): 1) the anterior tibial cortex line (ATC), 2) the tibial proximal anatomical axis (TPAA), 3) the posterior tibial cortex line (PTC) and 4) the fibular proximal anatomical axis (FPAA). The tangents for the medial and lateral tibial plateau were drawn on projection views of the medial and lateral half of the tibial plateau to rule out bias from the opposite site. In addition the slope was determined in a complete projection. Each measurement for the tangents and the axis was repeated 3 times by two orthopaedic surgeons on different days. The angles were measured by using ImageJ.

123

Fig. 1 Results: To evaluate the source of error, the measurements at the tibial shaft and plateau were analyzed separately. Tibial shaft: the variability for measurements done on the 2D projection view were in the average range between 0.8 (PTC) and 2.0 (FPAA) degrees. In contrast, the range for the mechanical axis based on 3D CT analysis was significantly less, with an average range of 0.3. Tibial plateau: the variability for 2D measurements was in the average range of 2.2, 2.7 and 3.5 for combined slope, medial and lateral slope, respectively. The average range for the 3D measurements at the tibial plateau was 0.5 and 1.5 for the medial and lateral slope, respectively. In summary, 3D measurements resulted in a combined average range of 0.8 and 1.8 using the mechanical axis and the CT measurements from the medial and lateral tibial plateau, respectively. The most favorable 2D combination using PTC and combined slope had a combined average range of 3.0. In contrast, the least favorable 2D combination (FPAA/lateral tibial plateau on radiograph) showed a huge combined average range of 5.5. Conclusions: Estimates of tibial slope from 3D CT were more reliable than those based on conventional 2D radiographs.

P18-844 High tibial osteotomy- clinical follow up after 24 months Rosslenbroich S.1, Zantop T.2, Herbort M.3, Raschke M.1, Petersen W.4 1 Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 2Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany, 3University of Muenster, Department of Traumatology, Muenster, Germany, 4Martin-Luther-Krankenhaus, Unfallchirurgie, Berlin, Germany Objectives: To evaluate the clinical results following high tibial valgus osteotomy in medial opening approach and internal plate fixation using an

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 angular stable plate system. He hypothesized that high tibial valgus osteotomy in medial opening approach provides good clinical outcome after 24 months. Methods: Patients with varus gonarthrosis were included in a prospective study design. Clinical diagnostic was performed using clinical examination, x-rays with Rosenbergs view and lomg standing x-rays. Prior to corrective osteotomy, arthroscopy was performed and microfracturing of the medial compartment was performed if indicated. Osteotomy was performed in medial opening technique and fixation using angular stable plate fixation (Tomofix, Synthes). Medial opening osteotomy of a height more than 12 mm were performed using autologous iliac crest bone graft. After 24 month follow up, clinical examinations and long standing x-rays were performed. Results: A total of 61 patients were included in the study. In 45 patients, microfracturing of the medial femoral condyle was performed. In 5 patients, ligament reconstructions (4 ACL reconstructions, 1 PCL reconstructions) were performed at the same setting. Mean medial opening value was 9.3 mm (range 5-21 mm). Postoperative long standing x-rays showed weigthbearing line at 63%of the tibial plateau. A total of 21 patients showed early postoperatively pain that was contributed to the plate system. In 18 patients an early plate removal was performed. In none of these patients a reposition lost was to be found. Conclusions: The results support our initial hypothesis. High tibial valgus osteotomy in medial opening approach provides good clinical outcome after 24 months. There seems to be plate associated pain syndrome in patients with less subcutaneous fat tissue. Patients should know of this fact and told that an implant removal may be necessary.

P18-847 Causes for failure and revision of high tibial osteotomy Zantop T.1, Rosslenbroich S.2, Lenschow S.2, Herbort M.3, Petersen W.4 1 Westfalian University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Mu¨nster, Germany, 2Wilhelms University Muenster, Department of Trauma-, Hand- and Reconstructive Surgery, Muenster, Germany, 3University of Muenster, Department of Traumatology, Muenster, Germany, 4Martin-Luther-Krankenhaus, Unfallchirurgie, Berlin, Germany Objectives: High tibial osteotomy is in the spot of interest. Due to new implants and angular stability of the locking plates, the medial opening approach is currently a wide used technique. However, there seems to be a high risk for complications when HTO are performed in centers with little clinical experience about this technique. Aim of the current study was to prospectively evaluate the reasons for failure after HTO and the surgical approach for revision. Methods: In a prospective study design, 20 consecutive patients following high tibial osteotomy were included. All primary osteotomies were performed at outside hospitals. Diagnostic approach was based on operative reports of the first surgery, clinical examination, long standing x-rays and CT scans to evaluate the osteotomy. Revisions were performed using a medial opening approach fixed with locking plate (Tomofix, Synthes). Results: Causes for failure were multifactorial. Among the common failures were fractures of the tibial plateau, compartment syndrome after injury to the popliteal artery, injury to the peroneal nerve (3 times lateral closing, 2 times medial opening approach). Most commonly were overand undercorrection of the varus deformity. In overcorrection there was a trend to overlooked lateral insufficiency. In one patient the varusdeformity after the first osteotomy was more pronounced than before surgery. All patients were treated with a medial opening osteotomy and medial opening (medial closing in 2 cases) approach. No complications did occur intraoperatively. X-ray analysis showed weigthbearing line at 60% of the tibial plateau. Conclusions: High tibial osteotomy in medial opening technique is a safe and secure procedure. However, the current analysis of revisions performed at our centre show that specific technical aspects should be respected. These include surgery without tourniquet in 20 knee flexion, accurate preoperative planning using long standing x-rays, surgery

S279 under image intensifier as well as careful clinical examination with regard to ligamentous injury. High tibial osteotomy in medial opening technique should be performed by surgeons with clinical experience in joint surgery.

P18-852 3D biomechanical analysis of walking before and after opening-wedge high tibial osteotomy Lind M.1, Webster K.2, Wittwer J.3, McClelland J.4, Feller J.5 1 Aarhus University Hospital, Sports Trauma Clinic, Aarhus, Denmark, 2 La Trobe University Musculoskeletal Research Center, Faculty of Health Sciences, Bundoora, Australia, 3Musculoskeletal Research Centre, Melbourne, Australia, 4Musculoskeletal Research Centre, Bundoora, Australia, 5La Trobe University Musculoskeletal Research Center, Melbourne, Australia Objectives: Medial opening wedge high tibial osteotomy (MOW-HTO) is widely used in the treatment of medial compartment osteoarthritis of the knee. The purpose of the procedure is to alter the coronal alignment of the knee to a slightly valgus alignment, thereby reducing symptoms from the medial compartment. The aim of this study was to investigate the functional biomechanical consequences of this alteration in alignment. Methods: Eleven male patients with medial compartment osteoarthritis underwent 3D gait analysis during level walking before and 12 months after MOW-HTO. Nine male control subjects of a similar age were also tested with the same protocol. Sagittal and coronal angles and moments in both knees were compared. Pre- and postoperative radiographic coronal plane alignment was also measured. Results: Walking speed increased significantly postoperatively (p=0.0001) and was not different from control subjects. Preoperatively there was reduced maximum flexion in stance compared to control subjects. Postoperatively this increased significantly (p=0.005) and was the same as in control subjects. The same improvement was also seen for the maximum external knee flexion moment. Preoperatively the mean maximum varus angle during stance was 13.5˚. This reduced to 5.4˚ postoperatively (p = 0.0001) and was not different from control subjects (6.8˚). The mean maximum adduction moment also reduced from 3.5 to 2.7 (% Bw/ht, p=0.02), compared to 3.6 in control subjects. Interestingly, the adductor moments in the non-operated knee increased postoperatively. The mean radiological mechanical alignment was changed from 8.7˚ varus preoperatively to 0.1 valgus postoperatively. Conclusions: Gait analysis is a useful tool for functional assessment following MOW-HTO. The procedure resulted in normalisation of various aspects of dynamic knee function and specifically reduced the adduction moment at the knee. The finding of an increased adduction moment in the non-operated knee could indicate an increased risk for the development of OA in this joint.

P18-1237 A short term clinical and radiographic review of medial high tibial opening wedge osteotomy Hook S.1, Melton J.1, Thomas N.1, Wilson A.1, Wandless F.1 1 Basingstoke and North Hampshire NHS Foundation Trust, Trauma and Orthopaedics, Basingstoke, United Kingdom Objectives: Improved surgical techniques and new fixation methods have revived interest in high tibial osteotomy surgery in recent years. Our aim was to review our first 59 cases including radiological changes and clinical outcome and to assess any correlation between them. Methods: All patients underwent radiological and clinical review including pre and post operative scores. Mean age at surgery was 43 (22-59) and mean follow up is 22 months. Results: The mean pre-operative limb alignment was 5.4 varus (range 116) with correction to 2 valgus (range -1 - 7). This equated to a mean change of 7.2 (2-15) achieved using an opening wedge at surgery of 10.2mm (4.5mm - 20mm).

123

S280 HTO is known to increase tibial slope, shifting the tibial resting position anteriorly and thus improving stability in the PCL deficient knee but potentially worsening instability in the ACL deficient knee. In this series the change in tibial slope from a -5.2 (95%CI: -6.36 to -4.07)) to -7.8 (-8.83 to -6.89) was statistically significant. p= 0.0014 (Mann Whitney). Patellar height is often reduced following opening wedge HTO and this is confirmed in our series. The Blackburne-Peel ratio changed from 0.74 to 0.58 and the Caton-Descamps from 0.83 to 0.7. Both were statistically significant at p\0.0001 and p=0.0001 respectively. HTO can also alter the angle of the joint line and we show a change of medial proximal tibial angle (MPTA) from 86.1 (81-92) to 91.3 (87-96). There were three superficial wound infections, and one non union which has been treated with grafting and re fixation. Six patients have had their plate removed, one is awaiting plate removal and a further 3 report discomfort from their plate. Range of movement of the knee was well preserved. The difference between the operated knee (mean 127) and the non operated knee (mean 138) was not statistically significant (Wilcoxon signed rank test). Heel height can be used as a measure of loss of hyperextension of the knee. In the operated knee this was found to be a mean of 34.8mm less than in the non operated knee. There was no correlation between this and change in tibial slope. All scores improved post operatively, the knee injury and osteoarthritis outcome (KOOS) from 48 (8-91) to 73 (27-96), the Oxford knee score (OKS) from 25 (3-47) to 37 (9-48), and the EQ5D from 189809 (1122132333) to 14138 (11111-22233) with the EQ5D VAS improving from 58 to 75. There was no correlation between change in limb alignment, tibial slope or patellar height and any of the scores used. Conclusions: Improvement in clinical scores in these patients confirms that medial opening wedge HTO is a reliable joint preserving procedure in the short term and our surgical technique is reproducible and consistent with other published series. Lack of correlation between radiological findings and scores suggests that alteration of tibial slope and patella height is not detrimental to clinical outcome.

P18-1332 High tibial and distal femoral opening-wedge osteotomies Panarella L.1, Puddu G.2 1 University of Rome Tor Vergata, Department of Orthopaedic Surgery, Rome, Italy, 2Valle Giulia Private Hospital, Department of Orthopaedic Surgery, Rome, Italy Objectives: This retrospective study describes the technical procedures and the post-operative results in a large group of patients operated between 1997 and 2008 using the Puddu Plate (ArthrexTM) to evaluate the effects of the knee axis correction using an Opening-Wedge High Tibial Osteotomy in varus knees or a Distal Femoral Osteotomy in valgus knees. Methods: We analyzed a cohort of 66 patients undergone to axial correction, 6 patients undergone to bilateral surgery (59 HTO and 13 DFO). All patients suffered of unicompartmental arthritis due to knee malalignment. The mean follow-up was 64 months (up to 200 months). The axial correction was checked by a bilateral standing long-leg radiograph, the patello-femoral arthritis was assessed by standard lateral view radiograph and the tibio-femoral arthritis by Rosenberg postero-anterior view. The HSS score and IKDC score was employed to evaluated long-term outcome of this technique. Results: In all patients we found an improved quality of life; pain was reduced and function of knee was satisfactory. Axial correction was maintained during follow-up controls and the articular space was globally preserved with no or little progression in cartilage degeneration. Conclusions: Correction of knee axis is a well-accepted procedure for management on young and active patients with unicompartmental arthritis due to a mechanical overload of the affected compartment. An accurate selection of patients and a careful pre-operative planning are essential for a good outcome. Post-operative management is also important for the final result. In our experience, we found both short-term and long-term patients’ satisfaction.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

Knee: total joint replacement P19-17 Preliminary results after bicompartimental knee prosthesis Maestro A.1, Salvador A.2, Rodriguez L.3, Fdez-Lombardia J.1, Olay M.J.4, Casero J.4 1 FREMAP, Orthopaedic Surgery, Gijon, Spain, 2Clinica Platon, Orthopaedic Surgery, Barcelona, Spain, 3Hospital Cabuen˜es, Orthopaedic Surgery, Gijon, Spain, 4FREMAP, Orthopaedic Surgery, Oviedo, Spain Objectives: To evaluate the early results of the Deuce Bicompartimental Knee prosthesis(Smith&Nephew). Methods: Have been included in the study a total of 12 knee operated on by the same team in the same number of patients (7 women and 5 men) with a mean age of 60.8 (6.1) years, 77.2 (5.1) Kgrs weight and 170(6.7) cm. 8 cases were primary surgery and 1 case was a revision surgery of femoro-patellar prosthesis implanted before. In 6 cases was the right knee and left in 3. Mean follow-up was 15.5(4.7) months. All cases were evaluated by clinical and radiological criteria of the American Knee Society. Results: The score of the AKS was 54.8(8.2) preoperatively and 87.5 (7.2) post. The alignment was satisfactory (between 38 of valgus and 28 of varus) in 8 cases and 1 case was the 38 of varus. One patient had a flexion deficit of more than 115 degrees, with full recovery of the extension. In three cases radiolucence lines were appreciated in the area 1 (A-P and Lat) in the tibial component. Highlights a case of complications as patelar subluxation and dysesthesia in the lateral proximal end of the leg. Conclusions: Our results are still short, but hopeful, as a functional surgery with fast recovery, requiring a learning curve and being a demanding technique. Evidence level: III. Comparative retrospective study.

P19-19 Long term results of LCS mobile bearing knee system in Korean Park H.1 1 Dankook University College of Medicine, Orthopaedic Surgery, Cheonan, Korea, Republic of Objectives: To evaluate the clinical and radiological outcomes of knee joint arthroplasty with LCS mobile bearing system (low contact stress, Depuy, USA) in 32 patients (33 cases) in whom a follow-up study was possible over ten years period and had korean activities. Methods: To evaluate the clinical outcome, we examined the flexion of contracture, and mean range of motion and the knee score based on the Knee Society Clinical Rating System (KSCRS) and the score system of HSS (Hospital for Special Surgery) recommended by Insall. Radiologic assessment: Based on the American Knee Society Roentgenographic Evaluation and Scoring system, component position, leg and knee alignment and the prosthesis-bone interface or fixation were evaluated on anteroposterior and lateral views of the knee. To define the alignment of lower limbs, the alignment and location of component were assessed by measuring femoral angle (a) and tibial angle (b) on anteroposterior view and femoral angle (c) and tibial angle (d). Results: Clinical outcome: The mean flexion contracture was 10.0 (range, 0 - 40) preoperatively and 3.7 (range, 0 - 30) postoperatively; this showed that flexion contracture significantly improved postoperatively (p\0.05). The mean range of motion was 104.0 (range, 20 - 135) preoperatively and 115.3 (rangem, 60 - 135) postoperatively; this also showed that the range of motion significantly improved postoperatively (p\0.05). At a final follow-up, the mean KSCRS score improved from 77.2 points (range, 57 - 98 points) preoperatively to 152.5 points (range, 138 - 193 points) postoperatively. The knee score and functional score showed more than good in 87.9% (29 cases) and 93.9% (31cases), respectively. In all cases, the HSS score improved from 49.5 points (range, 37 - 63 points) preoperatively to 80.1 points (range, 50 - 98 points) postoperatively. Based on the HSS score, the clinical outcome was assessed as ‘excellent’ in 60.6% (20 cases), ‘good’ in 27.3% (9 cases), ‘fair’ in 9.1% (3 cases) and ‘poor’ in 3.0% (1 case).

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Radiological outcome: The mean tibio-femoral angle was corrected 3.38 of varus (range, 188 of varus - 128 of valgus) preoperatively to 3.58 of valgus (range, 2 8 of varus - 12 8 of valgus) postoperatively. The location of component was assessed based on the guidelines of the American Knee society; this showed that mean a angle was 96.2, mean b angle was 87.8, mean c angle was 0.3 and mean d angle 86.1. The radiolucent zone was noted in 23.2% (8 cases). On lateral view of the femur, the radiolucent zone was detected in zone 1 (1 case, 3%) and zone 5 (1 case, 3%). On lateral view of the tibia, it was detected in zone 1 (1 case, 3%) and zone 3 (1 case, 3%). On anteroposterior view of the tibia, it was detected in zone 1 (2 cases, 6.1%), zone 3 (1 case, 3%) and zone 4 (1 case, 3%). The prosthesis-bone fixation was assessed as less than 4 points in 28 cases and 5*9 points in five cases. The loosening corresponding to 10 points was not noted. Conclusions: LCS TKA showed excellent results and good survival rate in Korean living activity, but more cases should be necessary for statistic significance.

P19-25 Comparison of the results between MCL complete detachment and medial epicondylar osteotomy for ligament balancing of varus deformity in TKA Lee B.1 1 Gachon University, Orthopedic Surgery, Incheon, Korea, Republic of Objectives: We compare the results between MCL complete detachment and medial epicondylar osteotomy for ligament balancing of varus deformity in TKA. Methods: We reviewed 8 cases of MCL complete detachment (group I) and 11 cases of medial epicondylar osteotomy (group II) which need extensive medial release during TKA in varus deformity (from February 2001 to December 2006). Average ages were 71.1-year-old in group I and 71.5-year-old in group II. Average follow-up periods were 41.1 months and 21.9 months respectively. Clinical outcome measures included Knee Society score (KSS), Function score (FS), and range of motion (ROM). Radiological measure were degree of medial instability by valgus stress radiograph, and alignment by whole extremity standing radiograph. Results: Group I had 4 neutral and 4 varus alignment and group II had 9 neutral, 1 varus and 1 valgus alignment. There were no significant differences in clinical results between both two groups, for KSS (95.1 vs 91.1), FS (82.5 vs 88.2), and ROM (0.6-115˚ vs 0-118.8˚) (p[0.05). However, medial instability compared with normal side were significant different between two Group. Group I had the medial instability of 4.1 degree at postoperative 3 months and 2.1 degree at final follow-up. Group II had the medial instability of 0.9 degree at postoperative 3 months and 0.4 degree at final follow-up (p\0.05). Conclusions: Medial epicondylar osteotomy in TKA could be better ligament balancing technique than distal release regarding the medial stability of the knee.

P19-43 The tibiofemoral contact point in knees with TKR and in normal knees: normal values and reproducibility de Jong R.J.1, Heesterbeek P.2, Wymenga A.1 1 Sint Maartenskliniek, Departement for Orthopaedic Surgery, Nijmegen, Netherlands, 2Sint Maartenskliniek, Research, Development & Education, Nijmegen, Netherlands Objectives: Flexion gap instability after cruciate retaining TKR allows paradoxical anterior movement of the femur during flexion. The tibiofemoral contact point (CP) moves anteriorly and produces a decrease in the lever arm of the extensor apparatus. This can provoke patellofemoral pain, tibiofemoral joint pain and instability. Standardized measurement of the CP could be a way to objectively document these phenomena. So far, no appropriate measurement methods or normal values are known. Therefore we developed a CP measurement technique for lateral radiographs that can be applied to natural knees and knees with a TKR, and determined the normal value for the CP in healthy knees.

S281 Methods: Conventional lateral radiographs were chosen to measure the CP in 90 flexion. To compare pre- and post TKR radiographs the posterior tibial cortex was chosen as fixed reference point and an artificial tibial cut of 7 mm for natural knees was introduced. The intra- and interobserver reproducibility of this measurement method was assessed by calculating the intraclass correlation coefficient (ICC) and by using the Bland and Altman method. The normal range of the CP in natural knees from 30 radiographs of healthy knees was determined. Results: The mean CP lies at 68% (SD 3.3), 95% CI [61.7-74.9%] measured from the anterior tibial cortex. The ICC ranged from 0.78-0.99 which was consistent with the small differences observed between measurements. The intra- and interobserver variability were good; the intraobserver variability was lower than the interobserver variability for both TKR and natural knee radiographs. CP measurement on radiographs of natural knees was slightly more reproducible compared to measurements on TKR radiographs. Conclusions: The contact point for a normal knee was determined at approximately 2/3 from the AP tibia distance at 90 degrees of flexion, measured 7 mm below the medial joint line. The contact point measurement technique as presented is reproducible on lateral radiographs of knees with or without a TKR. This CP measurement method can be used clinically to evaluate the CP after TKR as well as in patients with suspected (posterior cruciate) ligament lesions. The normal values can be used in TKR to aim for restoration of the natural contact point.

P19-53 Revision knee arthroplasty at a district general hospital: a case series analyzing incidence and causal factors Arshad M.-S.1, Bhutta A.2, Bari A.1, Shah N.3, Obeid E.-M.1 1 Tameside General Hospital, Trauma & Orthopaedics, Ashton-U-Lyne, United Kingdom, 2Royal Bolton Hospital, Taruma & Orthopaedics, Bolton, United Kingdom, 3Lancaster Royal Infirmary, Trauma & Orthopaedics, Lancaster, United Kingdom Objectives: Any revision surgery brings along with it despair and sleeplessness for the surgeon and widened expectation from the patient. Revision knee arthroplasty is no further from this truth. The 3-year revision rates for cemented knee arthroplasty were 1.4% in England between 2004-06 with loosening and infection being the commonest causes. Our rate over this time period was 4.8% thus, prompting further investigation. Methods: A retrospective study analysing revision knee arthroplasty at Tameside General Hospital for the past 15 years was undertaken. We aimed to review any causal factors and study the parameters used prior to embarking upon revision surgery, so that we may improve our understanding of the subject especially when dealing with patients in need of such procedures. Results: Of the 42 cases reviewed all were cemented and 76% were Kinemax prosthesis, others included Scorpio, Optetrak and Oxford Uni. Factors such as BMI, ASA grades and tourniquets times were analysed along with the cause for revision. For those with a diagnosis of infection we reviewed how the diagnosis was made and compared this to current best practice. The rate of revisions for the 15-year period was similarly found to be 4.5%. Our results indicated that loosening was the common cause for revision, especially the tibial component (55%). More than 1/3 of our patients had an ASA grade of 3 and a tourniquet time of greater than 90 minutes, with at least 17% having a combination of both. About half of our group had a BMI of greater than 30. For Infected cases at least two aspirations were needed with tissue biopsy being the gold standard method. All infected cases required 2 stage revisions. Conclusions: Our study highlighted the importance of patient selection and especially pre-operative assessment and optimisation. There was a general lack of evidence in current literature regarding the relationship between length of tourniquet application and revision surgery (during primary operation). Further studies concerning this are needed. For infected cases more importance should be given to aspirate WBC & neutrophil counts than aspirate cultures alone which, currently is widespread practice.

123

S282 P19-62 Total knee replacement in severely obese patients: outcomes and quality of life. A case-control study Nu´n˜ez M.1, Lozano L.2, Nu´n˜ez E.3, Segur J.M.4, Sastre S.5, del Val J.L.3, Macule´ F.4, Popescu D.2, Morales J.5, Suso S.4 1 Hospital Clı´nic de Barcelona, Rheumatology Department and Knee Unit, Barcelona, Spain, 2Hospital Clinic de Barcelona, Knee Unit, Barcelona, Spain, 3Institut Catala` de la Salut, SAP Suport al Diagno`stic i al Tractament, Barcelona, Spain, 4Hospital Clı´nic i Universitari de Barcelona, Knee Unit. Orthopaedic Surgery Department (ICEMEQ), Barcelona, Spain, 5Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain Objectives: Few short- or long-term studies have focused on health outcomes in severe and morbidly obese patients, and the few that exist are very heterogenous methodologically. In addition, the impact of obesity on TKR outcomes is unclear. Some studies have shown worse results in obese than in nonobese patients, while others have found no differences. The objective of this study was to evaluate HRQL preoperatively and at 12 months of follow-up in severe and morbidly obese patients with knee OA and a control group of nonobese patients undergoing TKR. Methods: Case-control study with 12 months follow-up. HRQL was measured using the WOMAC questionnaire. Sociodemographic variables, comorbidity, body mass index (BMI), degree of intra-operative difficulty (IOD), in-patient and postoperative medical data were collected. Statistical analysis: The effect size (ES) was measured for the different outcome variables. Comparison of the two groups after 12 months of follow-up was made using the t test. Results: Study group: Sixty patients (88% women) with a mean age of 70.2 years (SD 6.7) and a mean total WOMAC dimension score of 61.4 (SD 16.7). Control group: Sixty matched controls (88% women) with a mean age of 71.7 years (SD 6.7), and a mean WOMAC score of 58.2 (SD 13.4). There were significant improvements in all WOMAC dimensions compared to baseline (p\0.001) in both groups. There were no differences in WOMAC dimension scores between the two groups at 12 months. The study group had more IOD (p=0.014) and more-severe complications in the follow up. Conclusions: Severe and morbidly and nonobese patients had similar change scores and TKR outcomes in terms of HRQL at 12 months after TKR. Although obese patients had more intraoperative difficulties and more-severe postoperative complications, these aspects are susceptible to control.

P19-71 Health-related quality of life after debridement for acute total knee prosthetic joint infection Nu´n˜ez M.1, Vilchez J.2, Soriano A.3, Nun˜ez E.4, Castillo F.5, Lozano L.6, Torner P.7, Segur J.M.7, Popescu D.6, Sastre S.5, Macule´ F.7, MartinezPastor J.C.5 1 Hospital Clinic de Barcelona, Rheumatology Department and Knee Unit, Barcelona, Spain, 2Hospital Clinic, University of Barcelona, Knee Unit, Barcelona, Spain, 3Joint and Bone infections Unit, Hospital Clinic, University of Barcelona, Barcelona, Spain, 4Institut Catala` de la Salut, SAP Suport al Diagno`stic i al Tractament, Barcelona, Spain, 5Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain, 6Hospital Clinic de Barcelona, Knee Unit, Barcelona, Spain, 7Hospital Clı´nic i Universitari de Barcelona, Knee Unit. Orthopaedic Surgery Department (ICEMEQ), Barcelona, Spain Objectives: Open debridement without removing the implant has demonstrated a success rate higher than 70% in acute prosthetic joint infections. The objective of this study was to evaluate the functional status and health-related quality of life (HRQL) after 12 months of follow-up in patients undergoing open debridement for acute postsurgical knee prosthetic joint infection. Methods: Prospective 12 months study. HRQL was measured by the disease-specific WOMAC questionnaire and the general SF-36. Sociodemographic variables, comorbidity, body mass index (BMI), clinical manifestations, leukocyte count, C-reactive and protein level at admission for infection, surgical-treatment, microorganisms isolated, antimicrobial

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 therapy were measured. Statistical Analysis: The Wilcoxon matched pairs signed-ranks test was used to evaluate differences between mean scores at baseline and 12 months in the WOMAC dimensions. Differences between groups for categorical variables and WOMAC dimension scores were analyzed using the Mann-Whitney U-test. Spearman’s correlation coefficient was used to evaluate the relationship between WOMAC pain and function dimensions and SF-36 bodily pain and physical function dimensions. Results: Thirty patients (22 women, mean age 72.5 years (SD 7.88) were assessed at 12 months. The mean number of comorbidities was 2.0 (SD 1.2) and a mean BMI was 32.4 (SD 6.2). There was a significant improvement in the three WOMAC dimensions (p\0.01 for all comparisons). Staphylococcus aureus was the most frequently isolated microorganism, found in 14 patients (46.6%). In these patients, the SF-36 physical role, bodily pain, emotional role and mental health dimension scores at 12 months after infection were significantly lower than in patients affected by other pathogens (p\0.05). However, no differences were observed in the WOMAC scores according to the microorganism isolated. Conclusions: Ours results suggest that infections due to Staphylococcus aureus are more severe and need more prolonged antibiotic treatment that impacts on the health status. Patients with acute postoperative infection treated with open debridement and retention of the implant, achieved a reasonable functional status, measured by the WOMAC and SF-36.

P19-72 Health-related quality of life after acute infection in patients with total knee replacement: case-control study Nu´n˜ez M.1, Vilchez J.2, Soriano A.3, Nun˜ez E.4, Castillo F.5, Lozano L.6, Torner P.7, Segur J.M.7, Popescu D.6, Sastre S.5, Macule´ F.7, Martinez-Pastor J.C.5 1 Hospital Clinic de Barcelona, Rheumatology Department and Knee Unit, Barcelona, Spain, 2Hospital Clinic, University of Barcelona, Knee Unit, Barcelona, Spain, 3Joint and Bone infections Unit, Hospital Clinic, University of Barcelona, Barcelona, Spain, 4Institut Ca´tala de la Salut, SAP Suport al Diagno`stic i al Tractament, Barcelona, Spain, 5Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain, 6Hospital Clinic de Barcelona, Knee Unit, Barcelona, Spain, 7Hospital Clı´nic i Universitari de Barcelona, Knee Unit. Orthopaedic Surgery Department (ICEMEQ), Barcelona, Spain Objectives: To evaluate health-related quality of life (HRQL) preoperatively and at 12 months of follow-up in patients with acute total knee prosthetic joint infection treated with open debridement and antimicrobials and a control group of non-infected patients with total knee replacement (TKR) matched for age, gender, body mass index (BMI) and WOMAC dimension scores. Methods: Case-control study with 12 months follow-up. HRQL was measured using the WOMAC questionnaire. Sociodemographic data, comorbidity and clinical variables including BMI and comorbidity were collected before joint arthroplasty. Statistical Analysis. Groups were compared using the t-test for continuous variables and the B2 test for categorical variables. The Wilcoxon matched pairs signed-ranks test was used to evaluate differences between mean scores at baseline and 12 months in the WOMAC dimensions. Statistical analysis:Effect size (ES) was calculated for the outcome measure (WOMAC). Differences between groups for categorical variables and WOMAC dimension scores were analyzed using the Mann-Whitney U-test. Results: Study group: Thirty infected (22 female) patients with a mean age of 72.53 (SD 7.88) years. Twenty-nine patients (96.6%) had one or more comorbidity. Mean total WOMAC score was 59.97 (SD 12.64)). Twentyeight patients (93.33%) had overweight or obesity. Control group: Sixty matched controls. Forty-four were female, with a mean age of 70.82 (SD 7.30) years. The mean total WOMAC score was 60.07 (SD 12.19). There were significant improvements in all WOMAC dimensions compared to baseline (pB 0.003) in both groups. After 12 months of follow up, the study group had higher (worse) pain dimension scores compared with the control group, (30.17 [SD 24.30] vs 19.65 [SD 19.30], respectively) p=0.028. Conclusions: Infected patients treated with debridement plus antibiotics significantly improved their quality of life and function after 12 months of

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 follow-up. Although a significant difference was only observed in the pain dimension, the results in infected patients were worse than those obtained in non-infected patients. Our results support open debridement plus antimicrobial treatment as a good therapeutic approach in acute prosthetic joint infections in terms of function and quality of life.

P19-120 Clinical results according to patella management in revision TKA Seo S.-S.1, Ha D.-J.1, Kim C.-W.1, Park D.-H.1 1 Inje University, Orthopaedics, Busan, Korea, Republic of Objectives: To compare the clinical results among patella resurfacing or retention, nonresurfaced patella, and bony shell patella which is removed prosthetic patella in revision total knee arthroplasty(TKA). Methods: Between Mar. 1999 and Dec. 2007, 45 knees were performed revision TKA. Among them, 37 knees could be followed at mean duration of 2.4 years. Mean age was 67 years and 32 knees were female and 5 knees were male. 13 knees (Group III) were resurfaced the patella of which patella were resurfaced or retained and 24 knees were not resurfaced the patella, of which 10 knees (Group I) underwent nonresurfaced, just trimmed patella and 14 knees (Group II) had bony shell patella. The results were evaluated with KSS, Patella score. Results: Preoperative KSS and Function scores were 49.6±13.9 (mean±SD) and 44.3±15.6 in Group I, 51.4±14.8 and 42.5±13.7 in Group II, 46.8±13.1 and 42.9±14.4 in Group III, respectively. Postoperative KSS and Function scores were increased to 75.6±15.3 and 63.6±14.6 in Group I, 66.4±13.1 and 54.1±15.7 in Group II, 76.1±14.5 and 66.4±13.1 in Group III, respectively. The differences among groups were insignificant except postoperative Function score. The Patella score were 23.4±5.6, 20.1±4.7 and 24.6±5.4 in Group I, II and III, respectively. The Patella score in Group II was significantly lower than other groups. There was one patella fracture in Group II. Conclusions: This study showed that preservation of patella bone stock (nonresurfaced patella) or restoration of patellofemoral geometry (resurfacing or retaining patella prosthesis) gives us superior clinical results. Bony shell patella should be avoided in revision TKA as possible.

P19-156 Genetic polymorphisms of interleukin-1b (-511C/T) and interleukin-1 receptor antagonist (86-bp VNTR) in susceptibility to knee osteoarthritis in Croatian population — preliminary report Mihelic R.1, Jotanovic´ Z.1, Sestan B.1, Mulac Jericevic B.2, Dembic Z.3 1 Orthopaedic Clinic Lovran, Lovran, Croatia, 2Medical School University Rijeka, Dept. of Physiology, Rijeka, Croatia, 3University of Oslo, Oslo, Norway Objectives: We believe that one or more genes which are specific for a group of people would increase the risk of OA onset. Genetic research has discovered that there is a difference in morbidity between male and female patients and different joints. For this reason it is necessary to investigate the OA onset in both genders and different geographic population including different joints.Some studies have confirmed the presence of Interleukin-1 (IL-1) genetic cluster on the 2q-13 chromosome in patients with knee OA. The specific purpose of this investigation is to determine the possible connection of IL-1 genetic cluster with onset of knee OA in the Croat population of 500 male and female patients who were submitted to hip or knee prosthetic replacement. Methods: Genomic DNA will be extracted from blood samples prelevated at the surgery, using standard methods. Then the detection of polymorphism of nucleotides in these genes by polymerase chain reaction (using Taquman method - direct alleles detection) as well as the detection of polymorphisms of number of repeated sequences of these genes (indirect alleles detection) will be done. Data obtained by this investigation (genotype and alleles distribution) will be compared with the samples of 500 healthy individuals of the same age, representing the control group by means of hi-quadrate test.

S283 Results: Detailed specific results of the investigation will be presented at the congress. The analysis is in progress. Conclusions: The conclusions will be presented at the congress, analysis is in progress.

P19-175 Relationship between rotational alignment of the femoral component in total knee arthroplasty and soft tissue release during surgery or ligament balance in flexion after surgery Hatayama K.1, Terauchi M.1, Saito K.1, Yanagisawa S.2, Takagishi K.2 1 Social Insurance Gunma General Hospital, Department of Orthopaedic Surgery, Maebashi, Japan, 2Gunuma University Faculty of Medicine, Department of Orthopaedic Surgery, Maebashi, Japan Objectives: This study investigated whether rotational alignment of the femoral component in total knee arthroplasty (TKA) is related to the amount of medial soft tissue release during surgery and ligament balance in flexion after surgery. Methods: Sixty-three consecutive patients (70 knees) with primary varus osteoarthritic knee who underwent primary TKA were included. Cruciateretaining type TKA (PFCR, Depuy, Warsaw, IN) was performed using the independent cut technique in all cases. We attempted to align the femoral component parallel to the clinical epicondylar axis (CEA) and the need for medial soft tissue release was determined intraoperatively based on clinical observation of ligament balance. Medial release was performed by the following successive steps: (1) deep medial collateral ligament (MCL); (2) release of the posteromedial capsule and tibial attachment of the semimembranosus; (3) anteromedial tibial sleeve release of the superficial MCL above the insertion of the pes anserius; (4) anteromedial tibial sleeve release of the superficial MCL below the insertion of the pes anserius. Finally, ligament balance in extension and 90 flexion was assessed via Knee Balancer (Depuy, Warsaw, IN). The ligament balance in flexion was evaluated using modified Kanekasu’s epicondylar view radiographs with a distraction force of 2kg of weight at the patient’s ankle one year after surgery.

Modified Kanekasu’s epicondylar view radiograph The condylar twist angle (CTA) defined as the angle between the posterior condylar line (PCL) and CEA and the lift-off angle (LOA) defined as the angle between PCL and tibial cutting surface were measured. Furthermore, we investigated the influences of the amount of medial soft tissue release during surgery on postoperative CTA. Results: Postoperative CTA was 2.6±2.6 and LOA was 1.8±2.7. There was a significant positive correlation between postoperative CTA and LOA (r=0.55, P\0.0001).

123

S284

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: A retrospective radiographic analysis of 117 primary TKA cases was performed. Mechanical instrumentation was employed in all the cases. Postoperative AP hip to ankle and lateral knee radiographs were measured to evaluate component positioning. The following angles were measured: the hip-knee-ankle (HKA) angle, expressing the mechanical axis of the leg, the supplementary angle of the mechanical lateral distal femur angle (a), since we have determined it in the medial side, the medial proximal tibia angle (b), the femoral valgus angle (F) and the posterior tibial slope (PTS) measured in the posterior side. Results: Postoperative knee alignment: 66,7% of cases in the neutral group (HKA 1808 ± 38) (average HKA angle of the group was 179,158 ± 1,718), 29,1% in the varus group (HKA B 1768) (average HKA angle was 173,598 ± 2,038) and 4,2% in the valgus group (HKA C 1848) (average HKA angle of the group was 185,68 ± 0,898). The average HKA angle of the 117 TKAs was 177,818 ± 3,498. The average femoral valgus angle determined during the surgery was 5,89 degrees ± 0,63 degrees. The average postoperative femoral valgus angle was 6,778 ± 1,598. The average of PTS determined during the surgery was 84,448 ±±1,658. The average PTS of TKAs was 85,448 ± 2,948. We obtained statistically significant difference between F angle and PTS determined during the surgery and postoperative F angle and PTS. We have not obtained significant difference between different surgeons and different mechanical alignment systems. Conclusions: This study emphasizes that current mechanical instrumentation does not result in a high incidence of accuracy when each step of the procedure is measured.

Correlation between postoperative CTA and LOA The knees in which postoperative CTA was more than 3 required significantly more extensive medial release during surgery than knees in which it was less than 3 degrees (P=0.004).

Table 1 The amount of medial release More than 3 Less than 3 of Postoperative of Postoperative CTA(n=44) CTA (n=26) Deep-MCL

22 knees

5 knees

Posteromedial capsule

3 knees

1 knees

Sleeve release above the pes insertion 6 knees

5 knees

Sleeve release below the pes insertion 13 knees

15 knees

In 8 of 26 knees with more than 3 of postoperative CTA and 2 of 44 knees with less than 3 of postoperative CTA, postoperative LOA was more than 5. Conclusions: Internal rotation of the femoral component required extensive medial soft tissue release during surgery. Furthermore, it was difficult to achieve an appropriate balance in flexion after surgery in these cases. Our findings suggested that the CEA was a useful bone landmark for rotational alignment in order to achieve an appropriate balance in flexion and to minimize medial soft tissue release.

P19-196 Accuracy of current mechanical instrumentation Leo´n V.J.1 1 Hospital Vega del Rı´o Segura, Orthopaedic Surgery Department, Cieza, Spain Objectives: Recent studies are changing the premise that well-aligned TKAs have better survival than outliers. Till now it was considered that the long-term outcome of total knee arthroplasty (TKA) was related to the ability of the surgeon to achieve the desired alignment based on preoperative planning. Mechanical TKA alignment systems have fundamental limitations that limit their ultimate accuracy. The purpose of this study was to assess the accuracy of conventional mechanical intramedullary instrumentation systems.

123

P19-321 Driving reaction time before and after total knee arthroplasty: a prospective cohort study Liebensteiner M.1, Kern M.1, Haid C.1, Krismer M.1 1 Innsbruck Medical University, Innsbruck, Austria Objectives: Total knee arthroplasty (TKA) is a demonstrably successfully procedure being performed increasingly often in patients with osteoarthritis of the knee. Particularly the elderly are dependent on their driving abilities in order to execute activities of daily living. Those scheduled for TKA frequently wish to know when they can resume driving after the operation. Driving reaction time (DRT) is one of the most important factors responsible for driving safety. Two studies have addressed the effect of TKA on DRT but were effected by several limitations (e.g. patients with simultaneous bilateral TKA, small sample size). Given the high incidence of knee osteoarthritis which is intensified by an aging population, the increasing numbers of TKA being performed, and the restrictions of the above mentioned studies DRT before and after TKA has become a highly relevant issue. We hypothesized that DRT would differ on longitudinal comparison between preoperative, postoperative and follow-up conditions among patients undergoing TKA. Additionally it was hypothesized that the DRT of patients after TKA would significantly differ from that of healthy controls at each point in time. The third hypothesis was that there would be significant differences between preoperative, postoperative and FU DRT within certain subgroups (divided according to driving frequency, the side of the operated leg, and age). Methods: 31 consecutive patients (mean age 65.7 years, SD 10.2) receiving TKA were included. Data concerning the patients0 subjective driving frequency were obtained. The control group consisted of 31 healthy subjects (mean age 52 years, SD 7.7). Based on apparatuses described and validated in the published literature we devised an experimental device to measure DRT. When the accelerator was completely depressed a green lamp shone, indicating that the patient did not ‘drive’ in a ‘ready-to-brake fashion’. After 5 to 10 seconds the same investigator pushed an external trigger invisible to the patient, which activated a red lamp and the electronic time clock. Subjects were instructed to apply the brake ‘as quickly as possible’ with the right foot when the signal appeared. The time interval between the appearance of the signal and the subject operating the brake was taken as DRT and measured ten times by this procedure. Patients were tested preoperatively, postoperatively and after 8 weeks follow-up. Median (Md) and interquartile ranges (IQR) were calculated as descriptive statistics. For inferential

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 statistics the Friedman test was applied to test for differences in DRT. Wilcoxon tests were used for post-hoc analyses. For comparison with the control group Mann-Whitney-U tests were applied. Results: DRT increased from 620 msec (pre-op) to 640 msec (post-op) (n.s., p=0.06, increase in maximum and median values) and significantly decreased after 8 weeks FU to 617 msec (p\0.001). Patients driving ‘frequently’ or ‘very frequently’ showed no increase in their DRT postoperatively. Controls had a DRT of 487 msec, which was significantly different from the DRT of patients on all three test occasions (p\0.001). Conclusions: DRT was slightly impaired at discharge after TKA but was significantly improved after the 8-week FU. An 8-week period of driving abstinence may be deemed safe with reference to DRT in patients receiving TKA. We observed no relevant impairment in patients who rated themselves as frequent or very frequent drivers.

P19-331 Five- to nine-year clinical results of cemented zirconia-ceramic total knee arthroplasty Kitamura N.1, Kitayama S.1, Arakaki K.1, Kondo E.1, Tohyama H.1, Yasuda K.1 1 Hokkaido University, Dept. of Sports Medicine and Joint Surgery, Sapporo, Japan Objectives: Ceramic total knee arthroplasty (TKA) was introduced as a new-generation knee prosthesis in clinical practice, expected to reduce polyethylene wear due to its resistance to abrasion and lubrication. Alumina has been used in particular for the femoral component, however, the component is thicker than the equivalent metal implant and thus requires a larger bone resection. Using zirconia, which is a tougher ceramic material, we developed a zirconia-ceramic LFA-III total knee system in 1999, which is composed of a zirconia-ceramic femoral component and a titanium-alloy tibial component. The femoral component was designed using CT data of normal Japanese knees and its thickness is almost the same as that of the commonly used metal component. The purpose of this study was to evaluate clinical results in patients who had the zirconia-ceramic LFA-III system with a minimum five-year follow-up. Methods: Fifty-nine consecutive primary TKAs were performed in 49 patients between 1999 and 2004 using the cemented zirconia-ceramic LFA-III system. Four patients (six knees) had died of causes unrelated to the index arthroplasty, and five patients (five knees) were lost to followup. The remaining 48 knees were clinically and radiographically evaluated in the present study. The average follow-up time was 6.8 years (range, 5.0 to 9.9 years). The average age of the patients was 68.8 years (range, 50 to 84 years) at the time of the operation. Forty-six knees were in women, and two were in men. The diagnosis was osteoarthritis in 40 knees and rheumatoid arthritis in eight. Clinical assessment was conducted with the Hospital for Special Surgery Knee rating score and the Knee Society scoring system. The component alignment and bone-implant interface were evaluated according to the Knee Society Roentgenographic Evaluation System. Results: The mean postoperative knee score was 93.8 ± 8.0 and 90.8 ± 7.0 using the Knee Society and the Hospital for Special Surgery rating systems, respectively. Two knees were required reoperations within 5 years (one femoral loosening and one supracondylar fracture). Nonprogressive radiolucencies less than 2-mm thick were present in five tibias. These were located primarily on the medial side. A small (\1.5 cm) marginal erosion consistent with osteolysis was observed in the medial tibial plateau in two cases and in the anterior tibial plateau in one case. These implants did not migrate, subside, or have a circumferential radiolucency consistent with radiographic loosening. Conclusions: In this consecutive series of patients, the zirconia-ceramic LFA-III has performed well at an average 6.8 year follow-up. The lack of a ceramic breakage or patellar complications with this anatomical design was notable. Although we could not draw any conclusions regarding the superiority of the ceramic prosthesis with respect to wear and survivorship, the present study encouraged a further long-term follow-up study on ceramic prostheses.

S285 P19-338 Short term results with 101 consecutive journey total knee replacements Christen B.1, Neukamp M.S.2, Eichler P.3 1 Orthopa¨dische Klinik Bern, Salemspital, Bern, Switzerland, 2 Orthopa¨dische Klinik Bern, Bern, Switzerland, 3Orthopa¨die Frauenfeld, Frauenfeld, Switzerland Objectives: We report about a prospective, non-randomized series of 101 consecutive bicruciate stabilized Journey total knee arthroplasties. Expectations included better biomechanics and therefore deeper flexion and a more physiological kinematics without elevated complication rate. Methods: Since 1st of December 2006 till December 2008 101 consecutive TKR Journey Knees (Smith&Nephew) were implanted by a single surgeon in 73 women and in 28 men. The mean age at surgery was 68,9 years (46,5 - 84,6). In 88 knees with neutral or varus alignment a mini midvastus approach, in 13 cases a lateral approach including an osteotomy of the tibial tuberosity was necessary. The alignment was reached with the standard instrumentation with extramedullary aiming at the tibia and intramedullary at the femur. Routinely a tensioner was used for ligament balancing. All knees were examined 2, 4 and 12 months after surgery by the authors. All the TKR were documented at the Swiss Register SIRIS. Results: The mean flexion reached in mean 120 (80 - 140) 2 months after surgery with one outlier with 45 of flexion because of arthrofibrosis. Flexion improved to a average of 126  (115- 145) at 4 months and reaches 130  (85 - 145) after one year. All knees reached full extension after 4 months. 74% of the patients could kneel, 38 % were able to squat. In 3 knees a slight medial midflexion instability (\6) was recorded. 7 knees had to be revised because of mechanical problems. Four cases suffered from an irritation of the iliotibial band, three of them were revised arthroscopically. One lose patella button had to be revised. Out of two knee dislocations one case had to be revised completely. One revision had to be performed because of an infection due to an erysipelas. There were no cases of thrombosis, no vascular or nerve injury. Conclusions: The Journey Knee allows excellent mobility with a mean flexion of 116 after 2 months and 130 at one year. Patients were very satisfied in 80% of the cases. Complication rate was higher than expected and than reported by other authors but was decreasing with the experience of the surgeon. The surgical technique is demanding as the knee seems to be less forgiving than a standard prosthesis.

P19-351 Quantitative analysis of Polyethylene wear particles in synovial fluid by scanning electron microscopy. Comparative study between UHWMPE and crosslinked polyethylenes Lasurt S.1, Torner P.1, Francisco M.1, Prats E.2, Fontarnau R.2, Segur J.M.1, Lozano L.1, Martinez J.C.1, Castillo F.1, Sastre S.1, Popescu D.1, Suso S.1 1 Hospital Clinic i Universitari de Barcelona, Knee Unit. Orthopaedic Surgery Department (ICEMEQ), Barcelona, Spain, 2Scientific and Technic Services of University of Barcelona, Scanning Electronic Microscopy Department, Barcelona, Spain Objectives: Ultra High Molecular Weight Polyethylene (UHMWPE) has been used as a substitute surface in total knee replacements. It undergoes cyclic stresses in the tibial tray during load bearing activities resulting in fatigue fractures. Wear debris generated in vivo leads to osteolysis and aseptic loosening of the implant altering its survivorship. In recent decades, several studies have been carried out to discover ways to reduce polymer wear debris through crosslinking to improve mechanical properties. Currently, it is still controversial whether or not crosslinked polyethylene improves wear resistance compared with UHMWPE. To date, many in vitro studies have been published in the literature describing particles isolated from tissue samples in vivo that were taken at revision surgeries.

123

S286 Methods: A controlled pilot study involving 25 patients to whom bilateral total knee replacements were performed was carried out. The surgeries were performed between 2003 and 2004. Only one design of prosthesis was implanted in all knees. The type of polyethylene varied between either UHMWPE or crosslinked. Previous consent informed was obtained from these patients. Synovial fluid samples were collected by one of the surgeons of the department at 6 months, 1 year, 3 years and 5 years after surgical treatment. Synovial fluid samples were processed and observed by Scanning Electronic microscopy. The findings were confirmed by image analysis and Spectrometry. Polyethylene particles were identified, characterized and counted. Results: We collected a minimum of 1 milliliter synovial fluid from each patient for analysis. In our study, we did not find a significant difference between number of particles isolated in two types of polyethylene at one and at three years after surgery. The range of particles size varied considerably from less than 1 to 40 lm. The predominant morphologic features of these particles observed were rounded, elongated and fibrillar. Conclusions: In vivo analysis of synovial fluid demonstrates much less wear debris than what is noted in vitro. In vivo analysis requires an enzymatic digestion process and preparation in order to remove blood red cells and other organic components, such as bone or cartilage. Contamination by other type of particles should be taken into account with in vivo analysis. According to the literature submicron-sized particles are the most frequent and contribute to bone osteolysis. These particles are present not only in the synovial fluid but tissue surrounding the knee implant.

P19-357 Translation and adaptation to Spanish of the Knee Society Clinical Rating System (KSS) and the Hip & knee outcomes questionnaire in patients undergo to total knee arthroplasty Ares O.1, Castellet E.2, Amillo J.R.1, Hinarejos P.3, Montserrat F.4, Valenti J.R.5, SEROD (Sociedad Espan˜ola de la Rodilla) 1 Hospital de Viladecans, Viladecans, Spain, 2Hospital Vall d’Hebron, Barcelona, Spain, 3IMAS, Barcelona, Spain, 4Hospital de l’Esperanca (IMAS), Barcelona, Spain, 5Clinica Universitaria, Orthopaedic and Traumatology Surgery Dpt., Pamplona, Spain Objectives: The aim of the study was to perform the cross-cultural adaptation to Spanish version of the two scores. We have chosen two questionnaires regarding knee surgery to cross-cultural adaptation. We have chosen these questionnaires because their high interest: 1/ The knee society clinical rating system (KSS). This score was first published in CORR in 1989 and has become the standard clinical evaluation system for reporting results for patients undergoing total knee arthroplasty, and 2/Hip & knee outcomes questionnaire. This questionnaire has been endorsed more recently by prestigious scientific societies (American Academy of Orthopaedic Surgeons, American Knee Society, among others). The cross-cultural adaptation to Spanish is mandatory to validate these questionnaires in Spanish population. Methods: Cross-cultural Adaptation. We made a double translation for each of the questionnaires by 2 translators native Spanish. These translations were revised we made two versions in Spanish. These versions were mixed and only one new Spanish version (v0.1) was obtained. This version was translated again to English by native English translator and was compared with the original version. Posteriorly was written the second intermediate version in Spanish (v.02). Finally was made new single version (v1.0) in parallel of the second version intermediate in Spanish and the original version in English by native Spanish translator and one of the authors. The new version (v1.0) was analysed by an expert panel and a patient’s panel. The expert’s panel was composed by 4 expert’s surgeons in knee surgery and each of items was qualified regarding compressibility and relevance. The expert panel suggests modifications of the new version v1.0 of both

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 scores. The patient panel was composed by 20 patients who received the new version v1.0 of the AAOS H&K score. The patient panel was asked in compressibility and relevance. Results: We have created a Spanish version of the knee society clinical rating system (KSS) and Hip & knee outcomes questionnaire by a standard and validated method. The assessment of its metrics properties is being evaluated. Conclusions: The process followed assured maximal conceptual equivalence between the Spanish and the original Knee Society Clinical Rating System (KSS) and the Hip & knee outcomes questionnaire in patients undergo to total knee arthroplasty.

P19-369 Minimalinvasive implantation of polyaxial angle stable implants In the treatment of periprosthetisc fractures of the distal femur El-Zayat B.F.1, Zettl R.2, Efe T.1, Kru¨ger A.2, Eisenberg F.2, Ruchholtz S.2 1 University Hospital Giessen and Marburg GmbH, Department of Orthopaedics and Rheumatology, Marburg, Germany, 2University Hospital Giessen and Marburg GmbH, Department of Trauma, Hand and Reconstructive Surgery, Marburg, Germany Objectives: The periprosthetic fracture of femur mostly in elderly patients is still a challenging surgical procedure. Aim of the therapy is a fast and fullbearing mobilization of patients to get them back home as fast as possible. Aim of this study is the prospective evaluation of minimal invasive, percutaneous implantation of non-contact-bridging (NCB) plates at femur fractures in this specific population. Methods: In the period from January 2008 to September 2009, 35 osteosynthetic procedures with NCB-plates in femur fractures were registered prospectively. Out of them 29 were geriatric and osteoporotic patients. In this study the minimalinvasive fixation was performed percutaneously via a targeting frame. The data acquisition included intraoperative data (surgery duration, image-intensifier, blood loss), fracture consolidation, resilience and early complications. An x-ray and clinical follow up was performed after 6, 12 and 24 weeks. As additional parameter we assessed the Glasgow Outcome Score (GOS) before as well as 6, 12 and 24 weeks after trauma. Results: A total of n=29 patients with an average age of 74,5 (39-93) years and a mean ASA-score of 2,8 (2-4) were registered. 21 patients had a periprosthetic/-implant fracture, 10 in TKA, 6 in THA, 4 in DHS/PFN and 1 after TKA and THA implantation. 42 % of pat. (n=12) received an additional fixation e.g. by steel wires. The mean surgery time was 122 min. (84-175 min), the image intensifier time 2,6 min. (0,7-5,6 min.) and the blood loss 0,8 ECs (0-4). The early complications (11 % in total) included one secondary dislocation as well as one overlength of a screw, which had to be corrected and one plate breakage. The x-ray follow-up after 24 weeks showed in all pat. an adequate consolidation without secondary dislocation and reasonable function of the knee joint. A total of n=3 (11 %) of the pat. died after discharge due to the geriatric diseases 3, 4 and 12 weeks after surgery. The GOS 24 weeks postoperatively showed a decrease of two degrees in 4 pat. (14 %), of one degree in 8 pat. (28 %) and was in 58 % of pat. on the same level like before the trauma. Conclusions: The minimalinvasive, percutaneous, polyaxial, angle stable NCB-plate osteosynthesis in periprosthetic femur fractures is a safe alternative with a low complication rate. In these very sensitive and diseased pats. the early revision rate within the first 6 months is compared to similar procedures noticeable lower.

P19-376 TKR in severe and morbidly obese patients. Influence of complications and anthropometry of the knee in outcomes Lozano L.1, Nun˜ez M.2, Popescu D.3, Nun˜ez E.4, Torner P.1, Castillo F.1, Segur J.M.5, Sastre S.3, Martinez-Pastor J.C.1, Morales J.C.6, Suso S.5

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 1 Hospital Clinic University of Barcelona, Knee Unit, Orthopaedic Department, Barcelona, Spain, 2Hospital Clinic University of Barcelona, Functional Readaptation, Barcelona, Spain, 3Hospital Clinic University of Barcelona, Knee Unit, Barcelona, Spain, 4Institut Catala` de la Salut, SAP Suport al Diagno`stic i al Tractament, Barcelona, Spain, 5Hospital Clinic University of Barcelona, Knee Unit, Orthopaedic Surgery Department, Barcelona, Spain, 6Hospital Clinic University of Barcelona, Orthopaedic Department, Barcelona, Spain Objectives: Total knee arthroplasty (TKA) in patients with severe and morbid obesity is one of the current challenges in prosthetic knee surgery. The objective of this study was to describe the complications presented by patients with a BMI [ 35 kg/m2 undergoing TKR who require early reintervention and evaluate the influence of the anthropometric characteristics of the limb on health outcomes after TKR. Methods: We studied the initial evolution and early complications in 127 patients with severe or morbid obesity (Body Mass Index [ 35 kg/m2) undergoing total knee replacement (TKR) with one year of follow-up. The clinical and functional status was prospectively analyzed using the Knee Society Score (KSS) knee and functional scores and the quality of life by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) questionnaire preoperatively and at 12 months of follow-up. In all patients knee anthropometry of the knee requiring surgery were calculated. The influence of knee anthropometry and complications on functional results were evaluated. Results: There was a significant improvement in KSS knee and functional scores and an improvement in all WOMAC dimensions at 12 months. Seventeen (13.39%) patients underwent reintervention before 12 months. Nine of the seventeen patients were diagnosed with acute deep infection. There were statistically significant differences in WOMAC and KSS scores between patients with and without complications at 12 months. Linear regression analysis showed that the infrapatellar index inversely affected the WOMAC pain (p\0.011) dimension and the global WOMAC score (p\0.042) at 12 months. Conclusions: Postoperative complications and the infrapatellar index \75 correlated with worse functional outcomes

P19-389 Postoperative analgesia in total knee arthroplasty: a comparison of periarticular local anaesthetic infiltration and femoral nerve blockade McCleery M.1, Lomax A.1, Leach W.1, Rooney B.1, Hilditch G.2 1 Gartnavel General Hospital, Orthopaedics, Glasgow, United Kingdom, 2 Gartnavel General Hospital, Anaesthetics, Glasgow, United Kingdom Objectives: To compare post-operative pain, opiate use and outcomes after femoral nerve blockade (FNB) and periarticular local anaesthetic infiltration (PALAI). Methods: Cohort study of TKA patients between January and December 2008. All patients were operated under general (GA) or spinal (SP) anaesthetic under the care of the senior authors. From January to June patients received a FNB. Thereafter, all patients underwent PALAI using 20ml 1% xylocaine with adrenaline, 40mls 0.25% bupivacaine and 40mls normal saline. The postoperative (PO) opiate analgesia and rehabilitation was standardised for all patients. Patients not fulfilling these criteria were excluded. Data were collected for visual analogue scale pain (VAS) scores, nausea and vomiting (NV), antiemetic use, opiate use, flexion, mobility and Oxford scores. The data were compared using Student’s t-test and V2 testing for both methods for SP and GA patients. Significance level was pB0.05. Results: 133 patients were eligible, 58 patients underwent GA (24 PALAI and 34 FNB) and 75 patients underwent SP (47 FNB and 28 PALAI). There were no significant differences in patient demographics, rate of inflammatory arthropathies, preoperative opiate use or preoperative Oxford scores. Significantly less GA PALAI patients had NV in recovery (FNB 5 patients, PALAI 0 patients; p=0.049), this group also used less antiemetics on the 4th postoperative day (POD) (FNB 5 patients, PALAI 0 patients; p=0.049). Otherwise there were no significant differences in NV or antiemetic use in either group. GA PALAI patients had significantly lower median 7th VAS (Average 18hours PO.

S287 PALAI 0, FNB 0; p=0.034). There were no other significant differences in the median VAS or PO opiate use in either group. Significantly more patients in the SP FNB group required patient controlled analgesia (FNB 11 patients, PALAI 3 patients; p=0.0004). Significantly more patients were immobile on the 1st POD following FNB in both SP (FNB 15 patients, PALAI 5 patients; p=0.00005) and GA groups (FNB 15 patients, PALAI 0 patients; p=0.0002). There were significantly more patients with motor/ sensory blocks in the SP FNB group (FNB 12 patients, PALAI 7 patients; p=0.007). FNB patients continued to mobilise less ably, as significantly more FNB patients were mobile using a Zimmer frame with physiotherapist assistance on the 2nd POD in both the SP (FNB 12 patients, PALAI 10 patients; p=0.047) and GA groups (FNB 16 patients, PALAI 2 patients; p=0.002) and the 3rd POD amongst GA patients (FNB 7 patients, PALAI 0 patients; p=0.020). Range of motion was significantly greater for PALAI patients on the 1st POD, 2nd POD for the SP group and the 5th POD for the GA group. At discharge the mean ROM was significantly greater for SP PALAI patients (PALAI 83.3, FNB 75.6; p=0.010). However, there were no significant differences in the ROM by 3 months PO in either group. GA PALAI patients had a significantly shorter mean hospital stay (PALAI 4.67 days, FNB 6.53; p=0.007). There were no significant differences in the SP groups. There were no significant differences between the PALAI or FNB patients in the PO Oxford scores or mean difference in pre- to postoperative Oxford scores in either group. Conclusions: Periarticular local anaesthetic infiltration is comparable to femoral nerve blockade for effective postoperative analgesia with similar opiate/ antiemetic use and rate of NV. It allows earlier postoperative mobilisation and shorter hospital stay. This study supports its use in TKA.

P19-393 The usefulness of six different methods in determining the patellar height in relation to the joint line in navigated total knee arthroplasty Kircher J.1, Wettig M.1, Roin T.1, Zilkens C.1, Ja¨ger M.1, Krauspe R.1 1 University Hospital Du¨sseldorf, Department of Orthopaedics, Du¨sseldorf, Germany Objectives: Restoration of the joint line is of utmost importance in primary total knee arthroplasty (TKA). Measurement of the position of the patellar in relation to the joint line can be performed in different ways, but no standardized method has been established for the measurement in total knee arthroplasty when the bony anatomy widely changes due to the necessary bone cuts. The Miura-Index references to the distal femur, which is a great advantage, because this is close to the real center of movement but neglects the slope of the tibial plateau and was therefore modified by the authors. The objective of the study is the analysis of the restoration of the joint line using intraoperative navigation in TKA and secondary the validation of six different measurement techniques. Methods: Prospective clinical investigation of n=44 navigated primary total knee arthroplasties 2006-2009. Measurement of six different indices for determining the position of the patella: Insall-Salvati-Index (IS), modified Insall-Salvati-Index (mIS), Blackburne-Peel-Index (BP), CatonIndex (CI), Miura-Index (MI) and a modified Miura-Index (mMI) and the flexion angle on the Xray. Two independent measurements on standard lateral X-rays pre- and postoperative. Statistical analysis SPSS 17.0. Results: Mean age 63.26±21.05 (n=36 female, 61.23±22.75; n=8 male, 72.39±4.01, p=0.178). The mean indices pre- and postop were: IS 0.95±0.19, 1.07±0.21; mIS 1.22±0.23, 1.39±0.24; BP 0.67±1.72, 0.76±0.19; CI 0.68±0.20, 0.89±0.18; MI 0.74±0.21, 0.69±0.20 and mMI 0.63±0.19, 0.54±0.21 (all p[0.05). The correlation of pre- and postop indices was: IS r=0.847, p=0.001; mIS r=0.394, p=0.007; BP r=0.222, p=0.142; CI r=0.360, p=0.015; MI r=0.455, p=0.004; mMI r=0.368, r=0.013. Flexion angle on the Xray preop 39.23±9.89 and postop 27.94±11.22. There was no significant correlation for the height of the used PE-insert or the Xray flexion angles with neither of the indices. The definition of

123

S288 the BP and CI is unclear for TKA. Measurement at the level above or below the tibial component results in significantly different indices (p=0.001). Conclusions: All postoperative indices were within normal values (if defined). The data of the study show the best correlation of pre- and postoperative position of the patella for the IS. This index is independent of any bone cut necessary for TKA. Together with the fact, that all indices were without significant correlation to the height of the used PEinsert (e.g. the changed bony anatomy) results in the statement, that there was an excellent restoration of the joint line in all cases by navigated TKA. Little variances in the extent of knee flexion did not result in deterioration of the patella indices. The BP and CI do not have a clear definition for postop measurements and are significantly different using two described alternatives. The modified MI did not show an advantage for the description of the position of the patella.

P19-396 Revision total knee arthroplasty due to mechanical failure: mid to long term follow-up Soudry M.1, Nierenberg G.2, Greental A.1, Nikomarov D.1, Falah M.2 1 Rambam Health Care Campus, Orthopaedic Surgery, Haifa, Israel, 2 Rambam Health Care Campus, Haifa, Israel Objectives: To evaluate the results of revision Total Knee Arthroplasty due to mechanical failure. Methods: Between 1985 and 2009, 92 cases of knees after TKA were revised, most by the senior author. Fifty eight patients were females. The primary diagnosis was osteoarthritis in 90 patients, rheumatoid arthritis in 1 and post-traumatic in 1 patient. Five patients were revised after UKA. Postoperatively, patients were asked about their subjective satisfaction. Objective results were graded according to HSS score. The differences between preoperative and postoperative HSS score, arc of motion, average flexion and extension, were checked by t-test variance. Clinical evaluation was performed only after a minimum of five years of follow-up among 67 patients. Mean follow-up time was 10.9 years (range: 5-23Y). Mean age at revision was 73.3 Y (range:55-88 Y). Results: Time interval between primary and revision procedures was 94.6 months (range: 9-135 Mo). Loosening of tibial or femoral inserts were found to be as the cause of failure in 42 patients, instability in 11, wear of the polyethylene insert in 12 and bad surgical technique in 2 patients. Preoperative arc of motion was 101.8 degrees (between -2.7 extension to 104.5 flexion). Preoperative HSS score was 45.8. Intraoperative V-Y incision was needed in 2 patients (3%). The patella was revised in 53 patients (79%) and patellectomy was performed in 4 patients. Arc of motion at the end of revision averaged 104.5 degrees (between: -0.1 to 104.6). Implants used at revision were: 32 cases with CCK, 27 with TCPIII, 5 with Dual-articular, 2 with linked and 1 hinge prosthesis. Fifty patients were available for last FU, 15 died later on, and two patients were lost immediately after the revision. Complications were reported in 6 patients: 3 cases of instability and 1 loosening which underwent second revision with rotating hinge prosthesis, one case of dislocation of patella and one patello-tibial impingement which were treated with patellectomy. Subjective satisfaction was recorded by 59 patients (88%). Postoperatively, HSS score increased to 82.1 with 39% excellent result, 36% good, 16% fair and 9% poor results. Postoperative arc of motion was 90.55 degrees (between: -3.25 and 93.8). postoperatively, significant change was found in HSS score (p=0.019) but no significant change in average flexion (p=0.87) and arc of motion (p=0.34) were recorded in relation to preoperative status. Conclusions: In our hands, at 10.9 years follow-up in average with a minimum of five years, revision TKR due to mechanical failure resulted in 75% good and excellent results according to HSS score and 88% subjective satisfactory results, with salvageable complications.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P19-416 Posterior stabilised implants in primary total knee replacement Casteleyn P.-P.1, Ganzer D.2, Bayer-Helms H.3, Krause R.4 1 UZ Brussel, Orthopaedics and Traumatology, Brussels, Belgium, 2 Dietrich-Bonhoeffer-Klinikum, Altentreptow, Germany, 3Sint. JosefKrankenhaus-Hilden, Hilden, Germany, 4Orthopa¨dische Fachklinik Verein Oberlinhaus, Potsdam, Germany Objectives: There is an ongoing discussion concerning the effect of prosthetic design on the outcome of total knee prostheses. The purpose of this consecutive multicentre prospective case series study was to evaluate the short term clinical, functional, and radiological outcome of a posterior stabilised primary total knee arthroplasty (balanSys PS, Mathys Ltd Bettlach, Bettlach, Switzerland). Methods: From 01.09.2006 to 31.07.2008 256 (244 monolateral, n=6 bilateral) consecutive patients (m/w=1:2; mean age 70.0 yrs; range 44-87 yrs) were treated for osteoarthritis with a posterior stabilised total knee arthroplasty in 4 centres in Belgium and Germany. The total knee replacement was done by a medial approach in 96.5% of the surgeries. Patellar resurfacing was done in 59 % of the cases and mainly in two centres. The follow-up consisted of clinical as well as radiological evaluations, preoperatively, and at 6 weeks, 6 months, 1 and 2 years postoperative. Results: Follow-up information is until now available for 236 cases. The mean follow-up time was 14.9 months (range 1.3-30.3 months). The Knee Society Score increased from preoperatively 109.9 to 167.3 points. The mean VAS for pain (0-10) decreased from 7.4 (range 0-10) preoperatively to 1.7 (range 0-9). The mean VAS for satisfaction increased from 3.5 (range 0-10) preoperatively to 8.4 (range 2-10). Most striking with this design was the rapid recovery of the patients: – KSS, preoperative : 110, 3m :150, 6m : 170 – Knee score, preoperative : 57, 3m : 86, 6m : 93 – VAS pain, preoperative : 7.9, 3m : 3, 6m : 1.3 – VAS satisfaction, preoperative : 3.1, 3m : 7, 6m : 8 In our study patients without patella replacement had a better KSS (178.8 compared to 159.5, p \ 0.0001). The main difference was found in the function score but needs further analysis. Statistical analysis used the Wilcoxon 2 sided test and the Chi-square-test. As major complications we had to revise two patients for instability. 12 patients showed limited mobility and 1 patient is suspected of a radiological loosening. Conclusions: This type of posterior stabilised total knee prosthesis in combination with the described surgical techniques shows promising radiological and clinical short term results, with a rapid recovery and with an acceptable rate of adverse events. In our series patients without patella replacement show a significantly better function score than patients with a replaced patella, although this needs further analysis.

P19-457 Blood loss in total knee replacement: pharmacological vs mechanical hemostasis Antinolfi P.1, Innocenti B.2, Caraffa A.1, Cerulli G.1 1 University of Perugia, Perugia, Italy, 2European Centre for Knee Research, Leuven, Belgium Objectives: Assess the effect of a pharmacological action on post-op. bleeding after Total Knee Arthroplasty compared to mechanical hemostasis and a control group. Methods: Three groups were randomized in order to receive: 1) conventional intra-op hemostasis, post-op elastic compression 2) intraarticular injection of 500 mg acid tranexamic at the end of surgery and 3) 6 hours post-op bended knee. Only primary total Knee were included, all cemented and using the same Total Knee system. Assumption of Aspirin, warfarin or other pharmacological therapy as well as any co-morbidity influencing the coagulation systems were exclusion criteria for each group. We checked the drainage level on day ‘‘0’’, ‘‘1’’ and ‘‘2’’, we compared the variations in Hemoglobin and hematocrit value to the pre-op value

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

S289

P19-483 Small metal implants for restoration of knee articular cartilage defects Zazirnyi I.1, Kovalchuk V.1 1 Hospital ‘Feofania’, Centre of Orthopaedia, Trauma and Sport Medicine, Kiev, Ukraine Objectives: The aim of this low-invasive technique is to reduce pain and restore joint functionality by creating smooth and continuous surface able to sustain stress and pressure simultaneously filling the entire lesion. Methods: First study group included 11 patients (4 men, 7 women) aged 44 to 70 (mean age 57.9±4.6 years) having undergone metal implant local cartilage lesion prosthetics. Second group included 12 patients (7 men, 5 women) aged 40 to 65 (mean age 49.3±4.6 years) who underwent arthroscopic debridement of the knee joint. \bfr[ The criterion for inclusion into the study was an isolated cartilage damage of the medial or lateral femoral condyle load area of the III-IV grade of severity (Outerbridge) sized 1.13 lo 6.61cm2 (mean size 3.5±0.8cm2). The damaged area was located in the medial femoral condyle in 19 patients (82.6%) and in lateral condyle in 4 (17.4%) patients respectively. Results: Results of low-invasive cartilage endoprosthetics depended nonlinearly on the size of lesion and the size of implant. Direct correlation has been observed between good results and the lesion area sized 15 to 20 mm (1.8-3.14 cm2). Unsatisfactory results were observed in patients with a lesion sized more than 20 mm and less than 15 mm. Conclusions: Indications for the suggested technique are medial femoral condyle cartilage damages sized 15 to 20 mm (1.8-3.14 cm2) in patients aged 50 to 70 years.

P19-516 Patello-femoral function after total knee arthroplasty: gender related differences Giron F.1, Sensi L.2, Buzzi R.2, De Luca L.2, Aglietti P.3 1 First Orthopaedic Clinic - University of Florence, Firenze, Italy, 2 University of Florence, First Orthopaedic Clinic, Firenze, Italy, 3 University of Florence, First Orthopaedic Clinic, Florence, Italy Objectives: Patello-femoral complications are recognized as common after Total Knee Arthroplasty (TKA) and are a possible cause of revision surgery. It’s recognized that TKAs have more complications in women, including a higher incidence of anterior knee pain. The purpose of the study was to evaluate patellofemoral results in two comparable groups of male and female patients undergoing TKA. Methods: In order to include the same number of males and females, 50 consecutive males operated from January 2002 to July 2003 were included as well as the first woman operated just after each male. The two groups were comparable in terms of surgical technique, postoperative pain control, rehabilitation protocol and follow-up. The implant was posteriorly stabilized with a dome patellar replacement. The patients were studied using Knee Society knee and functional scores and the HSS patellar score. Radiological study included standard and weightbearing axial views and a CT scan to assess component rotation. Results: The average follow-up was 6 years (range 61-78 months). None of the patients was lost to follow-up. The two groups achieved a satisfactory and similar knee score; females had a significantly lower functional score (86.1 vs 96.5; p\.05) and patellar score (91.8 vs 96.6; p\.05). The incidence of patellofemoral pain after TKA was 8%. Patellofemoral pain was observed in 6 females and 2 males and it was mild in 5 cases and moderate in 3 (n.s.). There was no significant correlation between the incidence of patellofemoral pain and gender, BMI, preoperative deformity, and lateral retinacular release. The incidence of lateral patellar tilt, subluxation and lateral impingement was decreased in weightbearing views compared to non weight bearing. Medial patellar bony impingement was evident only in weight bearing views and correlated with pain (p\0.05). Femoral component rotation was ?1.4 on average with a range from 0 to ?6. Tibial component rotation was ?5.1 on average with a range from 0 to ?12. No femoral or tibial component was internally rotated. Conclusions: In conclusion females had significantly lower results in terms of functional and patellar scores than males after TKA. The medial patellar impingement during active quadriceps contraction should be considered as a possible cause of patellofemoral pain after TKA.

P19-510 The first 100 Zimmer Unicondylar Knee replacements - no learning curve Kalayci K.1, Nicolai P.1 1 West Suffolk Hospital, Orthopaedic, Bury St Edmunds, United Kingdom Objectives: Analysis of the first 100 Zimmer Unicondylar Knee replacements. Methods: Prospective cohort study in a District General Hospital with a follow-up of over 4 years. Demographic data, body mass index, side, Kellgren radiological score, implant sizes, status of the patello-femoral joint, tourniquet time, length of hospital stay and American Knee Society scores were collected by an independent assessor pre-operatively, and at 6 months, 1 year and 2 years postoperatively. Results: The average age at operation was 66 years. The average tourniquet time was 69 minutes and the average length of hospital stay 3.5 days. The average Knee score improved from 53.2 to 95.7 and the average Function score from 51.5 to 96.7 in 43 patients who were at least 2 years postoperatively. The scores continued to improve for up to 2 years postoperatively. Average knee flexion improved from 113 to 125 degrees. There were no major complications and no revisions to date. Two patients had further injuries resulting in lateral meniscal tears, which were treated arthroscopically. Conclusions: The results are similar if not better compared to other Unicondylar knee replacements at this stage. There was no learning curve, with knee and function scores of the first 10 patients being similar to the scores of the last 10 patients.

P19-525 Functional status of the patients treated with articulating spacer for infected total knee arthroplasty Martinez-Pastor J.C.1, Vilchez F.1, Segur J.M.1, Macule F.1, Lozano L.1, Torner P.1, Castillo F.1, Sastre S.1, Popescu D.1 1 Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain Objectives: Infection is one of the most devastating complications after total knee arthroplasty (TKA). Two-stage exchange arthroplasty remains the standard treatment of infection at site of a TKA. The temporary cement spacer is designed to deliver antibiotics to the area definitive second-stage surgery. We show our experience in 35 patients affected with chronic TKA infection treated with Spacer knee between the first and second time surgery. Methods: This is retrospective study; 35 patients who had undergone a 2stage exchange for septic total knee arthroplasty, between January of 2003 and July of 2007 were identified. The mean age at time of surgery was 73 years (range, 61- 88 years). There were 18 female and 17 male. The mean body mass index was 29.9(range, 24-37). In all patients prosthesis of antibiotic-loaded acrylic cement (Spacer) was implanted during the first stage after debridement. The diagnosis of infection was based on clinical examination, X-rays, C-reactive protein (CRP), erythrosedimentation rate (ESR) and joint fluid or tissue culture. The CPR mean at diagnosis was 12.09 mg/dl (range, 0.5-54) and ESR 70 (range, 15-125).

and we analyzed the variation in blood transfusions needed in each patient. Results: Patients who received pharmacological hemostasis showed a significative reduction of blood loss and need for transfusions. The mechanical compression didn’t help reduce the post-op bleeding compared to control group. At the moment of this writing, check of the last patients is still in progress but, as for now, incidence of complications does not seem to be influenced by the treatment received. Conclusions: Intra-op and post-op bleeding is an important factor in Primary TKA. Based on our experience, use of tranexamic acid is useful to reduce the blood loss and need for a transfusion, with a direct effect on possible complications related to acute anemia in these kind of patients.

123

S290 Results: The microorganism was isolated in 33 patients, 21 cases (63%) were Sthaphylococcus spp. The duration of antibiotic treatment between the first and second stage was 71 days (range, 23-240 days). The C-reactive protein at the moment of the second stage was 0.5 mg/dl (range, 0-1.6), the time mean between stages was 130 days (range, 40-330). The days of hospitalization for the infection were 40(range, 14-115), The mean follow up was 2.7 years (range, 1-5.6), the infection was controlled in 31 patients (88%), 4 patients (11.5%) the 2-stage exchange did not controlled the infection and was necessary rescue surgeries. The mean WOMAC and KSS preoperative was 55 and 53 and the postoperative was 88 and 76, respectively. Conclusions: The results of our investigation provide evidence that articulating knee mobility and function between stages without incurring additional risk of infection. We suspect that the Spacer system, by virtue of its articulating design, may offer advantages(bone preservation, improved function and take easy the second stage) over static spacers, but these advantages will need to be borne out appropriately designed comparative trials.

P19-527 Tibial tubercle osteotomy in septic revision total knee arthroplasty Segur J.M.1, Vilchez F.1, Martinez-Pastor J.C.1, Macule F.1, Lozano L.1, Sastre S.1, Popescu D.1, Castillo F.1, Torner P.1 1 Hospital Clinic, Universidad de Barcelona, Knee Unit, Barcelona, Spain Objectives: The septic revision total knee arthroplasty is a challenge for the surgeon because is a knee with multiple surgeries and the exposure can often be compromised because the limited range of motion, the peripatellar contracture and capsular inflammation. Extensile exposures are frequently used in these cases, to allow adequate operative exposure without compromising the extensor mechanism. The aim of this study was to report our experience with tibial tubercle osteotomy (TTO) and patient functional outcome after this approach in 2-stage exchange total knee arthroplasty. Methods: This is retrospective study; twenty six patients who had undergone TTO as part of the surgical exposure in 2-stage exchange septic total knee arthroplasty, between January of 2003 and July of 2007 were identified. The mean age at time of surgery was 73 years (range, 64- 88 years). There were 14 female and 12 male. The mean body mass index was 29.8(range, 37-24). In all patients prosthesis of antibiotic-loaded acrylic cement (Spacer) was implanted during the first stage. The duration of antibiotic treatment between the first and second stage was 67 days (range, 23-240 days). Then of the normalization of C-reactive protein, was done the second stage, mean 133 days (range, 40-291). Results: The microorganism was isolated in 24 patients, 18 cases (69%) were Sthaphylococcus spp. The mean follow up was 2.7 years (range, 15.6), the infection was controlled in all the patients. Cerclage wire fixation was used in 25 patients and in 1 patient the fixation was done with Ethibond. A lateral release was used in 13 of the 26 knees to improve central tracking. The satisfactory healing of the osteotomy with union occurred in 24 cases. The mean WOMAC and KSS preoperative was 55 and 53 respectively; and the postoperative was 88 and 76. Minor complications that did not affect the final clinical result occurred in 4 patients (15%), 2 patients had a stable fibrous union of the osteotomy without extension lag, other patient had a fracture of the tibial tubercle al level of the cerclage with a correct osseous union, and other patient had an extension lag of 5 grades. Conclusions: The TTO gives a good exposure in the revision total knee arthroplasty, prevent the no controlled lesion of the extensor mechanism, and with a correct technique is a safe approach. The active flexion an extension can be done in the early rehabilitation as well as the complete bearing of the extremity. The infection does not give an extra comorbidities in this approach, all the complications are minors and do not interfere with the final outcome.

P19-529 Matched pair TKA retrieval analysis: oxidized zirconium vs. CrCoMo Heyse T.1, Chen D.2, Kelly N.2, Boettner F.3, Wright T.2, Haas S.3

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 1

Universita¨tsklinikum Giessen und Marburg GmbH, Klinik fu¨r Orthopa¨die und Rheumatologie, Marburg, Germany, 2Hospital for Special Surgery, Department of Biomechanics, New York, United States, 3Hospital for Special Surgery, Adult Reconstruction Service, New York, United States Objectives: Oxidized zirconium (OxZr) is used as a ceramic surface for femoral components in total knee arthroplasty (TKA). The aim of this study was to investigate its performance by examining retrieved femoral components and their corresponding PE inserts in matched comparison with conventional chrome/cobalt/molybdenum alloy (CrCoMo). Methods: 11 retrieved posterior stabilized (PS) TKA with an OxZr femoral component were included. From a cohort of 56 retrieved TKA with CrCoMo femoral components, pairs were matched according to duration of implantation, patient age, reason for revision, and BMI. The retrieved tibial polyethylene (PE) inserts were analyzed for wear using the Hood classification. Femoral components were optically viewed at 8-32x magnification and screened for scratching, pitting, delamination, and striation. Profilometry was performed to measure surface roughness of the OxZr components using a non-contact white light profiler. Results: The prostheses were in situ for a mean of 18.5±10.6 (OxZr) and 19.5±14.3 (CrCoMo) months (p = 0.41). None of these cases were revised for problems directly related to the use of OxZr. There was one reaction to the implant in the CrCoMo group. The average wear of the tibial PE inserts was significantly lower with OxZr components (41.5±16.8 vs. 60.1±22.0, p=0.01). The average wear score in the visual analysis of the femoral components was significantly lower for the OxZr (1.6±1.3 vs. 9.5±0.6, p=0.005). Conclusions: Femoral components made of OxZr are less sensitive to in vivo wear damage than those of CrCoMo. PE inlays show significantly less in vivo wear damage in combination with an OxZr femoral component.

P19-532 Navigation of TKA: rotation of components and clinical results in a prospectively randomized study Schmitt J.1, Hauk C.1, Kienapfel H.2, Pfeiffer M.3, Efe T.1, Fuchs-Winkelmann S.1, Heyse T.1 1 University Hospital Marburg, Department of Orthopedics and Rheumatology, Marburg, Germany, 2Auguste Viktoria Klinik, Berlin, Germany, 3Helios Rosmann Klinik, Breisach am Rhein, Germany Objectives: Navigation was introduced into Total Knee Arthroplasty (TKA) to improve accuracy of component position, function and survival of implants. This study was designed to assess the outcome of navigated TKA in comparison with conventional implantation with the focus on rotational component position and clinical mid-term results. Methods: In a prospectively randomized single-blinded approach, 90 patients with primary gonarthrosis were assigned to three different groups. 30 patients each were assigned to NexGen LPS without and with navigation (groups 1 and 2), and 30 patients to navigation with the Stryker Scorpio PS (group 3). Clinical outcome was assessed by a blinded observer applying the Knee Society Score (KSS) and a visual analogue scale. CT scans and radiographs were conducted prior to, and 12 weeks after index surgery. Results: 79 patients were available for clinical evaluation at 3 ± 0.4 years follow-up. Four implants had to be revised for early loosening or infection (4.4%). Four patients had died and three patients were not able to follow the invitation for clinical assessment. Functional results in the KSS were significantly lower after navigated TKA. Operation time and incisions with navigation were significantly longer. Significantly less radiological outliers with navigation were found for coronal alignment of the femur, only. Conclusions: In this series, no beneficial effect for navigation in TKA could be shown assessing clinical data, as functional results in the presented series seemed to be lower after navigated TKA. The clinical mid- to long-term value of navigation remains to be evaluated in larger patient series or meta-analyses at longer follow-up.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P19-535 UKA after spontaneous osteonecrosis of the knee at 11 years follow-up Heyse T.1, Khefacha A.2, Fuchs-Winkelmann S.1, Cartier P.2 1 University Hospital Marburg, Department of Orthopedics and Rheumatology, Marburg, Germany, 2Clinique Hartmann, Institut de Genou, Neuilly sur Seine, Paris, France Objectives: Safety and efficacy of unicompartmental knee arthroplasty (UKA) in osteoarthritis has been shown in large patient series. It has been matter of discussion whether or not spontaneous osteonecrosis of the knee (SONK) can successfully be treated with UKA. Methods: A retrospective approach included 52 cases of UKA for SONK of the femoral condyles. Four implants were revised (7.7%), and seven patients had died. Nine patients were interviewed by telephone, 28 followed the invitation for clinical examination including clinical scores (KSS and WOMAC) and radiographies. Satisfaction of patients was recorded in four categories. Four patients (7.7%) were lost to follow-up. Results: Average follow-up was 10.9 ± 4.8 (4 - 25) years. Average age at operation was 66.6 ± 9.7 years. Two knees were mobilized within 3 weeks after the operation due to stiffness. The KSS score increased from a preoperative 85 ± 30 to 173 ± 27 (p \ 0.0001) at latest follow-up. WOMAC was 7.7 ± 11.4 at latest follow-up. Most patients were satisfied (21.6%) or very satisfied (75.7%) with the outcome of this surgical procedure. One patient was dissatisfied (2.7%). Conclusions: This study shows that spontaneous osteonecrosis of the knee (SONK) can successfully be treated with UKA and prevent TKA with a good mid- to long-term follow-up.

P19-545 Rotating hinge knee prostheses: primary and revision implant Heyse T.1, Hauk C.1, Pahrmann C.1, Fuchs-Winkelmann S.1, Schmitt J.1 1 University Hospital Marburg, Department of Orthopedics and Rheumatology, Marburg, Germany Objectives: The aim of this study was to elucidate clinical short- and midterm results of a rotating hinge knee prosthesis. Results of primary implants were compared with those of revisions. Methods: 113 patients with 121 rotating hinge knee prostheses were included with a mean follow-up of 51.7 ± 40.8 months (58 primary implants, 63 revisions). 45 patients with 49 prostheses were available for clinical assessment (21 primaries, 28 revisions). Results: The clinical results of both groups did not show significant differences in the applied scores (KSS, UCLA, Lequesne, VAS). Of 121 included implants 18 had to be explanted (14.9%), mainly due to infection. For primary implants the explantation rate of 5.2% compared to 23.8% in revision cases. Conclusions: The rotating hinge prosthesis allows good clinical and functional results in both primary and revision knee arthroplasty. The risk for revision interventions and implant failure is significantly higher in revision cases.

P19-562 In-vitro testing of loosening of the femoral component in comparison of several high-flex and conventional PS designs Bollars P.1, Luyckx, J.P.2, Innocenti B.3, Labey L.3, Victor J.4, Bellemans J.5 1 KU Leuven, Orthopaedic Surgery, Leuven Pellenberg, Belgium, 2MSc, European Centre for Knee Research, Leuven, Belgium, 3European Centre for Knee Research, Leuven, Belgium, 4St Lucas Bruges, Bruges, Belgium, 5 KU Leuven, Leuven, Belgium Objectives: High-flexion (HF) TKA designs were introduced in order to achieve greater flexion than with conventional TKA designs. Although early clinical results are promising, recent literature raises concerns about fixation and risk for early loosening of the femoral component, associated with weight-bearing during deep flexion. This study’s aim was to measure the loosening force of the femoral component in several PS-TKA designs during a simulated deep flexion activity.

S291 Methods: The loosening force of the femoral component of ten contemporary PS-TKAs, including five HF and five conventional designs from the major orthopaedic companies were evaluated. Each TKA was implanted in a femoral bone model and placed in a loading frame in 135 of flexion, with the tibia vertically. Loosening of the femoral component was induced by raising the tibial insert with constant displacement rate, maintaining the same flexion angle. The resisting force was recorded continuously. A stereo-photogrammetric system registered the relative motion between the femoral component and bone model. The loosening force was determined when a gap of 2 mm was observed. The influence of pegs on the loosening force was also investigated. Results: Generally, conventional femoral designs required higher forces before loosening occurred compared to HF designs (p\0.001). This is due to the load sharing between the posterior femoral condyles of the TKA and the femoral bone seen with conventional designs in deep flexion. In the group of the HF designs there was at least one statistically significant difference between the designs (p=0.015), due to the shape of the internal box cut. For some designs, the presence of pegs induced a statistically significant change in loosening force Conclusions: Several design characteristics of the femoral component can alter its resistance to loosening. In this in vitro study, it was shown that the shape of the internal box cut and the presence of pegs, as well as the geometry of the pegs, are important factors for the loosening force.

P19-594 A prospective, single blinded, controlled trial comparing the postoperative range of motion of the P.F.C.Ò Sigma RP-F Knee with the P.F.C. Sigma RP Knee in primary total knee arthroplasty. Results after 1-year follow-up Castillo F.1, Macule F.1, Popescu D.1, Martinez-Pastor J.C.1, Lozano L.1, Torner P.1, Sastre S.1, Segur J.M.1, Suso S.1 1 Hospital Clinic, Universitat de Barcelona, Knee Unit, Barcelona, Spain Objectives: The primary objectives of a TKA are to reduce pain, maximise range of motion (ROM) and provide stability through the gait cycle. Routine actives of daily life require flexion of 1008 to 1258 for Western adult populations. The challenge is that the goals of patients undergoing TKA are changing. It is no longer sufficient to provide normal, pain-free function but now, patients desire knees that meet more demanding and active lifestyle needs. The press fit condylar P.F.C.  Sigma RP-F (rotating-platform, high flexion) knee is designed to provide a range of motion (ROM) of 155 degrees without compromising wear, polyethylene contact stresses, patellofemoral tracking, or stability. The purpose of this clinical study is to determine if the P.F.C. Sigma RPF Knee has a greater post-operative range of motion compared with the P.F.C. Sigma RP Knee and this function is maintained without compromising implant survival. Methods: This is a randomised controlled single blinded study. The subjects were randomly assigned preoperatively either to the study device P.F.C. Sigma RP-F Knee or its comparator P.F.C. Sigma RP Knee. A total of 100 patients were included in the study, 50 patients for each group. Age, sex, body mass index, preoperative diagnosis, grade of varus/valgus deformity and preoperative ROM were analyzed to determine the effect of design on postoperative ROM. We also evaluated the functional results with the KSS and WOMAC Scores preoperatively and 1-year after surgery. Results: The mean age was 66 years (47-75), the mean BMI 30.04 (2437.1) and the mean preop. ROM was 1018, with no statistical differences between the groups. The mean increase in active ROM in the Sigma RP-F group was 19.3 degrees, compared with 7.8 degrees in the rotating-platform group (P=0.001). The mean increase in active ROM in patients who had less than 95 degrees of preoperative motion was 29 degrees in the Sigma RP-F group, compared with 10 degrees in the rotating-platform group (P = 0.002).

123

S292 Both groups showed a significant improvement in their KSS and WOMAC Scores at 1-year follow-up but we found no significant differences between them (P\0.05). Conclusions: Femoral components with a specific design have proved that an improvement of flexion in TKR can be achieved comparing to the conventional models. The most significant determinant of postoperative ROM is preoperative ROM. This is more noticeable in knees that have less than 958 flexion than in knees that already had a high preoperative ROM. With the new P.F.C. Sigma RP-F design, greater ROM can be achieved independent of preoperative ROM. Nevertheless, this is not reflected in the functional Scores like KSS or WOMAC.

P19-604 Impact of knee osteoarthritis on the quality of life avaluated by SF-36. Significance of age and sex Hinarejos P.1, Vila G.1, Leal J.1, Puig L.1, Montserrat F.1, Caceres E.1 1 Hospitals IMAS (Esperanza-Mar) Barcelona, Barcelona, Spain Objectives: Some instruments that measure the perceived health are an important information source for investigation on several diseases, and the usefulness of several treatment measures. SF-36 Health Questionnaire is the most employed instrument in the last decades. Its validity and reliability has been proved in several studies and populations. The aim of this study was to compare SF-36 scores in a sample of patients with knee osteoarthritis to the standardised scores by age and sex in the Spanish population. Methods: SF-36 questionnaire has been administered in the preoperative period in 804 consecutive patients that have signed informed consent. Patients have answered the questionnaire with the assistance of a trained nurse, and 782 valid questionnaires have been used to this study. The studied population was 604 female (77.2%) and 178 male patients, with a mean age of 71.3± 7.9 (range 43 to 90) years, and 68.9 % of the sample was between 66 and 80 years. The answers of the questionnaire are classed in 8 different subscales: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional role and mental health. Each subscale can score between 0 and 100, and higher scores represent better health. Scores of each subscale have been treated as normal distribution. All data has been analysed with SPSS for Windows program, using t-Student and chi-square for parametric variables. Level of statistic significance has been set at p\0.05. Results: The studied population has a statistically significant reduction in the score in 7 of the SF-36 subscales when compared with the standardised Spanish population (p\0.001). Only in the general health subscale there is no this reduction in the score. In two of the SF-36 subscales the reduction of the scores has been especially important (more than 35 points respect the standardised scores): physical functioning and bodily pain. When we have analysed SF-36 scores in both sexes we have found that women have lower scores in the 8 subscales of the SF-36 questionnaire (p\0.05). We have not found any significant correlation between SF-36 and the age in any of the questionnaire’s subscale (p[0.05). Conclusions: Women with knee osteoarthritis have a worst health perception than men, but this finding is also found the standardised population without osteoarthritis. In the standardised population there are decreasing SF-36 scores when increasing ages, but we have not found this correlation, probably because our sample is quite homogeneous in age, with most of the patients with ages between 66 and 80 years. Knee osteoarthritis causes a significant negative impact on the health perception in the studied population, causing a reduction in the SF-36 in almost all subscales, especially in the bodily pain and physical function.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P19-613 Comparison of three different polyethylene designs in artificial knee joints—first clinical results Krause R.1 1 Oberlinklinik, Potsdam, Germany Objectives: The aim this consecutive prospective study is the clinical and radiological comparison between three different polyethylene designs in primary total knee arthroplasty. All three designs were implanted with the step-by-step surgical technique. In two groups the polyethylene design was a fixed-bearing (group PS: posterior stabilised system; group fixedbearing: cruciate retaining inlay). In the third group the inlay was a rotating platform (group RP), a cruciate retaining system too. In all three groups the balan-Sys implant system was used, the main difference is in the polyethylene inlay. All patients were operated in a German clinic by the same group of surgeons. Methods: Between January 2005 and September 2007 129 patients were consecutively included in this study (PS N=43, RP N=44, fixed-bearing N=42). The follow-up examinations were performed three and twelve months postoperative. All except one person had osteoarthritis as indication for total knee ar-throplasty. Compared to the two other groups in the PS group less women were included (PS 60%, RP 70%, fixed-bearing 79%). Results: In the PS group the mean age was with 72.4 years significantly higher (p B 0.01) compared to the two other groups (RP 66.2 years, fixedbearing 67.6 years). In the PS group the mean passive flexion angle increased from preoperative 115 to 123 at twelve months postoperative. Already three months postoperative the average flexion angle for the patients in the PS group reached 120. In the two other groups the passive flexion angle was higher preoperative with 114/114 (fixed-bearing/RP group) than 12 months postoperative with 111/110. Three months postoperative the total Knee Society Score was higher in the PS group compared to the two other groups, at the twelve months follow-up the difference was less distinct (three months (mean/median): PS 180/185 points, RP 154/158, fixed-bearing 156/164; twelve months: PS 174/184, RP 168/169, fixed-bearing 172/176). In the RP group two patients needed a revision. One prosthesis had to be revised due to malalignment of the tibia component. The other patient in the RP group complained about retropatellar load pain. This patient was treated by a patella replacement and an inlay exchange. In the PS group one femur component was revised after six months due to arthrofibrosis. At this stage of the study no patient in the fixed-bearing group needed a revision. Conclusions: Our short term results are very satisfying and promising. Especially the patients with a posterior stabilised polyethylene design reached excellent total Knee Society Score, despite of the high age. Many patients achieved these high clinical scores even three months after surgery. Patients in the fixed-bearing and the RP group had comparable clinical results three months postoperative. The difference in the mean age of the three groups is a coincidence and not indication based. Long-term results will show, if the clinical difference will remain.

P19-619 Revision total knee arthroplasty: an analysis of 75 consecutive cases Saragaglia D.1, Grimaldi M.1, Mercier N.1 1 Grenoble South Teaching Hospital, Orthopaedic and Sport Traumatology, E´chirolles, France Objectives: The purpose of this study was twofold: to present and evaluate the radiological and clinical results of a continuous series of 75 total knee revision arthroplasties (TKAr) implanted between January 1993 and March 2007 to replace failed total knee arthroplasty (TKA) and to search parameters influencing the clinical and radiological results. Methods: The series was composed of 43 females and 29 males corresponding to 75 knees (3 bilateral). The mean age of the patients at revision was 72.2 ± 10.1 years (28-92) and the waiting time between the first implantation and revision was 83±48.6 months (6-213). The causes for revision were: 40 aseptic loosenings, 14 septic loosenings, 12 polyethylene

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 wears, 6 periprosthetic fractures, 2 instabilities and 1 rupture of the implant. We used the Anderson Orthopaedic Research Institute (AORI) bone defect classification proposed by Engh and Ammeen to evaluate the bone defects. Regarding the prostheses we implanted, we revised only the tibial plateau in 15 cases and we used only 9 hinged prostheses. All the other prostheses were more or less constrained PS prostheses. We retrospectively reviewed 65 TKAr (three patients died and seven were lost to follow-up) implanted in 62 patients. Patients were assessed clinically and radiologically using the IKS score at a mean follow-up of 50±31.9 months (24-146). Results: The mean IKS knee score improved from 51.2 points pre-operatively to 85 (p \ 0.0001, t-test). The mean IKS function score improved from 23.8 points pre-operatively to 70.5 (p \ 0.0001, t-test). The mean overall IKS Score also improved from 75.4 points pre-operatively to 155.5 post-operatively (p \ 0, 0001). 5 knees required another revision: 4, for aseptic loosening and 1 for infection. Of these 5 failures, 4 were related to a revision of the single tibial component. The mean post-operative alignment was 180, 6±2.7 (range 174 to 188). Restoration of the mechanical axis was achieved in 61 cases (81%) according to a HKA angle of 180±3 and in 70 cases (93.3%) according to a HKA angle of 180±4. We observed revision-free survival of 92.2% (95% confidence interval, 0.75 to 0.99) at 5 years and 70.5% (95% confidence interval, 0.68 to 0.98) at 10 years. Conclusions: In this study, the mechanical axis, the age as well as the type of bony defect (cavitary or segmentary) did not influence the IKS score. The Charnley type C deteriorated in a non significant way (p=0.056) the post operative IKS function score. The female gender modified pejoratively the overall IKS score (p=0.048). The tibial tubercle osteotomy significantly deteriored (p=0.009) the average flexion range of motion. Finally, the septic reason for revision deteriorated significantly (p=0.044) the Knee Society function score.

P19-623 Relevance of computer-assisted recording of femoral mechanical axis to give rotation to femoral implant Mercier N.1, Saragaglia D.1 1 Grenoble South Teaching Hospital, Orthopaedic and Sport Traumatology, E´chirolles, France Objectives: The most used mean to give external rotation to the femoral implant when performing computer-assisted TKA is to record the transepicondylar axis. But everybody knows it is not reproducible and this recording can be misleading. We have been recording the femoral mechanical axis (FMA) to give rotation according to this angle for 6 years. This recording is easier because, on one hand, it is a bony landmark and palpation of the distal femur is not hampered by soft tissue, and on the other hand, recording of the center of the femoral head is not biased by the surgeon’s action. We use the following rules: give external rotation when the FMA is over 3 of valgus ([ to 93), give internal rotation when the FMA is over 3 of varus (\to 87), and no rotation when the FMA ranges between 87 and 93 except for some cases in which the ligamentous balance is not satisfactory. The aim of this paper was to assess the clinical and radiological results of 49 cases in which external rotation was given according to our rules. Methods: Between March 2003 and October 2007 we recorded prospectively 476 TKA 476 TKA (e-Motion FP prosthesis, B-Braun-Aesculap, Tuttlingen, Germany). In this series, 49 cases were given external rotation, accounting for 10.3% of the cases. The series included 35 female and 14 male patients aged from 38 to 87 years (mean age : 74.57±9.07 years). We operated on 41 osteoarthritis cases, 4 posttraumatic arthritis, 2 severe chondrocalcinosis, 1 rheumatoid arthritis, and 1 Paget disease. The patient morphotype was a varus in 28 cases, a valgus in 17, and well aligned in 4 cases. The mean radiological HKA angle was 178.38 ± 8.41 (163195) and the mean FMA measured on the AP long leg X-Ray was 94.12 ± 2.84 (91-102). The patella was centered in 26 cases and subluxed in 23 (46.9%). The mean overall preoperative IKS score was of 89.34 ± 17 points (51-120). For operative procedure, we used the Orthopilot navigation device (B-Braun-Aesculap, Tuttlingen, Germany) in all the cases.

S293 The external rotation was given according to the navigated FMA which ranged between 92 and 102 (mean FMA : 95.28±2.64). The mean given external rotation was 3.41± 0.86 (2-7). Results: The mean follow up was 15.2 ± 13 months (2-58). For functional results, the mean Overall IKS score was 194 ± 7.15 points (170-200) and the mean flexion was 116 ± 13.5 (75-140). The radiological results showed a mean HKA angle of 179.14 ± 1.87 (175-184) with 94% of the cases between 177 and 183 (3 outliers). On skyline views the patella was well centered in 43 cases (88%) and slightly subluxed in 6 cases without any disability. Conclusions: Computer-assisted recording of FMA seems to be an accurate method to deal with external rotation of the femoral implant. The surgeon can give external rotation according to the resection of the medial condyle. If one resects more on the distal medial condyle in case of FMA in valgus, it is logical to resect more on the posterior medial condyle and therefore to give external rotation. Our clinical and radiological results support this approach but not too much external rotation should be given, particularly when the FMA is above 95. In our series, the maximum external rotation we gave was 7 but we cut a gash on the anterior lateral cortex of the distal femur.

P19-628 Does femoral cementing influence perioperative blood loss? A prospective randomized study about 107 total knee arthroplasties HLS noetos Demey G.1, Servien E.2, Lustig S.3, Ait Si Selmi T.4, Neyret P.5 1 Centre Albert Trillat, Lyon-Caluire, France, 2Centre Albert Trillat, Hospices Civils de Lyon, Lyon-Caluire, France, 3Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 4Croix-Rousse Hospital Centre Livet, Lyon, France, 5Hopital Croix-Rousse - Centre Livet, Chirurgie Orthopedique, Lyon, France Objectives: Between 2004 and 2005, we conducted a prospective randomised study of 130 consecutive primary total knee arthroplasty (TKA) to assess the influence of femoral cementing on TKA results. In this report, we analysed a subset of these patients to compare the perioperative blood loss of those patients with a cemented femoral component, to those receiving a cementless femoral component with Hydroxyapatite coating. Methods: The TKA used was the HLS Noetos. A cemented tibial component with mobile insert and a patellar resurfacing arthroplasty were performed in all cases. All patients were preoperatively randomly assigned in either the cemented group (group 1) or uncemented group (group 2). We selected the 107 TKA performed by medial parapatellar approach. Group 1 consisted of 42 women and 12 men (n=54). Group 2 consisted of 37 women and 16 men (n=53). There were no significant differences between the groups concerning anthropometric or demographic data. The surgical procedures were performed by the same surgical team using a standardized technique. At the time of surgery, two suction drains were inserted inside the joint, and the tourniquet time (TQ) was recorded. The haemoglobin and haematocrit levels were recorded preoperatively and 5 days postoperatively for each patient. The calculated blood losses were evaluated as described by Mercuriali. The volumes of postoperative suction drainage and incidence of blood transfusion were recorded. Results: The mean TQ was 63.8 min for group 1 and 65.5 min for group 2 (p=0.5). No difference was recorded in the patients’ initial haemoglobin and haematocrit levels. Postoperatively, the haemoglobin level was 9.7 g/dl for both groups; the haematocrit level was 29.4% for group 1 and 29.9% for group 2 (p=0.4). The total measured blood loss amounted to 1758.9 ml for group 1 and 1759 ml for group 2 (p=0.9). The average post-operative drainage was 1077 ml for group 1 and 1181 ml for group 2 (p=0.3). Following TKA, 18 patients from group 1 and 17 patients from group 2 received a blood transfusion. Conclusions: In our study, femoral cementing influenced neither perioperative blood loss, nor the need for subsequent transfusion. Some previous authors have found a relation between using cement and decreasing blood loss. However, many of them analyzed the use of tibial and femoral cementing at the same time. Furthermore, the comparison with other

123

S294 studies is difficult because several different methods have been used to evaluate the amount of blood loss.

P19-630 The effect of pelvic movement on the accuracy of hip centre location acquired using an imageless navigation system Lustig S.1, Servien E.2, Demey G.3, Neyret P.4 1 Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 2 Centre Albert Trillat, Hospices Civils de Lyon, Lyon-Caluire, France, 3 Centre Albert Trillat, Lyon-Caluire, France, 4Hopital Croix-Rousse Centre Livet, Chirurgie Orthopedique, Lyon, France Objectives: The aim of this study was to assess the accuracy of an imageless navigation system (PLEOS Knee) in localising the hip centre and to evaluate the effect of pelvic movement on the accuracy of hip centre acquisition. Methods: A phantom leg was constructed to simulate the hip joint and upper femur. A 3-D measurement device was used to measure precisely the coordinates of hip centre. A limber link simulated hip motion as used when defining the hip centre during computer-assisted surgery. The data generated by the 3-D measurement device and the image-free navigation system was compared with increasing amounts of simulated pelvic movement (none, less than 5mm, 15mm and more than 20mm). One hundred measurements were undertaken at each movement level. Results: For moderate pelvic movement (5 mm) the mean error of the hip centre was less than 4 mm and ranged between 1.5mm and 3.9mm with a corresponding angular error of between 0.25 and 0.64. In conditions of excessive pelvic movement (15 mm) the mean error was 11.7 mm corresponding to an angular error of 1.9. For critical pelvic movement, the navigation system did not allow acquisition. Conclusions: This in vitro study showed accurate acquisition of the centre of hip with a modern surgical navigation system. These data suggest that, during TKA using this computer assisted navigation system, the accuracy of insertion will not be significantly affected by moderate pelvic movement during data acquisition.

P19-631 Imageless computer assisted system; acquisition accuracy and implications for knee arthroplasty Lustig S.1, Fleury C.2, Servien E.3, Demey G.4, Neyret P.5 1 Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 2 Tornier SA, Saint-Ismier, France, 3Centre Albert Trillat, Hospices Civils de Lyon, Lyon-Caluire, France, 4Centre Albert Trillat, Lyon - Caluire, France, 5Hopital Croix-Rousse - Centre Livet, Chirurgie Orthopedique, Lyon, France Objectives: The majority of the current computer assisted systems (CAS) for knee prosthetic surgery require the acquisition of points using a tracker detected by an infrared camera. Two types of measurements are then essential: angles and distances. The goal of this study was to evaluate the accuracy of the data obtained during computer assisted surgery using an in vitro protocol. Methods: Two models were developed to locate precisely both points (120 acquisitions) and distance measurements (144 acquisitions) and angles (170 acquisitions) with an image-free CAS using an infrared optical camera. For validation, a precise coordinate 3D measurement device was used to assess the accuracy of CAS acquisitions. Results: The points, distances and angles had a mean error respectively of 0.638 mm (0.244 mm to 0.931 mm), 0.355 mm (0.001mm to 1.338mm) and 0.39 (0.06 to 0.69). For all these acquisitions, the mean error was statistically less than 1mm or 1 (p\0.001). Conclusions: By using a 3D measurement system, it was possible to determine the accuracy of the data obtained with the navigation system. This was more precise than using radiological measures or CT Scans. The precision assessed at less than 1mm or 1 corresponds with the accuracy needed in knee arthroplasty and with the use of CAS as a measurement tool.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P19-683 Preoperative nutritional status and early prosthetic joint infection in total knee replacements Lozano L.1, Font L.2, Forga M.T.3, Rios J.4, Soriano A.5, Casanova L.6, Garcia S.7, Llobet F.6 1 Hospital Clinic de Barcelona, Knee Unit, Barcelona, Spain, 2Hospital Clinic de Barcelona. University of Barcelona, Knee Section. Orthopaedic Surgery Department, Barcelona, Spain, 3Hospital Clinic de Barcelona. University of Barcelona, Endocrinology and Nutritional Unit, Bareclona, Spain, 4Hospital Clinic de Barcelona. University of Barcelona, Statistics & Methodology Support Unit. Clinical Pharmacology, Barcelona, Spain, 5 Joint and Bone Infections Unit, Hospital Clinic, University of Barcelona, Barcelona, Spain, 6Hospital Clinic de Barcelona. University of Barcelona, Knee Unit. Orthopaedic Surgery Department, Barcelona, Spain, 7Hospital Clinic de Barcelona. University of Barcelona, Orthopaedic Surgery Department, Barcelona, Spain Objectives: Although the influence of preoperative nutritional status on short term outcome in arthroplasty is well known, its relationship with early prosthetic joint infection (EPJI) in total knee replacement remains unclear. Our aim was to assess the effect of preoperative nutritional status on patients who went on to present with EPJI following total knee replacement surgery. This assessment was based on preoperative blood tests and anthropometric measurements. Methods: A total of 213 patients undergoing total knee replacement between December 2007 and May 2008 were included in the study. Patients with rheumatoid arthritis were excluded. For each patient we preoperatively checked haemoglobin level, CRP, ESR, total lymphocyte and protein count, albumin and pre-albumin concentration and triglicerids, cholesterol and creatinine levels. Triceps skindfold and arm/muscle circumference were measured the day before surgery. The body mass index was calculated based on the information contained in the anaesthetic chart. We also collected information about co-morbidities such as Diabetes, High blood pressure, ASA grading, age and gender. Information about early infections, both superficial and deep, was collected. A descriptive statistical analysis and logistic regression models approach for independent risk factors were performed. Results: The mean age was 71.5 years. There were 162 female and 51 male. Eleven patients (5.16%) had early wound infection: 5 deep EPJI and 6 superficial. When we compared infected vs non-infected group neither co-morbidities nor preoperatively laboratory test were associated with a high early infection risk. However, there were statistically significant differences in anthropometric variables as Arm circumference, triceps skinfold and Fat area (FA). Neither differences where found in the cases of muscular area and BMI. Patients with acute infections had longer inhospital stay than patients without prosthetic joint infection. Conclusions: A low triceps skindfold and FA were associated with an increment of risk of EPJI after a knee replacement. Although the relationship between some laboratory test as pre-albumin and lymphocyte account and wound healing and postoperatively complications is well known, we didn’t find it with EPJI in our group.

P19-687 Torque measures of common therapies for the treatment of knee flexion contractures Uhl T.1, Jacobs C.2 1 University of Kentucky, Department of Rehabilitation Sciences, Lexington, KY, United States, 2ERMI, Inc., Atlanta, GA, United States Objectives: Flexion contractures following TKA have been associated with increased pain, reduced functional ability, and more rapid progression of contralateral joint degeneration. The effectiveness of a given treatment to reduce flexion contractures is a function of the applied torque, as well as the duration and frequency of the treatment. Little is known about the torque being applied by the various therapies, and the purpose of this study was to measure the torque applied by physical therapists (PTs) during manual therapy, home therapy exercises, and 3 types of mechanical therapy devices (dynamic splint, static progressive stretch (SPS) device, and patient-actuated serial stretch (PASS) device).

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: An instrumented test leg recorded peak torque as 14 licensed PTs each performed 5, 10-second repetitions of knee extension mobilization.(Figure 1) Peak torque was also recorded after a single PT applied the 3 mechanical therapy devices to the test leg. The same PT applied the following home exercises: heel props, prone hangs, prone hangs with a shoe, and prone hangs with a 2 lb weight. An ANOVA was used to determine if differences existed between the 8 therapies.

Fig. 1 Instrumented test leg Results: The eight therapies were significantly different (p = 0.002), but pairwise comparisons could not be made due to a lack of inter-trial variability with the mechanical and home therapies. Mean peak torque applied by PTs was 62.2 ± 17.1 N-m, and the PTs were very consistent in their torque application as evidenced by the low coefficient of variation (mean = 5.6%, range = 1.5% to 13.5%). Peak torque of the home exercises were as follows: heel prop = 4.6 N-m, barefoot prone hang = 6.2 N-m, prone hang with shoe = 10.1 N-m, prone hang with shoe and weight = 12.4 N-m. Peak torque applied by the mechanical therapy devices were: dynamic splint = 7.2 N-m, SPS brace = 10.4 N-m, and PASS device = 53.0 N-m.(Figure 2)

Fig. 2. Results of torque testing Conclusions: The torque applied by PTs during mobilization and a PASS mechanical therapy device were greater than that applied during home therapy exercises and other forms of mechanical therapy. In order to maximize effectiveness, torque must be considered when developing appropriate treatment protocols for postoperative flexion contractures.

P19-703 Evaluation of the residual quadriceps muscle dysfunction after TKR using surface elektromyography Hajduk G.1, Kusz D.1, Wojciechowski P.1, Kopec´ K.1, Sobota G.2, Nowak K.2

S295 1

Medical University of Silesia, Department of Orthopaedics and Traumatolgy, Katowice, Poland, 2Academy of Physical Education in Silesia, Institution of Biomechanics, Katowice, Poland Objectives: Total knee replacement is currently the most accepted method of treatment for end-stage knee osteoarthritis. Despite a very good longterm results, we have notice that patient recovery and return to full physical activity may be prolonged up to a few months.Even in most clinically successful cases, many patients cannot achieve normal knee function over time. The aim of this study was to evaluate early outcomes of Total Knee Replacement performed via standard approach and to assess residual quadriceps muscle function abnormalities after TKR using surface electromyography. Methods: The study group included 20 patients who underwent TKR via standard approach. Clinical assessment was performed with the KSS and VAS scores before the surgery and then 6, 12 and 24 weeks postoperatively. Knee alignment was evaluated pre- and post surgery with the AP radiographs of the whole lower limb (including hip, knee and ankle) taken in the standing position. Electromyographic activity of the quadriceps muscle was measured with a 4-channel EMG MyoTrace400(TM) device, using an unidirectional linear dynamometre. During the study, patients were performing static tests as well as dynamic tests. Amplitude was standardized to the quadriceps MCV (maximal voluntary isometric contraction). EMG analysis was performed with the MyoResearch XP Master Edition software suite. Results: The follow-up time was 24 weeks. We found that over time the patients reported less pain (the VAS score decreased from 69 pts. before surgery to 35 pts. at 5-th post-op day and to 6 pts. 24 weeks after the surgery) and better knee function (the KSS score increased from 48 pts. preoperatively to 88 pts. at 24 weeks after the surgery). These results were statistically significant. Limb axis and implant positioning was correct in all of the patients. The analysis of the surface EMG performed with patients walking on a treadmill revealed incorrect activation profiles of all muscles comprising the knee extensor mechanism and prolonged phase of single stance. Conclusions: In conclusion, it seems that the disturbed EMG activity of the quadriceps muscle does not have a negative impact on the generally good early outcomes of TKR in the treatment of knee osteoarthritis.

P19-715 A no release extension-flexion gap balancing technique for TKA with the patella in place balancer Ghijselings I.A.M.1 1 AZ Alma Eeklo, Orthopedie, Eeklo, Belgium Objectives: Total knee arthroplasty (TKA) remains one of the most successful procedures in orthopaedic surgery. Complications certainly exist and are often related to failure of knee ligament balance. 1) Balancing of the joint gap in extension and flexion is a prerequisite for success of a TKA. 2) The joint gap is influenced by patellar position. 3) Extensor mechanism tightness may be important in achieving the optimal joint gap balance during total knee arthroplasty. 4) Posterior femoral condyle referencing did not provide proper femoral component rotation. 5) Rotation of the femoral component using a gap balancing technique resulted in better coronal stability suggesting improvement of functional performance and reduction of polyethylene wear. 6) Posterior cruciate ligament balancing is an important surgical aim for high-flexion TKA. 7) Significantly worse knee and function scores were noted in the group in which the PCL was released. 8) High wear rate and patellofemoral syndrome could be avoided by correct PCL and soft tissue balancing. 9) Preserving a maximum of posterior condylar offset is important for high-flexion. Methods: We have developed a new balancer (PIPB) for TKA designed to assist soft-tissue balancing avoiding releases and involving a reduced patellofemoral joint. It is tibial axis referencing, CR retaining method with

123

S296 a balanced tension flexion gap at 90 degrees knee flexion with the patella in its anatomical position. The femoral cutting block is put in place in order to do a posterior femoral bone cut, this in parallel with the tibial bone cut as well in a medio-lateral as in a antero-posterior direction. A maximum of posterior condylar offset and a well defined height of the flexion-extension gap is measured simultaneously before the femoral cuts are done.The femoral component can be placed in 0 or 3 of flexion and 4 or 6 of valgus. No ligament releases are required. Results: A 1 year follow up of 123 TKA procedures between February 2007 and February 2008 are presented. Conclusions: The technique of PCL retention is difficult because the normal configuration and tension needs to be reproduced with ligament tensioners. The PIPB is a reliable instrument to achieve the goal off ligament balancing without releases.

P19-721 Navigated revision total knee replacement. A comparison study with conventional revision Jenny J.-Y.1, Diesinger Y.2, Ehlinger M.3 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France, 2Hoˆpitaux Universitaires de Strasbourg, Centre de Chirurgie Orthope´dique et de la Main, Illkirch, France, 3University Hospital Strasbourg, Strasbourg, France Objectives: Navigation system might help improving the quality of implantation of a revision total knee replacement (TKR). Methods: 30 cases of revision TKR were operated on with an image-free system, and matched to 30 cases of conventional revision TKR. Quality of implantation was analyzed in both groups on post-operative long-leg X-rays. Following items were recorded: coronal femoro-tibial angle, coronal and sagittal orientation of femoral and tibial implants. The rate of globally satisfactory implanted prostheses and the rate of prostheses implanted within the desired range for each criterion were recorded in both groups and compared with a Chi2 test and an ANOVA test at a 5% level of significance. Results: We observed a significant improvement of all radiological items by navigated cases. Limb alignment was restored in 88% of the navigated cases and 73% of the conventional cases. Similar differences were observed for the coronal and sagittal orientation of the femoral and tibial implants. Overall, 78% of the implants were oriented satisfactorily for the four criteria for navigated cases, and only 58% for conventional cases. Conclusions: The navigation system enables reaching the implantation goals for implant position in the large majority of cases, with a rate similar to that obtained for primary TKA. The rate of optimally implanted prosthesis was significantly higher with navigation than with conventional technique. The navigation system is a useful aid for these often difficult operations, where the visual information is often misleading.

P19-723 A radiological and navigated study of the anatomical and mechanical femur axes in the lateral plane Jenny J.-Y.1, Barbe B.2 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France, 2University Hospital Strasbourg, CCOM, Illkirch, France Objectives: Data about sagittal orientation of the femoral component of a total knee replacement (TKR) are scarce, mainly because the definition of the femur axes on the lateral plane is not fully validated. Methods: We analyzed 60 patients scheduled for TKR. Following axes were drawn on pre-operative long leg lateral X-rays: distal anterior cortex axis, anatomic diaphyseal axis, and three different mechanical axes from the center of the femoral head: #1 to the lowest point of the Blumensaat line, #2 to the midportion of the femoral condyles, #3 to the junction between the anterior two-third and the posterior third of the femoral condyles. The cortical axis was considered as the reference, and the angles between this reference and the other axes were recorded (more flexion was considered positive).

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results: The mean orientation of the diaphyseal axis and the reference was ?0.6±3 (range, -1 to ?3). The mean orientation of the mechanical axis 1 was -0.8±2.1 (range, -5 to ?4). The mean orientation of the mechanical axis 2 was -0.6±2.1 (range, -5 to ?4). The mean orientation of the mechanical axis 3 was ?0.8±2.1 (range, -3 to ?5). Conclusions: There were few differences between the orientation of the different axes of the femur on the lateral view. The cortical axis has the lowest variance and may be the more reliable to document the femoral orientation on the lateral view. However this axis does not take into account the anteversion of the femoral neck.

P19-727 Axial knee rotation in mobile- and fixed bearing total knee arthroplasty during complex tasks Zu¨rcher A.W.1, van Hutten K.2, Harlaar J.2, Terwee C.B.3, Albers G.H.R.4, Poll R.G.5 1 Diakonessen Hospital, Orthopedic Surgery, Utrecht, Netherlands, 2 Vrije Universiteit Medical Center, Rehabilitation Medicine, Amsterdam, Netherlands, 3Vrije Universiteit Medical Center, EMGO Institute, Amsterdam, Netherlands, 4Tergooi Hospitals, Orthopedic Surgery, Hilversum, Netherlands, 5Slotervaart Hospital, Orthopedic Surgery, Amsterdam, Netherlands Objectives: The benefit of mobile-bearings in total knee arthroplasty has not yet been established. Clinical mid-term studies show comparable results to fixed-bearings, and long-term comparison studies are still not available. Kinematic studies comparing both designs are technically demanding and focused on relatively simple tasks. We present for the first time a drastic kinematic difference in favour of the mobile-bearing design, during the performance of more complex tasks. Methods: A group of 10 patients after fixed-bearing (FB) and a group of 19 patients after mobile-bearing (MB) total knee arthroplasty were compared, five year postoperatively. Both groups had comparable patient demographics, Knee Society Scores and patient derived questionnaires (WOMAC, SF-36). All were submitted to noninvasive optoelectronic motion analysis, using a femoral epicondylar frame for thigh-marker placement. They had to stand up from a chair and immediately start to walk, with and without crossover and sidestep turns. The range of axial rotation (sum of maximum tibia internal and external rotation) was compared between both groups. Results: During normal walking, the FB group had a range of rotation of 10 degrees, while the MB group had 12 degrees (p\ 0.01). This range increased to 12 degrees in the FB group and 19 degrees in the MB group, while performing a chair rise (p\ 0.01). There was still a difference in favour of the MB design, while performing chair rises with forced rotation: 15 degrees in the FB group and 20 degrees in the MB group (p=0.06). Conclusions: The amount of axial mobility in MB total knee arthroplasty is higher than in a FB design, and, is comparable to our earlier findings in healthy subjects. This supports the rational of the design, aiming at a reduction of loosening stresses by increased prostheses conformity and mobility.

P19-758 Hybrid cementation of tibial component in unidirectional rotating platform TKA is at risk of early loosening? Prospective study Rossi R.1, Bruzzone M.2, Bonasia D.E.3, Dettoni F.2, Marmotti A.4, Andrea F.5, Castoldi F.4 1 University of Torino, Mauriziano Umberto I, Torino, Italy, 2University of Torino - Mauriziano ‘Umberto I’ Hospital, Department of Orthopaedics and Traumatology, Torino, Italy, 3University of Torino, Torino, Italy, 4 Mauriziano Hospital, University of Torino, Department of Orthopaedics and Traumatology, Torino, Italy, 5University of Torino - Mauriziano ‘Umberto I’ Hospital, Torino, Italy Objectives: Controversy still exists regarding which cementation technique (full or surface) of the tibial component is most durable. Full cementation has shown excellent long-term outcomes and

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 surfacecementation using fixed-bearing designs provided excellent midterm results. Concerns have been expressed about the possible effect of rotary forces to the tibial rotating platform that may produce micromotion, when the tibial stem remains cementless with the risk of early loosening. Methods: The purpose of this prospective study was to evaluate 70 consecutive unidirectional rotating platform, posterior stabilized, total knee arthroplasties, using a surface tibial cementation and a press fit technique for the tibial stem. The Knee Society Score and the Knee Society Roentgenographic evaluation were used as scoring systems. The tibial plateau was divided in four zones in anteroposterior view and the cement penetration was evaluated in each zone on plain radiographs. Results: The mean follow-up was 43 months (range 37-61, SD 14). There were 46 women and 24 men with a mean age of 72 years (range 55-85, SD 8.4). The Knee Score averaged 52 (SD 15) preoperatively and 91 (SD 8) at the last visit follow-up. The mean Function score was 41 (SD 24) preoperatively and 86 (SD 17) at the last follow-up. The mean cement penetration under the tibial surface on the antero-posterior view was respectively 2.28 mm. (SD 0.7) in zone 1, 3.18 mm. (SD 0.6) in zone 2, 2.93 mm. (SD 0.7) in zone 3 and 2.62 mm. (SD 0.8) in zone 4. The overall average penetration in the whole tibial plateau was 2.75 mm. No early radiolucency lines were observed at the tibial plateau level. In five asymptomatic patients (7%) radiolucency was noted around the stem of the tibial component. Conclusions: The surface cementation technique of the tibial component, with adequate cement mantle, provided excellent results at short to midterm follow-up in unidirectional rotating platform total knee arthroplasties.

P19-771 Revision total knee arthroplasty using temporal spacers Zazirnyi I.1, Yevsyeyenko V.1 1 Hospital ‘Feofania’, Centre of Orthopaedia, Trauma and Sport Medicine, Kiev, Ukraine Objectives: The increase of the primary total knee arthroplasty (TKA) causes the increase of revision surgery. Aim: To analyze the revision TKA using temporal spacers. Methods: Over the period of 2000 - 20007, 14 revision surgeries of the earlier located implants were performed. In every case, the implants were removed due to the infection. After the primary implant removal and necrotomy, the created space between epiphysis of femoral tibia bones needed to be filled. For four patients (the surgery was performed in the period of 6-12 months after the primary TKA) we used the bone cement as a spacer with antibiotic (cefuroxime 1.5 mg). In the postoperational period the knee was fixed with a splint. For 10 patients (the surgery was performed in the period of 3-4 months after the primary TKA) we used the so called ‘‘functional’’ spacer - bone cement with antibiotic together with femoral component, articular surface of tibia component. The postoperative treatment regime was the same as for the patients after the primary TKA. Everyone was recommended the ambulation on crutches with no loading on the extremity over a period of 6 weeks after the surgery. The revision surgery was held within 3 months after the condition normalization (white blood cells, C-reactive protein). During the operation, the previous spacer was removed and instead the new implant was introduced. We used model LSSK ‘‘Zimmer’’ (4 patients), models for the primary TKA with posterior stabilization and stems to tibia plateau and/or femoral component (10 patients). Results: The treatment effect was evaluated over the period of 2 - 8 years using 100 points Knee Rating Scale. Thus, for 2 patients out of 4 (with LSSK ‘‘Zimmer’’ implants) the functioning was rated as good (70- 84 points), 2 patients were assessed as satisfactory (60-69 points). For 2 patients (out of ten who had the model for primary TKA with stems) the knee functioning was rated as excellent (more than 85 points), for 8 as good (70-84 points). Conclusions: 1. The earlier performed revision surgery with the infection of knee detected allows reducing the volume of the removed bone tissue. 2. The use of ‘‘functional’’ spacer after the primary implant removal (due to infection), allows to preserve the function of knee joint better.

S297 P19-777 Ligament referencing in total knee arthroplasty and its effect on patella tracking in varus knee Podesek S.1, Bogunia I.1, Horeglad S.1 1 Hospital for Rehabilitation and Orthopaedic Surgery ‘‘Go´rka’’, Busko Zdro´j, Poland Objectives: There are four methods to determine the femoral extrarotation: 1) Transepicondylar axis, 2) Whiteside’s line, 3) Posterior condylar line, 4) Extrarotation after ligament balancing. Many knee systems are ‘‘bone referenced’’ and use 3 degree external rotation for the femoral component. When ligament referenced systems are used (lamine spreader or tensioner in flex-ion), the amount of rotation will vary, but is assumed to be within a safe range. Malrotation of the femoral component can cause patella maltracking (patellar tilting, subluxation, early luxa-tion, late patellar prosthesis failure) and anterior knee pain. During TKA we are going to create stable knee with rectangular, symmetrical tibio-femoral space both in extension and flexion and restore level of the joint line. Methods: Our group consists of 207 knees in 187 patients operated on between march 2006 and march 2009. In 20 patients there was an operation on both knees. We reviewed our patients 6 weeks, 3 and six months postoperatively. In the clinical examination we especially focused on patello-femoral joint (PFJ) (pain during sitting, kneeling, stairs climbing and presence of pain while moving the patella medially and laterally). Each patient had a radiological control in anterior-posterior, lateral and Merchant/Laurin view. On axial view we measured lateral patellofemoral angle according to Laurin technique, congruence angle and lateral or medial translation. None of our patients had lateral release and patella resurfacing during prime proce-dure. Results: The average lateral patellofemoral angle was 12,3 degree (ranged 1 to 21 degrees). There were 60,3% knees within a normal value (16 degrees ±3,3 degrees). The average congruence angle was 6,7 degrees (ranged -38 to 73 degrees) with 72,5% knees within a normal value (-24 to ?8 degrees). The average value of lateral/medial translation was 0,95 mm (range -4 to 12 mm) and 65,9% of our patients had not had any displacement (neutral position). 5 patients presented with an anterior knee pain (2,4%) which passed after resurfacing of the patella. One of the patients was with a history of recurrent patella dislocations, one with advanced arthritic changes in the lateral part of the patello-femoral joint like in excessive lateral pressure syndrome, one with congruent PFJ and one with mildly incongruent PFJ. Conclusions: 1. In majority of TKA using soft tissue balancing technique as a reference point in bone cuts, we could restore congruent patellofemoral joint. 2. There is no statistical correlation between incongruent patello-femoral joint and anterior knee pain. 3. Arthrosis of the lateral part of the patello-femoral joint like in the excessive lateral pressure syn-drome and history of the patellar dislocations are risk factors for incidence of the anterior knee pain. 4. Good alignment of the patello-femoral joint does not preclude the incidence of the anterior knee pain. 4. The occurrence of the anterior knee pain after tibio-femoral arthroplasty is 1 over 40 patients . Resurfacing of the patella in risk patients can diminish the number of patients with anterior knee pain. 5. There was a positive correlation between congruence angle and lateral translation of the patella.

P19-793 Flexion instability after primary total knee arthroplasty: does revision lead to clinical and roentgenographic improvement. A prospective analysis Van Ochten H.1, Schimmel J.2, Van Tienen T.1, van Hellemondt G.1, Defoort K.1, Wymenga A.1

123

S298 1 Sint Maartenskliniek, Departement for Orthopaedic Surgery, Nijmegen, Netherlands, 2Sint Maartenskliniek, Research, Development and Education, Nijmegen, Netherlands Objectives: A recent study by Romero et al. established the existence of midflexion laxity after total knee arthroplasty (TKA). Goal of our study was to justify flexion instability after TKA as a reason for revision. We hypothesized that the revision procedure would lead to a significantly more stable knee joint in 70 degrees of flexion, better functional and clinical results, less pain and good patient satisfaction. Methods: We report a one year follow-up prospective analysis of results of revision TKA in a consecutive series of patients with proven midflexion instability of[10 on varus/valgus stress flexion fluoroscopy after primary TKA. Cases treated with a hinge prosthesis were left out this group. Preoperatively, three months and one year after the revision procedure different questionnaires were completed by physician and patient, i.e. Knee Society Score (KSS) clinical and functional scores, a 100-mm visual analogue scale (VAS) for pain and patient satisfaction. Stability of the joint was evaluated with the use of varus/valgus stress fluoroscopy in 70 degrees of flexion. Student’s paired t-tests were used to compare pre- and postoperative values. A total of 16 patients were included between 2004 and 2008, fifteen with [10 lateral laxity and one with a medial laxity of[10. Results: The mean age of the patient group was 64.6 ± 9.4 years. In one patient the insert was exchanged, in one patient the femoral component was revised and in fourteen patients the prosthesis was fully revised. One year after surgery the mean lateral laxity of the injured knee improved significantly on flexion stress radiographs to 6.7 ± 2.88 (p\0.05). Medial laxity did also improve, however not significantly (p=0.08). The KSS clinical and functional scores increased significantly one year postoperative (clinical from 49.2 ± 18.8 to 76.9 ± 16.4 and functional from 39.7 ± 17.7 to 62.5 ± 27.1, respectively). The patients did not report a significant decreased VAS pain score (62.5 ± 17.9 to 48.6 ± 27.0 after one year, p=0.25), however they were fairly satisfied (59.6 ± 20.1). Three patients received a thicker insert due to rest-instability, one patient underwent manipulation under anesthesia and one patient presented with an infrapatellar neurinoma. Conclusions: Revision of a primary TKA with a laxity of more than 10 degrees midflexion instability resulted in a more stable knee joint. Despite significantly improved clinical and functional results after one year, pain did not decrease and patients were only fairly satisfied. Therefore, we can conclude that revision arthroplasty may be considered as a valuable and effective treatment for mediolateral midflexion instability after primary TKA but did not improve subjective patient-based outcomes.

P19-836 Computer-assisted uni knee arthroplasty for genu varum deformity. Results of axial correction in a case-control study of 40 cases Plaweski S.1, Ayach A.1, Mercier N.1, Saragaglia D.1 1 Grenoble South Teaching Hospital, Department of Orthopaedic Surgery and Sport Traumatology, E´chirolles, France Objectives: The results of uni knee arthroplasties (UKA) are better when the indications are well chosen and when the procedure respects an under correction from 1 to 3. This is difficult to achieve when performing UKA using conventional ancillaries. Too much under or overcorrection leads to failure or loosening, increasing the rate of revisions. The aim of this study was to compare 2 series of 20 patients operated on for UKA: group A was computer-assisted and conventional surgery was used for group B. Methods: The series included 40 patients, 27 male and 13 female patients, aged from 55 to 83 years (mean age: 67.7 years). The two groups were comparable regarding age (67 ± 8.09 years, 55 to 82, for group A, and 68.4 ± 5.73, 62 to 83, for group B), gender (7 female and 13 male patients in group A, and 6 female and 14 male patients in group B), osteoarthritis stage according to Ahlba¨ck modified criteria (6 stage II, 10 stage III, 4 stage IV for group A, and 4 stage II, 9 stage III, 7 stage IV for group B), and HKA angle (174.55 ± 3.45, 167 to 180, for group A, and 172.65 ± 2.99, 165 to 177, for group B). In the operative procedure, for both groups we used the KAPS Uni (X’Nov Company, Belfort, France), a cemented mobile bearing prosthesis. The conventional technique was

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 based on an extra articular cutting guide for the tibia, the femur cut depending on the tibial one (no extra or intra medullary cutting guide). We used the Surgetics Station (Praxim, Grenoble, France) for the navigated group. We navigated only the tibial cut (varus, height of the resection, tibial slope) and we checked the HKA angle and the reducibility of the deformity on the computer to avoid medial instability, over correction or too much under correction. For both groups, the main goal of surgery was to obtain an HKA angle of 178 ± 1, and the secondary goals a tibial posterior slope of 3 ± 2 and a varus mechanical tibial axis (MTA) of 2±2. All the patients underwent a preoperative goniometry according to Ramadier’s protocol and the same protocol 3 months postoperatively. The navigated group was randomly identified from 100 non-navigated cases. Results: The HKA angle was 179. 3 ±1.05 (176 to 181) for group A, and 177.4 ± 1.93 (173 to 180) for group B (p=0,06). The preoperative goal (178±1) was reached in 85% of the cases for group A and in 60% of the cases for group B (p=0.01 - highly significant). The MTA difference was not significant (87± 2.70 for group A, and 84.8± 3,20 for group B - p=0.74). The difference was highly significant for the posterior tibial slope (3±1.33 - 0 to 5- for group A, and 1.05± 0,94 -0 to 3- for group B - p= 0.0001). Finally the IKS score difference was not significant (194.5 ?7.52 points -180 to 200- for group A, and 194.75± 5.49 points - 180 to 200 - for group B). Conclusions: Computer-assisted UKA allows reaching the preoperative goal more easily than conventional technique as published in other series. Nevertheless, after short-term follow-up, there is no statistical difference regarding the IKS Score. We need more follow-up to show that this new technology will improve the UKA survival rate.

P19-859 Comparison of outcomes of medial parapatellar vastus split versus lateral parapatellar subvastus approach with tibial tubercle osteotomy for primary total knee arthroplasty Hirschmann M.T.1, Hoffmann M.2, Forke L.3, Krause R.4, Friederich N.F.1 1 Kantonsspital Bruderholz, Department of Orthopaedic Surgery and Traumatology, Bruderholz, Switzerland, 2Kantonsspital Luzern, Orthopaedic Surgery and Traumatology, Luzern, Switzerland, 3Marienstift Arnstadt, Department of Orthopaedic Surgery, Arnstadt, Germany, 4 Oberlinklinik, Potsdam, Germany Objectives: The purpose of this prospective consecutive multicenter study was to investigate whether the type of surgical approach (medial parapatellar or lateral parapatellar with tibial tubercle osteotomy) influences the early clinical and radiological outcomes of primary total knee arthroplasty (TKA). Methods: Ligament balancing primary TKA with a rotating platform was performed in 133 patients with 143 knees (m:w=1:1.6; mean age 69±8 years). The TKA was done by a lateral parapatellar subvastus approach with stepcut osteotomy of the tibial tubercle (53%; n=76, group A) or medial parapatellar vastus split approach (47%; n=67, group B). The outcome was assessed at 1 and 2 years postoperatively by the American Knee Society score (KSS) and the knee society total knee arthroplasty roentgenographic evaluation and scoring system (TKA-RESS) including assessment of radiolucency, implant position and mechanical alignment. The patient‘s pain level and satisfaction was noted by a visual analogue scale (VAS). Data were analyzed using Stata version 8 with a level of significance of p\ 0.05. Results: Although having a lower degree of preoperative flexion (112±15 versus 115±15) patients in group A showed a significantly (p=0.027) higher degree of flexion (118±10) at their last follow-up than patients in group B (114±10). Patients in group A showed a significantly better mean VAS pain (p=0.0001) and satisfaction (p=0.0058) at 2 years follow-up. The pain free walking distance was significantly (p=0.036) longer for group A than group B. Patients treated with a lateral approach were significantly more stable in terms of valgus stress (p=0.049). The Knee society score was significantly (p=0.0009) higher at two years follow up in group A compared to group B. The postoperative mechanical alignment and positioning of the prosthesis were not significantly different. Patients in group B presented with significantly (p=0.0017) more tibial radiolucencies at their last follow-up than patients in group A. There

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 was no prosthesis related revision in either group. The revision rate in group A (4%) was higher than in group B (1.5%), which was mainly due to two cases of traumatic secondary displacement of the tibial tubercle and need for refixation. Conclusions: The lateral parapatellar approach with tibial tubercle osteotomy led to better functional results and less pain two years after primary TKA. It is however not clear if the improved outcome can outweigh the higher risk of early complications and revisions. Long-term studies are necessary to show whether there is any difference in prosthesis longevity between both types of approach.

P19-860 Post-operative oxygen saturations in computer navigated total knee joint replacements Chambers M.1, Rooney B.2, Leach W.2 1 Western Infimary, Orthopaedics, Glasgow, United Kingdom, 2 Gartnavel General Hospital, Orthopaedics, Glasgow, United Kingdom Objectives: The aim of this work was to compare the oxygen saturations in patients in the early period following total knee joint replacement surgery performed using either computer navigation or conventional intramedullary mechanical jigs. Methods: For a four month period twelve consecutive patients who had computer navigated total knee joint replacements were prospectively reviewed. A comparison group from the same period was made of twenty patients who had knee replacements performed using conventional jigs. Non-invasive oxygen saturations were measured and recorded as a percentage. Preoperative oxygen saturations measured at the assessment clinic were used a baseline. For the duration of the patients postoperative hospital stay oxygen saturations were measured five times a day as well as their oxygen requirements. Results: We found that the patients in the computer navigation group on average reached oxygen saturation levels on air equal to those measured in the assessment earlier than the intramedullary jig group. There was also a lower need for oxygen in the computer navigated group during this early post operative period. Conclusions: Previous studies using transcranial Doppler and transoesophageal echocardiograms have shown a reduction of systemic emboli with computer navigated total knee joint replacements. Using oxygen saturation monitoring we have shown there may be a considerable clinical advantage using computer navigated surgery over conventional intramedullary rod jigs in knee replacement surgery.

P19-866 Development of a standardized algorithm for evaluating patients after total knee arthroplasty using combined single photon emission tomography and conventional computerized tomography (SPECT/CT) Hirschmann M.T.1, Iranpour F.2, Konala P.2, Rasch H.3, Cobb J.2, Friederich N.F.1 1 Kantonsspital Bruderholz, Department of Orthopaedic Surgery and Traumatology, Bruderholz, Switzerland, 2Imperial College London, Musculoskeletal Surgery Department, London, United Kingdom, 3 Kantonsspital Bruderholz, Institute for Radiology and Nuclear Medicine, Bruderholz, Switzerland Objectives: Hybrid SPECT/CT has now been introduced in clinical practice for nearly a decade. Numerous series and case studies have been reported on the use of SPECT/CT in a variety of clinical scenarios such as oncology, cardiology, endocrinology and infectious diseases. The diagnostic possibilities, as well as the clinical value of simultaneously coregistered anatomic and functional data are being increasingly recognized. Along with the introduction of hybrid SPECT/CT systems, which are able to view the images separately or merged, the diagnostic accuracy as well as sensitivity and specificity has relevantly improved. SPECT/CT clearly displays not only the region of interest, but the specific anatomical location, which provides the surgeon or physician with information that is useful in determining appropriate clinical management. In patients with

S299 post arthroplasty pain it might directly affect the clinical management by confirming or excluding the need of further surgical treatment. Surprisingly, to date SPECT/CT has not become widely accepted or frequently used in orthopaedic patients, which is reflected by the paucity of available studies on SPECT/CT in orthopaedics. As SPECT/CT is relatively new to orthopaedic surgeons neither a standardized algorithm for evaluating patients with painful knees after primary total knee arthroplasty nor a localization scheme for tracer activity have been reported. Hence, it was the primary purpose of our study to develop a novel standardized SPECT/CT algorithm for evaluating patients with painful primary total knee arthroplasty and to evaluate its clinical applicability and usefulness as well as its inter- and intra-observer variation and reliability. Methods: A novel SPECT/CT localization scheme, which consists of 9 tibial, 9 femoral and 4 patellar regions on standardized transverse, coronal, and sagittal slices was introduced. It was assessed in 18 consecutive patients with painful knees after total knee arthroplasty. The localization and level of the tracer uptake on SPECT/CT were noted using a color coded 10 steps graded scale (0-100). The inter and intra-observer reliability were assessed. The femoral and tibial prosthetic component position was assessed in the CT images after 3D reconstruction and aligning them to standardized frames of reference. The average root mean square difference±standard deviations and ranges of these measured angles are presented along with the intraclass correlation coefficients for inter- and intraobserver reliability. Results: The localization scheme was useful and easily applicable in all 18 cases. The novel classification using the SPECT/CT for the femoral, the tibial and patellar region was reliable. The measurements of component position in SPECT/CT images were highly reliable and feasible in all cases with sufficient visibility of the landmarks. The mean intra-observer difference between the rotational alignment measurements of tibial and femoral components was less than 2 (2SD 1). The intra-observer variability for these measurements was less than 1 degree (2SD 1). Conclusions: The introduced algorithm using SPECT/CT in patients after total knee arthroplasty, which combines mechanical (assessment of 3D rotational alignment of the prosthesis in the inherent CT data) and metabolic data (SPECT/CT localization scheme), was highly reliable and useful. We propose its use in larger scaled clinical studies to investigate its clinical value.

P19-914 Secondary patellar resurfacing for anterior knee pain after primary total knee arthroplasty Mun˜oz-Mahamud E.1, Popescu D.1, Lozano L.1, Torner P.1, Nun˜ez M.2, Nun˜ez E.3, Sastre S.1, Vilchez F.1, Castillo F.1, Martinez-Pastor J.C.1, Segur J.M.1, Macule F.1 1 Hospital Clı´nic de Barcelona. Universitat de Barcelona, Orthopaedics and Trauma Surgery, Knee Unit, Barcelona, Spain, 2Hospital Clı´nic de Barcelona. Universitat de Barcelona, Functional Readaptation, Barcelona, Spain, 3Institut Catala` de la Salut, SAP Suport al Diagno`stic i al Tractament, Barcelona, Spain Objectives: The apparition of anterior knee pain is one of the most frequent complications after primary total knee arthroplasty (TKA). We performed a prospective study with the aim of evaluating the results of a secondary patellar resurfacing (SPR) for the treatment of this common complication. Methods: Between May 2004 and October 2007 we performed a total of 29 SPR with incapacitating anterior knee pain after a primary TKA. There were 26 women and 5 men with a mean age of 70.4 years old (range: 55-85). The mean time between the TKA and the SPR was 21 months (range: 8-38). Clinical results were evaluated using the anamnesis, physical exam and the KSS, WOMAC and femoropatellar scores. Complementary studies featured plain radiographs, Tc99 gammagraphy and computerised tomography in order to measure the rotation of the femoral component. The mean follow-up time was 17 months (range: 6-32). Results: A total of 17 patients (54.8%) referred a clinical improvement, whereas 12 did not notice any improvement and 2 referred being worse

123

S300 than before the SPR. On average, the KSS (knee / function) previous and after the surgery punctuated 72.1 / 71.0 and 82.2 / 77.9 respectively (p\0.001). The WOMAC (pain / stiffness / function) averaged 13.7 / 4.3 / 31.5 before de SPR and 7.3 / 2.4 / 16.4 after the surgery (p\0.001). The mean femoropatellar score changed from 37.6 to 67.1 (p\0.001). The patellar tilt changed from 10.88 to 10.78 (p=0.38) and the lateral patellar displacement from 4.9 mm to 2.4 mm (p=0.13). The study with Tc99 gammagraphy revealed an hypercaptation of the patella in 9 cases. As complications, there were 1 acute infection and 1 early patellar component loosening. According to the femoropatellar score, we found better results in those patients operated 18 months after the TKA. No correlation was found between the demographic, clinical or radiological parameters and the functional results. Conclusions: SPR for anterior knee pain after TKA remains a controversial issue. The SPR has achieved a clinical improvement in approximately the half of the patients with a low number of complications. It seems that the delayed timing of the SPR procedure may positively affect the final outcome.

P19-924 Femoral component rotation does not influence patellar tracking after total knee replacement with the balanced gap technique Heesterbeek P.1, Keijsers N.1, Wymenga A.2 1 Sint Maartenskliniek, Research, Development & Education, Nijmegen, Netherlands, 2Sint Maartenskliniek, Department for Orthopaedic Surgery, Nijmegen, Netherlands Objectives: With the balanced gap implantation technique, the rotation of the femoral component can vary freely on guidance of the ligaments. Although leading to a rectangular flexion gap, the free femoral rotation may lead to internal rotation of the femoral component. Besides preoperative patellar tracking, femoral component rotation is one of the factors influencing patellar tracking behaviour. Therefore, the goal of this study was to investigate whether femoral component rotation influenced patella position after primary total knee replacement with the balanced gap technique. Methods: In this prospective cohort study, a primary TKR was implanted in 49 patients using a balanced gap technique and a CT-free navigation system. Femoral component rotation was referenced from the posterior condyles and was calculated using the navigation data of the distal femur cut. All patients were followed up for two years and lateral patellar tilt and patellar displacement were scored on axial patella radiographs. A logistic regression model was built for both lateral patellar tilt (normal or tilted: [10 degrees) and patellar displacement (normal or subluxated: C4 mm). As independent factors rotation of the femoral component (continuous scale) and preoperative patellar tracking (dichotomous) or preoperative displacement (dichotomous) were chosen. Results: Mean femoral component rotation was 4.0 (SD 3.8) and ranged from 3.2 internal rotation to 12.2 external rotation. Only 4 of the 49 patients had a tilted patella ([10) at two years follow-up. Thirteen patients showed subluxation (C4mm) of whom 8 had mildly displaced patellae (4 or 5 mm). Of the four tilted patellae, three were also displaced. Neither femoral component rotation nor pre-operative patellar tracking alone could be related to patellar malposition. However, these factors together resulted into a statistically significant predictive model for both patellar tilt and patellar displacement. But, according to the model, the influence of femoral component rotation was low. Even with a 4 degrees internally rotated femoral component and a pre-operative patellar tracking problem the chance of postoperatively tilted or displaced patella is still low. Conclusions: In conclusion, although the balanced gap implantation technique resulted in a wide inter-patient femoral component rotation variability, this rotation could not significantly influence postoperative patellar maltracking, even with a pre-operatively existing patellar problem. Therefore, the balanced gap technique is considered safe for patella tracking behaviour.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P19-928 Patellar replacement vs non patellar replacement in total knee arthroplasty-A retrospective study Ribeiro da Silva M.1, Oliveira P.1, Rui P.1, Neves N.1, Rodrigues P.1, Simo˜es C.1, Ribeiro R.1, Tulha J.1, Trigo Cabral A.1 1 Hospital Sa˜o Joa˜o, Oporto, Portugal Objectives: Patellar replacement in total knee arthroplasty remains a controversial subject in orthopaedic surgery. In this study the authors present the different clinical results of a retrospective study between two groups of individuals that underwent total knee Arthroplasty with and without patellar replacement. Methods: Retrospective study of 46 patients who underwent total knee arthroplasty, 23 with patellar replacement and 23 without, randomized for age, sex, axial deviation, pre operative radiologic evaluation and post operative follow up time. The implants utilized during the study (PFC Sygma and Advance - Wright) allow its utilization either with or without patellar replacement. For each group was described Knee Society Score (Knee Score and Function) (KSS) pre and post operatively, Knee anterior pain and pain while climbing up stairs (VAS), Radiographic evaluation, patellar tilt, Insall-Salvatti ratio and TKARESS (The Knee Society Total Knee Arthroplasty Roentgenographic Evaluation and Scoring System), Statistical analysis with SPSS. Results: In this study 35 women and 11 men, mean age of 66 years (5379), were operated to 19 left knees and 27 right knees, and follow up for a mean of 39 months. There is a statistical significant improvement in favour of patellar replacement for KSS (score (p=0,008)) and function (p=0,006)), as it is for anterior knee pain (p\0,05), as it is shown by the Mann-Whitney test. There are no differences in the Insall-Salvatti ratio. Patelar Tilt in patellar replacement was 28 vs 48 without patellar replacement. In all patients TKARESS was \4. It was verified also a lesser number of revision procedures in the patients with patellar replacement (0 vs 5). Conclusions: Patients that underwent arthroplasty with patellar replacement present far better clinical results that than the ones that don’t. They also present less procedures of revision, at least in the time that went by during the study. The study doesn’t present enough time of follow up to analyse the complications that could be related with patellar replacement. The option for patellar replacement isn’t consensual, but this study presents some evidence favouring the replacement of the patella.

P19-935 Early results of the corin uniglide unicompartmental knee replacement from an independent centre Odumenya M.1, Dunn K.2, Spalding T.1, Thompson P.1 1 University Hospitals Coventry and Warwickshire NHS Trust, Trauma and Orthopaedic Surgery, Coventry, United Kingdom, 2Clinical Sciences Research Institute, Coventry, United Kingdom Objectives: The main objectives are: 1) To identify the early functional outcomes of the Uniglide at a minimum of 12 month 2) To identify the survivorship of the Uniglide prosthesis 3) To identify errors in component positioning 4) To identify any complications associated with the Uniglide prosthesis. Methods: A prospective consecutive case series of sixty-two Uniglide Unicompartmental Knee Replacements (UKRs) was carried out in fiftythree patients between January 2006 and March 2009. All operations were performed by or under direct supervision of the senior author (PT). Only patients with isolated unicompartmental tibiofemoral osteoarthritis were deemed suitable for this operation. Evidence of unicompartmental disease was defined on plain radiographic films. The data collected for this case series included primary outcome measures: 1) Pre-operative and post-operative (minimum 12 months) validated functional outcome scores: American Knee Society Score (AKSS), Oxford Knee Score (OKS), Western Ontario and McMaster Osteoarthritis Index (WOMAC). 2) Survivorship using revision as endpoint.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 The secondary outcome measures collected were X-ray Error Scores assessing component positioning and immediate, early and late complications. Results: No cases were lost to follow-up. Out of the 62 knees 2 were lateral and 60 were medial UKRs. The medial UKRs (51 patients; 18 men, 33 women) were included in the analysis. Five patients had bilateral Uniglide UKRs. The mean follow-up time was 19 months (12 to 36 months) and the mean age was 66 years (39 to 78 years). The functional outcomes and survivorship results are illustrated in Tables 1-3. Using paired sample ttests, the p-values for the differences in pre-operative to post-operative scores were p=0.000 for the difference between pre-op to 1yr for AKSS knee score, Oxford and Womac and p=0.001 for AKSS functional score. No other differences were found to be significant between 12 and 24 months and 24 and 36 months. All functional outcomes at 24 and 36 months showed significant improvement, p=0.001, compared to at the preoperative scores. The survivorship at one year was 95%. There were two revisions, one for aseptic loosening of the tibial component and the other for complete collapse of the posteromedial tibia resulting in aseptic loosening of the tibial component. The x-ray error score identified the most common errors made as the femoral flexion/extension angle[5 degrees of flexion or extension and anterior/posterior fit of the tibial component [2mm overhang.There were two intra-operative complications: a medial plateau fracture, which was fixed intraoperatively and an ACL partial avulsion. Neither patient had any residual problems. Early complications included one post-operative haematoma treated conservatively and one superficial wound infection successfully treated with oral antibiotics. The late complications were neuroma formation at the distal end of the scar, which was excised at 2.5 years with a good result and two revisions. Prosthesis type

Femur/tibia

Cemented

Uncemented

Hybrid

Mobile bearing

Fixed bearing

53/53

9/9

0/0

57

5

Functional outcome results Outcome

Pre-OP (SD) N=60 AKSS N=50

Post-OP 1 YR (SD) N=27 AKSS N=26

2 YRS (SD) N=18 AKSS N=15

3 YRS (SD) N=5

AKSS (knee rating)

56.5 (1.9)

83.4 (3.9)

88.4 (3.7)

85.6 (8.8)

AKSS (function)

49.4 (1.9)

72.1 (4.9)

74.3 (6.3)

77 (7.8)

OKS (12=best, 60=worst)

38.6 (1)

23.4 (2)

24.2 (2.2)

21.4 (4.4)

WOMAC

50.3 (2.3)

21.1 (4.1)

20.8 (4.3)

21.5 (11.7)

S301 P19-941 Cemented versus press-fit placed stems in revision total knee replacement—a randomized controlled trial with model-based RSA Heesterbeek P.1, Wymenga A.2, van Hellemondt G.2 1 Sint Maartenskliniek, Research, Development & Education, Nijmegen, Netherlands, 2Sint Maartenskliniek, Department for Orthopaedic Surgery, Nijmegen, Netherlands Objectives: The number of revision of total knee replacements (TKR) increases annually. Stability of the implants is important. So far, there is no consensus whether to cement the implant stems or to place them pressfit. This randomised-controlled trial was conducted to assess the primary stability of cemented versus press-fit placed stems. Methods: In this randomised controlled trial 32 patients will be included. All patients needed revision of their primary TKR with the Legion system (Smith & Nephew). Patients with large bony defects were excluded from the study (Anderson score type III or more), as were patients with infected TKR. Randomisation determined whether the stems of the implants were placed cemented or press-fit. Both cemented and press-fit stems were of the same length. Migration of the femoral and tibial implants was measured with model-based roentgen stereophotogrammetry (MB-RSA). Small tantalum balls (Æ 1 mm) were inserted in the femur and tibia (bone model) and scans were made of the implants (implant model). During the first 5 postoperative days the baseline RSA radiograph was obtained. Patients were followed up at 6 weeks, 3, 6, 12 and 24 months. Migration (translation and rotation) of the tibial and femoral implant (movement of implant model from bone model) in 3 dimensions was calculated at each follow-up moment compared to baseline measurement. Translation of 1 mm and rotation of 1 degree was considered clinically significant. At 6 weeks double RSA radiographs were obtained to assess the reproducibility of the measurement method (accuracy). Reproducibility was calculated by the Bland and Altman method. Results: So far, 25 patients received a revision TKR; 12 cemented and 13 press-fit placed stems. Reproducibility measurements at 6 weeks for the femur were obtained of 22 patients, for the tibia measurements of 19 patients were available. Reproducibility (95% prediction limit) of the MBRSA was excellent. Tibia translation could be calculated with at least 0.55 mm accuracy, tibia rotation with at least 0.79 degrees. For the femur, posterior-anterior translation was less precise (1.10 mm), as was anteriorposterior tilt (1.31). Six month follow-up was available of 18 patients. So far, no difference in migration was found between cemented and press-fit placed stems for both femoral and tibial components. Conclusions: Model-based RSA was feasible for assessing migration of revision TKR. The lower precision for posterior-anterior translation and anterior-posterior tilt of the femur could well be explained by the round surface of the implant in these projections; it is difficult for the software program to detect motion of a round implant. However, in the other directions the precision of the migration calculation is well below the clinical relevant value of 1mm/degree. So far, there was no difference in primary stability at 6 months between cemented and press-fit placed stems. However, this is a very preliminary conclusion, since not all patients had completed the follow-up.

Kaplan–Meier survivorship Proportion surviving this interval

Cumulative 95 % Confidence survival at end of interval (%)

O

60/60=1

100

2

40/42= 0.95

95

6.4

O

25/25=1

95

8.3

Interval (years)

At risk Censored At risk at end at start during of interval of interval interval

No. revised at end of interval

0-1

60

O

60

1-2

60

16

42

2-3

60

15

25

O

Conclusions: The Uniglide UKR provides a significant improvement in early knee functional outcome and satisfactory short-term survivorship with an acceptable complication rate.

P19-948 Is trans-patella tendon approach for lateral UKR associated with patella baja? Jacks D.A.1, Pandit H.1, Van Duren B.1, Fievez E.1, Weston-Simons S.1, Jackson W.1, Price A.1, Dodd C.1, Murray D.1 1 Nuffield Orthopaedic Centre, NDORMS, Oxford, United Kingdom Objectives: Patella baja has been noted to occur after TKR and has been associated with diminished functional outcomes. Patella tendon shortening of 10% or more is considered significant. Previous studies have not identified this association in medial unicompartmental knee replacements (UKR). However, no study has examined patellar tendon length after lateral UKR. Traumatic and/or ischaemic injury of the patellar tendon have been suggested as possible causes. The Oxford Domed Lateral UKR requires a

123

S302 vertical incision through the patellar tendon to facilitate the proper orientation of the proximal tibial saw cut. This may induce scarring or impair vascularity of the tendon and can cause shortening. The aim of the present study was to investigate the incidence of patella baja after domed lateral UKR and to compare the results to a matched cohort of medial UKR. Methods: We reviewed the radiographs of 50 consecutive patients undergoing domed lateral UKR with a minimum of one year follow-up and adequate serial x-rays including immediate post-operative and one year post-surgery radiographs. We also compared these with a matched cohort of medial UKR patients. The patella tendon length was measured on lateral radiographs using a graphical use interface in Matlab and the magnification was corrected by using the femoral component radius as the reference. Insall-Salvati ratio was also measured on all radiographs. Twenty sets of radiographs were assessed again (at two weeks’ interval) by one observer as well as by another observer to examine the repeatability and the reproducibility of the x-ray measurement technique. Results: The inter and intra-observer correlation was found to be excellent when assessing the patella tendon length using the two methods mentioned above. In the lateral UKR group, three cases demonstrated a change of more than 10% in the PTL at one year post-surgery as compared to the immediate post-operative x-rays. Out of these, two had patella baja (4%), and one had patella alta (2%). In the medial UKR group, no cases demonstrated a change in PTL of more than 10%. The results were similar with the Insall Salvati ratio. Chi square test did not demonstrate a significant difference between the two groups. Conclusions: This study demonstrates that despite a transpatellar approach to the proximal tibia performed during lateral domed UKR, patella baja is uncommon. UKRs do not significantly alter the joint line and therefore are unlikely to alter the extensor mechanism unlike TKRs.

P19-949 Posterior stabilised TKA: has the new generation improved the kinematic profile? van Duren B.H.1, Pandit H.1, Tilley S.2, Price M.3, Beard D.4, Gill H.S.1, Murray D.5, Thomas N.2 1 Nuffield Orthopaedic Centre, NDORMS, Oxford, United Kingdom, 2 North Hampshire Hospital, Basingstoke, United Kingdom, 3Basingstoke and North Hampshire NHS Foundation Trust, Trauma and Orthopaedics, Basingstoke, United Kingdom, 4Nuffeild Orthopaedic Centre, Oxford, United Kingdom, 5University of Oxford, Department of Orthopaedic Surgery, Oxford, United Kingdom Objectives: The cam-post mechanism of Posterior Stabilized Total Knee Arthroplasty (PS-TKA) should provide a constraint that limits anterior translation of the femur on the tibia in flexion and thereby ensure femoral roll-back. In a previous fluoroscopic study, we showed that the sagittal plane kinematics of a PCL substituting TKA (Scorpio PS) was suboptimal. The aim of the current study was to investigate the successor to this implant: the Triathlon TKA (Stryker, UK). Methods: Eleven patients with Triathlon PS-TKA underwent fluoroscopic assessment of the knee during a step-up exercise and deep knee bend. The data was analysed using a 3D model fitting technique. The minimum distance between cam and post as well as the femoro-tibial contact positions of the medial and lateral condyles were determined. The patella tendon angle (PTA) was calculated by measuring the angle subtended by patella tendon with the tibial axis and was plotted against knee flexion angle (KFA). The data was compared to kinematic data of a group of twelve patients with Scorpio PS TKA and also to twenty normal knees. Results: The PTA-KFA relationship of the Triathlon implant (although still abnormal) followed more closely the normal than the Scorpio implant. The cam/post engagement occurred in all the Triathlon knees, between 608 to 808 KFA. In the Scorpio group, cam/post did not engage in one patient and engaged between 708 to 1008 in the remaining. The Triathlon TKAs exhibited a medial pivot motion, compared to paradoxical anterior slide of both the condyles in the Scorpio group.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Conclusions: Our previous study of the Scorpio TKA kinematics concluded that femoral roll-back is inadequate and this was associated with paradoxical anterior slide. In contrast, the Triathlon TKA shows a kinematic profile closer to normal with better cam-post engagement and without paradoxical anterior slide with flexion.

P19-951 Bicruciate stabilised (BCS) journey knee total knee replacement: how effective are the added kinematic constraints? van Duren B.H.1, Pandit H.1, Tilley S.2, Fievez E.3, Price M.4, Gill H.S.1, Murray D.5, Thomas N.2 1 Nuffield Orthopaedic Centre, NDORMS, Oxford, United Kingdom, 2 North Hampshire Hospital, Basingstoke, United Kingdom, 3Nuffield Orthopaedic Centre, Oxford, United Kingdom, 4Basingstoke and North Hampshire NHS Foundation Trust, Trauma and Orthopaedics, Basingstoke, United Kingdom, 5University of Oxford, Department of Orthopaedic Surgery, Oxford, United Kingdom Objectives: Traditional TKR designs exhibit variable and often grossly abnormal kinematics with the femur subluxing posteriorly in extension and paradoxically sliding forward in flexion. Newer designs attempt to provide ‘‘guided motion’’ with the aim of reproducing normal knee kinematics. The Journey (Smith & Nephew) BCS TKR incorporates both an anterior and a posterior cam/post mechanism. This study assesses the in-vivo kinematics of Journey TKR and compares it with normal knee kinematics. Methods: Knee kinematics of ten patients with Journey-BCS TKR, all with excellent clinical outcome (average age: 65, range: 60-72 years) were analysed. Patients underwent fluoroscopic assessment of the knee during a step-up and deep knee bend exercise. Data was analyzed using a 3D model fitting technique. In addition, sagittal kinematics was assessed using patella tendon angle (PTA) and patella flexion angle (PFA). This data was compared to normal knee kinematics (n=20). Results: The anterior cam/post mechanism engaged between 20o to 30o knee flexion while the posterior cam/post mechanism engaged between 50o to 70o flexion. Medial tibio-femoral contact moved posteriorly during initial 20o knee flexion and afterwards remained stationary up to 80o flexion. The lateral tibio-femoral contact point showed a consistent posterior rollback from extension to 80o flexion (18mm). Both medial (6 mm) and lateral (7 mm) contact points moved further posteriorly during deep knee bend. Maximum knee flexion achieved was 130o. The PTA and PFA had similar trend to that of the normal knee. However, PTA was lower with BCS TKR (average 4o) and PFA was higher with BCS TKR (average 10o). Conclusions: Both cam-post mechanisms engaged and helped produce femoral rollback on lateral side with stationary medial contact from 20o to 80o flexion. However, lower PTA and higher PFA suggest failure to restore completely normal sagittal plane kinematics and this needs further investigation.

P19-957 The role and efficacy of medial unicompartmental knee replacement for partial thickness cartilage damage Gulati A.1, Pandit H.2, Gill H.S.2, Price A.3, Dodd C.4, Murray D.5 1 Nuffield Orthopaedic Centre, Oxford, United Kingdom, 2Nuffield Orthopaedic Centre, NDORMS, Oxford, United Kingdom, 3Oxford University, Nuffield Department of Orthopaedic Surgery, Oxford, United Kingdom, 4Nuffield Orthopaedic Centre, Department of Orthopaedic Surgery, Oxford, United Kingdom, 5University of Oxford, Department of Orthopaedic Surgery, Oxford, United Kingdom Objectives: Mobile bearing unicompartmental knee replacement (UKR) is an accepted treatment for patients with isolated medial unicompartmental knee osteoarthritis (OA) with a full thickness cartilage loss. The aim of this study was to determine if this recommendation was correct and if the procedure could be used for partial-thickness cartilage loss.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: 1053 Oxford medial UKRs were studied prospectively. The knees were divided into two groups; partial-thickness cartilage loss (PTCL) group and the full thickness-cartilage loss (FTCL) group. The primary outcome measure was the total Oxford Knee Score (OKS, 0 to 48) at the time of final follow up. The groups were also compared for the change in OKS (DOKS) and the proportion of patients that were considered to have benefited substantially from surgery (DOKS [5). Results: There were 21 knees in the PTCL group, with a mean age of 61 years (SD 13) and mean follow-up of 1.7 years (SD 1.0). There were 1032 knees in the FTCL group, with a mean age of 66 years (SD 10) and mean follow-up of 3.9 years (SD 2.5). The PTCL group had a significantly lower (p=0.03) total OKS (mean 35.8, SD 10.7) at final review than the FTCL group (mean 39.8, SD 8.5). There was no significant difference in preoperative or DOKS between the groups. However the proportion of patients that did not have substantial benefit from surgery (DOKS[5) was significant lower (p=0.04) in the PTCL group. Conclusions: The results confirm that a good outcome can routinely be achieved with UKR if there is bone-on-bone. If there is partial loss of cartilage the results are less predictable. Although some patients do well, a higher proportion do not benefit from the surgery. It may be that in a subgroup of these patients the pain did not initially arise from the partially damaged cartilage. We therefore recommend that the Oxford Knee should only be used for medial OA if there is bone on bone.

P19-987 Biological activity of different crosslinked polyethylenes in the murine knee joint Utzschneider S.1, Dedic M.2, Paulus A.1, Schroeder C.2, Sievers B.2, Sadoghi P.1, Gottschalk O.3, Jansson V.1 1 Ludwig-Maximilians-University, Department of Orthopaedics, Munich, Germany, 2Ludwig-Maximilians-University, Department of Orthopaedics, Laboratory of Biomechanics and Experimental Orthopaedics, Munich, Germany, 3Ludwig-Maximilians-University, Walter-Brendel-Centre for Experimental Medicine, Munich, Germany Objectives: Crosslinked polyethylenes (XPE) show reduced wear rates in knee arthroplasty. The biological activity of wear particles is an essential factor influencing aseptic loosening of prostheses. This study should clarify the biological activity of different crosslinked polyethylenes compared to ultra-high-molecular-weight-polyethylene (UHMWPE) particles in vivo. Methods: Wear particles of 3 XPE-inserts (1 sequential irradiated / annealed; 2 remelted) and 1 UHMWPE-insert were isolated from a knee joint simulator (20nm-nucleopore-filter;acid digestion method;ISO) and analysed by scanning electron microscopy (n=66000). After removing of endotoxin the particles were suspended in a phosphate buffered saline solution (0.1% vol/vol (particle volume/PBS volume)). Endotoxin levels were controlled using a standardised endotoxin detection method (Lonza) in all samples. 40 female Balb/c mice were randomly assigned to one of five treatment groups (according to the national guidelines of animal protection laws): XPE1 (95 kGy E-beam, remelted; n=8); XPE2 (65 kGy E-beam, remelted; n=8), XPE3 (3x30 kGy Gamma, annealed and sequential irradiated; n=8), UHMWPE particles (n=8) and control (n=8). 50 ll of the particle suspension were injected into the left murine knee joint under sterile conditions. The leukocyte-endothelial cell interactions and the synovial microcirculation were performed by intravital fluorescence microscopy one week after particle injection to assess the inflammatory reaction against the particles (by measuring the rolling fraction of leukocytes, the adherent cells and the functional capillary density (FCD)). Data analysis was performed using a computer-assisted microcirculation analysis system (Cap-Image). For the statistical analysis the Kruskal-Wallis test was used to determine differences within the groups, followed by an all pairwise multiple comparison procedure with a Bonferoni correction. The level of significance was set at p\0.05. Results: For all groups the particles were smooth, granular, irregular and less fibrillar. More than 85% of the particles were \1lm.

S303 The intravital microscopy showed a significantly (p\0.05) increased fraction of the rolling leukocytes, adherent cells and FCD in all biomaterials compared to control group without a significant difference between the UHMWPE and the XPE groups (p[0.05). Conclusions: Using crosslinked polyethylene in knee arthroplasty does not cause in a higher biological activity of the wear particles compared to conventional UHMWPE. According to the reduced wear rates crosslinked polyethylene could be recommended for the use in knee arthroplasty.

P19-1002 Diagnostic and therapeutic value of knee arthroscopy in symptomatic patients following knee replacements. A risk—benefit analysis Schro¨der K.1, Steimer O.1, Jung J.1, Kusma M.1 1 Saarland University Medical Center, Department for Orthopaedic Surgery, Homburg/Saar, Germany Objectives: Persistent pain after knee replacement (KR) is not uncommon and can cause diagnostic and therapeutic problems. Knee arthroscopy is regularly used as a diagnostic and therapeutic tool in these patients. The objective of this paper was to determine the diagnostic and therapeutic value of knee arthroscopy in symptomatic patients following knee replacement. Methods: From January 2000 to January 2009, 37 knee arthroscopies in 34 patients (21 female, 12 male, mean age 65.5 ± 8.8 years) following total (n=32) or unicondular (n=5) KR were performed. Of the 32 arthroscopies in patients with total KR, 11 had a patella resurfacing. The mean time between knee replacement and arthroscopy was 4.6 ± 3.6 years. The patient charts were evaluated for diagnostic findings, clinical consequences, therapeutic interventions, and complications associated with the arthroscopy. Results: Of the 37 knee arthroscopies, an arthroscopic diagnosis could be found in 28 (76%). A direct arthroscopic therapeutic procedure was performed in 20 (54%). A combined arthroscopic and open therapeutic procedure in 2 (5%) of the cases, a direct open intervention following diagnostic arthroscopy was performed in 3 (8%) of the cases. As a consequence of the diagnostic arthroscopy, a delayed open procedure was performed in 10 (27%) cases. As a non-operative consequence arising from the arthroscopy, radiosynovectomy was performed in 6 (16%) cases. However, in 9 (24%) cases arthroscopy could not give any additional diagnostic information leading to further treatment. No complications could be observed. Conclusions: Knee arthroscopy following total or unicondular KR is a safe procedure. It can be a valuable diagnostic tool in patients with symptoms, which cannot be clarified with non-invasive methods. In a certain number of patients, an adequate treatment can be performed arthroscopically.

P19-1031 A novel method to determine the lower limb mechanical axis in the coronal plane Collette M.1, Hohl N.2 1 Clinique Edith Cavell, Bruxelles, Belgium, 2Exactech Company, Niederbronn, France Objectives: The correct positioning of prosthetic implants has been widely recognised as one of the key factors in long term successful total knee arthroplasties. According to the literature, a malalignment error up to 3 in varus or valgus can be obtained in 75-80% of cases, using the classical method (intramedullary rod allowing bone cuts adjustment according to preoperative Xray plannings) and in 90-95% of cases using computer assisted surgery (CAS). Despite improving the alignment accuracy, the CAS method has not spread as widely as expected, mainly because it remains costly and somewhat time consuming. Our aim is to describe a new method to determine the lower limb mechanical axis in the coronal plane, as accurately but more simply and cheaply than the CAS method. Methods: According to Euclidean geometry basic principles, a perpendicular raised in the middle of any chord of a circle, passes through its center. When the knee is brought from abduction to adduction, it makes an arc of a circle turning around the centre of the hip.

123

S304 The femoral axis aiming device we present is made of a vertical bar, firmly fixed to the operating table, and a horizontal one overhanging the operated knee. Through a guiding tubes system, it allows to mark any point of the femoral condyle and to keep the ‘‘memory ‘‘ of its spatial location.The situation in space of a selected femoral mark (any point is suitable) is determined and kept in ‘‘memory’’ both in abduction and adduction position. A rod passing perpendicularly through the center of the segment joining those two positions will automatically cross the center of the hip. During 20 total knee replacements operated between march and June 2007,the coronal plane femoral axis was determined using this method and an image intensifier set over the hip joint as a reference. The results obtained with each method were compared and the differences measured. Results: In comparison with the ideal 0 reference axis: Mean error (in degrees) ± SD : 0.6 ± 0.6. Error B1: 80% of cases. Error B3: all cases. Conclusions: The described method allows to determine the lower limb mechanical axis in the frontal plane without violating the femoral medullary canal and without any computer assistance. It is quick, simple and inexpensive. It is compatible with minimal invasive procedures and could be also used in uni arthroplasties or osteotomies.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

Average cement penetration depth for each technique

5 different cementing techniques P19-1036 Cementing the tibial component in total knee arthroplasty: which technique is the best? Vanlommel J.1, Luyckx J.-P.2, Labey L.2, Innocenti B.2, De Corte R.3, Bellemans J.4 1 KU Leuven, Pellenberg, Belgium, 2European Centre for Knee Research, Leuven, Belgium, 3ECKR, Leuven, Belgium, 4KU Leuven, Leuven, Belgium Objectives: Aseptic loosening of the tibial component remains a major cause of failure in total knee arthroplasty and may be related, directly or indirectly, to micromotion. Therefore, good fixation of the tibial component is a prerequisite to achieve long-term success of the implant. Cementing technique is one of the factors that play a role in this respect. We investigated the effect of different cementing techniques on the cement penetration in the proximal tibia. Methods: We compared 5 different cementing techniques in an anatomical open pore sawbone model (n=25), using a contemporary TKA design and standard polymethylmetacrylate cement. In the first technique, 10 g of cement was applied in a thin layer on the lower surface of the tibial component. The component was then placed and impacted onto the tibia using the specific component impactor supplied by the manufacturer. In the second technique, 20 g of cement was applied in a thick layer on the lower surface of the tibial component. In the third technique, 20 g of cement was applied in equal parts, on both the tibial component and the tibial bone using a spatula. In the fourth technique, 20 g of cement was applied in equal parts on both the tibial component and the tibial bone, but it was fingerpacked into the bone. In the fifth technique, 20g of cement was applied to the tibial bone with the use of a cement gun. After making cuts in the medial and lateral oblique sagittal plane of the tibia, we used Corel PHOTO-PAINT 9 to quantify the cement penetration. Results: Technique 1 (thin layer of cement on the tibial implant only) and 2 (thick layer of cement on the tibial implant only) were not significantly different from each other in terms of penetration depth, but were both significantly different from the other techniques. The same was seen for technique 3 (two equal parts of cement on both the tibial component and the tibial bone using a spatula) and 4 (two equal parts of cement on both the tibial component and the tibial bone, using the fingerpacking technique). The penetration depth was highest for technique 5 (using a cement gun), which was significantly different from all the other techniques.

123

Example of a section with the measuring scale Conclusions: We demonstrated that applying cement to both the undersurface of the tibial baseplate and as well as onto the tibial bone, either by a spatula or fingerpacking technique, leads to an optimal cement penetration of 3-5mm. When cement is applied only onto the tibial component, penetration is insufficient. When a cement gun is used, cement penetration is too excessive.

P19-1050 Three-dimensional analysis of unicompartimental knee implant positioning Servien E.1, Lustig S.2, Demey G.3, Neyret P.4 1 Centre Albert Trillat, Hospices Civils de Lyon, Lyon-Caluire, France, 2 Centre Albert Trillat CHU Lyon Nord, Orthope´die, Caluire-Lyon, France, 3 Centre Albert Trillat, Lyon-Caluire, France, 4Hopital Croix-Rousse Centre Livet, Chirurgie Orthopedique, Lyon, France Objectives: The purpose of this study was to analyze the positioning of unicompartmental knee arthroplasty (UKA) and to compare between the medial and lateral compartments. Methods: All patients were examined postoperatively using computed tomography with three-dimensional analysis of the lower limb from the hip to the ankle. There were 18 lateral and 19 medial UKAs. All knees were analyzed using an image processing software that enabled 3D bone reconstructions and digitization. We measured the varus-valgus inclination and internal-external rotation of the femoral and tibial components.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339

S305

Results: The rotation of the femoral component was external (mean 3.2, SD 7.3) for the medial compartment, and internal (mean 5.8, SD 7.2) for the lateral compartment, and the difference was statistically significant (p \ 0.001). The rotation of the tibial component was external for the medial compartment (mean 6.5, SD 5.1) as well as for the lateral compartment (mean 7.3, SD 10.3), and the difference was statistically insignificant (p = 0.717). The inclination of the tibial component was in varus (5.5, SD 2.8) for medial compartment, and in valgus (mean 1.3, SD 4.0) for the lateral compartment. Conclusions: Our study presents the first 3D in vivo analysis of UKA positioning for both compartments, which remains a difficult surgical procedure.

after surgery. Touniquet was used after 200-300ml blood loss during the surgery. Blood loss during the surgery and total blood loss, decrease of Hb, and effect of CBC II system were investigated. Results: Mean blood loss during the surgery was 259ml and total blood loss was 901ml. Decrease of Hb was 3.4. Blood reinfusion was performed in 49 cases and mean volume was 333ml. Only 2 cases (3.4%) had homologous blood transfusion. One case was RA and another one case was the patient with severe renal failure. Conclusions: In our previous investigation using only CBC II system, six of 62 cases (9.7%) had the homologous blood transfusion. Combination of previously deposited autologous blood transfusion and reinfusion using CBC II system could reduce the homologous blood transfusion after TKA.

P19-1080 An in-vivo study of linear penetration in the Oxford unicompartmental knee replacement at twenty years Kendrick B.1, Simpson D.1, Bottomley N.1, Kaptein B.2, Garling E.2, Gill H.S.1, Dodd C.1, Murray D.1, Price A.1 1 Nuffield Orthopaedic Centre, NDORMS, Oxford, United Kingdom, 2 Leiden University Medical Center, Orthopaedics, Leiden, Netherlands Objectives: To investigate the linear penetration rate of the polyethylene bearing in unicompartmental knee arthroplasty at twenty years. Methods: The Phase 1 Oxford medial UKR was introduced in 1978 as a design against wear, with a fully congruous articulation. In 1987 the Phase 2 implant was introduced with new instrumentation and changes to the bearing shape. We have previously shown a linear penetration rate (LPR) of 0.02 mm/year at ten years in Phase 2, but that higher penetration rates can be seen with impingement. The aim of this study was to determine the 20 year in-vivo LPR of the Oxford UKR, using Roentgen Stereophotogrammetric Analysis (RSA). Six Phase 1 (5 patients, mean age 65.3 years) and seven Phase 2 (4 patients, mean age 63.5) Oxford UKR bearings, with an average time since surgery of 22.5 years and 19.5 years respectively, were assessed. Stereoradiographs were taken and penetration was calculated using a model-based RSA system. The penetration for each bearing was calculated by subtracting the measured thickness from the corrected nominal bearing thickness. Results: The measured LPR was 0.072 mm/year for Phase 1 (S.D. 0.028, range: 0.031 - 0.104 mm/year) and 0.028 mm/year for Phase 2 (S.D. 0.019, range: 0.014 - 0.07 mm/year). This difference in LPR for the two phases was statistically significant (p = 0.0028). Conclusions: The results show that in the knees studied there was a significant difference between the LPR in Phase 1 and 2. This is probably due to the change in instrumentation and bearing shape between Phases. This study demonstrates that very low penetration rates can be maintained to the end of the second decade after implantation. This is of particular importance when the device is used in younger patients.

P19-1105 Residual lateral laxity after TKA in varus knee affects medial stability In Y.1, Cho H.-M.1, Suhl K.-H.1 1 Uijongbu St. Mary’s Hospital, The Catholic University of Korea, Uijongbu-Si, Korea, Republic of Objectives: Although algorithmic balancing approaches for total knee arthroplasty (TKA) in varus knees have been documented, we could not get balanced gaps in some cases and had to allow extension lateral laxity. The purpose of this study was to evaluate whether or not knees with residual lateral laxity predispose to inferior clinical results to balanced knees after TKA. Methods: Between January 2004 and September 2007, we performed 711 primary PS TKAs. We have been using a step by step release approach to the varus knee which includes deep medial collateral ligament (MCL) release, posterior capsule and semimembranosus release, and superficial MCL and pes tendon release at tibial side. There were 23 bilateral TKA patients who had residual lateral laxity in one knee and balanced gap in the contralateral knee to be analyzed in this study. The stability of the knee, lower extremity alignment, range of motion, HSS and Knee Society scores were compared between knees with residual lateral laxity and knees with gap balance. The mean follow up was 3.1 years. Results: There were no significant differences in lower extremity alignment, range of motion, HSS and Knee Society scores between laterally lax knees and balanced knees. On valgus- varus stress radiographs, the laterally lax knees were significantly looser than the balanced knees not only laterally but also medially. But, no patient reported significant symptomatic instability. No knee had osteolysis or loosening. Conclusions: Knees with residual lateral laxity after TKA were looser not only laterally but also medially than knees with gap balance at mean 3.1 years follow up.

P19-1103 Postoperative autologous blood transfusion after total knee arthroplasty Shoda E.1, Ohuchi K.1, Maruyama S.1, Suzuki A.1, Kitada S.1, Sakurai A.2 1 Hyogo Prefectural Nishinomiya Hospital, Orthopaedic Surgery, Nishinomiya, Japan, 2Hyogo Prefectural Awaji Hospital, Orthopaedic Surgery, Sumoto, Japan Objectives: Reducing the need for homologous blood transfusion after total knee arthroplasty (TKA) is important because of the small but serious risk of viral disease and transfusion reactions. To avoid homologous blood transfusion, transfusion of previously deposited auotologous blood (400ml) and blood reinfusion using Consta Vac blood conservation system II (CBC II, Stryker Inc.) were performed after TKA. Methods: This study was retrospective and included 59 cases received cement TKA (NexGen CR-Flex, Zimmer Inc.). 400ml aoutologous blood was deposited 1 week before surgery and CBC II was used in all cases

P19-1120 The influence of prosthesis design on outcomes of TKA: minimum 5-year follow-up Lee M.C.1, Choi W.C.1, Lee S.1, Seong S.C.1 1 Seoul National University College of Medicine, Orthopaedic Surgery, Seoul, Korea, Republic of Objectives: Various prosthesis designs have been evolved to improve the clinical result and longevity of TKA. The aim of this study was to investigate the influence of prosthesis design on outcomes of TKA. Methods: A minimum of 5-year follow-up results of 189 primary TKAs in 138 patients with 4 different types of prosthesis serially evolved by same company were retrospectively reviewed. Group 1 knees(n=61) were implanted with a prosthesis that has a cruciate retaining(CR) multiradius(MR) femoral component and Group 2(n=43) with a CR singleradius(SR) component. In Group 3(n=31) and Group 4(n=54), a posterior stabilizing(PS) SR component were used in common, while a tibial insert with relaxed posterior slope was used in Group 4. The tibial component design was identical. Range of motion (ROM), American Knee Society Score (AKSS), and radiographic outcomes were evaluated. Subgroup analyses were performed to evaluate the effect of different radius(Group 1

123

S306 vs. 2), CR or PS (Group 2 vs. 3) and relaxed posterior slope of tibial insert (Group 3 vs. 4) on the outcomes. Results: Preoperative demographics and clinical or radiographic results were similar among groups except the duration of follow-up. ROM and AKSS were not significantly different among the groups(p=.283, p=.220) and neither tibiofemoral angle (TFA) nor patellar tilt angle (PTA) were different (p=.869, p=.404). Subgroup anlysis between Group 1 and 2 showed no differences in terms of ROM, AKSS, TFA and PTA(p=.516, p=.934, p=.070, p=.169). The extensor mechanism function was assessed by ability to rise from a chair, stair climbing, anterior knee pain and extension lag and the differences were not significant between the groups (p=.118, p=.330, p=.476, p=.247). Differences in ROM, AKSS, TFA and PTA between Group 2 and 3 (p=.640, p=.787, p=.174, p=.836) and Group 3 and 4 (p=.081, p=.644, p=.126, p=.137) were not significant. Revision rate was 5.2% (10/189) and the differences among the groups were not significant (p=0.613). Conclusions: TKAs with 4 different prosthesis designs resulted in similar outcomes with regard to ROM and clinical or functional results.

P19-1181 Strategic prophylaxis of surgical site infection with multi-resistant coaglase negative staphylococci associated with total knee arthroplasty Enomoto H.1, Ohno K.2, Kishino S.2, Niki Y.1, Matsumoto H.3, Otani T.4, Toyama Y.1, Suda Y.1 1 Keio University, Dept. of Orthopaedic Surgery, Tokyo, Japan, 2Meiji Pharceutical University, Tokyo, Japan, 3Keio University, Institute of Integrated Sports Medicine, Tokyo, Japan, 4Keio University, Faculty of Nursing and Medical Care, Tokyo, Japan Objectives: According to the surveillance regarding nosocomial infections in USA and Japan, multi-resistant coaglase negative staphylococci (MRCNS) such as epidermidis and the others dominate approximately 90% and 80% respectively. Actually the resistant rates are still increasing. During the 5 years between 2001 and 2006 we had treated surgical site infections caused by MRCNS associated with total knee arthroplasty (TKA-SSI), which we had never experienced before. Thus we investigated their clinical aspects for the purpose of early diagnosis, and introduced the new strategy to prevent SSI with MRCNS more efficiently. Methods: Six cases of TKA-SSI, out of 451 cases of TKA which were performed from 1995 to 2006, were analyzed with respect to bacterial strains, time from TKA to first salvage surgery, laboratory findings (serum CRP, counts of leukocyte and neutrophil), and localization of radiolucent area detectable on Xray. In 2007 we updated the prevention protocol by altering antibiotics from cephalosporin to teicoplanin to reduce the rate of infection with MRCNS. Pharmacokinetic studies were also conducted by measuring the level of teicoplanin (TEIC), which is administered just twice on the day of TKA, in synovium as well as plasma by high sensitivity HPLC-EDC. Results: The rate of TKA-SSI has risen from 0 to 3% in recent 5 years. All cases of TKA-SSI were found to be late infection, which time frame was 331±206 days. Among laboratory parameters, only the levels of serum CRP at 1 and 2 weeks after the surgery were statistically higher in TKASSI cases (1w; 10.4±4.0, 2w; 5.4±2.8 mg/dl) compared with non-SSI cases (1w; 3.7±2.7, 2w; 1.5±2.1 mg/dl) (p\0.01). In terms of radiological examinations, 50% revealed radiolucent area at zone 4 in tibia, and 80% at zone 1 in tibia. The level of TEIC in plasma during TKA (16.2±21.8 lg/ ml) and prior to the final venous administration (6.1±0.75 lg/ml), and its level in synovium during TKA (3.0±1.5 lg/g tissue) were all above the level of MIC90 (2.0 lg/ml). Conclusions: All cases of TKA-SSI in this study were found out to be late infection more than 3 months after the surgery. Serum CRP levels as well as localization of radiolucent area in tibia and femur is suggested to be helpful to observe with attention postoperatively and to diagnose earlier to avoid bone loss due to advancement of osteolysis originated from the infection. Based upon the pharmacokinetic study here and the fact that we have not yet experienced TKA-SSI after applying the new protocol, we

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 consider it to be beneficial to prevent TKA-SSI with multiresistant staphylococci.

P19-1218 Position of prosthesis in navigation TKA: limited correlation with radiographic measurements Lee M.C.1, Choi W.C.1, Lee S.1, Seong S.C.1 1 Seoul National University College of Medicine, Department of Orthopaedic Surgery, Seoul, Korea, Republic of Objectives: Navigation TKA is expected to increase the accuracy of component position and reproducibility of surgical techniques. This study investigated whether the navigation TKA leads to significantly more accurate component positions than the conventional technique. We also evaluated the correlation between navigation and radiographic alignment measurement. Methods: Consecutive 160 TKAs for osteoarthritic knees were enrolled in this prospective study and classified into two groups. Group 1 consisted of 80 TKAs with navigation and Group 2, 80 TKAs with conventional technique. We compared the mechanical leg axis, coronal femoral and tibial component alignment and number of outliers those exceeding 3 of varus or valgus deviation between the groups by radiographic assessment from two independent observers. Intraoperative navigated measurement and radiographic alignment were compared and the correlation was evaluated between them. Results: Femoral alignment of Group 1 was significantly more accurate than that of Group 2 (p=0.010), while mechanical leg axis and tibial alignment were not significantly different (p=0.845, p=0.094). The number of outliers in mechanical leg axis (25% vs. 25%) and tibial alignment (3.8% vs. 8.8%) showed no significant differences between the groups (p[0.999, p=0.328), however, Group 1 showed significantly less number of outliers with regard to femoral alignment (7.5% vs. 28.8%, p=0.001). Radiographic and navigation measurements showed weak correlation for mechanical leg axis and tibial alignment (r=0.347, r=0.267) and no correlation for femoral alignment (p=0.177). Conclusions: Femoral alignment was more accurate by radiographic assessment with navigation TKA than that of conventional TKA while mechanical leg axis and tibial alignment were similar between them. Navigation and radiographic alignment measurements were not or weakly correlated.

P19-1224 Clunk and painful patellofemoral crepitus after TKA: incidence, clinical feature, risk factors and treatment Lee M.C.1, Choi W.C.1, Lee S.1, Seong S.C.1 1 Seoul National University College of Medicine, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: Patellar clunk syndrome is generally believed to be a resolved problem with contemporary design TKA, however, painful patellofemoral (PF) crepitus is a newly emerged complication although the clinical importance is controversial. We evaluated the incidence, clinical feature, risk factors and treatment options of clunk or painful PF crepitus for which operative treatment was needed after TKA. Methods: A total of 944 patients who underwent primary TKAs with 6 different implants and followed-up for a mean of 5.3 years were retrospectively reviewed. The incidence and onset time of clunk or painful PF crepitus were assessed. The difference in range of motion, outcome scores and radiographic results including limb alignment, patellar position, and CT measurement for femoral component rotation were compared between patients with and without the complication. We also evaluated the treatment option and result. Results: Twenty patients (2.1%) underwent operative treatment for clunk or painful PF crepitus at average 7.1 months (range: 3 to 12 months) after TKA. TKAs with specific prosthesis design showed significantly higher incidence (9.73% vs. 0 to 1.69%, P\0.001). Incidence was higher in patients without patellar resurfacing (P=0.0078) and clinical, radiographic results showed no significant differences except increased patellar tilt (4.1

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 vs. 3.1, P\0.001) and decreased patellar height measured by BlackburnePeel ratio (0.76 vs. 0.8, P=0.049) in patients with clunk or painful PF crepitus. Patients suffered from clunk or painful PF crepitus were treated by arthroscopic debridement except 4 patients treated by resurfacing of patella. Symptom was alleviated and no recurrence was observed at average 4.1 years of follow-up. Conclusions: Clunk and painful PF crepitus have revisited in TKAs using specific implant. In such implant, it is recommended to resurface the patella and to avoid changes in patellar height or excessive patellar tilt. For the treatment of clunk or painful PF crepitus, arthroscopic debridement was effective to alleviate the symptom without recurrence.

P19-1234 Follow up result of combined single stage osteosynthesis and computer assisted total knee arthroplasty in a patient with psoriatic arthropathy Bhattacharyya M.1, Gerber B.1 1 University Hospital Lewisham, Orthopaedic Department, London, United Kingdom Objectives: We aim to describe the outcome of single stage supracondylar extramedually fixation and computer aided knee replacement in a case with proximal healed fracture and supracondylar femoral fracture with psoriatic arthropathy. We followed up for 4 years. We use CAOS as the proximal metal work in the femur and non-anatomical articular geometry did not allow arthroplasty with the conventional instrumentation. The secondary objective is to describe the complication following performing the procedure. Methods: A complex supra condylar femoral fracture with intra articular extension in an elderly female [fig 1], caused by a minor trauma without any other injury. She had an ipsilateral healed femoral neck fracture treated with dynamic hip screw. The patient also had past history of active psoriasis. Results: We had post operative wound infection and very delayed wound healing. Patient also sustained proximal tibial periprosthethetic fracture at 16 months after the index procedure. At 4 years follow up she has normal pre injury mobility. Conclusions: Computer aided surgery helps us to perform arthroplasty with extramedullary devices in situ [Fig 2, 3]. The precise surgical reconstruction of the mechanical axis of the knee and proper alignment of the component is achievable by using the ortho pilot system. The main advantage is the intraoperative visualisation of the leg axis, the ligament balancing and joint kinematics. The patient did not require any additional procedure of removal of metal work from proximal and distal femur. Psoriatic patient may suffer from increased wound healing problem.

P19-1243 A prospective randomised controlled trial of minimally invasive and standard surgical total knee arthroplasty; early results Tasker A.1, Hassaballa M.2, Murray J.1, Harries W.1, Porteus A.1 1 Avon Orthopaedic Centre, Department of Orthopaedics, Bristol, United Kingdom, 2Avon Orthopaedic Centre, Orthopaedic Surgery, Bristol, United Kingdom Objectives: To compare minimally invasive (MIS) and standard surgical total knee replacement technique through a prospective, randomised, single-centre, multi-surgeon, controlled trial. Methods: Between March 2007 and May 2009, 70 patients undergoing 73 total knee replacements were recruited. 42 operations were randomised to the MIS treatment arm, 31 to the standard control arm. Data were collected for mode of anaesthesia, American Society of Anaesthesiologists’ score (ASA), surgical time, Postoperative blood loss within surgical drains, length of stay and complications. Patients underwent surgery via a mini-mid vastus approach or medial parapatella approach (controls). All operations were performed using MIS instruments and the postoperative care pathway was identical between both groups. Postoperative radiographs were analysed to assess femoral and tibial component alignment. Results: There was relative homogenicity between MIS and control groups (average age 68.1 and 68.6, 65% and 55% female, combined regional and

S307 general anaesthesia 65% and 71% respectively, average ASA of 2 in both groups). Average operative time and post operative blood loss were 97 minutes and 714mls in the MIS patients versus 101 minutes and 705 mls in controls. Component alignment was similar between MIS and control groups. There were no deaths, pulmonary emboli or surgical site infections. 1 patient developed DVT and 1 required revision for pain and failure to regain flexion within 9 months of surgery, both in the control group. MIS surgery resulted in a shorter admission (4.5 days) when compared to standard surgical technique (5.6 days). Conclusions: Our study demonstrated a shorter duration of admission following MIS total knee arthroplasty when compared to standard surgical technique.

P19-1245 Loosening of the femoral component in the high-flexion TKA: the effects of cementing technique Lee M.C.1, Kim D.1, Lee S.1, Seong S.C.1 1 Seoul National University College of Medicine, Department of Orthopaedic Surgery, Seoul, Republic of Korea Objectives: Several studies have demonstrated improved clinical results after high-flexion design total knee arthroplasty (TKA) whereas there is a concern whether deep flexion will increase certain complications such as early loosening. The purpose of this study was to determine the effect of modification in the cementing technique on radiolucency of the femoral component after high-flexion TKA. Methods: Four hundred and twenty one knees (421 patients) that had been treated with high-flexion TKA between 2002 and 2005 were prospectively evaluated and divided into two groups on the basis of the modification in cementing technique for fixation of the femoral component. In Group 1 (217 knees) femurs were prepared with cement on the femur only without manual pressurization. In Group 2 (167 knees) femurs were prepared with manual pressurization of the additional lump of cement on anterior cut surface of the femur with thumb as well as cementing of the component. We analyzed postoperative radiographs at each follow-up interval to compare the radiolucent lines (RLLs). Survivorship analyses were performed using the life-table method with different definitions of end point. Results: Clinical outcome was comparable to other series in both groups and estimated survival rate at 6 years with revision for any reason was 97%. Survival rate was 98.3% at 7 years for Group 1 and 98.5% at 4 years for Group 2 in the life-table method with revision for femoral loosening or progression of RLLs as an end point, there was no significant difference between groups (p = 0.7948, Wilcoxon test). However, survival rate was 87.1% at 7 years for Group 1 and 98.8% at 4 years for Group 2, and there was a significant improvement (p \ 0.001) of the Group 2 over Group 1 with detection of RLLs as an end point. Of thirty knees with RLLs, one knee of a rheumatoid arthritis patient and two knees of one osteoarthritis patient in Group 1 were revised due to aseptic loosening, two knees in Group 2 showed progression of RLLs under the anterior flange of femoral component. Remain twenty-five femoral RLLs of both groups did not progress. Conclusions: Proper cementing technique for femoral component of highflexion TKA could reduce the occurrence of a radiolucent line in the early postoperative phase. This study could partially explain the clinical problem of early loosening of femoral components in high-flexion TKA despite of comparable clinical results.

P19-1261 Influence of posterior condylar offset on knee flexion after cruciatesacrificing mobile-bearing total knee replacement: a prospective analysis of 410 consecutive cases Bauer T.1, Biau D.1, Colmar M.2, Poux X.3, Hardy P.1 1 Hopital Ambroise Pare´, West Paris University, Boulogne, France, 2 Clinique Jeanne d’Arc, Saint-Brieuc, France, 3Laboratoire CERAVER, Roissy, France Objectives: The range of motion of the knee joint after Total Knee Replacement (TKR) is a factor of great importance that determines the

123

S308 postoperative function of patients. Much enthusiasm has been recently directed towards the posterior condylar offset with some authors reporting increasing postoperative knee flexion with increasing posterior condylar offset and others who did not report any significant association. Methods: Patients undergoing primary total knee replacement were included in a prospective multicenter study and the effect of the posterior condylar offset on the postoperative knee flexion was assessed after adjusting for known influential factors. The same cemented cruciate-sacrificing mobile bearing implant was used for all the patients. Clinical data, active knee flexion and posterior condylar offset were recorded preoperatively and postoperatively at a minimal one year follow-up. Univariate and multivariate linear models were fitted to select independent predictors of the postoperative knee flexion. Results: Four hundred and ten consecutive total knee replacements (379 patients) were included in the study. The mean knee flexion was 112 preoperatively and 108 postoperatively. The mean condylar offset was 28.3 mm preoperatively and 29.4 mm postoperatively. The multiple linear regression model identified only male gender and increased preoperative knee flexion as independent predictors of increased postoperative knee flexion. Conclusions: No correlation was found between the posterior condylar offset or the tibial slope and the postoperative knee flexion. The most significant predictive factor for postoperative flexion after posterior stabilized TKR without PCL retention was the preoperative range of flexion, with a linear effect.

P19-1284 Bilateral simultaneous knee replacement. Comparison with staged bilateral short term or long term procedure Jenny J.-Y.1, Ciobanu E.2 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France, 2Hoˆpitaux Universitaires de Strasbourg, Centre de Chirurgie Orthope´dique et de la Main, Illkirch, France Objectives: We wanted to compare the early results of bilateral simultaneous TKR, staged bilateral TKR and unilateral TKR performed by the same surgeon with the same operative technique and the same post-operative regimen. Methods: 60 cases scheduled for bilateral TKR were selected. 20 cases of staged short term bilateral TKR (2 week interval) were performed in the first part of the study (group A), and 20 cases of simultaneous bilateral TKR in the second stage (group B). 20 cases of unilateral TKR performed during the same period were randomly selected as a control group (group C), assuming that patients operated for staged long term bilateral TKR (more than 6 month interval) will experience twice the same procedure. Three-month results were compared in the three groups. Results: We observed no increase in the complication rate in both bilateral groups in comparison to unilateral group. 5 patients in each group received allogeneic transfusion. Total hospitalization time was shorter in group A (11days) than in group B (20 days), and was not increased in group A in comparison to group C (11 days), while it was almost doubled for group B. Total rehabilitation time showed similar differences. Hospital and surgeon reimbursement were significantly lower for group A. Conclusions: Simultaneous bilateral TKR did not show any medical inconvenient in comparison to staged short term or long term bilateral TKR. Total costs for the health care system might be lower due to the simultaneous hospitalization, rehabilitation and social consequences. However, hospital and surgeon reimbursements are lower.

P19-1289 Navigated, fully congruent floating platform unicompartmental knee replacement. A two year follow-up study Jenny J.-Y.1 1 University Hospital Strasbourg, Center for Orthopedic and Hand Surgery, Illkirch, France Objectives: Navigation has proved increasing the accuracy of unicompartmental knee replacement implantation. Mobile bearing prostheses have proved decreasing the polyethylene wear. However, the implantation technique is more demanding, and higher rates of complications have been reported. We wanted to test the following hypothesis: the combination of a

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 fully mobile and fully congruent polyethylene bearing and a navigated implantation will affect the functional and radiological results in comparison to a standard fixed bearing prosthesis. Methods: 80 cases of fully congruent, floating platform unicompartmental knee prostheses implanted with a navigation system were matched to 80 patients of an historical group of uncongruent, fixed bearing total knee prostheses implanted with the same navigation system by the same surgeon. Pre-operative KSS score, early post-operative X-ray axes and the 5 year follow-up KSS score and 5 year X-rays were recorded. Differences between both groups were analyzed with appropriate statistical tests at a 0.05 level of significance. Results: There was no significant difference between both groups in all pre-operative items. There was a significant increase in the clinical KSS score for floating platform prostheses, and this was clearly under the influence of a higher knee flexion. There was no difference between the two groups for all radiological items. There was no significant difference in the survival rates. There was no difference in the complication rate. Conclusions: This new model of total knee replacement showed a significant increase in the final knee flexion angle. This might be due to a better prosthesis kinematic with more physiological femoral roll-back as a consequence of the self-aligning possibility of the tibial gliding component. The short-term survival was not affected. This navigated, fully congruent floating platform unicompartmental knee replacement is a valuable alternative against conventional fixed bearing unicompartmental knee replacements.

P19-1320 Alteration of skin sensation over the front of the knee after knee arthroplasty. Impact on kneeling ability Hassaballa M.1, Porteous A.1 1 The Bristol Knee Group, Orthopaedic Surgery, Bristol, United Kingdom Objectives: Dermal dysaesthesia is a well-recognized consequence of knee replacement, but it has received little attention in the literature. Altered skin sensation is unpleasant for many patients at least initially, and may affect function, especially kneeling. The aim of this study was to compare postoperative skin dysaesthesia after midline, medial and short medial incisions. Methods: Eighty patients with 87 knees were examined for skin sensory alterations over the front of the knee at a minimum follow up of 18 months after knee arthroplasty. Seventeen knees had a long anteromedial incision, 21 had a midline incision, 40 had a short medial incision and 9 cases had a minimally invasive incision (MIS) for UKR. A purpose-designed grid was used to record sensory alterations in different areas. Results: The average length for the anteromedial incision was 18.5 cm with an average area of sensory alteration of 88.2 cm2. The midline incision average length was 17.7 cm with an average area of sensory alteration 56.5 cm2. The short medial incision used for UKR averaged 11.9 cm in length with an average area of sensory alteration of 49.3 cm2. Conclusions: The length of the incision correlated to the area of altered sensations, when comparing long and short medial incisions. Kneeling ability was related to the size and type of sensory alteration. Hypoesthesia improved with time while hyperaesthesia did not.

P19-1353 The rotaglide mobile bearing total knee arthroplasty results of 5 to 8 year follow-up Iosifidis M.1, Neophytou D.1, Liakos T.1, Sakorafas N.1, Karnatzikos G.1, Papantoniou E.1, Albanos D.1, Kyriakidis A.1 1 Papageorgiou G.H., Orthopaedic Surgery, Thessaloniki, Greece Objectives: Failure of a TKA is caused many times from the polyethylene debris or the mechanical forces which lead to loosening mostly to tibial component. The mobile meniscal knee prosthesis could provide solution as it simulates better normal knee function. The aim of our study is to present the midterm results of TKA using mobile bearing platform Rotaglide.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: During the period 2000-2004 we performed 261 TKA with the Rotaglide mobile polyethylene prosthesis (Corin Medical, UK). They are 235 women and 26 men, mean age 76.33 years, and the 93.7% of them (M=245) with primary osteoarthritis. The tibial component was cemented for all them, and for the femur was cementless for 146 cases (hybrid) and cemented for 115. None of them had patella replacement. One hundred five patients (59 hybrid and 46 cemented) were examined clinically and radiologically and the minimum follow-up time was 5 years (mean 6,6 / range 5-8 years). We use the Knee injury and Osteoarthritis Outcome Score (KOOS- range of scale for each subscore 0-100). Results: There was significant improvement of knee function and the majority of our patients were satisfied from the result. Specifically, the score for general symptoms and joint stiffness was 89.1, for pain was 83.3, for daily activities was 75.6 and for the quality of life was 72.5. One patient had revision TKA for femur component aseptic loosening. Conclusions: Our results indicate that the Rotaglide total knee arthroplasty is a great choice for primary knee OA with excellent functional result.

P19-1362 Norapine periarticular injection for pain management after total knee arthroplasty Adravanti P.1, Banchini P.1 1 Casa di Cura ‘Citta` di Parma’, Orthopaedic Department, Parma, Italy Objectives: Nowadays prosthetic surgery need to satisfy the always higher expectations of the patients, including a reduced post-operative pain. Indeed, following total knee arthroplasty (TKA) patients experience considerable pain which can also affect the functional post-operative results. In this study we analyze the efficacy of an intraoperative periarticular injection of Naropine 0.2%/Adrenaline on post-operative pain and NSAIDs assumption level. Methods: 100 consecutive consenting patients undergone PS TKA were randomly assigned either to group A (Naropine 0.2%/Adrenaline infiltration) or group B (saline infiltration). Inclusion criteria were primary or secondary knee arthrosis with indication for TKA; patients affected by severe heart diseases, artery pathologies, diabetes, neuropathic pain and neuropathologies were excluded from the study. Patients were operated by two senior surgeons. A sub-aracnoid anesthesia was performed, always using tourniquet and ischemic fascia. For all patients the following post operative protocol was used: for 48 hs from intervention Ketoprophene and Morphin by endovenous elastomeric pump; for 3 days Naropine 0.3% by perineural femoral catheter. All patients followed the same post-operative rehabilitation protocol. Patients were interviewed up to 7 days after intervention for pain and pain relief using VAS pain Scale (three times a day: morning, early afternoon and evening). Narcotic usage was also recorded. Results: After 24 hours from interventions the VAS score for group A was lower than for group B, but not significantly different (p[0.05); the difference between groups was appreciable until 72 hours, then patients from both groups experienced the same level of pain. The NSAIDs assumption was reduced in patients of group A until 72 hours from interventions. Conclusions: Our study seem to suggest that, although the intraoperative periarticular injection of Naropine 0.2%/Adrenaline results in decreased pain levels and narcotic consumption, it is not statistically significantly better than placebo. Other studies, including a higher number of patients, should be performed to better investigate the efficacy of this type of treatment.

P19-1369 Management of severe bone loss with tantalium cones in revision total knee arthroplasty Biserni M.1, Ginese A.2, Coppini R.1, Mugnaini M.3, Olivieri M.2 1 Ospedale Civile di Campostaggia, Orthopaedics and Traumatology, Poggibonsi, Italy, 2Ospedale Civile di Campostaggia, Poggibonsi, Italy, 3 Ospedale Civile di Poggibonsi, Orthopaedics and Traumatology, Poggibonsi, Italy

S309 Objectives: The present study details the indications, surgical technique, and clinical results for reconstructive strategy which uses the mechanical and biologic benefits of highly porous metal technology. in our experience, when the bone loss is severe, reconstruction of the supporting tibial or femoral platform and filling of the metaphyseal voids can be performed with newly developed highly porous metaphyseal cones along with bone graft and cement. Methods: In our series we have treated from February 2005 through July 2008 11 patients (age 75 through 86) that underwent knee revision for total knee prosthesis loosening associated to massive bone loss. 7 patients were males, while 4 females.The underlying pathology was primary osteoarthritis for 8 knees, rheumatoid arthritis for two and post traumatic osteoarthritis for one. Revision was for wear and osteolysis in 7 knees, for sepsis in 1, for aseptic loosening 3 cases. In 2 cases was associated a tibial periprosthetic fracture. The bone defect evaluated both radiografically and at surgery was AORI type 2 in 7 knees while in 4 was AORI type 3. All the cases but one and were treated with one step revision procedure, while in the patient with infection we performed a two steps revision. All the revisions were performed using the CCK system. The tibial bone loss was managed in 8 cases with tantalium cones as ‘‘structural graft’’ associated along with allograft bone chips. In all the procedures we used cement void-filling technique for the femoral side. The patients have been evaluated with standard x-rays at 1-3-6 and 12 months and then yearly; a CT-scan was performed at 6 months in those patients that underwent allograft with chips of bone. Results: Two patients were lost at follow up, while the remaining were followed for a mean follow-up of 32 months with dramatic improvement on clinical outcome. Radiografic and CT-scan evaluation revealed no sign of loosening and an increasing bone ingrowth and bone stock in the area that underwent both cones and bone chips reconstruction. Conclusions: The material properties of porous tantalum and the use of trabecular metal augments, wedges, and cones combined with press-fit intramedullary stems allow biologic fixation with the potential for rapid bony ingrowth and restoration of bone stock. Trabecular metal cones can help reconstruct large cavitary defects, and most bone deficiencies may be filled by the construct itself. These implants, along with offset stems when necessary, may eliminate the need for extensive bone grafting or allografting. The increased contact area between the implant, cone, and host bone serves as a mechanical platform and as support for the revision implants with less stress shielding and disuse atrophy of the surrounding bone.

P19-1490 Minimally invasive unicompartmental knee replacement: retrospective clinical and radiographic evaluation of 173 patients at 63 months follow-up Bruni D.1, Iacono F.2, Russo A.2, Zaffagnini S.1, Raspugli G.2, Marcacci M.1 1 Rizzoli Orthopaedic Institute, Biomechanics Lab, Bologna, Italy, 2Rizzoli Orthopaedic Institute - University of Bologna, Bologna, Italy Objectives: We performed a retrospective clinical and radiographic evaluation of 173 non consecutive cases operated in our institute between 1996 and 2007 with a mean follow-up of 63 months to assess the efficiency of Unicompartmental Knee Replacement (UKR) performed with a minimally invasive technique. The aim of this study was to correlate the clinical outcome with the pre- and post-op alignment, with implant positioning on coronal and sagittal plane and with demographics. Methods: 173 non consecutive patients (63 males, 110 females) underwent cemented UKR (De Puy Preservation Uni with all poly tibial component), both for medial OA (140 patients) and AVN of the medial femoral condyle (33 patients) . All patients were available at final follow up evaluation and they all presented an evident varus alignment at preoperative clinical and radiographic evaluation. At radiographic measurement, we considered a knee with Femoro-Tibial Angle (FTA) 175 as varus knee, 17090 for valgus knee and a TPA0 for varus knee. Clinical evaluation was performed using KSS, Womac score and Oxford score. Results: According to Hospital for Special Surgery (HSS) scoring system, at a mean follow-up of 63 months, 125 (74%) cases were excellent (80-

123

S310 100 points), 23 (18%) cases were good (70-79 points), 13 cases (8%) were sufficient (60-69 points) and 12 had bad results (0 points). In our series, patients with an excellent clinical result presented pre-operatively a mean varus deformity of 7.3 (3.6-10.8). According to literature, we demonstrated that a small amount of undercorrection with a small amount of residual varus deformity of 3-5 is the goal to be reached in order to avoid both rapid degeneration of the non-replaced compartment as well as the premature loosening of the replaced compartment. We performed a mean axial correction of 5 (standard deviation 3.9), leaving a mean axial varus deformity of 2.2 in the excellent group. In our series the group with excellent results also showed a post-operative PTS of 7.9 (2.4-11.6), while mean pre-operative PTS was 7.5 (2.7-10.3). Mean Womac score was 86.3 and mean Oxford score was 21.5. Global revision rate was 9%, with 16 aseptic and 2 septic loosening. Conclusions: In this study, results have shown that minimally invasive UKR producing a small amount of varus undercorrection in selected patients with medial tibio-femoral osteoarthritis or moderate avascular necrosis of the medial femoral condyle provides excellent clinical and functional results. Overcorrection of varus malalignment with a UKR may produce both rapid degeneration of the lateral tibio-femoral compartment and the early failure of the replaced compartment.

P19-1491 Medial unicompartmental focal resurfacing of the knee. Technique and clinical results at 2 years follow-up Bruni D.1, Zaffagnini S.1, Iacono F.2, Russo A.2, Marcheggiani Muccioli G.M.1, Marcacci M.1 1 Rizzoli Orthopaedic Institute, Biomechanics Lab, Bologna, Italy, 2Rizzoli Orthopaedic Institute - University of Bologna, Bologna, Italy Objectives: Unicompartmental knee arthroplasty (UKA) is one of the treatment options for relatively young and active patients with unicompartmental femoro-tibial early arthritis or moderate osteonechrosis, no severe axial deviation and normal joint stability. We developed a new prosthetic design for focal resurfacing (MaioR) which can be implanted with an arthroscopic minimally invasive technique and no bone sacrifice. Methods: It is an uncemented, focal resurfacing prosthesis that requires minimal bone sacrifice and utilizes a minimal invasive surgical (MIS) approach with or without arthroscopic assistance. Moreover, actually MaioR can be implanted both with a cementless technique, to obtain a completely biological integration, which is able to guarantee an optimal bone ingrowth, or with a conventional cemented technique, to obtain an immediate primary stability in selected patients. Thus, this new implant responds to the challenge to achieve a focal resurfacing which can be implanted with an arthroscopic technique and permits to maintain a normal knee kinematic without a severe bone sacrifice. We implanted 76 prosthesis and evaluated 33 patients at 2 years follow-up. Results: Subjective pain and joint function were assessed using Visual Analogue Scale (VAS) and Knee society scores respectively. VAS improved significantly at 1 year follow-up (p\0.05) and a trend towards further improvement was showed also at 2 years. HSS improved significantly both at 1 year and 2 years follow-up (p\0.05). We found a significantly higher improvement both at 1 year and 2 years follow-up in patients with a previous meniscectomy (p\0.05) and we observed a trend towards less satisfactory results in females. Conclusions: Early preliminary results at 8 months follow-up seem to show similarly good results also for cemented implants.

Leg/ankle P20-165 7 years after osteochondral transplantation of the talus - clinical and MR imaging evaluation Paul J.1, Ottinger B.1, Wo¨rtler K.2, La¨mmle L.3, Spang J.1, Imhoff A.1, Hinterwimmer S.1 1 Klinikum rechts der Isar, Department of Orthopedic Sports Medicine, Muenchen, Germany, 2Klinikum rechts der Isar, Institut fu¨r

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Ro¨ntgendiagnostik, Mu¨nchen, Germany, 3University of Augsburg, Department Psychology, Augsburg, Germany Objectives: Osteochondral lesions are often primarily treated by fragment excision, abrasion, drilling or microfracture. However, in order to treat both cartilage and the bone pathology in one step the major treatment options for osteochondral lesions of the knee, the talus and the elbow are autologous osteochondral transplantation with either mosaicplasty or the Osteochondral Autograft Transfer System (OATS). For the smaller joints such as the elbow and ankle current long-term results have not been obtained. Monitoring post-operative patients can include clinical outcome scores and magnetic resonance (MR) imaging to monitor the transplanted cylinders. Finally, it is unclear whether there is a correlation between the clinical outcome and MR imaging findings of cartilage repair tissue in patients treated with the OATS technique. The purpose of our study was to analyze the long term results of the OATS procedure for talar lesions utilizing clinical outcomes scores and MR imaging. Methods: We retrospectively analyzed 26 ankle procedures (25 patients) with an average follow up of 84 months (53 - 124 months). The patients completed the AOFAS- (American Orthopaedic Foot and Ankle Society) and Tegner score plus the visual analogue scale (VAS) preoperatively and at follow up. For 9 patients this OATS procedure represented a second procedure following clinical failure after drilling of the talus as a first line treatment. MR imaging examinations were conducted on a 1.5 Tesla whole-body magnet. T1-weighted Turbo Spin-Echo (TSE) sequence with a driven equilibrium (DRIVE) pulse and an intermediate weighted TSE sequence with a spectral fat saturation pulse (SPIR) were acquired in the coronal and sagittal planes. MR images were used to assess transplant congruency of the cartilage and subchondral bone, adjacent surface of the talus, cartilage lesions of the corresponding articular surface of the distal tibia and the presence or absence of a joint effusion. Results: Pre- and postoperative values showed significant increases for the AOFAS score (p=0.045), the Tegner score (p\0.01) and a significant decrease for the VAS (p\0.01). On MR imaging the joint surface congruity was normal or with minor incongruity in 81% of patients. The cartilage was normal or showed minor signal changes in 54% and small substance defects in 38% of our study group. Patients with normal integration or minor problems with integration of the transplant had significant better AOFAS scores (p = 0.03). However, for the other criteria no significant correlation between MR imaging results and clinical outcomes was found. Patients for whom the OATS represented a second procedure had significantly worse clinical scores (AOFAS and Tegner score) plus a worse VAS when compared with patients who had the OATS procedure as a first line treatment for a talus osteochondral injury. Conclusions: Long term clinical and MR imaging results after osteochondral transplantation are good and patients significantly profit from this surgery. Patients who had prior surgical procedure for a talar cartilage defect seem to have a worse outcome after subsequent osteochondral autograft transplantation. Therefore we conclude that OATS is a good treatment for OCLs of the talus. MR imaging should not be a routine control but rather appears to be indicated when clinical symptoms warrant further investigation.

P20-384 Osteochondral defects of the talus: a novel animal model van Bergen C.1, Kerkhoffs G.1, Marsidi N.1, Korstjens C.2, Everts V.2, van Ruijven L.2, van Dijk N.1, Blankevoort L.1 1 Academic Medical Center, Orthopaedic Surgery, Amsterdam, Netherlands, 2Academic Center for Dentistry Amsterdam, Amsterdam, The Netherlands Objectives: Osteochondral defects (ODs) of the talus are successfully treated by arthroscopic debridement and bone marrow stimulation. The results are less satisfactory in both large and secondary defects. An accurate animal model for testing new treatment methods for ankle ODs is not described in the literature. The goat’s ankle is a congruent joint that bears high loads similar to the human ankle joint, and it allows for the creation and treatment of relatively large ODs. The aim of the current study was to develop and evaluate a goat model for ankle ODs.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: A 6-mm OD was created in both tali of three adult Dutch milk goats (n = 6) through a posterolateral surgical approach. According to a randomized schedule, two defects were filled with autologous cancellous bone (AB) harvested from the iliac crest, two were filled with commercially available demineralized bone matrix (DBM), and two were left empty (control). The goats were full weight bearing within 24 hours, and were sacrificed after twelve weeks. The articular surfaces were macroscopically assessed using the ICRS cartilage repair assessment with a maximum score of 12. Histologic sections were stained with Goldner0 s trichrome method or toluidine blue. Histomorphometry was performed twice by an observer who was blinded to treatment allocation, analyzing areas that were representative for the centers and the margins of the ODs. lCT-scans were obtained, and the bone volume fraction (BV/TV) was quantified in the central 3-mm cylinders. Results: A standardized OD was created in all talar domes. The AB specimens scored a mean of 8, DBM 6, and control 7, according the ICRS cartilage repair assessment. Histology showed that bone had mainly regenerated from the margins and bottoms of the defects. Observed mineralization of fibrocartilage suggested endochondral ossification. Histomorphometry showed that most central bone tissue was formed in the AB group (mean, 39.8%), followed by DBM (14.7%), and control (0.1%). The intraobserver reliability was excellent for all measured parameters (Intraclass correlation coefficient [ 0.90, p \ 0.01). lCT-analysis showed that most bone tissue was present in the AB group (mean BV/TV, 30.0%) followed by control (9.6%) and DBM (8.4%). Conclusions: The developed goat model proofs effective for the evaluation of new treatment methods for ODs of the talus. The operative technique was well reproducible, and there was a clear difference in repair between treatment with AB and control. The 6-mm critical size defect was found appropriate because control defects did not heal. No firm conclusions can be drawn on the used treatment modalities because there are too few specimens. We currently undertake a study on the effectiveness of DBM hydrated with either normal saline or platelet-rich plasma, using the presented model with more goats and longer follow-up.

P20-442 Anteromedial impingement in the ankle joint—outcomes following arthroscopy Murawski C.1, Kennedy J.1 1 Hospital for Special Surgery, Orthopaedic Foot and Ankle Surgery, New York, United States Objectives: Anteromedial impingement (AMI) is an ankle joint condition characterized by the formation of talotibial osteophytes, consequently causing a mechanical obstruction to normal joint motion. Painful symptoms arise not from the osteophytes themselves but rather from the inflamed soft tissue that becomes fixed between the osteophytes along the anteromedial joint capsule. AMI is seen as a capsular injury, and may manifests through repetitive microtraumatic plantarflexion. Other hypotheses include direct and recurrent microtrauma to the anteromedial ankle joint. It is a common condition found in athletes, particularly soccer players, causing significant time lost to play at all levels of competition. The current authors present the retrospective results of arthroscopic resection of AMI in 75 patients. Methods: Between January 2005 and April 2009, 75 patients underwent arthroscopic debridement for AMI under the care of the senior author. Any patients with evidence of pre-existing degenerative arthrosis were excluded from the present study. Surgical procedure included debridement of all cicatrized soft tissue impingement in addition to resection of any osseous exostoses. Patients were seen for a minimum of 6-months post-operatively. All patients had pre- and post-operative AOFAS and SF-36V2 scores. Results: 95% of patients reported good to excellent functional outcomes and would recommend the procedure. Mean AOFAS score improved significant from 65 points pre-operatively to 93 points at follow-up. Mean SF-36V2 also improved and was compatible with age-matched controls of the healthy population. Return to competitive sport was seen at an average of 7 weeks. Complications include a single case of superficial peroneal nerve neuropraxia and a single case of arthrofibrosis. Two patients required an

S311 additional debridement procedure at 4 and 5 months following the initial surgery. These two patients identified were early in the current series and were from inadequate index resection. This was addressed in successive procedures. Conclusions: A treatment plan for AMI is often delayed, as diagnosis is not always made on standard radiograph. No single case series has described the outcomes of AMI following arthroscopy. The current author’s stress that the resolution of AMI may not only be in the treatment but in the diagnosis as well-it is often overlooked. Early recognition and thus treatment facilitates excellent functional outcomes, few complications, and an expedited return to competitive sport.

P20-484 Ankle fractures: impact of swelling on timing of surgery, length of hospital stay and the economic burden Sukeik M.1, Qaffaf M.1, Ferrier G.1 1 Cumberland Infirmary, Trauma & Orthopaedics, Carlisle, United Kingdom Objectives: Ankle fractures are among the commonest orthopaedic injuries. A delay in operating is often due to the swelling associated with such fractures. On the other hand, the delay in operative fixation beyond 24 hours from injury is associated with a lengthening of hospital stay which costs approximately £225 per patient per day for an acute trauma bed. The aim of this study was to analyse the relationship between the delay in surgical intervention of open reduction and internal fixation of ankle fractures from presentation due to ankle swelling, and the length of hospital stay and postoperative complications. Methods: A retrospective study of 145 consecutive patients treated for ankle fractures over a period of 12 months between January and December 2008. Results were collated excluding talar and pilon fractures. Emergency department presentation times were noted and time of anaesthetic to determine surgical delay. Notes were reviewed for inpatient stay and postoperative complications. Results: There were 62 male and 83 female patients with a mean age of 49 years. In total, 117 (80%) patients were operated on within 24 hours of presentation (early group). 28 patients’ surgery was delayed beyond 24 hours (delayed group). Of the 117 patients the mean inpatient stay was 3.79 days (± 2.39) whereas in the delayed group the mean stay was 8.57 days (± 6.54). Of the delayed group, 57% of the cases had swelling as the cause of a postponed operation, whereas other causes included lack of theatre time and lack of fitness for surgery. In the early group, 5 patients (4.27%) had wound infections and one patient had a chest infection (0.85%). Four patients (14.28%) from the delayed group developed wound infections all of whom were from patients with ankle swelling. Conclusions: We recommend that policies be put in place to provide early operative intervention for patients with fractured ankles prior to the development of swelling as this would result in improved patient outcome and significant financial savings. If an operation is not feasible within 24 hours of admission and the ankle is swollen resulting in a high operative risk, it maybe worth considering sending the patient home for a period of 5-7 days with advice on RICE and anticoagulation which would both permit surgery and cut down costs.

P20-751 In vivo model of critical bone defects regeneration by autologous adipose-derived stem cells (ASCs) loaded onto hydroxyapatite scaffold de Girolamo L.1, Arrigoni E.2, Lopa S.2, Stanco D.2, Domeneghini C.3, Brini A.T.2 1 Galeazzi Orthopaedic Institute, Orthopaedic Biotechnologies Lab, Milan, Italy, 2Universita´ degli Studi di Milano, Department of Medical Pharmacology, Milan, Italy, 3Universita´ degli Studi di Milano, Department of Veterinary Sciences and Technologies for Food, Milan, Italy Objectives: Bone defect can be provoked by several pathological conditions, like bone tumors, infections, major trauma with bone stock loss. Surgical

123

S312 techniques currently used for treating bone defects may count on different alternatives, including autologous or homologous vascularized bone grafts or prostheses, each one of them characterized by both specific complications and drawbacks, including donor site morbidity and limited available amount of transplantable tissue, possible immune response and viral transmission. Nowadays, an ideal alternative is the use of osteoconductive synthetic bone substitutes either alone or in combination with autologous cells able to enhance the regeneration process and thus to provide better results. Adipose-derived stem cells (ASCs), with their great availability and osteogenic potential may represent, in association with specific scaffolds, a novel and efficient approach for bone regeneration. Here we present the results of an experimental study using ASC alone or in association with custom-made hydroxyapatite scaffold (HA) to promote the repair of bone critical-size defect in a rabbit model. Methods: Twelve adult New Zealand White rabbits were divided into two experimental groups: – control group (n=6): lesion without treatment (right tibia) / lesion ? HA (left tibia) – experimental group (n=6): lesion ? rbASCs (right tibia)/lesion ? rbASCs ? HA (left tibia) rbASCs were isolated from interscapolar adipose tissue of each rabbit belonging to the experimental group; cells were then purified by plastic adherence and expanded in vitro. Two weeks after isolation, autologous rbASCs were reimplanted alone or in association with HA scaffold in the lesion site to evaluate their osteoregenerative potential. The bone regeneration process was monitored during time performing X-ray of the lesions sites. 8 weeks after surgical interventions, gross appearance, X-ray, BMD and histological analyses were performed on each samples. At the same time we in vitro characterized rbASCs: cellular yield (number of cells per ml of raw adipose tissue), proliferation rate, cellular viability, clonogenic ability and osteogenic potential were assayed for each population of cells. Results: Regarding the in vitro study, the number of rbASCs per ml of raw adipose tissue was 2.8x105±1.9x105 cells. One week after isolation, rbASCs started to proliferate (doubling time: 65±20 hours), showing the typical fibroblast-like morphology of mesenchymal stem cells. All the tested cell populations were able to express specific osteogenic markers, confirming their ability to participate in the bone regeneration process. No fractures, infections or other complications have been observed in rabbits during the experiment. At final follow up, the gross appearance analyses and radiographs showed a good lesion filling in all the samples, but with relevant differences between lesions treated with rbASCs?HA or rbASCs alone compared to not treated lesions or just treated with HA. This differences is strongly confirmed by histological analyses, which showed the presence of a regenerated tissue similar to native mature bone just in lesions treated with rbASCs. Conclusions: ASCs seem to be a valid alternative to treat severe bone defects. Indeed they could be easily harvested, expanded in vitro and then reimplanted in the lesion The in vitro data show their higher osteogenic potential and thus their ability in actively participating to the bone regeneration process.

P20-785 Triple arthroscopic decompression for soccer player ankle Molano Bernardino C.1, Flores F.J.2, Cancelo R.2, Lopez-Vidriero E.3 1 Ibermutuamur, Orthopaedic Surgery, Shoulder Unit, Sevilla, Spain, 2 Ibermutuamur, Orthopaedic Surgery, Sevilla, Spain, 3The Hospital of Ottawa. Canada, Fellow in Sports Medicine, Department of Orthopaedics, Ottawa, Canada Objectives: 1) Description of arthroscopic findings during arthroscopic treatment of anterior impingement in patients with the so called ‘‘soccer placer ankle’’: Bone spurs in tibia, and talus 2) Description of surgical procedure used for treatment. 3) Report of clinical results

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: Patient 38 years old suffering anterior impingement of his right ankle. Symptoms were anterior ankle pain in dorsal flexion and standing. Anterior bone spurs in tibia and talus were demonstrated in Xrays and CT exam. Intraarticular anterior soft tissue hypertrophy was detected in MRI. Diagnostic local injection of anaesthetic improved symptoms. Arthroscopy under spinal anaesthesia. Pictures of Xray, CT and MRI, as well as pictures of findings and surgical technique are provided. Talus spur was reached from within the joint Results: Arthroscopic findings were 1) soft tissue hypertrophy in anterolateral and anterior recess, 2) ostephytes in anterior tibial edge and in distal margin of talus hyialin cartilage margin. 3) Entrapment of soft tissue between tibial and talar spurs during dorsal flexion of the ankle. Surgical treatment included a triple decompression: 1) anterior soft tissue resection, 2) tibial anterior bone spur resection and 3) anterior talar ostephyte resection. Clinical result was range of motion recovery without pain. Dorsal flexion was recovered in 6 weeks. Conclusions: Arthroscopy is an effective method of treatment for painful ostephyte in talus and tibia. Although very distal in location, talus spurs are visible by arthroscpy from within the joint and its resection is possible and satisfactory.

P20-891 Analysis of complications arthroscopic ankle arthrodesis for 11 years Ruiz-Zafra J.E.1, Valencia-Garcı´a H.1, Chozas-Mun˜oz A.1, Egea-Gamez R.M.1, Monteagudo de la Rosa M.2 1 Hospital Universitario Fundacio´n Alcorcon, Traumatology, Alcorcon, Spain, 2Hospital Quiro´n, Traumatology, Pozuelo, Spain Objectives: To assess clinical outcomes and analyze the complications of ankle arthrodesis performed arthroscopically in our center. Methods: Over 11 years have been operated on 15 patients through 15 arthroscopic ankle arthrodesis (10 women and 5 men) with an average age of 62 years (range :38-84), of which only 12 patients were able to complete the study of at least 1 year follow up. Of those who completed at least 1 year of follow up, the diagnosis was 8-traumatic osteoarthritis (75%), 2 cases of ankle osteoarthritis 1st and 1 case of equine foot after N. Peroneal injury. Average hospital stay of 2.3 days (range 0-7) and a mean follow up of 65 months (range 12-128,5). We used a conventional technique of ankle arthroscopy with standard anterior portals and with posterolateral portal assistance in 2 cases. 91.6% of the arthrodesis were performed using 6.5-mm cannulated screws, half of the cases with 2 and half with 3 screws. Results: Arthrodesis was achieved in 11 out of 12 cases (91.6%) at an average time of 8 weeks with a good or excellent functional outcome in all patients. We present a total of 50% of complications from surgery. Most of them (5 cases, 83.2% of all complications) sought annoyance because of osteosynthesis material, proceeding to the extraction. One of the patients had surgical wound infection after open fracture of medial ankle grade IIIA. At 8 months, after failing talotibiofibular joint because septic arthritis is extracting the material arthroscopic osteosynthesis and arthrodesis and fixation with 2 cannulated screws is made. At 10 weeks of arthrodesis, medial screw is removed due to protrusion wound with signs of superficial infection. The wound is poor outcome, with resolution of soleus flap finally. Then lateral screw protrusion is observed, so the screw is removed. We report a case of nonunion in a patient operated for neurological clubfoot using 2 cannulated screws of 4.5 mm. It was resolved by removing the osteosynthesis material and conversion into open arthrodesis and fixation with retrograde nail. A single case of subtalar arthrosis treated conservatively. Conclusions: Arthroscopic ankle arthrodesis was first used in 1983 by Schneider. It offers a less hospital stay, less postoperative pain and low perioperative morbidity, with faster fusion and faster mobilization and

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 rehabilitation. Authors like Gougoulias have a 30.8% complications and 2.6% of nonunion. In the literature the rate of fusion is between 89% and 100%. Ours is 91.6%. Tellingly, the case of non-union which was used only 2 screws and also of 4.5 mm, which is clearly insufficient to ensure the consolidation. The rate of nuisance caused by the screws is not negligible. In the literature, there are figures of removal of the material up to 33%. In our series, 41.6% of patients required removal. Though a small sample of cases, it allows us to obtain a series of conclusions (in terms of technique and results) that encourage to continue using the ankle arthrodesis without gross deformities of the axis of loading.

P20-998 To assess the reliability of preoperative templating of implants in total hip replacements (THR) using digital x-rays Campbell J.1, Felice P.2, Murray P.1 1 The Galway Clinic, Galway, Ireland, 2University of South Carolina, South Carolina, United States Objectives: To assess the reliability of preoperative templating of implants in total hip replacements (THR) using digital x-rays. Methods: This was a single centre study of retrospective design. A random selection of X-rays of 38 consecutive primary total hip replacement patients, were reviewed. The x-rays were taken with the Philips digital radiological system (DRS), transferred to the Patient Archives Communication System (PACS) from which they were then printed. The pre-op templating was carried out using standardised templates. Results: Of the 114 acetabular and femoral components only in 28 cases were the implants correctly predicted and this was predominantly on the femoral side. Conclusions: Standard THR x-rays should be of adequate magnification, show the hip joint and femur, to allow for pre-operative templating using standardised templates. The errors in magnification observed were varied random.

P20-1047 Ankle arthroscopy for osteochondral defects of the talus: outcome after 8 to 20 years van Bergen C.1, Kox L.1, Maas M.2, Sierevelt I.1, van Dijk N.1 1 Academic Medical Center, Orthopaedic Surgery, Amsterdam, Netherlands, 2Academic Medical Center, Radiology, Amsterdam, Netherlands Objectives: The primary surgical treatment of osteochondral defects (OD) of the talus consists of arthroscopic excision and curettage (EC) with microfracturing or drilling (ECD). This treatment yields 84% short- to mediate-term success but there are few data on the long-term outcome. The objective of this study was to assess the long-term clinical and radiographic outcome of ankle arthroscopy for talar ODs. Methods: We contacted all patients with an OD who had been treated through anterior or posterior ankle arthroscopy at our institution between 1988 and 2000. Exclusion criteria were: previous ankle ligament reconstruction or medial malleolar osteotomy, concomitant lesion of the tibial plafond, ankle fracture, severe recent ankle trauma, and refusal to sign informed consent. The patients were assessed using the Ogilvie-Harris and Berndt & Harty scores, American Orthopaedic Foot and Ankle Society (AOFAS) Ankle-Hindfoot score, and Short Form-36 (SF-36). Weight-

S313 bearing radiographs of the affected ankles were obtained and compared with preoperative radiographs using an ankle osteoarthritis (OA) classification. Possible prognostic factors (gender, age, earlier OD surgery, traumatic etiology, symptom duration, follow-up, body mass index, lesion size, location, and classification) were recorded and analyzed by logistic regression. Results: Sixty-three (90%) of 70 eligible patients were included after a mean follow-up of 12 (range, 8-20) years. Forty-nine patients were treated by anterior ECD, ten by anterior EC, two by posterior ECD, and two by posterior EC. According to Ogilvie-Harris, 48 patients (76%) scored good or excellent, 13 fair, and 2 poor. According to Berndt & Harty, 46 patients (73%) rated their ankles good, 13 fair and 4 poor. The median AOFAS was 88 (range, 55-100). Median SF-36 scores ranged from 72 for bodily pain to 100 for role physical and role emotional components. The radiographs were classified as OA grade 0 in 51%, grade I in 44%, grade II in 5%, and grade III in 0%. Compared to preoperative OA classification, 68% of radiographs showed no progression, 31% showed progression by one grade, and 2% by two grades. There was no statistical significant difference between primary (n = 49) and secondary surgery (n = 11), i.e., 80% and 82% good or excellent results, respectively. However, none of the patients with two earlier procedures (n = 3) had a good or excellent result. The type of treatment did not significantly affect outcome, i.e., 77% good or excellent after curettage with drilling vs. 75% without drilling. None of the prognostic factors was significantly associated with outcome (p [ 0.05). Conclusions: Arthroscopic excision, curettage, and drilling of osteochondral lesions of the talus results in good to excellent clinical outcome in three-quarters of patients, and does not typically lead to advanced osteoarthritis in the long term. This type of treatment is also effective as secondary surgery.

P20-1051 Oblique medial malleolar osteotomy for exposure of the talus van Bergen C.1, Tuijthof G.1, van Dijk N.1 1 Academic Medical Center, Orthopaedic Surgery, Amsterdam, The Netherlands Objectives: A medial malleolar osteotomy is often indicated for surgical exposure of posteromedial osteochondral lesions of the talar dome and fractures of the talar body. The oblique osteotomy is an established procedure with the advantages of excellent exposure of the talus and preservation of the deltoid ligament. However, articular incongruence after fixation might lead to secondary osteoarthritis of the ankle. To obtain a congruent joint surface after fixation, the osteotomy is to exit perpendicularly at the intersection between the tibial plafond and articular facet of the medial malleolus. The purpose of this study was to determine the optimal osteotomy direction. Methods: We assessed anteroposterior mortise radiographs and coronal computed tomography (CT) scans of 46 ankles (45 patients) with an osteochondral lesion of the talus. Two observers independently measured the intersection angle (aa) between the tibial plafond and articular facet of the medial malleolus (Figure 1). The bisector of this angle (dashed line) indicated the optimal osteotomy (a) relative to the tibial plafond. The angle (b) between the osteotomy and the longitudinal tibial axis was assessed as an anatomic reference to the osteotomy. One observer measured all radiographic angles a second time in different order. Intraclass Correlation Coefficients (ICC) were calculated to assess reliability.

123

S314

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 a the same pack, this time without heating it. Target angles were 108 and 208 of inversion. Statistical analyses were done by Wilcoxon Signed Ranks and T-tests. Results: No significant differences were found between reproduction errors of target angles before and after application of packs (p[0.05). However, significant decreases were noted in reproduction errors following hot application in both target angles during active and passive testings (p\0.05). The differences ranged between 0.52 and 0.81 degrees (p\0.05). Conclusions: Results of this study showed that hot application increases the proprioceptive capability of the ankle. These findings should be taken into consideration for the prevention and treatment of sports injuries.

Fig. 1 Results: The mean intersection angle (aa) was measured 115.0 ± 6.9 degrees according to radiography and 113.0 ± 5.7 degrees according to CT, corresponding to an osteotomy (a) of 57.5 ± 3.4 and 56.5 ± 2.9 degrees relative to the tibial plafond, respectively. The mean angle (b) between the osteotomy and the longitudinal tibial axis was 30.2 ± 4.1 degrees. ICCs were 0.73 to 0.93 for radiography, and 0.65 to 0.91 for CT (p \ 0.001). Conclusions: The optimal oblique medial malleolar osteotomy is created in a 30-degree angle relative to the tibial axis to exit perpendicularly and obtain a congruent joint surface after reduction. The methods used are reliable and allow for preoperative planning of each individual patient.

P20-1069 Effect of hot application on ankle proprioception in normal individuals Akseki D.1, Baydar M.2, Unal A.M.3, Gu¨lbahar S.2, Erduran M.1, Akalin E.2, Pinar H.4 1 Balikesir University, Orthopaedics, Balikesir, Turkey, 2Dokuz Eylu¨l University, Physical Therapy and Rehabilitation, Izmir, Turkey, 3Ag˘ri Government Hospital, Orthopaedics, Ag˘ri, Turkey, 4Dokuz Eylul University, Orthopedics, Izmir, Turkey Objectives: Importance of proprioception in the treatment and prevention of sports injuries has become increasingly clear. Internal and external factors such as exercise, fatigue, and bandage or splint application have been shown to affect proprioceptive quality. Although it is well known that application of hot increases nerve conduction velocity, its effect on proprioceptive capability is not known. The purpose of this study was to investigate the effect of hot application on ankle proprioception in normal yerine healthy volunteers. Methods: Seventeen women, 13 men, total 30 healthy volunteers with no history of ankle pathologies, trauma or surgery with ages between 20 and 30 (av. 25.2) were included in the study. Proprioception was measured on dominant ankles before and after application of hot pack with the technique of both active and passive joint position sense by a isokinetic dynamometer. Same measurements were repeated one week later by using

123

P20-1149 Novel metal implantation technique for secondary osteochondral defects of the talus: preliminary results van Bergen C.1, Reilingh M.1, van Dijk N.1 1 Academic Medical Center, Orthopaedic Surgery, Amsterdam, Netherlands Objectives: Osteochondral ankle defects mainly occur in a young and active population. In 63% of cases the defect is located on the medial talar dome. Arthroscopic debridement and microfracture is considered the primary treatment of defects up to 15 mm. To treat patients with a secondary osteochondral defect of the medial talar dome, we were involved in the development of a novel 15-mm diameter metal implant (HemiCAP). The set of 15 offset sizes was designed to anatomically correspond with various talar dome curvatures. We initially performed a biomechanical cadaver study that provided sufficient rationale for clinical use, in terms of operative technique and intra-articular contact pressure. The present study was undertaken to evaluate the clinical effect of the metal implantation technique in secondary osteochondral lesions of the medial talar dome in a prospective study. Methods: We have treated 16 patients since October 2007 by implanting the prosthetic device through a medial malleolar osteotomy (Figure 1). Two patients were excluded from the study; one because of simultaneous bone grafting for a massive osteochondral defect, and one because of diabetes mellitus. Nine patients have less than one year follow-up at the time of writing and are therefore left out of this analysis. We report five patients with one year follow-up. All patients had had one or two earlier operations without success. On preoperative CT, the median lesion size was 17 x 10 (range, 15-19 x 7-13) mm. The patients were assessed preoperatively and at 3, 6 and 12 months postoperatively, using Numeric Rating Scales (NRS) pain, Foot Ankle Outcome Score (FAOS), American Orthopaedic Foot and Ankle Society (AOFAS) Ankle-Hindfoot score, and radiographs of the affected ankles.

Fig. 1 Results: All patients recovered well from their surgery and remained nonweight bearing for six weeks. Preoperatively, the median NRS at rest was 3 (range, 0-7), compared to 0 (0-4) after 1 year follow-up. NRS at walking significantly improved from a preoperative median of 6 (5-8) to 1 (0-4) at final follow-up (p \ 0.05). Most subscales of the FAOS improved with

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 each assessment (Figure 2). The median AOFAS improved from 71 (4775) before surgery to 88 (75-100) at 1 year follow-up (p \ 0.05). There were no clinical or radiographic complications.

Fig. 2 Conclusions: The metallic implantation technique seems to be a promising treatment for secondary osteochondral defects of the talus, but more patients and longer follow-up are necessary to draw any firm conclusions.

P20-1191 Functional results after surgical treatment of ankle fractures in athletes: review of 60 cases Vic¸oso Sousa Fernandes S.I.1, Barbosa A.T.1, Fraga Ferreira J.1, Cerqueira R.1, Correia Moreira A.1, Caetano V.1 1 Centro Hospitalar do Alto Ave, Orthopedics and Traumatology, Guimara˜es, Portugal Objectives: In athletes, around 10 to 20% of the lesions occur at the level of the ankle. Professional and amateur athletes, with unstable ankle fractures, treated surgically and submitted to an adequate programme of physical rehabilitation had better functional results. The objective of this study is to evaluate retrospectively the functional results, and, return to sports activity in athletes, after surgical treated ankle fractures. Methods: This study included 60 patients who between January 2000 and December 2007, had unstable ankle fracture while practicing sports activity, and were submitted to open reduction and internal fixation. Exclusion criteria included: \18 years; previous, multiple or exposed fractures; treatment without surgery and non attending follow-up.The patients were evaluated according to demographic, imagiological, clinical and functional variables. The patients were also assessed using the Short Musculoskeletal Function Assessment (SFMA) questionnaire and the American Orthopaedic Foot and Ankle Society (AOFAS) score. Results: There were 16 women and 44 men with medium age of 24,2 years (18-45 years) and with medium follow-up of 2 years (1-6 years). The lesions occur in 8 sports, 50% of the fractures occur while practicing soccer. The fractures were bimalleolar (n=30), medial malleolus (n=11), lateral malleolus (n=10), with sindesmotic lesion (n=6) and trimalleolar (n=3). 6 months after surgery 22% of the patients returned to sports activity ant at 12 months 43%. At 12 months the younger patients (p=0,0001) and men (p=0,001) returned earlier to sports activity. At one year 80% of the amateur and 20% of the professional athletes, had returned to sports practice. Fractures of the lateral malleolus returned earlier in 16,2 weeks than medial malleolus fracture in 59,5 weeks. The SMFA and AOFAS scores were high in all types of fracture. Conclusions: Correct treatment of instable ankle fractures in athletes, with anatomic reduction and preservation of the integrity of the articular surface, is crucial to the return to spots practice. The fractures that influence an earlier return to sports practice were younger age, male sex and less severe fracture, and negative predictors were older age and female sex.

S315 Athletes submitted to open reduction and internal fixation with adequate and precocious programme of physical rehabilitation, can return to the same level of sports practice, despite the seriousness of the fracture without pain and functional limitation.

P20-1203 Endoscopic treatment of calcaneo-fibular impingement after calcaneal fracture Bauer T.1, Deranlot J.1, Hardy P.1 1 Hopital Ambroise Pare´, West Paris University, Boulogne, France Objectives: Calcaneo-fibular impingement is a frequent long term sequelae of calcaneal fractures which can be treated by lateral exostosectomy and fibular tendons release with or without a subtalar arthrodesis by a lateral approach. The purpose of this study was to describe an endoscopic technique for the treatment of lateral calcaneo-fibular impingement with the first clinical and radiological results. Methods: We retrospectively reviewed 7 patients with calcaneo-fibular impingement treated endoscopically. All the seven patients studied had sustained thalamic type calcaneal fractures. There were 6 men and 1 woman and the mean age was 47 years (range 32 to 66 years). Clinically, the calcaneo-fibular impingement was diagnosed by: – Pain under the tip of the fibula increased by walking. – Edema around the tip of the fibula with an objective tender point. – Increased tenderness when a valgus force is applied on the hindfoot. All patients were assessed using the American Orthopaedic Foot and Ankle Society (AOFAS) hindfoot score. All interventions were performed by the same surgeon with a 2-portals technique around the tip of the lateral malleolus. the procedure consisted in a lateral debridement, lateral bone resection, peroneal tendons release and in 2 cases a subtalar joint arthrodesis was associated. A complete mobilization with immediate weight bearing according to pain was allowed from the day after the intervention. In both cases with resection of the lateral impinging bone combined with a subtalar joint arthrodesis, an immobilization with a cast during two months was prescribed. All patients were reviewed with a mean 2 years follow up (7 to 84 months). For all patients, the same clinical evaluation as in preoperative was conducted and a satisfaction survey was answered by the patients (very satisfied, satisfied, poorly satisfied and dissatisfied). Results: The AOFAS score improved in every case (Mean pre-surgical score = 49,5; post surgical=66,4). In all cases, the pain under the tip of the fibula disappeared. In the two patients who answered dissatisfied, pain was elicited when stress valgus forces were applied to the hindfoot. One of the patients required a new endoscopic procedure to remove the screws and complete the resection. One case of hypoesthesia of the lateral side of the foot was observed probably because of a fibular nerve stretching. Conclusions: There are three reasons of lateral impingement. • Bony contact between the lateral bulge of the calcaneus and the tip of the fibula. • Wound Stenosis of the retro-malleolar gutter • Compression of fibular tendons In the majority of cases, the patients were satisfied with the surgery performed by relieving them of their pain that was located at the tip of fibula. The poor functional improvement of the AOFAS score can be explained by the combined lesions (subtalar arthritis, stiffness, etc). The endoscopic treatment of lateral impingement of the ankle after calcaneal fractures seems a good alternative to the conventional lateral approach. The endoscopic procedure allows to perform a sub-malleolar release, a peroneal tenolysis, an exostosectomy and a subtalar arthrodesis if necessary. Endoscopic treatment is a less invasive technique which decreases the risk of wound complications, postoperative pain and therefore a shorter hospital stay.

123

S316 P20-1242 Arthroscopic resection of bony anterior ankle impingement with stiffness: a 13 cases series Bauer T.1, Breda R.1, Hardy P.1 1 Hopital Ambroise Pare´, West Paris University, Boulogne, France Objectives: The purpose of this study was to assess the short-term results on ankle pain and range of motion after arthroscopic treatment of bony anterior ankle impingement associated with tibio-talar joint stiffness. Methods: In this retrospective study, 13 cases with symptomatic bony anterior impingement of the ankle associated with a non tolerated stiffness were included. All the patient underwent surgery with the same arthroscopic procedure consisting in an anterior ankle synovectomy and debridement, osteophytes resection and large anterior capsular and ligament release. Physiotherapy was begun immediately after surgery under analgesic control. Clinical data were assessed preoperatively and postoperatively with the AOFAS ankle score. A satisfaction form was given to the patients at last follow-up. Results: After a mean 15 months follow-up, 10 patients on 13 were satisfied or very satisfied with the outcome and 3 were dissatisfied. Anterior ankle impingement symptoms completely disappeared in 12 cases. Five patients still had deep ankle pain after surgery. The range of motion in dorsal flexion of the ankle significantly improved from a mean 7 preoperatively to a 16 postoperatively (p\0.009). The range of motion in plantar flexion of the ankle significantly improved from a mean 20 preoperatively to a 34 postoperatively (p\0.004). The overall AOFAS score significantly improved from a mean 67/100 preoperatively to a mean 87/ 100 postoperatively (p\0.05). Conclusions: In the precise clinical situation with bony anterior ankle impingement associated with a non tolerated stiffness of the tibio-talar joint, it is possible to improve in the same time the painful symptoms due to the anterior impingement and the range of motion of the ankle joint with a simple arthroscopic procedure consisting in a large anterior synovectomy and debridement, a large anterior capsular and ligament release and a complete resection of anterior osteophytes. Physiotherapy must begin immediately after the procedure with analgesic control during a short hospital stay. The clinical short-term results are encouraging and show a significant functional improvement and an increased ankle range of motion.

P20-1262 Advanced indication for autologous osteochondral graft Usami N.1, Hiraishi E.2, Ikezawa H.1 1 Shiseikai Daini Hospital, Orthopaedic Surgery, Tokyo, Japan, 2 Eiju General Hospital, Orthopaedic Surgery, Tokyo, Japan Objectives: Recently autologous osteochondral graft has been frequently performed for osteochondral lesion of the talus, and are reported with hyaline cartilage regenerated. Utilizing this repair mechanism, we have treated various kinds of ankle disorders. Methods: Subjects were 38 feet (25males and 13 females). Their age at the time of surgery ranged from 23 through 59 years. Osteoarthritis after fracture-dislocation in 12 patients, aseptic necrose after talar fracture in 5,OA after sprain in 17, primary osteoarthritis in 3, and congenital club foot in 1. Affected side was medial side in 14 patients, lateral in 6, and extensive portion including medial and lateral sides in 18 patients. In 14 patients that showed narrowing of the joint space, in 6 patients that showed axial changes the dilatation was performed using an external fixator. In 6 patients that showed axial changes, the low tibial osteotomy and external fixator for opening of joint space. Follow-up period ranged from 1 year through 10 years and 4 months with an average of 4 year and 10 months. Results: Pain was alleviated in all the patients. One patient was unable to do sports, and 2 patients developed pain by one hour walking. There were no patients with limited range of motion of the joint. AOFAS scores were improved 62 points 90 points. On second look of the grafted bones that was performed showed survival of the grafted bones in all the patients, and formation of hyaline cartilage was observed. However, there was fibrous

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 cartilage between grafted bones, and synovial tissue was increased. There were no patients with increased horizontal diameter of the grafted bones. Conclusions: There are no reports describing that autologous osteochondral graft was performed for osteoarthritis or bone necrosis after trauma. This study showed excellent results for deformity or collapse of the talus. But in patients with severe failure of anatomical structure, pain remained. In patients who underwent corrective osteotomy, good results were obtained. So, in autologous osteochondral graft, it is necessary to restore the anatomical structure to the ordinary state.

P20-1352 Arthroscopy assisted surgical technique for retrograde autologous osteochondral grafting in talar osteochondritis dissecans: a cadaveric study Wajsfisz A.1, Naji O.1, Lintz F.1, Beaufils P.1 1 Centre Hospitalier de Versailles, Hopital Andre Mignot, Le Chesnay, France Objectives: Various techniques exist for the surgical treatment of medial talar dome osteochondral lesions. In both open and arthroscopic approaches, debridement and anterograde drilling can be performed. In case of unstable or displaced fragments, osteochondral transplantations (mosaicplasty) can be performed through a medial malleolar or anteromedial tibia osteotomy to reconstruct the defects. These procedures sacrifice tibial joint surface, which certainly affects the integrity of the joint. Approaches using a medial malleolar osteotomy allow an excellent exposure, but add significant morbidity to the procedure. Open techniques without medial malleolar osteotomy give limited access to the posterior half of the medial talar dome, and the direction of the drill is far from being orthogonal to the joint surface. We developed an arthroscopically assisted retrograde grafting technique for osteochondral lesions of the medial talar dome. The goal of this study is to describe the surgical technique and the results in cadaver experiments. Methods: A 3 step cadaveric study was performed. First, we simulated the procedure on saw bone models. Second, we performed the procedure on 2 cadaveric specimens on which ankles had been disjointed to study approaches and to develop the surgical technique. During the third step, we performed thearthroscopically assisted procedure : 9 ankles on seven cadaveric specimens conserved in formaldehyde/phe´nol were used. The procedure was based on mosaicoplasty autograft techniques, using only one osteochondral plug. First, arthroscopic examination of the posteromedial aspect of the talar dome was performed through standard antera` lateral and medial portals. An osteochondral plug 8 to 10mm in diameter and 20mm long was harvested through a short incision on the proximal and lateral trochlea, so that its chondral surface was orientated 30 degrees obliquely. A tunnel was drilled through the lateral talar tuberosity, through theposteromedial aspect of the talus using a core reamer. The osteochondral plug was inserted under arthroscopic control by retrograde approach, through the tunnel. Our objective was to obtain a complete procedure defined by correct joint congruency and positioning of the osteochondral graft. Results: The procedure was performed in seven cadaveric ankles out of nine. Two procedures were aborted because of an inadapted ancillary equipment. In all the other cases, we found no malpositioning of the graft. The position was medial in 5 cases, central in 2 cases. The joint congruency was satisfactory in all cases. We did not find any cartilage damage on the tibial roof except for superficial fraying. The subtalar joint capsule were opened during the procedures, which showed that the cartilage surfaces had been preserved. Conclusions: Due to the anatomic location of most osteochondral talar dome lesions, surgical exposure is either limited or involves significant comorbidity (malleolar osteotomy). We present a novel, arthroscopically assisted technique for the treatment of medial osteochondral lesions of the talus. A limitation of our technique may be that we can graft only one plug into the medial talar dome through a retrograde tunnel. Therefore this technique is limited to moderately sized defects. Osteochondral retrograde grafting of the medial talar dome under

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 arthroscopic control is therefore feasible but requires material upgrading and a technically demanding learning curve.

P20-1445 Reconstruction of lateral stability of the ankle joint by modified Karlsson technique Saji T.1, Sano T.1, Suzuki T.2, Matsuoka H.1 1 Shizuoka General Hospital, Department of Orthopaedic Surgery, Shizuoka, Japan, 2Osaka Red Cross Hospital, Osaka, Japan Objectives: There are several surgical procedures described for treatment of chronic lateral instability of the ankle joint. These days, anatomical reconstruction methods are recommended. We modified the anatomical reconstruction technique reported by Karlsson et al., using bioabsorbable suture anchors. Methods: Six patients with 6 ankles were treated operatively. The time between the original injury and the surgery was more than one year. Before reconstruction, we performed ankle arthroscopy for all the cases. A bone block was chiseled away from the anterior, inferior border of the lateral malleolus. Two of three bioabsorbable suture anchors were placed into the gutter and the lateral ankle ligaments were pulled and reinserted in the gutter. The mean postoperative follow-up period was 17.8 months. Results: Ankle arthroscopy revealed cartilage injury in all cases. In one case, international cartilage repair society (ICRS) score was grade 4, and in the other cases, grade 1 or 2. One patient had mild pain after the surgery, but the others did not feel pain. One had recurrence because of another sprain. All recovered full range of motion. All patients were satisfied with the result. Conclusions: Karlsson et al. have reported good result of their anatomical reconstruction method. The advantage using absorbable anchors was that it is easier to suture lateral ankle ligaments securely.

P20-1465 Arthroscopic treatment of a talar neck fracture: a case report Wajsfisz A.1, Guillou R.1, Lintz F.1, Beaufils P.1 1 Centre Hospitalier de Versailles, Hopital Andre Mignot, Le Chesnay, France Objectives: Fractures of the talus are rare and associated with high levels of complications. The gold standard treatment is an open reduction and internal fixation through anterior approach. The most frequent complication is post-traumatic talar avascular necrosis, due to poor vascular supply of this bone. The type of fracture and the degree of displacement are important risk factors, as well as the surgical approach which reduces talar vascularisation even more. Moreover with open procedure we cannot control subtalar joint after reduction wich can cause osteoarthritis in case of malunion. We propose here an alternative arthroscopic approach of the fractures to better respect the talar vascular supply and control subtalar joint. Methods: We report the case of a Hawkins type II talar fracture (displaced talar neck fracture with sub talar disjunction), in a 16 year old woman after falling out of a window. A closed reduction was performed at the Emergency Room after standard X Rays. A CT-Scan was also done to analyse the fracture and propose an appropriate treatment strategy. The fracture presented two main fragments and a third one incarcerated in the sub-talar joint. The patient was operated with the leg in prone position. Arthroscopy was carried out using two lateral sub-talar approaches and two standard antero medial and lateral approaches. A fluoroscopic control was used during surgery. First, the sub talar joint was explored, hemarthrosis evacuated and the third fragment debrided so as to allow correct reduction. Secondly, the tibio-talar joint was explored and fracture reduction was controlled and fixed using two divergent percutaneous K-wires. Two spongious, 7mm cannulated screws were placed using the K-wires to obtain compression of the fracture. After reduction, subtalar was controlled again to permit to debrided a new fragment.

S317 Post-operative management included immobilization using a simple plaster cast during 45 days, changed to a splint for passive rehabilitation. Weight bearing was not allowed for 4 months. Results: There was no soft tissue, vascular or nervous complication and there was no X-ray sign of avascular necrosis at 6 months follow up. The patient was able to walk without pain. Conclusions: Closed reduction and internal fixation is technically feasible under arthroscopic control in talar neck fractures, enabling a precise reduction and stable fixation. This approach is less invasive than open reduction, therefore it could reduce risks of avascular necrosis. Moreover, the subtalar joint control at the end of the procedure permit to fixe it if necessary

Foot P21-94 Treatment of an unicameral bone cyst of calcaneus with endoscopic curettage and percutaneous filling with corticocancellous allograft: a new technique Yildirim C.1, Mahirogullari M.2, Akmaz I.2, Kuskucu M.2 1 Tatvan Military Hospital, Orthopaedics and Traumatology, Bitlis, Turkey, 2Gata Haydarpasa Training Hospital, Orthopaedics and Traumatology, Istanbul, Turkey Objectives: We report a case wherein a different surgical intervention was used to treat a symptomatic unicameral cyst of the calcaneus with endoscopic curettage and percutaneous filling of the lesion with corticocancellous allograft. Methods: A 21-year-old man presented with pain in the left heel. Plain radiographs and computerized tomography revealed a unicameral cyst of the calcaneus. An endoscopically assisted technique is proposed for the curettage of a simple calcaneal bone cyst that takes advantage of direct visualization of the cyst wall and contents and permits accurate assessment of the extent of the lesion. After curettage, percutaneous filling of the defect with corticocancellous allograft makes the technique a complete minimally invasive surgical approach. This technique uses only two lateral portals (viewing lateral and working lateral portal). The portals are created under fluoroscopic control. Once the cyst has been located, the blunt trocar is inserted through the viewing lateral portal creating the first hole in the lateral wall of the calcaneus. The 30 arthroscope is placed into the cyst through this hole. Next, the second portal is created as the same way. Through these portals, the fluid contents of the cyst were evacuated, followed by fragmentation and removal of the cyst contents, and subsequent curettage of the inner surface of the cavernous cyst wall. Finally, complete packing of the previously cystic cavity with crushed corticocancellous allograft is performed under endoscopic visualization. Results: After two years, no recurrence was noted and the patient was symptom free. Conclusions: Simple bone cysts of the calcaneus are relatively uncommon. There is no clear consensus on either their etiology or management. In most cases, these lesions are asymptomatic and can be treated conservatively. The traditional method of treating unicameral cysts has been curettage with or without bone grafting. Our approach to the surgical treatment of a unicameral calcaneal cyst appears to be unique in the literature.

P21-121 Fracture/dislocation of the talus in the sequence of an equestrian trial - case report Vic¸oso Sousa Fernandes S.I.1, Barbosa A.T.1, Fraga Ferreira J.1, Cerqueira R.1, Basto T.1, Vasconcelos P.1, Lourenc¸o J.1 1 Centro Hospitalar do Alto Ave, Orthopedics and Traumatology, Guimara˜es, Portugal Objectives: The fractures at the talus are unusual, and represent about 3% of the total of foot fractures. Fractures without deviation of the talus can be treated without surgery, but if deviation is present they should be treated with surgery. The major complications of this type of lesion are avascular necrosis and secondary arthrosis of the joints involved.

123

S318 Methods: In this work we describe a case of fracture/dislocation of the talar neck with posterior dislocation. Results: Case report: It concerns a 19-years-old, who was victim of a fall from a horse in an equestrian trial and had left ankle trauma in 10/04/08. At the initial observation in the Emergency Room the patient had pain, swelling and deformity of the left ankle. Radiographs showed fracture of the talar neck, with posterior dislocation and rotation of the talar body type III of Hawkins Classification. The patient was submitted to surgical treatment 4h after the accident with anterolateral approach, reduction of the dislocation and fixation of the fracture with 2 parallel malleolar screws, and was immobilized with a cast during 6 weeks and posteriorly initiated physical therapy. The patient was evaluated in the postoperative period at 6 weeks, 3 months, 6 months and 9 months. At 6 months follow-up the patient walk without crutches, and had 0-358 de plantar flexion and 0-10 of dorsiflexion and with normal muscular force. At 9 months follow-up the fracture had healed and had no signs of avascular necrosis at the imagiological study. The patient didn’t had any restrictions in daily activities and returned to work has a secretary. He had maximal score 100 in the American Orthopaedic Foot and Ankle Society Scale for clinical evaluation of the ankle and hindfoot. Conclusions: Open surgical urgent reduction and internal fixation of a fracture/dislocation of the talus type III of Hawkins Classification have an important role in the prevention of the talus avascular necrosis and of the other complications that might occur in the sequence of this type of lesion. The prevention of these lesion sequels is necessary to maintain normal gait, normal sagittal and hindfoot mobility, normal ankle and hinfoot stability, normal alignment and normal life activities.

P21-431 Innovations in medical image processing for the design of custom implants and surgical guides Willems D.1, Boelen E.1 1 Materialise, Leuven, Belgium Objectives: There is a growing trend towards personalization of medical care, as evidenced by the latest developments in multislice CT imaging and ultra-fast MR imaging, personalized treatment planning in a variety of surgical disciplines and the development of more suitable implantable devices. To support this trend, the role of the biomedical engineer becomes increasingly important, as the operating theater becomes more and more a technical environment. Hospitals need multidisciplinary teams for the development of diagnostic tools, implants and tissue engineered materials, computer assisted surgery and rapid product development & virtual process simulation processes. Methods: Here, we will describe the use of 3D medical image information of individual patients as well as selected patient populations, combined with CAE tools and processes, in the rapid product development of custom and standard implantable devices. The combination of medical image information with CAE methods such as CAD, RP, FEA, CFD, Forward Engineering allows the engineer to develop implantable devices faster and better, with optimized designs tailored to the anthropometry of the targeted patient (population), using virtual instead of mechanical prototype testing. Results: Case studies from leading biomedical research institutes as well as medical device manufacturers will be demonstrated for a variety of orthopaedic subdomains such as knee, shoulder, foot and hip. Conclusions: The case studies clearly show the advantage of CAE methods on 3D medical image information in the development of custom and standard implantable devices. Keywords: CT/MRI, image processing, rapid product development, medical devices, custom design, patient-specific implant

P21-440 Ankle and subtalar joint range of motion following two lateral column lenghtening procedures for flexible flatfoot deformity Beimers L.1, Louwerens J.W.K.2, Tuijthof G.1, Jonges R.3, van Dijk C.N.1, Blankevoort L.1

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 1

Academic Medical Centre, Orthopaedic Surgery, Amsterdam, The Netherlands, 2Sint Maartenskliniek, Orthopaedic Surgery, Nijmegen, The Netherlands, 3Academic Medical Centre, Medical Physics, Amsterdam, The Netherlands Objectives: Lateral column lenghtening (LCL) has become an accepted procedure for the treatment of flexible adult acquired flatfoot deformity. The effect of LCL on ankle and subtalar joint kinematics is not clear. Residual ankle and subtalar joint range of motion was studied in patients who underwent an anterior open wedge calcaneal osteotomy (ACO) or a calcaneocuboid distraction arthrodesis (CCDA) for symptomatic flexible flatfoot deformity. Methods: Computed Tomography (CT) scanning was performed with the foot in eight extreme positions in five ACO and five CCDA patients. In both groups comparable additional soft tissue procedures had been performed during surgery. A bone segmentation and registration technique was applied to the distal tibia, talus and calcaneus in the CT images. Finite helical axis (FHA) parameters representing the range of motion of the ankle and subtalar joint were calculated for the motion between two opposite extreme foot positions of the tibia relative to the talus and calcaneus relative to the talus. A reference data set from a group of 20 normal non-matched subjects was used for comparison. Results: The range of motion of the ankle joint (extreme foot dorsiflexion to extreme foot plantarflexion) after ACO and CCDA was comparable in both groups (FHA mean rotation ACO 52.2 ± 12.4 degrees; CCDA 49.0 ± 12.0 degrees). The ankle joint motion following CCDA was statistical significantly reduced (p\0.05) when compared to the normal subjects (FHA mean rotation 63.3 ± 11.0 degrees). Subtalar joint range of motion (extreme foot eversion to extreme foot inversion) was reduced following both LCL procedures (FHA mean rotation ACO 22.8 ± 8.6 degrees; CCDA 24.4 ± 7.6 degrees) and differed significantly (p\0.05) from the normal subjects (FHA mean rotation 37.3 ± 5.9 degrees). FHA translations for the ankle and subtalar joint ranged from 0.0 to 2.2 mm with large variation. Conclusions: After an ACO or CCDA procedure for flexible flatfoot deformity, no difference could be detected in the ankle or subtalar joint range of motion. Subtalar joint range of motion after an ACO and CCDA procedure is smaller than that of normal subjects. The ACO procedure is not per definition the LCL procedure of choice for symptomatic flexible adult acquired flatfoot deformity as subtalar joint motion is not better preserved in comparison to the CCDA procedure.

P21-460 Subtalar joint arthrodesis using a 3-portal minimal invasive approach for talocalcaneal coalition Beimers L.1, de Leeuw P.1, van Dijk C.N.1 1 Academic Medical Centre, Orthopaedic Surgery, Amsterdam, The Netherlands Objectives: A symptomatic talocalcaneal coalition can be managed by a subtalar joint arthrodesis. Arthroscopic subtalar arthrodesis is technically challenging in patients with a talocalcaneal coalition as opening of the subtalar joint is restricted and the available workspace during surgery is reduced. We present a 3-portal minimal invasive approach for subtalar joint arthrodesis in patients with a talocalcaneal coalition. Methods: With the patient in the prone position, routine posterolateral and posteromedial portals are created for subtalar joint access. An accessory sinus tarsi portal allows for introduction of a large diameter blunt trocar for subtalar joint distraction. Following resection of the talocalcaneal coalition, all articular cartilage is removed from the posterior subtalar joint. Longitudinal grooves are cut in the subchondral bone to enhance fusion of the arthrodesis. Using fluoroscopy, two 6.5-mm lag screws are placed across the posterior subtalar joint. For post-operative treatment, a non-weightbearing lower leg cast is provided for 4 weeks followed by a walker boot. Results: Three female patients with a painful talocalcaneal coalition refractory to conservative treatment were operated using the technique as described. Bony union was seen on radiographs 6 weeks post-operatively. At time of follow-up (range 24-28 months), all patients were free of pain with unrestricted ambulation and all were satisfied with the result. No complications had occurred.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Conclusions: For arthroscopically assisted posterior subtalar joint arthrodesis in talocalcaneal coalition, sufficient subtalar joint distraction can be achieved by using a blunt trocar via an accessory sinus tarsi portal. The 3-portal minimal invasive approach was successfully used in three subsequent patients with a talocalcaneal coalition. With the 3-portal approach, a safe arthroscopic subtalar arthrodesis can be performed even in cases with limited subtalar joint space such as in talocalcaneal coalition.

P21-580 Percentage body-weight/weight-bearing (PBW/WB) values in increasing speeds from walking to running Kaplan Y.1 1 Jerusalem Physical Therapy and Sports Institute, Lerner Sports Center, Hebrew University, Jerusalem, Israel Objectives: Orthopedic surgeons and rehabilitation physicians frequently request limited weight-bearing for prolonged periods following certain bony or soft tissue pathologies as well as certain lower - limb surgical procedures. Jogging is a common activity that the injured athlete will seek to return to as soon as possible following injury or surgery. In order for the physician to be able to advise regarding weight-bearing in jogging, the PBW/WB values in increasing walking and jogging speeds must be known. These parameters as well as the gait distribution changes have eluded the rehabilitation community, mainly due to the technical inability to measure these paradigms. Methods: A revolutionary weight-bearing and gait analysis system (SmartstepTM was utilized to accurately measure the PBW/WB values and gait distribution patterns in a sample of 10 asymptomatic subjects between the ages of 18-36 years of age (Average = 27). The test measurements were conducted on a treadmill (TechnogymTM ‘‘run excite 700’’), in order to accurately control the speed changes. The tested speeds were 4,6,8,10,12,14,16,18 and 20 km/hr consecutively. Each subject ran for a 15 second time period for each speed. Results: Whilst analyzing PBW in the entire-foot, as the speed increases, there is almost no change in the PBW from 12km/hr onwards. As speed increases from 4km/hr to 20km/hr, the PBW more than doubles itself. As opposed to the PBW on the entire-foot, the PBW on the fore-foot increases to a maximum of 70%. The maximum fore-foot PBW is 30% higher to that of hind-foot. As the speed increases, the stance percentage reduces by 100% at 20km/hr with a concurrent increase in the swing percentage. The swing phase only increases by 6% from 10km/hr to 20km/hr, even though the speed increases by 100%. The main change occurs between 8km/hr and 10km/hr, as walking progresses to running. Conclusions: The most significant change in PBW occurs between 4km/hr and 12km/hr, as walking progresses to running. Therefore, this is what should be considered when recommending weight bearing activities to the athlete who intends returning to weight bearing following injury or surgery. Athletes can be encouraged to run at high speeds without concern of greatly increasing their PBW. The results should be cautiously interpreted until verified with ground reaction force studies.

P21-731 The new technique for the endoscopic partial plantar fascia release for the treatment of plantar fasciitis Innami K.1 1 Teikyo University, Department of Orthopaedic Surgery, Tokyo, Japan Objectives: Planter fascia release has been suggested to be of benefit for patients with symptoms of chronic unresponsive planter fasciitis. Endoscopic planter fascia release has been described by several authors. There methods were not able to resect a heel spur and release planter fascia unreliable. The purpose of this study was to investigate the surgical treatment outcome and limitation of the new technique for the endoscopic partial plantar fascia release for the plantar fasciitis. Methods: 6 feet of 5 patients that had undergone the endoscopic partial plantar fasciotomy were followed at least 6 months. All patients had undergone a conservative treatment for a minimum of 6 months. Their mean age was 35.8 years (range 22-66) at the time of surgery. There were

S319 3 men and 2 women. 3 patients were considered athletically active. 3 feet of 3patients had a heel spur. Mean follow-up period was 8.2 months (range 6-13). We performed the endoscopic surgery under spinal anesthesia. A medial portal is developed at the insertion of plantar fascia into calcaneus using the imageintensifire, and the endoscopic cannula was introduced into the area created at the anterior to the calcaneal tubercle and the superior to planter fascia. The lateral portal established by the inside-out technique. An arthroscopic shaver was used for visualization, taking care to shave soft tissue, synovium and a part of the flexor digitrum brevis. The heel spur was resected before releasing the plantar fascia using an arthroscopic burr. After exposing the planter fascia, the medial half of it was released using electric devices. After transaction, the cut ends of the planter fascia were examined as well as the heel fat pad. One week after operation, patients were allowed immediate full weight bearing on the foot. Results: There were no complications at final follow-up after surgical treatment. The mean AOFAS score was 62.2±10.0 at pre-operation and 91.4±7.6 at the final follow-up. The follow-up AOFAS score was better than that at pre-operation (p\0.05). It took mean 10.4 weeks (range 8-16) for walking with no pain. The mean return to sports activity time was 10.3 weeks (range 9-12). Conclusions: A short term outcome was favorable for the new endoscopic partial plantar fasciotomy. We regarded that the endoscopic partial plantar fasciotomy was less invading operation and postoperative pain was little. In fact, the return to sports activity time was 10.3 weeks and there was the limitation of early sports returning.

P21-1386 Assessment of plantar pressure in rheumatoid arthritis patients by static pedobarography Sagnak E.1, Euler G.1, Timmesfeld N.2, Wolf U.3, Fuchs-Winkelmann S.4, Skwara A.5 1 Universita¨tsklinikum Giessen und Marburg GmbH, Department of Orthopedics and Rheumatology, Marburg, Germany, 2Philipps-University Marburg, Institute for Medical Biometry and Epidemiology, Marburg, Germany, 3Universita¨tsklinikum Giessen und Marburg GmbH, Department of Physical Therapy, Marburg, Germany, 4University Hospital Marburg, Orthopaedic Surgery, Marburg, Germany, 5University Hospital Giessen and Marburg, Department of Orthopaedics and Rheumatology, Marburg, Germany Objectives: Rheumatoid arthritis (RA) patients often are impaired by deformities in the feet that may cause pain and immobility. For a better understanding of this connection, the present study assessed and evaluated plantar pressure characteristics of patients with rheumatoid arthritis during static stand. Methods: 82 patients suffering from rheumatoid arthritis (RA) with foot involvement and 80 healthy controls were included. Plantar pressure and its allocation between right and left foot as well as front and back part of foot were analyzed with Zebris pedobarographic system. Measurements were taken during bipedal as well as left and right monopedal stand. In RA patients, the Health Assessment Questionnaire (HAQ) and Foot Function Index (FFI) were administered to assess impairment and disability. Foot pain was examined clinically. To evaluate radiographs, the van-Heijde-modification of the established Sharp score was used. Results: In bipedal standing of RA patients, no significant difference in the allocation of plantar pressure between right and left foot, defined as deviation from an equal allocation, was found. The plantar pressure on the back part of the foot in bipedal standing did not differ between both groups. In single-legged standing, RA patients showed a significantly higher foot loading of their back part of foot (p \ 0.01). Conclusions: Pedobarography seems to be a useful tool to describe each patient’s individual plantar pressure characteristics. The higher plantar pressure values of RA patients’ back part of foot in monopedal standing, compared to healthy controls, might be a result of a disburdened front part of foot caused by deformities and pain, especially with less standing balance while standing on just one foot. Orthopaedic sole and shoes may improve the imbalance.

123

S320 P21-1393 Proprioception in the upper ankle joint in rheumatoid arthritis patients Sagnak E.1, Wingen S.1, Wolf U.2, Timmesfeld N.3, Fuchs-Winkelmann S.4, Skwara A.5 1 Universita¨tsklinikum Giessen und Marburg GmbH, Departement of Orthopedics and Rheumatology, Marburg, Germany, 2 Universita¨tsklinikum Giessen und Marburg GmbH, Departement of Physical Therapy, Marburg, Germany, 3Philipps-University Marburg, Institute for Medical Biometry and Epidemiology, Marburg, Germany, 4 University Hospital Marburg, Orthopaedic Surgery, Marburg, Germany, 5 University Hospital Giessen and Marburg, Department of Orthopaedics and Rheumatology, Marburg, Germany Objectives: The mobility of patients with rheumatoid arthritis (RA) of the foot is often impaired due to arthritis and erosive destruction of the foot and ankle joints. Standing and gait balance might be reduced due to decreased proprioception. The aim of the present study was to investigate the proprioception of the upper ankle joint in rheumatoid arthritis patients. Methods: In 82 patients with rheumatoid arthritis with foot involvement and 80 healthy controls the stance balance was examined by Zebris pedobarographic system. The center of mass was determined while bipedal as well as left and right monopedal stand. During the 10 seconds of measurement the movements of the center of mass were recorded and displayed as an ellipse. Foot pain was assessed by a self-rating scale. The Health Assessment Questionnaire (HAQ) and Foot Function Index (FFI) were used to evaluate patients’ impairments. The van-Heijde-modification of the established Sharp score was used to assess the grade of erosion in radiographs. Results: In RA patients with foot involvement, the area of the ellipse from bipedal standing was significantly increased compared with the control group (P‹0,03). In monopedal standing the area of the ellipse was very significantly larger in RA patients with foot involvement compared with the control group (P‹0,001) on the left as well as on the right foot. Conclusions: The increased area of the ellipse may be indicative of an impaired proprioception in the upper ankle joint in patients with RA. Impaired proprioception causes reduced standing and gait balance which influences the mobility of RA patients negatively.

P21-1455 Managment of complex trauma of the foot in pediatric patients Eberl R.1, Singer G.1, Hoellwarth M.E.1, Weinberg A.1 1 Medical University of Graz, Department of Pediatric and Adolescent Surgery, Graz, Austria Objectives: Complex injuries of the foot in the paediatric population present difficult treatment challenges. While standardized protocols exist for the adult population to achieve an optimal result in the treatment of such injuries, therapy in paediatric patients must be managed without a firm treatment algorithm. Methods: The medical records of all patients with a complex trauma of the foot treated at our Department over a period of 13 years were evaluated. A complex trauma of the foot was defined using the scoring system developed by Zwipp. We developed a simple algorithm for the treatment of complex injuries of the foot in paediatric patients. After a careful clinical examination and interpretation of the conventional X-rays in two planes the course of treatment was determined according to the underlying injury. Patients with compartment syndrome and open fractures were brought immediately to the operation room. Initial operative interventions were kept to a minimum. In case of closed fractures further diagnosis with CT scan was performed and definitive treatment realized with delay depending on the situation of the soft tissues. Treatment and outcome were analyzed. Results: Twenty-nine patients were included in the study (79% m; 21% f, average age 12.1 years, ranging 2-16 years). Traffic accidents were the most common mechanism (n=14; 48.3%), followed by a fall from a

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 height in five patients (17.2%). Lawn mower injuries were found in another 3 patients (10.3%) and other mechanisms of injury in 7 patients (24.2%). The mean score according to Zwipp was 5.8 points (range 5-8 points). While closed fractures were diagnosed in 20 (69%) patients, 9 patients (31%) presented open fractures. Operative intervention was necessary in 24 patients (82.8%). Fracture stabilisation could be realised using K-wires in 13 cases (54.2%), screws in 3 cases (12.5%) and plate fixation in one case (4.2%). Combined techniques including external fixation were applied in another 7 (29.1%) cases. The mean time between injury and latest follow up examination was 5.7 years (range 13 months to 13 years). The mean functional outcome was 47.6 (29-56) points for the OAFQ, 15.1 (0-69) points for the FFI and 82.3 (59-100) points for the AOFAS Score. Conclusions: A complex trauma of the paediatric foot is a rare and challenging injury. To regard the maxims in treating complex injuries and open fractures in the growing skeleton we developed a simple treatment algorithm for complex foot injuries in order to provide preservation of the soft tissue envelope, avoidance of infection, restoration of the axis and the articular surface. Despite the severity of trauma we found a favorable outcome in the majority of cases. Long time follow-up is essential to detect complications.

Muscle

P22-648 Proximal avulsion of all three hamstring tendons. Correlation between the result of operative treatment and the delay to surgery, the age of the patient, and the number of suture anchors used Sarimo J.1, Lempainen L.2, Mattila K.3, Orava S.4 1 Mehila¨inen Sports Trauma Research Center, Mehila¨inen Hospital and Sports Clinic, Turku, Finland, 2Department of Orthopaedic Surgery and Traumatology, University Hospital of Turku, Turku, Finland, 3 Medical Imaging Centre of Southwest Finland, University Hospital of Turku, Turku, Finland, 4Mehilainen Hospital, Turku, Finland Objectives: A complete (three tendon) proximal hamstring avulsion is a serious injury that can cause considerable morbidity and it is often associated with significant functional loss. In a complete avulsion of a tendon the muscle becomes non-functional. If the continuity is not restored the muscle starts to atrophy and fatty degeneration occurs. In previous studies no difference in the result of surgery has been found between cases operated in the acute versus the chronic phase. In our study we wanted to evaluate whether early surgical treatment (within 3 months from injury to surgery) in complete proximal hamstring avulsions would result in a better outcome than surgery in the chronic phase. We also evaluated whether the age of the patient or the number of suture anchors used in surgery would have an effect on the result. Methods: The study consisted of forty-one patients with complete three tendon avulsions of the proximal hamstrings. All patients were treated surgically and suture anchors (2 to 4) were used to reattach the torn tendons to the ischial tuberosity. The cases were retrospectively analyzed and statistical correlations were calculated between the result and the delay from injury to surgery, the age of the patient as well as the number of suture anchors used. The average follow-up in the study was 37 months. Results: The delay from the injury to surgery averaged 2.4 months in the patients that were rated as having excellent or good results. In the group of patients with a moderate or poor result the delay averaged 11.7 months. The difference was statistically significant (P \ 0.001). Statistically significant differences between these two groups (excellent/good vs. moderate/poor) were not found in regarding the age of the patient or the number of suture anchors used. Conclusions: Early operative treatment (within 3 months from the injury) of proximal complete three tendon avulsions of the hamstrings leads to better results than late surgery and is therefore recommended.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 P22-916 Ultrasonographic findings in 51 football players with acute hamstring injuries Petersen J.1, Thorborg K.1, Nielsen M.B.2, Skjødt T.3, Bolvig L.4, Bang N.4, Ho¨lmich P.1 1 Amager Hospital, University of Copenhagen, Department of Orthopaedic Surgery, Copenhagen, Denmark, 2Rigshospitalet, University of Copenhagen, Department of Radiology, Copenhagen, Denmark, 3 Vejle Hospital, Department of Radiology, Vejle, Denmark, 4 Aarhus Sygehus, Aarhus University Hospital, Department of Radiology, Aarhus, Denmark Objectives: Hamstring injury is the most common injury in association football (soccer) and accounts for 12-16% of the total number of injuries. Most injuries are grade I and II injuries to the musculotendinous junction. Full ruptures or avulsions (grade III) are seldom. Because of the accessibility and reduced cost ultrasonography has in the recent years been increasingly used in musculoskeletal imaging. It is therefore suspected that an increasing number of hamstring injuries will be investigated using ultrasonography. Hence, it is important to know the expected findings and the practical consequences of these. The aims of this study were therefore to investigate the characteristic sonographic findings in acute hamstring injuries in football players. Furthermore, we wanted to compare the mean injury severity in injured players with and without sonographic verified abnormalities. Finally we wanted to correlate the size of the injured area and injury severity in order to investigate if ultrasonography can be used in the prognosis of return to pre-injury level. Methods: Players from 50 teams participating in one of the top-five Danish football divisions were followed in the period from January to December 2008. A total of 51 players sustained an acute hamstring injury and underwent an ultrasound examination of the injured thigh and were included in this study. Results: The ultrasound examinations were performed 1-10 days after the injury (mean 5.2 days, SD 3.0). Of the 51 performed ultrasound examinations rupture, haematoma and/or oedema were found in 31 (61%) cases. None full rupture injuries (grade III) were documented. The 51 acute hamstring injuries resulted in 6-74 days absence from football per injury (mean 25.4, SD 15.7). No significant differences in injury severity between players with or without US verified abnormalities were documented: rupture (p=0.91); haemorrhage (p=0.41); rupture and haemorrhage (p=0.36); oedema (0.54); verified pathology defined as rupture, haemorrhage or oedema (0.36). Statistic significant correlations existed between injury severity and length (r=0.42, p=0.04) and size (r=0.49, p=0.02) of rupture. Conclusions: This study questions the relevance of ultrasonography as a tool to predict recovery time in athletes with grade I or II hamstring injuries.

P22-1056 Mechanical therapy for postoperative loss of knee extension Dempsey A.1, Branch T.1, Mills T.1, Karsch R.1 1 University Orthopaedic Clinic, Decatur, GA, United States Objectives: Postoperative flexion contractures have been associated with increased pain and a reduced ability to perform activities of daily living. High-intensity home mechanical therapy is often used to treat flexion contractures; however, few studies have evaluated the efficacy of this treatment. The purpose of this retrospective study was to determine if adjunctive home mechanical therapy improved the degree of knee extension for patients with postoperative flexion contractures. Methods: Forty-three patients (12 women, 31 men, age = 55 ± 14.9 years) with postoperative flexion contractures were treated with high-intensity home mechanical therapy in addition to outpatient physical therapy. Mechanical therapy was only prescribed for those patients whose motion had reached a plateau when treated with physical therapy alone. Patients were asked to perform six 10-minute bouts of end-range stretching per day with the device. Passive knee extension was recorded during the postoperative visit that mechanical therapy was prescribed, 3 months after beginning mechanical therapy, and at the most recent follow-up. We used

S321 a repeated measures ANOVA to evaluate the change in passive knee extension over time, with p\.05 considered significant. Results: Passive knee extension deficits were significantly improved from 10.4 ±5.4 at the initial visit to 2.4 ± 3.1 at the 3 month visit (p\0.001). The degree of extension was maintained at the most recent follow-up (2.2± 3.2), which was also significantly greater than the initial motion (p\0.001). The average length of follow-up was 8.3 months. Conclusions: Previous studies have reported that patients with postoperative flexion contractures demonstrated significantly decreased outcome scores for pain, walking, stair-climbing, and over-all function. By limiting a patient’s ability to properly accept weight during gait, flexion contractures of even one degree have been reported to negatively affect clinical outcomes. We conclude that adjunctive high-intensity mechanical therapy resulted in significantly improved knee extension, which was maintained at the patients’ most recent follow-up. Future studies will determine if sustained gains in knee extension as a result of adjunctive mechanical therapy improve long-term outcomes for these patients.

P22-1075 Triceps brachi muscle tears Orava S.1, Sarimo J.2, Lempainen L.3, Heikkila¨ J.4, Rawlins M.1 1 Mehilainen Hospital, Turku, Finland, 2Mehila¨inen Sports Trauma Research Center, Mehila¨inen Hospital and Sports Clinic, Turku, Finland, 3 Department of Orthopaedic Surgery and Traumatology, University Hospital of Turku, Turku, Finland, 4SportsClinic and Hospital Mehila¨inen, Turku, Finland Objectives: This report is a retrospective series of surgically treated triceps brachi muscle tears in athletes. Different types of muscle - tendon tears are described and different surgical repair methods are discussed together with results of surgery. Methods: During years 1998 - 2008 eleven triceps muscle tears were treated surgically. The location of the tears as well as the injury mechanisms were cleared. Different types of muscle tears required different surgical repair methods. The end results were asked by mail, phone or personal contact approximately 6 years after surgery (1.5 to 9 years). Results: There was one tear (avulsion) proximally at the scapular origin of the long head of triceps muscle. There were 4 tears at the muscle tendon area in the body of the muscle, 4 tears distally in the tendinous part proximal to olecranon and 2 avulsion tears from olecranon. The sports represented were: power lifting 7, weight lifting 2 and accidental 2. The two last injuries occurred to musculous individuals outside gym, in falling and pushing heavy load at work. Reinsertion with suture anchors of the torn / avulsed tendon was done in 5 patients and saturation of the muscle tendon in 6 patients. Once superficial wound infection occurred postoperatively. It was treated successfully with antibiotics. In three cases considerable muscle atrophy remained in tears of the muscle mass. In 3 cases excision of the prominent olecranon bony spur was done during the surgery. The end result was excellent or good in 7 cases, moderate in 3 cases and poor in one case. In the last case there was atrophy, nerve pain and strength loss that stopped the power training of the athlete. Conclusions: Triceps muscle tears are rare muscle injuries. They are tendon avulsions or tears or ruptures of the muscle tendon area. These are difficult to find early, they may become chronic and the delay of surgery may be long. Tears can be treated successfully with proper surgical methods.

Tendon P23-50 Acute microcirculatory effects of topical glyceryl trinitrate on the previously ruptured Achilles tendon Knobloch K.1, Osadnik R.2, Vogt P.1 1 Hannover Medical School, Plastic, Hand, and Reconstructive Surgery, Hannover, Germany, 2Orthopedic Surgery, Aachen, Germany Objectives: Topical glyceryl trinitrate treatment has demonstrated short- to mid-term efficacy in chronic noninsertional Achilles tendinopathy.

123

S322 However, the underlying mechanisms are far from being understood. Some patients develop Achilles tendon pain even after surgical repair of the Achilles tendon. Given the fact that almost all patients suffering Achilles tendon ruptures demonstrate signs of tendon degeneration and neovascularisation in histology, one might suggest microcirculatory tendon changes in this patient subgroup, too. We hypothetised that Achilles tendon capillary blood flow changes immediately after topical glyceryl trinitrate treatment. Methods: 15 patients (55±15 years, VAS 5.8±2.3) with current midportion Achilles tendon pain 36 months after open surgical Achilles tendon repair for tendon rupture were included. 1.2mg topical glyceryl trinitrate was sprayed on the Achilles mid-portion. Microcirculatory monitoring included capillary blood flow, tendon oxygen saturation and postcapillary venous filling pressures at the insertion and 2cm, 4cm and 6cm above the insertion using a combined laser Doppler & spectrophotometry system (Oxygen-to-see, Germany). Results: Baseline capillary blood flows of the painful vs. the uninjured tendon were increased (108±46 vs. 81±20 (2cm above the insertion), 104±40 vs. 76±20 (4cm above the insertion), 111±53 vs. 90±21 (6cm above the insertion, p\0.05)). However, topical glyceryl trinitrate did not change capillary blood flow at 2mm and 8mm tissue depths at at the painful Achilles tendon or the healthy tendon. Tendon oxygenation was not changed at the painful or the healthy Achilles tendon. Postcapillary venous filling pressure was reduced at 8mm at the mid-portion in the painful Achilles tendon only (113±37 vs. 95±31, p=0.030). Conclusions: Acute topical glyceryl trinitrate facilitates capillary venous outflow in painful Achilles tendons. However, capilllary blood flow and tendon oxygenation remain unchanged following acute topical glyceryl trinitrate application. Elevated capillary blood flow at the entire midportion is encountered at baseline in previously ruptured painful Achilles tendons even three years after surgical repair of the Achilles tendon indicating an altered microcirculatory flow pattern.

P23-293 A comparison of the histological and ultra-structural characteristics of the internal obturator tendon in patients with arthritis of the hip and patients with fracture of the collum femoris Meknas K.1, Johansen O.1, Olsen R.2, E. Steigen S.3, Kartus J.4 1 University Hospital of North Norway, Department of Orthopedic, Tromsø, Norway, 2University of Tromsø, Department of Electron Microscopy, Tromsø, Norway, 3University of Tromsø, Institute of Medical Biology, Department of Pathology, Tromsø, Norway, 4NU- Hospital Organization, Trollha¨ttan/Uddevalla, Department of Orthopaedics, Trollha¨ttan, Sweden Objectives: The aim of the study was to evaluate and compare the histological and ultra-structural characteristics of the internal obturator tendon in patients with either arthritis of the hip or fracture of the collum femoris (FCF). Methods: Nine patients (4 male, 5 female; median age 60 (48-75) years) with arthritis of the hip and ten patients (2 male, 8 female; median age 82.5 (6090) years) who had suffered an FCF (Garden III or more) underwent an open biopsy procedure in conjunction with a total hip replacement. A minimum of two biopsies were obtained from each patient. From all cases representative tissue material was obtained from the tendon. The tissue samples were prepared and stained using standard procedures for light and transmission electron microscopy (TEM) respectively. The light microscope was used to evaluate the cellularity, vascularity, fibre structure and the presence of Glycos-amino-glycans (GAGS), scar tissue and calcaria. TEM was used to evaluate the fibril diameter distribution. Results: All tissue samples in the arthritis group had limited areas of scar tissue, the corresponding was found in 4/8 patients in the FCF group (p=0.02). There were significantly more GAGs (p=0.009) and calcaria (p=0.001) in the samples from the obturator internus in the arthritis group than in the FCF group. Furthermore, within the arthritis group there were significantly more vascularity (p=0.04) and significantly more deteriorated fibre structure (p=0.02) found in the scar tissue than in the non-scar tissue. The corresponding was not found in the FCF group. The ultra-structural evaluation of the fibril diameter distribution was more heterogenous in the samples from the arthritis group and there were

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 significantly less small and medium sized fibrils found in the arthritis group than in the FCF group (p\0.0001). Conclusions: The tissue samples from the internal obturator tendon in the arthritis group revealed a more degenerative appearance in the light microscope as well as an altered fibril diameter distribution in the TEM compared with the samples from the internal obturator in the FCF group. These finding can be used in future comparisons with the pathology found in e.g patients with retro-trochanteric pain.

P23-356 Patellar tendon rupture in systemic lupus erythematosus: a case report Sano T.1, Matsuoka H.1, Nakayama K.1, Saji T.1, Hamamoto Y.1 1 Shizuoka General Hospital, Department of Orthopaedic Surgery, Shizuoka, Japan Objectives: Tendon rupture in patients with systemic lupus erythematosus (SLE) is a rare complication. We report a case of patellar tendon rupture with SLE. Methods: The patient was a 54-year-old man with a 28-year history of SLE who had received prednisone for 26 years. He was brought by ambulance to an emergency department because of left knee pain and inability to extend his leg after missing his footing on the stairs. Plain radiographs showed patella alta without any fractures. Magnetic resonance imaging revealed left patellar tendon rupture. The operation was performed eleven days after the trauma. The patella tendon was completely ruptured in the middle part of the tendon, and was repaired using nonabsorbable suture reinforced with the Leeds-Keio artificial (LK) ligament. The LK ligament was passed through the quadriceps muscle tendon just above the patella, and crossed over the patella. The end of the LK ligament was fixed to the tibia with two metallic staplers. The knee was immobilized for three weeks after the surgery. Results: One year after the surgery, the patient had full passive extension and 145 degrees of flexion. Conclusions: The LK ligament was useful to treat the patellar tendon rupture with SLE.

P23-358 One and two tendon proximal hamstring avulsions in athletes - when is surgery indicated? Lempainen L.1, Sarimo J.2, Mattila K.3, Orava S.2 1 University Hospital of Turku, Department of Orthopaedic Surgery and Traumatology, Turku, Finland, 2Mehila¨inen Sports Trauma Research Center, Mehila¨inen Hospital and Sports Clinic, Turku, Finland, 3 Medical Imaging Centre of Southwest Finland, University Hospital of Turku, Turku, Finland Objectives: Hamstring injuries are common in sports. They can be serious and difficult to treat, resulting often in impaired athletic performance and long rehabilitation times. Previous studies considering treatment of these problems are scarce. A common guideline regarding hamstring injuries is that in the presence of a complete proximal avulsion of the hamstring musculotendinous complex should surgical repair be considered. However, also more active recommendations have been recently shown up. This study was designed to investigate the effect of surgery on one and two tendon proximal hamstring avulsions in athletes. Methods: Forty-seven patients (32 men, 15 women) with one or two tendon proximal hamstring avulsions treated surgically were included. All included patients were actively involved in sports, and twenty-eight of them were professional or competitive level athletes. The average age of the professional and competitive level athletes was 25 years, and the most common sports among them were soccer and sprinting. The diagnosis was confirmed using an MRI. The operation was performed after failed conservative treatment on average 13 months after the injury. At the time of surgery, none of the patients were satisfied with their athletic performance because of their symptoms resulting from the injury and therefore surgery was indicated. In surgery avulsed proximal hamstring tendon(s) was reattached to the ischial tuberosity by suture anchors. The sciatic nerve was explored and in cases in which there were minor adhesions around it they were freed. A four category rating system was

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 used to evaluate the overall result. At follow-up the patients were asked about possible symptoms and their ability to return to sports. Results: The average follow-up was 36 months. Of the 47 patients, 41 had excellent or good results and they were able to return to their former level of sport after an average of five months. All 47 patients felt that they had benefited from the surgery, and their performance as well as the strength of the operated hamstrings had improved after the operation. The proximal tendon of the biceps femoris was the most commonly avulsed from the hamstring complex. Conclusions: It seems that conservatively treated one and two tendon proximal hamstring avulsions can lead to decreased performance in athletes demanding highly coordinated neuromuscular efforts in their sports. Given the good functional outcome and low complication rate, the authors advocate surgical treatment in one and two tendon proximal hamstring avulsions in top-level athletes. It should be remembered, however, that a proximal one or two tendon avulsion can be a career ending injury for an athlete. Further studies are needed to evaluate the role of conservative treatment in these injuries.

P23-402 High resolution magnetic resonance imaging of the semitendinosus tendon regeneration after harvesting for ACL reconstruction using a microscopy coil Hirano A.1, Fukubayashi T.2, Manmoto T.1, Mamizuka N.1 1 Mito Kyodo General Hospital, Orthopaedic Surgery, Mito, Japan, 2 Waseda University, Sports Science, Tokorozawa, Japan Objectives: The purpose of this study is to evaluate the regeneration process and the quality of semitendinosus (ST) tendon after harvesting for ACL reconstruction using the high resolution MRI. Methods: Subjects were 27 patients (13 male and 14 female; mean age, 23.9 years; range, 14 to 54 years) who had undergone ACL reconstruction using a ST tendon or both ST and gracilis tendons. All MR images were obtained on a MAGNETOM SYMPHONY 1.5T with microscopy coil. The MRI protocols were fast spin echo proton density weighted axial image and GRE T2 star weighted axial image. We performed MRI longitudinally until 12 months after the operation. The MR axial images at the level of 2 to 3 cm above the joint line were analyzed about signal characteristics of regenerated tendon. Results: MRI showed an apparent regeneration of ST tendon until 3 to 4 months after the operation except one case. 1) One to 2 weeks after graft harvest, MRI showed remaining of the tendon sheath and heterogeneous internal signal, probably hematoma. 2) Three to 8 weeks after, MRI showed harvest sites were swelling, covered the membranous tissue forming multiple layers. Images of internal regenerated tendon showed increased signal on proton density weighted images and intermediate signal on T2 star weighted images. 3) Three to 4 months after, the decreasing signal intensity on T2 star weighted images were shown in the internal regenerated tendon bundle. 4) Five to 9 months after, the internal signal decreased on both weighted images. The regenerated tendon was mature, however more thicker than normal ST tendon. 5) One year after, the regenerated tendon had normal MRI signal. Conclusions: Following harvested site of ST tendon, high-resolution MR images demonstrated the regeneration beginning from early stage after the operation. Five to 9 months after graft harvest, the regenerated tendon was mature, however more thicker than normal ST tendon.

P23-550 Functional outcome one year after surgically treated Achilles tendon rupture. No effect of autologous platelets Kvist J.1, Norrman H.1, Schepull T.2, Aspenberg P.2 1 Institute of Medicine and Health Sciences, Physiotherapy, Linko¨ping, Sweden, 2Section for Orthopaedics and Sports Medicine, Department of Clinical and Experimental Medicine, Linko¨ping, Sweden

S323 Objectives: We examined the objective and self-rated Achilles tendon function, as well as quality of life, in patients with total Achilles tendon rupture, one year after surgery. We also to examined the effect of autologous platelet concentrate (PRP) applied during surgery, on Achilles tendon function. Methods: Thirty patients with total Achilles tendon rupture were recruited consecutively. At the time of surgery, the patients were randomly allocated so that one group received a PRP injection. One patient in the PRPgroup suffered a re-rupture 2 months after cast removal. The remaining 29 patients were examined one year after surgical treatment. Muscle function was evaluated by Heel-Raise index (HRI), which is the product of maximum number of heel raises and maximal height at heel raise. In addition, range of motion, calf circumference, height at one leg vertical jump and peak force at toe-off during walking (measured on a Kistler force platform) were also evaluated. The patients estimated their function with the Achilles Tendon Total Rupture Score (ATRS) and quality of life with EuroQol - 5 Dimensions (EQ-5D). The mechanical properties of the healing Achilles tendon were measured at 7 weeks (cast removal), 19 weeks and 1 year after surgery. These results are presented in another abstract. Results: Range of motion, calf circumference and muscle function measured by HRI and one leg jump, were significantly reduced on the injured side compared to the non-injured. Peak force at toe-off during walking did not differ between legs indicating a normalization of function during light activities of daily living. Patients0 self-rated function with ATRS was good (median 84, range 59-98). The quality of life corresponded to values for the normal population. No differences in function were found between the PRP group and the control group, except for ATRS which was rated lower in the PRP group indicating less good function. Although significant, p=0.02, this does not allow firm conclusions, as it was a secondary variable. Conclusions: Achilles tendon function remained impaired one year after surgery for total tendon rupture. However, the patients reported good function, suggesting that the impairment was small enough not to limit the patient in daily activities.

P23-567 The process of tendon regeneration in an Achilles tendon resection rat model: a macroscopic, histological, and biomechanical study Otoshi K.1, Sekiguchi M.2, Konno S.2 1 Fukushima Medical University School of Medicine, Fukushima, Japan, 2 Fukushima Medical University School of Medicine, Orthopaedic Surgery, Fukushima, Japan Objectives: Regeneration of the medial hamstrings after harvesting for anterior cruciate ligament reconstruction has been reported in several studies. However, the process of tendon regeneration has not been well described. The purpose of this study is to clarify the mechanism of tendon regeneration by investigating the process of tendon regeneration macroscopically, histologically, and biomechanically. Methods: Fifty, adult female, Sprague-Dawley rats were used. After anesthesia, the Achilles tendon in the left hind limb was removed totally using the tendon-stripping device. Rats were sacrificed at 2, 7, 30, 90, and 180 days after surgery, and the regenerate tendons were dissected. Contralateral Achilles tendons were used as normal controls. Gross anatomic changes, microscopic remodeling, and recovery of biomechanical properties of regenerate tendons were investigated. The expressions of type I collagen, type III collagen, and transforming growth factor-b1 (TGF-b1) were also investigated by immunohistological observation. Results: The regenerate tendons formed in all specimens. In the early phase, hematoma and soft granulation tissue were uniformly observed at the harvest defect. These gradually matured with time, and the microscopic structure became quite similar to normal 180 days after surgery. However, the ultimate tensile strength and stiffness were significantly inferior to the normal tendon (p\0.05). TGF-b1 was well co-localized with inflammatory cells and fibroblasts in the regenerate tendon. The type I/type III collagen ratio in the regenerate tendon was significantly decreased in the early phase (p\0.05), but gradually increased with time.

123

S324 Conclusions: Tendon regeneration and maturation occurred uniformly along the length of regenerate tendons, rather than in a proximal to distal fashion. The hematoma that initially occupies the harvest defect may act as a scaffold for fibroblast precursor cells from the surrounding peritendinous tissue and tendon sheath.

P23-658 Traumatic proximal tibiofibular joint dislocation: a case report Sano T.1, Matsuoka H.1, Nakayama K.1, Saji T.1 1 Shizuoka General Hospital, Department of Orthopaedic Surgery, Shizuoka, Japan Objectives: Proximal tibiofibular joint dislocation is a rare injury. Methods: A 36-year-old male was brought to the emergency department with the right lower limb pain following a traffic accident. Plain radiographs showed anterolateral dislocation of the right proximal tibiofibular joint, open fracture of the right tibial shaft and posterior dislocation of the right hip joint with acetabular fracture. Open reduction and internal fixation was performed for all fractures. We used a 4.0mm cannulated cancellous screw to stabilize proximal tibiofibular joint. An above-knee plaster was applied and patient was asked to remain non-weight-bearing for 5 weeks. Five weeks after the operation, the screw was removed and functional rehabilitation started. Results: Patient achieved full range of motion of the knee joint one year after the operation. Conclusions: Proximal tibiofibular joint dislocation is a rare condition. Early diagnosis and appropriate treatment will lead full recovery of the knee joint.

P23-682 The use of injected platelet enriched plasma (PEP) for treatment of tendinosis. A prospective case serie to study indications, technique and limitations in daily orthopedic practice Sybesma T.1, van der Weegen W.2 1 St. Anna Hospital, Gelrop, Netherlands, 2St. Anna Hospital, Orthopedic Surgery, Gelrop, Netherlands Objectives: To evaluate the effectiveness of PEP treatment in four different indications (fasciitis plantaris, achilles tendinosis, patellatendinosis and epicondylitis lateralis). Secondary objectives are to evaluate the safety of PEP injections, to study the limitations of its use, to evaluate the technique for administering PEP and patient satisfaction. Methods: Patients with failed conservative treatment for four different types of tendinosis (fasciitis plantaris, achilles tendinosis, patellatendinosis and epicondylitis lateralis) were prospectively included in a single centre case series. Short term follow up was at 3 months and long term follow by questionnaire at 12 months. Pain was measured using a VAS scale and a self constructed questionnaire was used to evaluate patient satisfaction and how patients perceived the PEP injection. Treatment location was painguided and the tendon was infiltrated 4 to 6 time. Any complication or adverse event was recorded. Data was analysed with standard descriptive statistics and with a 2-tailed Student T-test where applicable. Results: PEP was used to treat 116 patients with chronic tendinosis. Eight patients for patellatendinosis, 14 for epicondylitis lateralis, 38 for achillestendinosis and 56 for fascitis plantaris. Symptoms were present for more than 12 months in 94 patients. In 22 patients, symptoms were present for 6 to 12 months. Three months after treatment, the mean VAS score for pain decreased to 3.8 (6.2 before treatment, p 0.003). At long term follow up, mean pain scores decreased to 2.8. Changes in pain per indication are presented in table 1. With longer follow up, patient satisfaction increased (42.9% at 3 months, 56.2% at 11 months. Most patients (59.1%) perceived treatment as painful. After treatment, 54.1% of all patients experienced an increase in symptoms (more pain: 29.6%; increased swelling: 9.2%; pain and swelling: 10.2%; other symptoms: 5.1%). The duration of these symptoms ranged from 1 to 14 weeks (median: 6). One patient, treated for achillestendinosis, suffered a DVT. Initially, patients were prescribed a soft cast for 2 week. After this event casts were no longer prescribed.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Patients were kept non-weight bearing for 2 days using crutches and were instructed to gradually resume activities after 2 days. Pain scores Mean pain (VAS)

Before treatment 3 months

p

All

6.2 (2.1)

3.8 (2.9) .003

2.8 (3.1) .005

Epicondylitis lateralis

6.6 (1.8)

2.9 (3.7) .008

2.6 (3.7) .434

Fasciitis plantaris

6.6 (2)

3.8 (2.9) .0001 2.9 (3.1) .066

Achillestendinosis 6.1 (1.9)

4.2 (2.9) .002

Patellatendinosis

2.1 (1.8) .225

3.3 (2.1)

11 months

2.6 (3)

p

.017

3 (2.1) .596

Conclusions: While other conservative treatments for tendinosis are either labour intensive (physiotherapy, exercises) or have unwanted side effects (corticosteroids), PEP treatment is easy to administer and without serious unwanted side effects. Pain decreases significantly and patient satisfaction is high regarding the characteristics of this patient population (chronic symptoms and failed conservative therapy). More research, using a control group, is needed.

P23-745 Involvement of the plantaris tendon in Achilles tendinopathy: an anatomical observational study van Sterkenburg M.N.1, Kerkhoffs G.2, van Dijk C.N.1 1 Academic Medical Center, Orthopaedic Surgery, Amsterdam, The Netherlands, 2AMC, Orthopaedics, Amsterdam, The Netherlands Objectives: The source of pain and the background to the pain mechanisms associated with mid-portion Achilles tendinopathy have not yet been clarified. It is questionable if degeneration of the tendon itself is the main cause of pain, since intratendinous changes are found in up to 34% of people without complaints. Pain is often most prominent on the medial side. During Achilles tendoscopy, we find that the plantaris tendon is fixed to the Achilles tendon at the level of complaints. In a healthy situation the plantaris tendon is attached to the calcaneus or Achilles tendon at its insertion, but can glide in relation to the Achilles tendon at its mid-portion. We postulate that in chronic inflammation, adhesions between Achillesand plantaris tendon are formed. The plantaris tendon is bi-articular whereas the medial portion (soleus) of the Achilles tendon is monoarticular. The Achilles tendon is involved in plantarflexion, whereas the plantaris tendon also causes ankle inversion. These opposite forces may cause complaints when adhesions are present. The purpose of this study was to anatomically assess the relationship of the plantaris tendon with the Achilles tendon. Methods: Seventy-six lower extremities were dissected. The plantaris tendon was identified proximally. After opening the superficial fascia and peritendineum, the plantaris tendon was bluntly released from the Achilles tendon moving distally. When with maximum pulling force the plantaris tendon could not be released, it was defined as an ‘adhesion’ to the Achilles tendon. Presence of the plantaris tendon, site of insertion onto the calcaneus, and ‘adhesions’ were documented (photo & video). Results: Mean age was 83 years (range 56-99), 42 were male, 27 female and 5 unknown. Thirty-four left-, and 40 right lower legs were dissected. In all a plantaris tendon was identified. Insertion of the plantaris tendon onto the calcaneus was medially in 36 (49%) specimens; anteromedially in 23 (31%); dorsomedially in 9 (12%); 2 (3%) inserted anteriorly into the Achilles tendon; 3 (4%) inserted medially onto the calcaneus as a tripod; and one (1%) inserted in the deep fascia. In 5 of 23 fresh frozen cadavers (22%) ‘adhesions’ were found. Three were a retinaculum- like structure, transversally constricting the Achilles and plantaris tendon 32-88 millimetres from the insertion; two inserted on the mid-portion of the Achilles tendon; and one adhered to the mid-portion of the Achilles tendon by a solid cord of connective tissue. One adhesion was found in the fixated

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 group, with the plantaris tendon inserting anteriorly into the mid-portion of the Achilles tendon. Conclusions: In 6 specimens adhesions between Achilles- and plantaris tendon were found, confirming our hypothesis that the plantaris tendon could play a role in maintaining complaints in patients with Achilles tendinopathy. An interesting secondary finding was that in contrast with the literature, in all specimens a plantaris tendon was identified.

P23-748 Achilles tendinopathy: stripping the plantaris tendon results in a good to excellent outcome. A report of 3 cases van Sterkenburg M.N.1, Kerkhoffs G.M.M.J.1, van Dijk C.N.1 1 Academic Medical Center, Orthopaedic Surgery, Amsterdam, The Netherlands Objectives: Achilles tendinopathy is generally difficult to treat. The source of pain and the background to the pain mechanisms associated with chronic Achilles tendinopathy have not been scientifically clarified. Often, intratendinous degenerative changes are addressed, however it is questionable if these changes are the main cause of the pain in tendinopathy. Intratendinous changes in up to 34% of people without complaints have been reported. Achilles tendinopathy and -paratendinopathy often co-exist. The pain is frequently most prominent on the medial side of the midportion of the tendon. This is where the plantaris tendon is also located. The plantaris tendon is bi-articular whereas the medial portion (soleus) of the Achilles tendon is mono-articular. During tendoscopy, the plantaris tendon seems fixed to the Achilles tendon at the level of complaints; solely releasing the peritendineum and cutting the plantaris tendon renders good results. We hypothesized that isolated stripping of the plantaris tendon from the symptomatic leg, and therefore release of the plantaris tendon from the Achilles tendon is sufficient to relieve complaints in patients with combined Achilles tendinopathy and paratendinopathy. Methods: Three patients with complaints of Achilles tendinopathy were included (2 female, 1 male). Mean age was 47 years (range 43-50). Patients experienced pain and stiffness around the mid-portion of the Achilles tendon. Pain was most prominent on the medial side. MRI indicated Achilles tendinopathy with involvement of the plantaris tendon. Maximum conservative treatment did not relieve complaints. In all three patients, the plantaris tendon was bluntly retrieved and excised with a tendon stripper through a 3 centimetre incision in the proximal calf. After- treatment consisted of a compression bandage for 3-4 days and functional recovery, allowing weightbearing when possible. Pre- and postoperatively VISA-A questionnaires and VAS for pain were completed; the post- operative questionnaire also included four- point scales for complaints and subjective outcome. Results: Pre-operatively, mean VAS for pain was 62 mm (range 58-69); VISA-A scores calculated 29 (range 19-44). At 3 months and 1 year follow-up, two had excellent outcome and no complaints and one reported less complaints and good outcome. Mean VAS for pain was 13 mm (range 4-22); mean VISA-A measured 59 points (range 38-83) at three months. The two sports active resumed their normal activities at 6 weeks postoperatively. Mean VAS for pain was 8 mm (range 3- 12) after one year; VISA-A measured 80 points (range 55- 97). Conclusions: We report a good to excellent outcome of stripping the plantaris tendon in three patients. The finding that pain is most prominent on the medial side in a large group of patients with Achilles tendinopathy may be based on involvement of the plantaris tendon in the process.

P23-801 Cells in the chronic painful Achilles tendon produce signal and pain substanses-stem cell characteristics Alfredson H.1 1 University of Umea˚, Sports Medicine Unit, Umea˚, Sweden Objectives: To characterise the cells in the chronic painful Achilles tendon. Methods: Biopsies from chronic painful tendinosis tendons and pain-free tendons were analyzed using immunohistochemical and in-situ hybridization techniques.

S325 Results: Cells in the hyper-cellular chronic painful Achilles tendinosis tendon were shown to produce acetyl cholin, katecholamines, Substans-P and glutamate. Conclusions: Cells in the hyper-cellular chronic painful Achilles tendinosis tendon, generally considered to be tenocytes, were shown to have stem cell-like characteristics, with production of signal substances related to the nervous system. These substances are involved in cell proliferation, pain mechanisms, collagen production, vasoregulation and apoptosis, all known to be involved in tendinosis. The findings stimulate to further evaluation of the tendon cella, and might have implications on the treatment of tendinosis.

P23-921 An anatomical and radiological study of the fascia cruris and paratenon of the Achilles tendon Carmont M.1, Highland A.1, Paling E.1, Rochester J.2, Davies M.1 1 The Sheffield Foot and Ankle Unit, Sheffield, United Kingdom, 2 Medical Teaching Unit, Sheffield, United Kingdom Objectives: Minimally invasive repair of the Achilles tendon has been shown to be a safe and effective means of repair. The Achillon jig permits the placement of sutures deep to the fascia cruris and the paratenon through the substance of the ruptured tendon itself, facilitating an intraparatenon repair and minimising risk of sural nerve damage. Occasional placement of the branches of the jig outside the paratenon suggests that these two layers may not be as clearly delineated as first thought or may merge at an, as yet, undefined level. Methods: We performed an anatomical and radiological study of the layers of tissue superficial to the Achilles tendon: the fascia cruris and the paratenon, in 4 paired and 4 non-paired cadaveric specimens comparing ultrasound and magnetic resonance imaging findings with anatomical dissection. Results: The mean distance for the confluence of the fascia cruris and paratenon from the Postero-Superior Calcaneal Tubercle (PSCT) was found to be 37.3mm (range 17-58mm). Ultrasound was less distinct than MRI scanning and the mean distances of the confluence to the PSCT were found to be 35.3mm (range 27.8-44mm) and 38.1mm (range 23.253.7mm) respectively. The mean thickness of the Achilles tendon was found to be 5.59mm at 3cm proximal to the PSCT. The sural nerve intersected the lateral border of the Achilles tendon at a mean distance of 85mm (range 55-123mm) from the PCST. Conclusions: We recommend close scrutiny for these layers when performing surgery on the Achilles tendon and careful placement of the branches of the Achillon jig beneath the paratenon, when using this method to repair the tendon.

P23-1018 Biomechanical evaluation of a medial knee reconstruction with comparison of bioabsorbable interference screw constructs and optimization with a cortical button Wijdicks C.1, Brand E.1, Nuckley D.1, Johansen S.2, LaPrade R.1, Engebretsen L.2 1 University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, United States, 2Ulleva˚l University Hospital, Orthopaedic Center, Oslo, Norway Objectives: Surgical treatment of severe acute injuries may be necessary in some circumstances to prevent chronic medial knee instability. Current fixation techniques predominantly utilize interference screws alone for soft tissue graft fixation in medial knee reconstructions. Some authors have proposed a cortical button as back up fixation to an interference screw, although these hybrid techniques have not been biomechanically validated. Our purpose was to biomechanically evaluate two tibial superficial MCL graft fixation techniques which consisted of an interference screw alone and in combination with a cortical button. Furthermore, we aimed to compare interference screws of different constructs.

123

S326 Methods: Twenty-four porcine tibias (average bone mineral density of 1.30 g/cm2; ± 0.18 g/cm2; range, 1.02 - 1.59 g/cm2, measured by DEXA scan), were divided into 4 groups of six specimens each. Group Ia: A 7 x 23 mm PLLA interference screw. Group Ib: A 7 x 23 mm PLLA interference screw and a 2.6 x 12 mm two-holed titanium cortical button. Group IIa: A 7 x 23 mm composite [70% poly(L-lactide-co-D, L-lactide) (PLDLA) and 30% biphasic calcium phosphate (BCP)] interference screw. Group IIb: A 7 x 23 mm composite interference screw and a 2.6 x 12 mm two-holed titanium cortical button. The reconstructed tibial specimens were fixed at the base of an Instron 5865 tester (Instron Systems, Norwood, Massachusetts). The grafts were isolated and preconditioned from 10 to 50 N at 0.1 Hz for 10 cycles, and then immediately subjected to cyclic loading, under repeated loads, for 500 cycles between 50 and 250 N at a frequency of 1 Hz. The grafts were further displaced at 20 mm per minute until failure. Statistical analysis was performed using a 2-way analysis of variance. Post hoc Tukey HSD tests were conducted to assess if there was a significant difference among testing parameters for each fixation techniques group. Results: Cyclic Testing There was a significant increase in the cyclic stiffness for the PLLA hybrid fixation (group Ib) (29.6 ± 6.9 N/mm) when compared to the PLLA screw-only (group Ia) (21.2 ± 3.8 N/mm) (p \ .05). While hybrid fixation groups (Ib and IIb; 5.2 (± 2.0) N/mm and 5.1 (± 1.3) N/ mm respectively) had lower changes when compared to the screw only groups (Ia and IIa; 7.1 (± 1.9) N/mm and 6.9 (± 1.7) N/mm respectively), the differences were not significant. Failure Testing In load-to-failure testing, a statistical comparative analysis was performed on the ultimate fixation loads of the present study to the previously published ultimate loads of Robinson et al. who reported the average strength of the entire superficial MCL to be 534 N (± 85). There were no significant differences in ultimate failure compared to the PLLA screw-only group (445.0 ± 72.2 N) or for the PLLA screw hybrid fixation group (511.0 ± 78.5 N) compared to the native superficial MCL. However, there was significantly less observed ultimate load for the composite screw-only group with an average ultimate failure load of 407.8 (± 77.9) N when compared to that the native superficial MCL (p \ 0.05). Conclusions: In conclusion, the current tibial superficial MCL reconstruction graft technique utilizing a PLLA interference screw alone serves as an adequate recreation of the native tibial superficial MCL strength and stiffness. In addition, a hybrid fixation of the PLLA screw with a cortical button lends additional stiffness during cyclic loading to its fixation and may be advisable for use in suboptimal cases.

P23-1037 Patellar tendon reconstruction with semitendinosus and gracilis tendon autografts van der Zwaal P.1, van Arkel E.1 1 MC Haaglanden, Orthopedic Surgery, Den Haag, The Netherlands Objectives: Retrospectatively evaluate reconstruction of acute and chronic patellar tendon ruptures using ipsilateral semitendinosus and gracilis (STG) tendon autografts. Methods: 8 patients underwent STG reconstruction. The grafts are harvested with a standard tendon stripper and need to be at least 18 cm in length. A horizontal tunnel is drilled through the distal half of the patella and through the tibial tuberosity with a canulated burr over a K-wire. The inner diameter of the tunnels equals the diameter of the bundled tendons, usually approximately 7 mm. The grafts are placed in the proximal tunnel and inserted crosswise through the tunnel in the tibial tuberosity. The patella is mobilized to its anatomical position by taking the knee through the range of motion from 08 to 908 and using the contralateral patella as a control. The grafts are tightened and sutured onto each other. Now they are fixed in the distal tunnel using a bioabsorbable interference screw. The screw diameter is 2 mm wider than diameter of the tunnel. If there is less torque than 15 inch/lbs, additional staples are used for securing the graft in the tibia. Patients were evaluated using Kee injury and Osteoarthitis Outcome Score.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Results:

Sex

Side

Age

Acute/chronic

Follow up

KOOS

Flexion

1.

male

R

52

acute

5y

68

130

2.

male

R

43

chronic

4y

72

134

3.

female

L

35

chronic

3y

82

140

4.

female

R

50

chronic

3y

78

128

5.

male

L

25

chronic

1y

66

90

6.

male

L

58

acute

1y

76

122

7.

male

R

88

acute

2y

48

95

8.

female

R

36

acute

5y

84

138

Conclusions: STG autograft reconstruction is a straight forward and effective operative technique in the case of extensor lag caused by patellar tendon rupture.

P23-1099 Functional bracing versus rigid immobilization after modified percutaneous Achilles tendon repair under local anesthesia Cretnik A.1 1 UKC Maribor, Department of Traumatology, Maribor, Slovenia Objectives: General complication rates were reported to be the highest in patients treated with percutaneous repair and early mobilization (15,6%). Modified (own) percutaneous suture techniques has shown significantly greater strength in biomechanical studies.Purpose of the study was to analyze the results of two ways of immobilization after treatment with the modified (own) percutaneous Achilles tendon repair under local anesthesia. Methods: In prospective, randomized study from the year 2001 to 2004 with 3 years follow-up there were 31 patients (32 ruptures) in the functional bracing group (Group 1) and 30 patients in rigid immobilization (Group 2), both for the period of 6 weeks and operated on using the same method. Results: There were similar data in Group 1 and 2 regarding gender (3 vs. 2 women), average age (41,9 vs. 42,2), side (16 vs. 13 right sided ruptures) and sports activity during injury (18 vs. 19). There was no re-rupture in Group 1 vs. one (3,3%) in Group 2 (1,6% altogether), two (6,2%) transient sural nerve disturbances in Group 1 and one (3,3%) in Group 2 (4,8% altogether), one suture extrusion problem in Group 2 (3,3%) with no other major or minor complications (6,2% of altogether complications in Group 1 vs. 10% in Group 2). Patients in Group 1 had slightly thicker but less rigid healed tendons, were (subjectively) more satisfied with the comfort (treatment), reached sooner final range of motion without limping and had higher average AOFAS score (96,9 vs. 95,7) with no statistical significant difference (p=0,38). Conclusions: The results of study support the choice of modified percutaneous Achilles tendon repair under local anesthesia with functional bracing and early mobilization.

P23-1157 Late partial patellar tendon rupture after arthroscopic cruciate ligament procedure Vega J.1, Redo´ D.1, Codina D.1, Golano´ P.2, Aguilera J.M.1 1 Hospital Asepeyo Sant Cugat, Orthopaedics, Sant Cugat del Valles, Spain, 2University of Barcelona, Laboratory of Arthroscopic and Surgical Anatomy, L’Hospitalet de Llobregat, Spain Objectives: Patellar tendon rupture following cruciate ligament reconstruction appears to be an infrequent but severe complication, especially in the early postoperative period. We present a case of partial patellar tendon rupture more than 4 years after uneventful failed posterior cruciate ligament reconstruction.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: A 24-year-old man, twists the left knee during his sport activity. Two months later, the patient was referred to our institution for evaluation of a progressive deficiency of strength at the extension knee. Five years ago, the patient underwent arthroscopic PCL reconstruction using bone-patellar tendon-bone autograft, obtained from the central third of patellar tendon. One year later, because subjective instability and knee pain during daily activity, a new arthroscopic PCL reconstruction using hamstring autograft was performed. The postoperative rehabilitation was uneventful, with the patient obtaining full range of motion and subjective stability. The patient remained without complaint for the next 4 years. Results: A partial patellar tendon rupture was detected in ultrasonography and MRI studies. During surgery, we observed a rupture of the medial and central thirds of the patellar tendon. The lateral third of the tendon was elongated but no rupture was detected. A reconstruction of the medial and central thirds of the patellar tendon was performed using bone-patellar tendon-bone allograft. At three months of follow-up, graft integration was demonstrated by a histological and ultrasonography study. Fifteen months after the operation, the knee had a complete range of motion with good stability and the patient restored full muscle volume and full strength of knee extensio´n. Conclusions: Late patellar tendon rupture following cruciate ligament reconstruction is an infrequent but severe complication. In the reported case, we suggest that use of hamstring in previous surgery has influenced in partial patellar tendon rupture. Hamstring tendons are important for the internal rotation of the limb, and this function is difficult to compensate after its harvesting. Maintained dynamic external rotation of the tibia may be the cause of a stress rupture in the medial and central thirds of the patellar tendon. In conclusio´n, bone-patellar tendon-bone allograft is a valid alternative for patellar tendon rupture treatment. We recommend the use of allograft for the treatment of failed cruciate ligament reconstruction in contrast to autologous graft.

P23-1215 Reconstruction of Achilles tendon using the artificial ligament Usami N.1, Hiraishi E.2, Ikezawa H.3 1 Shiseikai Daini Hospital, Foot and Shoe Center, Tokyo, Japan, 2Eiju General Hospital, Orthopaedic Surgery, Tokyo, Japan, 3Shisei, Shoe, Foot & Ankle Center, Tokyo, Japan Objectives: In the case of patients with an chronic rupture of Achilles tendon or patients with failed initial surgery, reconstruction is needed. Conventional methods require a longer time for cast or brace due to autologous tissue, resulting in a delayed return to social activities. We have reconstructed Achilles tendon using artificial ligament since 1987. Methods: The subjects were 38feet in 38 patients. The age was 24 to 58 years. The follow-up period ranged from 1 year 4 months to 12 years 9 months (average 5 years 6 months). Patients were a cast for 2 to 4 weeks postoperatively, started active exercise. At 3 to 4 weeks, partial weight bearing was started and 6 to 8 weeks, total weight bearing. Results: No patients showed restriction on range of motion. In sports activities, 31 out of 38 patients returned to the initial level, and the other 7 patients returned with slightly level-down. Infection and re-rupture were not observed in any patients. This method doesn’t require any autologous tissue, so it is not caused dysfunction postoperatively. It is operative indication to the cases with large defect, too. Because Achilles tendon is extra-articular tissue, it can be expected reconstructed tendon is high quality. Conclusions: Because the method using artificial ligament doesn’t sacrifice the autologous tissue, it is possible to return early to social and sports activities. It can be expected the reconstruction of high quality new-tendon because to extra-articular tissue.

P23-1266 A fibroblast-collagen membrane composite for tendon repair: an in vitro study Mangiavini L.1, Sosio C.2, Scotti C.1, Digiancamillo A.3, Domeneghini C.3, Fraschini G.2, Peretti G.4

S327 1

University of Milan, Residency Program in Orthopaedic and Traumatology, Milan, Italy, 2San Raffaele Scientific Institute, Department of Orthopaedics and Traumatology, Milan, Italy, 3University of Milan, Department of Veterinary Sciences and Technologies for Food, Milan, Italy, 4University of Milan, Faculty of Exercise Sciences, Milan, Italy Objectives: Tendon repair represents a current challenging clinical problem, as tendons have a poor intrinsic healing potential. Cell-based therapy could represent a valid therapeutic solution for this issue. The aim of this study was to identify an in vitro model of biological membrane seeded with fibroblasts, as a potential tool for improving the biological and biomechanical properties of the repair tissue. Methods: Achilles tendons’ specimens were surgically harvested from young pigs. The specimens were cut in small pieces of approximately 1 mm of diameter, cultured in vitro, in order to allow the cells to leave the specimens and then to reach the confluence (approximately 1 month). The fibroblasts were then enzymatically isolated, resuspended and expanded since confluence was reached once more. The cells were seeded onto membranes of collagen type I and III of 6 mm of diameter at 3 different concentrations. The membranes were cultured in vitro at standard culture conditions for 2 and 5 additional weeks, then retrieved from culture for macroscopic, histological and SEM analysis. Results: Macroscopically, the seeded membranes showed shrinkage and reduced biomechanical integrity compared to the unseeded membranes. The histological examination demonstrated the presence of vital cells within the membranes with a certain amount of matrix production. Conclusions: The results from this study demonstrate that the swine fibroblasts can be seeded onto a collagen scaffold. These cells remain vital during in vitro culture. The shrinkage of the experimental samples was probably due to an enzyme produced by the fibroblasts. Further studies will be schedule in order to demonstrate the survival of the cells and the reparative potential of fibroblast transplantation in an orthotopic in vivo model. We believe this model could be a valuable tool for tendon lesions, working either as a cell-carrier and a as patch augmentation.

P23-1312 Patellar tendon rupture repair with wire cerclage protected suture (Chandler’s Technique) - evaluation of results Loureiro M.1, Sousa A.1, Raposo F.1, Valente L.1, Moura-Gonc¸alves A.1, Sa˜o-Sima˜o R.1, Pinto R.1, Trigo-Cabral A.1 1 Hospital de Sa˜o Joa˜o, Orthopaedics, Porto, Portugal Objectives: Patelar tendon rupture without patelar fracture are relatively rare lesions that require surgical treatment in order to restore anatomical integrity of the extensor apparatus of the knee. It0 s more frequent in patients under 40 years, especially in athletes. Systemic illness like lupus or diabetes, as well as steroids injection are risk factors. At our center we use direct repair and re-insertion of the tendon in the patella and protect the repair with a wire cerclage(Chandler0 s Technique). The objective of this study is to evaluate the functional outcomes of patients treated using this technique in our Hospital. Methods: Retrospective study of a cohort of patients with patellar tendon rupture operated using Chandler0 s technique between 2001 and 2006: 13 patients with a mean age of 47.2 ± 15.9, 11 men and 2 women, with an average follow-up of 40 months. Knee function was evaluated with Knee injury and Osteoarthritis Outcome Score (KOOS) and International Knee Score (IKS), and the presence of systemic illness was registered. Statistical study with SPSS. Results: 6 resulted from sports accident, 5 personal injury and 2 labor related. Comorbidities - 2 patients with Chronic Renal Failure on hemodialysis and 2 patients with type II Diabetes (30.7% with systemic disease). KOOS means and sd, median: Pain - 76.7 ± 15.9, 86; Symptoms - 73 ± 28.2, 86; Daily activities (AD) - 86.1 ± 20.7, 94; Sport - 32.3 ± 22.6, 35; Quality of Life - 61.7 ± 23.7, 69. ROM (mean): 3-120. Conclusions: It is confirmed in this work some prevalence of this type of lesion in patients with systemic disease. The results were good except for 3 patients with severe comorbidities. The low score of sport, reflects a significant change in recreational and sports of patients who suffer this injury.

123

S328 The cerclage of Chandler is a good option for the repair of acute patellar tendon rupture, allowing for early rehabilitation and good functional results.

P23-1324 Ultrasound imaging of the Achilles enthesis fibrocartilage with histological correlation Basci O.1, Aydin S.Z.2, Bas E.3, Direskeneli H.2, Celikel C.3, Atagunduz P.2, Filippucci E.4, Wakefield R.5, Mcgonagle D.5, Benjamin M.6, Karahan M.1 1 Marmara University Faculty of Medicine, Orthopedics and Traumatology, Istanbul, Turkey, 2Marmara University Faculty of Medicine, Rheumatology, Istanbul, Turkey, 3Marmara University Faculty of Medicine, Pathology, Istanbul, Turkey, 4Universita` Politecnica delle Marche, Clinica Reumatologica, Ancona, Italy, 5University of Leeds and Chapel Allerton Hospital, Rheumatology, Leeds, United Kingdom, 6 Cardiff University, Connective Tissue Biology, Cardiff, United Kingdom Objectives: Ultrasound (US) is useful for the assessment of enthesitis, but its ability to visualize fibrocartilage at the enthesis has not been explored. Since fibrocartilage is thought to be a common target for pathology, its visualisation is of great potential clinical interest. To validate US as a technique for visualizing Achilles tendon entheseal fibrocartilage (EF) by imaging the bovine Achilles insertion and performing histological comparison and by imaging the Achilles tendon in patients with spondyloarthropathy. Methods: Achilles enthesis US of 18 bovine hindfeet was performed using a MyLab 70 US machine (Esaote Biomedica, Genoa - Italy), equipped with a broadband 6-18 MHz linear probe. The presence of tissue with fibrocartilage characteristics was documented and histological confirmation was performed on 5 randomly selected sections using Masson trichrome staining. US of the Achilles tendon was performed in 21 patients with spondyloarthropathies and 8 healthy controls to visualize the same region. Results: The bovine EF could be visualized by US in all cases and seen as a thin, uncompressible, anechoic/hypoechoic layer between the hyperechoic bone and the hyperechoic fibrils of the enthesis both in longitudinal and transverse scans. This region corresponded to EF on histological examination. The same pattern of low signal corresponding to fibrocartilage location could be seen in 19/21 SpA patients and in normals. The anechoic layer was mostly continuous in healthy controls but discontinuous around erosions and enthesophytes in SpA patients. Conclusions: US can visualize EF of the Achilles tendon which has implications for a better understanding of enthesopathy.

P23-1330 Bilateral rupture of quadriceps tendon in an athlete - case report Loureiro M.1, Sousa A.1, Raposo F.1, Valente L.1, Moura-Gonc¸alves A.1, Pinho A.1, Pinto R.1, Trigo-Cabral A.1 1 Hospital de Sa˜o Joa˜o, Orthopaedics, Porto, Portugal Objectives: The rupture of the quadriceps tendon is a relatively uncommon lesion that usually occurs in patients over 40 years and has a strong association with systemic disease such as diabetes, gout, renal failure, steroids and hyperthyroidism. It is rarely bilateral. Early repair is important because soft tissue fibrosis and contraction tends to difficult the repair at later stage. The authors present a case report with bilateral rupture of quadriceps tendon in an amateur runner. Methods: Male, 37 years, amateur runner, trying to avoid falling of stairs suffered an indirect trauma with immediate severe pain and functional disability. Ultrasound revealed bilateral rupture of the quadriceps tendon. Comorbidities: Diabetes mellitus non-insulin-dependent medicated and controlled. Repair 6 hours after acute injury. Reinsertion of the tendon in the upper pole of the patella and suture of the peritendon. Results: Discharge from Hospital in the 3rd postoperative day, no complications. Intermittent passive mobilization began at 4 weeks with splints

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 and muscle strengthening beginning at 6 weeks. At 3 months started walking, ROM of 0-110. At 8 months resumed running activity. Conclusions: Surgical repair of acute rupture of the quadriceps tendon provides a good functional recovery. In the case presented a bilateral lesion of lower limb resulted in a delayed recovery.

P23-1337 Popliteus tendon dislocation preventing arthroscopic reduction in a tibial plateau fracture Bilgic E.1 1 Kahramanmaras Sutcu Imam University, Medical School, Kahramanmaras, Turkey Objectives: Arthroscopically assisted reduction and fixation techniques are being used with increased frequency for the treatment of tibial plateau fractures. We report a case of popliteus tendon dislocation between fracture fragments in a Schatzker type II tibial plateau fracture, which necessitate open reduction. Methods: A 34 years old woman has been admitted to our clinic after a fall from stairs with a diagnosis of Schatzker type II tibial plateau fracture. Arthroscopically assisted reduction and fixation was planned for her definitive treatment. During diagnostic arthroscopy it was noted that anterior and middle horns the lateral meniscus has been detached peripherally and displaced in posterior and medial direction. With this displacement the popliteus tendon has also been displaced and trapped between fracture fragments. Though every effort has been made it was impossible to deliver the popliteus tendon to its original location arthroscopically. Open reduction has been performed with an anterolateral incision. It was possible to relocate the popliteus tendon after displacing the anterolateral fracture fragment anteriorly. Fracture has been fixed with a periarticular locking plate, defects were filled with cancellous bone and lateral meniscus has been sutured. Results: In her control after one year from the operation the patient had no complaints though her knee flexion has been limited to 100 degrees. Conclusions: To our knowledge this is the first report in literature about popliteus tendon preventing arthroscopic reduction in a tibial plateau fracture.

P23-1345 A original surgical technique to treat neglected ruptures and reruptures of the Achille tendon that use hamstring graft: 5 case reports Lefevre N.1, Bohu Y.2, Herman S.3 1 Institut de l’Appareil Locomoteur Nollet, Paris, France, 2 La Pitie´-Salpe´trie`re Hospital, Orthopedic and Sport Surgery, Paris, France, 3Centre Me´dico-chirurgical Paris V, Paris, France Objectives: For patients with a neglected rupture or rerupture of the Achille tendon, the usual treatment is surgical. Many surgical techniques have been published with more or less good results. We propose an original technique using a hamstring graft to realize a reconstruction with 6 or 8 strands hamstring tendons. We treated 5 male patients with a neglected rupture or rerupture of the Achille tendon. Methods: The first case concerned a male of 50 years old with a neglected rupture, over a year, with retraction and a 12 cm defect of the tendon. The second case concerned a male of 35 years old who was initially and unsuccessfully operated by a percutaneous technique, with a rerupture and a lack of substance of 6 cm. The third case was a patient of 60 years old who had rerupture 3 months after a bone reinsertion that needed a 6 cm graft. The fourth case was a man of 62 years with a complete rupture undiagnosed and neglected for over 5 months with a defect of 8 cm The fifth case was a male of 77 years with a complete rupture undiagnosed for 1 year with a defect of more than 10cm. The surgical technique was the same for each of the 5 patients. We took a hamstring graft on the homolateral knee (gracilis and semitendinous

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 tendons). We adapted the size of the graft to the lack of substance and realised a 6 or 8 strand reconstruction. Then we sutured the graft to the proximal and distal ends of the Achille tendon in 4 cases, and and we did a bone reinsertion in one case. Results: The follow-up needed an immobilization in plaster during 2 month minimum (3 successive casts with decrease in equine and progressive loading) At mean follow up of 36 months (6 to 96), the results are satisfying in the 5 cases, the patients have no walking limit, a normal mobility and a steady support. In 3 cases a resumption of sporting entertainment was possible after the 6th month. Conclusions: In conclusion, this original technique is able to repair a major lack of substance. The use of hamstring graft to treat neglected ruptures and reruptures of the achille tendon rendered a good over-all subjective and objective outcome of the patients.

P23-1399 Using biological scaffold, surgical reconstruction of traumatic rupture quadriceps tendon with more than 4cm defect Bhattacharyya M.1, Sakka S.1 1 University Hospital Lewisham, Orthopaedic, London, United Kingdom Objectives: Surgical management of quadriceps tendon ruptures, acute or chronic are well described. However, literature reports no consensus as to the best repair technique. Direct tendon repair with various surgical approaches, as well as using allograft to repair chronic tendon rupture, are documented. Our objective was to describe the use of biological scaffold in the reconstruction of quadriceps tendon and the surgical outcome. Methods: Thirty one patients had a quadriceps or patellar ligament rupture, admitted to our senior authors unit at our institution from January 2003 to September 2009. All these patients had clinically and radiologically confirmed quadriceps or patellar tendon rupture and received primary treatment in the Accident and Emergency department with above knee cast immobilization. Only three patients fit the inclusion criteria, who had, chronic quadriceps rupture, more than four weeks duration and with greater than four cm soft tissue defect and no bony deficiency, as confirmed clinically and radiologically, in the extensor mechanism of the knee. These patients had open repair of the extensor mechanism using biological scaffold. The affected knee was opened with a midline incision, the soft tissue defect in the quadriceps aponeurosis was identified. The scaffolding material (graft jacket) was placed between the defect and the superior pole of the patella. The biologic scaffold was then sutured with the host tissue. Subsequently the patients were followed up in the outpatient clinic with an MRI scan. Results: At 3 months follow up, none had any surgical site infection, mean flexion arc was 70 degrees. All Patients began knee exercises after three(3) months. As soon as the patients reach 100 degrees of knee motion, they were allowed bearing full weight on the operatively treated limb. All patients achieved full pre injury level of range of motion Conclusions: We describe the early clinical results of biologic scaffolding material to repair the chronic quadriceps rupture. Our aim was to preserve the biological integrity of the extensor mechanism to facilitate biological integration and minimize distal patella migration. In addition, implanting a scaffold is inductive, stimulating the regeneration of the desirable tissue by releasing growth factors and cytokines, as confirmed by late MRI scan. This method enhances load sharing among these fibers, thereby increasing the overall strength of the repair. This study conducted with non-elite individuals showed stable knee joints with a range of motion necessary for activities of daily living. Early results suggest that this folding material may act as a bridge to provide physiological joint biomechanics. The absence of the synthetic scaffold materials may prevent host graft response and thereby prevent late mechanical failure, which could account for poor surgical outcome. However, further research is necessary to

S329 understand the physiologic response following augmented quadriceps repair.

P23-1430 Using absorbable reinforcement, surgical reconstruction of traumatic rupture patellar tendon - a bridge to provide physiological joint biomechanics Bhattacharyya M.1 1 University Hospital Lewisham, Orthopaedic, London, United Kingdom Objectives: Surgical management of patellar tendon ruptures, acute or chronic are well described. However, literature reports no consensus as to the best repair technique. Direct tendon repair with various surgical sutures to repair acute tendon rupture, are documented. Our objective was to describe the use of the absorbable reinforcement tape for control tension at the suture line in the repair patellar tendon and the surgical outcome. Methods: Thirty five patients had acute patellar ligament rupture, admitted to our senior authors unit at our institution from January 2003 to January 2009. All these patients had clinically and radiologically confirmed patellar tendon rupture and received primary treatment in the Accident and Emergency department with above knee cast immobilization. These patients had open repair of the patellar ligament extensor mechanism using biological scaffold. They were prospectively followed up for an average period of 21 months. The affected knee was opened with a midline incision, the ruptured ligament was identified. The ligament was directly repaired with sutures passed through the transosseous tunnel in the patella and tied at the superior pole of the patella. The absorbable reinforcement tape was then sutured with the host tissue circumferentially drilling a coronal tunnel in the proximal tibia. The patients were followed up in the outpatient clinic clinically and radiologically. Results: All patients were allowed to weight bear immediately in a protected knee splint. None had any surgical site infection. All patients achieved full pre injury level of range of motion. The patients resumed their pre-injury activities at an average of 5.2 months. The active knee movement averaged 0-140 degrees compared to 0-150 degrees in the contralateral knee. Radiologically no patella alta, patella baja or degenerative changes in the patellofemoral joints were noted. Conclusions: The clinical results of absorbable suture tape material to repair the acute patella tendon rupture instead of cerclage wiring gives a good outcome. The aim was to reduce the tension at the suture and provide the biological integrity of the extensor mechanism to facilitate biological integration and minimize patella migration. The reported complications of the repaired patellar tendon have been attributed to the influence of the mechanical environment on the healing process. This study postulates that the healing complications can be minimised through tension regulation at the suture line using an absorbable reinforcement tape. In addition, this method enhances load sharing among these fibers, thereby increasing the overall strength of the repair. This study conducted with non-elite individuals showed stable knee joints with a range of motion necessary for activities of daily living. Early results suggest that this material may act as a bridge to provide physiological joint biomechanics.

Sport specific injuries P24-49 Biomechanical analysis of double plate vs LISS fixation for complex tibial plateau fractures Rademakers M.V.1, Blankevoort L.1, Kerkhoffs G.1, Marti R.2 1 Academic Medical Centre Amsterdam, Orthopaedics, Amsterdam, The Netherlands, 2Academic Medical Centre Amsterdam, Amsterdam, The Netherlands

123

S330 Objectives: In this biomechanical study we analyzed the load tolerance, stiffness and failure mechanism of bicondylar tibial plateau fractures in cadaver knees treated by double plating versus locked unilateral plating. Methods: Eight pairs of fresh frozen human knee specimens were used in this study. In all specimens the collateral and cruciate ligaments were preserved. Pairs were randomized for fixation method (double plating or LISS). After osteosynthesis the specimens were placed in a pressure bench. Cyclic loading was applied with a frequency of 1 Hz. The force increased with 200 N every ten minutes until failure of the construct. The stiffness and failure mechanism were analyzed making use of threedimensional analysis. Results: No statistical significant differences in bone mineral density within the matched pairs were observed (p=0.138). The load until failure was significantly higher in double plating constructs (p=0.017) and showed less subsidence when compared LISS fixation (p= 0.041 a t 400N). All LISS constructs failed at the proximal plate-to-screw fixation points. Conclusions: Biomechanical analysis of highly unstable tibial plateau fractures show a lower initial stability and a higher risk of secondary dislocation with LISS fixation when compared to double plating.

P24-54 Arthroscopic repair for recurrent anterior shoulder instability in professional soccer players Vaquerizo Garcı´a V.1, Cugat R.2, Alvarez P.3, Steinbacher G.3, Seijas R.4, Ares O.5, Cusco´ X.2, Samitier G.3 1 Prı´ncipe de Asturias University Hospital, Alcala de Henares, Spain, 2 Hospital Quiron, Barcelona, Spain, 3Catalonian Soccer Federation, Orthopedic Surgery and Traumatology, Barcelona, Spain, 4 Hospital Quiron, Orthopaedics and Traumatology, Barcelona, Spain, 5 Hospital de Viladecans, Orthopaedic Surgery, Barcelona, Spain Objectives: Since the 80s the treatment of anterior shoulder instability begins with arthroscopic repair, the results have improved close to the development of new implants and the surgical technique. The anterior instability of shoulder supposes a functional limitation for the sports practice in professional soccer players. The aim of the surgical treatment is to recover the shoulder stability for reincorporation to patient’s sports activity of the. The purpose of our study was to value the results obtained in the treatment of the recurrent anterior shoulder instability in professional soccer players. Methods: We realized a retrospective study of 52 patient professional soccer players who had recurrent anterior shoulder underwent arthroscopic reconstruction between 2000 and 2005. The minimal follow-up was 2 years. All the patients were operated by same surgeon. In all the cases it was possible to realize a follow-up of the mobility and force compared with contralateral shoulder, evaluating the stability and recurrence rates. Finally we compared final sport activity level of the patients with the previous injury level. Results: All patients were professional soccer players belong to the Catalonian Soccer federation. The average follow-up was 4.13 years. The mean age was 21.1 years. In all patient we used knotless suture anchor, kinsa 51.9%, Biokontless 48.1%. At the end of the follow-up we observed a significant increase of the stability, using apprehension test to evaluate results, 3 cases of recurrence after 1.6 years after indirect traumatism. we observed neither a significant decrease of the mobility nor muscular force. Comparing the sports level at the end of the follow-up with the previous we observed 100% of patients returned to practise soccer as the previous level. Conclusions: Though exists multiple studies in the literature that value the treatment of the recurrent anterior shoulder instability in athletes, even in contact sports in general, there are no studies with professional soccer players as in our study. One of the principal problems that the professional athletes present is the risk of recurrence, also the possibility of reincorporation to the previous sports practice. In the literature we observe that exists a significant incidence of patients who practise sports of contact and therefore sports of risks that do not return to play or they practise to a low level. In our study we can observe all the patients return to play soccer at

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 the same level before the injury. We believe that the arthroscopic repair for treatment of recurrent anterior shoulder instability is indicated in professional soccer players.

P24-55 Application of botulinum toxin type a in dynamic osteopathy of the pubis Vaquerizo Garcı´a V.1, Vaquerizo Garcı´a E.2, Gala´n Novella P.3 1 Prı´ncipe de Asturias University Hospital, Alcala de Henares, Spain, 2 Infanta Elena Hospital, Valdemoro, Spain, 3Virgen de la Poveda Hospital, Villa del Prado, Spain Objectives: Dynamic osteopathy of the pubis (DOP) is an inflammation of the insertion of the adductor muscles of the lower limbs in the pubis following an imbalance between these and the abdominal muscles. Rehabilitation therapy usually obtains good results but surgery is occasionally necessary. The objective of this preliminary study is to assess the efficacy of treatment with Botulinum Toxin type A (BTX A) in dynamic osteopathy not responding to rehabilitation. Methods: We have studied five patients, 4 males and one female, diagnosed as having DOP confirmed by MRI. Three of the men associated the symptoms with sports (soccer and jogging), the other with work activities (packing unloading) and the woman with childbirth (she had a natural birth and pubis diastasis). All of them had received rehabilitation therapy (kinesitherapy, magnetotherapy, laser therapy, ultrasound treatment and analgesic electrotherapy) without any appreciable improvement in the pain or function. Patients with symptoms associated with sports were also treated before with corticoid and anaesthetic local infiltration twice in one case and three times the other two, without symptoms relieve. We used an analogic visual scale to pain assessment. Patients were followed up for one year and evaluated three and twelve weeks after infiltration. Results: The mean age was 28.6 ± 3.2 years. The mean time since onset was 9.4 ± 4.3 months. In three patients, the pain had disappeared completely three weeks after administration of BTX A. Initial AVS was 8.8 ? 0.6, after administration of BTX A AVS was \2 in three patients and the others refereared pain relieve but AVS was [7. After 12 weeks, these two patients AVS was [5 and the other three remained asymptomatic. Two patients required re-infiltration 12 weeks later due to partial persistence of the symptoms. The four males remained asymptomatic one year after administration of BTX-A and have returned to their regular sporting and employment activities. The woman continued with pain and is waiting surgery. Conclusions: The DOP is a very common injury among athletes, although rehabilitation in many cases requires surgery. In the literature we observe the good results that show surgical treatment even though is not without complications. Infiltration with botulinum toxin seems to be an effective treatment for DOP and may avoid the need for surgery. Larger prospective studies with professional soccer players have to be performed to confirm this treatment with botulinum toxin type A to DOP.

P24-140 Prevalence of bowlegs among adolescent soccer players Massada M.1, Pereira A.1, Sousa R.1, Massada L.2 1 Hospital de Santo Anto´nio, Servic¸o de Ortopedia, Porto, Portugal, 2 Faculdade de Desporto, Universidade do Porto, Traumatologia do Desporto, Porto, Portugal Objectives: Men’s anthropometric variables are determined by genetic factors and Little is known about the interactions of sports-related demands and those variables, in particular on musculoskeletal features, during growth. This study pretends to focus on the relationship between soccer and lower limb alignment, namely on the variation of the anatomic angle of the soccer player’s knee, during the fast and terminal growth phases.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Methods: Participants: 580 male child/adolescent soccer players aged 12 to 17 years. Interventions: All athletes completed a demographic questionnaire and underwent physical examinations, which included height, weight, knee, ankle, foot and spine axis, hip range of motion, tibial torsion and Q angle. Main Outcome Measurement: Varus/valgus axis was determined by the definition of anatomic angle according to Salenius et al. Results: A significantly higher prevalence of knee varus was found among the groups. The 17 years old player age group showed a mean left varus knee of 2.28 ± 3.5, with a prevalency of right varus knee of 34.9% and left varus knee of 51.5%. Conclusions: This anthropometric facet observed in the soccer player may depend of the asymmetric mechanical stresses acting on the femoral and tibial fisis during the growth phase. Further research is needed to explore the rationale of this phenomenon.

P24-154 Central aponeurosis tears of the rectus femoris: practical sonographic prognosis Maestro A.1, Balius R.2, Pedret C.3, Rodriguez L.4, Fdez-Lombardia J.5, Garcia P.6 1 FREMAP, Sports Medicine, Gijon, Spain, 2FIACT, Sports Medicine, Barcelona, Spain, 3Unitat Medicina Esport. Fundacio´ Hospital Comarcal Sant Anto, Vilanova i la Geltru´, Spain, 4Hospital Cabuen˜es, Gijon, Spain, 5 FREMAP, Orthopaedic Surgery, Gijon, Spain, 6Hospital Cabuen˜es, Radiology, Gijon, Spain Objectives: This study is a statistical analysis to establish whether or not a correlation exists between the level and degree of rectus femoris (RF) central tendon injury and the amount of time that athlete is unable to participate subsequently referred to as ‘‘sports participation absence’’ (SPA). Methods: 35 players from two high-level Spanish soccer teams with an injury to the central tendon of the RF based on clinical and ultrasound (US) criteria. Main outcome measurement: Ultrasound examination was performed with an 8-12 MHz linear multi-frequency transducer. All studies included both longitudinal and transverse RF sections. Results: At the proximal level the SPA time was 45.1 days when the injury length is 4.0 cm. This value increases by 5.3 days with each 1 cm increase in the length of injury. In the case of distal level injury, SPA time was 32.9 days when the injury length is 3.9 cm. This value increases by 3.4 days with each 1 cm increase. In the total representative sample, SPA time when the injury length is 4.2 cm corresponds to 39.1 days. This value increases by 4.2 days per length unit. Conclusions: RF central tendon injury at the proximal level is associated with a greater SPA time than at the distal level. Patients with a grade II injury have a SPA time longer than those with a grade I whether the injury is located proximal or distal.

P24-159 Stress reaction of the tibia in recreational athletes. Is there a golden standard for the medical treatment? Pu¨tz O.1, Scha¨ferhoff P.1, Dewitz H.1 1 MediaPark Klinik, Orthopaedics and Traumatology, Cologne, Germany Objectives: The purpose of this observational study was to investigate the frequency of stress reactions of the tibia in sports. In our department different orthopaedic surgeons reported an increasing number of stress reactions of the tibiaplateau in recreational athletes. There are some case reports concerning this diagnosis in athletes in literature. Nevertheless there is no golden standard for the medical treatment. Methods: 31 patients (9 female, 22male) with a diagnosis of stress reaction of the tibia were seen in our orthopaedic department between May 2008 and September 2009. All athletes had increased the intensity of their training a few weeks before their stress injury. There were no advices for a stress reaction in x-ray, but it could be detected in MRI examinations in all cases. All patients underwent conservative treatment without extra medicamentous therapy. Dependent on localisation of the stress injury an

S331 elimination in sports, magnetic field therapy, physiotherapy and nonweight bearing conditioned by pain were exhibited in all athletes. 15 patients got a leg brace (varus/valgus) in addition to their non-weight bearing. Weekly follow-ups and an increase of load dependent upon the clinical findings have been carried out. Results: The average of the medical treatment until full weight bearing was 6 weeks for the athletes with leg brace and 8 weeks for the other 16 patients. Similar to these findings the pain conditioned non-weight bearing had a shorter term for the patients with orthosis. Conclusions: Stress reactions are relatively common overuse injuries seen in recreational athletes, particularly in running athletes. The group with leg brace show earlier pain conditioned weight bearing in contrast to the control group without orthosis. By this means a faster mobilisation could be carried out. Furthermore athletes treated with a leg brace are able to take part in their competition or athleticism significant earlier than others. It is very difficult to define a golden standard regarding the variations of stress reactions and the patient compliance. MRI examination has emerged as a highly sensitive method for detecting stress reactions of bone. We would suggest in patients with unicondylar stress reactions other than that mentioned medical treatment a valg./var. orthosis. Regarding all the cases in our orthopedic department it seems to be a very important point that a lot of recreational athletes misjudge their training intensity and scale.

P24-172 The Enduro-motorcyclists wrist and other overuse injuries in competitive Enduro-motorcyclists: a prospective study Sabeti-Aschraf M.1, Serek M.1, Pachtner T.1, Auner K.1, Graf A.1 1 General Hospital of Vienna, Vienna, Austria Objectives: According to public opinion Enduro- motorcycling is a dangerous sport. Little is known about the anatomical region and the kind of overuse injuries appearing in a competition. This study was carried out to analyse off road motorcycle sport immanent overuse injuries. Methods: A study consisting of two phases was initiated using preformed questionnaires in volunteering Enduro- motorcyclists, who were examined immediately before and after the race. In phase one a prospective field study at the Prolog of the Erzberg-Rodeo was carried out to evaluate the overused anatomical regions. Basing on phase’s one results, the next year’s phase two study was designed as a prospective randomised investigation. There, clinical examinations for pain in the hand/ wrist and forearm region using the Visual Analogue Scale, for Raynaud Syndrome, Carpal Tunnel Syndrome, loss of grip strength in the forearm muscles were recorded. Results: Overall 298 athletes took part in this investigation. The predominately overused regions were found at the hand/wrist and forearms in phase one. Nearly 50% of all riders complained about pain, or paraesthetic sensations in the hand and wrist. In phase two the results concerning pain in the hand/wrist were confirmed with significant increase of pain during the race, but no significant difference within professionals and hobby athletes. Concerning pain in the forearms, professionals had significantly less pain. After the second run 38.7% of all riders ha carpal tunnel like symptoms. Conclusions: More than half of all analysed athletes complained about overuse injury. There is strong evidence that Enduro- motorcycling has a high potential to cause overuse syndromes, especially in the upper extremity. Due to this study’s findings transient carpal tunnel like symptoms are obviously a sport immanent overuse injury occurring in professionals and non professionals in comparable numbers.

P24-176 An anatomic posterolateral reconstruction of the knee: results after one year Van Ochten H.1, Van Tienen T.1, Luites J.2, Wymenga A.1 1 Sint Maartenskliniek, Department for Orthopaedic Surgery, Nijmegen, Netherlands, 2Sint Maartenskliniek, Research, Development and Education, Nijmegen, Netherlands

123

S332 Objectives: We report a one year follow - up prospective analysis of results of an anatomic posterolateral reconstruction technique using an Achilles tendon allograft in a consecutive series of fourteen patients with chronic symptomatic instability and pain of the knee. All procedures were performed between 2004 and 2008. Goal of the study was to asses the stability of the knee after this procedure and we hypothesized that this reconstruction technique would lead to a stable knee joint comparable to that of the opposite uninjured knee and that the patients would return to their pre-injury activity. Methods: All fourteen patients underwent a posterolateral corner reconstruction as described by Laprade et al.. If a cruciate ligament rupture was present and not addressed earlier, these were reconstructed as well. One year after surgery the stability of the joint was evaluated in terms of lateral laxity using varus stress radiographs and clinically (VAS, Noyes, IKDC, Lysholm and Tegner). All fourteen patients underwent complete one year postoperative followup. No patients were last to follow-up. Four patients underwent a reconstruction of the PCL (one in two tempi), two patients underwent an ACL reconstruction (one in two tempi) and three patient had previously undergone a reconstruction of one of the cruciate ligaments. Results: Mean lateral laxity of the injured knee was significantly less one year after surgery on varus stress radiographs compared to the pre-surgery values; from 9.4 ± 4.2 to 6.2 ± 3.5 8 (p \ 0.01). Postoperative lateral laxity, 6.2 ± 3.5, did not return to the level of the uninjured knee, 3.5 ± 1.6 (p \ 0.01). The external rotational component was evaluated clinically and improved to normal level in all subjects. All subjective score forms improved significantly one year after surgery. Functional VAS improved from a mean of 29.7 to 59.2 (p \ 0.01), Tegner improved from a mean of 1.8 to 3.2 (p = 0.033), Lysholm improved from a mean of 51.6 to 78.2 (p \ 0.01), Noyes score improved from 51.9 to 74.3 (p \ 0.01) and IKDC subjective knee form 39.1 to 56.6 (p \ 0.01). Preoperative clinically assessed laxity did not correspond to laxity as found on varus stress radiographs. However, when clinical residual laxity was found one year postoperatively, it was found on varus stress radiographs as well. Conclusions: We conclude that laxity of the posterolateral corner can be treated effectively by this anatomic reconstruction technique according to Laprade et al.; we were able to improve the varus and the external rotational stability but the reconstructed knee did not become as stable as the non injured knee and hence patients did not return to their original level of activity.

P24-184 The use of platelet growth factor in the treatment of chronic Achilleus tendonitis Diotti E.1, Manzini C.2, Pozzi L.2, Gifuni P.2, Pelucchi R.2 1 Azienda Ospedaliera di Desio e Vimercate, Medicina Fisica e Riabilitazione, Giussano, Italy, 2Azienda Ospedaliera di Desio e Vimercate, Ortopedia e Traumatologia, Giussano, Italy Objectives: Aim of the study: In this study we evaluated the use of platelet-rich-plasma (PRP) as a treatment of the chronic tendinopaty of the Achilles tendon. Purpose of this study was to get an immediate reduction of inflammation and pain and therefore a best compliance of the patient and an early beginning of the rehabilitation protocol for a precocious return to the sporting activity and the specific athletic practice. Methods: Thirty young sportsmen range between 18 and 31 years old (soccer players, tennis players, runners) with chronic tendinopathy and/or tendinosis of the Achilles tendon were enrolled in this study. All patients before the infiltration treatment obtained with Recover Kit, have been ecographically submitted and after the injection, they followed a suitable rehabilitation protocol. All the patients have been submitted to a therapy in the precedents months with physical therapy (laser therapy, US, ionophoresis, shock-wave) and physiokinesitherapy in the preceding months without benefit. The patients have been submitted ecographically to pre-operating evaluation and 30 -90 days, and they were evaluated clinically and with VAS score for the assessment of the pain.

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 Rehabilitation protocol: The patients begins a stretching program (Level One) from the third day after the injection and progress to a strengthening program (Second level) three weeks later. The aim of the rehabilitation program keeps in mind the necessity to restore tendon functionality. This pass through a job of reconditioning with increasing loads and the correction of the predisposing factors. (The load increase is performed both increasing the external resistance and the speed of movement). From the third week, in accord with literature studies, eccentric exercises have been proposed: the eccentric job represents a great mechanical load in comparison to the concentric and the isometric ones and therefore it is able to maximally solicit the muscle-tendineous structure. From the firth week, proprioceptive exercises were introduced. Results: Ecography: 30 days post-injection, we found an initial reduction of the hypoechogenic spots. 90 days post-injection, in the 87% of the cases we found ecographic signs of improvement of the tendon’s ecostructure. VAS Score: 30 days after the procedure, 78% of the patients reported a pain reduction; 90 days after the treatment, 92% of the patients reported a pain reduction. Conclusions: The use of the GFs has allowed us to get an immediate postoperating reduction of the edema and the inflammation and subsequently a pain reduction. This allows a precocious return to the specific athletic practice thanks to a best compliance of the patient to an early beginning of the rehabilitation treatment and above all an immediate reduction of the pain. To the satisfactory results, the great advantage of this technique is the nearly complete absence of complications and collateral effects. Our study evidence the effectiveness of the infiltrative treatment with PRP as inductor of the process of tissue recovery. Further investigations and analysis on this innovative therapy are still necessary to verify and to confirm the real effectiveness of the PRP to promote tendinous regeneration. Because of his rational convincing and the therapeutic successes that we have gotten, we propose the PRP as valid therapeutic option in the patients with chronic not responsive tendinosis.

P24-232 Influence of testing position on the isokinetic Harmstring and Quadriceps performance by elite skiers Ziltener J.-L.1 1 Hoˆpitaux Universitaires Geneve, Geneve, Switzerland Objectives: During isokinetic knee measurements, mode of contraction and testing position are of great importance. This preliminary study has 2 objectives: – determine if two different testing positions (upright sitting and semireclined sitting) modify the torques for quadriceps (Q) and hamstrings (HS) – determine if the type of sport also affect the HS/Q ratios (HS conc / Q conc and HS ecc / Q conc). Methods: 10 adult international level skiers (age: 18-30 yrs), 20 young national level skiers (13-17 yrs), 15 young multisports athletes (13-17 yrs) and 15 adult active athletes (control group: 18-40 yrs) were enrolled. They were tested for the knee on an isokinetic device concentrically and eccentrically. Every test included the 2 described sitting positions. Results: – A significant difference for all groups between conventional concentric ratio and functional ecc/conc ratio was observed. – There was no difference for the functional ratio between adult professional skiers and young national skiers, in both testing positions. – By comparing young skiers with young multisports athletes and adult skiers with controls, the functional ratio decreased significantly among skiers, when measured in the semi-reclined position. Conclusions: – The functional ratio (HS ecc/ Q conc) is of great interest. – The hamstring eccentric torque is reduced in semi-reclined position, predominantly by skiers.

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 – Some significant differences are seen between skiers and symmetrical sports athletes, but not between professional adult skiers and young high level skiers, who showed the same muscle profile.

P24-303 Value of the clinical anterior drawer test of the talus in comparison to stress sonography in the diagnosis of ligament lesions of the ankle Wiebking U.1, Omar-Pacha T.1, Knobloch K.2, Hankemeier S.1, Krettek C.1, Jagodzinski M.1 1 Medical School Hannover, Department of Traumatology, Hannover, Germany, 2Medical School Hannover, Department of Plastic Surgery, Hannover, Germany Objectives: Ankle sprains and ligament tears are the most common conditions in sports traumatology. Correct diagnosis require assessment tools with high diagnostic accuracy. The aim of this study was to develop a measurement tool that can be used immediately after injury and to determine the diagnostic accuracy. Methods: The method was compared with clinical assessment including talar drawer and tilt tests and stress sonography. The new device investigated the talar drawer with 100N and maximum manual testing using a side to side evaluation. As gold standard, high resolution sonography was used to investigate the integrity of the lateral ankle ligaments.30 patients were recruited for this study with a mean age of 35,4 ±14,2 years. There were 15 patients with a rupture and 15 patients with an ankle sprain. Results: Sensitivity and specificity were as found as follows: The sensitivity for clinical investigation was 0,93 with an specificity of 0,71. Using a cut-off value from 3,95mm, the diagnostic gadgetry without using ultrasound showed a sensitivity of 0,8 and a specificity of 0,4; by contrast, ultrasound assisted gadgetry with a cut-off-value of 3mm showed a sensitivity of 0,26 and a specificity of 0,86. In the group of patients with a rupture the mean value measuring with ultrasound assisted gadgetry was 1,2 ± 2,2mm in contrast to 5,2 ± 1,8mm using the gadgetry without assisted ultrasound. Conclusions: Clinical examination under adequate analgesia remains the most reliable tool to investigate ligament lesions of the ankle. An arthrometer showed higher accuracy than stress ultrasound in the current study. Further clinical studies with higher case numbers are necessary in order to further evaluate arthrometric measurements following ankle injuries.

P24-308 High prevalence of GIRD in asymptomatic and oligosymptomatic world class junior javelin-throwers Brucker P.1, Beitzel K.1, Zandt J.2, Amereller M.2, Timpert K.3, Schwirtz A.2, Imhoff A.4 1 Technical University of Munich, Klinikum rechts der Isar, Department of Orthopedic Sports Medicine, Munich, Germany, 2Technical University of Munich, Faculty of Sports Science, Munich, Germany, 3 University Hospital Munich, Ludwig-Maximilian University, Department of Radiology, Campus Grosshadern, Mu¨nchen, Germany, 4 Technical University Munich, Department of Orthopaedic Sports Medicine, Munich, Germany Objectives: Glenohumeral internal rotation deficit (GIRD) is often diagnosed in the dominant shoulder of overhead athletes and been associated with the development of secondary shoulder lesions. Javelin is one of the throwing disciplines with repetitive maximal loading pattern of the throwing shoulder. However, there is a lack of clinical and isokinetic findings in high performance competitive junior javelin throwers. Therefore, the aim of the study was to evaluate clinical, isokinetic, and imaging findings in the throwing shoulder of world class junior javelin athletes compared to their non-dominant shoulder. Methods: Five asymptomatic and one oligosymtomatic world class junior javelin throwers with a mean age of 17 (16-18) years were included. The subjects perform javelin for 2,8 years and throwing activities for 5,3 years on average. The right arm was the dominant side for all athletes. No major trauma or surgery was detected in medical history. All athletes underwent

S333 questionnaire-based interview, clinical examination of both shoulders, and shoulder scoring systems (Constant Score, Rowe Score, Visual Analog Scale, Athletic Shoulder Outcome Rating Scale). In addition, bilateral shoulder isokinetic testing (ISOMED) and bilateral imaging were performed using a native 3T-MRI for exclusion of structural alterations. T-test was used for statistical analysis of the isokinetic testing. Results: 5 of 6 athletes demontrated distinct GIRD of equal or less or than 10 (1 with 10, 3 with 15, 1 with 25). Only 1 athlete (GIRD positive) was oligosymptomatic (Visual Analoge Scale of 5 at throwing activities), all other athletes were asymptomatic (Visual Analoge Scale of 0). All athletes showed excellent Constant Scores (mean 95,7 p.), Rowe Scores (mean 99,2 p.), and Athletic Shoulder Outcome Rating Scale (mean 99,2 p.). Peak torque in isokinetic testing revealed 104±20,5 and 97,7±15,7 Nm for concentric internal rotation and 46,3±13,7 and 45,6±9,8 Nm for concentric external rotation of the right shoulder in comparison to the adominant left shoulder, respectively. The difference between the right and left shoulder was not significant (p=0,57 for isokinetic internal rotation, p=0,92 for isokinetic external rotation). Interestingly, all athletes present cyst formation at the infraspinatus insertion area of the throwing shoulder. Conclusions: The results indicate a high prevalence of GIRD in asymptomatic and oligosymptomatic world class junior javelin athletes. Since isokinetic testing could not demonstrate a significant difference between dominant and adominant shoulder, the role of GIRD is unclear for quantitative shoulder strength analysis. In contrast, our study show that GIRD and structural alterations of the shoulder other than the tightness of the posteroinferior capsule may co-exist. GIRD may cause secondary structural changes of the shoulder even in the beginning of a throwing career. If so, an early intervention strategy for GIRD in young and asymptomatic throwing athletes must be postulated.

P24-456 Refractary patellar tendinopathy treated with platelet rich plasma inyections. A preliminary report of 8 cases Ruiz-Iban M.A.1, Diaz-Heredia J.1, Gonzalez-Liza´n F.1, Moros S.1, Del Cura Varas M.S.1 1 Hospital Ramon y Cajal, Dep. de Cirugı´a Ortope´dica y Traumatologı´a, Madrid, Spain Objectives: Patellar tendinopaty (or jumper0 s knee) is a frequent problem that affects active young adults. In some cases the different conservative treatment options are innefective and surgical treatment is considered. The purpose of this study is to determine if repeated intratendinous injections of platelet rich plasma (PRP) are effective for the treatment of these refractory cases. Methods: Eight consecutive patients (4 males and 4 females, mean age 24±5,9) who presented refractary patellar tendinopathies were included. All patients had presented symptoms for at least 6 months and had received treatment for at least 3 months. All patients had been subjected to activity limitation, physical therapy, NSAID0 s and laser and ultrasound therapy. In 3 cases corticosteroid injections had been used. The subjects were assessed before treatment and 3 months and one year later with a Visual Analoge pain Scale (0 to 100mm, VAS), the Victorian Institute of Sport Assessment Patellar tendinopathy assessment scale (VISA-P) and the Lysholm score. Treatment consisted of 3 infiltrations (one week apart) of 3 cm3 of PRP extracted from their own blood with the GPS system (Biomet, Warsaw, Indiana, U.S.A). The PRP was infiltrated at the level of the tender tendon and immediately behind the tendon at the proximal tendinous insertion and 1 cm distal to it through a single cutaneous puncture. Results: Of the 8 patients, 7 presented a significant increase (more than 20 points) in the VISA-P score and 1 did not present any noticeable improvement. No complications related to the injections were observed. The VISA-P score increased from a pretreatment mean of 29 ± 10.7 to 79 ±10.7 at one year (significant differences, p\0.001). A similar decrease was observed in the VAS pain score (pretreatment values of 75±28 to one year values of 21±19). There were not significant differences in the Lysholm score. Conclusions: PRP seems to be a possible alternative to surgical treatment in refractary patellar tendinopathy.

123

S334 P24-578 A radiographic study of the relationship between jumper’s knee and patella alta Koyama H.1, Takahashi M.1, Matsubara T.1, Hanada M.1, Furuhashi R.1 1 Hamamatsu University School of Medicine, Orthopaedic Surgery, Hamamatsu, Japan Objectives: Our previous radiographic study of Insall-Salvati Index (ISI) revealed the patella alta could not be evaluated accurately because ISI depends on the patella shape. The purpose is to investigate the relationship between jumper’s knee and patella alta using a radiographic index that is independent of the shape of the patella. Methods: Subjects comprised 23 jumper’s knees of 18 patients (10-20 years old, average 14.1 years old). Control subjects were randomly selected from patients with knee contusions (89 knees and subjects, 10-25 years old, average 19.1 years old). Then, 19 knees with post epiphyseal closure of the tibial tuberosity were chosen from the control subjects as an age matched control group. Radiographic evaluations were carried out using the epiphyseal closure of the tibial tuberosity, the patella alta (ISI, Blackburne-Peel Index (BPI) and Caton-Deschamps index (CDI)), and the elongation of the patella (Patella morphology ratio (PMR)). The ISI, which is based on the length of the patella, depends on the patella shape. On the other hand, the BPI and CDI, which are based on the length of the patellae articular surface, are independent of the patella shape. Results: 11 knees in the jumper’s knee group and 77 knees in the control group showed closure of the epiphyseal line of tibial tuberosity. ISI was 1.07, BPI was 0.91, CDI was 0.99 and PMR was 1.40 in the jumper’s knee group. Meanwhile, ISI was 1.03, BPI was 0.80, CDI was 0.92 and PMR was 1.38 in the control group. The BPI and CDI in the jumper’s knee group were significantly higher than those of the control group (P\0.01). In the control group, inverse-correlation was observed between the IPE, the CDI and the PMR and age. The younger subjects of the control group had higher BPI, CDI and PMR, and the age in the jumper’s knee group was lower than that of the control group. Hence we investigated the differences before and after closure of the epiphyseal line in both groups. However, comparison of the indices before and after the closure in the jumper’s knee group did not show any significant differences. There was no significant difference in the ISI between the jumper’s knee group and the control group after closure. Contrary to this, jumper’s knee BPI and CDI were significantly higher than those in the control group after closure (P=0.02 and P=0.03). In jumper0 s knee with patella alta, it is possible that ISI becomes low due to elongation of the patella. In this event, we believe that patella height in jumper’s knee should be evaluated with indices that are independent of the shape of the patella. Conclusions: After closure of the epiphyseal line of tibial tuberosity, BPI and CDI of the jumper’s knee group were significantly higher than those in the control group. Our study showed that the patella alta should be determined as the position of the patella as it is independent of the patella shape in radiographical evaluation. Patients with jumper’s knee had patella alta evaluated using a radiographic index independent of the shape of the patella.

P24-655 Role of angiogenesis in muscle derived stem cell mediated regeneration of injured medial collateral ligament Nishimori M.1, Matsumoto T.2, Ota S.2, Kopf S.3, Fu F.3, Ochi M.1, Huard J.2 1 Hiroshima University, Orthopaedic Surgery, Hiroshima, Japan, 2 Children’s Hospital of Pittsburgh & University of Pittsburgh, Growth and Development Laboratory, Department of Orthopaedic, Pittsburgh, PA, United States, 3University of Pittsburgh, Dept. of Orthopaedic Surgery, Pittsburgh, Pennsylvania, United States Objectives: Rapid revascularization of injured ligament is essential for early restoration of function. Our laboratory has shown that muscle derived stem cells (MDSCs) have a multilineage differentiation potential including endothelial cell lineages. We have also shown that MDSCs promote tissue repair mainly through neovascularization. Thus, we

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 hypothesized that transplantation of MDSCs as a reservoir of secreting molecule, could contribute to ligament healing via vasculogenesis/angiogenesis once transplanted into an injured ligament. Methods: Murine MDSCs obtained via preplate technique were retrovirally transduced to express: 1) vascular endothelial growth factor (VEGF) and nLacZ transgene (MDSC-VEGF), 2) soluble fms-like tyrosine kinase-1 (sFlt1, soluble VEGF-specific receptor/antagonist) and nLacZ transgene (MDSC-sFlt1) and 3) nLacZ transgene (MDSC). After a complete section of the midsubstance of the medial collateral ligament (MCL) in 8-week-old female immunodeficient rats, 5 x 105 cells were transplanted into the injured site using fibrin glue as a scaffold. Additionally to the three above mentioned groups (n=34 for each group), a control group was established, where only fibrin glue and phosphatebuffered saline (PBS) was applied. Animals were sacrificed and the MCLs were harvested for RT-PCR analysis at week1, for immunohistological evaluations at week 1 and 2 and for biomechanical testing and macroscopic assessment at week 2 and 4. Immunofluorescent staining: In order to detect the presence of the mouse cells in the rat, at 1 week after surgery double immunohistochemistry was performed with nLacZ and the murine-specific endothelial cell (EC) marker, Isolectin B4-FITC conjugate. We also performed immunohistochemical staining for mouse anti-rat CD31 antibody, a rat specific EC marker, to detect regenerated capillaries and/or neovascularity at week 2 after surgery. After staining, we also evaluated the capillary density to estimate the capillary numbers in 5 randomly selected fields (250x250lm) of the tissue at the peri-injury site of MCL by using Northern Eclipse software. The ratio of the capillary within the peri-injury sites was quantified using a previously described protocol. RT-PCR: One week after surgery, four knees from each group were selected for RT-PCR. RT-PCR analysis was performed with Rat vascular endothelial growth factor (rVEGF), Rat tenomodulin (rTeM), Rat collagen1A2 (rCol1A2) and Rat glyceraldehyde-3-phosphate dehydrogenase (rGAPDH). Results: Immunohistochemical staining at the injury site demonstrated that double positive cells for Isolectin B4 and b-galactosidase (b-gal) was higher in the MDSC-VEGF group compared to the other groups at week 1. Reverse transcriptase-polymerase chain reaction (RT-PCR) at the injury site demonstrated that molecular expression of rat tenomudulin (TeM) and VEGF were higher in the MDSC-VEGF group compared to the other groups at week 1. Capillary density at the injury site was significantly higher in MDSC-VEGF groups than the other groups at week 2. In addition, ligament healing assessed by macroscopic, histological, and biomechanical examination was significantly enhanced by MDSC-VEGF transplantation compared to the other groups. Conclusions: MDSC-VEGF has a potent vasculogenesis/angiogenesis mechanism in the MCL injury-induced environment. Our data strongly suggest that local transplantation of MDSC-VEGF might augment the ligament healing process by promoting a favorable environment through neovascularization.

P24-666 Virtual arthroscopy with MSCT in pre-operative assessment of athlete’s knee and shoulder in sports injury Walter F.1, Nuehrenboerger C.2, Page P.1, Pape D.3, Seil R.3 1 CHL Clinique d’Eich, Radiology, Luxembourg, Luxembourg, 2 CHL Clinique d’Eich, Sports Medecine, Luxembourg, Luxembourg, 3 CHL Clinique d’Eich, Orthopedic Surgery, Luxembourg, Luxembourg Objectives: To demonstrate and illustrate the usefulness of virtual arthroscopy before surgical arthroscopy of athlete’s painful shoulder or knee. Methods: 38 athletes (23 men, 15 women) aged 17-66 years (mean : 35.5) underwent CT-arthrography of the knee (22) or of the shoulder (16) before surgical arthroscopy. CT-arthrography was performed with 0.5mm collimation (Emotion 6, Siemens) and multiplanar reformations were systematically realised. Additional 3D virtual arthroscopic reconstructions

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 were also performed with the same data images on specific workstation (Wizard, Leonardo, Siemens). The localisation, severity, and extension of the different lesions were assessed with usual criteria. Datas were compared with surgical arthrocopic findings (Cohen0 s kappa test) and also with MR images when available. Results: There were 8 rotator cuff tears, 5 labral lesions and 3 cartilage lesions for the shoulders. Knees0 exams revealed 12 cartilage lesions and 10 meniscal tears. In all cases, there was a good agreement between CT arhrography results and surgical arthroscopic findings for all criteria analyzed. Moreover, virtual arthroscopy allowed to anticipate, to plan and to guide the arthroscopic treatment. Conclusions: Virtual Arthroscopy with MSCT is an innovative 3D reconstruction method wich can be useful in pre-operative assessment of athlete0 s shoulder or knee before surgical arthroscopy.

P24-707 Demonstrate the injury patterns of Alpine skiing and snowboarding in Kongsberg, Norway Ahrabi V.1, Østheim E.1, Ekeland A.2 1 Kongsberg Sykehus, Ortopedisk AVD, Kongsberg, Norway, 2Martina Hansens Hospital, Department of Orthopaedic Surgery, Oslo, Norway Objectives: During 5 seasons (2001 -2005), 430 patients with different injuries related to skiing were admitted to Kongsberg Hospital .The patients came from different ski areas around the city, the majority from Funklia Ski Center after passing the only county medical center or directly to us. Methods: We used questionnaires to collect the data from the patients and required the following information: name, age, sex, location, type av skiing activity and skill level. 81 patients related to alpine skiing and 224 related to snowboarding, all coming from Funklia ski center, had correctly completed supplemental survey forms and could be used in the study. Results: The female/male ratio was 37.5 /62.5% for snowboarders and 48/ 52% for alpine skiers. The commonest snowboarding injuries were fractures (56%), followed by contusions(35.%), sprain (3%) and dislocations (3%).The most frequent alpine injuries were contusions (43%), followed by fractures (40%), sprain (11%) and lacerations (4%). Snowboarders were statistically younger(19.5 years; range, 6-60 years) than skiers (21.8 years; range 9-41 years). The most common location of injury/body part for snowboarder was lower arm/wrist and for alpinists was knee region. The most frequently fractured site in skiers was the lower leg (27.3% of all fractures) followed by the clavicle (24.2%), and in snowboarders, it was the wrist (60.5% of all fractures). Except for advanced riders and skiers with 30% fracture rate, with elevated skill level snowboarders have higher fracture rate and skier have lower fracture rate, ending up 50% fracture rate among intermediary group i both groups. Conclusions: Snowboarders and Alpine skiers are equally prone to injury.Snowboarding accidents are typically less severe and show significantly different patterns than skiing accidents. For injury prevention wrist guards for snowboard riders and leg/shoulder protectors for alpine skiers are recommended.

P24-789 High-resolution MR imaging using a microscopy coil for tibial stress injuries Mammoto T.1, Mamizuka N.1, Hirano A.1 1 Mito Kyodo General Hospital, Orthopaedic Surgery, Mito, Japan Objectives: Medial tibial pain in athlete has been diagnosed as either a shin splint, chronic compartment syndrome or as a stress fracture. It is often difficult to distinguish physically between them. The aim of this

S335 study was to investigate the high-resolution MR imaging using a microscopy appearance of tibial stress injury and evaluate the relative involvement of bone and soft tissues. Methods: A total of 15 patients (male 7, female 8; mean age 16 years) with exercise induced tibial pain were evaluated. A microscopy surface coil (SIEMENS flex coil small) was put at the point with tenderness on medial tibial aspect. MR images were obtained on a 1.5T unit (SIEMENS Magnetom Symphony). The MR imaging protocol consisted of the protondensity weighted images, T2* weighted images, and the fat suppressed T2 (T2FS) weighted images. All images were obtained in axial plane (FOV 60X60). The image evaluation included abnormal signal of the periosteum, cortex, bone marrow and presence of a fracture line. Edema along fascial structures and involving muscle were noted. Results: For the 14 tibiae, high intensity area in periosteum was found in T2* and T2FS. Of the 14 with positive periosteal high intensity, 10 had positive findings showing high intensity within bone marrow. High intensity area along with fascia and muscles in flexor digitorum longus(FDL), tibialis posterior(TP) muscle was found in 3 patients. High intensity in bone marrow as well as periosteum was observed in the 5 with shin splints. In the 8 stress fractures, a hypertrophy of the periosteum, striations and cavity were found in the cortex. One showed high signal within FDL and TP without neither periosteal nor bone marrow abnormalities. Conclusions: The MR images have been shown to be sensitive method for the detection of tibial stress injury. This high-resolution MR imaging showed the detail of the changes in the tibiae. Periosteal high intensity was seen in 14 of 15 patients. In the shin splints, bone marrow abnormality with periosteal changes was found. Cortical abnormalities with periosteal and bone marrow changes were found in the stress fractures. These findings suggest periosteal intensity change is the initial reaction of the tibia to the stress, and then extend to the bone marrow, finally affected on the cortex.

P24-810 A case of restriction of ankle dorsiflexion after trauma of gymnastics Takigami J.1, Hashimoto Y.2 1 Yodogawa Christian Hospital Orthopaedic Surgery, Osaka, Japan, 2 Osaka City University Graduate School of Medicine, Orthopaedic Surgery, Osaka, Japan Objectives: Traumatic gastrocnemius contractures are quite rare. We present a rare case of sole gastrocnemius contracture after the trauma with gymnastics. Methods: The patient was a 17 years-old boy. His left leg was injured during gymnastics by hitting horizontal bar. He realized that his left leg was talipes equines after injury and had a medical examination but there was no neurological evidence in his leg. 6 months later after trauma, he consulted to our sports clinic. His symptoms were gait disorder with talipes equines and restriction of ankle dorsiflexion. Funiculus about 7cm length was palpated with tenderness in the proximal medial gastrocnemius. The dorsiflexion of the left ankle (range of motion)was -30 degrees with knee extension and 10 degrees with knee 120 degree flexion. Abnormal findings were not seen on Xray and blood test and electromyography. MRI indicated low intensity area about 50mm in T1-weighted images and T2-weighted images along the proximal medial gastrocnemius. We diagnosed talipes equines by sole traumatic gastrocnemius contractures and then performed gastrocnemius recession with Burk’s approach. Fascia of gastrocnemius was normal but heterotopic tendon (scar) about 60mm was found in proximal medial gastrocnemius. We excised it and identified the dorsiflexion of the ankle (range of motion) was 0 degree with knee extension. Histologic findings of this scar looked like normal tendon in hematoxylin and eosin stain and polarizing microscopes. This finding agreed with MRI finding. At three months after operation, the patient had no gait disorder and could heel-up gait and running without pain.

123

S336 At six months after operation, the dorsiflexion of the ankle (range of motion) was 20 degree with knee extension and no heterotopic tendon and scar tissue were found in the proximal medial gastrocnemius with MRI. Results: Only three cases have reported the traumatic gastrocnemius contractures in Japan. In all three cases, scar and talipes equines occurred after trauma and talipes equines of all cases improved after gastrocnemius recession and lengthening of the tendoachilles. The release of the gastrocnemius tendon was first described by Vulpius in 1913. Strayer further delineated the procedure and described its indications in 1950. Stephen et al reported isolated gastrocnemius release (Strayer procedure) was useful for clinically significant gastrocnemius equines contracture. In this case, we speculated that heterotopic tendon (scar) was generated in gastrocnemius by trauma of gymnastics because low intensity area in T1-weighted images on MRI and histologic examination of excised tissues looked like normal tendon. In operation, we performed only heterotopic tendon recession, because after the recession, dorsiflexion of the ankle was 0 degree with knee extension. This showed almost no contracture in soleus muscle. Since the operation, restriction of ankle dorsiflexion has never reccured but we need followup hereafter. Conclusions: We presented a rare case of sole gastrocnemius contracture after the trauma with gymnastics. We need follow-up hereafter.

P24-849 Open single stage complete reconstruction of traumatic knee dislocation in sport professionals - what can we expect? A mid to long-term follow-up study Hirschmann M.T.1, Amsler F.2, Mu¨ller W.3, Friederich N.F.1 1 Kantonsspital Bruderholz, Department of Orthopaedic Surgery and Traumatology, Bruderholz, Switzerland, 2Amsler Consulting, Biel-Benken, Switzerland, 3Riehen, Switzerland Objectives: A traumatic knee dislocation represents a serious injury, particularly for athletes having highest demands on their knee function. Our aim was to analyze the long-term outcome and return to sport after traumatic knee dislocation in elite athletes treated surgically according to a standardized treatment protocol and to identify predictive factors for a successful outcome. Methods: Retrospective consecutive study of all elite athletes treated surgically (n=26) underwent an open complete single stage reconstruction/ refixation or repair of the cruciates collateral ligaments, meniscal lesions, capsular structures including the posterolateral corner. Preinjury and postoperative activity level as well as return to sport were recorded. With a mean follow up of 10±8 years 24 patients (92% follow up rate) were evaluated for visual analogue scale (VAS pain, satisfaction), instrumented anterior-posterior laxity (KT-1000), International Knee Documentation Committee form (IKDC), American Knee Society Score (KSS) as well as Lysholm and Tegner score. Standard weight bearing and stress radiographs were performed. Results: 79% of patients (n=19) returned to their previous sport after a median time of 5.5 months (range 1.5-36 months) with eight of them returning to preinjury levels. 8% had a VAS pain [3. 13% of patients showed a flexion deficit [15 and 8% an extension deficit [10. The mean side-to-side differences for anterior and posterior laxity were 2.3mm (range 1-5mm) and 2.0mm (range 2-7mm), respectively. The total IKDC score was normal in 4, nearly normal in 12, abnormal in 6 and severely abnormal in 2 patients. The median Lysholm score was 91.8 (range 37-100). The median Tegner score decreased from 9 (7-10) to 7 (2-10). The median Knee Society score was 190 (range 162-200). The median radiological anterior/posterior side-to-side differences in 30 and 90 flexion were 1mm (range 1-6mm)/ 1 mm (range 0-11mm) and 1mm (range 0-7mm)/ 3 mm (range 0-11mm), respectively. Patients treated more than 40 days post-injury had a poorer outcome on the satisfaction and relative Tegner scores. This group was also less

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 successful in returning to sport compared to patients treated earlier. Financial support by the Swiss National Insurance Trust (SUVA) is greatly appreciated. Conclusions: Athletes treated by early open complete single stage reconstruction within 40 days from injury showed favorable results to those treated later. Although 2/3 of patients had good functional outcome scores and stress radiographs proved ligamentous stability and longevity, only 2/5 of athletes reached their identical pre-injury sport activity level.

P24-927 Simultaneous rupture of the anterior crutiate ligament and patellar tendon, a rare injury of alpine ski: 2 cases report Lefevre N.1, Cascua S.2, Herman S.3 1 Institut de l’Appareil Locomoteur Nollet, Paris, France, 2Centre de Traumatologie du Sport Paris-la De´fense, Puteaux, France, 3Centre Me´dico-chirurgical Paris V, Paris, France Objectives: The simultaneous rupture of the patellar tendon (PT) and the anterior crutiate ligament (ACL) is a rare ski injury. The initial diagnosis is difficult and the treatment is not well defined. We present 2 cases of knee injury at alpine skiing. Methods: The first case concerned a female of 43 years old after an alpine ski injury of the right knee. She presented severe knee pain, hemarthrosis and she was unable to walk. The initial clinical examination concluded of a isolated ACL tear. The treatment was functional with brace for 6 week but unsuccessful. The MRI on the 28 th day showed a complete rupture of the ACL, a discontinuity of the patellar tendon and also a partial tear of the lateral and medial meniscus. The patient had rapidly a first surgery to repair the PT, to regularize the remnants of the ACL and to treat the meniscus with suture. Eight months later, using a hamstring tendon graft, the torn ACL was reconstructed. The second case was very similar and concerned a female of 24 years old. The initial diagnosis was just a rupture of the ACL. But 10 days later, a MRI showed an associated rupture of the patellar tendon, ACL tendon, a partial tear of the lateral and medial meniscus and also a partial tear of medial collateral ligament. At first the surgical treatment repaired the PT and meniscus. Six months later, was realised the ligamentoplastie. Results: At 3 years follow-up, for the first case and 2 years follow-up for the second case, the functional result was satisfying. Conclusions: !The combined rupture of the PT and the ACL is very rare and unknown. Very few reports of such injury have been published. The initial diagnosis needs a MRI in emergency. Two surgeries separated by 6 months are required to avoid arthrofibrosis. Delayed reconstruction of the ACL, after the PT repair, is believed to provide a more satisfying functional outcome.

P24-1367 Complete paralysis of the quadriceps muscle caused by traumatic psoas hematoma: a case report Lefevre N.1, Bernard L.2, Herman S.3 1 Institut de l’Appareil Locomoteur Nollet, Paris, France, 2Centre du Sport, Toulouse, France, 3Centre Me´dico-chirurgical Paris V, Paris, France Objectives: The femoral nerve palsy secondary to compression of psoas hematoma in the pelvis is a well known adverse effect occurring with an overdose of antivitamin K. It has been described after hip and pelvic surgery. However post-traumatic cases are uncommon. Methods: We described a case of a 16 year-old male, who presented a lack of knee extension after a hip and knee trauma. The patient had a skateboard accident. He reported a knee injury against the patella and the

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 ipsilateral hip (direct impact on the greater trochanter). The initial clinical examination was normal, but a painful knee and hip. A simple rest was prescribed. Six weeks after the initial trauma, the patient consulted again for a lack of active knee extension. The initial discussed diagnosis was a patella tendon rupture after a direct trauma. The MRI who was normal didn’t confirm this diagnosis. A careful examination showed a complete paralysis of the quadriceps muscle due to a femoral nerve paralysis. An MRI of the pelvis showed a large psoas hematoma compressing the femoral nerve. Results: The patient had rapid surgery with evacuation of the hematome. The patient was reeducated for a year with gradual and complete recovery of the quadriceps. Electromyography performed at 3 months showed reinnervation about 50% of the territory of the quadriceps. A year later, the electromyogram was normal. Conclusions: This is an exceptional case of femoral nerve palsy secondary to post-traumatic psoas hematoma, without risk factors such as antivitamin K overdose. The diagnosis of psoas hematoma wasn’t initially evocated because of the knee trauma. Only the MRI of the pelvis allowed the diagnosis.

P24-1429 Bone bruise pattern on magnetic resonance imaging in anterior cruciate ligament injuries - where are they found Widjaya R.1, Lee Y.H.D.2, Chin Y.C.2, Chang H.C.3 1 Changi General Hospital, Orthopedic Surgery, Singapore, Singapore, 2 Changi General Hospital, Singapore, Singapore, 3HC Chang Orthopaedic Surgery, Singapore, Singapore Objectives: To investigate the incidence and pattern of bone bruising in ACL injuries on magnetic resonance imaging (MRI) of the knee. To differentiate the pattern of Bone bruise in isolated ACL injuries and multiligament injuries Methods: The MRIs of 75 patients with ACL injuries over a 6 month period were reviewed. The bone bruise pattern was correlated with the mechanism of injury, ie contact versus non-contact. Results: Our cohort had 43 isolated ACL injuries, 10 combined ACL & MCL injuries, 5 combined ACL & LCL injuries, 8 combined ACL& MCL& LCL injuries, 5 combined ACL & PCL injuries and 3 combined ACL, PCL & MCL injuries and 1 combined ACL, PCL & MCL injury. The most common site of bone bruising that occurs with isolated ACL injuries occurs in the lateral femoral condyle (73%) and the lateral tibia (32%), ie the lateral compartment of the knee. This is especially true with isolated non-contact mechanism of knee injuries. With combined multi-ligament with injuries, the pattern was that more bone bruising was noted in the medial compartment of the knee as well as in the patellofemoral joint. The mechanism of injury in such cases are usually contact injuries or from landing on the knee. (See table) Conclusions: The entity of bone bruises associated with ACL are significant as they can cause persistent knee pain. The most common site of bone bruising in isolated ACL injuries is in the lateral compartment of the knee. Recognition of the pattern will help with the understanding and rehabilitation of knee ACL injuries.

S337 P24-1452 Stress fracture of medial malleolus: results of operative treatment Heikkila¨ J.T.1, Sarimo J.2, Orava S.3 1 Mehila¨inen Sports Trauma Research Center, Hospital and Sports Clinic, Turku, Finland, 2Mehila¨inen Sports Trauma Research Center, Mehila¨inen Hospital and Sports Clinic, Turku, Finland, 3Mehilainen Hospital, Turku, Finland Objectives: To study the results of operatively treated medial malleolus stress fractures. Methods: All patients treated operatively for stress fracture of medial malleollus from 1995 until 2008 at our hospital. There were 9 patients, 4 females and 5 males. Mean age was 21 years, range from 17 to 33. They were all competitive high level track and field athletes or runners. Symptoms had persisted from 1,5 to 18 months (mean 4,4 months). The diagnosis was verified with MRI. All patients had cortical fissure of medial malleolus in MRI and one patient developed acute fracture of medial malleolus. The follow-up time ranged from one to eleven years (mean 6.2). The operation was performed with two parallel AO cancellous compression screws in 8 patients and with just one screw in one patient. The screws were removed not before 12 months. After the operation partial weight bearing and immediate mobilization of the ankle was allowed according to pain. After 4 to 6 weeks water running, bicycling and gym training was allowed, and running after 8 to 12 weeks and competitive activities 4 to 5 months after the operation. The result was considered excellent or good if the patient could return to sports with no complaints, moderate if he could practice sport, and poor if he was not able to do that. Results: The result for all patients was excellent or good in 7 patients, they returned to previous sporting level 4 to 5 months postoperatively. One patient was operated 18 moths after the onset of the symptoms when an acute-on-chronic fracture occurred. The fracture healed well in 3 months, but caused deformity of the medial corner of the distal tibia joint-surface. Arthrofibrosis complicated the healing, and arthroscopy of the ankle. Conclusions: MRI is necessary for the early diagnosis for patient with chronic pain in medial malleolus. Initial conservative treatment is recommended, but operation should be considered for non-responders after three months and for patients with a cortical fissure line in plain films or in MRI.

Injury prevention P25-86 The physical characteristics of the senior high school baseball players Hashiba T.1, Watanabe N.1, Kobayashi T.2 1 KKR Hokuriku Hospital, Rehabilitation, Kanazawa, Japan, 2 KKR Hokuriku Hospital, Orthopaedics, Kanazawa, Japan Objectives: The flexibility evaluation of the hip joint is important to prevent shoulder disorders in overhead athlete. The limitation of hip joint motion gives glenohumeral(GH) joint over load during throwing motion. Our hypothesis is that there may be difference in physical flexibility between the first grader and the second grader in the senior high school

Summary of bone bruise locations No of patients

Medial Femoral Condyle

Lateral Femoral Condyle

Both Femoral Condyles

Medial Tibia Plateau

Lateral Tibial Plateau

Both Tibial Plateaus

Patella region

Isolated ACL

43

12 (28%)

25 (58%)

3 (7%)

11 (26%)

26 (61%)

3 (8%)

1 (2%)

Multi-ligament pattern

32

12 (37.5%)

23 (72%)

7 (22%)

13 (41%)

16 (25%)

6 (19%)

4 (12.5%)

123

S338 baseball players. The purpose of this study is to examine characteristics in the difference of the school year of range of motion (ROM) in GH joint and hip joint, and relations of ROM in GHjoint at the throwing arm side and hip joint of both throwing arm side and non-throwing arm side from the aspects of quantity of exercise, experience, and the physical development. Methods: Asymptomatic 71 men who were a member of baseball club of three senior high schools in Ishikawa Japan were participated in this study. They were 35 first graders(15.2±0.4y.o), 36 second graders (16.1± 0.3y.o).The measurements of passive ROM were performed in throwing arm and non-throwing arm respectively. We measured the ROM of hip internal rotation(HI), hip external rotation(HE), shoulder internal rotation (2nd IR), and shoulder external rotation (2nd ER). Measurement of hip joint was in prone position with keeping the pelvis. That of GH joint was in supine position with keeping the scapular fixed .The average ROM in GH joint and hip joint at throwing arm and non-throwing arm in first graders and second graders were calculated respectively. Moreover the average grand total rotation angle in GH joint and hip joint were evaluated. Statistical analysis were performed using t-tests and the Pearson coefficient of correlation. Results: The ROM of the second graders were significantly smaller than that of the first graders in HI, 2ndIR, hip total arc(HTA), and GH total arc(GTA)in throwing arm side and non-throwing arm side (p\0.05). GTA in throwing arm and HTA at throwing arm side had correlation in the first graders(r=0.59, p\0.01). GTA in throwing arm and HTA at non-throwing arm side had correlation in the first graders (r=0.56, p\0.01).GTA in throwing arm and bilateral HTA did not have any correlation in the second graders. Conclusions: All of ROMs of the second graders were smaller than that of the first graders. In particular, GTA and HTA at throwing arm side and HTA at non-throwing arm side in the second graders were about 10 degrees smaller than that of the first graders. These results revealed that in the first graders with shoulder disorders close examination in hip joint must be needed with that in GH joint. On the other hand, in the second graders the ROM (flexibility) except for shoulder and hip joint have an effect on shoulder disorders.Further researches are necessary to prevent throwing shoulder injuries and disabilities in these adolescent baseball players.

P25-240 The effect of gender on force, muscle activity and frontal plane knee stabilisation during maximum eccentric leg-press exercise Liebensteiner M.1, Szubski C.2, Platzer H.-P.2, Burtscher M.2, Raschner C.2 1 Innsbruck Medical University, Innsbruck, Austria, 2 University of Innsbruck, Innsbruck, Austria Objectives: It was reported that tears of the anterior cruciate ligament (ACL) often occur during sports in a non-contact mechanism during cutting maneuvers with deceleration and landing from jumps which are eccentric tasks. Additionally, altered neuromuscular control (eg. decreased hamstrings co-contraction) and increased knee abduction (valgus alignment) were reported as pathomechanism. When we put together the above mentioned pathomechanisms on the one hand and the fact that ACL tears are more common in females on the other it might be speculated that women in general have altered kinematics and neuromuscular control of the knee, especially during eccentric contractions. So it was hypothesized that men and women show significant differences in key parameters of force, knee stabilisation and muscle activity when exposed to maximum eccentric leg extension. Methods: 13 females and 13 males were matched for age and physical activity. They were positioned in a force-measuring leg press system and had to perform maximum eccentric quadriceps contraction against a stable footplate when it was moved towards them and to rest the leg passively on the footplate when it was removed. This task was repeated 6 times. After five minutes, the same task was performed with an unstable footplate, which permitted rubber-damped inversion and eversion of the ankle joint. This was done to simulate functional conditions by challenging the ability of the subjects to stabilise adjacent segments of the leg. Earlier studies

123

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 indicated that such perturbational test conditions might be relevant in the field of ACL injuries. During the eccentric leg press task we simultaneously recorded force, frontal plane knee stabilisation (motion analysis) and muscle recruitment (surface electromyography). Muscle activity was determined for vastus medialis, vastus lateralis, semitendinosus, semimembranosus, gastrocnemius and peronaeus longus. Results: The ‘force stabilisation deficit’ (difference between maximum forces under normal and perturbed leg pressing) did not differ significantly between genders. Similarly, for the parameters of muscle activity and frontal plane leg stabilisation (mean leg alignment and amplitude of knee excursions) we did not ascertain significant differences between male and female participants. Conclusions: The findings of the current study are partly consistent with and partly in contrast to previous research. We suppose, that some of the discrepancies with other investigators could be related to the different test procedures. Whereas many others tested jumping and cutting manoeuvres, we decided for a very standardized exercise with an eccentric leg-press exercise with footplates of varied stability. This study is novel in that gender differences in parameters of force, muscle activity and leg kinematic were investigated during functional conditions of eccentric legpressing. In summary, no gender differences in muscle activity, frontal plane knee stabilisation and ‘force stabilisation deficit’ were observed during normal and perturbed eccentric leg-pressing. However, many previous studies revealed gender differences in kinematics and muscle activity during other activities like cutting and jumping.

P25-473 Hamstring - quadriceps ratio of Chinese population with chronic ACL rupture Chan W.L.1 1 Kwong Wah Hospital, Dept. of O&T, Hong Kong, Hong Kong Objectives: Hamstring-quadriceps (H/Q) ratio with a ‘‘normal’’ value ranges from 0.43-0.90 has been reported. There seems to be little consensus of a normative value for this conventional ratio. The hamstringquadriceps ratio between normal and ACL injured knees in local Chinese population is reported in the present study. Methods: 194 patients receiving ACL reconstruction from 2004 to 2009 were recruited for preoperative isokinetic assessment using Biodex dynamometer. The peak torques and mean power of hamstrings and quadriceps were measured and the percentage of quadriceps, hamstring deficit, and H/Q ratio was calculated at different angular velocities (60, 180, 300 deg/sec). Results: The quadriceps peak torque difference in injured and uninvolved knee were 17.84%, 14.75% and 12.89%, and for hamstring were 14,17%, 11,16% and 5.99% with respective to the different angular velocities mentioned. The mean power deficit of quadriceps in injured knee were 16.18%, 13.86% and 13.61% and of hamstring were 13.03%, 11.18% and 10.76%. The H/Q ratio (peak torque) of ACL deficient knee were 0.40, 0.56 and 0.45 and of normal knee were 0.47, 0.3 and 0.52. The H/Q ratio (mean power) of ACL deficient knee were 0.41, 0.33, 0.36 and of the normal knee were 0.35, 0.40 and 0.32. Conclusions: The H/Q ratio of both the normal and injured knees approached to lower spectrum of the ‘‘normal’’ value in our study. Hamstring deficits are more prominent in the local population and the results will serve as a baseline to re-evaluate our present regime of ACL reconstruction and rehabilitation.

P25-571 The epidemiology of injuries in American Flag Football (AFF): two-season injury-exploratory study results Kaplan Y.1 1 Jerusalem Physical Therapy and Sports Institute, Lerner Sports Center, Hebrew University, Jerusalem, Israel Objectives: Although AFF is considered a non-tackle sport, many moderate-serious contact-type injuries have been reported. Although one injury-exploratory study involving female athletes has been previously

Knee Surg Sports Traumatol Arthrosc (2010) 18 (Suppl 1):S123–S339 documented, there are no studies describing the epidemiology and prevention of injuries in this international sporting activity. Methods: 1791 amateur players (consisting of 4 leagues- mens, womens, male high school and co-ed) participated in a two-season prospective injury-exploratory study. 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: 1005 games were played over the two seasons. This amounts to a total of athlete-exposures = 1799955. There were 179 injuries recorded, with an Injury rate = 0.099 per 1000 athlete-exposures (95% CI: 0.08, 0.11). The average probability that any player will sustain at least one injury during the course of one season was 9.6% (95% CI: 8.29%, 11.02%). All injuries were traumatic in nature.Over 75% of injuries were caused by contact with the ground. 30% of the contact injuries were to the fingers, thumb and wrist, 20% to the knee, 16% to the head and face and 12% to the ankle. 41% were moderate injuries (8-28 days off play), whilst 17% were severe ([28 days). Conclusions: These results reflect both the high number of moderate to severe type-injuries (60%) in this so-called ‘‘non-tackle’’ sport. Despite the fact that most of the injuries (68%) resulted from either direct contact with the ground or involving another player, very few players (19%) used any form of protective equipment. Even though nearly one-third of all the injuries involved the wrist and hand, only 2 players (2.1%) used protective equipment related to prevent injury in this anatomical location, both of them after their injury. The men0 s league seems to be more aggressive, with a three times higher incidence of injuries compared to the women. These results form part of a two-season prospective, epidemiological study of injuries in the AFI, which will be followed by a two-season prospective, longitudinal randomized- controlled injury prevention trial over a further two-season period.

P25-978 Relationship between dynamic balance and core strength in young female handball and basketball players Shima Y.1, Kitaoka K.1, Miyata T.1, Ueshima K.1, Goshima K.1, Hayashi M.1, Takahashi R.1, Tsukagoshi S.2, Yoneda Y.2, Moriyama S.2, Tomita K.1 1 Kanazawa University, Orthopaedics Surgery, Kanazawa, Japan, 2 Kanazawa University Hospital, Physical Therapy, Kanazawa, Japan Objectives: A three-year prospective cohort study has been initiated in Japanese young female handball and basketball players to identify risk factors for non-contact ACL injuries. Although core strength has been proposed as a key component of the trunk control and dynamic balance, few studies have examined this relationship. To investigate the relationship between dynamic balance and core strength, we examined star excursion balance test (SEBT) and core strength test (CST) in young (15-16 y.o.) female handball and basketball players. Methods: 80 of 104 young female players (handball: 26 players, basketball: 55 players) were participated in this control study. The anterior, posteromedial, and posterolateral SEBT reach distances and limb lengths were examined, then the distance scores (cm) for each direction of the SEBT grid were averaged over the 3 trials and normalized to leg length (reach distance/leg length x 100 = percentage of leg length). Abdominal endurance test, right side-bridge, left side-bridge, and frontal-bridge as the

S339 CST were also examined and we divided the players into 5 groups according to their CST score. Results: No differences were found between handball players and basketball players. Although the players with good result in frontal-bridge test scored significantly better result in posterolateral and summed SEBT reach distances (103% vs. 97%, p=0.0035, 280% vs. 267%, p=0.0159, respectively), no significant differences in each SEBT reach distances were found between the groups according to CST score. Conclusions: Although there was certain tendency that the players with good core strength had better dynamic balance, we could not determine the significant relationship between dynamic balance and core strength. Since core strength is multifactorial, careful interpretation is needed and further investigations including an examination the interrelationship between lower extremity function and core strength should be carried out.

P25-1223 Prevention of muscle and tendon injuries: survey of training methods among basketball, handball and volleyball coaches in central Italy Cerulli G.1, Caraffa A.1, Ponteggia F.1, Conti V.2, Benvenuti E.2, Battaglioli A.1, Orlando D.1 1 University of Perugia, Physical Medicine and Rehabilitation Residency Program, Perugia, Italy, 2Italian Olympic Committee, Umbria Section, Perugia, Italy Objectives: To assess the training methods and consequently the knowledge and the application of injury prevention programmes. Methods: Coaches of 176 basketball, handball and volleyball teams (in total 2936 players, 1315 males and 1621 females : average age 15.6 years) in the Umbria region (central Italy) answered an anonymous questionnaire regarding training exercises (warm-up, stretching, weekly session). Additional information on prevention methods (pre-season quantitative evaluation, balance training, landing technique, sport specific movements, video analysis, rapid deceleration and cutting manoeuvre, proprioceptive exercises during rehabilitation) was obtained. Results: 86% of the teams performed 3 session of training a week (2 weekly sessions in 6%, 4 times a week in 6% of the teams while only in 2% of the teams there was daily training). The length of warm-up was 10 minutes in 39% of the teams; 15 minutes in 39%, 20 minutes in 18% and 30 minutes in 6% of the teams. Warm-up was performed exclusively without use of ball by 83% of the team. Stretching exercises were performed only before the training session by 37% of the teams, only after the session by 6% of the teams, before and after the session by 47% of the teams (in the remaining 10% of the teams no stretching exercises were performed). Only 2% of the teams performed routinely pre-season quantitative evaluation of the flexibility in the players. Conclusions: There is a great variability in the training methods. The flexibility of the muscle-tendon units is an important parameter for the performance of the athletes and also for the prevention of injury. In our opinion large prospective randomized controlled studies should be performed to establish the necessary quantitative and qualitative amount of warm-up and stretching exercises according to sport-specific movement, level of practice, age and sex of the athlete, both to improve performances and diminish injuries. Moreover, quantitative evaluation performed preseason, during and after the season in each athlete are useful to discover risk factors and consequently prevent injuries and also to help during and at the end of a rehabilitation period when the player should safely return to play.

123