Eur Spine J (2015) 24 (Suppl 6):S743–S800 DOI 10.1007/s00586-015-4131-7

ABSTRACTS

POSTER PRESENTATIONS EUROSPINE 2015

P2 THE INSERTION TECHNIQUE OF GUIDE WIRE FOR THE CERVICAL CANNULATED PEDICLE SCREWS Hisanori Ikuma, Ryoji Joko

CERVICAL SPINE Department of Orthopaedics Surgery, Kagawa Rosai Hospital, Marugame, Japan P1 MINIMAL INVASIVE APPROACH FOR VENTRALLY LOCATED INTRADURAL SMALL LESIONS IN THE CRANIOVERTEBRAL JUNCTION Sven O. Eicker, Klaus Christian Mende, Lasse Du¨hrsen, Nils Ole Schmidt Department of Neurosurgery, University Hospital HamburgEppendorf, Germany Background: Surgical management of lesions anterior to the neuroaxis in the craniovertebral junction (CVJ) and upper cervical spine is challenging. Most published reports focus on the treatment of huge lesions in this region. This technical note addresses the small lesions in the ventral aspect of the neuroaxis in the CVJ. Method: Data for the length of the posterior atlanto-occipital membrane and the posterior atlanto-axial ligament were analyzed on the basis of 100 computerized tomography. Between 01/2012 and 05/2014 six symptomatic patients with ventrally located intradural small lesions in the CVJ were treated in our institution using a minimal invasive dorsal approach. Results: Atlantooccipital distance ranged from 3 mm to 17 mm in supine neutral position CT scans (mean 8.98 mm). Atlantoaxial distance ranged from 5 mm to 17 mm (mean 10.56 mm) without significant difference between female and male (p = 0.14 atlantooccipital and p = 0.72 atlantoaxial). The six treated patients (mean age 54,7 years) were administered to our institution due to slight neurological pathology in terms of sensible deficit in the lower extremities during inclination, which disappeared after surgery. The postoperative course was in all six patients uneventful. The neuropathological findings confirmed a meningotheliomatous meningioma (WHO grade I) in five cases and an extramedullary cavernous hemangioma in one case. MRI confirmed complete resection of the lesions. Conclusions: Minimal invasive approach for ventrally located intradural small lesions in the craniovertebral junction provides a direct exposure to the lesion avoiding an anterior transoral and broad dorsal approach. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

Background: A cannulated pedicle screw has been widely used to prevent the displacement of pedicle screw for thoracic and lumbar spine. However it is not a common procedure for cervical spine. We report the insertion technique of the guide wire for the cervical cannulated pedicle screws. Methods: We assessed 26 cases have undergone to the posterior fixation of cervical spine using the cannulated pedicle screw with CT navigation system. For the first 2 cases, we made the screw hole through the pedicle for the guide wire using T-handle drill (3.0 mm) by the hand control method before the guide wire is inserted, and for the next 24 cases, a guide wire was inserted directly using a power tool. After the guide wire is located, a cannulated pedicle screw was inserted. Multidetector-row CT was performed postoperatively and the number of screws penetrated 2 mm and over was evaluated. Results: 104 screws have been implanted (9 were in T handle group, 95 were in the direct guide wire group). The total ratio of pedicle perforations was 5.8 % (6 screws). There were 3 screws in T handle group (33.3 %) and 3 screws in the direct guide wire group (3.2 %). All of these screws penetrated toward to the transverse foramen. Although we had experienced the high perforation ratio in the first 2 cases with T handle group, we could reduce it substantially to use the direct power tool insertion of guide wire. Conclusions: The direct insertion of guide wire using a power tool can reduce the friction power compared to a T-handle drill technique. The possibility of pushing down the cervical spine that can cause the misplacement of screw can be low in the direct guide wire group, and it can make the accuracy of CT navigation better and maintain. The using of guide wire for the cervical cannulated pedicle screw with power tool under CT navigation can be a safe method for cervical pedicle screwing. Disclosures: author 1: none; author 2: none. P3 INFLUENCE OF POSTERIOR CERVICAL CAGE ON CERVICAL FORAMINAL AREA Piotr Janusz, Kris Siemionow Department of Orthopaedics, University of Illinois at Chicago, USA Introduction: Foraminal stenosis is a common cause of cervical radiculopathy. Posterior cervical cages can indirectly increase foraminal area and decompress the nerve root.

123

S744 Purpose: The aim of this study was to assess the influence of a posterior cervical cage on the surface area and shape of the neural foramen. Methods: Thirty-five patients who underwent fusion with posterior cervical cage for cervical spondylotic radiculopathy were retrospectively enrolled into the study. A total of 69 posterior cervical cages were used. The patients had CT scans taken preoperatively, at 6 months, and at 12 months postoperatively. The measurements were performed, after orthogonal multi-planar reconstruction of cervical CT scans. Sagittal scans were angled 45 degrees, crossing through the middle of the two adjacent pedicles. The foraminal area, height, the distance between the infero-posterior edge of the superior vertebrae and the superior margin of the inferior facet (DSI) and the distance between the supero-posterior edge of the inferior vertebrae and the superior margin of the superior facet or the tip of implant were measured (DIS). Level of significance p \ 0.05. Comparison with repeated ANOVA was performed with P-value significance level \0.05. Results: Foraminal area and height were significantly greater following placement of the implant. The mean foraminal area before surgery was 0.3954 ± 0.091 cm2, 6 months after surgery 0.4344 ± 0.097 cm2 and 12 months after surgery was 0.4177 ± 0.091 cm2; p \ 0.0001. The mean height before surgery was 0.883 ± 0.11 cm, 6 months after surgery 0.943 ± 0.10 cm and 12 months after surgery was 0.924 ± 0.11 cm; p \ 0.0001. The mean DIS before surgery was 0.700 ± 0.11 cm, 6 months after surgery 0.733 ± 0.14 cm and 12 months after surgery was 0.711 ± 0.13 cm; p = 0.039. Decrease in DSI was not statistically significant. The mean DSI before surgery was 0.591 ± 0.14 cm, 6 months after surgery 0.589 ± 0.13 cm and 12 months after surgery was 0.568 ± 0.13 cm; p = 0.1559. Conclusions: Implantation of posterior cervical cages can increase foraminal area and may indirectly decompress the nerve roots. This technique can be useful in treating select patients with nerve compression and cervical radiculopathy secondary to spondylosis. Disclosures: author 1: employee; Company=Department of Orthopaedics, University of Illinois at Chicago; author 2: grants/research support; Company=DePuy, consultant; Company=DePuy, captureproof.com, Providence Medical Technologies, stock/shareholder; Company=Providence Medical Technologies, royalties; Company=Amedica, other financial report; Company=Globus, Providence Medical Technologies.

P4 COMPARATIVE STUDY ON THE ZERO-PROFILE CAGE WITH SCREWS SYSTEM AND STAND-ALONE RECTANGULAR PEEK CAGE SYSTEM-AUGMENTED FUSIONS AFTER SINGLE LEVEL ANTERIOR CERVICAL DISCECTOMY Jong-Tae Kim, Doo-Yong Choi Incheon Saint Mary’s Hospital, The Catholic University ot Korea, Incheon, Korea Purpose: A new Zero-profile Implant for Anterior Cervical Interbody Fusion is known as implant which may provide strong fixation and reduce the complications associated with other high profile cervical fixation systems. We evaluated and compared radiological results, clinical outcomes after the Zero-Profile cage with screws system and stand-alone rectangular peek cage system-augmented fusions in patients undergoing single level anterior cervical discectomy and fusion (ACDF).

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Methods: The two groups compared were: 30 cases of anterior interbody fusion with zero-profile cage with screws system (group A); 26 cases of anterior interbody fusion with stand-alone rectangular PEEK cage system (group B). Mean follow-up period were 28.3 months in group A and 30.2 months in group B. All patients could be followed-up for at least 12 months. Radiological (construct height and height loss which means disc height loss, segmental angle, cervical lordotic angle, especially, subsidence rate) and clinical assessments were performed preoperatively, immediately postoperatively, and at 1, 3, 6 months and final follow up postoperatively. Results: The mean construct heights (distance between upper endplate of upper body and lower endplate of lower body of the fused segment) were 30.1 mm preoperatively, 34.5 mm at 1 day, 33.3 mm at 1 months, 33.1 mm at 6 months and 32.5 mm at final follow-up after surgery in group A, and 34.0, 36.9, 35.8 and 33.4 in group B, respectively. In group B, early, progressive and significant disc height loss and delayed fusion processing were observed during postoperative follow- up period. Subsidence (defined as any settlement in disc height of at least 3 mm on lateral radiographs) rate were 10 % (3/30 patients) in group A, 26.9 % (7/26 patients) in Group B. Clinical results according to the Odom’s criteria were excellent and good in 90.0 % (group A), 77.0 % (group B). There were no significant complications associated with both systems except a case of minor (2.5 mm) pull-out of left lower screw of zero-profile system. Conclusions: In this comparative study, the authors demonstrated that implantation of the zero-profile cage with screw systems maybe have the advantages of prevention of significant undesirable subsidence and related complications and ameliorate the clinical outcome compared to implantation of the stand-alone rectangular cage. Disclosures: author 1: none; author 2: none.

P5 THE EFFECT OF HUBBING ON THE PULL-OUT STRENGTH OF LATERAL MASS SCREWS IN THE CERVICAL SPINE: A BIOMECHANICAL EXPERIMENT Ki-Hyoung Koo, S. Tim Yoon, Jung-Hoon Kim, William C. Hutton, Keun-Tae Cho Department of Orthopaedic Surgery, Dongguk University Ilsan Hospital, Gyeonggi, Korea Study design: This was a cadaveric biomechanical experiment. Objective: To compare the pull-out strength of polyaxial screws that are either hubbed or not hubbed when inserted into the lateral mass. Summary of background data: It has been shown in a study on pedicle screws in the thoracic spine that ‘‘hubbing’’ the head of the screw against the dorsal laminar cortex results in significantly lower pull-out strength of the screws. Materials and methods: Fifteen segments of the human cervical spine (from C3 to C7) were prepared. Polyaxial screws 3.5 mm in diameter were used. On one side screws 12 mm in length were inserted until the screw head touched the lateral mass; they were then turned 2.5 more times until they were fully hubbed (hubbed screws). On the other side screws 14 mm in length were inserted until the screw head just touched the lateral mass (nonhubbed screws). The 2 mm difference in length was to ensure that the screws were buried to the same length. All screws inserted into the lateral masses underwent tensile pull-out by applying a tensile force down the long axis of the screw. The difference in pull-out strength between the 2 groups was evaluated using a nonparametric paired test (the Wilcoxon signed rank test), which compared side to side on each vertebra. Results: One specimen was excluded because of cement breakage during the biomechanical test. A total of 14 vertebrae were tested.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Four vertebrae in the hubbed group showed small fractures or cracks around the screw hole after screw insertion. In a side to side comparison, the hubbed screws had significantly lower pull-out strengths as compared with the nonhubbed screws (P = 0.033). Conclusions: Hubbing of lateral mass screws lowers the potential pull-out strength of the screws as compared with the pull-out strength of nonhubbed screws. Thus, hubbing of lateral mass screws, on the basis of the parameters applied in this study, is not recommended. Disclosures: author 1: none; author 2: grants/research support; Company=Depuy Synthes Spine, Biomet, consultant; Company=Stryker, Biomet, Meditech, stock/shareholder; Company = Meditech, Medyssey, royalties; Company=Stryker, Biomet, other financial report; Company=AOSpine, Biomet, Nuvasive; author 3: none; author 4: none; author 5: none.

P6 POLYETHERETHERKETONE CAGE WITH DEMINERALIZED BONE MATRIX CAN REPLACE THE ILIAC CREST AUTOGRAFTS FOR ANTERIOR CERVICAL DISCECTOMY AND FUSION IN SUBAXIAL CERVICAL SPINE INJURIES Jung-Kil Lee, Moon-Soo Han Department of Neurosurgery, Chonnam National University Medical School and Research Institute of Medical Sciences, Gwangju, Korea This study was designed to compare clinical and radiologic outcomes of patients with subaxial cervical injury who underwent anterior cervical discectomy and fusion (ACDF) with autologous iliac bone graft or polyetheretherketone (PEEK) cages with demineralized bone matrix (DBM). Autologous iliac crest grafts were used in 33 patients (Group I), whereas 37 patients received ACDF using PEEK cage filled with DBM (Group II). Plain radiographs were used to assess bone fusion, interbody height (IBH) and segmental angle (SA), overall cervical sagittal alignment (CSA, C2-C7 angle), and development of adjacent segmental degeneration (ASD). Clinical outcome was assessed using a visual analogue scale (VAS) of neck pain and Frankel grade. The mean follow-up duration was 28.9 and 25.4 months in group I and group II, respectively. All patients of both groups achieved solid fusion during follow-up period. The IBH and SA of fused segment, and CSA were well maintained in group II compared to group I. Radiologic ASD was developed in 9 cases of group I and 2 cases of group II, respectively. There were no statistical significant differences of the VAS scores for neck pain and Frankel grade. This study showed that the PEEK cage filled with DBM and plate fixation is at least as safe and effective as traditional ACDF using autograft with well maintenance of cervical alignment. With advantages of no donor site morbidity and no graft-related complications, the PEEK cage filled with DBM and plate fixation provides a promising surgical option for treating traumatic subaxial cervical spine injuries. Disclosures: author 1: none; author 2: none.

S745 Purpose: To evaluate cervical adjacent segment range of motion (ROM) and instantaneous center of rotation (ICR) after anterior cervical discectomy and self-locking cage fusion and to explore the influence on cervical kinematics. Methods: 122 patients underwent anterior cervical discectomy and self-locking cage fusion between Jan. 2012 and Dec. 2014, including 56 males and 66 females with ages ranging from 40 to 65 years (average 55 years). 52 cases underwent 2-level and 70 cases were 3-level. The follow-up averaged 6 months (range 1-21 months). Flexion and extension cervical X rays were obtained in the standing position before surgery and at the final follow-up. Several parameters, including cervical adjacent segment ROM and ICR, total ROM, flexion and extension ROM, VAS, JOA and NDI were measured. Results: Cervical total ROM was 49.22 ± 10.90° and 29.72 ± 9.02° for baseline and final follow-up respectively with statistical significance (P \ 0.05). Cervical flexion ROM was -4.25 ± 2.55° and -4.31 ± 3.09° for baseline and final follow-up respectively with statistical significance (P \ 0.05). Cervical extension ROM was statistical significance among pre-operative and final follow-up. The cervical flexion ROM and extension’s were significant after operation. Cervical adjacent segment superior ROM and inferior ROM showed no significant change between preoperation and postoperation respectively. There were significant between Cervical adjacent segment superior ROM and inferior ROM after operation. No statistical differences were observed among adjacent segment ICR compared the post-operative with pre-operative. Cervical adjacent segment inferior ICR was located closer to the superior half of the lower vertebral body height than superior. VAS was 8.98 % ± 5.75 % and 0.80 % ± 0.50 % for baseline and final follow-up respectively with statistical significance (P \ 0.05). JOA was 12.32 % ± 2.47 % and 16.62 % ± 0.05 % for baseline and final follow-up respectively with statistical significance (P \ 0.05). NDI was statistical significance among pre-operative and final follow-up. Conclusions: After anterior cervical discectomy and self-locking cage fusion, patients with clinical symptoms were improved vastly, reducing the cervical total ROM of, increasing cervical flexion ROM, decreasing extension ROM obviously. Cervical adjacent segment ROM did not change with comparing to preoperation. Adjacent segment ICR of final follow-up did not change obviously, cervical adjacent segment biomechanical conditions did not change obviously. Disclosures: author 1: grants/research support; Company=National Natural Science Foundation of China (NO. 81171764); author 2: grants/research support; Company=Nation Natural Science Foundation of China (NO. 81171764); author 3: grants/research support; Company=National Natural Science Foundation of China (NO. 81171764).

P8 ANTERIOR SCREW FIXATION IN TYPE II ODONTOID FRACTURES: COMPARISON BETWEEN THE YOUNG AND THE ELDERLY Woo-Kie Min

P7 KINEMATIC ANALYSIS OF THE CERVICAL SPINE AFTER ANTERIOR CERVICAL DISCECTOMY AND SELF LOCKING CAGE FUSION AT AN ADJACENT LEVEL Baoge Liu, Dongmei Li, Yao Zhang Beijing TianTan Hospital, Capital Medical University, Beijing, China

Kyungpook National University Hospital, Daegu, Korea Background: Optimal surgical treatment in elderly patients with type II odontoid fracture, which is the most common injury of the axis, is still controversial. Purpose: The purpose of this study is to compare the outcomes of treatment between age groups in type II odontoid fracture using anterior screw fixation.

123

S746 Study design/setting: A retrospective study to validate the usefulness of anterior odontoid screw fixation, especially in the elderly with type II odontoid fracture. Patient sample: Twenty-nine patients who had Type II odontoid fracture were treated consecutively by anterior odontoid screw fixation. The patient group comprised 23 men and 6 women. There were 13 patients aged 60 years or older (Group 1) and 16 patients younger than 60 years of age (Group 2). Outcome measures: The medical records, simple X-rays and CT scans of all patients were reviewed and fusion rate, union time, incidence of perioperative complications, neurological outcome, and mortality were compared between two groups. Methods: Patients were grouped as follows: Group1: 60 years or older (mean age, 64.5 yr); and Group2: younger than 60 years (mean age, 38.4 yr). Radiological and clinical follow-up were performed on each group and compared. Statistical analysis was performed with Fisher’s exact test. A P value of \0.05 was considered statistically significant. Results: All patients were treated with anterior odontoid screw fixation by use of one compression screw. Mean follow-up was 18.3 months. Fusion rates were 77 % in group 1 and 81 % in group 2. If follow-up studies revealed pseudoarthrosis, additional dorsal fixation with transarticular C1-C2 screws or C1-C2 posterior screw-rod fixation with bone graft was performed. Mean union time was 17.1 weeks in group 1 and 14.4 weeks in group 2. Neurological status at admission and after treatment was similar in both groups. Statistical analysis did not show significant differences for other factors between two groups except union time. Conclusions: Outcome after anterior odontoid screw fixation in type II odontoid fracture, especially in the elderly, is comparable to that of the younger patients. Therefore, anterior odontoid screw fixation can be a useful method to treat the type II odontoid fracture in the elderly. Disclosures: author 1: none.

P9 RADIOGRAPHIC DIMENSIONAL ANALYSIS OF OPEN DOOR LAMINOPLASTY WITH PLATE IN CERVICAL SPONDYLOTIC MYELOPATHY Woo-Kie Min Kyungpook National University Hospital, Deagu, South Korea Background context: Many surgeons perform open door laminoplasty in cervical spondylotic hinge angle and those of relation via computerized tomography (CT) in single door cervical laminoplasty with pre-contoured plate. Purpose: To analyze radiological dimensional change of single door cervical laminoplasty with pre-contoured plate using computerized tomography (CT) scan. Study design/setting: Retrospective case series. Patient sample: 23 patients (18 males and 5 females) with multilevel cervical spondylotic myelopathy who underwent single door cervical laminoplasty were included in the case series. Outcome measures: A total of 81 laminae were evaluated with preoperative and postoperative CT. Measurement of dimension and diameter were performed with CT axial image. Methods: CT analysis were performed for measurement of dimension, diameter, hinge angle. Both hinge and open side dimension, change of dimension and diameter according to hinge angle were also investigated. For analysis of dimensional increment according to hinge angle, it were categorized into Group A (0-10°), B (10-20°), C (20-30°) with range of hinge angle and it were divided into both group which are below 10° (group I) and above 10° (group II).

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Results: Both side demonstrated different increase rate. Side of open increased to 162.2 ± 27.6 mm which means a 62.5 % rate of increase, whereas side of hinge demonstrated expansion to 141.1 ± 21.8 mm which is relatively a less rate of 44.4 % in comparison to the opposite side.(P=0.000). The mean of hinge angle in postoperative CT were 17.2 ± 8.8°. Group C exhibited the highest rate of increase (56.3 %) in dimension among Group A, B, C (p = 0.001). On the contrary, there was no definite difference of increase rate in diameter between range Group B and range Group C (p = 0.081). Increase rate of diameter was 38.8 % in group I and 46.5 % in group II (p = 0.039), respectively. Dimension of spinal canal increased by 41.2 % in group I and 53.1 % in group II (p = 0.033). Conclusions: Single door cervical laminoplasty with pre-contoured plate enables the spinal cord to be properly decompressed while increasing dimension (50.5 %) and diameter of spinal canal (4.46 mm). Laminoplasty is recommended with hing angle range from 10 to 20°. Increase of dimension in symptomatic side (open side) is higher than those of non symptomatic side (hinge side), which suggest that opening of symptomatic side have advantage compared to opposite side in terms of dimension and decompression. Disclosures: author 1: none.

P10 EFFECT OF PREOPERATIVE SAGITTAL BALANCE ON CERVICAL LAMINOPLASTY OUTCOMES IN ELDERLY PATIENTS AGED ‡ 65 YEARS Yasushi Oshima Department of Orthopaedic Surgery, The University of Tokyo, Japan Introduction: Sagittal imbalance increases with age. Also, previous studies have reported that the surgical effect of cervical laminoplasty can be affected by age. Because sagittal imbalance can affect quality of life, a higher proportion of those patients with sagittal imbalance may be one of the causes of poorer surgical outcomes in the elderly. The purpose of this study was to investigate the effect of preoperative sagittal balance on cervical laminoplasty outcomes in elderly patients aged C65 years. Materials and Methods: We evaluated the cervical laminoplasty outcomes of 92 patients. Radiographic analysis included the measurement of cervical Cobb angles and range of motion (ROM) between C2 and C7 and the C7-sagittal vertical axis (SVA). Primary outcomes were Japanese Orthopedic Association (JOA) scores, the Short Form 36 (SF-36) Health Survey [physical and mental summary scores (PCS and MCS, respectively)], and the Neck Disability Index (NDI). Results: There were 61 men and 31 women with a mean age of 65 years (range, 32-82 years), and the mean follow up period was 27 months (range, 12-84 months). The C7-SVA was B5 cm in 65 patients and [5 cm in 27. Of the 51 patients aged C65 years, the C7SVA was B5 cm in 30 and [5 cm in 21. There were no significant differences in age, sex, C2/7 Cobb angle, and ROM between the two groups. The pre and postoperative PCS and JOA scores were significantly poorer in patients with a C7-SVA of [5 cm than in those with a C7-SVA of B5 cm. However, the improvement rate of those scores after the surgical intervention was not significantly different between the two groups. Discussion: Among the patients aged C65 years, the presence of sagittal imbalance itself may have led to the poorer preoperative and postoperative scores in those with a C7-SVA of[5 cm, independent of the degree of myelopathy. However, the improvement rate in each outcome was not significantly different, although the postoperative scores themselves were inferior in patients with a C7-SVA of[5 cm, which may indicate that the effect of surgical intervention itself on myelopathy was not

Eur Spine J (2015) 24 (Suppl 6):S743–S800 affected by sagittal imbalance. Nonetheless, the influence of sagittal imbalance on surgical outcomes should be considered in elderly patients both pre and postoperatively, because such patients might have poorer scores, independently of the degree of myelopathy. Disclosures: author 1: none.

P11 DOES PREOPERATIVE NECK PAIN DECREASE AFTER CERVICAL LAMINOPLASTY? Yasushi Oshima Department of Orthopaedic Surgery, The University of Tokyo, Japan Introduction: The purpose of this study was to determine whether the intensity of preoperative neck pain, but not scapular pain, decreases after cervical laminoplasty in patients with cervical myelopathy. Methods: We retrospectively evaluated the outcomes of 93 patients (60 men and 33 women) who underwent cervical laminoplasty for cervical compression myelopathy. The mean follow-up period was 25 months. Primary outcomes were Japanese Orthopedic Association (JOA) scores, numerical rating scale (NRS, 0-10) for each part of the body (neck pain indicates area 1), the Neck Disability Index (NDI), the Short Form 36 (SF-36) Health Survey [physical and mental summary scores (PCS and MCS, respectively)], and satisfaction with treatment. Cervical alignment (C2-7 Cobb angle), range of motion (ROM), C7 slope, and cervical tilt were evaluated as radiographic parameters. A postoperative change of C2 in NRS for neck pain was considered significant. Results: The mean pre- and postoperative JOA scores were 10.1 and 13.3, respectively, with a mean recovery rate of 46 %. The average pre- and postoperative NRS for neck pain was 3.4 and 3.0, respectively. Neck pain improved in 30 patients and deteriorated in 24. We then focused on the 47 patients with a preoperative NRS of C3 for neck pain; 23 showed improvement (improved group) and 24 showed deterioration or no change (not improved group). A comparison of surgical outcomes between the improved and not improved groups showed JOA score recovery rates of 56 % and 44 %, respectively. NRS for the arm, leg, and neck and PCS showed better scores after surgery in the improved group. Satisfaction with treatment was significantly better in the improved group. With regard to radiographic parameters, a greater number of patients in the not improved group had a larger C7 slope and cervical tilt. Conclusions: Of the patients with preoperative neck pain, 50 % showed improvement after cervical laminoplasty. Although we cannot completely rule out the possibility of improvement in radiculopathy, these patients had better surgical outcomes in general, which indicates that the decrease in neck pain was related to an improvement in myelopathy. On the other hand, 50 % patients showed no change or worsening of preoperative neck pain and had radiographic malalignment, which suggests the involvement of posture. Therefore, laminoplasty should be avoided in patients with neck pain and cervical malalignment. Disclosures: author 1: none. P12 OUTCOMES AND REVISION SURGERY OF FRENCH DOOR LAMINOPLASTY Itaru Oda, Eihiro Murota, Hirohito Takeuchi, Masaru Suzuki, Masanori Fujiya Hokkaido Orthopaedic Memorial Hospital, Sapporo, Japan

S747 Although French door laminoplasty is one of the standard procedures for cervical myelopathy, there have been only a few reports of revision surgery for failed laminoplasty. The objectives of this study were to evaluate the outcomes of French door laminoplasty and to investigate the causes and results of revision surgery for recurrent myelopathy. Methods: From April 2009 to March 2013, 84 of 256 patients who underwent French door laminoplasty were involved in this study. Inclusion criteria were patients with functional X-rays, CT, and MRI at more than 6 months follow-up. There were 66 male and 18 female. The average age at surgery was 63.3. The primary disease were CSM in 70 patients and OPLL in 14 patients. Clinical evaluations included C2-7 sagittal angle, C2-7 range of motion (ROM), and lamina closure defined as loss of more than 5 degrees of lamina-vertebral angle on CT. Rate and causes of revision surgery were also evaluated. Results: Average C2-7 angle and ROM showed significant decrease after surgery and remained unchanged at the final follow-up. Lamina closure were observed in 43 % of the patients. Between the lamina closure and the non-lamina closure groups, no significant difference was detected in age, gender, primary disease, follow-up period, surgical levels, and average C2-7 angle except for average C2-7 ROM. In the lamina closure group at the final follow-up, C2-7 ROM showed significant decrease compared with preoperative one. Revision cases were 2 of 256 patients, so revision rate was 0.8 %. The causes of recurrent myelopathy included kyphosis, instability, epidural scar formation, and lamina closure. In both revision cases, recurrent myelopathy was not caused by a single factor, and posterior decompression and fusion with pedicle screws was performed. JOA score showed improvement in the both cases after revision. No major complication including neurovascular injury was observed. Discussion: Lamina closure following French door laminoplasty were observed in 43 % of patients, however, most cases were asymptomatic. Although risk factors of lamina closure were still unknown, the lamina closure group showed significant decrease in postoperative C2-7 ROM. When multiple factors such as kyphosis, instability, epidural scar formation, and lamina closure occur simultaneously, they may result in recurrent myelopathy. In this study, all the revision surgery were performed posteriory and showed excellent results. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

P13 ANATOMICAL EFFECT OF ROTATION ON ATLANTOAXIAL JOINT: A PILOT MRI STUDY Narendra Rath, Kathleen Lyons, Xin Yang, Len Nokes, Michael J. H. McCarthy University Hospitals of Wales, Bridgend, UK Background: There is a paucity of MRI studies looking at the anatomical changes that occur during spine motion especially at the atlantoaxial joint in the cervical spine (C1-C2). The majority of the literature focuses on range of movement and does not investigate into isolated rotation and subluxation at atlantoaxial motion segment. Aim: To provide normal variant data for better understanding of effect rotational at atlantoaxial joint. MRI based pilot study. Method: Ten volunteers with inclusion criteria male, age 20-30 years, height 5’6’’ - 6’, BMI 20-30 and no previous back problems. The EQ5D, Neck Disability Index, Oswestry Disability Index, PHQ 9 and GAD 7 score were within normal range in the sample population. MRI based radiological measurement of rotation of C1 over C2 vertebra and subluxation of lateral masses of C1 over C2.

123

S748

Eur Spine J (2015) 24 (Suppl 6):S743–S800

Standard supine MRI cervical spine in neutral and right and left maximum rotation of the head was performed using 1.5 tesla scanner. Sagittal and axial images were analysed by spinal surgeon and musculoskeletal radiologist. Results: Mean age 22 years (range 20-25). The mean rotation of C1 over C2 on right and left rotation was 31 degrees (95 % CI = +/3.3). The mean forward subluxation of opposite lateral mass to side of rotation is 66 % (95 % CI = +/- 2.8), whereas the same side lateral mass subluxes about 80 % (95 % CI = +/- 2.4) backwards. Conclusion: The atlantoaxial joint accounts for a large proportion of total neck rotation and demonstrates more than 60 % forward and 80 % backward subluxation. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: consultant; Company=Globus medical education.

Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

P14 CERVICAL SPINE FRACTURE IN PATIENTS WITH DIFFUSE IDIOPATHIC SKELETAL HYPEROSTOSIS. A REPORT OF 7 CASES

Objective: The majority of patients sustaining cervical spinal cord injuries (SCI) are over the age of 50 years in Japan. Life expectancies for persons with SCI are known to be lower than those with no SCI. Identifying predictors of the prognosis following SCI is important to make treatment decision. In the present study, we examined the impact of patient age, extent of neurological injury, diffuse idiopathic skeletal hyperostosis (DISH), and other factors on the life prognosis with traumatic cervical SCI. Methods: Between 1999 and 2013, 126 patients underwent surgery for cervical spinal cord injury (101 men and 25 women; mean age, 64 years). Nineteen patients had upper cervical spine injuries, while 107 patients had subaxial cervical spine injuries. After calculating the survival rate by using the Kaplan-Meier method, a univariate analysis using the log-rank test and a multivariate analysis by using the Cox regression analysis were performed, with age, sex, severity of injury, American Spinal Injury Association impairment scale (AIS), fracture type, comorbidity of ossification of the posterior longitudinal ligament/DISH, and modified Charlson Comorbidity Index (CCI) as independent variables, and survival period as the outcome variable. Results: Comorbid DISH was observed in 8 patients (6 %). The overall postoperative 5-year cumulative survival rate was 76.4 %. Of the 23 patients (18 %) who died, 9 patients died within 6 months after the surgery, and pneumonia was the most common cause of death. The log-rank test revealed significant differences in advanced age (age C 70 years), CCI (C5 points), the presence of DISH, and AIS (A or B). The Cox regression analysis showed that CCI (C5 points; p = 0.0445; hazards ratio [HR], 2.965), the presence of DISH (p = 0.0056; HR, 5.670), and AIS (A or B; p = 0.0002; HR, 4.818) significantly affected the survival period. Conclusion: In the present study, the poor prognosis associated with DISH in the patients with cervical spinal cord injury supports the findings of recent case series. The presence of DISH in elderly patients with cervical spinal cord injury is a high-risk factor of a poor life prognosis, as well as motor complete injury and severe comorbidities. As comorbid DISH will increase in the coming decades, surgeons need careful consideration of life prognosis before surgery. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

Hiroaki Sameda, Jun Shimbo, Sumio Ikenoue, Kan Takase, Yukio Someya, Hiroto Kanamoto, Masaya Mimura Department of Orthopaedic Surgery, Funabashi Municipal Medical Center, Funabashi City, Japan Introduction: Fractures of the spine in diffuse idiopathic skeletal hyperostosis (DISH) have rarely been reported. Pre-existing DISH causes adjacent intervertebral stability and mechanical power to concentrate on fractured parts, which cause the fractured parts to become pseudarthrosis easily and difficult to control with an external brace. The DISH patients render high neurological risk after spinal fracture. Moreover, it is difficult to identify fracture just on plain radiographs. Purpose: The purpose of this study was to examine the surgical outcome of cervical spine fracture of DISH at our institution. Materials and methods: We have collected seven cases of pre-existing DISH and vertebral fractures through an area of ankylosed spine at the cervical level. There were three men and four women with age ranged from 72 to 86 years old (average, 78 yrs). All of the seven patients were treated operatively. Plain radiographs, magnetic resonance images (MRI) and computerized tomographic scans (CT) were made for all seven patients. Results: One fracture was caused by high-energy trauma (felling down the stairs) and six fractures by low-energy. Cervical fractures of all anterior, middle, and posterior column were identified on MRI and CT in all patients, but not on plain radiographs clearly in all patients. MRI and CT of the injured vertebral segments demonstrated destabilizing injuries of all three columns of the spines in all patients. The spinal fracture caused complete quadriplegia in three patients, incomplete quadriplegia in three patients, and no neurologic deficit in one patient. Four cases were treated with posterior fixation and three cases were treated with anterior and posterior fixation. At the time of operation, the patients were positioned with great care in order to maintain the pre-injury cervical alignment. All patients had healing of the fracture with anatomical alignment of the spine and without severe postoperative complications. In all incomplete quadriplegia patients, paralyses were improved. Conclusions: Treatment of cervical spine fracture in DISH requires rigid stabilization of the vertebral bodies as soon as possible. Treatment of this rare injury should be early stabilization of the spine to avoid complications of nonunion, deformity, neurologic injury. Early diagnosis and treatment are important and careful check up of MRI and CT is mandatory for DISH patients with persistent or severe cervical pain.

123

P15 POOR PROGNOSIS ASSOCIATED WITH DIFFUSE IDIOPATHIC SKELETAL HYPEROSTOSIS IN CERVICAL SPINAL CORD INJURY: ANALYSIS OF PROGNOSTIC FACTOR IN PATIENTS WITH CERVICAL SPINAL CORD INJURY Kimihiko Sawakami, Takui Ito, Ikuko Takahashi, Seiichi Ishikawa Department of Orthopaedic Surgery, Niigata City General Hospital, Japan

P16 SIGNIFICANT REDUCTION IN THE INCIDENCE OF C5 PALSY AFTER CERVICAL LAMINOPLASTY USING CHILLED IRRIGATION WATER Shota Takenaka, Noboru Hosono, Yoshihiro Mukai, Kosuke Tateishi, Takeshi Fuji Japan Community Healthcare Organization Osaka Hospital, Osaka, Japan

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Purpose: Basic research suggests that upper limb palsy (ULP) results from thermal damage to the nerve roots by friction-generated heat during bone drilling, but no reports have shown that chilled irrigation saline use during laminoplasty significantly decreases clinical ULP (a reduction of one grade or more on manual muscle testing [MMT]) incidence. This study aimed to determine whether the use of chilled irrigation saline decreases the incidence of clinical upper limb palsy. Method: Irrigation saline for drilling was used at room temperature (RT, 25.6 °C) in open-door laminoplasty in 400 patients (RT group) and was chilled to a mean temperature of 12.1 °C during operations for 400 patients (low temperature [LT] group). We assessed the strength of the deltoid, biceps, and triceps brachii muscles by MMT. In addition, we evaluated the ULP onset, ULP affected side, difference in MMT for the most deteriorated muscle (DMMT), and ULP recovery period. ULP occurring within 2 days postoperatively was categorized as early-onset palsy, and ULP occurring subsequently was defined as late-onset palsy. Results: The incidence of ULP was significantly smaller for the LT group than for the RT group (4.0 % vs. 9.5 %, p = 0.003). The incidence of early-onset ULP cases in the LT group was significantly less than that in the RT group (1.0 % vs. 5.5 %, p \ 0.001); the incidence of late-onset ULP cases did not significantly differ between the groups (3.0 % vs. 4.0 %, p = 0.565). Multivariate analysis indicated that RT irrigation saline use, concomitant foraminotomy, and opened side were significant predictors for ULP. The ULPs which did not recover within two years of surgery were all early-onset palsies. Discussion: Using chilled irrigation saline during bone drilling significantly decreased the ULP incidence, particularly for the earlyonset type, which was likely to be severe. Chilled irrigation saline can thus be recommended as a simple method for preventing ULP. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: consultant; Company=Daiichi-Sankyo, royalties; Company=Century Medical, Showa Ika-Kogyo.

S749 Results: Without decompression, FAD (preop./postop.) was 3.6 ± 1.4 mm/3.7 ± 1.4 mm: 3.7 ± 1.4 mm/3.8 ± 1.3 mm without IFS (120 roots) and 1.3 ± 0.4 mm/1.0 ± 0.4 mm with IFS (5 roots). Both preoperative and postoperative FADs were significantly smaller with than without IFS (p \ 0.001). With foraminotomy, FAS was 2.0 ± 0.9 mm/3.3 ± 1.5 mm (p \ 0.001): 2.1 ± 0.9 mm/3.5 ± 1.5 mm without IFS (48 roots) (p \ 0.001) and 1.7 ± 0.3 mm/1.6 ± 0.2 mm with IFS (4 roots). Preoperative FADs were not significantly different with or without IFS; however, postoperative FAD was significantly smaller with than without IFS (p \ 0.001). With facetectomy, FAD was 2.0 ± 1.0 mm/6.4 ± 2.5 mm (p \ 0.001): 2.0 ± 1.0 mm/6.6 ± 2.3 mm without IFS (26 roots) (p \ 0.001) and 1.1 mm/1.8 mm with IFS (1 root). Preoperative FAD with facetectomy was not significantly different from that with foraminotomy; however, postoperative FAD with facetectomy was significantly larger than that with foraminotomy (p \ 0.001). Discussion: Foraminotomy primarily decompresses only the inlet of the foramen. Therefore, it may cause IFS unexpectedly by creating rigid fixation, increased lordosis, or posterior translation in the degenerative cervical spine. However, facetectomy decompresses the entire foramen, allowing full visualization of the nerve roots. Facetectomy can widen the FAD more than foraminotomy. Therefore, facetectomy is preferred for securing sufficient FAD. Disclosures: author 1: consultant; Company=Alphatec Spine; author 2: none; author 3: none; author 4: none.

P18 STAND-ALONE ANCHORED SPACER VERSUS ANTERIOR PLATE FOR MULTILEVEL ANTERIOR CERVICAL DISKECTOMY AND FUSION Lili Yang, Wen Yuan

P17 FACETECTOMY CAN WIDEN FORAMINAL AXIAL DIAMETER MORE THAN FORAMINOTOMY TO PREVENT IATROGENIC FORAMINAL STENOSIS AFTER CERVICAL PEDICLE SCREW FIXATION Akiyoshi Yamazaki, Tomohiro Izumi, Tatsuki Mizouchi, Hideki Tashi Niigata Central Hospital, Niigata, Japan Introduction: Foraminotomy has been recommended for preventing iatrogenic foraminal stenosis (IFS) after cervical pedicle screw (PS) fixation. However, postoperative foraminal axial diameter (FAD) should be [ 1.9 mm, which can be obtained by performing facetectomy rather than foraminotomy to prevent IFS. This study aimed to investigate the FAD preoperatively and postoperatively and compare foraminotomy and facetectomy. Methods: PS fixation was performed in 43 patients (mean age 62 years, 102 disc levels) from C4/5 to C7/T1. Out of 204 nerve roots (C5, C6, C7, and C8 in 72, 58, 48, and 26 roots, respectively) in the fusion area, 125 roots underwent no foraminal decompression, 52 underwent foraminotomy, and 27 underwent facetectomy. Overall incidence of IFS was 4.9 % (10/204 roots; C5 in 7, C6 in 1, and C8 in 2): it was 4 % without decompression, 7.7 % with foraminotomy, and 3.7 % with facetectomy. The preoperative and postoperative FADs on computed tomography scans were compared between patients who underwent foraminotomy and facetectomy.

Department of Spine Surgery, Shanghai Changzheng Hospital, Shanghai, China The purpose of this study was to compare the clinical outcomes and radiological changes of 3- and 4-level anterior cervical diskectomy and fusion with stand-alone anchored spacers and with traditional anterior plates. A total of 51 consecutive patients with cervical spondylotic myelopathy who underwent 3- or 4-level anterior cervical diskectomy and fusion were divided into 2 groups: group A (n = 23) received anchored spacers and group B (n = 28) received an anterior plate. Mean follow-up was 14.6 months. Solid fusion was achieved in all patients at final follow-up. No significant difference existed between multilevel anterior cervical diskectomy and fusion with stand-alone anchored implants and with an anterior cervical plate in achieving clinical symptomatic improvement, fusion rate, and lordotic curvature improvement. The dysphagia rate of group A at 2-month follow-up was significantly lower than that of group B. No statistically significant difference existed between the 2 groups at the other time points. Swallowing Quality of Life of group A at 48 hours and 2 months postoperatively were significantly higher than those of group B. The thickness of the prevertebral soft tissue at 48 hours and 2 months postoperatively were significantly lower in group A than in group B. Compared with using an anterior plate, anterior cervical diskectomy and fusion with a stand-alone anchored spacer achieved a similar clinical outcome with less irritation to the prevertebral soft tissue and a lower dysphagia rate in the first 2 months. Disclosures: author 1: none.

123

S750 P19 THE INFLUENCE OF OCCIPITAL ORIENTATION ON CERVICAL SAGITTAL ALIGNMENT: A PROSPECTIVE RADIOGRAPHIC STUDY ON 354 NORMAL SUBJECTS Weiguo Zhu, Zezhang Zhu, Zhen Liu, Bangping Qian, Yong Qiu From the Department of Spine Surgery, Drum Tower Hospital of Nanjing University Medical School, Nanjing, China Background: Previous studies had demonstrated that cervical sagittal alignment was correlated with the thoracic alignment. However, the relationship of occiput and cervical has not been revealed. Objectives: To explore the features of occipitocervical alignment and evaluate the correlations between occipital orientation and cervical sagittal alignment in normal subjects. Materials and methods: From January 2007 to January 2013, a total of 354 individuals (139 males and 215 females) between 10 and 60 years of age with lateral radiography were recruited. Occipital incidence (OI), occipital slope (OS), occipital tilt (OT), angles for occipital to 2nd cervical vertebrae (C0-C2), sagittal alignment of 2nd to 7th cervical vertebrae (C2-C7), angles for occipital to 7th cervical vertebrae (C0-C7) and C2 standard sagittal balance parameters (C2SVA) were measured for every individual. Comparisons of these parameters among three age groups (group A: \20, group B: 20*40, group C:[40) were performed with ANOVA and differences of these parameters between females and males were analyzed with Student t tests. In addition, the relationships among these parameters were assessed using the Pearson correlation analysis. Results: The mean values of occipital parameters of the cohort were 34.61°± 3.07 for OI, 31.42°± 7.37° for OS, and 6.75°± 3.29° for OT, respectively. The mean OI was constant through three age groups. The OS increased and the OT decreased in group B and C compared with group A, however, no significant differences of OS and OT were noted between group B and C. In terms of gender, no significant differences of the three parameters were noted between males and females. In addition, OI was strongly correlated with OS (r = 0.279), C0-C2 angle (r = 0.573), C2-C7 angle (r = 0.240) and C0-C7 angle (r = 0.589). OS was also strongly correlated with OT (r = 0.885), C0-C2 angle (r = 0.327), C2-C7 angle (r = 0.300) and C0-C7 angle (r = 0.478). Meanwhile, strong correlations were found between OT and C0-C2 angle (r = 0.172), C2-C7 angle (r = 0.230) and C0-C7 angle (r = 0.313). Conclusion: Occipital orientation was an important factor that influenced the cervical sagittal alignment. The upper cervical sagittal alignment was mainly compensated by occipital orientation, while lower cervical sagittal alignment was mainly compensated by occipital orientation and thoracic alignment. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none. BASIC SCIENCE, BIOLOGY

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Directly reprogrammed autologous neural stem cells and regenerative matrix for complete spinal cord injury O.V. Durov1, J.E. Ahlfors2, A.V. Averyanov1, M.A. Tikhonovsky1, M.A. Konoplyannikov1, N.N. Sovetnikov1 1 Federal Research Clinical Center under FMBA of Russia, Moscow, Russia 2 New World Laboratories Inc., Laval, Canada Spinal cord injury (SCI) remains a challenging issue of medicine. The spinal cord virtually does not regenerate after injury. However, regenerative medicine, namely stem cells transplantation, offers a means for the restoration of spinal cord function. Objective: To evaluate the safety and potential efficacy of directly reprogrammed autologous neural stem cell (drNSC) transplantation for subacute and chronic spinal cord injury. Methods: We have developed a technique for spinal cord restoration after injury. It includes implantation of drNSC and Regeneration Matrix (RMx) into a spinal cord defect, followed by repeated intrathecal injections of drNSC. At the moment 7 patients 1F/6M, aged 25-54 with complete thoracic spinal cord injury, ASIA score below 200, types A and B have been enrolled. The patients underwent surgery including laminectomy, durotomy, filling of the spinal cord defect with RMx, and direct drNSC injections into the white and grey matter of the spinal cord adjacent to the defect. In specified time intervals, the patients have been intrathecally injected with drNSC. The study was registered at ClinicalTrials.gov (NCT02326662). Results: We have 3 months follow up results in 4 operated patients. There were no complications or adverse events associated with drNSC and RMx transplantation. All patients demonstrated improvements in their neurologic state. One patient was excluded from the study because of a severe hydrocephalus decompensation during 3 weeks after the surgery (unrelated to the trial). Conclusion: The original technique of drNSC and RMx transplantation have demonstrated the safety in the first patients and presents potential benefits for the treatment of subacute and chronic spinal cord injury patients. Author(s) disclosures: Jan-Eric Ahlfors is the inventor of the drNSC and RMx used in the study. Disclosures: author 1: none; author 2: stock/shareholder; Company=New World Laboratories Inc., employee; Company=New World Laboratories Inc.; author 3: none; author 4: none; author 5: none; author 6: none.

P21 G-PROTEIN COUPLED RECEPTOR 128 IS REGULATED BY ESTRADIOL Florina Moldovan, Amani Hassan, Eduardo Bagu, Julie Couillard, Shunmoogum A. Patten CHU Sainte-Justine, Montreal, Canada; Faculty of DentistryStomatology, Montreal, Canada

P20 DIRECTLY REPROGRAMMED AUTOLOGOUS NEURAL STEM CELLS AND REGENERATIVE MATRIX FOR COMPLETE SPINAL CORD INJURY Oleg Durov, Jan-Eric Ahlfors, Alexander Averyanov, Mihail Tikhonovsky, Mihail Konoplyannikov, Nikolay Sovetnikov Federal Research Clinical Center under FMBA of Russia, Moscow, Russia; 2 New World Laboratories Inc., Laval, Canada

123

GPR128 is an orphan member of the adhesion subfamily of G-protein coupled receptors with a long serine/threonine-rich N-terminus, The GPR128 and another member of the same subfamily (GPR126) were recently associated with the Adolescent Idiopathic Scoliosis (AIS). Variants in GPR128 and GPR126 were reported in French and Japanese AIS patients with suggested role in the growth and ossification of the developing spine. Here we investigated the regulation of GPR128 by estradiol. Methods: In-silico analysis for potential estrogen response element (ERE) in the GPR128 promoter was done using MatInspector and

Eur Spine J (2015) 24 (Suppl 6):S743–S800

S751

ECR-Browser and then several promoter fragments were cloned in PGL3 vector upstream of the luciferase gene. Cells were then transiently transfected with the GPR128 promoter constructs. In the presence or absence of 17-beta-estradiol, promoter study (luciferase assay), mRNA (qPCR), and protein (western blot) expression studies were performed. Results: In Huh7 cell lines that were transfected with the GPR128 promoter constructs, treatment with estradiol for 24 hours following over-expression of ER-a led to a 2.5 fold increase in the promoter activity, confirming the regulation by estradiol. We located the ERE in the GPR128 promoter to the 61bp sequence in the 390bp fragment. Likewise the mRNA and GPR128 was highly up-regulated in the presence of estradiol in both MCF7 and osteoblasts. Our study demonstrated that estradiol is involved in the expression of GPR128. Pertinence: These results could help to understand the possible contribution of GPR128, and molecular mechanisms involved in AIS progression during the pubertal growth spurt. Disclosures: author 1: grants/research support; Company = Yves Cotrel Foundation; author 2: none; author 3: none; author 4: none; author 5: none.

transverse processes. Analysis of radiographic imaging highlights the superiority of fusion quality in the DBM treated rats compared to the ICBG treated rats. Radiographic grading according to the Lenke scale revealed 75 % of rats in both DBM groups were graded as having large and trabeculated bilateral fusion masses compared to only 25 % of the ICBG treated rats. Overall, specimens treated with either DBM demonstrated better bone formation and fusion consolidation than those treated with ICBG. Comprehensively, histological analysis revealed the main findings of the study and demonstrated that DBM A exhibited greater biologic activity at 8 weeks than DBM B. Disclosures: author 1: none; author 2: none; author 3: none; author 4: grants/research support; Company=Integra; author 5: grants/research support; Company=Integra.

P22 EFFICACY COMPARISON OF TWO COMMERCIALLY AVAILABLE DEMINERALIZED BONE MATRICES AGAINST AUTOLOGOUS BONE IN A RAT POSTEROLATERAL SPINE FUSION MODEL

Tito Bassani, Sara Zacchetti, Claudia Ottardi, Francesco Costa, Marco Brayda-Bruno, Hans-Joachim Wilke, Fabio Galbusera

Antonio Brecevich, Paul Kiely, Byung Jo Victor Yoon, Frank Cammisa, Celeste Abjornson

Introduction: The radiographic images examination of coronal and sagittal planes allows quantifying anatomical properties, i.e. spine curves and vertebral rotations, but does not provide information about biomechanical aspects, i.e. load forces acting on intervertebral discs. The opportunity of non-invasively evaluating spine biomechanics would provide essential information on spinal loads acting maintaining specific postures and would support clinicians in planning the best strategy for spinal instrumentation, e.g. in case of scoliosis correction. The present study proposes a semi-automatic software approach to reconstruct 3D patient-specific spine mechanical model from radiographic digitized images in coronal and sagittal planes simultaneously acquired with the EOS Imaging System (EOS Imaging, Paris, France). Methods: Five adolescent subjects with mild scoliosis (Cobb angles \24°) were evaluated in standing position. For each vertebra a set of landmarks were manually identified on the radiographic images via an in-house dedicated software. Coordinates of landmarks were processed allowing to calculate geometrical parameters of vertebral configuration in the 3D space: (1) placement; (2) dimensions; (3) rotation around coronal, sagittal and axial axes. Spherical joints were set between vertebral pairs simulating intervertebral discs. Thoracic joints were kept unable to move in order to model the high stiffness of the trunk, whereas lumbar joints were not constrained. Body weight distribution, muscles forces and muscle insertion points on vertebrae were placed according to physiologically-anatomically appropriate values. Inverse static analysis, able to calculate joints reactions in maintaining assigned spine configuration, was performed with AnyBody software (AnyBody Technology, Denmark). Joint reaction forces were computed to quantify load forces acting on lumbar intervertebral discs, i.e. axial compression, frontal and lateral shear forces. Results and conclusions: Maximum axial compression ranged from 766N to 1025N. Maximum anterior shear force ranged from 299N to 475N. Maximum lateral shear force was found related to scoliotic

Hospital for Special Surgery; New York, NY, USA Currently, the gold standard for stabilizing fusion masses in spinal fusion procedures involves utilizing the osteogenic, osteoinductive, and osteoconductive properties of autologous iliac crest cortico-cancellous bone (ICBG). However, considerable morbidity is associated with harvesting the autologous graft. Donor site complications including infection and pain have been reported at rates as high as 50 %. The objective of the study is to compare the quality and rate of fusion using a new generation of Demineralized Bone Matrix (DBM A) against an established DBM (DBM B) and the ‘gold standard’ bone graft, ICBG, in the athymic rat posterior intertransverse process model. The objective will be achieved by assessing the fusion quality and rate in radiographic appearance, nondestructive mechanical testing and histological analysis of the fusion mass produced by each treatment group. A comparison of the quality and rate of fusion between DBM A and DBM B, with the ICBG as the control, was performed using the established posterolateral intertransverse process athymic rat model. Materials were evaluated for fusion by several criteria: manual palpation, radiographic imaging, micro-CT and histological analysis. 51 athymic rats received a bilateral intertransverse process fusion, using either bone from the iliac crest, DBM A or DBM B. 12 rats (4 from each group) were sacrificed at 3 weeks post-operatively, while the remaining 36 were sacrificed at 8 weeks postoperatively. Manual palpation results indicated success of the test articles in forming a solid union; and nondestructive mechanical testing confirmed that fusions had occurred. These findings were consistent with high-resolution radiography and micro-CT, both of which show significant bone formation and bridging across the region between the

BASIC SCIENCE, BIOMECHANICS

P23 BIOMECHANICAL PATIENT-SPECIFIC 3D SPINE MODELING FROM BIPLANAR RADIOGRAPHIC IMAGES

IRCCS Galeazzi, IRCCS Humanitas, Politecnico di Milano, Italy and Ulm University, Ulm, Germany

123

S752 curve degree, ranging from 132N to 145N. Although preliminary, spinal loads predicted for the five subjects were found comparable with values known from literature. The proposed approach is therefore potentially appropriate in describing patient-specific spine mechanics and providing new input to assist clinicians. Disclosures: author 1: grants/research support; Company=Italian Ministry of Health (GR-2011-02351464); author 2: employee; Company=IRCCS Istituto Ortopedico Galeazzi; author 3: grants/research support; Company=Italian Ministry of Health (GR-2011-02351464); author 4: none; author 5: grants/research support; Company = Medtronic - K2M, consultant; Company=Depuy Spine - Implanet Spineguard; author 6: none; author 7: grants/research support; Company=Italian Ministry of Health (GR-2011-02351464), employee; Company=IRCCS Istituto Ortopedico Galeazzi.

P24 EVALUATING SAGITTAL SPINAL POSTURE DURING FUNCTIONAL TASKS: CAN KINEMATICS DIFFERENTIATE BETWEEN NON-SPECIFIC CHRONIC LOW BACK PAIN (NSCLBP) SUBGROUPS AND HEALTHY CONTROLS? Rebecca Hemming, Liba Sheeran, Robert van Deursen, Valerie Sparkes School of Healthcare Sciences, Cardiff University, Cardiff, UK Background: Identifying homogeneous NSCLBP subgroups is crucial for establishing targeted interventions. The multidimensional classification approach (MCA) (O’Sullivan, 2005) proposes specific motor control impairment (MCI) subgroups exist, with differences in sitting lumbar angle observed between subgroups (Flexion Pattern (FP) and Active Extension Pattern (AEP)) and healthy subjects (Dankaerts et al. 2006). Classification-based cognitive functional therapy (CB-CFT) approaches are proposed to be effective for these subgroups (Fersum et al, 2013) however, subgroup spinal kinematics during functional tasks is yet to be established. Methods: An observational, cross-sectional study investigated spinal kinematics of 50 NSCLBP subjects (28 FP, 23 AEP) and 28 healthy controls using 3D motion analysis (Vicon) during sit-to-stand (STS), reach up, step up, box lift and pick up a pen tasks. Mean sagittal angle for total lumbar (Lx), total thoracic (Tx), upper thoracic (UTx), lower thoracic (LTx), upper lumbar (ULx) and lower lumbar (LLx) regions between groups were compared. Significance level was p \ 0.05. Results: No significant differences in Tx or Lx angles were observed with the exception of STS. Significant differences were consistently observed in the LTx and ULx regions between the AEP and FP groups during all tasks (with the exception of LTx during reach up). Significant differences were observed between the FP and control group in the LTx during the STS and pick up a pen tasks, and in the ULx during the step-up and pick up a pen tasks. All significant between group regional differences demonstrated the FP group to operate in comparatively greater flexion. No significant differences were observed between the AEP and control groups, or in the UTx or LLx regions during any task. Conclusion: Subdivided regional spinal angles appear to be key in identifying between MCI sub-group differences, with the spinal region between T6 to L3 able to discriminate between FP and AEP, and FP and healthy subjects during functional tasks. These findings further validate the proposed MCA and can inform the development of specific CB-CFT interventions for NSCLBP subgroups. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 P25 BIOMECHANICAL COMPARISON OF TRADITIONAL ILIAC SCREW (TIS) FIXATION VERSUS DISTAL ILIAC SCREW (DIS) FIXATION: A CADAVERIC STUDY Meric Enercan, Mutlu Cobanoglu, Sinan Kahraman, Bahadir Gokcen, Sinan Yilar, Tunay Sanli, Erden Erturer, Cagatay Ozturk, Ahmet Alanay, Azmi Hamzaoglu Istanbul Spine Center, Istanbul, Turkey Summary: This cadaveric study demonstrated that distal iliac screw (DIS) fixation technique which does not require any cortical bone resection for entry point biomechanically provided higher insertional torques, stiffness, axial-pull out and toggle forces than traditional iliac fixation (TIS). More distal starting point (posterior inferior iliac spine) enables greater and longer screw purchase with low profile fixation than traditional iliac screw fixation. Design: Comparative biomechanical study. Introduction: During insertion of the TIS, resection of the posterior superior iliac spine in order to avoid implant prominence will result in loss of the cortical bone, which will lead to a decrease in the insertional torques of the iliac screw and its primary stability. In order to overcome problems related with TIS fixation, we introduced a new freehand technique ‘‘DIS’’ fixation with a starting point located more distally at the posterior inferior iliac spine, which does not require any decortication or osteotomy for entry point and with a trajectory courses very close to the rigid subcortical bone over the sciatic notch. The aim of this study is to biomechanically evaluate and compare TIS fixation vs DIS fixation technique with a cadaveric study. Methods: Eight fresh human (4F, 4M) lumbopelvic spines were tested and each specimen was assigned a TIS fixation on one side and DIS fixation on the contralateral side. The insertional torque forces were recorded with a digital torque wrench through placement and the axial pull-out and toggle tests were conducted using a MTS test system. All specimens were radiographed, and 3D images were taken using O-Arm system prior to and after testing. Results: Mean peak insertion torque was 2.48 ± 1.84 Nm for traditional and 3.98 ± 2.40 Nm for new trajectory (p \ 0.008). DIS fixation achieved higher maximum axial pull-out force and higher stiffness than TIS fixation. At the 1st and 100th load cycle with toggle displacement of 5 mm, DIS achieved higher toggle forces than the TIS fixation. All screws placed in the new trajectory were longer in length compared to the screws placed in the traditional trajectory without any cortical breech. Conclusion: DIS fixation technique provided higher insertional torques, stiffness, axial pull-out and toggle forces and longer screw length than TIS fixation. The results of this study encourages the clinical application of DIS fixation technique in primary or revision adult deformity surgery. Disclosures: author 1: none; author 2: no indication; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: none; author 9: grants/research support; Company=Depuy Synthes, consultant; Company=Stryker; author 10: consultant; Company=Medtronic.

P26 FACET JOINT CAPSULAR LAXITY IN DEGENERATIVE LUMBAR SPONDYLOLISTHESIS ASSOCIATED WITH THE INCREASED EXPRESSION OF FRACTALKINE (CX3CL1)/ CX3CR1 CHEMOKINE In-Soo Oh, Kee-Yong Ha

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Incheon St. Mary’s Hospital, The Catholic University of Korea, Incheon, Korea CX3CL1 and its receptor (CX3CR1) are part of a chemokine system involved in leukocyte recruitment and adhesion in chronic inflammatory disease. From previous study the CX3CL1 and CX3CR1 activity has been investigated in ligament flavum, synovial membrane, and intervertebral discs, but not with the facet joint capsule related issue. The purpose of this study is to investigate the role of fractalkine (CX3CL1)/CX3CR1 chemokine on facet joint capsular laxity in degenerative lumbar spondylolisthesis and correlation between expression of CX3CL1/CX3CR1 chemokine and degree of slippage by lumbar spondylolisthesis. The mRNA concentrations of CX3CL1/CX3CR1 chemokine were analyzed in the surgically obtained facet joint capsule specimens from grade 1 spondylolisthesis (n = 12), grade 2 spondylolisthesis (n = 12) and more than grade 3 spondylolisthesis (n = 11) by real-time PCR. Grade 1 to 3 is being decided upon degree of slippage which is, less than 5 mm, between 5 to 10 mm, and more than 10 mm. The localization of CX3CL1/ CX3CR1 chemokine within the facet joint capsule was determined using immunohistochemical study. Plasma level of soluble fractalkine (sFKN) was measured by enzyme-linked immunosorbent assay (ELISA), respectively. The cells that shows higher CX3CL1/ CX3CR1 chemokine expression ratio in the facet joint capsule are observed in tissues acquired from patients with higher degree of slippage by lumbar spondylolysthesis (P = 0.000, 0.000). In ELISA, the plasma levels of sFKN was significantly higher in the group with more severe degree of slippage (P = 0.002). There was greater CX3CL1/CX3CR1 expression in higher grade spondylolisthesis as quantified by RT-PCR (P = 0.000, 0.003). Degree of slippage in DLS patients was significantly correlated with serum CX3CL1 level (R2 = 0.451, P = 0.000) and with mRNA expression of CX3CL1/ CX3CR1 (R2 = 0.360, P = 0.000) (R2 = 0.205, P = 0.006). This study identified for the first time that increases in CX3CL1 and CX3CR1-expressing cells are significantly related to facet joint capsular laxity, which may provide new conceptual and therapeutic approaches for treating spinal spondylolisthesis. Disclosures: author 1: none; author 2: none.

P27 TO WHAT EXTENT LUMBOPELVIC POSTURE INFLUENCES THE MYOELECTRIC SILENCE OF THE ERECTOR SPINAE DURING TRUNK FLEXION ´ ngeles Sarti Carlos Barrios, Rut Expositor-Rodrı´guez, M. A Universidad Cato´lica de Valencia, Spain; Universidad de Valencia, Spain Background: The myoelectric relaxation (silence) of the erector spine occurs nearby spine full flexion. Accordingly, the risk of injury on the spine tissues increases when the trunk goes into extreme ranges of flexion. Objective: To determine whether the occurrence of the myoelectric silence of the erector spinae during trunk flexion movement is related to the lumbopelvic posture in standing upright. Methods: 42 healthy subjects participated in this study [average 23.8, SD 3.9 years]. Lumbar curvature (LC) and sacral inclination (SI), in relaxed upright stance, and lumbar motion during forward bending were calculated in the sagittal plane with an electro-goniometer (degrees). This consists of two receivers located at L1 and S1 spinous processes. Lumbar curvature is the angle formed between L1 and S1 receivers. Sacral inclination is the angle given by S1-receiver in the

S753 sagittal plane (pelvic angulation in standing). LC and SI in standing reflect the lumbopelvic posture. During sagittal flexion, the lumbar spine angular displacement was recorded continuously and simultaneously with the surface EMG (lV) from the right erector spinae at L3 level. The start of the myoelectric relaxation of the erector spinae was defined as the spine flexion degrees whereby the electrical activity was less than that in erect standing, which were expressed as the range of spine flexion from upright position. Results: the average and (SD) values for LC were [-40.2°, (16.9°)] and for SI [20.4°, (15.1°)]. The myoelectric silence started at [51.3°, (11.4°); 87.3 % (10.2 %)]. Lumbopelvic posture in standing and the range of spine flexion at the start of the myoelectric silence of the erector spinae were significantly (p \ .000) correlated (r = .70, R2 = .50). Conclusions: The start of the myoelectric silence of the erector spinae in terms of the spine flexion could be explained up to 50 % by the lumbopelvic posture in standing. To predict the appearance of the myoelectric silence of the erector spinae from lumbopelvic posture may be useful as a preventative measure in clinical and surgical settings. Disclosures: author 1: none; author 2: none; author 3: grants/research support; Company=FIS2001-0070-01. TIC 2001-2786-C02-02.

THORACO-LUMBAR SPINE: DEGENERATIVE, DEFORMITY

P28 DEGENERATIVE LUMBAR SPINAL STENOSIS AND PARASPINAL MUSCLES CONDITION Janan Abbas1,2, Kamal Hamoud2,3, Natan Peled4, Israel Hershkovitz1 1

Department of Anatomy and Anthropology, Sackler Faculty of Medicine Tel-Aviv University, Israel; 2Department of Physical Therapy, Zefat Academic College, Zefat, Israel; 3Spine Unit, Baruch Padeh Poriya Medical Center, Tiberias, Israel; 4Department of Radiology, Carmel Medical Center, Haifa, Israel Background: As life expectancy increases, degenerative lumbar spinal stenosis (DLSS) becomes a common health problem among the elderly, leading to low back pain, radiculopathy, and intermittent claudication. The paraspinal muscles play an important role in the stability and functional movements of the lumbar vertebral column. Although spinal instability has been shown to play a role in DLSS, the part of paraspinal muscles has been ignored. Purpose: To shed light on the association of paraspinal muscles condition e.g. muscles cross sectional area (CSA) size and density and DLSS. Materials and methods: Two groups were studied: the first included 165 individuals with DLSS (age range: 40-88 years; sex ratio: 80 males, 85 females) and the second 180 individuals (age range was 40-90 years, sex ratio: 90 males, 90 females) without DLSS related symptoms. Measurements were performed at the middle part of L3 vertebral body, using CT images (Philips Brilliance 64). Muscles density was measured in Hounsfield units (HU) using a 50 mm2 circle of the muscle mass at three different locations and the mean density was then calculated. The CSA size was also measured using the quantitative CT angiography (Q-CTA) method. Results: Both males and females in the stenosis group had higher muscle density compared to their counterparts in the control group (P \ 0.05). The CSAs for the erector spinae and psoas muscles were significantly greater in the stenosis males compared to their

123

S754 counterparts in the control group (P \ 0.05). Additionally, densities of multifidus (both sexes) and erector spinae muscles (males) were found to be significant predictors for DLSS development. Conclusions: Paraspinal muscles densities and CSA sizes in the DLSS were remarkably deviate from the norm. Density of the multifidus muscles may be an indicator for the onset of symptomatic DLSS. Clinical relevance: As the increased paraspinal density in DLSS individuals was a specific radiological finding for this disorder (not related to other degenerative lumbar spine disease), we suggest this measurement could be used as a radiological marker for detecting clinical syndrome of DLSS. Keywords: Degenerative lumbar spinal stenosis, paraspinal muscles condition e.g. density and cross-sectional area (CSA) size, CT images. Disclosures: author 1: none; author 2: none; author 3: no indication; author 4: grants/research support; Company=Tassia and Dr. Joseph Meychan Chair for the History and Philosophy of Medicine; author 5; author 6; author 7; author 8; author 9; author 10.

P29 HOW RELIABLE IS THE SURGEON’S ABILITY TO DIFFERENTIATE BETWEEN IDIOPATHIC AND DEGENERATIVE DEFORMITY IN ADULTS; WHAT PARAMETERS HELP THEM DECIDE? Emre Acaroglu, Umit Ozgur Guler, Selim Ayhan, Sule Yakici, Montse Domingo-Sabat, Ferran Pellise, Francisco Javier Sanchez Perez-Grueso, Ahmet Alanay, Ibrahim Obeid, Frank Kleinstu¨ck, European Spine Study Group (ESSG) 1,2,3,4

Ankara Spine Center, Ankara, Turkey; 5Raquis, Fundacio´ Institut de Recerca Vall d’Hebron, Barcelona, Spain; 6Spine Unit, Hospital Vall d’Hebron, Barcelona, Spain; 7Spine Unit, Hospital Universitario La Paz, Madrid, Spain; 8Acibadem University School of Medicine, Comprehensive Spine Center, Acibadem Maslak Hospital, Istanbul, Turkey; 9Spine Unit, Bordeaux University Hospital, Bordeaux, France; 10Spine Unit, Schulthess Klinik, Zu¨rich, Switzerland; 11Fundacio´ Institut de Recerca Vall d’Hebron

Background: Adult spinal deformity (ASD) may be classified as idiopathic (I) or degenerative (D) (or other) based on classifier’s perception, the reliability of and factors inherent to which remain unknown. Purpose: To evaluate the inter and intraobserver reliability of surgeons’ perception in differentiating I from D ASD and to identify the determinants of it. Patients and methods: From a multicentric prospective database, 179 pts were identified with the diagnosis of I (n = 103) or D (n = 76); no previous surgery; and a lumbar coronal curve [ 20. Standing AP and lateral X-Rays were sent to five experienced spine surgeons to be identified as D or I (or other); followed by a second round after reshuffling. Weighted Kappa statistics was used, after which the patients were stratified by number of agreements as perfect (10/10) and very good (C8/10); these were further compared for additional radiological parameters. Results: Four observers completed both rounds while the 5th did only the first (a total of 10 observations/pt including the database record). Agreement levels were moderate to good for intra (0.566 - 0.638), but fair to moderate for interobserver (0.144 - 0.611) comparisons. There were 42 perfect and 80 very good agreements for I patients but only 6 perfect and 17 very good agreements for D. Upon comparison of these, it was seen that they were different for some coronal

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 parameters such as lumbar Cobb angle (larger in I, p \ 0.001), CSVL modifier (C more common in I, p = 0.007) and presence of rotatory subluxation (less common in D, p = 0.017) and very different for most sagittal parameters (Lumbar Lordosis, Sagittal Vertical Axis, T2-Sagittal tilt, Pelvic Tilt, Sacral Slope and Global Tilt; increased sagittal imbalance in D, all p B 0.001). Conclusion: Surgeons in this study demonstrated reasonable intraobserver agreement but only fair agreement amongst them. These findings suggest that especially in patients with significant coronal curves, determination of curve etiology with only radiological data may not be accurate. In patients with good agreement, the most consistent radiologic determinant appeared to be the presence of sagittal imbalance. Disclosures: author 1: grants/research support; Company = Medtronic, Depuy Synthes, stock/shareholder; Company = IncredX; author 2: none; author 3: none; author 4: grants/research support; Company=Depuy Spine; author 5: grants/research support; Company=Depuy-synthes; author 6: grants/research support; Company=DePuy Synthes, K2M, consultant; Company=DePuy synthes, Biomet; author 7: grants/research support; Company = DePuy Synthes, consultant; Company=DePuy Synthes; author 8: grants/research support; Company=Depuy, consultant; Company=Stryker; author 9: grants/research support; Company = Depuy Synthes, consultant; Company=Depuy Synthes Medtronic; author 10: none.

P30 EVALUATION OF CELL BINDING PEPTIDE (P15) WITH SILK FIBRE ENHANCED HYDROXYAPPATITE BONE SUBSTITUTE FOR POSTEROLATERAL SPINAL FUSION IN SHEEP Martin Axelsen, Stig Mindedahl Jespersen, Søren Overgaard, Ming Ding Orthopaedic Research Laboratory, Department of Orthopedic Surgery and Traumatology, Odense University Hospital, Institute of Clinical Research, University of Southern Denmark, Odense C, Denmark Background: Spinal fusion is indicated in the surgical management of various spinal disorders. To ensure stabile fusion, bone graft materials are essential. Traditionally allo- or autograft has been used, but both are associated with limitations. Synthetic bone graft materials that reassemble today’s gold standard are highly desired. Uninstrumented posterolateral fusion (PLF) is one of the most challenging bone graft procedures because of large graft size and lack of external stability. P15 is a synthetic 15 amino acid peptide sequence, identical to the biding site for a2-b1 integrin on the surface of bone forming cells. The binding initiates natural intra- and extracellular signalling pathways, inducing production of growth factors, bone morphogenic proteins and cytokines. P15 peptide has previously shown to improve osteoinductive properties when coated on graft materials. Purpose: In this study, we compared fusion rates between silk fibre enhanced anorganic bovine derived hydroxyapatite matrix (ABM) with and without P15 peptide coating in uninstrumented PLF in a preclinical setting. Study design: Randomised prospective study in sheep. Method/materials: Twelve Tex/got sheep underwent open two level uninstrumented PLF at level L2/L3 and L4/L5. Levels were randomised to receive silk fibre enhanced ABM graft with or graft without P15 coating. The sheep were sacrificed after 4.5 months. Levels were harvested and evaluated with Micro-CT 50 scanner and qualitative histology. Fusion rates were assessed with 2D sections and

Eur Spine J (2015) 24 (Suppl 6):S743–S800 3D reconstruction images and fusion was defined as intertransverse bridging. Results: Spinal fusion was found in 72 % of levels receiving silk fibre enhanced ABM/P15 graft material. In levels with silk fibre enhanced AMB fusion rate was 41 %. These findings are significant different (P \ 0.05). No significant differences in three dimensional microarchitecture properties were found. Qualitative histology indicated more bone formation in levels with P15 coated graft material then without. Mostly woven bone was present but areas with laminar initiation were found. In both groups good osteointegration between graft and host bone was found. Conclusion: P15 peptide enhanced bone formation and fusion rates when coated on silk fibre enhanced ABM bone graft. Reported fusion rates were similar to previous reported rates using allo- or autograft bone. This preclinical study suggested that ABM/P15 might be a potential graft material for clinical use. Disclosures: author 1: none; author 2: none; author 3: grants/research support; Company=Cerapidics; author 4: none.

P31 MULTIPLE REGRESSION ANALYSIS OF FACTORS AFFECTING THE MENTAL COMPONENT SCORE CONSTITUENTS OF SF-36 IN ADULT SPINAL DEFORMITY Selim Ayhan, Selcen Yuksel, Asli Niyazi, Vugar Nabiyev, Umit Ozgur Guler, Emre Acaroglu, Montse Domingo-Sabat, Ferran Pellise, Ahmet Alanay, Francisco Javier Sanchez Perez-Grueso, Ibrahim Obeid, Frank Kleinstu¨ck, European Spine Study Group (ESSG) 1,4,5,6

Ankara Spine Center, Ankara, Turkey; 2Department of Biostatistics, Yildirim Beyazit University School of Medicine, Ankara, Turkey; 3Department of Psychology, Middle East Technical University Northern Cyprus Campus, Turkish Republic of Northern Cyprus; 7Raquis, Fundacio´ Institut de Recerca Vall d’Hebron, Barcelona, Spain 8Spine Unit, Hospital Vall d’Hebron, Barcelona, Spain; 9Acibadem University School of Medicine, Comprehensive Spine Center, Acibadem Maslak Hospital, Istanbul, Turkey; 10Spine Unit, Hospital Universitario La Paz, Madrid, Spain; 11Spine Unit, Bordeaux University Hospital, Bordeaux, France; 12Spine Unit, Schulthess Klinik, Zu¨rich, Switzerland; 13Fundacio´ Institut de Recerca Vall d’Hebron, Barcelona, Spain Background: As surgical decision-making and preoperative planning for adult spinal deformity (ASD) need strongly be interrelated to health related quality of life (HRQOL), there are multiple studies focusing on factors with an impact on it. Based on the general perception of association between the treatment results and the psychological condition of patients with ASD, analyzing the factors governing the baseline psychological status of this group may be worthwhile. Aim: To develop an understanding of which factors have a greater impact on the SF-36 mental component score (MCS) and establish a hierarchy of these parameters through multiple regression analysis. Patients and methods: Prospectively collected data from a multicentric adult deformity database was analyzed using multiple regression analysis. SF-36 MCS was detected as dependent variable; demographical, radiolographical and the HRQOL parameters were assigned as independent variables. The regression model was started with a correlation analysis between SF-36 MCS and all independent variables then conducted by introducing the variables with the highest correlation with SF-36 MCS sequentially.

S755 Results: A total of 229 patients (181F, 47M) with a mean age of 49.4 (range: 18-85) years were analyzed. A strong correlation between SF36 MCS and Scoliosis Research Society (SRS)-22, Oswestry Disability Index (ODI), gender, and diagnosis were found (p \ 0.05). Regression analysis showed that a one unit increase in SRS-22 increases SF-36 MCS by 0.403 (p \ 0.001) and a one unit increase in thoracic (T2-T12) kyphosis (TK) decreases SF-36 MCS by 0.194 (p = 0.012). In addition, being male increases SF-36 MCS by 0.197 compared to being female (p = 0.013). The overall R2 of this model was 0.254 (p \ 0.001). Conclusion: This study has demonstrated that, among the evaluated parameters, the overall HRQOL (SRS22 and ODI) as well as TK and gender are the most important parameters affecting the mental component summary of SF-36 in ASD population. Although the strong association with SRS-22 and/or ODI was to be expected, less strong associations with TK (as a token of appearance?) and gender (due to different mechanisms of coping with disability?) were less expected and may warrant further consideration in our understanding of the population of ASD. Disclosures: author 1: none; author 2: consultant; Company = Yildirim Beyazit University; author 3: none; author 4: grants/research support; Company=Medtronic, Depuy Synthes; author 5: none; author 6: grants/research support; Company=Medtronic, Depuy Synthes, stock/shareholder; Company=IncredX; author 7: grants/research support; Company=Depuy-synthes; author 8: grants/research support; Company=DePuy Synthes, K2M, consultant; Company=DePuy synthes, Biomet; author 9: grants/research support; Company=Depuy, consultant; Company=Stryker; author 10: grants/research support; Company=DePuy Synthes, consultant; Company=DePuy Synthes.

P32 COGNITIVE IMPAIRMENT FOLLOWING ADULT SPINAL DEFORMITY SURGERY Selim Ayhan, Vugar Nabiyev, Selcen Yuksel, Montse Domingo-Sabat, Ferran Pellise, Ahmet Alanay, Francisco Javier Sanchez Perez-Grueso, Frank Kleinstu¨ck, Ibrahim Obeid, Emre Acaroglu, European Spine Study Group (ESSG) 1,2,10

Ankara Spine Center, Ankara, Turkey; 3Department of Biostatistics, Yildirim Beyazit University School of Medicine, Ankara, Turkey; 4Raquis, Fundacio´ Institut de Recerca Vall d’Hebron, Barcelona, Spain; 5Spine Unit, Hospital Vall d’Hebron, Barcelona, Spain; 6Acibadem University School of Medicine, Comprehensive Spine Center, Acibadem Maslak Hospital, Istanbul, Turkey; 7Spine Unit, Hospital Universitario La Paz, Madrid, Spain; 8 Spine Unit, Schulthess Klinik, Zu¨rich, Switzerland; 9Spine Unit, Bordeaux University Hospital, Bordeaux, France; 11Fundacio´ Institut de Recerca Vall d’Hebron, Barcelona, Spain Background: Elderly patients undergoing major surgery may experience cognitive deterioration due to lesser plasticity in their brain tissue. This so called postoperative cognitive dysfunction (POCD) syndrome is characterized with non-specific dysfunction in memory, concentration and analysis skills. It is not known whether adult spinal deformity (ASD) surgery is associated with POCD. Purpose: To analyze the cognitive abilities of older patients undergoing spinal deformity surgery before and after the surgery so as to understand whether ASD surgery is associated with POCD. Methods: A prospective longitudinal study was performed on surgical patients older than 50 years enrolled in a prospective multicentric database. Mini mental state examination (MMSE) was performed to assess cognitive functions in addition to the health related

123

S756 quality of life (HRQoL) tests (SF-36, ODI and SRS-22) at preoperative, post-operative 6th week and 6th month points. Demographics, preoperative health status, comorbidities, surgical characteristics were also analyzed. Descriptive statistics and repeated measures of variance analysis were performed. Results: A total of 90 patients with a mean age of 67.4 ± 8.2 years were enrolled in the study; all had 6th week and 58 had both 6th week and 6th month follow-up MMSE evaluations. The averages (SD) of surgical time, estimated blood loss, number of levels fused and hospital stay were 240.1 (111.9), 1621.2 (1058.7), 11.2 (4.4) and 14.2 (11.45) respectively. On analysis, it was seen that there was even a slight increase in mean MMSE score between time points [for n = 90; Baseline = 26.9 ± 2.7; 6th week = 27.1 ± 2.5 (p = 0.16)], [for n = 58; Baseline = 26.9 ± 2.8, 6th week = 27.3 ± 2.6 (p = 0.99); 6th month = 27.6 ± 2.5 (p = 0.06)]. But, a decrease of [2 points (3 or 4 points) was also noted in 6 patients (6.7 %) at both time points. Conclusion: Although ASD surgery in older patients is recognized as challenging, this study suggests that it is not necessarily associated with a significant deterioration in the cognitive abilities of patients undergoing it. These results are different compared to those reported for other major surgical interventions. This may be due to the relatively minor influence of ASD itself on the cognitive abilities of the patients involved as well as to the relatively stable hemodynamic conditions obtainable during modern ASD surgery. Disclosures: author 1: none; author 2: grants/research support; Company = Medtronic, Depuy Synthes; author 3: consultant; Company=Yildirim Beyazit University; author 4: grants/research support; Company=Depuy-synthes; author 5: grants/research support; Company=DePuy Synthes, K2M, consultant; Company = DePuy synthes, Biomet; author 6: grants/research support; Company=Depuy, consultant; Company=stryker; author 7: grants/research support; Company=DePuy synthes, consultant; Company=DePuy Synthes; author 8: none; author 9: grants/research support; Company=Depuy Synthes, consultant; Company=Depuy Synthes Medtronic; author 10: grants/research support; Company = Medtronic, Depuy Synthes, stock/shareholder; Company=IncredX.

P33 CORONAL IMBALANCE IN DEGENERATIVE LUMBAR SCOLIOSIS: PREVALENCE AND INFLUENCE ON SURGICAL DECISION-MAKING FOR SPINE OSTEOTOMY Hongda Bao, Feng Zhu, Yong Qiu, Zezhang Zhu The Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing, China Introduction: Currently there is paucity of information on the prevalence of pre-operative coronal imbalance in patients with DLS and its influence on post-operative surgical outcomes. This study aims to investigate in DLS, the prevalence of coronal imbalance, to propose a novel classification system based on these radiographic parameters and to investigate whether pre-operative coronal imbalance affects clinical outcomes following osteotomy. Methods: A total of 284 DLS patients were recruited into this twostage study, among which 69 patients were treated by posterior-only correction and the remaining 215 patients received conservative treatment. Using pre-operative long-cassette standing coronal X-ray films, all patients were classified based on coronal balance distance (CBD) defined as the horizontal distance between C7 plumb line and central sacral vertical line: Type A, CBD \ 3mm; Type B, CBD [ 3mm and C7PL shifts to the concave side of the curve; Type C, CBD

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 [ 3mm and C7PL shifts to the convex side of the curve. The prevalence of pre-operative coronal imbalance was calculated for stage I study. Post-operative CBD in patients who received surgery was also measured to evaluate clinical outcomes for stage II study. Results: 34.8 % of the 284 patients presented pre-operative coronal imbalance with an average CBD of 48.5 ± 18.7 mm. More patients with Type B malalignment were observed compared to patients with Type C malalignment (62 vs. 37 cases). In stage II, 30.4 % (21/69) patients were identified with post-operative imbalance. More patients in Type C group presented with post-operative imbalance than patients in Type A or B group (P \ 0.001). At 1-year follow-up, less improvement was observed in terms of SF-36 PCS and VAS (P = 0.034 and 0.025, respectively) in Type C patients. Conclusion: For the first time, the present study reported with a large sample size that the prevalence of coronal imbalance in DLS was 34.8 %. Patients with Type C coronal malalignment may be at greater risk for post-operative coronal imbalance following posterior osteotomy. This novel classification based on coronal imbalance provides valuable insight during surgical decision-making in DLS. In addition, less correction of distal lumbosacral fractional curve in patients with pre-operative convex decompensation may lead to post-operative coronal imbalance. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P34 CHANGES IN THE SAGITTAL ALIGNMENT AND RANGE OF MOTION OF THE LUMBAR SPINE AFTER OBLIQUE IMPLANTED TOTAL LUMBAR DISC REPLACEMENT: A 2-YEAR PROSPECTIVE STUDY OF 52 CASES Eduardo Hevia, Juan Solaz, Alberto Caballero, Jesu´s Burgos, Carlos Barrios Spine Surgery Unit, Hospital La Fraternidad-Muprespa, Madrid, Spain; Institute for Research on Musculoskeletal Disorders, Valencia Catholic University, Valencia, Spain Introduction: Oblique implantable Total disc replacements (TDR) have been developed in an attempt to partially resect the anterior longitudinal ligament (ALL), together with additional partial resection of lateral annulus fibers. To date, the literature has not addressed the impact of the TDR oblique implantation on the lumbar spine sagittal alignment. This study was aimed at analyzing the following hypothesis: TDR at the L4-L5 level does not change the sagittal alignment and the range of motion of the lumbar spine when the implant is placed in accurate position. Methods: Prospective single-center radiological investigation of L4/5 TDR inserted through an oblique approach for the treatment of disc disease. A series of 52 patients with a minimum of 2-year FU after oblique TDR at L4/L5 level was analyzed for radiological changes in sagittal alignment and range of motion of the lumbar spine. The total sagittal lumbar lordosis (TSLL), the segmental sagittal lumbar lordosis (SSLL) of the operated level, and the range of motion of the TDR implant were determined in pre- and postoperative functional X-rays. The accuracy of the implant position was also evaluated. Results: A total of 52 patients (mean age, 42.7) were available. There were no revision surgeries for general and/or device-related complications. Only a 28.8 % of cases (n = 15) showed a satisfactory position. Off-center lateralized implants were the most common misplacements. Axial malrotated TDR accounted for the 28.1 % of cases. From 3 to 24 months of FU, differences in range of motion were found in the total L1-S1 flexion, and in the mean range of

Eur Spine J (2015) 24 (Suppl 6):S743–S800 motion of the implant both improving significantly. TDRs showing unsatisfactory implantation in the radiological studies (71.8 %) demonstrated similar lumbar and segmental range of motion in comparison to properly implanted TDRs. Conclusions: Oblique implanted L4/L5 TDR significantly increases total lordosis while retaining segmental lordosis, independently of the accuracy of its intervertebral position. Oblique TDR maintains anteroposterior segmental and total balance in most cases. Further studies should evaluate whether this finding has any implication for the longterm outcome. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

P35 SAGITTAL PLANE CORRECTION IS CORRELATED WITH QUALITY OF LIFE AT EARLY FOLLOW-UP IN ADULT DEFORMITY PATIENTS Evalina Burger, Andriy Noshchenko, Cameron Barton, Sean Molloy, Michael Chang, Vincent Fiere

S757 correlation between these treatment effects was weak, and other confounding factors should be studied. Disclosures: author 1: grants/research support; Company=Aesculap, consultant; Company=Medicrea, paradigm Signus; author 2: employee; Company=Colodo University Denver, USA; author 3: grants/research support; Company=Medicrea; author 4: grants/research support; Company=Metronic, consultant; Company=Zimmer, K2M, royalties; Company=Medicrea; author 5: grants/research support; Company=Medicrea, consultant; Company=Medtronic, Stryker; author 6: consultant; Company=Medicrea Clariance, royalties; Company=Medicrea clariance.

P36 TRANSDISCAL SCREW VERSUS PEDICLE SCREW FIXATION FOR HIGH-GRADE L5-S1 ISTHMIC SPONDYLOLISTHESIS IN PATIENTS YOUNGER THAN 60 YEARS. A CASE-CONTROL STUDY Isabel Collados, Alejandro Lizaur-Utrilla, Paloma Bas, Teresa Bas Hospital Universitario Y politecnico La Fe, Valencia, Spain

University of Colorado, Anschutz Medical Campus, USA; Centre Orthopedique Santy, Lyon, France; The Royal National Orthopaedic Hospital, Stanmore, U.K.; University of Arizona, Sonoran Spine Center, USA Introduction: Adult spinal deformity (ASD) is a debilitating condition defined by alteration of normal spinal curvature and is associated with decreased quality of life. Deformity correction often requires intensive surgery and has potential for major complications that may be outweighed by overall improvement in quality of life (QOL). The purpose of the study is to evaluate the correlation between sagittal plane correction and QOL for ASD patients treated with instrumented posterior fusion. Study design: Multi-center prospective cohort study. Methods: Seventy-two consecutive patients who underwent multilevel (4-18) posterior surgical correction for degenerative scoliosis and/or kyphosis were enrolled between 2012 and 2014. Sixty-two patients (48 female, and 14 male), mean age 60.6 years (Standard Deviation = 11.7) were eligible for analysis after exclusion. Patient quality of life questionnaires (SRS22 and ODI) and sagittal radiographic parameters (sagittal vertical axis (SVA), pelvic incidence minus lumbar lordosis (PI-LL), T1 pelvic angle (TPA) and pelvic tilt (PT)) were collected preoperatively and at 6 months follow-up. Questionnaire scores were compared to radiographic parameters to determine correlation and treatment effect (TE). Outcome measures: TE was defined as mean pre to post difference (MD); standardized mean difference (SMD) was defined as the ratio of MD to preoperative standard deviation. Results: Fifty three patients were available at 6 month follow up: compared to preoperative status, the ODI MD was -14 (St.D, 8.2), SMD = -0.94, p \ 0.001; the SVA MD was -15.3mm (St.D, 51.4), SMD = -0.28, p = 0.005; the rate of sagittal imbalance decreased to 24 % [95 %Cl: 12.2; 35.8]. The SVA TE had significant correlation with preoperative SVA (r = 0.78, p = 0.005) and postop status: SRS Function (r = 0.423, p = 0.002) and SRS Total (r = 0.314, p = 0.026). Correlation between SVA and ODI/SRS TE in the whole cohort did not meet significance (Spearman’s correlation = 0.24, p = 0.1). Statistical power analysis suggests that increasing of the number enrolled subjects to 100-120 would make these results statistically significant. Conclusions: Results of current study suggest that multilevel posterior instrumentation allows successful restoration of sagittal alignment in ASD patients. At 6 months follow-up, sagittal plane correction corresponded with improvement of QOL. However, the

Background: Surgical management of high-grade spondylolisthesis remains controversial, and posterolateral fusion with pedicle fixation is the gold standard. Transdiscal L5-S1 fixation has emerged as a treatment option based on a potential mechanical advantage but currently there are no comparative clinical studies. The purpose of this study was to compare outcomes between both fixation types. Methods: This was a retrospective case-control study with patients prospectively followed. Twenty-five consecutive patients (mean age, 36.7 years) who underwent transdiscal fixation, and thirty-one other patients (mean age, 42.0 years) underwent pedicle fixation were clinically and radiographically compared. Clinical assessment was performed using Oswestry Disability Index (ODI), Core Outcomes Measures Index (COMI), Short-Form 12 (SF-12), and pain visual analogue scale (VAS). Radiographic spinopelvic parameters were also evaluated. The mean follow-up was 2.7 years (range, 2.0 to 5.3). Results: Preoperative data were comparable between groups. Surgery time, blood loss, and hospital stay were similar between groups. At last follow-up, clinical and radiographic outcomes were significantly improved in both groups. Both lumbar and leg pain VAS were similar, but ODI (20.2 vs. 31.6, p = 0.010), COMI (1.6 vs. 2.8, p = 0.012), and SF-12 physical (84.3 vs. 61.5, p = 0.004) and mental (81.5 vs. 69.4, p = 0.021) scores were significantly better in the transdiscal group. The complication rate was similar in both groups. There were four pseudoarthroses in the pedicle group, and none in the transdiscal group. Conclusion: In this study, transdiscal fixation provided better functional outcomes at medium-term than conventional pedicle fixation for high-grade spondylolisthesis. Disclosures: author 1: grants/research support; Company=Sociedad Espan˜ola Cirugia Ortopedica y Traumatologia (SECOT); author 2: grants/research support; Company=Sociedad Espan˜ola de Cirugia Ortopedica y Traumatologia (SECOT); author 3: none; author 4: none.

P37 CAN SPINO-PELVIC PARAMETERS PREDICT HARDWARE FAILURE IN SCHEUERMANN’S KYPHOSIS PATIENTS? Eyal Behrbalk, Ofir Uri, Hossein Mehdian, Bronek Boszczyk, Masood Shafafy, Luigi Nasto, Radek Kaiser, Michael Grevitt Queen’s Medical Centre NHS Trust, Nottingham, UK

123

S758

Eur Spine J (2015) 24 (Suppl 6):S743–S800

Objectives: Proximal junction failure is one of the most common complications following surgical correction of Scheuermann’s Kyphosis (SK). When thoracic-kyphosis is corrected a compensatory decrease in lumbar-lordosis occurs. This study investigates the relationship between patients’ spino-pelvic characteristics and occurrence of proximal junctional complications. Methods: Retrospective case series of 29 patients. Spino-pelvic characteristics of 8 patients (age 20 ± 5 years, 7 males) who developed proximal-junction complications after surgical correction of SK were compared to those of 21 patients (age 21 ± 6 years, 18 males) who did not have complication after similar operation. Results: The preoperative and postoperative magnitude of the thoracic-kyphosis and lumbar-lordosis were similar in the complication and non-complication groups (p = ns). However, the pelvic-incidence and the preoperative sacral-slope were significantly higher in the complications group (52 ± 11 vs. 42 ± 10 and 38 ± 9 vs. 27 ± 7 respectively; p \ 0.05). Similarly, the preoperative sagittalvertical-axis was significantly more positive in patients who developed proximal-junction complications (24 ± 29 cm vs. -16 ± 29 cm; p = 0.002). Conclusion: Patients who developed proximal-junction complications had higher pelvic-incidence, higher sacral-slope and bigger preoperative sagittal-vertical-axis. This study shows that the compensatory reduction in lumbar-lordosis post SK correction in patients with high PI causes spino-pelvic mismatch and proximal junction failure. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: grants/ research support; Company=KSPINE.

calculated by subtracting the postoperative scores from the baseline scores. Multiple regression analyses were conducted to examine the relationship of the duration of symptoms and relevant MCID ratio values while controlling for independent variables. Further, a comparison between 2 groups of patients was performed to analyze the changes of clinical outcomes for the patients who underwent fusion within \ 24 months vs. [ 24 months. Results: It was determined that the duration of symptoms was a significant predictor of better leg pain resolution (p = 0.018), but not back pain resolution (p = 0.27), or improvement in ODI (p = 0.10) and SF-36 PCS scores (p = 0.19). The patients with shorter duration of symptoms had significantly better radicular symptom resolution (p = 0.032) compared to patients who waited at least 24 months or longer to undergo fusion. Conclusion: A shorter duration of symptoms was found to be a statistically significant predictor for better resolution of radicular symptoms in patients undergoing TLIF for painful degenerative disc disease, stenosis and spondylolisthesis. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P38 THE TIMING OF SURGERY AND SYMPTOM RESOLUTION IN PATIENTS UNDERGOING TRANSFORAMINAL LUMBAR INTERBODY FUSION FOR LUMBAR DEGENERATIVE DISC DISEASE AND RADICULOPATHY

Department of Health Sciences, Faculty of Earth and Life Sciences, EMGO+ Institute for Health and Care Research, VU University, Amsterdam, Netherlands

Sigita Burneikiene, Alan Villavicencio, Alexander Mason, Sharad Rajpal Boulder Neurosurgical Associates, Boulder, USA Introduction: Success rates of surgical interventions for lumbar disorders vary significantly depending on multiple factors and among them the duration of symptoms. It is not clear whether there is a ‘‘cutoff’’ time when decompression and fusion surgery becomes less effective in the conditions with chronic nerve root compression symptomatology. The main objective was to analyze if duration of symptoms has any effect on clinical outcomes and primarily resolution of radicular pain symptoms due to degenerative disc disease and stenosis with or without spondylolisthesis in patients undergoing transforaminal lumbar interbody fusion (TLIF). Methods: This is a prospective observational study. Eighty-four patients with radicular symptoms due to degenerative disc disease, stenosis or spondylolisthesis with no previous fusion surgeries and undergoing one- to three-level TLIF surgery were enrolled. Fifteen patients (18 %) were lost to follow-up and were excluded from this analysis leaving a total of 69 patients. Standardized questionnaires were used to analyze clinical outcomes and were administered preoperatively within 3 months of scheduled surgery, postoperatively at 3, 6, 12 and 24 months. To emphasize the change in clinical outcome scores, the relevant scores were calculated as the ratio of minimal clinically important difference (MCID) values and change scores. The change scores were

123

P39 DEVELOPMENT OF A CORE OUTCOME SET TO UPDATE THE EXISTING SET OF OUTCOME DOMAINS FOR LOW BACK PAIN Alessandro Chiarotto, Caroline Terwee, Maarten Boers, Raymond Ostelo

Background: Heterogeneity in outcomes reporting in clinical trials conducted in patients with non-specific low back pain (NSLBP) can hinder comparison of findings and reliability of systematic reviews. A core outcome set (COS) can address this issue as it defines a minimum set of outcomes that should be reported in all clinical trials. In 1998, Deyo et al. recommended a standardized set of outcome domains for LBP clinical research. The aim of this study was to update the existing set by determining which outcome domains should be included in a COS for clinical trials in NSLBP. Methods: An international Steering Committee worked on the development of this COS following the methodology suggested by initiatives like COMET and OMERACT. The OMERACT Filter 2.0 framework was used to draw a list of potential core domains that was presented in a three-round multidisciplinary and multi-stakeholder Delphi survey. Delphi participants had to judge which domains were core and a-priori criteria for achievement of consensus were established before each round. Comments provided by panellists on importance, overlap, aggregation and/or addition of potential core domains were also analysed. The Steering Committee discussed the final results and made final decisions. Results: A total of 280 experts was invited to participate in the Delphi survey; response rates in the three rounds were 52 %, 50 % and 45 %, respectively. Of 41 potential core domains presented in the first round, 13 had sufficient support to be presented for rating in the third round. Overall consensus was reached for the inclusion of three domains in this COS: physical functioning, pain intensity and healthrelated quality of life. Consensus on physical functioning and pain intensity was robust and consistent across all stakeholders, while health-related quality of life was not supported by the patients, and all the other domains were not supported by two or more groups of

Eur Spine J (2015) 24 (Suppl 6):S743–S800 stakeholders. Weighting all possible argumentations, the Steering Committee decided to include in the COS the three domains that reached overall consensus and the domain number of deaths. Conclusions: The following outcome domains were included in this updated COS: physical functioning, pain intensity, health-related quality of life and number of deaths. The next step in the development of this COS will be to determine which measurement instruments should be selected to measure the core outcome domains. Disclosures: author 1: none; author 2: none; author 3: none.

P40 A PROPOSED SET OF METRICS FOR STANDARDIZED OUTCOME REPORTING IN THE MANAGEMENT OF LOW BACK PAIN Rutledge Clement, Adina Welander, Caleb Stowell, ICHOM Working Group in Low Back Pain, Peter Fritzell The International Consortium for Health Outcomes Measurement, Capel Hill, USA Background: Outcome measurement has been shown to improve performance in several fields of healthcare. This understanding has driven a growing interest in value-based healthcare. Value is generally defined as outcomes achieved per money spent, so to understand and compare value, we must be able to measure outcomes and costs in standardized formats. However, while low back pain (LBP) constitutes an enormous burden of disease, no universal set of metrics has yet been accepted to measure and compare outcomes in clinical practice. Here, we aim to define such a set. Methods: An international group of 22 specialists in multiple disciplines of spine care was convened by the International Consortium for Health Outcomes Measurement (a non-profit joint venture between Harvard Business School, the Boston Consulting Group and the Karolinska Institute). The group included spine surgeons, non-operative spine providers such as physiatrists and pain specialists, registry leaders, nurses, researchers and a patient advocate. Members, who represented 9 countries from Europe, North America, Australia and Asia, reviewed existing literature and selected LBP outcome metrics through a 6 round modified-Delphi process. Results: Inclusion criteria for the proposed outcome set span degenerative lumbar conditions. Patient-reported metrics include numerical pain scales, lumbar-related function using the Oswestry Disability Index, health-related quality of life using the EQ-5D-3L questionnaire, and questions assessing work status and analgesic use. Specific common and serious complications were included. Recommended follow-up intervals include 6, 12 and 24 months after initiating treatment, with optional follow-up at 3 months and 5 years. Metrics for risk stratification were selected based on pre-existing tools such as the Charlson Comorbidity Index and questions used by large spine registries. Conclusions: The outcome measures recommended here are structured around specific etiologies of LBP, span a patient’s entire cycle of care, and allow for risk adjustment. Thus, when implemented, this set can be expected to facilitate meaningful comparisons and ultimately to provide a continuous feedback loop, enabling ongoing improvements in quality of care as well as cost evaluations linked to treatment effects. Much work lies ahead in implementation, revision, and validation of this set, but it is an essential first step towards a LBP community focused on maximizing value for its patients. Disclosures: author 1: none; author 2: employee; Company= Boston Consulting Group; author 3: none; author 4: none; author 5: none.

S759 P41 SIGNIFICANCE OF AQUATIC FUNCTIONAL TRAINING IN THE THERAPY OF CHRONIC LOW BACK PAIN Stefan Dalichau, Torsten Mo¨ller BG Unfallambulanz und rehazentrum Bremen, Germany Purpose: The aim of this investigation was the evaluation of the effects of an Aquatic Functional Training compared to alternative physiotherapeutic treatments in the therapy of chronic low back pain. Material and methods: In a controlled prospective study 96 male patients aged 36 to 49 years suffering chronic back pain ([ 2 years) were randomised to a control group (CG) without physiotherapeutic treatment and 4 test groups (TG), who carried out a back training program of 90 min over a period of five weeks twice weekly. The TG started with an warming up of 30 min (cycling, stretching, tonicizing) following by an special treatment of 60 min. TG1 completed the training program by means of auxotonic training devices, TG2 was supplied with a land program to improve coordinative abilities. TG3 passed through a combined program consisting of an aquatic and auxotonic training and the patients of TG4 practised therapy waterbased exclusively. Results: With regard to the selected criteria it was realized a homogenous level for all groups at the beginning of the investigation (T0). While the data obtained for the CG remained virtually unchanged over the period of investigation, in all TG a decrease in back pain (VAS) as well as in restrictions for activities of daily living (Oswestry LBPDQ) could be proved at the end of the intervention in T1 (5 weeks after T0). Furthermore an increase in balance of isometric peak torque of the trunk musculature (extension/flexion) and in quality of spinal proprioception (senses of position and strength) was analysed (p \ .05/.01). Without any further accompanied physiotherapeutic treatment the positive results for the TG could be stabilized 5 months later (T2). The strongest effects in T1 and T2 were evaluated for the combined program consisting of aquatic and auxotonic training, followed by training of coordinative abilities, water-based training and finally by auxotonic training. Conclusions: In the sample investigated the Aquatic Functional Training take effect especially as a treatment preparing effective auxotonic training of muscle strength. However, the problem of the exact measuring out the required dose of physical strain intensity by means of aquatic training needed for an active stabilization of the low back is not solved sufficiently up to now. Disclosures: author 1: none; author 2: none.

P42 IS THE ROUTINE USE OF MAGNETIC RESONANCE IMAGING INDICATED IN PATIENTS WITH SCOLIOSIS? Varun Dewan, Stephen Forster, Jacob Matthews, Matthew Newton Ede, Jwalant Mehta, Jonathan Spilsbury, David Marks, Adrian Gardner Royal Orthopaedic Hospital, Birmingham, UK Background: The increased prevalence of neural axis anomalies in scoliosis compared with the background population is well described. Despite this, the routine use of Magnetic Resonance Imaging (MRI) in scoliosis is controversial, with some surgeons reserving its use for particular patient criteria. At our institution, all patients with spinal deformity undergo MRI of their whole spine.

123

S760 The aim of the study was to define the prevalence of spinal and extraspinal anomalies in scoliosis patients at a single institution. Methods: The MRI reports were reviewed for all paediatric patients with scoliosis over the past 6 years. Information collected included patient demographics, curve pattern, spinal anomalies and extraspinal anomalies. Results: 854 (643F, 211M) scans were reviewed. The average age was 14 years. 143 neural axis abnormalities were identified in 114 patients (84F, 30M). The most common abnormalities seen were cerebellar tonsillar descent (n = 73, 8.5 %) and the presence of a syrinx (n = 47, 5.5 %). Neither age, sex, nor curve pattern were found to be predictors of neural axis anomalies in the juvenile and adolescent populations. 17 patients (14F, 3M, average age 13.8 years) were found to have anomalies unrelated to their spine. These incidental findings were most commonly in the abdomen and pelvis and some altered clinical management. Conclusion: This is the largest single centre study of the routine use of MRI in patients with scoliosis. We found that the prevalence of neural axis abnormalities is greater than previously reported. Further, none of the predictors of neural anomalies described in the literature, used by some surgeons as indications for selective MRI were found to be predictors in our study. Additionally we defined a number of extra spinal anomalies. We advocate the routine use of MRI in all patients with scoliosis as incidental spinal and extra-spinal findings may alter clinical management. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: consultant; Company=K2M; author 6: none; author 7: consultant; Company=Depuy Synthes Spine / Medtronic, royalties; Company=Depuy Synthes Spine, other financial report; Company=Stryker Spine / K2M; author 8: grants/research support; Company=Medtronic Spine.

P43 VALIDITY AND RELIABILITY OF AN OBJECTIVE MEASUREMENT OF FUNCTIONAL IMPAIRMENT IN LUMBAR DEGENERATIVE SPINE DISEASE: THE TIMEDUP-AND-GO-TEST (TUG-TEST) Oliver P. Gautschi, Nicolas R. Smoll, Marco V. Corniola, Holger Joswig, Gerhard Hildebrandt, Karl Schaller, Martin N. Stienen Department of Neurosurgery and Faculty of Medicine, University Hospital Geneva, Geneva, Switzerland Background: The indication for surgical treatment of lumbar degenerative spine disease is generally based on patient-related factors such as pain, functional disability and reduced health-related quality of life (hrQoL) in view of corresponding radiological imaging findings. In addition to subjective scores, objective measures of functional disability may be helpful in the process of clinical decision-making. Methods: In a prospective two-centre study, functional disability was determined in 253 patients scheduled for lumbar spine surgery and a representative cohort of n = 111 healthy subjects as the control group by a simple objective test, the timed-up-and-go (TUG) test. The TUG test outcome parameter (time in seconds (sec)) was correlated to validated subjective measures of pain (Visual analogue scale), functional disability scores (Oswestry, Roland Morris, etc.) and hrQoL (Short Form-12, Euro Quol etc.). Results: The TUG test showed excellent intra- (intraclass-correlation coefficient (ICC) 0.97) and inter-rater reliability (ICC 0.99) with a standard error of measurement of 0.21 and 0.23 sec, respectively. The

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 validity of the TUG test is demonstrated by appropriate relationships with the Visual Analogue Scale (VAS) back (Pearson’s correlation coefficient (PCC) 0.25) and VAS leg pain (PCC 0.29), functional impairment (Roland-Morris-Disability- (PCC 0.38) and OswestryDisability-Index (PCC 0.34)) as well as with health-related quality of life (SF-12 MCS (PCC -0.25); SF-12 PCS (PCC -0.32); and Euro-Qol (PCC-0.28)). The upper limit of normal was determined to be 11.52 seconds. Mild (\33th percentile), moderate (33th to 66th percentile) and severe disability ([66th percentile) as determined by the TUG test were determined as \13.4 seconds, 13.4-18.4 seconds and [18.4 seconds, respectively. Conclusions: The TUG test is a quick, easy-to-use, valid and reliable tool to objectively evaluate the functional impairment in patients with degenerative spine disease. Patients in the clinical setting with a TUG test time greater then 12 seconds can be considered to have functional impairment. Disclosures: author 1: none; author 2: no indication; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

P44 MICROSURGICAL ANULAR RESTORATION WITH A ZSUTURE FOLLOWING LUMBAR MICRODISCECTOMY: A RETROSPECTIVE COMPARATIVE STUDY ON 763 PATIENTS WITH A MEAN FOLLOW-UP OF 12.5 YEARS Oliver P. Gautschi, Martin N. Stienen, Eliane Rohner, Matthias Rohner Department of Neurosurgery and Faculty of Medicine, University Hospital Geneva, Geneva, Switzerland Background: Microdiscectomy for lumbar disc herniation (LDH) is one of the most common procedures performed by spine surgeons worldwide. It usually significantly reduces pain and restores both functional disability and health-related quality of life. The incidence of a postoperative recurrence, however, is encountered in 10-25 %. A large anular defect is known to increase the risk of recurrence. A perioperative anular restoration may thus decrease the risk of recurrence. Methods: A retrospective study was performed comparing the risk of recurrence after lumbar single-level microdiscectomy with (study group) or without (control group) anular restoration using a Z-suture. Included patients were consecutively operated between 05/1990-04/ 1994 (control group) and 09/1998-12/1998 (study group). Subsequent LDH recurrence was recorded analysing the patients’ records and rereferrals until a mean follow-up of 12.5 years. Results: A total of n = 763 patients were included, of which 374 patients received a Z-suture (study group) and 389 patients served as controls (control group). The study groups did not differ in age (mean; 44.8 vs. 45.1 years; p = 0.86), gender (54.8 and 51.7 % males; p = 0.38) or distribution of the lumbar segments (p = 0.46). Surgical time was significantly longer in the study group (81.6 vs 69.3 min., p \ 0.0001). The rate of re-operation for a local LDH recurrence was significantly lower in the study group at three months (0 % vs. 1.8 %, p = 0.02) and remained to show a clear trend at twelve months postoperative (2.1 % vs. 4.6 %, p = 0.07). At the time of last followup (mean; 8.3 vs. 16.6 years, p \ 0.0001), local recurrence was observed in 21 out of 374 (5.6 %) in the study group compared to 36 out of 389 (9.3 %) in the control group (p = 0.06). The subsequent risk reduction of reoperation for local recurrence was 100 %, 54.3 %, and 39.8 % after three months, twelve months, and at long-term follow-up using a Z-suture. There were no complications associated with the Z-suture in this series.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Conclusions: Our data indicate that a Z-suture as anular restoration is safe and prolongs the surgery by about 12 minutes. It reduces the recurrence rate at three and twelve months after microdiscectomy, while its impact on the long-term follow-up has to be interpreted with caution due to the longer follow-up of our control group. Therefore, anular defect restoration should be considered as an option to reduce the local recurrence rate for patients undergoing lumbar microdiscectomy. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P45 DOES OBESITY AFFECT OUTCOMES AFTER DECOMPRESSIVE SURGERY FOR LUMBAR SPINAL STENOSIS? - A MULTICENTER OBSERVATIONAL REGISTRY-BASED STUDY Charalampis Giannadakis1, Ulf Nerland1, Ole Solheim1, Clemens Weber2, Sasha Gulati1 1

Department of Neurosurgery, St. Olavs University Hospital, Trondheim, Norway; 2National Advisory Unit on Spinal Surgery Center for Spinal Disorders, St. Olavs University Hospital, Trondheim, Norway Objective: The primary aim was to evaluate the association between obesity (body-mass index C30) and outcomes at one year using the Oswestry Disability Index (ODI) after laminectomy or microdecompression for single and two-level lumbar spinal stenosis (LSS). Materials and methods: Prospective collection of data from the Norwegian Registry for Spine Surgery. Results: All patients (n = 1473) improved in ODI at one year (16.7 points, 95 % CI 15.7-17.7, p \ 0.001). The improvement in ODI was 17.5 points in non-obese and 14.3 points in obese (p = 0.007). Obese patients were less likely to achieve a minimal clinically important difference in ODI than non-obese patients (62.2 vs 70.3 %, p = 0.013). Obesity was identified as a negative predictor for ODI improvement in multiple regression analysis (p \ 0.001). Non-obese patients experienced more improvement in both back pain (0.7 points, p = 0.002) and leg pain (0.8 points, p = 0.001) measured by numeric rating scales. Duration of surgery was shorter for non-obese patients for both single (79 vs 89 minutes, p = 0.001) and two-level (102 vs 114 minutes, p = 0.004) surgery. There was no difference in complication rates (10.4 % vs 10.8 %, p = 0.84). There was no difference in length of hospital stays for single (2.7 vs 3.0 days, p = 0.229) or two-level (3.5 vs 3.6 days, p = 0.704) surgery. Conclusion: Both non-obese and obese patients report considerable clinical improvement one year after surgery for LSS, but improvement was less in obese patients. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

P46 IMPROVED OUTCOME OVER TIME AFTER SURGERY FOR DDD. A REPORT FROM THE NATIONAL SPINE REGISTER IN SWEDEN; SWESPINE Olle Ha¨gg, Peter Fritzell, Bjo¨rn Stro¨mqvist, Bjo¨rn Knutsson Spine Center Go¨teborg, Neuroorthopedic department, Ryhovs Hospital, Jo¨nko¨ping; Department of Orthopedics, Lund University

S761 Hospital, Lund; Department of Orthopedics, Sundsvall Hospital, Sundsvall, Sweden Introduction: The Swedish spine registry, Swespine, was launched nationally in 1998 and has a coverage of approximately 90 %, and a completeness of 75 %-80 %, with yearly reports and availability for all surgical units to compare their results with the national average. Since 2007 we offer public benchmarking of outcome from participating spinal units at the site of the Swedish Society of Spine Surgeons (www.4s.nu). Has this changed the outcome? Material: Between 2000 and 2012 the yearly number of recorded primary procedures for DDD increased from 150 to 375. This increase reflects both a true increase in numbers, and an increase in recording. Swespine contains 4298 primary procedures for DDD. Of these 2975 completed the 1-year follow up. Primary outcome measure was patient rated Global Assessment of pain with reply alternatives ‘‘Completely gone’’, ‘‘Much better’’, ‘‘Somewhat better’’, ‘‘Unchanged’’ and ‘‘Worse’’. Success is defined as ‘‘Completely gone’’/ Much better’’. Secondary outcome measure was Satisfaction with treatment outcome, with the alternatives ‘‘Satisfied’’, ‘‘Unsure’’ and ‘‘Dissatisfied’’. Searching for predictors of outcome, we used multivariate analysis of regression and to analyze changes of predictors over time we divided the observation time into two periods, 2000-2005 and 2006-2012. Results: In 2000 the success rate was 58 %. There has been a slight continuous, improvement with 72 % success in 2012. Outcome measured as satisfaction showed a similar trend, from 68 % to 79 % satisfied with treatment outcome in 2012. The multivariate analysis showed male gender, smoking, preop sick leave and low preop scores of the EQ-5D significantly associated with poor outcome. Comparing the two periods we found fewer smokers, fewer patients on sick leave preop, higher preop scores of EQ-5D and more disc replacement procedures during 2006-2012. Discussion: According to national registry data, during the studied period, outcome after spine surgery for DDD has improved, despite a concurrent increase in the number of procedures performed. Still, however, 6 % have no benefit at all and 7 % have worse pain after surgery. These failures are the future challenge. Disclosures: author 1: stock/shareholder; Company=Spine Centre Go¨teborg AB, employee; Company=Spine Centre Go¨teborg AB; author 2: none; author 3: none; author 4: none.

P47 THE RADIOLOGICAL AND CLINICAL TREATMENT EFFECT OF A PERCUTANEOUS INTERSPINOUS DEVICE IN SPINAL STENOSIS - A COHORT STUDY WITH A ONEYEAR FOLLOW-UP Christian Hagelberg, Wisam Witwit, Christer Johansson, Hanna Hebelka Bolminger, Helena Brisby, Adad Baranto Institution of Clinical Sciences at Sahlgrenska Academy University of Gothenburg and Sahlgrenska University Hospital, Gothenburg, Sweden Study design: Prospective cohort study. Objective: The present study aims to contribute to the knowledge of the radiological outcome regarding the central spinal canal and foraminal area, the patient reported clinical outcome and the relationship between these entities after implantation of a percutaneous interspinous device (IPD).

123

S762 Summary of background data: Open decompressive surgery for spinal stenosis mostly provide symptom relief, but it is also associated with complications, both related to the procedure itself and worsening of co-morbidities. Percutaneous interspinous devices are a less invasive treatment alternative where divergent results has been reported. Limited data is available regarding the effect on widening of the spinal canal. Methods: 31 patients with a posture dependent pseudo claudication and a MRI verified cross-sectional dural area less than 100 mm2 (Mean 56, 7) received a percutaneous IPD on one or two levels. The narrowest level was considered clinically significant and was used for analysis. The Radiological treatment result was monitored through a post-op MRI and patient related outcome parameters via the Swedish Spine Registry. Primary outcome measures was the cross-sectional dural area and VAS leg pain, the secondary outcome measures was the foraminal area, Oswestry disability index, EQ-5D, back pain VAS, patient reported satisfaction with outcome and reoperation rate. Results: The cross-sectional dural area increased with 11,85 mm2 (P = 0.015). Also clinical parameters such as leg pain improved, decrease of -20,83 VAS units (P = 0,006), ODI -13,3 (P = 0,001) and an increase of EQ-5d index 0,26 (P = 0,001). Patient satisfaction however remained low with 36,7 % of the patient being ‘‘satisfied’’, 23,3 % ‘‘doubtful’’ and 40 % ‘‘dissatisfied’’ No correlation was seen between radiological and clinical treatment effect. 3 patients (9,7 %) required open decompression during the follow-up time. Conclusion: The IPD increased the dural area and the patients reported a moderate clinical benefit but interestingly no clear correlation between radiological treatment effect and the clinical outcome were detected. Somewhat surprising, the satisfaction with the treatment was low despite the clinical and radiological effects. Level of evidence: 3 Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

P48 CLINICAL RESULTS AND COMPLICATIONS OF SURGICAL TREATMENT FOR THORACIC MYELOPATHY CAUSED BY OSSIFICATION OF THE POSTERIOR LONGITUDINAL LIGAMENT: A MULTICENTER RETROSPECTIVE STUDY Mitsumasa Hayashida, Katsumi Harimaya, Ken Maeda, Hideki Ohta, Kenzo Shirasawa, Kuniyoshi Tsuchiya, Kazumasa Terada, Kozo Kaji, Tsuyoshi Arizono, Yukihide Iwamoto Department of Orthopaedic Surgery, Graduate School of Medical Sciences, Kyushu University Purpose: The Surgical outcome for T-OPLL have not necessarily improved and the possibility of postoperative neurological deterioration remains a major risk. The purpose of this study was to evaluate the surgical outcome and related risk factors of T-OPLL in recent years. Methods: This study included 57 patients with thoracic myelopathy caused by T-OPLL who underwent operations at institutions in the Fukuoka Spine Group from 2000 to 2010. Of these patients, 32 were females and 25 were males, and the mean age was 56.0 years at the time of surgery. The mean follow-up period was 5.3 years. The modified Japanese Orthopaedic Association scoring system and modified Frankel score were used to evaluate the clinical outcome of the patients. We obtained the patient data, including sex, age, disease

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 duration, BMI, comorbidity, ossification type, level of maximum ossification, surgical methods, decompression and fusion levels, surgical time, total blood loss and complications. To determine the most important prognostic factors related to the surgical outcome, these factors were evaluated by uni- and multivariate analyses. Results: The recovery rate was 50 % or higher in 20 patients (35.1 %) . A neurological status of at least one grade according to the modified Frankel score was improved in 33 patients (57.9 %), remained unchanged in 14 patients (24.6 %) and deteriorated in 10 patients (17.5 %). Cerebrospinal fluid (CSF) leakage was observed in 14 patients (24.6 %). The use of instrumentation was significantly related with a better outcome as previously reported. The vertebral fusion levels with instrumentation not less than the decompression levels led to a better clinical outcome compared with the fusion levels less than the decompression levels. Eight out of 14 patients (57.1 %) with CSF leakage were neurologically deteriorated, which was not associated with the type of surgical procedure. None of the other factors was found to be associated with the surgical outcome. Conclusions: Our findings demonstrated that the use of instrumentation should be considered with posterior decompression for T-OPLL and the vertebral fusion levels with instrumentation should be not less than the decompression levels, due to the enhancement and preservation of the decompression effect. We therefore strongly believe that the complication of CSF leakage should be avoided as far as possible due to the high possibility of postoperative deterioration, regardless of the surgical procedure used. Disclosures: author 1: none; author 2: none; author 3: none; author 4: no indication; author 5: none; author 6: none; author 7: none; author 8: none; author 9: none; author 10: none.

P49 SEGMENTAL STABILITY FOLLOWING MINIMALLY INVASIVE DECOMPRESSIVE SURGERY WITH TUBULAR RETRACTOR FOR LUMBAR SPINAL STENOSIS Naoki Higashiyama, Taku Sugawara Department of Spinal surgery, Research Institute for Brain and Blood Vessels-Akita Introduction: Minimally invasive decompressive surgery using microscope with tubular retractor (MISMT) has been widely performed for the treatment of lumbar spinal stenosis (LSS). Postoperative spinal instability affects the clinical outcomes after decompressive surgery, however, there were few reports regarding the segmental stability following this surgery. In this study, clinical and radiological outcomes were evaluated. Methods: The study involves a retrospective analysis of patients who underwent microscopic bilateral decompression via a unilateral approach with tubular retractor for LSS between June 2008 and Feb 2013. Clinical outcomes were evaluated using the Japanese Orthopedic Association (JOA) score, visual analogue scale (VAS), and pain vision scale (PVS) before surgery and at the final followup. Radiographic parameters were assessed using percentage slip on standing lateral radiographs, range of motion (ROM) of the intervertebral disc space, and posterior intervertebral angle on flexion. Results: One hundred and forty-four patients were followed for at least 2 year after MISMT for LSS. The mean patient age was 68 years (range 34-85 years), and there were 89 men and 55 women. The mean follow-up period was 42 months (range 24-80 months). Clinical

Eur Spine J (2015) 24 (Suppl 6):S743–S800 outcomes showed significant improvement of JOA score (from 15.5 to 24.8), VAS (from 49.8 to 14.4) and PVS (from 96.0 to 20.7) at the final follow-up. Radiographic parameters were not significantly altered (percentage slip from 7.1 to 7.4, ROM form 7.4 to 7.3, posterior intervertebral angle from-1.0 to 1.8) at final follow-up. Conclusion: In this study, MISMT showed excellent clinical outcomes and preservation of spinal segmental stability. The use of tubular retractor for microsurgical decompression for LSS is a safe and effective treatment modality, but long-term observation is necessary. Disclosures: author 1: none; author 2: none.

S763 P51 THE EFFECT OF FUSION LEVEL ON THE RADIOLOGIC AND FUNCTIONAL OUTCOMES IN THE SURGICAL TREATMENT OF ADULT DEFORMITY IN PATIENTS OLDER THAN 65 YEARS-OLD Erden Erturer, Sinan Yilar, Meric Enercan, Sinan Kahraman, Bahadir Mutlu, Mutlu Cobanoglu, Tunay Sanli, Mercan Sarier, Cagatay Ozturk, Azmi Hamzaoglu Istanbul Spine Center, Istanbul, Turkey

P50 OUTCOMES OF MINIMALLY INVASIVE LUMBAR FUSION (MILIF) IN PATIENTS WITH STENOSIS: SUBGROUP ANALYSIS Ulrich Hubbe, Roberto Assietti, Khai Lam, Hamid Khoshab, Kai Scheufler, Salvador Fuster, Joerg Franke Universita¨tsklinikum Freiburg, Germany Objective: To compare the outcomes of MILIF for Degenerative Lumbar Disorders (DLD) in a one year multicenter prospective observational study: Patients with (S)/without stenosis (NS) and between stenosis patients with (S+D)/without decompression (S+ND). Patients and methods: Two hundred and fifty two patients from a multicenter 1 year prospective observational study (NCT01143324), underwent 1- (83.0 %) or 2-level (17.0 %) MILIF (TLIF: 95.0 %; PLIF: 5.0 %) for treatment of leg pain (52.0 %), back pain (38.9 %) or claudication (9.1 %) due to DLD, including spondylolisthesis (52.8 %), stenosis (71.4 %), and/or disc pathology (93.7 %). Time (days) to first ambulation (TFA) and postsurgical recovery (TPSR), VAS back/leg pain and ODI pre-/post-surgery (4 weeks: 4w, 12 months: 12m) and changes from baseline [all variables: medians and interquartile ranges (IQR), Mann-Whitney U-test] were compared for S vs NS (N = 180 vs N = 72) and S+D vs S+ND (N = 142 vs N = 8) patients. Results: All outcome measures improved significantly in all groups. TFA was similar for all subgroups [S/NS 1.0 (0.5)/1.0 (1.0) p = 0.0702; S+D/S+ND 1.0 (0.0)/1.0 (1.0) p = 0.9473)]. TPSR was higher for S vs NS [3.0 (2.5)/2.0 (1.0) p = 0.0021] and lower for S+D vs S+ND [2.0 (2.0)/4.0 (3.0) p = 0.0017)]. At baseline, VAS leg pain was significantly higher in S vs NS [7.0 (3.0)/5.0 (4.0) p = 0.0014 but similar at 4w [2.0 (5.0)/1.0 (4.0) p = 0.3805] and at 12 m, [1.0 (4.0)/ 1.0 (5.0) p = 0.6425] due to a more pronounced drop from baseline in S (4w: 3.9 vs 3.0, p = 0.0784; 12 m: 5.0 vs 3.0, p = 0.0185). ODI improved slightly more (n.s.) in S vs NS and VAS back pain was equal in these groups. VAS leg pain, ODI and VAS back pain pre-/post-surgery as well as changes from baseline were similar in S+D compared to S+ND. Conclusions: Stenosis patients need longer time for surgery recovery, present more leg pain at baseline but better improvement after MILIF. The non-stenosis patients present similar outcomes at 4 w and 12 m. Decompressed stenosis patients recover sooner from surgery than non-decompressed patients. Disclosures: author 1: grants/research support; Company=Medtronic, consultant; Company=Medtronic; author 2: none; author 3: none; author 4: none; author 5: consultant; Company=Medtronic; author 6: none; author 7: grants/research support; Company=Baxter, Medtronic, Zimmer, Relievant, consultant; Company=Medtronic, Silony, Medacta, Exp Orthopedics, royalties; Company=Ohst AG.

Summary: Adult deformity patients who underwent long fusion up to T2 demonstrated better clinical and radiological outcomes at the end of a 3 year follow-up. Although prophylactic vertebroplasty was performed in order to prevent PJK, patients with short fusion (UIVT10) demonstrated higher rate of PJK (26.6 %) than patients with long fusions (UIV-T2). Design: Retrospective. Introduction: In this study, we aimed to demonstrate and compare the treatment outcomes of patients with adult deformity surgery and who underwent either short (T10-S1) or long (T2-S1) level fusion. Methods: 75 pts, [ 65 yrs, underwent fusion surgery for adult spinal deformity between 2008-2013 were reviewed. The patients were separated into 2 groups based on their fusion levels. Group 1 included 30 pts (22F, 8M) with upper instrumented vertebra (UIV) at T10 and Group 2 included 45 pts (40F, 5M) in whom the fusion level was stopped at T2. SRS 22, ODI and VAS were used for clinical evaluation. Radiologic studies included measurement of TK, LL, PI, PT, SVA, TPA and PJK. Radiologic and clinical results were compared. Results: Mean age was 70.1 (65-81) and mean f/up was 36.4 m (24-85) in G.1 Mean age was 70.4 (65-84) and mean f/up was 37.3 m (24-88) in G2. 6 pts (20 %) in G1 and 4 pts (8.8 %) in G2 underwent revision surgery. Indications of revision in G1 were development of PJK in 2, implant failure/pseudoarthrosis in 4 pts. In G2 the indications were development of PJK in 1, implant failure/pseudoarthosis in 3 pts. Clinical results assessed at f/up were significantly better in G2. Operative time and amount of bleeding were greater in G2. Radiologic evaluation showed that the deformities in all planes were corrected in the early period in both groups. F/up measurements showed that the correction could be preserved more in the G2. Radiologically, 8 patients (26.6 %) had PJK/PJF in G1, and 3 patients had PJK in G2 (10 %). Conclusion: Despite the application of prophylactic vertebroplasty, the development of PJK was greater in patients with UIV at T10 (26.6 %), compared to T2 (10 %). The clinical and radiologic outcomes obtained during the early period were similar in both groups, however at the end of a 3 year f/up those patients in whom the fusion was stopped at T2 had higher rates of maintaining the corrections in the sagittal plane and also had better clinical outcomes. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: one; author 6: none; author 7: none; author 8: none; author 9: none; author 10: consultant; Company=Medtronic.

P52 DISTAL ILIAC SCREW (DIS) FIXATION TECHNIQUE: AN ALTERNATIVE ILIOPELVIC FIXATION TECHNIQUE IN ADULT DEFORMITY SURGERY Meric Enercan, Sinan Kahraman, Bahadir Gokcen, Tunay Sanli, Mutlu Cobanoglu, Erden Erturer, Cagatay Ozturk, Ahmet Alanay, Mercan Sarier, Azmi Hamzaoglu Istanbul Spine Center, Istanbul, Turkey

123

S764 Summary: Distal iliac screw (DIS) fixation technique which has a more distal starting point than traditional iliac screw fixation provided rigid stability for lumbosacral fusion with very low rate of implant related complications even in osteoporotic patients. Design: Retrospective study Introduction: We start to use a freehand distal iliac screw (DIS) fixation with a more distal starting point (posterior inferior iliac spine) as an alternative lumbopelvic fixation technique in adult deformity surgery. DIS fixation does not require any cortical bone resection for entry and has low profile than traditional iliac and S2AI fixation. DIS fixation biomechanically provided greater insertional torques, axialpull out and toggle forces than traditional iliac fixation in our cadaveric study. The main disadvantage of the technique is the additional distal soft tissue exposure for the placement of the screw. The purpose of this study is to evaluate the clinical outcomes of DIS fixation in adult deformity surgery. Methods: 61 patients (43F, 18M) who underwent a long fusion ([5 levels) to the sacrum with DIS fixation were reviewed. Preop, postop, f/up standing AP/L, pelvis AP were reviewed for radiological data. Results: Mean age was 61.8 yrs (47-84), mean f/up was 28.8 months (24-38). Ave instr. level was 9.6 levels (5-16). In 42 patients (69 %) with BMD \-2.5 T score, cement augmented fenestrated pedicle fixation technique (except S1 and DIS) was performed to augment posterior fixation. Mean iliac screw length was 95.2 mm (80-100 mm). Iliac screw diameters were 7, 5 mm in 11 pts, 8, 5 mm in 26 pts and 9.5 mm in 4 pts. In addition to lumbopelvic fixation, interbody fusion for L5-S1 level was performed in 70 % (43 pts.) of the patients. Posterior instr. was augmented with multi-rod fixation in 41 pts (67 %). Complications related to DIS were; 6 screws (4.9 %) had loosening [ 2 mm in 3 pts. There were no pseudoarthrosis or implant failure related to lumbosacral joint. ODI showed a significant decrease from 75 .6 to 28.4 and VAS scores improved 7.8 to 4.2 postoperatively. Conclusion: DIS fixation provided the required stability for lumbosacral fusion and demonstrated very low rate of complications even in osteoporotic patients. DIS fixation technique is a good alternative for lumbosacral fixation in adult deformity surgery. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: grants/ research support; Company=Depuy Synthes, consultant; Company=Stryker; author 9: none; author 10: consultant; Company=Medtronic.

P53 CLINICAL RESULTS OF DYNAMIC STABILIZATION ADJACENT TO FUSION LEVEL: A NEW LUMBAR HYBRID INSTRUMENTATION Meric Enercan, Bahadir Gokcen, Sinan Kahraman, Mutlu Cobanoglu, Sinan Yilar, Tunay Sanli, Erden Erturer, Cagatay Ozturk, Mercan Sarier, Azmi Hamzaoglu Istanbul Spine Center, Istanbul, Turkey Summary: New hybrid instrumentation with PEEK rod system provides rigid stability for fused levels and dynamic portion helps to prevent adjacent segment degeneration at the end of 2 years followup. Design: Retrospective. Introduction: Adjacent segment degeneration is a common (34 %) problem following posterior spinal fusions in long term f/up. We have been using a new hybrid design which has a dynamic portion made of silicone and PEEK aiming motion preservation and fusion portion is

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 entirely made of PEEK. The aim of this study is to evaluate the efficiency of dynamic portion of the PEEK rod system in preventing adjacent level problems in the surgical treatment of multilevel lumbar degenerative disease. Methods: 54 patients (28F, 26M), mean age 48,2 yrs (26-65) with 84 levels of TLIF’s with more than 2 years of f/up were reviewed retrospectively. Preop, postop AP/L x-rays were measured for pelvic and sagittal parameters. Disc angles, ROM, anterior disc height (ADH) and posterior disc height (PDH) were measured for adjacent (AL) and supraadjacent (SAL) levels. All patients were evaluated with EOS images, dynamic x-rays and 3D CT scan at the final f/up. Clinical evaluation was done with ODI and VAS. Results: Mean f/up was 26,3 months (24-38).Ave instr. levels was 3,33 (2-5) and ave fused levels was 1,66 (1-3). TLIF’s were at L5-S1 in 42 pts, L4-5 in 35 pts, L3-4 in 6 pts and L2-3 in 1 patient. TLIF’s were single level in 10 pts, 2 levels in 28 pts and 3 levels in 6 pts. Preop LL was restored to 42.7° and 49.3° at final f/up. There were no significant differences in ADH, PDH and disc angles between preop and f/up for AL and SAL levels. Preop ave ROM for SAL of 5,85° changed to 6.57°. Preop ave ROM of 6,72° was decreased to 5,07° at AL with a limitation of 24,6 % postoperatively. 3D CT evaluation revealed solid fusions for all TLIF levels. Mean of 43,51 % ODI was improved to 18,93 and preop VAS score 7,2 was improved to 2,2. Conclusion: New hybrid lumbar instrumentation with PEEK rod system is effective in the treatment of multilevel degenerative lumbar disc disease. Dynamic portion of the hybrid system limits ROM by 24.6 % at adjacent level. AL and SAL did not demonstrate any significant facet or disc degeneration at the end of min 2 yrs f/up. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: one; author 6: none; author 7: none; author 8: none; author 9: none; author 10: consultant; Company=Medtronic.

P54 THE OUTCOMES OF INSTRUMENTED POSTEROLATERAL LUMBAR FUSION IN PATIENTS WITH RHEUMATOID ARTHRITIS Ji-Hoon Shim, Seung-Pyo Suh, Chul-Woong Kim, Chang-Nam Kang Hanyang University, Seoul, South Korea To evaluate the clinical and radiologic outcomes of instrumented posterolateral fusion (PLF) performed in rheumatoid arthritis (RA) patients, forty patients with RA who underwent instrumented lumbar PLF for spinal stenosis at our institute were matched for age, sex, bone density, smoking status, and number of fusion segments to a contemporaneous 134 patients without RA. Visual analogue scale (VAS) score and Korean Oswestry disability index (KODI) were evaluated before, 1 year after, and 2 years after the surgery to compare clinical outcomes and investigate patients for infection, neurological damage, and revision. Bone union in patients was examined through Lenke classification on a plain radiograph, and implant problems and adjacent segment disease (ASD) were also examined. The average age of the RA group was 64.3. The group included 1 male and 39 female patients, the average number of fusion segments was 3.0, and the average follow-up period was 36.4 months. The average age in the non-RA group was 65.3, with the group including 17 male and 117 female patients, the average number of fusion segments was 2.9, and the average follow-up period was 39.1 months. There were no significant differences between the two groups concerning age, sex, bone density, smoking history, diabetes history, surgery time, and the number of fusion segments. Both groups showed significant improvement in clinical symptoms after surgery,

Eur Spine J (2015) 24 (Suppl 6):S743–S800 while the RA group exhibited a deterioration of clinical outcomes due to early complications during the 2nd year after surgery. The complications occurred at a significantly higher rate in the RA group (19 cases, 47.5 %) than in the non-RA group (23 cases, 17.1 %) (p \ 0.001). The RA group showed symptomatic nonunion in 7 cases, implant problems in 6 cases, ASD in 7 cases, infection in 4 cases, and neurologic complications in 1 case. A total of 15 cases (37.5 %) required revision due to complications. On the other hand, the nonRA group showed symptomatic nonunion in 2 cases, implant problems in 7 cases, ASD in 3 cases, and infection in 4 cases. However, there were no cases of neurologic complications, and 12 cases (8.9 %) required revision (p \ 0.001). A number of approaches from many angles to reduce complications in RA patients are required. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P55 EXPANDABLE TECHNOLOGY IN MINIMALLY INVASIVE TLIF: A MULTICENTER CLINICAL AND RADIOGRAPHIC ANALYSIS OF 202 PATIENTS WITH TWO YEAR FOLLOW UP Choll Kim, James Lindley, Todd Doerr, Phillip St. Louis, Ingrid Luna, Piotr A. Kowalski, Gita Joshua Musculoskeletal Education and Research Center (MERC), a division of Globus Medical Inc., Audubon, USA Background: Static interbody cages require impaction for insertion while cages with incremental expansion allow for ease of insertion and optimized endplate contact. This study served to document clinical and radiographic outcomes in patients who had a minimally invasive Transforaminal Lumbar Interbody Fusion (TLIF) with a device which offers controlled in-situ expansion. Purpose: The purpose of this clinical study was to assess clinical and radiographic outcomes in patients implanted with an expandable interbody fusion device. Study design: Multicenter, retrospective analysis consisting of 24 month follow up postoperatively. Patient sample: 202 patients with clinical evidence of degenerative lumbar disc disease at 1 or 2 level(s). Outcome measures: Patient demographics and intraoperative measures were quantified. Patient outcomes including radiographs, Visual Analog Scale (back and legs) (VAS), Oswestry Disability Index (ODI), and Odom’s Criteria and complications were recorded preoperatively and at 6, 12 and 24 months postoperatively. Methods: A total of 202 patients comprise the basis of this retrospective analysis using an expandable interbody spacer combined with transpedicular posterior stabilization. Device-related complications were defined as implant breakage, migration, subsidence, and revision surgery at the index level. Results: Mean VAS and ODI scores decreased significantly from preoperative to the 24 month postoperative interval (p \ 0.05). Intervertebral disc heights (0.6 ± 0.1 vs 1.1 ± 0.2 cm) and neuroforaminal heights (1.7 ± 0.4 vs 2.0 ± 0.3 cm) increased significantly and were maintained throughout 24 months (p \ 0.05). There were no cases of device failure, however, asymptomatic migration or subsidence was present in 12 (5.9 %) patients and the overall reoperation rate at the index level was 2.97 % (n = 6), which was secondary to pedicle screw failure and pseudoarthrosis. Conclusions: The current study served to document the use of expandable interbody implant technology for the treatment of lumbar discogenic pathology based on clinical outcomes. Significant improvements were observed in ODI and VAS scores along with

S765 intervertebral disc and neuroforaminal height restoration from preoperative levels to all postoperative time points. Disclosures: author 1: grants/research support; Company=Globus Medical, consultant; Company=Globus Medical, royalties; Company=Globus Medical; author 2: royalties; Company=Globus Medical; author 3: none; author 4: none; author 5: employee; Company=Globus Medical; author 6: stock/shareholder; Company=Globus Medical Inc., employee; Company=Globus Medical Inc.; author 7: stock/shareholder; Company=Globus Medical inc, employee; Company=Globus Medical Inc.

P56 IDENTIFYING THORACIC COMPENSATION AND PREDICTING RECIPROCAL THORACIC KYPHOSIS AND PJK Renaud Lafage, Themistocles Protopsaltis, Bassel Diebo, Justin Smith, Eric Klineberg, Douglas Burton, Christopher Ames, Shay Bess, Frank Schwab, International Spine Study Group International Spine Study Group, New York, USA Hypothesis: Patients with more baseline thoracic compensation have more post-operative reciprocal thoracic kyphosis and PJK. Design: Retrospective analysis. Introduction: Adult spinal deformity requires patients to recruit compensatory mechanisms like pelvic retroversion and knee flexion. However, thoracic hypokyphosis is a less recognized and poorly defined compensation. Methods: 219 patients undergoing TLD correction were included with fusions to the pelvis and UIV of T9-L1. Patients were divided into those with post-operative reciprocal thoracic kyphosis (RK: D unfused TK C15°) and those who maintained thoracic alignment (MT). Thoracic compensation was defined as theoretical thoracic kyphosis (tTK) minus preop TK. The tTK was calculated from LL = 2(PI + TK) + 10 where LL = PI + 10 for PI B 40°, LL = PI for PI 40°-70° and LL = PI-10 for PI C 70°. Results: For RK (n = 117), the mean change in unfused TK was 21.7° and the mean PJK was 17.6° vs 6.1° and 5.7° (p \ 0.001) for MT (n = 102). RK and MT were similar in age, BMI, gender, and comorbidities. RK had larger preop PI-LL mismatch (30.7 vs 23.6 p = 0.008) and less pre-operative TK (22.3 vs 30.6 p \ 0.001), otherwise SVA, PT and TPA were similar. RK patients had more PILL correction (29.8 vs 17.3, p \ 0.001) and more pre-operative thoracic compensation (29.9 vs 20.0, p \ 0.001). There were no differences in pre-operative HRQOL except RK had worse SRS appearance (2.2 vs 2.5, p = 0.005). Using a logistic regression model, the only predictor for RK was more thoracic compensation. Postoperatively the RK and MT groups were well aligned by all SRS Schwab modifiers. The RK group had 76 patients with PJK and 39 without. There were no differences between these patients in thoracic compensation, PI-LL correction, or pre/post-operative alignment except PT. HRQOL were not different for any group at 6 wk and 1 y. Both younger RK and older ([65 y) had larger thoracic compensation and more correction than MT young and old. The tTK was similar to the postop TK for all groups. Conclusions: Postoperative reciprocal thoracic kyphosis can be anticipated and incorporated into pre-operative planning of thoracolumbar deformity correction by calculating tTK. Reciprocal thoracic kyphosis was predicted by the magnitude of preoperative thoracic compensation. Disclosures: author 1: none; author 2: grants/research support; Company=Zimmer, consultant; Company=Medicrea, Biomet,

123

S766 AlphaSpine; author 3: none; author 4: grants/research support; Company=Depuy-Synthes, ISSGF, consultant; Company=Biomet, Nuvasive, Cerapedics, Medtronic, royalties; Company=Biomet; author 5: grants/research support; Company=AO Spine, Depuy, OREF, other financial report; Company=Speakers Fees Depuy AO Spine; author 6: grants/research support; Company=depuy, consultant; Company=depuy, royalties; Company=depuy; author 7: grants/ research support; Company=depuy synthes spine, consultant; Company=stryker medtronic depuy synthes, stock/shareholder; Company=Baxano Surgical, Doctor’s Research Group, royalties; Company=stryker biomet, Aesculap, employee; Company=ucsf; author 8: grants/research support; Company=depuy, medtronic, k2, innovasis, stryker, consultant; Company=k2, allosource, nuvasive, royalties; Company=pioneer; author 9: none; author 10: grants/research support; Company=J&J, MSD, Inovasis, Stryker, Biomet.

P57 AFTER NINE YEARS OF THREE-COLUMN OSTEOTOMIES, ARE WE DOING BETTER? PERFORMANCE CURVE ANALYSIS OF 573 SURGERIES WITH 2-YEAR FOLLOW UP Frank Schwab, Bassel Diebo, Virginie Lafage, Munish Gupta, Christopher Ames, Robert Hart, Richard Hostin, Ibrahim Obeid, International Spine Study Group NYU Hospital for Joint Diseases, New York, USA Hypothesis: Performance of 3CO surgeries improves over years of practice. Design: Retrospective review. Introduction: In spinal deformity treatment, increased utilization of three-column (3CO) osteotomies reflects a greater comfort and better training. This study aims to evaluate performance curve and adverse events for multi-center group following a decade of 3CO. Methods: Inclusion criteria were patients underwent 3CO for spinal deformity with intra/post-operative and revision data collected up to 2-year only. Patients were categorized evenly into four groups and reviewed retrospectively. Demographics, baseline deformity/correction, and surgical metrics were compared using Student T-Test. Postoperative and revision rates were compared using Chi-Square analysis. Results: 573 patients stratified to: G1 (N = 143, Feb 2004-Apr 2008), G2 (N = 142, Apr 2008- Sep 2009), G3 (N = 144, Sep 2009-Nov 2010), G4 (N = 144 Nov 2010-Jan 2013). Demographics, primary/revision, baseline radiographic parameters, and 3CO levels were similar. Most recent patients were significantly more disabled by ODI (G4 = 49.2 vs. G1 = 38.3, p = 0.001), and received a larger resection (G4 = 26° vs. G1 = 20°, p = 0.011). There was significant decrease in revision rate (45 %, 35 %, 33 %, 30 %, p = 0.039), notably revisions for pseudoarthrosis (16.7 % G1 vs. 6.9 % G4, p = 0.007). Major complications rate significantly decreased (56.6 %, 50.3 %, 45.8 %, 38.3 %, p = 0.023), as did Excessive blood loss ([4 liters, 27.2 vs. 16.6 % p = 0.023), and bladder/bowel deficit (4.2 % vs. 0.7 % p = 0.002). Successful rate (no complications or revision) significantly increased (25.2 %, 34.5 % 40.3 %, 48.6 % p\0.001). OR time significantly decreased (441, 431, 445, 397 mins, p = 0.024). Intra-operative complications were similar. Conclusions: After 9 years, the surgeries performed are operating on a more disabled population with better technical ability reflected by increased bone resection angle. These surgeries are performed diminishing OR time by 48 minutes and reducing revisions and complications rate by 15 % and 18.3 respectively.

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Disclosures: author 1: none; author 2: none; author 3: grants/research support; Company=Depuy Synthes Spine, Medtronic, SRS, stock/ shareholder; Company=Nemaris Inc, other financial report; Company=J&J, MSD, Nuvasive, Medicrea; author 4: grants/research support; Company=Depuy medicrea, consultant; Company=Orthofix depuy, royalties; Company=Depuy; author 5: grants/research support; Company=depuy synthes, consultant; Company=depuy medtronic stryker, stock/shareholder; Company=Baxano Surgical, Doctor’s Research Group, royalties; Company=biomet, stryker, employee; Company=ucsf; author 6: grants/research support; Company=Medtronic, ISSGF, consultant; Company=DepuySynthes, Globus, Medtronic, stock/shareholder; Company=Spine Connect, royalties; Company=Seaspine, DepuySynthes, other financial report; Company=DepuySynthes; author 7: none; author 8: grants/research support; Company=Depuy Synthes, consultant; Company=Depuy Synthes Medtronic; author 9: grants/research support; Company=J&J, MSD, Inovasis, Stryker, Biomet.

P58 UNLOCKING TPA’S CLINICAL AND SAGITTAL SIGNIFICANCE BY ANALYZING ITS RELATION TO PELVIC TILT Virginie Lafage, Renaud Lafage, Jonathan Oren, Shaleen Vira, Bradley Harris, Matthew Spiegel, Bassel Diebo, Themistocles Protopsaltis, Thomas Errico, Frank Schwab NYU Hospital for Joint Diseases, New York, USA Hypothesis: A single T1 Spino Pelvic Angle (TPA) can be associated with drastically different HRQOL but it can easily be supplemented to convey both global alignment and meaningful clinical outcome. Design: Retrospective cohort. Introduction: TPA is a valuable perioperative planning tool that accounts for both pelvic tilt (PT) and trunk inclination. However, it is limited as a standalone parameter because it does not distinguish patients’ ability to compensate with pelvic retroversion. For a given TPA, patients who are unable to recruit compensatory mechanisms may have significantly worse HRQOL scores than those who can. Can TPA be augmented to better describe a patient’s global alignment and more accurately predict HRQOL? Methods: Single-center study of patients with full body X-ray, HRQOL and TPA C10°. Proportions of PT to TPA (PTp = PT/TPA) and T1SPi to TPA (T1SPip = T1SPi/TPA) were calculated and investigated against increased values of TPA. Then, 2 sub-groups were created (HighPT and LowPT) based on mean (PTp) ± 0.5 standard deviation. HighPT and LowPT were compared across the entire cohort using an unpaired T-test. Results: 230 patients were included (58.7 ± 15.5 y, 60 %F). Mean sagittal parameters included: PI-LL 12.3 ± 16.3°, SVA 41 ± 49 mm, TPA 21.9 ± 10.1° and PT 24.4 ± 8.6°. The analysis of PTp distribution revealed a decrease in PT recruitment as TPA increases (137 ± 39 % for patients with TPA \15°, 87 ± 15 % for patients with TPA[40°). Comparing LowPT (n = 57) with HighPT (n = 69) revealed that for a similar TPA (24.1 vs. 22.1°, p = 0.308), patients with LowPT (and therefore little compensatory PT) had significantly worse HRQL scores in terms of ODI (45 vs. 32 in HighPT; p = 0.002) and EQ-5D (9.7 vs. 8.5 in HighPT, p = 0.003). Conclusions: While TPA captures the severity of deformity, disability is a product of deformity severity and the inability to recruit compensatory mechanisms. TPA measures the severity of the thoracolumbar deformity separate from pelvic compensation. Therefore for a complete picture of standing sagittal alignment, TPA should be considered in

Eur Spine J (2015) 24 (Suppl 6):S743–S800 conjunction with PT to convey the full radiological and clinical picture. Failing to do so potentially masks a patient’s disability. Disclosures: author 1: grants/research support; Company=Depuy Synthes Spine, Medtronic, SRS, stock/shareholder; Company=Nemaris Inc, other financial report; Company=J&J, MSD, Nuvasive, Medicrea; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: grants/research support; Company=Zimmer, consultant; Company=Medicrea, Biomet, AlphaSpine; author 9: grants/research support; Company=OMEGA, Fridolin Trust, Paradigm Spine, royalties; Company=Fastenetix, K2M, other financial report; Company=K2M; author 10: stock/ shareholder; Company=Nemaris Inc.

P59 CHAIN OF RELAXATION: HOW SAGITTAL CORRECTION AFFECTS SPINO-PELVIC, LOWER LIMB, AND GLOBAL ALIGNMENT PARAMETERS Frank Schwab, Jonathan Oren, Shaleen Vira, Barthelemy Liabaud, Bassel Diebo, Elizabeth Tanzi, Matthew Spiegel, Renaud Lafage, Jensen Henry, Virginie Lafage NYU Hospital for Joint Diseases, New York, USA Hypothesis: Mechanisms of relaxation after sagittal correction may differ from the compensation mechanisms of sagittal deformity development. Design: Retrospective cohort. Introduction: Adult spinal deformity patients recruit compensatory mechanisms to maintain alignment with increasing deformity. While attention has been paid to pre-operative compensatory adaptations, little is known regarding the sequence of relaxation of these parameters based upon the amount of residual PI-LL mismatch. This study details the progression of compensatory mechanism relaxation as PILL mismatch improves. Methods: Single site review of post-operative full-body x-rays of patients, at least 9 months after surgery. Radiographic measurements were obtained with dedicated spine software and included PT, knee flexion (KA), ankle dorsiflexion (AD), pelvic shift (PShift), T1 spinopelvic inclination (T1SPi), SVA and T1 pelvic angle (TPA). Patients were stratified by their remaining need for lordosis based on ageadjusted normative published values. Group comparisons were carried out via ANOVA analysis. Results: 262 patients were included, mean age 61.5, mean BMI 28, and males 31 %. PI-LL groups were significantly (p \ 0.05) different in terms of PT, KA, PShift, TPA, SVA and AD (Fig). Analysis of the sequence of correction revealed that the majority of age-adjusted offset in terms of SVA and PT was corrected (59 %, 57 %) during the first stage of correction ([[30°] to [20°-30°] of remaining need for lordosis). This stage was associated with significant changes in PShift and T1SPi (p \ 0.05). Additional stages of corrections were associated with gradual decreases in SVA, while no additional decreases in PT were observed until the post-operative lordosis was near ideal. Conclusions: When improving from severe to moderate age-adjusted PI-LL mismatch, there is a correction in age-adjusted SVA to normal range. Nevertheless, these patients still exhibit high degree of compensation with respect to the pelvis and lower limbs to maintain alignment. PT and SVA can be corrected to an age-adjusted ideal when PI-LL is also corrected. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: none; author 9: none; author 10: grants/research support; Company=Depuy Synthes Spine, Medtronic, SRS, stock/shareholder; Company=Nemaris Inc, other financial report; Company=J&J, MSD, Nuvasive, Medicrea.

S767 P60 NORMATIVE VALUES FOR THE L5 INCIDENCE ANGLE AND ITS CORRELATION WITH OTHER SPINO-PELVIC PARAMETERS: A DATABASE OF 147 ASYMPTOMATIC SUBJECTS Jean-Charles LeHuec, Dennis Dominguez, Arnaud Cogniet, Antonio Faundez 1

Spine Unit 2, Bordeaux University Hospital, CHU Pellegrin 33076 Bordeaux, France; 2Division of Orthopaedics and Trauma Surgery, Geneva University Hospitals, Geneva, Switzerland

Introduction: Sagittal balance analysis has become a mandatory step before any therapeutic decision related to lombo-spinal pathologies. The determination of pelvic parameters such as Pelvic Incidence (PI) is paramount. In case of lumbosacral transitional vertebrae, or certain forms of spondylolisthesis with dome shaped sacrum, classical measurement of PI is not reliable. Upper endplate of L5 is less subject to anomalies, but no normative values for pelvic and spinal parameters based on L5I have been described yet. The purpose of this study was to provide normative data of the L5 incidence angle (L5I) and their correlation with standard spino-pelvic parameters. Materials and methods: The protocol was approved by the local ethics committee. 147 asymptomatic volunteers underwent an EOS low-dose full spine Xray (EOS Imaging, Paris, France). SterEOS software was used to obtain 3D full spine reconstructions. From the 3D reconstructions, L5I was calculated similarly as the standard Pelvic Incidence (PI), but using the perpendicular line of the upper L5 endplate instead of the S1 endplate. We calculated L5I, L5 tilt, L5 slope and L5 % of PI for the 147 subjects (group A). We estimated statistical correlations between: L5I and Pelvic Incidence, and L5I with L1-L5 lordosis. For a subgroup of these subjects we only calculated L5I, L5 tilt and L5 slope (20/147 subjects, subgroup B). For these patients, the PI cannot be measured. ODI and VAS were calculated. Results: For group A, L5I, L5 tilt, L5 slope and L5 % of PI mean values were respectively 22.43, 4.65, 17.73 and 40.72. For group A, Pearson correlation between L5I and pelvic incidence was 0.83, L5I and lordosis L1-L5 was 0.62. A mathematical relationship between L5I and lordosis L1-L5 was obtained: Lordosis L1-L5 = 0.67 * L5I + 30.7 (p \ 0.05). ODI and VAS showed that patients are asymptomatic. For subgroup B, L5I, L5 tilt and L5 slope mean values were respectively 29.1, 7.0, and 22.3. Discussion and conclusion: This study is the first to provide functional status for an asymptomatic population as well as spino-pelvic parameters. We propose L5I as a new spino-pelvic parameter. L5I is highly correlated with S1 incidence in a normal population and it can be always calculated even in cases when PI cannot be calculated. The normative values provided in this study will help to control/reestablish the lordosis restoration for patients with lombo-sacral variations where L5S1 is anatomically not mobile due to L5 sacralisation. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P61 THE ROOTOGRAM TO SUCCESS? Kiran Lingutla, Suribabu Gudipati, Michael McCarthy University Hospital Of Wales, Cardiff, UK Summary: The current study has observed that about 60 % of patients experienced good pain relief after a successful rootogram.

123

S768 Optimal needle position, visualization of medial epidural spill of the dye and good quality imaging have been identified to be positive prognostic factors associated with better pain relief. Hypothesis: Rootogram has a no effect on pain relief in patients with lumbar radiculopathy Design: Prospective observational cohort study of consecutive patients over a 2-year period at a regional tertiary spinal unit performed by a single surgeon. Introduction: Nerve root blocks are widely used as a non-operative treatment option for both diagnostic and therapeutic purposes to treat radicular leg symptoms. It is widely reported in literature that patients with a successful rootogram have superior clinical outcomes. Methods: Patients with incomplete data, absence of images on system, missed follow up appointments and those moved out of the catchment area were excluded. An independent radiological analysis for the position of the needle, adequacy of the rootogram by the path of the dye and medial pedicle spill was performed. Both pre-operative and post-operative visual analogue scores for pain and Oswestry disability index scores were used as outcome measures. Results: 106 patients were eligible with equal distribution of males and females with a mean age of 53.5 years. Needle position was good in 100 % whereas the image quality was deemed poor in 16 %. There was a clear rootogram in 57 % and medial spill in 50 %. Successful pain relief was achieved in 60 % but pain recurred in 1/3rd of these patients. Surgery was subsequently performed in 30 % of the cohort. Only epidural spill was associated with successful pain relief (p = 0.02). Lack of rootogram (p = 0.02), lack of epidural spill (p = 0.03) or lack of both (p = 0.006) was associated with subsequent surgery. Conclusion: Medial epidural spill at the time of rootogram during nerve root block appears to be associated with successful pain relief. Disclosures: author 1: none; author 2: none; author 3: consultant; Company=Globus medical education.

P62 CLINICAL OUTCOME IN LUMBAR DECOMPRESSION SURGERY FOR SPINAL CANAL STENOSIS IN THE AGED POPULATION Frank S. Kleinstu¨ck, Nils Ulrich, Christoph Woernle, Sebastian Winkelhofer, Jakob Burgstaller, Mazda Farshad, J. Oberle, Francois Porchet, Kan Min, LSOS group Department of Orthopedics, University Hospital Balgrist, University of Zurich, Switzerland; Department of Orthopedics and Neurosurgery, Spine Center, Schulthess Clinic, Zurich, Switzerland; Department of Radiology, University Hospital Balgrist, University of Zurich, Switzerland; Horten Centre for Patient Oriented Research and Knowledge Transfer, University of Zurich, Switzerland; Department of Neurosurgery, Kantonsspital Winterthur, Switzerland; and On behalf of the LumbSten Research Collaboration, Zurich, Switzerland Introduction: Lumbar decompression surgery without fusion has been shown to improve quality of life in degenerative lumbar spinal stenosis (LSS). In the population older than 80 years, treatment recommendations for LSS show conflicting results. We aimed to examine whether outcome and quality of life improved after decompression surgery for LSS even in patients older than 80 years and to compare data with a younger patient population from the same patient collective. Methods: Eight centers in the metropolitan area of Zurich, Switzerland agreed on the classification of LSS, surgical principles, and

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 follow-up protocols. Patients were evaluated at baseline, and at 6 and 12 months postoperatively using 5 different instruments Swiss Spinal Stenosis Measure, Feeling Thermometer, Numeric Rating Scale, EuroQoL 5D-3L, and Roland and Morris Disability Questionnaire. Results: Thirty-seven patients with an average age of 82.5 ± 2.5 years reached the 12-month follow-up. Spinal Stenosis Measure scores, the Feeling Thermometer, the Numeric Rating Scale, and the Roland and Morris Disability Questionnaire showed significant improvements at the 6-month and 12-month follow-ups (P \ 0.001). One EQ-5D-3Lsubgroup ‘‘anxiety/depression’’ showed no significant improvement (P = 0.109) at the 12-month follow-up. The minimal clinically important change score for the ‘‘Symptom Severity scale’’ in the Spinal Stenosis Measure was achieved by 70 % of the older patient population. Conclusion: Patients 80 years or older can expect a clinically meaningful improvement after lumbar decompression for symptomatic DLSS. Our patient population showed significant improvements in quality of life in short- and long-term follow-ups. Disclosures: author 1: none; author 2: none; author 3: none; author 4: no indication; author 5: none; author 6: none; author 7: none; author 8: none; author 9: none.

P63 THE EFFECT OF EPIDURAL STEROID INJECTION ON POSTOPERATIVE OUTCOME IN PATIENTS FROM THE LUMBAR SPINAL STENOSIS OUTCOME STUDY Tamas F. Fekete, Christoph Woernle, Anne F. Mannion, Ulrike Held, Frank S. Kleinstu¨ck, Nils Ulrich, Daniel Haschtmann, Hans-Ju¨rgen Becker, Francois Porchet, Robert Theiler, Johann Steurer, LSOS group Schulthess Klinik, Zu¨rich, Switzerland; Balgrist University Hospital, Department of Orthopedics, University of Zurich, Switzerland; Horten Centre, University Zurich, Zu¨rich, Switzerland; Triemli City Hospital, Zurich, Switzerland Introduction: In patients with lumbar spinal stenosis the use of epidural steroid injections (ESI) is very common. ESIs reduce symptoms in the short-term, but according to a subgroup analysis from the Spine Patient Outcomes Research Trial (SPORT) they seem to reduce the degree of improvement after subsequent surgical or nonoperative treatment. This was a retrospective analysis of prospectively collected data from 281 patients participating in the Lumbar Spinal Stenosis Outcome Study (LSOS), a prospective cohort study. The aim of this study was to assess clinical outcomes after surgical or nonoperative treatment in patients with and without prior epidural steroid injections. Methods: The data of 281 patients with lumbar spinal stenosis who had completed baseline and 6-month follow-up assessments were analyzed. Patients completed the Spinal Stenosis Measure (SSM), The Roland Morris Disability Questionnaire, pain intensity (numeric rating scale NRS) and EuroQoL five dimensions (EQ-5D). Changes in the SSM scores from baseline to 6 months’ follow-up were compared between patients with and without ESI prior to enrolment in the study, for both the surgical and non-surgical treatment groups separately. Results: The mean (SD) age of the patients was 75.0 (8.7) years. In total, 229 patients underwent surgery and 111 of these had received an ESI in the 12 months before surgery. Of the 52 patients treated nonoperatively, 29 had received an epidural steroid injection in the 12 months before enrolment. The unadjusted changes (improvement) in the SSM-symptom scores between baseline and 6 months’ follow up

Eur Spine J (2015) 24 (Suppl 6):S743–S800 were: surgery and prior ESI 0.95, surgery and no prior ESI 0.78 (p = 0.15); no surgery and prior ESI 0.28, no surgery and no prior ESI 0.29 (p = 0.85). When adjusted for confounding factors, the reduction in SSM-symptom score at 6 months was greater for surgery than for non-operative treatment by 0.41 points (p \ 0.001); the effect of having had an ESI prior to study entry was -0.08 (p = 0.40). Conclusions: The analysis of outcomes in the LSOS cohort provided no evidence that ESIs have a negative effect on the short-term outcome of surgery or non-operative treatment in patients with lumbar spinal stenosis. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: none; author 9: none.

P64 CAN SPINOPELVIC PARAMETERS PREDICT PROXIMAL JUNCTION KYPHOSIS FOLLOWING CORRECTION OF SCHEUERMANN’S KYPHOSIS? Luigi Aurelio Nasto, Ana Belen Perez Romera, Eyal Behrbalk, Saggah Tarek Shalabi, Hossein Mehdian The Centre for Spinal Studies and Surgery, Queen’s Medical Centre University Hospital, Nottingham, UK Background context: Surgical correction of SK is complicated by a relatively high incidence of early (22 %) and late (53 %) PJK. Recent studies show that SK patients have smaller PI values than age-matched controls. This study aims to evaluate if pre-operative PI can be used to predict PJK. Study design: Retrospective analysis of spinopelvic parameters in patients treated for Scheuermann’s kyphosis. Objectives: This study evaluates the influence of pelvic incidence (PI) on the prediction of PJK following Scheuermann’s kyphosis (SK) correction. SK patients with larger preoperative PI and pelvic tilt (PT) may be at increased risk of developing PJK due to potential postoperative mismatch between lumbar lordosis (LL) and PI. Our findings suggest that a smaller degree of thoracic correction may reduce the likelihood of development of PJK in patients with a larger PI. Methods: This is a retrospective analysis of 35 patients, with an average age of 20 (15-26), who were treated surgically for SK. Radiological variables were assessed in order to predict the possibility of implant failure. The average followup was 7.4 (2-12 years). Radiographic measurements were carried out before and after surgery (2 months post operatively), and at final follow up. Results: 23 % of the patients developed PJK (average 7.4 years). There was no significant difference in pre-operative thoracic kyphosis between the two compared groups (failure vs non-failure, 73° vs 74°). Preoperative PI and PT were found to be significantly higher in PJK group (51° vs 45° and 18.5° vs 11° respectively; p \ 0.05). Postoperatively, lumbar lordosis (LL) was smaller for PJK group (48.7° vs 59.3°; p \ 0.05). In addition, we found that post-operative PT values in failure vs non-failure group were significantly higher (21.7° vs 13.5°). Conclusions: This study shows the importance of the mismatch between PI/PT and LL in predicting the possibility of development of PJK following SK correction. A high degree of thoracic kyphosis correction can lead to a greater reduction of LL and increased PT. Our experience suggests that less correction of thoracic kyphosis in patients with higher preoperative PI values (C45°) should be attempted and this may reduce the risk of developing PJK. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

S769 P65 TWO-YEAR CLINICAL RESULTS WITH AN INVESTIGATIONAL LUMBAR ZYGAPOPHYSIAL JOINT RESTORATION DEVICE Hans-Jo¨rg Meisel, Konrad Seller, Achim Lu¨th, Karin Bu¨ttner-Janz, Peter Stosberg, Alexander Mo¨ser, Luiz Pimenta BG Clinic Bergmannstrost, Department of Neurosurgery, Halle/Saale, Germany Background: The zygapophysial (facet) joint is the primary pain generator in one third of chronic low back pain cases. Current treatment options include conservative, nonsurgical therapy, facet joint infiltrations, facet denervation, and, rarely, lumbar arthrodesis. The purpose of this study was to assess the safety and effectiveness of a novel minimally invasive implant intended to resurface the facet joint and restore facet joint function in patients with chronic lumbar facetogenic pain. Methods: This prospective, multi-center feasibility study enrolled patients with confirmed lumbar facetogenic joint pain at 1 or 2 levels who underwent at least 6 months of non-operative care. Patients received a minimally invasive implant intended to resurface the facet joint, restoring facet joint function while preserving the native anatomy. Primary outcomes were measured as a decrease in back pain severity of at least 20/100 points using a visual analog scale, and decrease of at least 15 percentage points using the Oswestry Disability Index (ODI). Adverse events were adjudicated by an independent Clinical Events Committee. Results: Of 40 enrolled patients, 37/38 (97.4 %) patients had a successfully attempted implant of the facet restoration device. Twentyeight (28) patients completed 2 years of follow-up. At 2 years, back pain severity decreased 38 % (p = 0.002) on average with 60.0 % of patients reporting at least a 20/100 mm decrease in lumbar pain. ODI decreased 35.6 % (p = 0.0003) from baseline, and 64.0 % of patients experienced a decrease in disability of at least 15 percentage points at 2 years. Freedom from a device- or procedure-related serious adverse event through 2 years was 86.2 %. Implant migration was observed in 3 patients and implant expulsion from the facet joint occurred in 3 patients. In total, 3 (8.1 %) patients underwent removal of the implant through 2 years post-treatment. Conclusions: A minimally invasive facet restoration implant is a promising treatment option in select patients with chronic lumbar zygapophysial pain who continue with pain despite nonsurgical treatments. The device can be safely implanted and post procedure migration and expulsion were addressed with modifications of the device. Therapeutic benefit of this device persists 2 years after implantation. Disclosures: author 1: consultant; Company=Zyga; author 2: none; author 6: grants/research support; Company=zyga; author 7: no indication.

P66 THE RISK OF GETTING WORSE: PREDICTORS OF DETERIORATION AFTER DECOMPRESSIVE SURGERY FOR LUMBAR SPINAL STENOSIS - A MULTICENTER OBSERVATIONAL STUDY Ulf Skule Nerland, Asgeir Jakola, Charalampis Giannadakis, Ole Solheim, Clemens Weber, Øystein Petter Nygaard, Sasha Gulati Department of Neurosurgery, St.Olavs Hospital, Trondheim - Norway

123

S770 Objective: To investigate the frequency and predictors of deterioration following decompressive surgery for single and two-level lumbar spinal stenosis. Material and methods: Prospectively collected data were retrieved from the Norwegian Registry for Spine Surgery. Clinically significant deterioration was defined as an 8-point increase in Oswestry disability index (ODI) between baseline and 12 months follow-up. Results: There were 2181 patients enrolled in the study. Out of 1735 patients with complete 12-months follow-up 151 (8.7 %) patients reported deterioration. The following variables were significantly associated with deterioration at 12-months follow-up; decreasing age (OR 1.02, 95 % CI 1.00-1.04, p = 0.046), tobacco smoking (OR 2.10, 95 % CI 1.42-3.22, p = 0.000), ASA grade C3 (OR 1.80, 95 % CI 1.07-2.94, p = 0.025), decreasing preoperative ODI (OR 1.05, 95 % CI 1.02-1.07, p = 0.000), previous surgery in the same level (OR 2.00, 95 % CI 1.18-3.27, p = 0.009), and previous surgery in other lumbar level(s) (OR 2.10, 95 % CI 1.19-3.53, p = 0.009). Conclusion: Overall risk of clinically significant deterioration in patient-reported pain and disability following decompressive surgery for lumbar spinal stenosis is about 9 %. Predictors for deterioration are decreasing age, current tobacco smoking, ASA grade C 3, decreasing preoperative ODI, and previous surgery in same or different lumbar level. We suggest that these predictors should be emphasized and discussed with the patients prior to surgery. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

P67 ABILITY OF THE SRS-SCHWAB ADULT SPINAL DEFORMITY CLASSIFICATION TO IDENTIFY PATIENTS WITH SEVERE DISABILITY Dennis Hallager Nielsen, Lars Valentin Hansen, Casper Rokkjær Dragsted, Nina Caroline Peytz, Martin Gerhchen, Benny Dahl Spine Unit, Department of Orthopaedic Surgery, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Introduction: The Scoliosis Research Society (SRS)-Schwab Adult Spinal Deformity Classification includes sagittal modifiers considered important for health related quality of life (HRQOL): Pelvic Tilt (PT), Pelvic Incidence minus Lumbar Lordosis (PI-LL) and Sagittal Vertical Axis (SVA). Each modifier is graded 0, + or ++ with increasing abnormality. The cut offs dividing patients with normal from abnormal sagittal modifiers have been determined based on the ability to predict ODI scores of 40 with linear regressions. The precision of the classification to identify patients with severe disability determined by an ODI score of at least 40 has not previously been evaluated. Objective: To evaluate the accuracy of the SRS-Schwab Classification modifiers regarding identification of patients with severe disability defined as an ODI score of at least 40. Methods: Between March 2013 and May 2014 patients at least 18 years of age with sufficient radiographic images taken at one outpatient clinic to evaluate a spinal deformity were prospectively enrolled. Exclusion criteria were deformity surgery within 6 months or missing ODI questionnaires. Patients were classified as having normal or abnormal sagittal modifiers based on the SRS-Schwab Classification cut points: 20° for PT, 4 cm for SVA and 10° for PI-LL. Observers were unaware of patient scores. Diagnostic statistics were calculated with ODI of at least 40 as a dichotomized measure of severe disability. Results: A total of 460 patients were eligible and 286 were included. 30 had deformity surgery within 6 months and 144 did not return the

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 questionnaire. 123 (43 %) had abnormal PI-LL, 156 (55 %) had abnormal PT and 148 (52 %) had abnormal SVA. The study population prevalence of severe disability was 45 %. Abnormal PT, SVA or PI-LL identified severe disability with a sensitivity of 57 %, 65 % and 65 % respectively; specificity was 54 %, 59 % and 69 % respectively. Positive predictive values (PPV) were 54 % for PT, 57 % for SVA and 60 % for PI-LL. Negative predictive values (NPV) were 65 % for PT, 67 % for SVA and 66 % for PI-LL. Having at least one abnormal modifier predicted severe disability with a sensitivity of 80 %, specificity of 40 %, PPV of 53 % and NPV of 70 %. Conclusion: Precision for each of the SRS-Schwab Classification modifiers to identify patients with severe disability determined by the ODI score was considered weak, and moderate when the modifiers were combined. Disclosures: author 1: grants/research support; Company=Globus Medical Inc.; author 2: none; author 3: none; author 4: none; author 5: grants/research support; Company=, consultant; Company=; author 6: grants/research support; Company=Medtronic, Globus Medical, K2M.

P68 RISK FACTORS AND CLINICAL IMPACT OF EARLY UNANTICIPATED REVISION SURGERY IN ADULT SPINAL DEFORMITY Susana Nu´n˜ez Pereira, Ferran Pellise´, Alba Vila Casademunt, Montserrat Domingo Sabat, Emre Acaroglu, Ahmet Alanay, Francisco Sa´nchez Pe´rez-Grueso, Frank Kleinstu¨ck, Juan Bago´, European Spine Study Group ESSG Vall d’Hebron Research Institute, Barcelona, Spain Introduction: Risk factors associated with revision surgery in adult spinal deformity (ASD) and its impact on patient outcomes are still poorly understood. This study aims to assess potential risk factors associated with first-year unanticipated revision surgery in ASD and to evaluate its impact in HRQOL outcome scores. Methods: Retrospective analysis of prospectively collected data from consecutive surgically treated patients included in a multicenter, international ASD database. All surgical patients with more than 1 year of follow-up were stratified in 2 groups: R (revision surgery within the first year) and NR (no revision). Demographic, surgical, radiographic and HRQOL data were assessed. Statistical analysis included chi-square and Student t-test. Results: 301 patients met inclusion criteria, 46 were included in group R. Reasons for revision were: 18 (54.2 %) implant failure, 13 (28.6 %) infection and wound problems, 8 (16.7 %) adjacent problems and 5 (10.4 %) neurological complications. 43.5 % of revisions occurred within the first month, 67.4 % within the first three months. R group patients were older (58.1 vs 47.7, p \ 0.001), had higher BMI (27.5 vs 24.7, p \ 0.001), higher ASA Score (p = 0.002), and worse sagittal alignment (SVA 73.1 mm vs 26.1 mm, p \ 0.001; PT 26.8° vs 20.1°, p \ 0.001, Global Tilt 35.4 vs 22.9). R group patients were more likely to undergo decompression (46.8 % vs 27.6 %; p = 0.004), osteotomies (62.9 % vs 44.4 %, p = 0.009) and pelvic fixation (67.7 % vs 33.5 %, p\0.001). Preoperative HRQOL scores in R patients were worse than in NR (ODI p \ 0.001, back pain; p = 0.001, leg pain; p\0.001, SF-36 physical component; p = 0.002, and SRS22; p = 0.002). At 6 months follow-up mental health in R group did deteriorate (SF-36 MCS, p = 0.004 and SRS 22 MH, p = 0.007) but was restored at 1 year. The same improvement in all other HRQOL dimensions was observed in both groups during the first year of follow-up.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Conclusion: Early unanticipated revision surgery has a negative impact on mental health at 6 months, but no clear impact on function and disability improvement at 1 year follow-up. Delaying surgery over time might be associated with deformity progression, higher chances of early unanticipated revision surgery and a worse clinical outcome. Disclosures: author 1: grants/research support; Company=Depuy Synthes; author 2: grants/research support; Company=dePuy Synthes, K2M, consultant; Company=dePuy synthes, Biomet; author 3: grants/ research support; Company=Depuy Synthes; author 4: grants/research support; Company=Depuy-synthes; author 5: grants/research support; Company=Medtronic, Depuy Synthes, stock/shareholder; Company=IncredX; author 6: grants/research support; Company=DePuy, consultant; Company=Stryker; author 7: grants/research support; Company=DePuy Synthes, consultant; Company=DePuy Synthes; author 8: none; author 9: grants/research support; Company=Depuy Synthes; author 10: grants/research support; Company=Depuysynthes.

P69 CAN A CASE MANAGER REDUCE FUNCTIONAL DISABILITY AND ABSENCE FROM WORK FOR LUMBAR SPINAL FUSION PATIENTS? A CLINICAL RANDOMIZED STUDY WITH A TWO YEARS FOLLOW-UP

S771 median reduction of the ODI was -11 (-19;0) in the CM group and -12 (-25;-5) in the control group (NS). After 2 years 15 patients (CM group: 8, control group: 7) had ended on disability pension, and an additional 7 patients (CM group: 3, control group: 4) on old age pension. Of the remaining 61 patients 26 patients in the CM group and 22 patients in the control group had RTW. The median number of sick leave in the CM group was 38 (19; 64) compared to 36 (19; 64) in the control group (NS). The RR of RTW was 1.15 (95 % CI: 0.8; 1.7) favouring the CM group (NS) Conclusion: The addition of a case manager did not seem to have an effect on LSF patients’ functional disability, pain or spell of sick leave. Though the relative cumulative incidence of RTW seemed to favour the CM group, these differences were not statistically significant. Disclosures: author 1: grants/research support; Company=The Danish Rheumatism Association (Gigtforeningen); author 2: none; author 3: grants/research support; Company=Danish Strategic Research Council; author 4: none; author 5: none; author 6: none; author 7: none.

P70 AGE AND PRO-INFLAMMATORY GENE POLYMORPHISMS INFLUENCE ADJACENT SEGMENT DISC DEGENERATION MORE THAN FUSION IN PATIENTS TREATED FOR CHRONIC LOW BACK PAIN

Lisa Gregersen Oestergaard, Finn Bjarke Christensen, Cody Eric Bu¨nger, Randi Holm, Peter Helmig, Rikke Sogaard, Claus Vinther Nielsen

Ahmad Omair, Anne F. Mannion, Gunnar Leivseth, Jeremy Fairbank, Marit Holden, Olle Ha¨gg, Peter Fritzell, Jens I. Brox

1

1

Introduction: The return to work (RTW) rate following a lumbar spinal fusion (LSF) has been reported as low as 35 % looking at 1 year follow-up and between 50 % and 70 % looking at 2-3 years follow-up. Long absence from work has been documented to be a risk factor for exclusion from the work force. Aim: To examine the effect of a case manager on functional disability, pain, and RTW for LSF patients. Material and methods: A clinical randomized study with 2 years follow-up, including a total of 83 LSF patients diagnosed disc degenerations, or spondylolisthesis grade I-II. Primary outcome: Oswestry Disability Index (ODI). Secondary outcome: RTW rates, weeks of sick leave, and back and leg pain. The patients were randomized 1:1 to an intervention group (CM group) with a case manager assigned, or a control group. All patients received the usual physical post-surgical rehabilitation. In the CM group the case manager and the patients met prior to surgery and during the postdischarge period to discuss and plan the patients’ RTW. The medians difference from baseline to 2-years follow up was compared between the groups using the Wilcoxon rank sum test. Median (25; 75 percentile) is reported. The relative cumulative incidence (RR) of RTW was analyzed in a generalized linear regression model adjusting for old age pension and disability pension. Results: The two groups were comparable at baseline. Looking at disability and pain no difference was found between the groups. The

Introduction: Does lumbar fusion lead to accelerated adjacent segment disc degeneration (ASDD) or is it explained by genetics and aging? The influence of genetics on ASDD remains to be explored. This study assesses whether the disc space height adjacent to a fused segment is associated with candidate gene single nucleotide polymorphisms (SNPs). Methods: European patients with low back pain from 4 RCTs (N = 208 had fusion; 77, non-operative treatment) underwent standing plain radiography and genetic analyses at 13 ± 4 years follow-up. Disc space height was measured using a validated computer-assisted distortion compensated roentgen analysis technique and reported in standard deviations from age and gender adjusted normal values. Genetic association analyses included 34 SNPs in 25 structural, inflammatory, matrix degrading, apoptotic, vitamin D receptor and OA-related genes, relevant to disc degeneration. The SNPs were each analysed for their association with disc space height (after adjusting for age, gender, smoking, duration of follow-up and treatment group) and then analysed together in a stepwise multivariable model, with variable entry based on the Akaike information criterion selection procedure. Results: Two SNPs from the IL18RAP gene (rs1420106 and rs917997) were each associated with a lower disc space height at the adjacent level (B = -0.34, p = 0.04 and B = -0.35, p = 0.04, respectively) and the MMP-9 gene SNP rs20544 was associated with a greater disc space height (B = 0.35, p = 0.04); age (p \ 0.001) and

Department of Physiotherapy and Occupational Therapy, Aarhus University Hospital, Noerrebrogade 44 DK-8000 Aarhus C, Denmark; 2 Department of Orthopaedic Surgery Aarhus University Hospital Noerrebrogade 44 DK-8000 Aarhus C, Denmark; 3Region Hopsital of Silkeborg, Orthopaedic department DK-8600 Silkeborg, Denmark; 4 Department of Public Health and Quality Improvement, Central Denmark Region, DK-8200 Aarhus N; 5Department of Social Medicine and Rehabilitation, School of Public Health, Aarhus University DK-8000 Aarhus C, Denmark

Oslo University Hospital Rikshospitalet, Oslo, Norway; 2Spine Center Division, Schulthess Klinik, Zu¨rich, Switzerland; 3 Neuromuscular Diseases Research Group, University of Tromso, Norway; 4Nuffield Orthopaedic Centre, University of Oxford, Oxford, United Kingdom; 5Norwegian Computing Centre, Blindern, Oslo, Norway; 6Spine Center Go¨teborg, Gothenburg, Sweden; 7Neuroortopedic center, La¨nssjukhuset Ryhov, Jo¨nko¨ping, Sweden; 8Oslo University Hospital Ulleval, University of Oslo, Oslo, Norway

123

S772 fusion (p \ 0.008) were also significant variables in the respective analyses. The total explained variance in disc space height was in each case 13-14 %, with 11-12 % of this being accounted for by the given SNP, 64-67 % by age and 19-22 % by fusion. In stepwise multivariable regression analysis (with 9 SNPs selected for entry into the model, along with age, gender, smoking, duration of follow-up and treatment group) the total explained variance in disc space height was 23 %, with the 9 SNPs, age and fusion accounting for 45 %, 45 % and 7 % of this, respectively. Conclusions: Age was the most significant determinant of adjacent segment disc space height followed by genetic factors, specifically inflammatory genes. Fusion explained a statistically significant but small proportion of the total variance. Much of the variance remained to be explained. Other determinants, such as BMI and AGEs, and gene-environment interactions may be important to consider in future studies. Disclosures: author 1: grants/research support; Company=Norwegian Research Council, Hansjorg Wyss Research; author 2: none; author 3: no indication; author 4: none; author 5: no indication; author 6: stock/ shareholder; Company=Spine Centre Go¨teborg AB, employee; Company=Spine Centre Go¨teborg AB; author 7: none; author 8: none.

P71 RISK FACTORS FOR DEVELOPMENT OF ADJACENT SEGMENT DISEASE AFTER MINIMALLY INVASIVE TRANSFORAMINAL INTERBODY FUSION Koichiro Ono, Kazuo Ohmori Center for Spinal Surgery, Nippon Kokan Hospital, Kawasaki, Japan Introduction: Minimally invasive transforaminal interbody fusion using percutaneous pedicle screws (MIS-TLIF, PPS) for lumbar degenerative disease is becoming popular in Japan. As this is a relatively new procedure, there is little information about complications such as adjacent segment disease (ASD). Here we investigated the risk factors for ASD after MIS-TLIF, PPS. Methods: From Feb 2009 to Sept 2013, 82 patients received L4/5 single-level MIS-TLIF, PPS in our facility. Of these 82 patients, eight were excluded from this study; four had additional microendoscopic laminotomy (MEL) on an adjacent level, two had nonunion, one died during follow up, and one was intractable. Hence, 74 patients were retrospectively investigated. Thirty eight male and 36 female, average age 62.4 years old (range 25-80 years), mean follow up period 725.8 days (range 366-1,824 days). Radiographic ASD is defined as development of spondylolisthesis[3 mm, decrease in disc height[3 mm, or intervertebral angle at flexion of less than -5°. Patients were divided into group A (ASD) or N (nonASD), and age, recovery rate of Japanese Orthopaedic Association score, postoperative serum creatine phosphokinase elevation, lumbar lordosis (LL), segmental lordosis angle, disc height, PPS angle, and disc degeneration (Pfirrmann grade) were compared between the two groups. Results: Radiographic ASD was found in five out of 74 cases (6.8 %). Four of these ASD cases were symptomatic, and two cases had additional surgeries (MEL and MIS-TLIF, PPS). Mean LL was significantly decreased in group A (36.8° to 31.8°) compared to group N (40.8° to 41.1°) (p \ .05). Mean postoperative segmental lordosis angle was smaller in group A compared to group N (10° vs 14.2°, p = .062). There was no difference in any other parameter between the two groups. Conclusion: After MIS-TLIF, PPS, patients develop ASD with similar frequency to conventional fusion surgery. Decreased LL and small postoperative segmental lordosis angle were risk factors for

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 developing ASD after MIS-TLIF, PPS, indicating that decent lordosis needs to be gained during MIS-TLIF, PPS. Further research is necessary to elucidate optimal lordosis. Disclosures: author 1: none; author 2: none.

P72 IMPACT OF THE CAGE POSITION ON THE LUMBAR SEGMENTAL LORDOSIS AFTER LATERAL LUMBAR INTERBODY FUSION Bungo Otsuki, Shunsuke Fujibayashi, Hiroaki Kimura, Mitsuru Takemoto, Shuichi Matsuda Graduate School of Medicine, Kyoto University, Kyoto, Japan Object: The precise effect of minimally invasive lateral retroperitoneal lumbar interbody fusion (MI-LLIF) on spinal alignment, especially in the sagittal plane, remains unclear. The aim of this retrospective study was to decide the factors affecting lumbar segmental lordosis in MI-LLIF. Methods: A total of 60 patients with degenerative lumbar disease were included. These patients underwent MI-LLIF with posterior percutaneous pedicle screw fixation without any posterior osteotomy. Computed tomography was performed before and within two weeks after the surgery in all patients, and segmental lordotic angle after surgery (Post-SLA) was predicted using multiple regression analysis. Explanatory factors considered in this study included segmental lordotic angle before surgery (Pre-SLA), disc height before surgery, cage position (CageP; distance between the center of the cage and the center of the disc, where a positive value indicates an anterior cage position), cage angle (CageA), cage height, and level fused. Results: A total of 91 levels were analyzed. Average Post-SLA and the change of the lordotic angle (Post-SLA - Pre-SLA) were not significantly different between 10° cages and 6° cages. Segmental lordosis decreased after surgery in 21 levels, and the CagePs of these levels were significantly smaller (posterior cage position) than that of other levels (-1.22 mm vs. 3.87 mm, respectively; p \0.01). Multiple regression analysis showed that the model using three independent variables, namely CageP, Pre-SLA, and CageA, was the final model, and the adjusted coefficient of determination was 0.734. Of the three independent variables, CageP had the largest impact on the Post-SLA, and as the position of the cage changed anteriorly by 5 mm, Post-SLA increased by 3.2°. Pre-SLA had the second largest effect on the PostSLA, and CageA had the smallest effect. Post-SL could be estimated using the following formula: Post-SL = 0.64 CageP + 0.65 Pre-SL + 0.26 CageA + 1.1. Conclusion: This study revealed that CageP has the largest impact on the Post-SLA, and surgeon should regulate the position of the cage properly to gain appropriate segmental lordosis in MI-LLIF surgery. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

P73 THE FUTURE OF SPINAL FUSION SURGERY - THE VERY OLD PATIENT (80+ YEARS): TECHNIQUES, COMPLICATIONS AND OUTCOME Lukas Panzenboeck, Sebastian Kreuzer, Alexander Tuschel, Philipp Becker, Michael Ogon Speising Orthopedic Hospital Vienna, Austria

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Introduction: The patient population undergoing spinal fusion surgery was increasingly getting older over the past years. This led to a whole new group of patients 80-years and older, with few or none; comorbidities, having high demands regarding postoperative function and quality of life. Aim: Evaluate the outcome and complication rate of patients aged 80 years or older after spinal fusion surgery in the degenerative thoracolumbar spine Patients and methods: Retrospective evaluation of 140 patients, 80 years or older at time of surgery, who underwent thoracolumbar spinal fusion surgery (posterior approach) between 2008-2013. Excluded were cases of revisions, with previous spinal surgery, spondylodiscitis or acute vertebral fractures. Postoperative outcome parameters: lumbar and leg pain (VAS) and Oswestry disability index (ODI) at 6-weeks and 1-year postoperatively. We evaluated types and incidence of complications, surgical details as well as demographic information. Results: The median age at surgery was 82 years, 71 % were women, 29 % men. The median length of instrumentation was 1.5 levels (range 1-8), PLIF was performed in 96 % of all patients. Most commonly performed procedure was mono-segmental PLIF. The overall peri- and postoperative complication rate was 44.3 %. In 16.4 % of the cases revision surgery was necessary (35 % of all cases with complications). The incidence of dural tears was 11 %, infections 20 % (spinal 6 %, non-spinal 14 %), hematoma 12 %, nonsurgical/internal 16 %, prolonged postoperative sciatica 8 %, wound healing disorder 3 %, cauda equina symptoms 2 %, paresis/hypaesthesia 9 %. One patient deceased due to postoperative seizure and cerebral bleeding/edema (0.7 %). Scores for back and leg pain (VAS) improved significantly after 6-weeks as well as one year (\0.05, Wilcoxon-test). Similarly the ODI was significantly improved throughout follow-up (\0.05, t-test). Conclusion: As previously shown in other studies, properly selected older patients are achieving a very good postoperative outcome with significantly improved functional and pain scores. Nevertheless, patients 80-years and older show a higher complication rate compared with younger patients who underwent the same surgery at the same hospital. Especially wound infections, bleeding- and non-surgical complications were far more common with possibly deleterious effects. Disclosures: author 1: other financial report; Company=DePuy Synthes Spine; author 2: none; author 3: no indication; author 4: no indication; author 5: grants/research support; Company=DePuySynthes, Medtronic, consultant; Company=DePuySynthes, AOSpine.

P74 THE INFLUENCE OF OBESITY AND AGE ON OUTCOMES OF MINIMALLY INVASIVE LUMBAR FUSION (MILIF): A SUBGROUP ANALYSIS OF A 1 YEAR PROSPECTIVE MULTICENTER OBSERVATIONAL STUDY Wolfgang Senker, Ulrich Hubbe, Paulo Pereira, Khai Lam, Salvador Fuster, Neil Manson Landesklinikum Mostviertel Amstetten, Austria Objective: This study aims to investigate whether outcomes of MILIF for degenerative lumbar disorders (DLD) are affected by age or weight using data from our multicenter 1-year prospective observational study (NCT01143324). Patients and methods: A total of 252 DLD patients from a multicenter 1 year prospective observational study underwent 1- (83.3 %) or 2-level (16.7 %) MILIF (TLIF: 95.0 %; PLIF: 5.0 %) for treatment

S773 of predominant leg pain (52.0 %), back pain (38.9 %) or claudication (9.1 %), including spondylolisthesis (52.8 %), stenosis (71.4 %), and/ or disc pathology (93.7 %). Outcomes measured: time (days) to first ambulation (TFA) and postsurgical recovery (TPSR), VAS back/leg pain, ODI and EQ5D (baseline, 4 weeks, 6 and 12 months) [all variables: medians back/leg pain, ODI and EQ5D (baseline, 4 weeks, 6 and 12 months) [all variables: medians and (interquartile ranges), Kruskal-Wallis test]. Age (50yrs: N = 102; 51-64 yrs: N = 102; 65 yrs: N = 48) and weight groups (min BMI - 25.0: N = 79; 25.1 29.9: N = 104; 30.0 - max: N = 69). Baseline to 12 mo (MannWhitney U-test) and linear regression analyses performed for TFA/ TPSR and age or BMI. Results: All groups showed improvement in the clinical outcomes at 12 months compared to baseline (p\0.0001). TFA was similar for all subgroups [age groups: 1.0 (1.0)/1.0 (1.0)/1.0 (0.5) p = 0.8707; weight classes: 1.0 (1.0)/1.0 (1.0)/1.0 (0.0) p = 0.1013)]. TPSR was higher for older and heavier patients [age groups: 2.0 (1.0)/2.0 (2.0)/ 3.0 (3.0) p = 0.0662; weight classes: 2.0 (1.0)/3.0 (2.0)/3.0 (3.0) p = 0.1591)] with a significant linear relationship between TPSR and age (p = 0.0028) and TPSR and weight (p = 0.0024). ODI, VAS back and leg pain and EQ5D were similar in all subgroups at every time point. Conclusions: Older, heavier and older heavier patients need more time to recover from MILIF surgery, however the additional time needed on average remains acceptably below 24 hrs. The MILIF approach for spine surgery gives good results for subjects of all age groups and weight classes with no significant differences in clinical endpoints between subgroups. Disclosures: author 1: consultant; Company=Medtronic; author 2: grants/research support; Company=Medtronic, consultant; Company=Medtronic; author 3: consultant; Company=Medtronic, DePuy Synthes; author 4: none; author 5: none; author 6: grants/research support; Company=Medtronic Canada, consultant; Company=Medtronic Canada, Halifax Biomedical Inc.

P75 MEP AND SEP INTRAOPERATIVE NEUROMONITORING WITH DEFORMITY CORRECTION SURGERY. SINGLE CENTRE EXPERIENCE OF 8 YEARS Thomas Pfandlsteiner, Ahmed Ezzat Siam, Elsayed Shaheen, Cornelius Wimmer Scho¨n Klinik Vogtareuth, Spine Surgery and Scoliosis Centre, Vogtareuth, Germany Introduction: The aim of this study is the prospective evaluation of the risk of paraplegia in deformity correction surgeries monitored by intraoperative motor evoked potential (MEP) and sensory evoked potential (SEP). Methods: A prospective study from 6/2006 to 5/2014 with consecutive patients undergoing surgical correction of deformities, accompanied with intraoperative neuromonitoring with MEP and SEP. This included 572 patients; 326 females and 246 males. The indications for surgery were: idiopathic scoliosis in 244, neuromuscular scoliosis in 221, hyperkyphosis in 62, lytic spondylolisthesis and spondyloptosis (grade II to V Meyerding) in 45 cases. A signal change with a decrease of latency of 10 % and reduction of amplitude of 50 % or more has been considered as neurologically relevant, which leads to persistent abnormality. Results: Wake-up test has been done in 4 cases because of functional problems of the monitoring system. Transient significant signal changes during surgery have been documented in 33 cases, which

123

S774

Eur Spine J (2015) 24 (Suppl 6):S743–S800

recovered within 5 minutes. In one case, a postoperative incomplete paraplegia was diagnosed within 2 hours after surgery, despite normal intraoperative neuromonitoring. This was treated conservatively until neurological recovery in 2 weeks, followed by surgical revision and reduction of deformity correction. Discussion: The validity of intraoperative neuromonitoring with SEP and MEP is significantly higher than wake-up test, especially by neuromuscular scoliosis (compliance difficulties). The sensitivity of neuromonitoring from our experience is 94.5 % (33 cases). The neuromonitoring showed false negative situation in 0.174 %. The wake-up test should only be done when neuromonitoring fails. The neuromonitoring should be controlled at least for 30 minutes after deformity correction (MEP and SEP), because a reduction of correction will be needed in case of persistent amplitude reduction and decrease of latency, otherwise neurological complications are to be expected. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

Disclosures: author 1: grants/research support; Company=Medtronic; author 2: grants/research support; Company=Medtronic; author 3: none; author 4: none; author 5: grants/research support; Company=Medtronic, consultant; Company=Medtronic, royalties; Company=Medtronic; author 6: grants/research support; Company=OREF, consultant; Company=Medtronic, Nuvasive, royalties; Company=Medtronic.

P76 THE EFFICACY AND ACCURACY OF CONE-BEAM CT NAVIGATION ON SCREW POSITION IN PRIMARY CASES OF ADULT MAJOR DEFORMITY SURGERY

Kae Sian Tay, Anupreet Bassi, William Yeo, Wai Mun Yue

John Street, Jason Strelzow, Daniel Mendelsohn, Nicolas Dea, Marcel Dvorak, Charles Fisher University of British Columbia, Vancouver, Canada Hypothesis: Intra-operative cone-beam CT navigation would result in increased pedicle screw accuracy compared to non-navigated techniques. Design: Ambispective matched cohort of adult major deformity patients undergoing spinal instrumentation Introduction: Intra-operative cone-beam CT navigation systems may provide an opportunity to improve precision and accuracy of pedicle screw placement. Adult spinal deformity provides unique anatomical challenges potentially amenable to spinal navigation. Our study examines the efficacy and safety of intra-operative cone-beam CT navigation for pedicle screw placement in complex spinal deformity cases Methods: We identified patients treated at our institution with spinal fusion for major adult deformity between January 2008 and December 2012 in whom intra-operative cone-beam CT navigation was used (NAV). A historic control cohort (NonNAV) was matched based on age, number of levels, curve type and size and previous fusion. The number and timing of screw malposition, the need for revision screw placement was recorded, along with direction and anatomic level of misplaced screws. All patients had a minimum of 1 year follow-up. Quantitative statistical analysis compared screw placement between cohorts. Results: Fifty-six patients met inclusion criteria in each cohort. The mean number of screws placed in each group was not significantly different (p = 0.75). 38 (34 %) patients in the NonNAV group had misplaced screws compared to 21 (19 %) in the NAV group (p = 0.002). The need for intra-operative screw revision favoured navigation (p \ 0.03). The number of adverse events and length of stay were not significantly different. Mean number of post-operative CT scans was significantly fewer in the NAV group (p = 0.004) while mean OR time was statistically different between groups (492 mins vs. 408 mins, p = 0.002). Conclusion: Intra-operative cone-beam CT-guided navigation provides an equally safe and more accurate tool for pedicle screw placement than traditional techniques in adult spinal deformity surgery. There were more intra-operative screws adjusted and fewer post-operative screws revised with navigation.

123

P77 COMPLETE REDUCTION DOES NOT RESULT IN BETTER OUTCOMES IN LUMBAR LOW GRADE SPONDYLOLISTHESIS (LGS) WITH NEUROGENIC SYMPTOMS AFTER SINGLE LEVEL MINIMALLY INVASIVE SURGERY TRANSFORAMINAL LUMBAR INTERBODY FUSION (MISTLIF) - A 5 YEAR FOLLOW-UP STUDY

Singapore General Hospital, Singapore Rationale: To determine if complete reduction of lumbar LGS with neurogenic symptoms will result in significantly better clinical and radiological outcomes at 5 years in patients after single-level MISTLIF. Methods: Fifty six patients who underwent single-level MISTLIF between 2004 and 2009 for LGS with neurogenic symptoms with 5-year follow-up were included. Prospectively collected data were retrospective analysed. Patients with multiple level, high grade lumbar spondylolisthesis, previous lumbar surgery, tumours, infections and acute trauma were excluded. Depending on achievement of complete reduction intra-operatively, 26 cases of Non-Reduction Group (NRG) and 30 cases of Reduction Group (RG) were compared for age, sex, Body Mass Index (BMI) and the levels operated. Oswestry Disability Index (ODI), Neurogenic Symptom Score (NSS), the 36-Item Short Form Health Survey (SF-36) and Numerical Pain Rating Scale scores (NPRS) for back and leg pain were used to assess clinical outcomes. Radiological findings included spondylolisthesis grading, fusion, implant failure/loosening and Adjacent Segment Disease (ASD). Patients were evaluated preoperatively, at 6 months, 2 years and 5 years postoperatively. Results: Demographic factors (mean age, gender, race distribution and BMI) were similar in both the groups (p [ 0.05). Spinal levels operated, etiology (degenerative or isthmic), type of graft used (autograft or allograft), blood loss (ml), fluoroscopic time (sec), morphine used (mg), time to ambulation (days), length of operation (min), length of hospitalization (days) and complications were similar (p [ 0.05). Pre-operative clinical outcomes were similar in both groups (p [ 0.05). Clinical outcomes (ODI, NSS, SF-36, NPRS for back and leg pain) improved significantly (p \ 0.05), compared to pre-operatively and grade 1 fusion (Bridwell classification) was achieved in 100 % in both groups at 5-year follow-up. Complications (p[0.05) were found in 7 (26.9 % - 7 minor and 0 major) patients in NRG and in 4 (7 % - 1 major - misplaced screw required re-operation and 3 minor) patients in RG. All 10 minor complications (NRG+RG) and 3 new ASD cases (RG) were managed conservatively. Conclusion: Complete reduction does not result in better clinical, operative and radiological outcomes in lumbar Low Grade Spondylolisthesis (LGS) with neurogenic symptoms in patients undergoing single-level Minimally Invasive Surgery Transforaminal lumbar Interbody Fusion (MISTLIF) at 5-year follow-up. Disclosures: author 1: none; author 2: none; author 3: none; author 4: grants/research support; Company=Medtronics, Depuy-Synthes,

Eur Spine J (2015) 24 (Suppl 6):S743–S800 consultant; Company=Medtronics, Depuy-Synthes, other financial report; Company=Nuvasive.

P78 COST-UTILITY STUDY AND LIFE QUALITY IMPROVEMENT AFTER INSTRUMENTED LUMBAR FUSION IN ELDERLY PATIENTS OVER 80 YEARS OLD ´ ngel R. Pin˜era, Javier Melchor Duart Fe´lix Tome´-Bermejo, A Clemente, Luis Alvarez, Marta Martı´n Ferna´ndez Spine Department. Fundacio´n Jime´nez Dı´az University Hospital, Madrid, Spain Objective: Lumbar surgery has become one of the most rapidly growing surgical disciplines in medicine substantially increasing health care costs with a significant socio-economic burden. The value of lumbar spine fusion in elderly patients is not well documented and remains matter of considerable controversy. In a time of limited financial resources, the cost of the improvement in life quality health related (HRQOL) is an important consideration for the allocation of resources. In our country, any health intervention with an approximate incremental cost [30,000€ per QALYS is likely to be rejected as not be considered cost effective. The aim of our study is to determine the clinical, functional improvement and QALYS gained in elderly patients over 80 years old undergoing lumbar instrumented arthrodesis during a 2-year period. Material and method: Retrospective study with 25 consecutive elderly patients aged [80 years old suffering from symptomatic spinal stenosis undergoing one or two levels instrumented lumbar fusion with cement augmented screws. Preoperative comorbidity assessment according to the scale of the American Society of Anesthesiology (ASA). Outcome measures included Oswestry Disability Index (ODI), the onset of complications and reoperation. To determine QALY, the Short Form 6D (SF-6D), a utility index derived from the Short Form (36) Health Survey (SF-36) was used. Results: Mean age of 82.2 years. 40 % (10 patients) had a preoperative ASA-III. Perioperative complications were recorded in 6 patients (24 %). Two patients died of unrelated causes. Statistically significant improvement in the ODI (p \ 0.05) as shown from an average preoperative value of 61.8 to 31.8 at final follow-up. The accumulated mean health utility value over the two postoperative years showed a 0.60 QALY improvement. Conclusions: This study demonstrates significant functional improvement after instrumented lumbar fusion in elderly patients[80 years old despite a high postoperative complication rate. There is a growing demand not only to examine the efficacy of the treatment modalities but also to consider the related economic and social costs. Disclosures: author 1: none; author 2: none; author 3: none; author 4: grants/research support; Company=Biomet, consultant; Company=Spineart; author 5: none.

P79 DO SURGICAL EXPECTATIONS CHANGE DEPENDING ON FIRST TIME SURGERY OR RE OPERATION? A PROSPECTIVE COHORT STUDY IN LUMBAR SPINE SURGERY Gemma Vila` Canet, Sergi Rodrı´guez-Alabau, Augusto Covaro, Ana Garcı´a de Frutos, Maite Ubierna, Enric Ca´ceres Hospital Universitari Quiro´n-Dexeus, Barcelona, Spain

S775 Introduction and objective: Patients satisfaction after spinal surgery can be influenced by the fulfillment of their expectations. The objective of the present study was to assess if there is any difference on patient expectation depending on the first time surgery or reoperation for the same condition in degenerative lumbar spine procedures. A second objective was to determine if surgical expectations can be influenced by mood disorders. Material and methods: A consecutive prospective cohort of patients listed to be operated was selected including different lumbar degenerative etiologies. The study group was divided into two groups. Group 1: patients going to be operated for the first time and Group 2: patients that already had been operated for the same condition some months before. All patients completed a set of self-report questionnaires including: modified version of the NASS lumbar spine expectation questionnaire (8 items with an increasing range of expectations going from 1 to 5), lumbar and radicular VAS, ODI and Zung depression scale (ZDS). Results: Seventy-seven patients were enrolled in the study. 56 patients in group 1 and 21 in group 2. No differences were found in ODI, VAS and Zung results between two groups. We couldn’t find statistically significant differences in surgical expectations between the two groups (p [ 0,05). Both groups showed really high expectations, expecting about 80 or 100 % improvement in most of the 8 questionnaire items. Patients with depressive symptoms showed higher ODI values (p 0,001) but did not show less surgical expectations than non-depressed group (p [ 0,05) Discussion and conclusion: A previous failed lumbar surgical procedure did not worsen patient expectations for the following operation, neither do depression. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: consultant; Company=Surgival.

P80 INTERVERTEBRAL SPACE RESTORATION AFTER LUMBAR DISCECTOMY: IS VERTEBRAL END PLATE PERFORATION THE ANSWER? Miha Vodicar, Robert Kosak, Matevz Gorensek, Robert Korez, Tomaz Vrtovec, Jadran Koder, Rok Vengust 1

University medical centre Ljubljana, Department of Orthopedic Surgery; 2University of Ljubljana, Faculty of Electrical Engineering; 3 University medical centre Ljubljana, Institute of radiology, Ljugljana, Slovenia Low back pain is the commonest long-term complication after lumbar discectomy. It is mainly caused by intervertebral disc space loss, which promotes progressive degeneration. This is the first study to test the efficiency of a previously described method (vertebral endplate perforation) that should advocate for annulus fibrosus reparation and disc space restoration. We present the results of a pilot single-centre, stratified, prospective, randomized, double-blinded, parallel-group, controlled study. The aims were to determine whether vertebral end-plate perforation after lumbar discectomy causes annulus reparation and intervertebral disc volume restoration; to determine that after six months there would be no clinical differences between the control and study group; to determine whether the amount of disc tissue removal correlates with disc space loss. We selected 30 eligible patients according to inclusion and exclusion criteria and randomly assigned them to the control (no end-plate perforation) or study (end-plate perforation) group. Each patient was

123

S776 evaluated in 5 different periods where data was collected (preoperative and six months follow up MRI and functional outcome data (VAS back, VAS legs, Oswestry disability questionnaire). Intervertebral space volume (ISV) and height (ISH) were measured form the MRI images. Statistical analysis was performed using paired t-test and linear regression. P\0.05 was considered statistically significant. We found no statistically significant difference between the control group and the study group concerning ISV (p = 0.6808) and ISH (p = 0.8981) six months after surgery. No statistically significant differences were found between ODI, VAS back and VAS legs after six months between the two groups, however there were statistically significant differences between these parameters in different time periods. Correlation between the volume of disc tissue removed and preoperative versus postoperative difference in ISV was statistically significant (p = 0.0020). The present study showed positive correlation between the volume of removed disc tissue and decrease in postoperative ISV and ISH. There were no statistically significant differences in ISV and ISH between the group with end-plate perforation and the control group six months after lumbar discectomy. Clinical outcome and disability were significantly improved in both groups three and six months after surgery. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

P81 RISK FACTORS ASSOCIATED WITH DEEP SURGICAL SITE INFECTIONS AFTER PRIMARY ADULT DEGENERATIVE LUMBAR SCOLIOSIS: SYSTEM ANALYSIS OF 1727 ALDS Dingjun Hao, En Xie Department of Spine Surgery, Hong Hui Hospital, Xi’an Jiaotong University College of Medicine, Xi’an, China Background: Deep surgical site infection following ADLS is a devastating complication. Patient and surgical risk factors for this complication have not been thoroughly examined. The purpose of this study was to evaluate risk factors associated with deep ALDS surgical site infection.. Methods: A retrospective review of a prospectively followed cohort of primary ADLS recorded in a total spine surgery from 2007 to 2013 was conducted. Records were screened for deep surgical site infection with use of a validated algorithm, and the results were adjudicated by chart review. Patient factors, surgical factors, and surgeon and hospital characteristics were identified. Cox regression models were used to assess risk factors associated with deep surgical site infection. Results: A total of 1727 ADLS were identified; 63.0 % were done in women, the average age of the patients was 67.7 years (standard deviation [SD] = 7.7), and the average body mass index (BMI) was 27 kg/ m2 (SD = 7). The incidence of deep surgical site infection was 0.77 % 7. In a fully adjusted model, patient factors associated with deep surgical site infection included a BMI of C35 (hazard ratio [HR] = 1.33), diabetes mellitus (HR = 1.27), male sex (HR = 1.77), Protective surgical factors included use of Gentamicin in flushing (HR = 0.77), a bilateral procedure (HR = 0.57). Surgical risk factors included

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Surgical wound exposure (HR = 3.77) and the use of Hydrogen peroxide solution (HR = 1.77). In a subanalysis, operative time was a risk factor, with a 9 % increased risk per fifteen-minute increment. Conclusions: Use of a comprehensive infection surveillance system, combined with an ADLS registry, identified patient and surgical factors associated with infection following ADLS in a large sample. High-risk patients should be counseled, and modifiable clinical conditions should be optimized. Disclosures: author 1: none; author 2: none.

P82 COMPARISON OF STAGED RECONSTRUCTION WITH EXTREME LATERAL INTERBODY FUSION (XLIF) AND MULTILEVEL CORRECTIVE PLIF/TLIF FOR ADULT THORACOLUMBAR KYPHOSCOLIOTIC DEFORMITY Tokumi Kanemura, Kotaro Satake, Naoki Segi, Hidetoshi Yamaguchi Department of Spine and Orthopedic Surgery, Konan Kosei Hospital, Konan-City, Japan Background: We compared radiological and clinical outcomes between two-stage reconstruction with multilevel XLIF (group X) and one-stage multilevel corrective PLIF/TLIF (group P) for degenerative deformities in Japanese adult patients. Methods: A consecutive series of 30 patients fulfilled the following criteria: age [50 years; scoliosis [30° or [20° and lumbar lordosis (LL) \10°; and one- or two-stage reconstruction surgery from the lower thoracic spine to the pelvis. The group P (n = 15) underwent one-stage posterior fusion with multilevel corrective PLIF/TLIF. The group P (n = 15) underwent a first-stage multilevel XLIF and a second-stage procedure involving posterior fixation. The symptoms were evaluated using the Japanese Orthopedic Association (JOA) score and the JOA Back Pain Evaluation Questionnaire (JOABPEQ). Radiologic parameters included coronal Cobb angle (CC), sagittal vertical axis (SVA), pelvic tilt, pelvic incidence, and LL. Results: There was no difference between the groups regarding preoperative demographic data or radiographic parameters. No significant differences were observed in the recovery rate of the JOA score or improvement of JOABPEQ. Radiologically, no significant differences were present in the correction of deformity; the preoperative CC averaged 32.5° in the group P and 35.5° in the group X, which were corrected to 6.4° and 8.4°, respectively, after surgery. The preoperative LL averaged 20.7° in the group P and 12.1° in the group X, which was corrected to 42.3° and 49.6°, respectively. The preoperative SVA averaged 8.3 cm in the group P and 10.3 cm in the group X, which was corrected to 3.3 cm and 1.5 cm, respectively. The average total operative times were similar; however, in the group P, the mean EBL (3515 ml and 1362 ml), the rate of admission to the ICU (33.3 % and 6.7 %,), and the major complication rate (46.7 % and 6.7 %) were significantly greater compared with those in the group X (p \ 0.01). Conclusions: This retrospective study indicates that the multilevel XLIF procedure is a less invasive technique than the multilevel corrective PLIF/TLIF procedure for the treatment of degenerative scoliosis and kyphoscoliosis in adult patients who are [50 years. We

Eur Spine J (2015) 24 (Suppl 6):S743–S800 conclude that staged reconstruction for adult degenerative deformities is a reasonable approach, and affords effective correction with acceptable complication rates. Disclosures: author 1: consultant; Company=Madtronic, DePuy Synthes, AOSpine, NuVasive; author 2: none; author 3: none; author 4: none.

P83 DOES THE CURVE DIRECTION REMAIN CORRELATED WITH THE DOMINANT SIDE OF TONSILLAR ECTOPIA WHEN SYRINGOMYELIA IS ABSENT? AN ANALYSIS OF SCOLIOTIC CURVE PATTERNS IN CHIARI I MALFORMATION WITHOUT SYRINGOMYELIA Huang Yan, Zezhang Zhu, Zhen Liu, Bangping Qian, Yong Qiu From the Spine Surgery, the Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing, China Objective: To conduct a systematic and quantitative assessment of the correlation of tonsillar ectopia with scoliosis. Summary of background data: There was an observation that curve direction tended to be on the same side as the dominant tonsillar ectopia. However, the evidence for this relationship was obviously inconclusive due to the impact of associated syrinx. Methods: A retrospective study was conducted on patients with a scoliosis secondary to Chiari I malformation without syringomyelia (CMI-only). The curve direction, curve pat-tern/features and side of the dominant tonsillar ectopia were recorded and assessed quantitatively. Based on the measurement results, the correlations of the asymmetrically displaced tonsils, curve convexity and curve patterns/ features were analyzed. Results: A total of 26 patients, consisting of 7 males and 19 females with an average age of 15.4 years (range, 6 to 28 years), were included in the current study. In 19 patients with asymmetrically displaced tonsils, the concordance between the dominant side of the asymmetrically displaced tonsils and curve direction was 78.9 % (15/ 19). A statistically significant association was found between the dominant side of the tonsillar ectopia and the convex side of scoliosis according to Fisher’s exact test (P = 0.045). In addition, it was noted that there was a high incidence (52.9 %) of atypical scoliotic cases in CMI-only patients. A significantly high incidence of atypical features with a superior shift of either the apical or the end vertebrae was found in 85.7 % of thoracic curves and 40 % of lumbar curves. Conclusions: Asymmetrical cerebellar tonsillar herniation is an underappreciated radiological feature that may often have clinical correlations with curve direction and specific curve pat-terns/features. The evidence from this study supports the idea that the direct compression of the dorsal columns of the spinal cord by the impacted cerebellar tonsils, could interfere with spinal cord function, thus providing impetus to initiate and exaggerate the scoliosis. Keywords: Chiari I Malformation without syringomyelia (CMIonly), scoliosis, tonsillar ectopia, asymmetry. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

S777 P84 WHICH FACTORS INFLUENCE THE SURGERY VS. NONSURGERY DECISION FOR ADULT IDIOPATHIC SCOLIOSIS PATIENTS WITH GRAY ZONE (40-55 DEGREE) MAIN THORACIC CURVES? Caglar Yilgor, Meric Enercan, Azmi Hamzaoglu, Ferran Pellise, Paco Sanchez Perez-Grueso, Emre Acaroglu, Ibrahim Obeid, Frank Kleinstuck, Ahmet Alanay, ESSG European Spine Study Group Acibadem University School of Medicine, Istanbul Spine Center, Turkey; Spine Unit, Hospital Valle Hebron Barcelona; Hospital De La Paz, Madrid, Spain; Orthopedic Spine, Ankara Spine Center, Turkey; CHU Bordeaux Pellegrin Hospital, France; Department of Orthopaedics Schulthess Klinik, Zurich, Switzerland; Fundacio´ Institut de Recerca Vall Hebron, Barcelona, Spain Summary: The treatment decision-making process for gray zone adult idiopathic scoliosis (AdIS) patients is controversial. Analysis of 44 non-surgical and 20 surgical consecutive, multicenter patients revealed that decreased function, pain and self-image created a predilection towards surgery. Hypothesis: Functional status, pain and self-image play a role in surgical decision making process of AdIS patients. Design: Retrospective analysis of a multicenter, prospective, consecutive patient series Introduction: To analyze the factors that may influence surgical vs non-surgical treatment for AdIS patients within gray zone (40-55o) main thoracic (MT) curves. Methods: A retrospective analysis of a multicenter, prospective, consecutive patient series. Inclusion criteria were: AdIS, C18 years of age, major curve to be MT, Cobb between 40 and 55o. Sixty-four patients (44 Non-surgical and 20 Surgical) were included. Non-surgery group had 34 F and 10 M; mean age: 26.8 (18-47), mean Cobb: 46.9 (40-55). Surgery group had 18 F and 2M; mean age: 28.0 (1871), mean Cobb: 49.5 (43-55). All patients completed SF-36, SRS-22 and ODI when they were first seen in the clinic. AP and lateral Cobb measurements and sagittal plane parameters were measured. Independent samples t-test was used to compare groups. Using all the demographic, radiographic and patient-reported outcome data, a variable importance analysis was done using classification and regression tree algorithm to predict factors that influence surgical decision. Results: The two groups were matched according to age, sex, MT and Lumbar curve Cobb, coronal balance, trunk shift, shoulder parameters, sagittal Cobb, SVA, pelvic parameters and leg length discrepancy (p [ 0.05). Most important variable that created a tendency towards surgery was SRS-22 functional status followed by SRS-22 pain and self image scores, ODI and SF-36 PCS. Conclusion: AdIS patients having a curve magnitude in the gray zone (40-55o) with decreased SRS-22 function, pain, self-image and SF-36 PCS and increased ODI scores have a predilection towards surgery. Most important variable was SRS-22 function score. Disclosures: author 1: none; author 2: none; author 3: consultant; Company=Medtronic; author 4: grants/research support; Company=DePuy Synthes, K2M, consultant; Company=DePuy synthes, Biomet; author 5: grants/research support; Company=DePuy Synthes,

123

S778 consultant; Company=DePuy Synthes; author 6: grants/research support; Company=Medtronic, Depuy Synthes, stock/shareholder; Company=IncredX; author 7: grants/research support; Company=Depuy Synthes, consultant; Company=Depuy Synthes Medtronic; author 8: none; author 9: grants/research support; Company=DePuy, consultant; Company=Stryker; author 10: grants/ research support; Company=Depuy-synthes.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 support; Company=Medtronic, Depuy Synthes, stock/shareholder; Company=IncredX; author 7: grants/research support; Company=Depuy Synthes, consultant; Company=Depuy Synthes Medtronic; author 8: none; author 9: grants/research support; Company=Depuy Synthes, consultant; Company=Stryker; author 10: grants/research support; Company=Depuy-synthes. TRAUMA, TUMOR

P85 MENTAL HEALTH AND SELF IMAGE PERCEPTION OF NON-DISABLED ADULT IDIOPATHIC SCOLIOSIS PATIENTS HAVING MILD TO MODERATE CURVES COMPARED TO NORMAL POPULATION Caglar Yilgor, Meric Enercan, Azmi Hamzaoglu, Ferran Pellise, Paco Sanchez Perez-Grueso, Emre Acaroglu, Ibrahim Obeid, Frank Kleinstuck, Ahmet Alanay, ESSG European Spine Study Group Acibadem University School of Medicine, Istanbul Spine Center, Turkey; Spine Unit, Hospital Valle Hebron Barcelona; Hospital De La Paz, Madrid, Spain; Orthopedic Spine, Ankara Spine Center, Turkey; CHU Bordeaux Pellegrin Hospital, France; Department of Orthopaedics Schulthess Klinik, Zurich, Switzerland; Fundacio´ Institut de Recerca Vall Hebron, Barcelona, Spain Summary: This study that includes patients with curves between 20-55° suggests that the self image and mental status are unaffected by mild and moderate curves if the patient has no disability due to pain. Hypothesis: Mild-moderate AdIS does not affect mental health and self-image in non-disabled patients. Design: Retrospective analysis of a multicenter, prospective, consecutive patient series Introduction: There is little information about the effects of scoliotic curves on self image (SI) and mental health (MH) based on validated questionnaires. Aim was to analyze the effect of scoliosis on MH and SI in a non-disabled adult idiopathic scoliosis (AdIS) population having curves under surgical indication threshold (Main Thoracic (MT) Cobb (B55°), and Thoracolumbar/Lumbar (TL) curve (B45°). Methods: A retrospective analysis of a multicenter, prospective, consecutive patient series. Inclusion criteria were: non-operated AdIS, C18 years of age, MT Cobb 20-55°, TL Cobb 20-45°, ODI \20, SRS 22 Pain score[4. ODI and SRS22 pain score were used to distinguish patients that have pain and disability from the ones that do not. SRS22 and SF-36 normative data for different age groups were used for comparison. AP and lateral Cobb measurements, sagittal plane parameters and demographic data were analyzed in terms of correlations with SI and MH parameters. Results: 76 patients (64F, 12M) met the inclusion criteria. Mean age was 25.9 (18-44), mean MT Cobb was 38.5° (21-55) and TL Cobb was 34.6° (25-44). SRS-22 MH, SI and other domains were similar with SRS normative data (p [ 0.05). SF-36 MCS and PCS domains for age groups 18-24, 25-34 and 35-44 were not significantly different than normative data (p [ 0.05). None of the demographic and radiographic parameters were correlated with changes in SRS-22 function, SI, MH and SF 36 MCS, PCS parameters. Conclusion: AdIS does not affect mental health and self-image in non-disabled patients with curve magnitudes below the surgical threshold. Disclosures: author 1: none; author 2: none; author 3: consultant; Company=Medtronic; author 4: grants/research support; Company=DePuy Synthes, K2M, consultant; Company=DePuy synthes, Biomet; author 5: grants/research support; Company=DePuy Synthes, consultant; Company=DePuy Synthes; author 6: grants/research

123

P86 THE USAGE OF THE METHODS OF PREVENTION OF VERTEBRAL FRACTURES ADJACENT TO THE TRANSPEDICULAR VERTEBRAL SYSTEM Igor Basankin, Karapet Takhmazyan, Sergey Malakhov, Asker Afaunov, Dmitry Ptashnikov, Nikolay Gavryushenko, Olga Ponkina, Vladimir Shapovalov Research Institute Krasnodar Regional Hospital 1, Krasnodar, Russia Thematic justification: Compression stress fractions of the vertebrae in the upper section fitted with steel structure is a serious problem, especially in extensive fusion and in patients with osteoporosis. Nowadays there is the lack of information on the fracture of the vertebrae adjacent to the steel structure preventive methods. Purpose: The determination of the possible ways of the adjacent vertebrae fractures prevention in conditions of cement pedicle fixation in the setting of osteoporosis. Study design: Biomechanical cadaver study. Materials and methods: We used cadaver material obtained from females aged 66-83 years. The length of the blocks withdrawn from Th10 to L4 (7 vertebrae). The fracture L1 was simulated in all the blocks by its mechanical destruction. In the study there was a total of 15 blocks divided into 3 groups of five units each: 1 - control (PSF Th12-L2 with cement); 2 - with preventive vertebroplasty Th11 and L3 above and below the PSF Th12-L2 with cement. 3 - PSF Th12-L2 with cement + laminar fixation Th11 ‘‘Universal Clamp’’ We used the axial load by an universal servo-hydraulic testing machine ‘‘Walter + bay ag’’ LFV-10-T50 (Switzerland). Compression of the blocks was carried out before the graphic display signs of fracture. We performed X-ray of blocks before and after the study. Results: The control group of blocks revealed the appearance of vertebral fractures just above the metal structure (Th11). The first graphic signs of fracture identified in the range of 0.78-0.94 (average 0.84 ± 0.39831) kN. Group of blocks with prophylactic vertebroplasty Th11 and L3 revealed fractured vertebrae adjacent to the vertebroplasty (Th10). In this case, the first graphics oscillations corresponding fracture identified in the range 1,78-2,05 (average 1.91 ± 0.40566) kN. Group of blocks with laminar fixation ‘‘Universal Clamp’’ revealed a fracture Th10 and Th11. The first graphic oscillations corresponding fracture identified in the range 0,96-1,48 (average 1.21 ± 0.192666) kN . Conclusion: Prophylactic vertebroplasty of overlying vertebra is an effective way to prevent the pathological fracture (Th11), and also increases the stability of the overlying (Th10) vertebra to fracture in 2 - 2.5 times. The usage of laminar fixation ‘‘Universal Clamp’’ does not prevent vertebral fracture over the pedicle system (Th11), but allows to increase the stability of the vertebral fracture in 1.5 times Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: none.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 P87 THE EPIDEMIOLOGY OF SURGICALLY TREATED SPINAL FRACTURES IN EASTERN DENMARK Thomas Pensbo-Madsen, Kiran Anderson, Corrado Lucantoni, Ram Babu, Martin Gehrchen, Fin Biering-Sørensen, Benny Dahl Spine Unit, Department of Orthopaedic Surgery, Rigshospitalet and University of Copenhagen; Department of Spinal Cord Injuries, Rigshospitalet and Glostrup Hospital and University of Copenhagen Introduction: The epidemiology of spinal fractures is of relevance to monitor the impact of this injury and if possible initiate preventive measures. Few studies are representative of a complete population, and can therefore be affected by referral bias. In October 2010 the first Spine Unit in Denmark with neurosurgical and orthopaedic spine surgeons was established. Consequently, all patients operated for a spinal fracture in Eastern Denmark were operated in one facility. The purpose of the present study was to To describe the epidemiology of surgically treated spinal injuries in a tertiary referral center serving a population of 2.5 million people, and compare the findings with results from the literature. Material and methods: All patients undergoing surgical treatment of spinal injuries were registered in a prospective database from October 1st, 2010 through December 31st, 2012. All fractures were classified according to location (C0-C2, C3-C7, T1-T9, T10-L2, L3-L5, and complex injuries). The database was retrospectively reviewed for information regarding age, gender, surgical strategy, and SCI. Results: A total of 275 patients were operated during the 27 months study period, corresponding to an overall incidence of 4.9/100.000/ year. 66 % of the patients were secondarily referred from another hospital. Most fractures were located in the cervical and thoracolumbar region. Patients with cervical injuries were significantly older than patients with thoracolumbar fractures, with an average age of 58 years compared to 46 years (P = 0.000, Mann-Whitney). 59 % of the patients had no other injuries. A total of 40 patients (15 %) had neurological injuries (paraplegia, tetraplegia or central cord syndrome). This corresponds to an SCI incidence of 7.1 per million per year. The majority of patients were treated in one surgical procedure whereas 32 patients were treated with combined procedures; typically an initial posterior instrumentation followed by an anterior corpectomi and cage insertion. Discussion and conclusion To our knowledge this is the first epidemiological study, in a Danish trauma population, focusing on surgical treatment of spinal injuries. The incidence of surgical procedures is comparable to other North European studies, and the incidence of SCI is low, compared to other reports. Disclosures: author 1: none; author 2: none; author 3: no indication; author 4: no indication; author 5: grants/research support; Company=Medtronic, Globus Medical, K2M, consultant; Company=Medtronic, Globus Medical, K2M; author 6: none; author 7: grants/research support; Company=Medtronic, Globus Medical, K2M.

P88 IS IT SAFE THE PERCUTANEOUS PEDICLE SCREW FIXATION AFTER SPINAL CANAL DECOMPRESSION IN MAGERL TYPE A3 THORACOLUMBAR BURST FRACTURES? Moon Soo Han, Jung-Kil Lee Chonnam National University Medical School & Research Institute of Medical Sciences

S779 Objective: Although there are some controversy about the safety of PPSF in thoracolumbar burst fractures (TLBFs), but PPSF without grafting seems to be suitable for minimally displaced Magerl Type A1, A2, and B2 fractures. Recently, we performed PPSF with decompression in Magerl type A3 TLBFs with neurologic deficits corresponding with fracture fragment. So, in the present study, we investigated the surgical results PPSF after spinal canal decompression via small laminotomy in the treatment of Magerl type A3 TLBFs with neurologic symptoms. Methods: This study included 25 patients, who underwent PPSF after spinal canal decompression for stabilization of Magerl type A3 TLBFs with symptomatic canal encroachment between January 2009 and December 2012. PPSF was performed as the short segment fixation including fractured vertebra. For spinal canal compression by bony fragment, disc material, or hematoma with corresponding neurologic deficit, decompression was performed via small laminotomy. To assess the reduction of spinal canal, computed tomography (CT) scan was performed before discharge. For patient’s pain and functional assessment, visual analogue scale (VAS), Frankel grading system, and Greenough Low Back Outcome Score (LBOS) were measured. Results: The average follow up period was 16.3 months. The average preoperative Cobb angle was 19.2 ± 8.1 degrees, which significantly decreased to 7.2 ± 4.9 degrees after surgery (P \ 0.05). The average vertebral body compression ratio and vertebral wedge angle were 38.4 ± 11.5 % and 25.3 ± 11.5, which improved significantly to 15.3 ± 8.4 % and 11.8 ± 5.9 after surgery (P \ 0.05). In last followup, regional Cobb angle was slightly increased to 10.9 ± 5.4 degrees. Preoperative spinal canal encroachment was 41.3 ± 11.6, and it was decreased to 18.3 ± 10.6 at immediately after surgery. In Magerl type A3.3 TLBFs, correction loss of kyphosis was developed more large compared to Magerl type A3.1 and 3.2 TLBFs. Clinically, there was no newly developed neurological deficit after surgery and during follow-up period. VAS and LBOS was also significantly improved at the last follow-up compared to immediately after surgery. Conclusions: PPSF after spinal canal decompression via small laminotomy may be applied as the effective primary surgery in Mageri type A3 TLBFs with neurologic symptoms Correction loss of kyphosis was developed larger in type A3.3 TLBFs than type A3.1 and A3.2 TLBFs Disclosures: author 1: none; author 2: none.

P89 BLOOD LOSS IN SPINAL TUMOUR SURGERY: EVALUATION OF INFLUENCING FACTORS Naresh Kumar, Aye Sandar Zaw National University Health System, Singapore Background: Surgical management of spinal tumours is often of prolonged duration, which makes the anticipation and replenishment of blood lost during the surgery quite significant. Till date, there exists a general consensus that spinal tumour surgeries results in significant blood loss. As a result, blood products are often ordered preoperatively in excessive quantities due to the non-existence of specific blood ordering guidelines. This study was conducted to investigate the extent of blood loss associated with various surgical approaches and vertebral levels for different types of spinal tumours so that it would facilitate pre-operative planning and management of blood transfusion. Materials and methods: We retrospectively analyzed the patients who underwent spinal tumour surgery in our institution from 2005-2014. Types of primary tumour (I: highly vascularised tumour,

123

S780 II: moderately vascularised tumour, III: hematological tumour and IV: primary bone tumour), types of surgery (1: cervical corpectomy and stabilization, 2: thoracolumbar posterior decompression and instrumentation, 3: thoracolumbar corpectomy and 4: minimally invasive surgery), level of instrumentation (\6, 6-9 and [9 levels) and decompression (0, 1-2. 3-4 and [4) on the influence of blood loss were analyzed. Generalized linear model was used for analysis. Results: A total of 259 patients were evaluated. Mean blood loss was 660 ml (range: 20-6000 ml). Multivariate analysis revealed that intraoperative blood loss rose significantly in proportion to increasing levels of decompression as well as instrumentation. Compared to highly vascularised tumour group, there was significantly reduced mean blood loss in moderately vascularised and haematological tumour groups (-617 ml and -436 ml respectively). Compared to open cervical approach, there was significantly less mean blood loss in minimally invasive surgery (-614 ml) and a borderline increase in thoracolumbar corpectomy surgery (319 ml). Significant association between blood loss and blood transfusion was also observed. Conclusions: There was a significant variation in the amount of blood loss based on primary tumours and types of surgery. This study would help in pre-operative planning to address the significant problem of blood loss during various spine tumour surgeries. Disclosures: author 1: none; author 2: none.

P90 VERSATILITY OF PERCUTANEOUS PEDICULAR SCREW FIXATION IN METASTATIC SPINE TUMOUR SURGERY Naresh Kumar, Aye Sandar Zaw, Rishi Malhotra, Pang Hung Wu, Milindu Makandura National University Health System, Singapore Background: Minimally invasive surgery has been evolved to address the problems associated with metastatic spine diseases (MSD). Posterior percutaneous spinal fixation (PPSF) is one of the main modalities of minimally invasive surgery. There have been a number of studies which evaluated the outcomes of PPSF but none of them assessed in detail regarding the versatility of PPSF in MSD. This study was designed to evaluate feasibility and spectrum of application of PPSF in management of MSD, highlighting its clinical advantages. Hypothesis: The spectrum of PPSF is quite versatile in management of MSD, even in patients with predicted poor survival prognosis Methods: Twenty-seven consecutive patients with MSD treated with PPSF in our institution from January 2011 to June 2014 were studied. All patients, after a multi-disciplinary assessment, were considered for surgical intervention due to clinical presentation of either neural deficit, skeletal instability, or both. Some of these patients belonged to poor prognostic category based on survival prognostic scoring systems i.e Tokuhashi and Tomita. The patients were categorized into seven groups depending on the modality of PPSF used. Demographic data, operative details, and clinical outcomes were investigated for each category and compared pre and postoperatively. Results: The median age was 60 years (range: 49-78 yrs). Generally, all patients either maintained or improved their neurological status and achieved pain alleviation. none of the patients in any group showed worsening in neurological status in postoperative period. Ambulatory status and ECOG scores were also improved using any modality of PPSF. Pure stabilization group had the lowest amount of mean blood loss (92 ml), shortest operative time (180 minutes), ICU (1 day) and hospital stays (10 days) while long construct group was

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 observed to have greatest amount of blood loss (355 ml), longest operative time (305 minutes) and ICU stay (2.5 days). Conclusion: For patients with MSD, even with predicted poor prognosis on survival prognostic scoring systems, it is possible to improve functional outcomes and quality of life with PPSF keeping surgical morbidity to a minimum. PPSF allows addressing patients with pure spinal instability successfully with least morbidity. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

P91 FUNCTIONAL AND RADIOLOGICAL RESULTS OF TRANSPEDICULAR ENDPLATE REDUCTION AND BONE AUTOGRAFT WITH POSTERIOR INSTRUMENTATION FOR AO TYPE A3.1 THORACOLUMBAR BURST FRACTURES Michael Osti, Julia Wolfram, Werner Ploner, Karl Peter Benedetto Academic Hospital Feldkirch, Austria Background: The objective of this study was to evaluate the radiological and functional outcome following transpedicular endplate reduction and autologous bone grafting in AO/ASIF Type A3.1 fractures and compare the alternatives of open short-segment and percutaneous multilevel instrumentation. Methods: 58 patients (m:w = 39:19, mean age 34.1 ± 13.3 years) and 20 patients (m:w = 17:3, mean age 37.4 ± 13.7 years) were operated in posterior open short-segment (Group A, 80 % thoracolumbar junction) and posterior percutaneous multilevel technique (Group B, 70 % thoracolumbar junction). Reconstruction of the anterior column in both groups was accomplished by transpedicular autologous cortico-cancellous bone grafting following balloon-assisted endplate reduction. Radiological evaluation included monoand bisegmental kyphotic angles (MKA, BKA) and sagittal vertebral body index (VBI) at the time of trauma, post-surgery, implant removal and follow-up. Functional assessment consisted of OswestryDisability-Index (ODI), visual analogue scale (VAS) and Dennis work and pain scales at the time of trauma and follow-up. Results: After 4.2 ± 1.8 years (Group A) functional results averaged to ODI 11 (pre-OP 3), VAS 80 (pre-OP 94), Dennis pain 2 (pre-OP 1) and Dennis work 2 in 27 patients. Occupational capacity and leisure activities averaged to 75 % of the pre-traumatic status. All patients were free of complaints regarding graft harvest. Average loss of correction in Group A was 3.1° (MKA), 5.0° (BKA) and 0.0 (VBI) between surgery and IR and 8.4° (MKA), 12.0° (BKA) and 0.1 (VBI) between surgery and FU, respectively. In Group B radiological differences averaged to 3.1° (MKA), 5.6° (BKA) and 0.0 (VBI) between surgery and IR. Restoration of MKA and VBI between posttraumatic and postoperative images averaged to 16.7° and 0.3 in Group A and to 14.3° and 0.3 in Group B. Operation time was significantly lower and intraoperative radiation time significantly higher in Group B. Conclusion: Yielding the benefit of a segment preserving, biological and rational reconstruction for A3.1 fractures, the presented technique results in favorable patient satisfaction as well as occupational and recreational reintegration. The instrumentation technique did not affect radiological results between surgery and IR. Dynamic loss of correction levels off at a maximum of 8.4°. Despite a considerable sintering over time, radiological measurements did not correlate to inferior functional results. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 GROWING SPINE

P92 THORACIC SAGITTAL PLANE VARIATIONS BETWEEN PATIENTS WITH MODERATE THORACIC ADOLESCENT IDIOPATHIC SCOLIOSIS AND HEALTHY ADOLESCENTS Alberto Nu´n˜ez Medina, Javier Pizones, Felisa Sa´nchez-Mariscal, Lorenzo Zu´n˜iga Go´mez, Enrique Izquierdo Hospital Universitario de Getafe, Madrid, Spain Introduction: Idiopathic thoracic curves are believed to be induced in part by thoracic hypokyphosis. Our hypothesis is that the sagittal thoracic plane of patients with moderate thoracic idiopathic curves is similar to that of healthy patients. Our aim was to analyze the sagittal parameters of different types of progressive thoracic Adolescent Idiopathic Scoliosis (AIS) and compare them with the same parameters in healthy adolescents. Methods: 231 patients were included in a prospective comparative analysis of two cohorts. 115 AIS patients with moderate main thoracic curves (Cobb angle: 59.4 ± 12.7) were prospectively compared with 116 healthy adolescents. AIS and Control (C) groups were homogeneous in terms of age and gender. Standing sagittal X-rays were analyzed for differences in T5-T12 kyphosis, T5-T8 and T9-T12 segmental kyphosis, the change between these two angles (TK Change) and the double rib contour sign (DRCS). Statistical analyses were performed using the Chi2, One-way ANOVA, Mann Whitney-U and Student’s-t tests with SPSS 20. Results: The variables: Age (AIS: 14.9 ± 1.9 vs C: 14.8 ± 0.9) and gender (AIS-Female: 80 %; C-Female: 70 %) were homogeneous (P [ 0.05) between groups. The sagittal parameters of Lenke 1 curves did not differ from healthy adolescents (T5-T8: 17.1 ± 10 vs C: 16 ± 7); (T9-T12: 6.3 ± 7 vs C:7.9 ± 5); (T5-T12: 23.9 ± 14 vs 23.9 ± 8). Compared with controls, Lenke type 3 curves were globally more hypokyphotic (T5-T12: 18.9 ± 12 vs C: 23.9 ± 8, P = 0.027) due to a ‘‘lordosis’’ of the lower thoracic segment (T9T12: 0.9 ± 10 vs C: 7.9 ± 5, P = 0.001). Type 2 curves tended to have more pronounced upper thoracic kyphosis (T5-T8: 20.7 ± 12 vs C: 16 ± 7). For these reasons, both types 2 and 3 require a marked TK change in the transition between the upper and lower thoracic segments (type 2: 15.1 ± 12 and type 3: 16.1° ± 15 vs C: 8.7 ± 8, P \ 0.02) to compensate for global (T5-T12) kyphosis. Conclusions: In moderate AIS, Lenke 1 curves have normal thoracic sagittal parameters, which questions the impact of lordosis in the development of single thoracic curves. Lenke 3 curves show lower thoracic segmental hypokyphosis, while type 2 showed upper segmental hyperkyphosis. These results should be considered when planning a surgical strategy. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none

S781 Zentralklinik Bad Berka, Germany; Assiut University Hospitals, Egypt Introduction: In recent years the incidence of spinal infection in western societies has been increasing. Paediatric Spinal infection (SI) is a rare disease with unknown incidence, due to the few case series available the literature. Patients and methods: Retrospective study of prospectively collected patients with SI in the time between 1995 and 2010. The analysis revealed 16 patients under the age of 20 years (1.41 %) of totally 1137 cases of SI treated in our center, (9 males and 7 females, mean age: 15 years (4-20) with mean follow-up of 5.8 years. We analysed the clinical presentation, radiological and laboratory findings, treatment methods and outcomes. Results: The mean interval between the onset of symptoms and the definite management was 52 days for pyogenic and 207 days for tuberculous infection. Back pain was the main presenting symptom in 15 patients (94 %), fever in 9 (56 %), neurological affection in 2 and spinal deformity in two patients. Haematogenous infection occurred in 14 patients, one developed postoperatively and one due to contiguous spread. The most commonly affected region was thoracic in 9 patients (56 %) and lumbar in 5 patients. Non-contiguous infection; thoracic and lumbar in one patient, all 4 spinal regions were affected in one patient with tuberculosis. Pyogenic infection in 11 cases (of them 7 cases with Staph. aureus), three patients had Mycobacterial tuberculosis and in two cases no organism could be isolated. At the time of diagnosis, mean CRP was 85 mg/L (6.9 at the last FU), WBS was 10,000/mm3 (7 at last FU) and ESR of 77 mm/ hr (26 at the last FU). Debridement and fusion surgery was necessary in 12 patients, less invasive surgeries in 8 patients of them. In four cases biopsy to identify the causative organism with antimicrobial susceptibility testing and antibiotic therapy was sufficient to manage the infection. The clinical outcome at the last FU was excellent in 10 patients, good in 5 and fair in one patient with residual neurological deficits. Conclusion: The onset of SI in children is basically different from that in adults in terms of its course and severity. It presents a nonuniform picture of nonspecific and usually mild symptoms, making it difficult to diagnose. The disease is often benign and self-limiting. High level of suspicion is needed supported with radiological and laboratory investigations to early diagnose the SI in this age group. Disclosures: author 1: none; author 2: none; author 3: none; author 4: royalties; Company=Medicon eG.

P94 1030 PATIENTS TREATED WITH MINIMALLY INVASIVE OZONE DISCOLYSIS IN LUMBAR DISC HERNIATION: 12 MONTHS FOLLOW-UP Alejandro Ortiz, Cesar Alcantara-Canseco, Eduardo Rivero-Sigarroa, Jose Luis Hernandez-Oropeza Hospital Medica Sur, Mexico City, Mexico

NEW TECHNIQUES, IMAGING, PATIENT SAFETY, INFECTION, COMPLICATIONS

P93 SPINAL INFECTION IN CHILDREN AND TEENAGERS Hamdan Abdelrahman, kais Abu Nahleh, Mootaz Shousha, Heinrich Boehm

Ozone is a made up of 3 oxygen atoms, with a peculiar smell produced by a medical ozone generator and it is not always toxic. Biochemical property of a gas mixture of oxygen-ozone is an important complementary treatment option for vertebral spine pathology; is performed by a percutaneous minimally invasive intradiscal, periganglionic and periradicular injection in treatment for lumbar disc herniation. Ozone therapy has been performed in Europe for more than 35 years. In the past, empiricism has been an obstacle, but in last decade mechanisms of action and biochemistry in blood, fluids and different structures have been studied, so that ozonetherapy has become a scientific discipline so as classical medicine.

123

S782 Objective: Examine the use of ozone discolysis as a treatment option for lumbar disc herniation. Materials and methods: It is an observational, prospective and descriptive study. 1030 patients were treated with ozone discolysis in a third level private hospital in Mexico City. The group was studied with lumbar spine MRI, and treated with discolysis between 2006 and 2013, 12 months’ follow-up. Descriptive statistics for categorical variables with frequencies and percentages for numerical variables with its corresponding central tendency and dispersion, analyzed with Stata 13. Results: 1030 patients treated with lumbar ozone discolysis, 577 males and 453 females, age of 52 ± 15.99, BMI of 26 ± 2 females and 28 ± 3 males. In 982(95.33 %) patients the ozone discolysis worked, 48(4.67 %) patients did not come back and/or ended in surgery. Lumbar segments treated were L1L2 (2 %), L2L3 (10 %), L3L4 (11 %), L4L5 (43 %) and L5S1 (34 %). Patients were released from the study upon reporting the cessation of pain taking into consideration SF-36 health survey, Oswestry Scale and Numeric Analogue Scale statistically analyzed. Patients treated with discolysis have statistically significant symptoms improvement between good and excellent in [95 % which prevails in 94 % of the patients with a follow-up of 12 months without pain medication. Conclusions: Results suggests that ozone therapy is a useful modality in the treatment of lumbar disc herniation. Discolysis has an important and beneficial effect as a treatment for disc herniation; it also shows statistically significant improved efficacy with a reduced treatment time. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P95 LOW PROFILE PELVIC FIXATION: ANATOMIC PARAMETERS FOR NEW EXTRA.ARTICULAR ILIAC FIXATION VERSUS SACRAL ALAR-ILIAC FIXATION Veronica Familiar, Julia Montoya, Marta Martin-Fernandez, Angel R. Pin˜era, Felix Tome, Javier M. Duart, Francisco M. Garzon, Luis Alvarez Fundacio´n Jime´nez Dı´az, Madrid, Spain Introduction: Long anchors projecting into the ilium provide optimal pelvic fixation. The traditional starting point in the PSIS requires muscle dissection and connectors or rod bends. Iliac fixation through the S2 ala provides a starting point in line with S1 pedicle anchors and implant prominence is minimized. However, pelvic morphology varies between individuals, and transfixiation of the sacroiliac joint is not always possible. We describe a new entry point for low profile iliac screws that consider the perpendicular axis to the narrowest width of the ilium, considering the width of the screw. Methods: We review 30 patients with low profile pelvic fixation. There are two groups; I.- 30 screws with the starting point at S2, and II.- 30 screws with an entry point through medial ilium cortical wall. Ideal trajectory considering the width of the screw though the narrowest width of the ilium was obtained with a three-dimensional CT program (Alma), by 2 radiologists, and the results were compared in both groups with the real trajectory of the crews placed. Results: The new trajectory described, that considers the width of the screw, correlates statistically better with the new entry point described in all parameters, except for distance from the skin and sagittal plane angulation, were they are similar. The number of violations of external cortical wall decreases.

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Conclusion. The new entry point described has the advantage of low profile pelvic fixation, with an starting point in line with S1 pedicle anchors and low implant prominence, but adapts better to the morphological features of the pelvis of each individual. Disclosures: author 1: none; author 2: grants/research support; Company=, none; Company=none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: grants/ research support; Company=Biomet, consultant; Company=Spineart.

P96 ARE TRADITIONAL RADIOGRAPHIC METHODS ACCURATE PREDICTORS OF PEDICLE MORPHOLOGY? Siddharth Badve, William Lavelle, Nathaniel Ordway, Brandon Clair, Stephen Albanese State University of New York & Upstate University Medical Center, Mumbai, India Introduction: Traditional radiographic imaging techniques overestimate minimal pedicle diameter (MPD) and measurement depends on pedicle orientation. This study aims to examine reliability of MPD based on coronal and transverse CT reconstructions. Methods: Thoracolumbar spines (N = 5, T1-L4, 48-59 years) were CT scanned using standard protocols. At each vertebral level, coronal and transverse thin cut 1mm slice reconstructions acquired orthogonal to pedicle axis in the sagittal plane. MPD was measured for both reconstructions. Additional measurements collected for coronal and transverse pedicle orientation. Data from 128 lumbar and thoracic pedicles organized into groups by vertebral level to assess trends and variations between transverse and coronal pedicle diameter measurements and angle variations. A paired t-test performed to compare MPD based on reconstruction type. Regression analyses between the difference in MPD from transverse and coronal images examined against pedicle orientations. Results: MPD was significantly less (p \ 0.001) based on transverse compared to coronal images. Mean coronal and transverse pedicle orientation were 8.4 degrees and 12.2 degrees, respectively, but had a large range based on vertebral level. There was significant correlation between difference in MPD and coronal pedicle orientation (r2 = 0.19, p \ 0.001). As coronal orientation increased, MPD from transverse image got larger than that from coronal image. There was significant correlation between the difference in MPD and transverse pedicle orientation (r2 = 0.21, p \ 0.001). As transverse orientation increased, minimal coronal diameter got larger than minimal transverse diameter. Conclusion: The determination of the MPD is more difficult than previously described, attributable to the existence the coronal pedicle angle. The angulation of the pedicle in three dimensions causes the radiographic measurements to overestimate true pedicle diameter in both transverse and coronal images. MPD measurements are most accurate when taken from coronal images at midthoracic levels and from transverse images at lumbar and upper thoracic levels. In severe deformities where abnormal pedicle morphology is suspected, the most accurate approach to determine the MPD is measurement in both transverse and coronal images. Accurate determination of MPD using imaging techniques is only possible when the coronal and transverse pedicle orientation is small. Disclosures: author 1: none; author 2: grants/research support; Company=Stryker, Medtronic, Stryker, vti; author 3: grants/research support; Company=Depuy Spine; author 4: none; author 5: none.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 P97 MINIMALLY INVASIVE TRANSFORAMINAL LUMBAR INTERBODY FUSION [MI-TLIF]: A TAILOR MADE SOLUTION FOR SELECTED FAILED BACK Shashidhar Bangalore Kantharajanna, Arvind Gopalrao Kulkarni Bombay Hospital and Medical Research Center, Mumbai, India Purpose of the study: To study the outcomes of MI-TLIF in selective failed lumbar spine surgeries in comparison to primary MI-TLIFs. Methods: In a study period that from January 2011 to May 2012, 60 consecutive MI-TLIFs [20 failed back and 40 primary] were retrospectively reviewed, to elucidate differences in perioperative data including operative time, estimated blood loss, complications, Oswestry Disability Index [ODI] and Visual Analogue Scale [VAS] for back and leg pain before surgery and at last follow. The results were analyzed using the SPSS software [version20]. Results: The mean-age was 50.61 years, with M: F: 1.8:1 in failedbacks and 49.52 years with M: F: 1.7:1 in primary surgeries. Nineteen revision surgery cases were compared with 36 primary surgeries. One failed back and 4 primary surgery patients were excluded during analysis, due to non-availability of complete data. The mean follow up was 28 months and 24 months in primary and failed back groups respectively. The mean pre-operative and post-operative ODI were 53.18 and 20.23 in primary and 52.01 and 25.72 in failed-back MITLIFs [P value 0.304]. The pre-op and post-op VAS for back pain were 4.77 and 1.75 in primary and 4.1 and 2.0 in failed-back MITLIFs. The pre-op and post op VAS for leg pain were 6.52 and 1.27 in primary and 9.5 and 1.375 in failed-back MI-TLIFs. The percentage change in VAS for back pain was found to be statistically significant [P-0.027] whereas the percentage change for VAS for leg pain was not found to reach statistical significance [P value- 0.538]. There was no statistical significant difference [P-0.406] in terms of time for surgery [3.4 and 3.2 hrs.] and estimated blood loss [110 and 100 ml]. No complications were seen in both groups. Conclusions: The inherent technique of MI-TLIF is tailor made for failed backs, since it exploits the intact para-median corridor, which is the gateway for TLIF. The authors recommend MI-TLIF in selected failed-back surgery. Disclosures: author 1: none; author 2: none.

P98 SURGICAL OUTCOMES OF SPINAL SURGERY FOR PATIENTS WITH LONG-TERM HEMODIALYSIS Shunichi Toki, Masahiro Kashima, Takefumi Nakagawa, Masaru Nakamura, Shunji Nakano, Takashi Chikawa Tokushima Municipal Hospital, Tokushima, Japan Introduction: Many patients treated with long-term hemodialysis (HD) suffer from long-term HD related spinal disorder. Some reports mentioned that there are two groups those with radiological changes of destructive spondyloarthropathy (DSA) and those without the changes of DSA. The purpose of this study is to review clinical outcomes, including survival rate and to discuss the benefit of surgical treatments for patients treated with long-term HD in both two groups. Materials and methods: We retrospectively reviewed 81 long-term HD patients who underwent 44 cervical, 4 thoracic and 33 lumbar spinal surgeries. According to the radiological findings, we divided them into following two groups; DSA group (with radiological changes of DSA) and non-DSA group (without radiological changes

S783 of DSA). There were 45 patients of DSA group and 36 patients of non-DSA group. Age, gender, affected lesion, pre JOA score are similar in both groups. Decompression surgery was performed for the non-DSA patients, whereas decompression and spinal fusion was performed for DSA patients. Duration of HD, operative time, estimated blood loss (EBL), improvement ratio of Japanese Orthopaedic Association (JOA) score, deep surgical site infection (SSI) rate and survival rate. Results: Duration of HD of DSA group was significantly longer than non-DSA group. The operation time was significantly longer and EBL was significantly larger in the DSA group. Improvement ratio of JOA score was 46.4 % in DSA group and 51.5 % in non-DSA group. Five patients in DSA group had SSI. No complication such as SSI, dural tear and neural involvement was observed in non-DSA group. Thirteen patients consisted of 8 patients in DSA group and 5 patients in non-DSA group died during follow-up. Two patients in DSA group died because of cardiac insufficiency within 2 months after the surgery. Remaining 11 cases died because of HD-related complications between 10 and 49 months after the surgery. Kaplan-Meier two-year survivor rate was around 85 % in both groups. The final survivor rate of DSA group and non-DSA group were 79.5 % and 57.9 %, respectively. Conclusions: Spinal surgeries for patients with HD related spinal disorder obtained neurological and functional improvement. The survival rate of two groups was not significantly different. Thus, even for the DSA patients, surgical reconstruction should be considered, if need. Disclosures: author 1: none; author 2: employee; Company=Tokushima Municipal Hospital; author 3: none; author 4: none.

P99 PATTERNS OF LUMBAR PAIN: DOES REALLY EXIST ANY DIFFERENCE BETWEEN DE MUSCULAR, THE FACET SYNDROME AND NEUROLOGIC PATTERNS? A COST SUBANALYSIS OF THE 6 DIFFERENT PATTERNS OF LUMBAR PAIN GROUPED BY THREE RETROSPECTIVE COHORTS OF 1251 PATIENTS Jaime Diaz de Atauri Bosch, Oscar Zabalza Mantilla, Mikel Ayala Garcı´a Clinica Ercilla. Mutualia Bizkaia; Hospital San Jose. Mutualia Araba, Bilbao, Spain Objective: We present a cost analysis study of three different groups of lumbar pain in primary care in a working population by grouping patients according to different pain patterns. The aim is to assess the cost and effectiveness of the three groups, whether differences exist between them and which one can generate more spending Material and methods: A three retrospective cohorts study of patients treated for lumbar pain at our Worker’s Compensation insurance clinic in 2014. The first group was treated as muscular pattern (G1), the second group was treated as facet syndrome (G2) and the third group was the neurologic pattern (G3). Diagnosis, number of sick days and mean duration, sick leave indication, number of additional tests, hospital admissions, number of medical visits, surgeries, referrals to physical therapy (duration and type of therapy) and all their costs were studied. A statistical analysis was performed using SPSS, a Kolmogorov-Smirnov test was done and then a Pearson‘s chi-squared test, a Fisher’s exact test and a Kruskal-Wallis test were done with a sensitivity of 95 %. RESULTS We reviewed 1251 patients: 900(G1), 159(G2) and 192(G3). The sick leave mean was 12,16(G1), 12,31(G2) and 26,58(G3); the sick leave indication was

123

S784 54,56 %(G1), 76,73 %(G2) and 72,40 %(G3); total sick leave/day cost per patient was 378,42€(G1), 538,73€(G2) and 1097,53€(G3); the hospital admission was 0,56 %(G1), 0,63 %(G2) and 4,17 %(G3); the hospital stay/day cost was 2113,68€(G1) and 176,14 €(G2) and 7574,02€(G3); the surgical procedures were 4(G1),0(G2) and 9(G3), with a cost of 4005,82€(G1),0€(G2) and 11474,56€(G3), the complementary tests mean was 1,11(G1),1,61(G2) and 1,83(G3), a cost of 35204,89€(G1),11471,09€(G2) and 22616,26€(G3); the mean number of medical visits was 9,88(G1),12(G2) and 20,01(G3), a cost of 341932,66€(G1),76823,65€(G2) and 108627,26€(G3); referral to physical therapy was 8,67 %(G1), 12,58 %(G2) and 15,1 %(G3); the mean of treatments was 46,73(G1),28,47(G2) and 63,14(G3) with a cost of 37924,74€(G1), 5532,12€(G2) and 19488,15€(G3). These variables had a statistically significant difference (P\0.01). The total cost per patient 869,13€(G1) and 1158,8€(G2) and 2032,02€(G3). Conclusions: We find differences in both cost and studied variables between the 3 groups being these differences between G1 and G3 statistically significant with a clinical correlation. However in G2, data are more scattered, not reflecting what was expected either because they were not assigned to the right pattern or because they hadn’t been assigned properly. Disclosures: author 1: none; author 2: none; author 3: none.

P100 INTRAOPERATIVE NEUROMONITORING (IONM) MAY NOT PREDICT ALL POSTOPERATIVE NEUROLOGICAL ADVERSE EVENTS IN ADULT DEFORMITY SURGERY. A SCOLI-RISK-1 TRIAL SUBANALYSIS Ferran Pellise´, Montse Domingo-Sa`bat, Dulce Moncho, Kimia Rahnama, Michael Fehlings, Lawrence G. Lenke, Christopher I. Shaffrey, Branko kopjar, Kenneth Cheung, L. Carreon, F. Schwab, O. Boachie-Adjei, K. Kebaish, C. Ames, Y. Qiu, Y. Matsuyama, B. Dahl, H. Mehdian, S. Lewis, S. Berven Spine Unit, Hospital Valle Hebron, Vall Hebron Institut de Recerca; University of Toronto; Washington University School of Medicine; University of Virginia Health System, University of Washington; The University of Hong Kong; Norton Leatherman Spine Center, NYUHospital for Joint Diseases, Hospital for Special Surgery, Johns Hopkins University, University of California San Francisco; NanJing University Medical School, Hamamatsu University School of Medicine; Copenhagen University Hospital, Rigshospitalet; Center for Spinal Studies and Surgery, University Health Network, University of California-San Francisco, USA Introduction: IONM is routinely used to detect intraoperative neurological adverse events (NAE). The Scoli-Risk-1 trial identified a lower extremity motor score (LEMS) decline at discharge in 23 % of patients following complex Adult Spinal Deformity (ASD) surgery. Some studies have reported NAE not detected by IONM following ASD surgery. The purpose of this study is to assess the association between unexpected NAE at the time of surgery and IONM in ASD surgery. Methods: Retrospective analysis of prospectively collected ScoliRisk-1 data (15 centers from 3 continents). All IONM changes were reported if: SSEPs amplitude drop [50 % and/or increased latency [10 %; MEPs amplitude drop [80 % and/or increase in latency [10 %. NAE was only reported in case of postoperative neurological deterioration. A Clinical End Point Committee (CEC) evaluated all reported NAEs in order to obtain uniform judgments. NAEs were classified by the surgeons as expected or unexpected based on intraoperative findings and IONM.

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Results: 273 patients underwent 328 surgeries. IONM was used in 299 surgeries (91 %) on 267 (98 %) patients: SSEPs in 293, MEPs in 281 and EMG in 237. A total of 25 surgeries on 25 patients had IONM (SSEPs and/or MEPs) changes. There were 36 intra-operative NAEs in 35 surgeries; 16 (44 %) of which were unexpected NAE. In the 29 surgeries were IONM was not used, no NAE was reported. In the 54 surgeries that had SSEPs+MEP, 10 had NAE, and 6 were unexpected. In the 227 surgeries that had SSEPs+MEPs+EMG, 24 had 25 NAEs, and 9 were unexpected. In patients with 3-column osteotomies, 82 % of unexpected NAEs occurred when osteotomy was performed below L1. Conclusion: In patients undergoing complex ASD surgery, intraoperative findings including IONM do not accurately identify all patients sustaining intra-operative NAE (specially at cauda equine level), as more than 40 % of postoperative NAE were unexpected. Disclosures: author 1: grants/research support; Company=DePuy Synthes, K2M, consultant; Company=DePuy Synthes, Biomet; author 2: grants/research support; Company=DePuy-synthes; author 3: employee; Company=Vall d’Hebron University Hospital; author 4: employee; Company=Vall d’Hebron University Hospital; author 5: none; author 6: no indication; author 7: grants/research support; Company=AO, NREF, NACTN, consultant; Company=Biomet, Medtronic, Nuvasive, Stryker, stock/shareholder; Company=Nuvasive, royalties; Company=Biomet, Medronic, Nuvasive; author 8: none; author 9: grants/research support; Company=Ellipse technologies; author 10: grants/research support; Company=Norton Healthcare, AO Spine, OREF, SRS, other financial report; Company=OREF, NIH, University of Louisville, Center for Spine Surgery and Research, Region of Southern Denmark, Children’s Tumor Foundation, Nuvasive, Medtronic, employee; Company=Norton Healthcare.

P101 ADVERSE EVENTS IN ADULT SPINAL DEFORMITY PROCEDURES Evalina L Burger, Michael S. Chang, Sean Molloy, Vincent Fie`re University of Colorado SOM, US; Sonoran Spine, US; NHS Royal National Orthopaedic Hospital UK; Centre Orthope´dique Santy, France Introduction: Surgical correction of ASD is usually associated with a high complications rate (up to 40 %) and a revision rate up to 25 %. This study investigates the relationship between surgical strategies and early adverse events (AE) in a retrospective review of multicenter prospective database. Materials and methods: 72 patients requiring posterior surgical correction for adult degenerative scoliosis or kyphosis between Sept 2012 and March 2014 were included in 4 centers and followed for 2 years. 62 patients reaching 6 months follow-up are eligible for analysis. Demographics, surgical metrics (OR time, blood loss, grafts, inter-body fusion (IBF), UIV, LIV), complications, revision surgery, as well as baseline and 3-6 month post-operative radiographs and HRQL were collected. Patients were categorized into the AE group (n = 22): those who experienced peri/post-operative complication and/or revision; and the no AE group (n = 40) with no peri/postoperative AE or revision. Chi-square and unpaired t-test analysis with a 0.05 level of significance were used for group comparisons. Results: Groups were not different at baseline for demographics, HRQL, sagittal radiographic parameters as the number of levels fused, use of osteotomy or autograft. The AE group was more likely to receive BMP (77.2 % vs 22.7 %, p = 0.018) and less likely to

Eur Spine J (2015) 24 (Suppl 6):S743–S800 receive allograft (41 % vs. 59 %, p = 0.009) or DBM (22.7 % vs. 77.2 %, p = 0.001). There was no significant difference in percentage of patients with IBF (AE 77 % vs. 85 %). There was, however, significant difference in terms of approach (AE: 40.9 % ALIF, no AE: 70 % DLIF). The AE group was less likely to have a fusion extended to the pelvis (22.7 % vs. 51.2 %, p = 0.029). This group also had longer OR time (374 vs. 257 mn p = 0.003) and length of stay (21.4 vs. 8.6 days, p = 0.023), without difference in EBL. Post-operatively, there was no significant difference in HRQL. Patients in the AE group were less likely to reach an SVA of \50 mm (64 % vs 100 %, p \ 0.001). Conclusions: This study demonstrates that surgical metrics, particularly the fusion approach, length of instrumentation and achievement of a balanced spine significantly influence adverse event occurrence. Disclosures: author 1: grants/research support; Company=Aesculap, consultant; Company=Medicrea, Paradigm, Signus; author 2: grants/ research support; Company=Medicrea, consultant; Company=Medtronic, Stryker; author 3: grants/research support; Company=Metronic, consultant; Company=Zimmer, K2M, royalties; Company=Medicrea; author 4: consultant; Company=Medicrea Clariance, royalties; Company=Medicrea Clariance.

S785 dural tear as an intra-operative complication (0.06 %). None of the patients had an intra-operative or early postoperative neurologic complication. There were no listhesis nor vertebral collapse observed during intraoperative osteotomy correction, nor at the first post-operative follow-up. Conclusion: In this preliminary study, the new osteotomy reduction plier proved efficient to achieve the preoperatively calculated vertebral correction angle when performing pedicle subtraction osteotomies. Further prospective studies are needed to confirm these results. Disclosures: author 1: consultant; Company=Medtronic; author 2: none; author 3: none; author 4: grants/research support; Company=Medtronic, Globus Medical, K2M, consultant; Company=Medtronic, Globus Medical, K2M.

P103 PREOPERATIVE ASSESSMENT OF URETER USING DUALPHASE ENHANCED 3D-CT FOR LATERAL LUMBAR INTERBODY FUSION Shunsuke Fujiayashi, Bungo Otsuki, Hiroaki Kimura, Shuichi Matsuda

P102 OPTIMIZING PEDICLE SUBTRACTION OSTEOTOMY TECHNIQUES: A NEW REDUCTION PLIER TO INCREASE TECHNICAL SAFETY AND ANGULAR REDUCTION EFFICIENCY Antonio Faundez, Jean Charles Le Huec, Lars Hansen, Martin Gehrchen Division of Orthopaedic Surgery, Geneva University Hospitals, Switzerland; Spine Unit 2, University Victor Segalen, Bordeaux, France; Department of Orthopaedic Surgery, National University Hospital of Copenhagen, Denmark Purpose: To present a new reduction plier for pedicle subtraction osteotomies of the thoraco-lumbar spine. Methods: A consecutive series of 17 patients has been treated at three different European University Hospitals, by two surgical teams. All the patients underwent a pedicle subtraction osteotomy (PSO) of the lumbar spine, to treat major sagittal imbalance, due to physiologic degeneration or post-fusion major sagittal imbalance. The amount of vertebral angular correction needed was calculated using the Full Balance Integrated (FBI) method. A special reduction plier, that allows to safely control the angular correction, was used intraoperatively to close the osteotomy gap. The intraoperative correction was verified on C-arm images. Preoperative and early postoperative (during hospital stay) global sagittal balance parameters were compared, to verify adequacy between the amount of correction aimed for and the final result. The instrument is secured on two pedicle screws above and two below the PSO vertebra. The main advantages of the instrument are: safe, precise and efficient reduction, made possible through 3 articulations, which allow obtaining a rotation of the pedicle screws close to the osteotomy line, thus avoiding collapse of the osteotomized vertebra and lack of correction, complications usually seen with the conventional reduction technique. Results: The mean preoperative calculated osteotomy correction angle (FBI) was 33.8°, the mean post-operative correction angle obtained was 32.1°. Global lumbar lordosis (LL) was statistically greater than pre-operatively (55.8 degrees versus 19.4 degrees, p \ 0.0001). The global sagittal balance was improved, as shown by the increase of the spino-sacral angle (SSA) from 122° average pre-operatively, to 128° post-operatively (p = 0.0547). One patient had a

Department of Orthopaedic Surgery, Graduate School of Medicine, Kyoto University; Kyoto, Japan Introduction: Minimally invasive lateral lumbar interbody fusion (LLIF), such as XLIF (extreme lateral interbody fusion) and OLIF (oblique lateral interbody fusion), are increasing in popularity among spine surgeons. Although these surgical corridors are a little different, both are close to retroperitoneal important structures. Recently, direct injury to the ureter and kidney has been described as access related surgical complications of XLIF. These injuries are one of the devastating complications that should be concerned during LLIF. In general, surgeons have a low threshold of suspicion of these injuries. In the current study, we introduce dual-phase CT and assess the location of ureteral and vascular structures simultaneously, and discuss the risk of these injuries during both procedures. Materials and methods: 12 consecutive OLIF cases were enrolled in this study. Preoperatively, intravenous contrast material was administered with a split-bolus technique before CT scanning to enhance ureters and abdominal vessels simultaneously. The location of ureter was evaluated bilaterally at L2/3, 3/4, 4/5 levels using axial and 3D images. The location of ureter was divided into six parts as reference to psoas muscle (PM) and vertebral body (VB). Total 72 ureters were evaluated and compared with intraoperative findings. Results: 54 ureters were successfully evaluated radiologically, excluding 5 unclearly enhanced and 13 located in kidney. In reference to PM, 92.5 % was located at anterior to PM. In reference to VB, 59.2 % was located at anterior to VB and remaining 40.7 % was located at lateral to VB. At L2/3 level, 54.1 % of kidneys were located at lateral to VB. Intraoperative findings indicated that all ureters were successfully retracted anteriorly with peritoneal contents. Discussion: Usually, OLIF corridor is created at anterior border of PM, whereas XLIF corridor is created at midportion of VB. Results of this study indicated that during the procedure of OLIF, ureter is likely to be closest to the surgical corridor. On the other hand, during the procedure of XLIF, ureter may be close to the surgical corridor. Surgeons should keep in mind that complete retraction of peritoneal content during both procedures is mandatory and will avoid ureteral injury. Conclusion: Preoperative dual-phase enhanced CT will provide important information about retroperitoneal structures for LLIF. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

123

S786 P104 THE ADVANTAGEOUS EFFECT OF IODINE-COATED IMPLANTS IN PYOGENIC SPONDYLITIS Moriyuki Fujii, Hideki Murakami, Satoru Demura, Satoshi Kato, Katsuhito Yoshioka, Takashi Igarashi, Noritaka Yonezawa, Hiroyuki Tsuchiya Department of Orthop. Surgery Kanazawa University, Kanazawa, Japan Background and purpose: Treatment for pyogenic spondylitis is generally conservative; surgery with spinal instrumentation may be necessary in cases with neurological symptoms, bone destruction, and/ or instability. Conventionally, spinal instrumentation of the infected spine has been contraindicated due to the susceptibility of metal implants to biofilm formation and its adverse effect on antibiotic therapy. We have developed the technology to impart antibacterial properties to titanium implants by applying an iodine (T2) coating. The T2-supported implants have been used prophylactically and in active infections, making it possible to shorten the time to fusion. We report the one-stage treatment results of 11 cases of active pyogenic spondylitis using spinal stabilization with T2-supported implants. Materials: Eleven consecutive patients diagnosed with pyogenic spondylitis who underwent debridement and spinal instrumentation using iodine-supported implants from November 2010 to 2013, were prospectively followed. There were three males and eight females with a mean age of 72.4 years (range 57-85 years). The mean followup period was 21.7 months (10-38 months). The interval between the onset of symptoms and surgery averaged 14.1 weeks (range 3-45 weeks). Surgery was performed after conservative treatment was deemed unsuccessful. The isolated microorganisms included methicillin-sensitive Staphylococcus aureus in two patients, E. coli in two, methicillin-resistant S. aureus, Pseudomonas aeruginosa and Streptococcus pneumoniae in one each, and unknown organisms in four. The method of fixation was anterior in six cases, posterior in four, and anterior/posterior in one. The extent of instrumentation was 1-level in four cases, 2-levels in four and C3-levels in three. Result: In 10 of 11 cases, there were no incidences of recurrence after one-stage surgery using T2-supported implants. One case of anterior fixation required additional posterior fixation for screw loosening; after posterior stabilization, the patient recovered unremarkably. Conclusion: We demonstrated the efficacy of T2-supported titanium implants in conjunction with thorough debridement and instrumentation in the one-stage treatment of pyogenic spondylitis. Despite the conventional contraindication to using spinal instrumentation for stabilization in active spinal infections, we believe the use of T2supported implants to be advantageous in the one-stage treatment of pyogenic spondylitis. Disclosures: author 1: none; author 2: none; author 3: no indication; author 4: none; author 5: none; author 6: none; author 7: none; author 8: none.

P105 ARTIFICIAL NEURAL NETWORKS AS A TOOL FOR THE AUTOMATED ANALYSIS OF MEDICAL IMAGES OF THE SPINE Fabio Galbusera, Tito Bassani, Marco Brayda-Bruno, Francesco Costa, Guglielmo Cannella, Alberto Zerbi, Hans-Joachim Wilke IRCCS Istituto Ortopedico Galeazzi, Milan, Italy; Ulm University, Ulm, Germany

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Introduction: Artificial neural networks are biologically inspired tools to perform nonlinear classification and regression, and are currently used for the automatization of many tasks in which a robust, fault tolerance performance is required. In this work, the feasibility of using neural networks for various tasks involving spinal imaging (i.e. vertebral landmark recognition, grading of disc degeneration and identification of endplate defects) is evaluated. Materials and methods: Radiographic databases of lumbar vertebrae were created by collecting images obtained with the EOS Imaging System in anteroposterior and lateral projections of 50 young patients suffering from mild adolescent idiopathic scoliosis. Landmarks at the projected corners of the vertebral endplates and in the pedicles were manually identified in all images by a trained operator. The annotated images were used to train and test artificial neural networks in the automatic recognition of such landmarks. In parallel, MRI scans of 300 subjects suffering from low back pain were retrospectively collected. For all images, a radiologist performed the Pfirrmann grading of all lumbar intervertebral discs and the identification of endplate defects such as notches and Schmorl’s nodes. Neural networks were then trained to automatically determine the degree of disc degeneration and the possible presence of endplate abnormalities. Results: The neural networks used for the landmark recognition demonstrated a solid performance with respect to semi-automatic sterEOS software (EOS Imaging, Paris, France), which was extensively validated in previous literature, although with some exceptions. The predicted positions of the vertebral landmarks were satisfactory by visual inspection. Most problematic identifications were related to the recognition of the L5 landmarks in the sagittal projection. Similarly, high accuracy ([85 %) was obtained in the grading of disc degeneration and in the identification of endplate defects. Conclusions: Artificial neural networks seem to represent a viable tool for the automatic analysis of medical images of the spine, if training sets of good quality and size are available. Disclosures: author 1: grants/research support; Company=Italian Ministry of Health (GR-2011-02351464), employee; Company=IRCCS Istituto Ortopedico Galeazzi; author 2: grants/research support; Company=Italian Ministry of Health (GR-2011-02351464); author 3: grants/research support; Company=Medtronic - K2M, consultant; Company=Depuy Spine - Implanet – Spineguard; author 4: none; author 5: no indication; author 6: none; author 7: none.

P106 AUGMENTED AND VIRTUAL REALITY APPLICATIONS TO SHOW COMPLEX SPINE PATHOLOGIES: FIRST VERSION Marcelo Galvez, Gonzalo Rojas, Jorge Fuentes, Takeshi Asahi, Carlos Montoya, Aaron Vidal, William Currie, Andres Chahin Department of Radiology, Clı´nica las Condes, Santiago, Chile Visualization of complex human body structures such as the spine and his malformations such as herniated disk, foraminal stenosis, and spine stenosis is a challenge. Augmented Reality is the combination of a real scene supplemented (or augmented) by computer generated information, such as image, video, sound, etc. Android and iOS based mobile devices such as smartphones and tablets are being increasingly massive adoption and with higher performance characteristics. Sagital 2D T2w TSE sequence MRI (TR 4420 ms, TE 130 ms, FoV 320 9 320, Matrix 448 9 448, Voxel Size 0.7 9 0.7 9 3 mm) and sagital CT (kVp 140, 1 mm slice thickness, 0,30 9 0,30 mm pixel spacing, B31s kernel) images were scanned in a 57 years old male patient. Both images were co-registered using ‘‘Linear registration’’

Eur Spine J (2015) 24 (Suppl 6):S743–S800 algorithm of 3D Slicer v 3.6 software. CT image was segmented using threshold option (113, 1303 Hounsfield units) of 3D Slicer and then L4-L5 vertebral bodies (VB) was extracted manually using ImageJ v1.49n version. The L4-L5 intervertebral disk (ID) was segmented using ‘‘simple region growing’’ algorithm of 3D Slicer first, and then was cleaned manually using ImageJ software. Mesh of L4-L5 and ID was generated using ‘‘Model Maker’’ algorithm of 3D Slicer with default options, and HC Laplacian smoothing of MeshLab software. The iOS-Android application were created using C# language and software tools: Unity 4.5x, Mono develop, Blender, Gimp, Qualcomm Vuforia for Unity Android and iOS. By focusing the printed target image using iOS-Android app, the 3D spine of the patient was shown with two VB and ID. The user can rotate the spine using the onscreen controls, by rotating the view angle over the target image, or rotating the target image itself. The size of the spine can be changed by modifying the distance between the mobile device and the target image. Virtual reality demonstrated new views from inside spine and relationship of normal structures and of different pathologies. The application described here demonstrates the fusion of augmented and virtual reality techniques with medical images of patients with spine pathologies. The application shows how to use ubiquitous mobile devices to display advanced medical information in a 3D interactive application. The application is useful to show, explain and describe in a 3D and interactive way the spine pathology that the patient has. More work was done to create an automatic CT-MRI registration and segmentation of VB and ID. Disclosures: author 1: grants/research support; Company=Corfo 14IAET-28677; author 2: grants/research support; Company=Corfo 14IAET-28677; author 3: none; author 4: none; author 5: none; author 6: no indication; author 7: none; author 8: none.

P107 MORBIDITY AND MORTALITY AND SURGICAL STRATEGY IN PATIENTS WITH ANKYLOSING SPONDYLITIS SPINE FRACTURES Jens Gempt, Elisabeth To¨ro¨k, Martin Vazan, Thomas Huber, Bernhard Meyer, Yu-Mi Ryang 1

Neurochirurgische Klinik und Poliklinik; 2Abteilung fu¨r Neuroradiologie, Klinikum rechts der Isar, Technische Universita¨t Mu¨nchen, Germny Objective: Spine fractures in patients with ankylosing spondylitis can often occur after minor trauma. Moribund patients and complex surgical procedures make the treatment of these patients a challenging task. Methods: We reviewed our clinical database for patients with ankylosing spondylitis who underwent surgical treatment for spine fractures between 03/2008 and 8/2014. Hospital-/ICU-stay, co-morbidities, peri-/postoperative complications and surgical strategy were assessed. Results: 31 pts (mean age 74 yrs, range 46 yrs - 90 yrs) were operated for traumatic Bechterew fractures (24 cervical, 6 thoracic, 1 lumbar). Mean hospital stay was 27 d (6 d - 70 d). 21 pts needed ICU treatment (mean stay 20 d, range 1 d - 63 d) due to medical deterioration. 20/31 pts received a 360° fusion with posterior stabilization +/laminectomy and ventral corpectomy/discectomy. 10 pts were treated by dorsal approach, and 1 pt by ventral approach only. Preexisting poor medical condition and clinical deterioration obviated a secondary procedure.

S787 Neurological status of our pts according to ASIA classification was preoperatively: 4 A, 3 C, 6 D, 16 E, 2 n/a, and postoperatively 2 A, 4 C, 8D, 14E, 3 n/a. Revision surgeries were necessary in 5/31 pts. 3/31 pts died during hospital stay from severe medical complications. Median survival of all pts was 1139 d (+/- 428,6) with 13 pts already deceased during follow-up. Conclusion: Morbidity-rate is very high in patient with ankylosing spondylitis spine fractures with long hospital and ICU-stay and high medical expenses. Overall survival in these patients is similar to patients harboring malign tumor-associated diseases. Surgery-associated complications were comparable rare. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: grants/research support; Company=MDT DPY ulrich, consultant; Company=MDT DPY ulrich; author 6: consultant; Company=BrainLAB.

P108 INFLAMMATORY BLOOD LABORATORY LEVELS AS MARKERS OF DEGENERATIVE LUMBAR SCOLIOSIS SURGERY INFECTION: A SYSTEMATIC REVIEW AND META-ANALYSIS Dingjun Hao, En Xie Department of Spine Surgery, Hong Hui Hospital, Xi’an Jiaotong University College of Medicine, Xi’an, China Background: The preoperative diagnosis of Degenerative Lumbar Scoliosis surgery infection in patients may rely in part on the use of systemic inflammation markers. These markers have unclear accuracy. The objective of this review was tantamount to summarize the evidence on the accuracy of the peripheral white blood-cell count, the erythrocyte sedimentation rate, serum C-reactive protein levels, and serum interleukin-6 levels for the diagnosis of Degenerative Lumbar Scoliosis surgery infection. Methods: We searched electronic databases (MEDLINE, EMBASE, Cochrane Library, Web of Science, and Scopus) from 1950 through 2014. Eligible studies evaluated the accuracy of the white blood-cell count, erythrocyte sedimentation rate, serum C-reactive protein level, and serum interleukin-6 level for the interpretative diagnosis of Degenerative Lumbar Scoliosis surgery infection at the time of revision arthroplasty. Two reviewers, working independently extracted study characteristics and data to estimate the diagnostic odds ratio and 95 % confidence intervals for each result. Results: We included thirty eligible studies that included 6717 revision total Degenerative Lumbar Scoliosis surgical sites. The prevalence of Degenerative Lumbar Scoliosis surgery infection was 2.64 % (177 of 6717). The accuracy of assessed inflammation markers, represented with a diagnostic odds ratio, was 271.7 (95 % confidence interval, 113.0 to 876.8) for interleukin-6 (3 studies), 11.7 (95 % confidence interval, 7.9 to 21.7) for C-reactive protein level (17 studies), 7.6 (95 % confidence interval, 3.7 to 10.7) for erythrocyte sedimentation rate (27 studies), and 4.7(95 % confidence interval, 2.7 to 7.7) for white blood-cell count (18 studies). Conclusions: The diagnostic accuracy for Degenerative Lumbar Scoliosis surgery infection was best for interleukin-6, monitored by C-reactive protein level, erythrocyte sedimentation rate, and white blood-cell count. Have regard to the limited numbers of studies assessing interleukin-6 levels, further investigations assessing the accuracy of interleukin-6 for the diagnosis of Degenerative Lumbar Scoliosis surgery infection are warranted. Disclosures: author 1: none; author 2: none; author 3.

123

S788 P109 C2 NERVE DYSFUNCTION ASSOCIATED WITH POSTERIOR UPPER CERVICAL SPINE SURGERY: A CASE REPORT OF 21 AND A LITERATURE REVIEW Da-Geng Huang, Ding-Jun Hao, Bao-Rong He, Tuan-Jiang Liu, Xiao-Dong Wang, Qi-Ning Wu, Hua Guo Department of Spine Surgery, Honghui Hospital, Xi’an Jiaotong University Health Science Center; Xi’an, China Introduction: C2 nerve dysfunction induced by C1 lateral mass screw insertion has been reported in several literatures. However, we found that postoperative C2 nerve dysfunction could happen in others patients who underwent C1 pedicle screw insertion instead of C1 lateral mass screw insertion or even occipitocervical fixation with no screw placement in C1. To the best of our knowledge, there is no literature to discuss postoperative C2 nerve dysfunction associated with all posterior upper cervical spine surgery yet. Purpose: To report 21 cases of postoperative C2 nerve dysfunction associated with posterior upper cervical spine surgery, review related literatures and discuss the causes and prevention of this complication. Methods: Twenty-one patients who developed C2 nerve dysfunction after posterior upper cervical spine surgery were analysed. The severity of C2 nerve dysfunction was assessed with a visual analog scale (VAS) questionnaire. The natural history of this complication was analysis. The causes and prevention were discussed. Related literatures were reviewed as well. Results: Of the 21 patients, 15 patients underwent bilateral C1 lateral mass screw insertion (C2 nerve impingement was confirmed by postoperative CT images in 13 of the 15 patients), 3 underwent bilateral C1 pedicle screw insertion with posterior autograft fusion, 1 underwent bilateral C1 pedicle screw insertion with posterior structural autograft fusion, 1 underwent occipitocervical fixation with no screw placement in C1, 1 underwent a debridement and had C1 and C2 pedicle screws and rods removed due to infection. The initial VAS score ranged from 3 to 10, and dropped down slowly with the time being. At the final follow-up, the VAS score ranged from 0 to 5. Conclusions: Postoperative C2 nerve dysfunction can happen not only in patients with C1 lateral mass screw insertion, but also some patients who underwent posterior upper cervical spine surgery without C1 screw insertion. The causes of this complication are complicated, the potential causes includes traction during surgery, irritation or impingement from C1 lateral mass screw, irritation or impingement from morcellized autograft, accidental injury during incision, suture and debridement. Prevention of this complication is possible with understanding of the potential causes. The severity of postoperative C2 nerve dysfunction varies in different patients. However, in most patients, the symptom can relieve itself with the time being. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

P110 COMPARISON BETWEEN ANTERIOR AND POSTERIOR APPROACH OF DEBRIDEMENT AND INSTRUMENTED FUSION FOR TREATING LUMBAR SPINAL BRUCELLOSIS Jun-Song Yang, Ding-Jun Hao, Li-Min He, Tuan-Jiang Liu, Yuan-Ting Zhao Department of Spinal Surgery, Hong-Hui Hospital, Medical College of Xi’an Jiaotong University, Xi’an, China

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Purpose: To compare the outcome of patients with symptomatic lumbar spinal brucellosis treated with debridement and instrumented fusion using the anterior approach to those treated with the posterior approach. Method: From March 2001 to March 2012, a total of 30 consecutive patients with symptomatic lumbar spinal brucellosis who underwent debridement and instrumented fusion using anterior approach (15 patients in group A) or posterior approach (15 patients in group B) were enrolled. The average follow-up time is 18 months. We compared the operative time, total blood loss, erythrocyte sedimentation rate, fusion time of allograft and length of total hospital stay between the anterior and posterior approaches. Oswestry Disability Index (ODI) and visual analog scale (VAS) were also applied for the evaluation of clinic outcome. Results: Mean operative time in group A was 4 h 10 min versus 3 h 50 min in group B (P [ 0.05). More mean blood loss (600 ml vs. 400 ml, P \ 0.05) and longer mean hospital stay (11.9 days vs. 7.5 days, P \ 0.05) were observed in group A. The average bone healing time was 4.1+0.6 months in group A and 4.5+0.8 months in group B. Erythrocyte sedimentation rates (ESR) returned from 37.6 mm/h (group A) and 35.7 mm/h (group B), respectively, to normal levels 8 to 12 weeks postoperatively. When considering the ODI and VAS, postoperatively, the clinical outcomes of the two approaches were significantly improved, but the differences between the two approaches were not significant. Five surgery-related complications were observed among all enrolled patients. Conclusions: In our study, there is no significant difference in the clinical outcomes between the two approaches. When considering the operative time, total blood loss, and length of total hospital stay, posterior approach may be preferable. However, a relative long fusion time of allograft is needed to be considered. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

P111 THE CLINICAL VALIDITY OF DEGENERATIVE DISC DEGENERATION ON MAGNETIC RESONANCE IMAGING FOR THE LUMBAR RELATED SYMPTOMS Masahiko Kanamori, Taketoshi Yasuda, Kayo Suzuki, Kazuo Ohmori University of Toyama, Toyama City, Japan Study design: Retrospective comparative study. Purpose: Low back pain and sciatica are not always reflected on the magnetic resonance imaging (MRI) features. Therefore we want to clarify the clinical validities of new grading system for the lumbar disc degeneration. Overview of the literature: Various kinds of MRI classification criteria have been widely used as an evaluation for the disc pathology. However, still much controversy exists regarding the factors that affect the accurate diagnosis on MRI. Methods: The evaluation data were collected from the 570 lumbar discs in 114 patients. A new grading system on T2-weighted MRI was developed, and assessed the clinical validities of the lumbar disc degeneration. The symptom was categorized according to the Japanese Orthopaedic Association low back pain score (JOA score). Intraobserver reproducibility and interobserver reliability of a new grading system were compared with Schneiderman’s criteria which are commonly used. Results: Disc degeneration can be graded reliably on routine T2weighted MRI using two evaluation systems, by two blinded observers, repeatedly. There were 197 Grade I, 103 Grade II, 234 Grade III,

Eur Spine J (2015) 24 (Suppl 6):S743–S800 and 36 Grade IV discs (evaluated by the first author: M.K.) and there were 192 Grade I, 137 Grade II, 212 Grade III, and 29 Grade IV discs (evaluated by the second author: M.N.) at the 1st evaluation by the new grading system which has developed on the basis of the shape and intensity of nucleus pulposus. Total JOA score was corresponded to a grading system of the disc degeneration on T2-weighted MRI. The new grading system reflected the leg symptom (p \ 0.05) and walking ability (p \ 0.05), but not low back pain, straight leg lasing test, sensory disturbance nor muscle weakness in the factor analysis The findings of MRI reflected the leg symptom and walking ability, but not low back pain, straight leg lasing test, sensory disturbance nor muscle weakness in the item of the JOA score. New grading system is suitable to detect the shape of the nucleus pulposus as well as signal intensity. Intraobserver complete reproducibilities of new grading system and Schneiderman’s criteria were 85.0 % and 81.5 %, respectively. Interobserver complete reliabilities were 60.0 % and 58.5 %, respectively. Conclusion: The new grading system of the disc degeneration on MRI mostly reflected the leg symptoms and walking ability. Interobserver and intraobserver complete agreement of our grading system were superior to Schneiderman’s criteria. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P112 DOES LUMBAR INTERVERTEBRAL DISC DEGENERATION AFFECT PSOAS MAJOR MUSCLES? Teruo Kita, Kazushi Takayama, Hiroaki Nakamura, Fumiaki Kanematsu, Toshiya Yasunami, Kazuya Nishino, Hideki Sakanaka, Yoshiki Yamano 1

Department of Orthopaedic Surgery, Seikeikai Hospital, Osaka, Japan; 2Department of Orthopaedic Surgery, Osaka City University Graduate School of Medicine, Osaka, Japan; 3Department of Orthopaedic Surgery, Saiseikai Nakatsu Hospital, Osaka, Japan Purpose: In this study, we evaluated whether degeneration of the disc close to the psoas major (PM) muscles might affect the CSA and % fat of this muscle. Methods: This study was conducted in 240 patients aged 60 years or over (80 males; 160 females; mean age, 73 years) in whom atrophy of the psoas major muscles had been evaluated based on the presence/ absence of paraspinal atrophy in plain lumbar MRI. The body mass index (BMI) of the subjects was measured. Using an image measurement software, the CSA and % fat of the psoas major muscles at the L2/3 to L4/5 level were measured on T2-weighted axial MR images, and the degree of lumbar lordosis (L1 to S1) was measured on sagittal MR images. The severity of the disc degeneration at the L2/3 to L4/5 levels was assessed using the Pfirrmann grading system (Grade I to V, a higher grade indicating more severe degeneration), and the association between the severity of the disc degeneration and the CSA/% fat of the psoas major muscles was evaluated. Results: The mean BMI was 22.4 kg/m2, and the mean lumbar lordosis angle was 31.1°. In the assessment of the severity of the disc degeneration at L2/3, none of the patients showed Pfirrmann Grade I disc degeneration (PG I), 25 patients showed PG II, 73 patients showed PG III, 99 patients showed PG, and 43 patients showed PG V. In the same assessment conducted at the L3/4 level, 8, 65, 127, 40 patients showed PG II, III, IV, V respectively. For the case of L4/5, 1, 49, 123, 67 patients showed PG II, III, IV, V. The mean PM CSA at

S789 L2/3 was 406, 429, 405, 460 mm2 in the PG II, III, IV, V patients. The mean PM CSA at L3/4 was 658, 694, 647, 698 mm2 in the PG II, III, IV, V patients. The mean PM CSA at L4/5 was 780, 899, 844, 921 mm2 in PG II, III, IV, V patients. The mean % fat of PM at L2/3 was 4.7, 4.6, 5.5, 6.5 % in PG II, III, IV, V patients. The mean % fat at L3/ 4 was 4.1, 5.0, 4.9, 4.4 % in PG II, III, IV, V patients. The mean % fat at L4/5 was 4.2, 4.0, 4.9, 3.7 % in PG II, III, IV, V patients. No significant difference was noted among the values in any of the patients. While both disc degeneration and atrophy of the PM muscles advance with age, atrophy of the PM muscles appears to be scarcely affected by the severity of disc degeneration in patients aged 60 years and over, suggesting that disc degeneration and PM muscle atrophy advance independently. Conclusions: This study showed that disc degeneration scarcely affects the CSA and % fat of the PM muscles. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: none.

P113 THROMBOEMBOLIC COMPLICATIONS FOLLOWING SURGICAL TREATMENT FOR DEGENERATIVE SPINAL DISEASE Casper Winther Larsen, Karsten Thomsen, Lars Peder Sørensen, Morten Jenstrup, Frederik Birkebæk Thomsen Institute of Health and Medical sciences, University of Copenhagen; Department of Spinal Surgery and Department of Anaesthesiology, Aleris-Hamlet Hospital; Department of Urology, Frederiksberg Hospital, Copenhagen, Denmark Background: We examined the risk of a confirmed thromboembolic complication, i.e. deep vein thrombosis (DVT) or pulmonary embolism (PE), in patients undergoing surgery for degenerative conditions in the cervical or lumbar spine. Material and methods: All patients undergoing spinal surgery at Aleris-Hamlet Hospital from August 1, 2008 through December 31, 2013. All patients suffered from a degenerative disease in either the cervical or lumbar region predominantly herniated disc, spinal stenosis, and/or degenerative disc disease. The surgical procedures in the cervical spine were conventional anterior approach including discectomy, neural decompression, and intervertebral fusion using cage-stabilization. In the lumbar region all procedures were performed via a conventional minimal open posterior approach, subperiosteal dissection, and neural decompression with or without instrumented fusion with pedicle screw fixation. Patients identified to have an increased risk of thromboembolic complications were per local protocol treated with medical thromboembolic prophylaxis (dabigatran) (age older than 40 years and prolonged surgery time, previous thromboembolism, disposition for thromboembolic disease, current malignant disease, Morbus Cordis, cardiac arythmia, above 75 years of age, or a body mass index higher than 40). Based upon the Personal Identity Number we retrieved information from the National Patient Register on hospital admission in the first 6 months after surgery for DVT or PE, using the ICD10 codes: DI260, DI269, DI802, DI803, DT817C, DT817D. DVT or PE was confirmed by personal communication and patient chart review. Results: In total, 6276 patients underwent surgery for a degenerative spinal disease of which 808 operations were cervical and 5468 lumbar. Altogether 12 patients (0,2 %) were examined suspecting a thromboembolic event. In 8 (0,1 %) patients this was confirmed:

123

S790 seven patients were diagnosed with DVT and 1 patient with PE. The incidence of DVT was 0.1 % following cervical (1 patient) and 0,1 % following lumbar (6 patients). The patient diagnosed with PE had undergone lumbar decompressive surgery and died 69 days postoperatively. All patients who experienced a thromboembolic complication and had increased risk of such (n = 3) received medical prophylaxis per protocol. Conclusion: In a big cohort the risk of a diagnosed thromboembolic event following surgical treatment for a degenerative spinal disease was 0.1 % with a mortality rate of 0.02 %. Disclosures: author 1: none; author 2: grants/research support; Company=Alexis Research Fond - To the LPR; author 3: no indication; author 4: none; author 5: grants/research support; Company=AlerisHamlet hospital.

P114 HOW WELL CAN THE CLINICIAN APPRAISE THE SEVERITY AND IMPACT OF A PATIENT’S BACK PROBLEM DURING THE CLINICAL CONSULTATION? Anne F. Mannion, Urs Mutter, Sabrina Donzelli, Monia Lusini, Frank S. Kleinstu¨ck, Salvatore Minnella, Stefano Negrini, Fabio Zaina 1

Spine Centre, Schulthess Clinic, Zuerich, Switzerland; 2ISICO (Italian Scientific Spine Institute), Milan, Italy; 3University of Brescia, Italy; 4IRCCS Don Gnocchi, Milan, Italy

Introduction: The main concerns of patients with back problems are pain and its impact on function and quality of life (QoL). These are subjective phenomena, and should be probed during the clinical consultation so that the physician can ascertain the extent of the problem, its impact on the patient’s life, and hence the most appropriate treatment. This study evaluated the agreement between clinician and patient ratings of the patient’s status on the main domains of the multidimensional Core Outcome Measures index (COMI). Methods: This was a two-centre international study involving 5 spine specialists (1 surgeon, 4 physicians) and 108 patients. Prior to the consultation, the patient completed the COMI and returned it to an independent investigator. After the consultation the clinician also completed a COMI, imagining the likely answer given by the patient, based on the preceding discussion and medical history taking. Concordance was assessed by % agreement, Kappa values and Spearman rank correlation coefficients. Results: Agreement between patient and doctor about the ‘‘main problem’’ (back pain, leg/buttock pain, or neurological disturbance) was 83 %, Kappa = 0.70 (95 % CI 0.58-0.81)). Moderate to strong correlations were found between the doctors’ and patients’ COMIitem ratings (back pain, 0.49; leg pain, 0.66; back pain minus leg pain, 0.74; function, 0.63; symptom-specific well-being, 0.48; QoL, 0.58; all p \ 0.0001), although the doctors systematically underestimated leg pain (p = 0.002) and overestimated symptom-specific well-being (p = 0.002). Discussion: The doctors were able to ascertain the location of the main problem and the differential between back and leg pain with good accuracy, but systematically underestimated the severity of leg pain and the patient’s satisfaction with their current symptom state. More detailed questioning on these domains during the consultation might deliver a better impression of the impact of the back problem on the patient’s daily life. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: stock/shareholder; Company=ISICO (Italian Scientific Spine Institute); author 8: none.

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 P115 EFFECT OF TERIPARATIDE ON SUBSEQUENT VERTEBRAL FRACTURE AFTER LONG INSTRUMENTED FUSION SURGERY FOR OSTEOPOROTIC VERTEBRAL COLLAPSE Toshiya Tachibana, Shinichi Inoue, Fumihiro Arizumi, Shinichi Yoshiya Department of Orthopaedic surgery, Hyogo College of Medicine; Nishinomiya, Japan Introduction: Vertebroplasty (VP) or kyphoplasty have been reported as effective treatments for persistent painful osteoporotic vertebral compression fractures though these procedures increase the risk of new vertebral fracture. Vertebroplasty and instrumented fusion surgery often requires osteoporotic vertebral collapse (OVC) with neurological deficits. Long instrumented fusion surgery may increase the risk of subsequent vertebral compression fracture. Teriparatide has been reported to decrease the risk of new vertebral fracture. The objective of this study was to assess the incidence and effect of teriparatide on subsequent vertebral fracture after long instrumented fusion surgery for OVC. Methods: A total of 47 patients who underwent VP with instrumented fusion surgery ([ 3 levels) for OVC between 1999 and 2013 were included in the study. The mean age at surgery was 76 years (range, 58-88years) with 20 males and 27 females. The mean follow-up period was 23 months (range, 12-59). The types of surgery included VP with posterior spinal fusion in 36 cases, anterior and posterior spinal fusion in 3 cases, posterior spinal fusion without VP in 2 cases, posterior lumbar interbody fusion in 2 cases, and osteotomy and posterior spinal fusion in 4 cases. The average of fused vertebrae was 4.9 (range, 3-7 vertebrae). Teriparatide was used in 19 patients. The occurrence of subsequent vertebral fracture was estimated by KaplanMeier analyses and was compared between the teriparatide group and the non-teriparatide using the log-rank test. Risk factors were evaluated using a Cox proportional hazards model. Results: The overall occurrence of subsequent vertebral fracture was 21 % at 1 year and 32 % at 2 years. The median time of vertebral fracture from index surgery was 4 months. A higher rate of subsequent vertebral fracture was observed with the non-teriparatide group. There was no significant difference in age, fused levels, presence of prevalent fracture, or correction loss between non-teriparatide and teriparatide group. The occurrence of subsequent vertebral fracture was lower in the teriparatide group (16 % vs 54 %, P = 0.014). Cox proportional hazard model revealed use of teriparatide is only a protective factor of subsequent vertebral fracture after instrumented fusion surgery for OVC (Hazard ratio 0.281, P = 0.047). Discussion and conclusion: Teriparatide significantly reduced the occurrence of subsequent vertebral fracture after instrumented fusion surgery for OVC. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P116 PREVALENCE AND MECHANISMS OF ADJACENT SEGMENT DISEASE FOLLOWING LUMBAR SPINE FUSION FOR DIALYSIS-ASSOCIATED SPONDYLOSIS IN LONGTERM HEMODIALYSIS PATIENTS Toshiya Tachibana, Shinichi Inoue, Fumihiro Arizumi, Shinichi Yoshiya Department of Orthopaedic Surgery, Hyogo College of Medicine

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Introduction: Dialysis-associated spondylosis (DAS) is a serious complication which includes destructive kyphosis, vertebral slippage, and extradural amyloid deposit in long-term hemodialysis patients. Instrumented fusion surgery is often required to stabilize these conditions. However, the adjacent segment pathology after instrumented lumbar fusion surgery for DAS is currently unclear. The objective of this study was to assess the prevalence and mechanisms of adjacent segment disease following lumbar instrumented fusion surgery for DAS in long-term hemodialysis patients. Methods: A consecutive series of 36 long-term hemodialysis patients who underwent lumbar instrumented fusion surgery for DAS between 1999 and 2013 were included in this study. The mean age at surgery was 65 years (range, 49-79 years) with 21 males. The mean follow-up period was 4 years (range 1-10 years). The average length of hemodialysis was 22.5 years (range, 10-35 years). The types of DAS consisted of destructive kyphosis in 4 cases, vertebral slippage in 17 cases, Type A and B in 4 cases, extradural amyloid deposit in 5 cases, and degenerative scoliosis in 6 cases. A total of 14 patients underwent a 1-level fusion, 22 patients underwent a multi-level fusion (2-level in 9, 3-level in 10, and 4-level in 3). ASD was defined as adjacent disc collapse, adjacent segment stenosis. The Japanese Orthopedic Association score (JOA score), recovery rate (Hirabayashi method), and reoperation were reviewed to evaluate the clinical outcome. Results: The mean JOA score significantly increased from 13.5 before surgery to 21.3 at the final follow-up. The mean recovery rate was 51.4 %. Six of the 36 patients died within 1 year after index surgery. One patient died due to perioperative complication. More than 1 year of follow-up after surgery was available for 30 patients. Symptomatic ASD occurred in 43 % (13 of 30) of the cases. Of the 13 cases, 5 had adjacent segment disc degeneration and 8 had adjacent segment spinal stenosis. Three cases (10 %) required reoperation due to proximal adjacent segment spinal stenosis. Multilevel fusion surgery increased the risk of ASD compared with 1-level fusion surgery (61 % vs. 38 %). The recovery rate was significantly lower in the ASD group than the non-ASD group (61 % vs. 38 %). Discussion and conclusion: Symptomatic ASD occurred in 43 % of the cases. Three cases (10 %) required revision surgery for ASD. Multilevel fusion surgery increased the risk of ASD in DAS. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P117 A MEASURE TO AVOID PLEURA INJURIES IN XLIF AT UPPER LUMBAR LEVELS Yong Kim, Masabumi Miyamoto Department. of Orthopedic Surgery, Nippon Medical School, Chiba Hokusoh Hospital, Inzai-City, Chiba, Japan Objective: To report the review of the problems and solutions of pleural injury associated with the procedure at the upper lumbar levels. Methods: Of 60 patients who underwent XLIF between September 2013 and December 2014, we reviewed 31 patients with the topmost fusion vertebrae proximal to the 2nd lumbar vertebra. The patients consisted of 19 males and 12 females with the mean age of 68.7 (30 to 82) years. The diseases treated consisted of lumbar instability in 9 patients, degenerative scoliosis in 6, spinal kyphosis in 6, lumbar foraminal stenosis in 5, lumbar degenerative spondylolisthesis in 4, and lumbar burst fracture in 1. We investigated the occurrence of intraoperative pleural injury, postoperative pneumothorax, etc.

S791 Results: One patient with the topmost fusion at Th12 experienced intraoperative pleural injury, pleural suture, and intraoperative chest drain insertion, 1 of 8 patients with the topmost fusion at L1 developed postoperative pneumothorax and received chest drain insertion on the next day of the procedure, and 1 of 22 patients with the topmost fusion at L2 experienced pleural injury and received intraoperative pleural suture, but postoperative pneumothorax was noted and a chest drain was inserted on the day of procedure. In all of these patients, the drain was able to be removed within 1 week. Discussion: There is a concern about pneumothorax due to pleural or diaphragmatic injury in XLIF at the upper lumbar levels. Even at the L2/3 level, we should keep the risk of injury in mind. Recently, we resect the 11th rib lateral to the vertebral body (if there seems to be a problem in the approach, we also resect the 12th rib and use it for bone transplantation), then, although we make an opening for approach to the retroperitoneal space at a distal site to the 12th rib and insert a retractor in an oblique direction, it is easy to raise the retractor vertical to the lateral side of the interbody because the 11th rib is resected. Since this method was started, pleura injuries haven’t been experienced. Additionally this method gave advantages to avoid angle technique, and to cancel donor site pain of illiac bone which should be considered since bone harvests are still common in Japan. We consider that this technique can apply to XLIF at up to the L1/2 level, but at the TH12/L1 level, further investigations are required. Disclosures: author 1: none; author 2: none.

P118 LEARNING CURVE OF THE SURGICAL TECHNIQUE OF XLIF Yong Kim, Masabumi Miyamoto Department of Orthopedic Surgery, Nippon medical School, Chiba Hokusoh Hospital, Inzai-City, Chiba, Japan Objective: To report the review of the first consecutive cases of XLIF performed by the same surgeon and the discussion of acquisition of the XLIF surgical technique. Subjects and methods: We reviewed the first 60 cases of XLIF for 16 months since September 2013 (33 males, 27 females, mean age 66.9 [22 to 80] years). The diseases treated consisted of lumbar degenerative spondylolisthesis in 16 patients, lumbar instability in 15, lumbar disc herniation/vertebral foraminal stenosis in 11, degenerative scoliosis in 8, spinal kyphosis in 6, trauma in 3, and pyogenic spondylitis in 1. We retrospectively investigated the duration of surgery and intraoperative blood loss in the first 20 cases (group A), intermediate 20 cases (group B), and last 20 cases (group C). In addition, the three levels of L2/3, L3/4, and L4/5 were compared between the first 10 patients and last 10 patients. Note that the procedure was performed using the left lateral approach with the use of NVM5 spinal monitoring in all the patients. Results: The mean total duration of surgery including the postural change and posterior procedure was 299 minutes in group A, 255 minutes in group B, and 292 minutes in group C. The mean duration of XLIF was 115 minutes in group A, 80.2 minutes in group B, and 74.7 minutes in group C (XLIF per level: 74.4 minutes in group A, 41.1 minutes in group B, and 35.5 minutes in group C). The mean total blood loss during XLIF was 18.6 mL in group A, 6.5 mL in group B, and 7.7 mL in group C. Moreover the mean blood loss per level was 12 mL in group A, 3.3 mL in group B, and 3.6 mL in group C. The comparison between the first and last patients showed that the duration of XLIF per level and the blood loss were decreased from 48 to 36 minutes and 4.2 to 2.0mL, respectively, for L2/3; 50 to 30.5

123

S792 minutes and 8.9 to 2.3 mL, respectively, for L3/4; and 78 to 38 minutes and 13 to 4.4 mL, respectively, for L4/5. Discussion and conclusion: At the upper lumbar vertebrae, placement of a retractor is hampered by the ribs and caution needs to be exercised for occurrence of pleural or diaphragmatic injury. In consideration of the learning curve in our review, XLIF should be performed for the L3/4 level at the early stage of introduction and after the surgeon becomes familiar with the surgical technique, they should start to perform the procedure for L2/3 and L4/5, which appeared to be a shorter way to acquire the XLIF technique. Disclosures: author 1: none; author 2: none.

P119 GERIATRIC RISK IN THE SURGICAL MANAGEMENT OF INFECTIOUS SPONDYLITIS Bong-Soon Chang, Jae Hong Ha, Choon-Ki Lee, Hyoungmin Kim, Jong-Hun Jung, Doohyun Kwon, Guang Bin Zheng, Jae Woo Park Department of Orthopaedic Surgery, Seoul National University Hospital, Seoul, Korea Hypothesis: The purpose of this study was to evaluate and compare clinical outcomes in patients over and under 65 years of age who have undergone surgery for infectious spondylitis, and to identify any prognostic factors. Design: Retrospectively, we reviewed 97 consecutive patients who received surgical treatment for infectious spondylitis. Patients with a previous history of spine surgery or spinal intervention within 1 month were excluded. Among the 60 patients enrolled, 26 were over 65 years of age and 34 were under 65. The minimum follow up period was 1 year, with 86.7 % of patients fulfilling this criteria. Primary outcome measures were postoperative complications and control of infection upon follow up. Introduction: Although elderly patients account for approximately 40 % of total patients treated surgically for infectious spondylitis, there are few studies comparing clinical outcomes in elderly and younger patient groups. Methods: Retrospectively, we reviewed patients’ comorbidities, preoperative serum albumin levels, postoperative complications, and control of infection upon follow up. Results: Postoperative complication rates (p = 0.764) and infection control rates (p = 0.275) were not significantly different between the two groups. Univariate analysis did not show correlation between age and clinical outcome, while BMI (p = 0.04), CCI (p = 0.017), ASA grade (p = 0.006), and serum albumin (p = 0.003) were associated with overall postoperative complications. BMI (p = 0.002) and CCI (p = 0.000) were also associated with postoperative fatalities. ˆI§2 test for trend also showed that CCI (p = 0.018), ASA grade (p = 0.007), low serum albumin (\3.5 mg/dL; p = 0.004) were associated with postoperative complications. Logistic regression analysis showed that ASA grade (p = 0.034) and BMI (p = 0.044) were related to overall postoperative complications. ROC curve analysis using ASA grade and BMI to predict major postoperative complications and fatality showed an area under curve value of 0.793 (p = 0.001) and 0.942 (p = 0.002), respectively. Clinical outcomes in the surgical management of infectious spondylitis in elderly patients were comparable to those in younger patients. Body mass index (BMI), American Society of Anesthesiologists (ASA) grade, Charlson Comorbidity Index (CCI) scores, and serum albumin levels, rather than age, were significantly associated with clinical outcomes. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: none.

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 P120 IS ASYMMETRICAL PEDICLE SUBTRACTION OSTEOTOMY AN EFFECTIVE METHOD TO RECONSTRUCT CORONAL AND SAGITTAL BALANCE FOR ANKYLOSING SPONDYLITIS PATIENTS WITH THORACOLUMBAR KYPHOSCOLIOTIC DEFORMITY? Bangping Qian, Yong Qiu, Mingliang Ji, Jun Hu Spine Surgery, the Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing, China Introduction: Ankylosing spondylitis (AS) patients with thoracolumbar kyphoscoliotic deformity may present with fixed sagittal and coronal deformity. To date, asymmetric pedicle subtraction osteotomy (APSO) has been reported as an optimal choice to correct a combined coronal and sagittal imbalance in adult scoliosis patients. However, the effectiveness of APSO on AS patients with biplanar imbalance has not yet been addressed. Objective: To investigate the influence of APSO on the reconstruction of coronal and sagittal balance in AS patients with thoracolumbar kyphoscoliotic deformity. Materials and methods: Between October 2005 and June 2012, sixteen AS patients with thoracolumbar kyphoscoliotic deformity undergoing APSO were included in this study. There were 13 males and 3 females with a mean age of 35.4 years (range, 22-48 years). The average time of follow-up was 36 months. Full-length antero-posterior and lateral spine radiographs of the pre-operative, post-operative, and the final follow-up were available. Coronal and sagittal parameters were measured, including Cobb angle, central sacral vertical line (CSVL), global kyphosis (GK), sagittal vertical axis (SVA), thoracic kyphosis (TK), lumbar lordosis (LL), pelvic tilt (PT), sacral slope (SS), and pelvic incidence (PI). Results: The mean Cobb angle was improved from 25.8° preoperatively to 7.6° postoperatively (P \ 0.001), and the correction rate was 70.5 %. The CSVL was corrected from 5.6 cm to 1.8 cm (P \ 0.001). The mean GK was decreased from 76.8° to 25.6° (P \0.001), and the correction rate was 66.7 %. The SVA was restored from 15.1 cm to 3.8 cm (P \ 0.001). In addition, LL, PT, and SS were improved from -0.4°, 33.6°, and 10.3° to 44.1°, 22.3°, and 20.9°, respectively (P \ 0.001). In terms of Cobb angle, CSVL, GK, SVA, LL, PT, and SS, no significant differences were observed between postoperative and the last follow-up (P [ 0.05). Conclusion: APSO can achieve successful realignment of biplanar balance by correcting thoracolumbar kyphosis and scoliosis simultaneously. In addition, APSO provided excellent long-term radiographic outcomes without a significant loss of correction during follow-up. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P121 RISK FACTORS OF NEW SYMPTOMATIC VERTEBRAL COMPRESSION FRACTURES IN OSTEOPOROTIC PATIENTS UNDERGONE PERCUTANEOUS VERTEBROPLASTY Hai-Long Ren, Jian-Ming Jiang, Jian-Ting Chen, Ji-Xing Wang Nanfang Hospital, Southern Medical University, Guangzhou, China Purpose: This study evaluated the risk factors of new vertebral compression fractures (VCFs) following PVP.

Eur Spine J (2015) 24 (Suppl 6):S743–S800

S793

Methods: From June 2005 to January 2011, patients with osteoporotic VCFs (OVCFs) who were treated with PVP and met this study’s inclusion criteria were retrospectively reviewed. Observed parameters were age, sex, BMD, BMI, amount of bone cement, cement leakage into the disk, preoperative kyphosis, preoperative degree of anterior vertebral compression, preoperative degree of middle vertebral compression, kyphosis correction, anterior vertebral height restoration, middle vertebral height restoration, and number of initial symptomatic fractures (levels treated). The data were analyzed by univariate and multivariate analysis for the emergence of new fractures after PVP to determine related risk factors. Results: A total of 182 patients met the inclusion criteria. There were 155 female and 27 male patients with a mean age of 69.7 years (range, 49-91 years). The follow-up period was 24-50 months (average, 26.4 months). A total of 294 VCFs among 182 patients were observed, twenty-eight new VCFs occurred in 21 patients (21/182, 11.5 %) during the follow-up period. Statistical analysis indicated that higher BMI (P = 0.004) and a greater number of initial symptomatic fractures (P = 0.017) were significantly associated with new VCFs after PVP. It is the most obvious that the risk of new fractures increased 2.518-fold (95 % CI, 1.176-5.395), when the number of initial VCFs increased by one level. Conclusions: The incidence of new symptomatic VCFs after PVP was higher in osteoporotic patients with initial multiple-level fractures. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

aurus were cultured in 17 (52 %), the rate of implant removal was not different in the group with Staph aurus (4 of 17) compared to other microbes (5 of 16). We found no significance for age or number of instrumented vertebrae. In the first 16 patients ranked by date of operation the implant was retained in 9 (56 %), in the 17 last patients the implant was retained in 15 (88 %), (p = 0.06, Fisher exact). Discussion: In most patients with early SSI after instrumented spinal surgery the infection can be resolved and implants retained with surgical debridement and proper use of antibiotics. A non-significant tendency for a higher rate of implant retention in the last part of the study period may reflect a greater confidence that infections can be resolved with retained implants. Disclosures: author 1: none; author 2: none; author 3: consultant; Company=DePuySynthes; author 4: none.

P122 IMPLANTS CAN BE RETAINED IN MOST CASES OF EARLY SURGICAL SITE INFECTION AFTER INSTRUMENTED SPINAL SURGERY

Introduction: In LIF transforaminal approach with kidney shaped cages is among the most frequent used techniques. TLIF cages share in common: they are small compared to the vertebral body and rest on the weak central part. This can lead to cage subsidence or cage migration with loss of correction and pseudoarthrosis. When ALIF or XLIF approach is used where the cage design provides more stability a second approach for pedicle screw fixation is needed. Objectives: To overcome the possible disadvantages of existing TLIF cages and the need for a second approach in ALIF and XLIF a novel cage design was introduced and the operative technique modified. The novel ACRON cage rests on the apophyseal rings due to his dimensions and shape and can be inserted via the same approach as in traditional TLIF with just a small modification: the extraforaminal facetectomy approach. this EF-LIF approach includes complete resection of the medial facett and resection of the cranial part of the lateral facett. Dissection of the intertransverse ligament and muscle is added to be able the mobilise the traversing nerve root - ganglion if necessary. This leads to a straight route the the far lateral aspect ot the disc. After clearance of the disc space the ACRON cage can be inserted without any retraction of dural sac. Methods: EF-LIF with ACRON cage was done in a consecutive series of patients all having criteria for lumbar spinal fusion. These include: degenerative spondylolisthesis grade 1 or 2, spinal canal stenosis, DDD, postlaminectomy syndrome. Adjacent segment instability and degenerative lumbar scoliosis. Orthopedic and neurologic examination pre- and postoperatively and radiologic control was done in the usual clinical setting. In 110 patients 75 single level, 23 two level and 12 three level LIF was performed. Results: In all consecutive cases the operation did not show any complications intraoperatively. Correct placement of the ACRON cage in a transverse manner from the EF-LIF approach was achieved and did not show any change during radiologic controls after 1 week, 1 month and 3 months. Correction of deformity and realignment with sagittal balance was achieved as seen on ap and lateral X-ray. The clinical condition of the patients improved concerning pain and functional status.

Ivar Rossvoll, Øyvind Sletten, Torsten Bra¨uer, Rainer Knobloch Trondheim University Hospital, NTNU, Trondheim, Norway Introduction: The reported incidence of surgical site infection (SSI) after instrumented spinal surgery varies from 2.2 to 12 %. Implant retention rates from 40 to 100 % are reported after debridement and use of antimicrobial agents. The aim of this retrospective study was to evaluate to which extent implants could be retained in early (\30 days) SSI? Methods: Patients operated for deep (below fascia) SSI within 30 days after instrumented spinal surgery (1.1.2005 to 31.12.2012) were identified in our operation registry. Patients with positive bacteriological culturing operated for deformity, fracture, or degenerative conditions in the thoracolumbosacral column, and residing in our health region, were included. Patient data were collected from the electronic patient journal. Results: 33 patients (19 women), mean age 46(14-84) years were included. Time from index operation to first debridement was 16(728) days. A mean of 1.8(1-4) debridements were performed. Intravenous antibiotics were started after biopsy for culturing were taken, and were continued mean 2.6(1-6) weeks after last debridement followed by per oral antibiotics for mean 8.0(4-12) weeks in patients in whom implants were retained. Implants were retained in 24 (73 %), in 6 of these part or all of the original implant was replaced. Implants were removed in 9 (4/11 fractures, 1/7 deformities, 4/15 degenerative). All infections were resolved; no reoccurrence was registered during a follow up of 59(range 27-110) months. In patients where the implant was removed there were more debridements (mean 2.4 vs 1.5, p = 0.008) and preoperative CRP was higher (mean 268 vs 143, p = 0.01). ASA was higher in implant removals (p = 0.04). Staph

P123 EF - LIF: (EXTRAFORAMINAL FACETTECTOMY LIF) A NEW MODIFICATION AND IMPLANT DESIGN FOR LUMBAR SPINAL FUSION. CLINICAL AND RADIOLOGIC RESULTS OF THE FIRST 110 CASES Ronald J. Sabitzer Orthopedic Department, Otto Wagner Hospital (OWS) Vienna, Austria

123

S794 Conclusions: The new technique of EF-LIF using the ACRON cage implant shows to be an alternative to traditional TLIF ALIF and XLIF procedure. Multicenter trial study should be done. Disclosures: author 1: none.

P124 RESULTS OF LONG FUSIONS TO THE SACRUM FOR ADULT SPINE DEFORMITY Charles Sansur, Andrew Frost, Niall Craig, John Schmidt, Jennifer McCool, Megan Dumas University of Maryland Medical Center, NHS Grampian, Leesburg, USA Introduction: The purpose of this study is to assess the adverse events resulting from long fusions to the sacrum from procedures correcting adult spinal deformity. Methods: Data from 2344 patients was entered either prospectively or retrospectively into a multicenter database from 2009-2013. 48 separate sites entered data including 5 overseas sites. Patients were included in this study if they were 18+ years old and had a long fusion C 5 levels fused) to the sacrum correcting any adult deformity of the spine. Results: 408 patients were included in this study that had long fusions to the sacrum. 129 patients had adult deformities with long fusions to the sacrum. 57 patients had 2+ years of follow up. Patients were analyzed on the presence of adverse events. 40 patients (31 %) with a mean of 9.91 ± 3.71 levels fused experienced adverse events related to the procedure. 44 patients required revision surgery (34.1 %) with 21 (16.3 %) of these patients undergoing revision surgery for implantrelated complications. The most common implant complication types were rod fracture (15 pts, 11.6 %) and screw interface disengagement (9 pts, 7 %). Pseudarthrosis was only reported in 2 patients with adult deformities (1.6 %). Conclusion: A previous study analyzed the prevalence of pseudarthrosis in adult deformity fusions to the sacrum and found a rate nonunion rate of 24 % [1]. Another similar study found a 24 % pseudarthrosis rate in their cohort [2]. However, in the current study we found the rate of pseudarthrosis to be low at only 2 %. From this data, surgeons should not worry about fusing long and including the L5/S1 level if needed. 1. J. Weistroffer, J. Perra, J. Lonstein, J. Schwender, T. Garvey, E. Transfeldt, J. Ogilvie, F. Denis, R. Winter, J. Wroblewski, Spine Vol 33, p. 1478-83, 2008. 2. Y. Kim, K. Bridwell, L. Lenke, S. Rhim, G. Cheh, Spine Vol 31, p. 2329-36, 2006. Disclosures: author 1: consultant; Company=Depuy synthes; author 2: none; author 3: grants/research support; Company=Cerapedics, consultant; Company=Q Spine; author 4: employee; Company=K2M; author 5: stock/shareholder; Company=K2M, inc., employee; Company=K2M, inc.; author 6: employee; Company=K2M, inc.

P125 COMPARISON OF ADVERSE EVENTS IN SHORT AND LONG CONSTRUCTS IN BOTH POSTERIOR AND ANTERIOR SPINAL SURGERIES Charles Sansur, Andrew Frost, Niall Craig, John Schmidt, Jennifer McCool, Megan Dumas University of Maryland Medical Center, NHS Grampian, Leesburg, USA

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Introduction: The purpose of this study is to assess whether there is a difference in adverse events from patients having either a long or short construct from either a posterior or anterior surgery. Methods: Data from 2344 patients was entered either prospectively or retrospectively into a multicenter database from 2009-2013. 48 separate sites entered data including 5 overseas sites. Patient selection was based on no prior surgery and presence of adverse events (AEs). Patients were then divided into either anterior only or posterior only surgeries with short (\ 5 levels) or long C 5 levels) constructs. Results: 801 patients are included in this analysis. The mean length of constructs was 8.65 ± 3.11 (long) and 2.26 ± 1.27 (short) levels. 180 related AEs were reported in 125 patients. There were 350 patients with long constructs, 59 of which (16.86 %) experienced 87 AEs. 451 patients had short constructs with 66 (14.63 %) experiencing 92 AEs. Out of the 173 patients with anterior only surgeries, 30 (17.34 %) experienced 38 AEs while 95 of the 628 patients with posterior only surgeries (15.13 %) experienced 142 AEs. Out of the 13 patients with instrumentation failure, 6 had long constructs (46.15 %) and 7 had short constructs (53.85 %). Two of those instrumentation failures were from anterior only surgeries (15.38 %) while 11 were from posterior only surgeries (84.62 %). Instrumentation failure (13 pts), dural tears (16), back pain (21), ileus (14) and decrease in MEPs (10) were the most common AEs. Conclusions: We anticipated that there would be a greater amount of complications in the long fusion group when compared to the short. However, the rates of AEs in long versus short constructs (16.86 % vs 14.63 %) were virtually the same. There also appears to be no significant difference between the AE reported in anterior versus posterior surgery (17.34 vs 15.13 %). Instrumentation failures are also similar in short versus long constructs (53.85 % vs 46.15 %). However, posterior instrumentation has a far greater propensity to instrumentation failure than anterior instrumentation (84.62 % v 15.38 %). The data presented suggests that surgeons should not be concerned about using long constructs as they do not seem to have a significantly increased occurrence of AE. Disclosures: author 1: consultant; Company=Depuy synthes; author 2: none; author 3: grants/research support; Company=Cerapedics, consultant; Company=Q Spine; author 4: employee; Company=K2M; author 5: stock/shareholder; Company=K2M, inc., employee; Company=K2M, inc.; author 6: employee; Company=K2M, inc.

P126 COMPUTER ASSISTED ROBOTIC SURGERY IN OCTOGENARIANS A CASE CONTROLLED STUDY Josh Schroeder, Eyal Itshayek, Yair Barzilay, Leon Kaplan Hadassah Hebrew University Medical Center, Jerusalem, Israel Introduction: The proportion of population over the age of 80 undergoing spinal surgery is growing. The use of robotic computer assisted surgery in these patients is challenging, due to poor bone quality, and at times complex pathology and anatomy. We are present a consecutive series of octogenarian patients who underwent robotic computerized guided spine surgery. The aim of the study is to evaluate the success of robotic navigation spine surgery in patients 80 years old or older. Methods: Prospective data in spine referral center was retrospectively collected and analyzed. All patients who were 80 years or older at the time of surgery were identified. These patients were matched to 120 patients under the age of 80 (a 3:1 ratio). Patients’ age, sex and indication for surgery were documented. Time of procedure, accuracy execution, fluoroscopy time, and any instrumentation related complication were documented.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Results: Between 2007 and 2013, 192 trajectories were executed in the octogenarian patients and 568 trajectories in the younger patients. The average age was 83.9 years vs 61.2 (P \ 0.05); BMI was 25.4 in the octgenrians vs 27.2 in the younger patients. 12 patients were males in the octogenarian vs 50 in the more younger patients. The upper instrumented in both groups vertebra was T5 and the lowest was S1. Percutaneous procedures were performed in 25 octogenarian patients and 73 in the younger group of patients. Average robotic usage time was 6 min and 40 seconds in octogenarians, compared to 5 min and 5 seconds in younger patients (P \ 0.05). Radiation exposure time was 16.3 seconds per screw in the octogenarian’s vs 9.3 seconds in the younger patients (P \ 0.05). 182 (94.5 %) executed trajectories were accurate vs 545 (95.9 %) accuracy in patients under the age of 80 (NS). No robotic assistance or any hardware related complications occurred. Conclusion: Spine surgery in the octogenarians is challenging. The combination of osteoporotic bone and complex spinal pathologies, results in longer robotic procedures and in higher radiation exposure time, compared to younger patients. However, execution accuracy and safety is identical to younger group of patients, allowing to apply optimal instrumentation in these frail patients. Disclosures: author 1: none; author 2: grants/research support; Company=MFAST - local distributer of medical devices; author 3: consultant; Company=MazorRobotics.

P127 NARCOTIC USE AMONGST SPINE PATIENTS - TRENDS AND BELIEFS Josh Schroeder, Jennifer Shue, Leon Kaplan, Joseph Nguyen, Frederico Girardi Hadassah Hebrew University Medical Center, Jerusalem, Israel Introduction: There is a growing body of evidence regarding the dangers involved with the use of narcotics. With increasing numbers of patients taking narcotics on a regular basis, complications from narcotics may be greater than complications from other groups of medications. Methods: An internet-based survey was distributed to AOSpine members in North America and Europe, and spine surgeons in South America. Spine surgeons and practitioners were asked to answer the questions regarding proportion of patients using narcotics prior to presenting to practice and specific conditions and when narcotics should be introduced for cervical radiculopathy (CR), lumbar disc herniation (LDH), spinal stenosis, spondylolisthesis, and osteoporotic compression fracture. Finally, surgeons were asked if patients treated with narcotics have an inferior outcome. Results: 204 surgeons from around the world completed the survey. 125 of respondents were from Europe and 57 from North America. 54 % of surgeons were more than 10 years in practice. Narcotic use was higher in the USA than Europe with surgeons in practices with approximately one to two thirds of patients on narcotics. USA surgeons tend to give narcotics for the care of acute problems (i.e. LDH and CR), and tend to keep patients for longer periods of time on narcotics after surgery. Significantly more surgeons thought that narcotics should never be started in chronic conditions (i.e. spinal stenosis and spondylolisthesis) versus acute pain conditions (p \ 0.05), with less surgeons thinking that narcotics are a good bridge to surgery in these patients. When assessing if surgeons think that narcotics hinder surgical outcome, 37 % responded that it hinders outcome, 20 % answered that they do not know, and 43 % thought it does not affect the surgical outcome. However, the surgeons that do not give narcotics in chronic conditions feel that narcotics hinder the outcome of spine surgery.

S795 Conclusions: Narcotic use is a universal phenomenon with 40 % of spine surgeons claiming that over one third of the patients they see are on narcotics after family doctors and primary care physicians treat them. The surgeons feel that narcotics should be used as a bridge to surgery when needed in patients with acute conditions. Even so, this use is believed to be safe by only 43 % of surgeons. Guidelines and education on the narcotics are necessary to increase patient safety. Disclosures: author 1: none; author 2: none; author 3: no indication; author 4: none.

P128 SPINAL EPIDURAL ABSCESS: PATHOGENESIS, CAUSATIVE ORGANISMS AND TREATMENT ALGORITHM Ahmed Shawky, Christian Bickel, Ali Ezzati Helios Hospitals Erfurt - Germany; Assist University Hospitals, Egypt Introduction: Spinal epidural abscess is a relatively rare but lifethreatening condition for affected patients. Polymorbid patients and those with immune suppression are commonly affected. Typically, spinal epidural empyema is due to gram-positive bacterial infection. Spontaneous epidural abscess with gram-negative bacilli are very rare. Gas-forming organisms is very rare and is usually associated with high morbidity and mortality rates. Patients and methods: Retrospective analysis of our hospital records between January 2005 and January 2014 for patients diagnosed as spinal epidural abscess revealed 75 patients. Analysis of these cases regarding epidemiological data, aetiology, causative organisms, affected anatomical region and treatment was done. Results: The mean age was 63 years. 42 patients were females while 33 were males. In 18.67 % the causative organism could not be detected. In 25.33 % the epidural abscess was post-intervention while in 74.67 % it was spontaneous. Gram-negative organisms were the causative organism in 8 %, while gram-positive were detected in 73.33 %. Gram-negative gas-forming organisms were detected in two cases. Staphylococcus aureus was the most common organism representing about 50.66 %. Two cases showed epidural gas formation. In 90 % (68 patients) operative drainage was done. All cases received antibiotics for 6 weeks postoperative. Seven Cases (9 %) died perioperative, three cases preoperative and two cases postoperative, and two late postoperative ([6 months postoperative) Conclusion: Spinal epidural abscess is a rare but devastating condition commonly caused by gram-positive pathogens. Staphylococcus aureus is the main causative organism without remarkable changes in pathogens profile over the last ten years. Surgical drainage and culture-sensitivity-based antibiotic therapy are the gold-standard treatment. Gas-forming pathogens are associated with high morbidity and mortality rates. Disclosures: author 1: none; author 2: none; author 3: none.

P129 PATIENT AND SURGEON RADIATION EXPOSURE DURING SPINAL INSTRUMENTATION USING INTRAOPERATIVE CT-BASED NAVIGATION John Street, Daniel Mendelsohn, Jason Strelzow, Juliet Batke, Nicolas Dea, Marcel Dvorak, Charles Fisher University of British Columbia, Vancouver, Canada

123

S796 Hypothesis: Use of intra-operative cone-beam CT navigation alters patient and surgeon exposure to radiation and decreases the number of post-operative radiographs and CT scans. Design: Ambispective review of surgical cases using intra-operative cone-beam CT navigation. Introduction: Imaging modalities to optimize visualization of spinal anatomy include plain x-rays, fluoroscopy with C-arm and intra-operative cone-beam CT which all expose the patient and surgeon to ionizing radiation. Methods: Intra-operative x-rays, fluoro and CT dosages were recorded and standardized to effective doses. The number of peri-operative images was compared to a matched cohort involving only intra-operative x-ray and fluoro. Results were compared to literature values for fluoro-guided spine instrumentation. Results: Radiation emission data was available for 73 surgical cases involving an average of 5.44 levels. Thoracolumbar instrumentation was associated with the highest radiation emission compared to cervical cases (5.65 mSv vs. 2.19 mSv). Major deformity and degenerative cases involved more radiation exposure than trauma or oncology (5.81 mSv vs. 3.54 mSv). On average, the patient was exposed to 5.9 times more radiation than the surgeon. Total patient exposure was 3 times the values reported in the literature for thoracolumbar fusion without intraoperative CT navigation. Radiation exposure to the surgeon was reduced by 48 % compared to conventional fluoroscopy facilitated open thoracolumbar fusions and 63 % less than minimally invasive thoracolumbar fusions with navigation. The average total patient radiation exposure was 6.05 mSv, less than a single routine lumbar CT scan (7.5 mSv to 10 mSv). Use of intra-operative CT did not reduce the number of post-operative x-rays or CT scans. Conclusion: Intra-operative cone-beam CT navigation increases radiation exposure to patients and reduces radiation exposure to surgeons compared to values reported in the literature. Thoracolumbar instrumentation for major deformity and degenerative diagnoses are associated with the highest levels of radiation exposure. Intraoperative cone-beam CT navigation improves the accuracy of spine instrumentation with acceptable patient radiation exposure and reduced surgeon exposure. Disclosures: author 1: grants/research support; Company=Medtronic; author 2: none; author 3: grants/research support; Company=Medtronic; author 4: none; author 5: none; author 6: grants/ research support; Company=Medtronic, consultant; Company=Medtronic, royalties; Company=Medtronic; author 7: grants/ research support; Company=OREF, consultant; Company=Medtronic, Nuvasive, royalties; Company=Medtronic.

P130 RADIATION-FREE PEDICLE SCREW INSERTION METHOD FOR THE CERVICAL AND THORACIC SPINE-THE SCREW GUIDE TEMPLATE SYSTEM Taku Sugawara, Naoki Higashiyama, Shuichi Kaneyama, Masato Takabatake, Masatoshi Sumi Research Institute for Brain and Blood Vessels Akita, Japan Introduction: Cervical and thoracic pedicle screw fixation can provide immediate rigid intervertebral fixation, but carries a potential risk for injury to the arteries, nerve roots, and dural sac. We recently developed an intraoperative screw guiding method using patientspecific laminar templates and verified the accuracy of our multi-step procedure in the cervical and thoracic spine. Methods: Preoperative bone images of the computed tomography (CT) scans were analyzed using three-dimensional (3D)/multiplanar

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 imaging software and the trajectories of the screws were planned. Plastic templates with screw guiding structures were created for each lamina by 3D design and printing technology. Three types of templates were made for precise multi-step guidance, and all templates were specially designed to fit and lock on the lamina during the procedure. Plastic vertebra models were also generated and preoperative screwing simulation was performed. Surgery was performed using this patient-specific screw guide template system and the placement of screws was postoperatively evaluated by CT. Results: Ninety patients with cervical and/or thoracic pathology were selected to verify this novel procedure. This method is used to insert a total of 431 cervical and thoracic pedicle screws (C2: 91 screws, C36: 88, C7: 78, T1-4:104, T5-8: 53, T9-12: 17). Preoperatively, each template was found to exactly fit and lock on the lamina of the vertebra models, and screwing simulation was successfully performed. Intraoperatively, templates also fit and locked on the patient lamina and screw insertion was completed successfully. Postoperative CT scans showed that three screws (0.7 %) slightly violated the cortex of the pedicles, and the mean deviation of the screws from the planned trajectories was 0.71 ± 0.25 mm at the coronal midpoint section of the pedicles. Conclusions: The multi-step, patient-specific screw guide template system is useful for intraoperative pedicle screw navigation in the cervical and thoracic spine. This simple method substantially improves the accuracy of screw insertion and reduces operation time, and does not require expensive equipment, such as image-guided navigation machine or intraoperative CT scanner. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

P131 A NOVEL EVALUATION METHOD FOR NEURAL FUNCTIONS OF CEREBROSPINAL TRACT IN HUMAN USING FUNCTIONAL MAGNETIC RESONANCE IMAGING TECHNIQUE Eiji Takasawa, Mitsunari Abe, Haku Iizuka, Kenji Takagishi, Takashi Hanakawa Department of Orthopaedic Surgery, Gunma University Graduate School of Medicine, Maebashi, Japan Introduction: To data, spinal disorders have been diagnosed mainly by anatomical images (e.g. X-ray, CT scan or MRI), and neurological examination. However, there exist clinical discrepancies between the severity of images and clinical symptoms, which may cause a misdiagnosis and failed spinal surgeries. To solve this clinical problem, it is necessary to assess the actual neural function of central nervous system in an objective manner. The purpose of this study was to develop a novel functional MRI (fMRI) scanning technique which enables simultaneous measurement of neural activities in the brain and the cervical spinal cord, and to assess the neural function of cerebrospinal tract (CST). Methods: Thirteen right-handed, young adult healthy volunteers participated in this study. Subjects performed finger-opposition tasks with the right and left hand. The fMRI data were acquired on a 3.0 T MRI scanner (Siemens Verio) during the tasks. Scanning was performed in the sagittal plane covering from the primary motor cortex (M1) down to the cervical spine. The fMRI data were preprocessed with SPM 8 software (UCL, London, UK). Region of interests (ROIs) data were extracted from the M1 and the cervical spinal cord and then neural activities in the two regions were detected. Regression analysis

Eur Spine J (2015) 24 (Suppl 6):S743–S800 was also computed to verify the functional coupling between the M1 and the hand motor segment C7-T1. Results: Neural activities in the brain and the cervical spinal cord were successfully detected by our cerebrospinal fMRI method. Higher neural activities were found in the contralateral M1 compared to the ipsilateral M1 during hand movements. Stronger spinal activities were also found at the C7-T1 segment with the localization on the ipsilateral side. Notably, our regression analysis showed an functional and effective coupling between the M1 and the spinal cord, and the spinal activities were precisely tuned by bilateral M1s through the crossed and uncrossed CSTs. Conclusion: A developed cerebrospinal fMRI scanning technique enabled us to measure neural activities in the M1 and the spinal cord (i.e. CSTs), simultaneously. We first demonstrated a cerebrospinal functional coupling in human, noninvasively. Our developed method could offer the potential to provide clinically useful information for not only the functional diagnosis but also prognostic parameters of neural recovery in the patients with spinal cord injury, cervical myelopathy, and the other spinal cord diseases. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

S797 group showed lower EQ-5D scores (LSS+ group: mean 0.778 ± 0.215 vs. LSS- group: 0.840 ± 0.198, p = 0.004) with the disability of mobility and usual activities (e.g. work and housework), and the participants with LSS had more pain/discomfort. Conclusion: LSS was associated with lower muscle mass of legs, which may cause walking disability and increase the risk of a fall. The presence of LSS was also associated with a decline in the HRQOL. The current study suggests that early diagnosis and therapeutic intervention (e.g. physiotherapy) may reduce the risk of a fall and improve the HRQOL in the aged population with LSS. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

P133 PREOPERATIVE COMPUTED TOMOGRAPHIC (CT) SCAN TEMPLATING IN ADOLESCENT IDIOPATHIC SCOLIOSIS (AIS): IS IT REALLY NECESSARY? Mark Tan, Reuben Soh, Wilona Lee KK Women’s and Children’s Hospital, Singapore

P132 DECREASED MUSCLE MASS OF LOWER EXTREMITIES IN LUMBAR SPINAL STENOSIS: IS IT ASSOCIATED WITH THE INCREASED RISK OF A FALL AND THE LOSS OF HEALTH-RELATED QUALITY OF LIFE? Eiji Takasawa, Haku Iizuka, Yoichi Iizuka, Yasunori Sorimachi, Tokue Mieda, Kenji Takagishi Department of Orthopaedic Surgery, Gunma University Graduate School of Medicine, Maebashi, Japan Introduction: Lumbar spinal canal stenosis (LSS) is a very common in the aged population. LSS can lead to walking disability and a fall, which may result in loss of health. However, there is still little information regarding the association between the risk of a fall and LSS. We hypothesized that low muscle mass, function and strength (e.g. sarcopenia) may be a notable factor of walking disability and a fall in LSS. The purpose of this study was to investigate the association between LSS, muscle mass and a fall with the assessment of the heath-related quality of life (HRQOL). Methods: Medical checkups were conducted among the residents of a mountain village. The participants consisted of 187 residents, including 69 males and 118 females with a mean age of 69.0 ± 9.8 years (from 51 to 90-year-old). LSS was diagnosed by a diagnostic support tool for LSS (Konno, et al. 2007), and the participants were divided into two groups (LSS+ group; LSS- group). The participants completed a questionnaire including information on age, gender, the presence of low back pain and fall-experiences in the past year. We analyzed the muscle mass amounts in their body and extremities by Tanita MC-780 body composition analyzer (Tanita Corp., Tokyo, Japan), and we calculated the standardized muscle index (SMI). The HRQOL was also assessed by EuroQol (EQ-5D/VAS) questionnaire. All statistical analyses were conducted with the SPSS Statistics 19 software (IBM Japan, Ltd, Tokyo, Japan), and the critical value for significance was set at p \ .05. Results: The prevalence of LSS was 20.3 %, and LSS increased with age. Gender and low back pain had no association with LSS. In the LSS+ group, the SMI significantly decreased, especially in the lower extremities, and fall-experiences in the past year were more frequent (LSS+ group: 64.2 % vs. LSS- group: 38.9 %, p = 0.045). LSS+

Background: Surgeons may utilise CT scans of the thoracolumbar spine in the preoperative planning for spinal fusion in AIS. Information provided by these scans includes pedicle height, width, pedicle length and the degree of vertebral rotation. This aids cephalocaudal and medial-lateral pedicle trajectory planning. There are concerns of the routine use of the CT scans in AIS patients due to the high dose of ionizing radiation to a developing child. There is also criticism that intraoperative pedicle trajectories differ as scans are often performed in supine position without general anaesthesia or a surgical support cushion. Aims: The purpose of this study was to compare pedicle width, height and length measurements, and apical vertebral rotation angles obtained on plain radiographs (x-rays) to those evaluated on CT scans. Methods: All AIS patients who underwent spinal fusion at our center between 2012 and 2014 were retrospectively reviewed. Each had standard standing, lateral and lateral bending posterior-anterior radiographs, as well as CT scans performed prior to surgery. The largest T12 and L1 pedicle width and height amongst the standing and lateral bending x-rays was measured and compared to the same pedicle on the CT scan. The pedicle length (distance between the L1 lamina and anterior border of the vertebra on lateral film) was correlated to the pedicle length on axial CT cuts. Results: A total of 54 patients were analyzed. The average cobb of the primary curve was 61.49 degrees (39.4 to 96.4 degrees). Measurements of pedicle height and width at the T12 and L1 vertebrae on x-rays showed good correlation with those measured on CT scans. (p \0.001). The T12 pedicle width showed a mean difference of 0.57 mm and the pedicle height showed a mean difference of 3.11 mm. The L1 pedicle width had a mean difference of 0.38 mm, while the L1 pedicle height showed a mean difference of 2.89 mm. The mean difference between the pedicle length measured on x-rays and CT scan was 5.50 mm. Bland-Altman plots showed acceptable variation in values of these parameters measured. Conclusion: This study supports the use of preoperative x-rays to obtain measurements in the estimation of pedicle screw size and insertion. Given that CT scans have the disadvantage of ionizing radiation and cost, it is reasonable to avoid routine CT imaging of the spine in children during the preoperative planning of spinal fusion in AIS. Disclosures: author 1: none; author 2: none; author 3: none.

123

S798 P134 EVALUATION OF THE INFLUENCE OF DIFFERENT ARMS POSITION ON SAGITTAL SPINAL ALIGNMENT USING VIDEO-RASTERSTEREOGRAPHY Lucrezia Tognolo, Massimiliano Mangone, Teresa Venditto, Marco Paoloni, Valter Santilli Department of Physical and Rehabilitative Medicine, University ‘‘Sapienza’’, Rome, Italy Despite x-rays represent the gold standard to assess three-dimensional spinal alignment, a number of new, alternative technologies are now available for spinal postural assessment. Among them, video-rasterstereography is a valid and reliable technique, based on light emission, which allows the reconstruction of the back surface of the body and the calculation of frontal, transverse and sagittal spinal parameters. The aim of the study was to evaluate by means of rasterstereography the influence of different arm positioning on sagittal alignment of the spine in a population of healthy volunteers. A total of 30 adult volunteers were enrolled in this study. Each subject has been consecutively evaluated in normal standing position, with arms along the body, then with active arm elevation in 30°, 60°, 90°, and 120° position. Values for each test position were compared within subjects using repeated measures ANOVAs. All experimental position resulted in a negative shift in trunk inclination from a positive value of 14 mm to a negative one of -28 mm. The fle`che cervicale value progressively decreased from 60° position (41mm) compared with control position (60 mm). Kyphotic angle significantly decreases from 50,4° to 40,9°. Finally, the fle`che lombaire increases in value with the change in arm positioning (from 43 mm to 71 mm). Although x-rays represent the gold standard to assess three-dimensional spinal alignment, the elevation of the arms during acquisition of a lateral spinal radiograph produces significant changes in sagittal spinal parameters. The analysis of sagittal alignment performed by X-ray is influenced by the position of the arms, so it is impossible to compare tests performed in different positions. With rasterstereography we overcome these technical difficulties and we evaluate the same parameters of full spine lateral radiograph. We demonstrate that two spinal parameters, the trunk inclination and the fle`che cervicale, are hardly influenced by arms positioning. There is also a progressive decrease of the fle`che lombaire, which becomes more marked with 120° shoulder flexion position. These changes are probably due to the compensatory trunk extension which counterbalances arms elevation in order to maintain the Center Of Mass over the base of support. Between tested positions, the 30 ° seems to be the more representative of the reference position and the most suitable also for the radiographs to describe the sagittal alignment. Disclosures: author 1: none; author 2: none; author 3: grants/research support; Company=Sapienza University; author 4: none; author 5: none.

Eur Spine J (2015) 24 (Suppl 6):S743–S800 Spine Department. Fundacio´n Jime´nez Dı´az University Hospital, Madrid, Spain Objective: There is no consensus on the treatment of chronic low back pain of disc origin in the medical literature. Recent publications have highlighted the success of intradiscal injection of methylene blue (IIMB), but it is not clear whether the results of these previous studies are of sufficient strength to warrant specific recommendations. The aim of this study is to evaluate and compare the efficacy of IIMB, caudal epidural steroid injections (CESI) and oral non-steroidal antiinflammatory drugs (NSAID) in reducing pain and improving the associated disability. Material and method: This is a quasi-experimental, ambispective comparative study of historical control groups of 73 patients with chronic low back disc pain and magnetic resonance evidence of degenerative disc disease according to Pfirrmann et al. description. 27 patients treated by means of IIMB (IIMB-Group), were compared with 25 patients treated with NSAIDs/rehabilitation (NSAIDs-Group) and 21 who received fluoroscopy guided caudal epidural steroid injections (CESI-Group) from a previous study published by the authors. All patients were clinically evaluated at 4, 12 and 24 weeks Results: The three groups were comparable with respect to demographic characteristics. Back pain measured by visual analog scale (VAS) reflected an improvement in all groups, being significant only in patients included in the IIMB-Group (p \ 0.05). Regarding the Oswestry Disability (ODI) questionnaire, the IIAM-Group of patients had a greater level of pretreatment disability (43.5 IIMB; 27.94 CESI; 31.26 NSAIDs). At the end of the follow-up, patients included in the IIMB-Group demonstrated greater functional recovery (35.00) that was statistically significant (p \ 0.05); patients included in the CESIGroup showed no significant improvement (26.24) (p [ 0.05); and patients included in the NSAIDs-Group showed functional worsening (35.04). At the end of follow-up, patients undergoing Interventional therapies (IIMB/CESI) showed a much higher level of global satisfaction than those who received only anti-inflammatory analgesic treatment, and even a high percentage of them would undergo the same treatment again. Conclusions: The results from the present study suggest that IIMB could be an effective alternative for the treatment of chronic low back pain of disc origin, achieving superior results in pain relief, quality of life and function improvement than those obtained with other therapies. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: grants/research support; Company=Biomet, consultant; Company=Spineart.

P136 BLEEDING PROFILE AND TRANSFUSION REQUIREMENTS IN ADULT SPINAL DEFORMITY SURGERY Lidia Mora, Alba Vila-Casademunt, Ferran Pellise´, M Jose´ Colomina, Montse Domingo-Sa`bat, Francisco Javier S. Pe´rez-Grueso, Ahmet Alanay, Emre Acaroglu, Joan Bago´, ESSG European Spine Study Group

P135 INTRADISCAL METHYLENE BLUE IN THE TREATMENT OF CHRONIC DISCOGENIC LOW BACK PAIN. COMPARATIVE AMBISPECTIVE STUDY

Hospital Vall Hebron, Barcelona; Vall Hebron Institut de Recerca, Barcelona, Hospital De La Paz, Madrid, Spain; Acibadem University School of Medicine, Istanbul; Ankara Spine Center, Ankara, Turkey

´ ngel R. Pin˜era, Fe´lix Tome´-Bermejo, Javier Cervera-Irimia, A Javier Melchor Duart Clemente, Luis Alvarez

Introduction: ASD surgery can be considered a complex procedure in terms of bleeding and its related morbidity and mortality. Since bleeding and transfusions were previously associated with more

123

Eur Spine J (2015) 24 (Suppl 6):S743–S800 postoperative complications, haemostatic balance maintenance is crucial to avoid high bleeding and unnecessary transfusions. The main objective of this study is to perform a bleeding and transfusion descriptive profile in ASD surgery. Methods: A retrospective analysis of prospectively collected data of all the patients from a single center [18y, with coronal Cobb[20° or Sagittal Vertical Axis[5cm or Pelvic Tilt[25° or Thoracic Kyphosis [60°, who consecutively underwent surgery was performed. Preoperative and postoperative laboratory variables, intraoperative data and transfusion requirements were evaluated. Total (intraoperative and drainages) Estimated Blood Loss (EBL, % of estimated blood volume: % of estimated blood volume: men 75 ml/kg, women 65 ml/kg) and transfusion requirements were stratified according to the sample distribution. EBL was divided into four quartiles: B30 %, 30.1-50 %, 50.1-80 %, [80 %, and transfusion requirements into two groups: limited or extensive (0-3 vs C4 units). Results: 75 patients, 57 females, mean age 50.1 y (SD 19.5) and who underwent 88 surgeries, met inclusion criteria. Osteotomies were performed in 46.6 % of cases; the mean number of fused levels was 10. Surgical time was 405.5 min (SD 141.8), with a mean EBL of 57.5 %. Surgical time correlated with intraoperative EBL (p \ .0001) and transfusion rate (p \ .0001, r = -.58). At least 1 unit was used in 70 % of cases. However, while 24.1 % of the sample was in the upper quartile ([80 %), only 18.1 % were included in the extensive transfusion group. All the lab variables were in range before surgery and only haemoglobin was below the normal range 24h postoperatively. Of all of them, only preoperative platelets correlated with intraoperative bleeding (p \ .05). Conclusion: Although ASD surgery is related to significant bleeding, with 24.1 % of the sample losing [80 % of their blood volume, only 18.1 % of them were included in the extensive transfusion group. Proper haemostatic balance maintenance is fundamental. Disclosures: author 1: none; author 2: grants/research support; Company=Depuy Synthes; author 3: grants/research support; Company=DePuy Synthes, K2M, consultant; Company=DePuy Synthes, Biomet; author 4: none; author 5: grants/research support; Company=DePuy-synthes; author 6: grants/research support; Company=DePuy Synthes, consultant; Company=De Puy Synthes; author 7: grants/research support; Company=Depuy Synthes, consultant; Company=Stryker; author 8: grants/research support; Company=Medtronic, Depuy Synthes, stock/shareholder; Company=IncredX; author 9: grants/research support; Company=Depuy Synthes; author 10: grants/research support; Company=DePuysynthes.

P137 RISK FACTORS OF GASTROINTESTINAL HEMORRHAGE AFTER LONG POSTERIOR INSTRUMENTATION AND FUSION IN DEGENERATIVE ADULT LUMBAR SCOLIOSIS Dingjun Hao, En Xie Department of Spine Surgery, Hong Hui Hospital, Xi’an Jiaotong University College of Medicine, Xi’an, China Study design: Retrospective database analysis. Objective: To determine rates of gastrointestinal hemorrhage (GIH) after lumbar fusions, a population-based database was analyzed to identify the incidence, mortality, and risk factors associated with anterior (ALF), posterior (PLF), and simultaneous anterior/posterior (APLF) lumbar fusions. Summary of background data: GIH after Long Posterior Instrumentation and Fusion in Degenerative Adult Lumbar Scoliosis is a

S799 rare complication that can have devastating consequences. Incidences of GI bleeding after lumbar fusion are not well characterized in the current literature. Methods: Data were obtained from 2002 to 2013. Patients undergoing long Posterior Instrumentation and Fusion in Degenerative Adult Lumbar scoliosis were identified and the incidence of GIH was evaluated. Patient demographics, Charlson Comorbidity Index, length of stay, costs, and mortality were assessed. SPSS version 20 (IBM; Armonk, NY) was used to detect statistical differences between groups and perform logistic regression analyses to identify independent predictors of GI bleeding. A P value of \0.001 denoted significance. Results: A total of 7871 Long Posterior Instrumentation and Fusion in Degenerative Adult Lumbar scoliosis were identified from 2002 to 2013. Of these, patients with GI bleeding demonstrated greater Charlson Comorbidity Index scores, length of stay, costs, and mortality (P \ 0.001). Logistic regression analysis demonstrated independent predictors of GIH including advanced age ([65 yr), male sex, blood loss anemia, fluid/electrolyte disorders, metastatic Neoplasm, and weight loss (P \ 0.001). Conclusion: The results of our study demonstrate very low complication rates of GIH after Long Posterior Instrumentation and Fusion in Degenerative Adult Lumbar Scoliosis. . Across all surgical procedures, the presence of GI bleeding complications was associated with greater comorbidity, length of stay, cost, and mortality. We strongly advise physicians to perform stringent perioperative assessments of risk factors and to provide prompt medical attention to minimize the impact of GI bleeding complications. Disclosures: author 1: none; author 2: none.

P138 RISK FACTORS OF SURGICAL SITE INFECTION IN ADULT DEGENERATIVE LUMBAR SCOLIOSIS: DETECTION AND MANAGEMENT BASED ON SERIAL PROEALEITONIN MEASUREMENTS: AN OPEN-LABEL RANDOMISED TRIAL En Xie, Dingjun Hao Department of Spine Surgery, Hong Hui Hospital, Xi’an Jiaotong University College of Medicine, Xi’an, China Object: The goal of this study was to determine the diagnostic significance of the PCT as a detector for early onset of surgical site infection in ADLS and to discuss effective medical treatment through clinical interpretation and application of the measured PCT values. Methods: An open-label randomised study was performed in 4787 consecutive cases involving patients who underwent ADLS between January and September 2014. Blood samples were obtained preoperatively and on postoperative days 1, 3, 5, 7 and 14 in patients undergoing single-level decompression surgery. An additional blood specimen was obtained at postoperative day 7 in patients requiring more extensive surgeries. Recorded laboratory results were compared with the patients’ clinical course to determine the diagnostic significance of the PCT. All of the patients received intravenous prophylactic antibiotic therapy. Once an abnormal response of PCT, indicated by a tendency toward continuous elevation, was important to note on Day 5 or Day 7, the prophylactic antibiotics were replaced with another regimen and administration was resumed along with careful observation for signs of surgical site infection. Results: Monitoring of PCT revealed a characteristic increase and decrease pattern in 4571 of 4787 patients (95.49 %) showing a normal clinical course with regard to early infectious complications. The mean measured PCT (reference range \ 0.05 ng/ml) averaged 0.157

123

S800

Eur Spine J (2015) 24 (Suppl 6):S743–S800

0.217 ng/ml on Day 1, 0.171 0.277 ng/ml on Day 3, and 0.78 0.171 ng/ml on Day 5. In contrast, there were 216 cases (4.51 %) of abnormal PCT responses resulting in the resumption of intravenous antibiotic treatment, which included a second rise (in 162 cases) and a steady rise (in 54) in the PCT value. Five (1.4 %) of 216 patients experienced infectious complications related to ADLS. 43 patients (0.9 %) received long-term antibiotic therapy for 4-6 weeks; however, all patients recovered with medical treatment alone and did not experience gross wound disruption or subsequent disciples. As a predictor for primary wound infection, the sensitivity, specificity, positive predictive value, and negative predictive value for abnormal PCT responses were calculated as 100 %, 96.7 %, 39.1 %, and 100 %, respectively. Conclusions: The above results demonstrate that PCT screening is a simple and reliable test for the detection of early infectious complications after ADLS. Disclosures: author 1: none; author 2: none.

Conclusions: The overall incidence of rod breakage following long adult scoliosis instrumentation and fusion to S1 was 15.3 %. Preoperative hyperkyphosis, increased number of fused level and lower implant density could increase the risk of rod breakage. Keywords: Rod Breakage; Incidence; Adult Scoliosis; Long Fusion Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

P139 ROD BREAKAGE IN LONG FUSIONS TO THE SACRUM FOR SCOLIOSIS IN ADULTS AGED FORTY YEARS OR OLDER: INCIDENCE AND RISK FACTORS

Wajokai Eniwa Hospital, Eniwa City, Hokkaido, Japan

Leilei Xu, Zezhang Zhu, Feng Zhu, Bangping Qian, Yong Qiu Spine Surgery, Drum Tower Hospital of Nanjing University Medical School, Nanjing, China Background: Long fusions to the sacrum for the treatment of adult spine deformity have always been a technical challenge for the surgeons. Several risk factors can be associated with rod breakage following long spinal fusion to sacrum in patients with adult scoliosis. Purpose: To determine the incidence and risk factors of rod breakage following long spinal fusion to sacrum in patients with adult scoliosis. Methods: 59 patients who underwent posterior spinal fusion surgeries for adult lumbar scoliosis from May 2005 to June 2011 were reviewed. The patients were classified into two groups, with patients in Group A having rod breakage and with patients in Group B having no rod breakage or other type of implant failure. A retrospective chart review was performed to collect baseline characteristics of the patient including age, gender, weight, height, etiology of scoliosis, type of instrumentation, levels of instrumentation, implant density, and type of posterior arthrodesis. 7 radiological parameters, obtained from preoperative and final follow-up, were evaluated for each subject, including curve magnitude, coronal balance (CB), global kyphosis (GK), lumbar lordosis (LL), pelvic incidence (PI), pelvic obliquity and sagittal vertical axis (SVA).20 Statistical comparisons were made between the two groups to determine the risk factors of rod breakage. Results: Rod breakage was identified in 9 patients (15.3 %), consisted of 8 female and 1 male. The average number of fused vertebrae in patients with rod breakage was 13.6 (range 7-16). Implant density was averaged 56.9 % ± 12.1 %. The mean time from surgery to the development of rod breakage was 29.8 ± 30.4 months (range 8 - 96 months). The result of comparison between the rod breakage group and non-breakage group showed that preoperative GK was significantly higher in the rod breakage group than in the non-breakage group (52.2°± 10.0° vs. 24.2° ± 23.8°, p = 0.007). Besides, patients in the rod breakage group were found to have significantly more fused vertebra (13.6 ± 2.9 vs. 10.1 ± 3.7, p = 0.018) and remarkably lower implant density (56.9 % ± 12.1 % vs. 76.5 % ± 14.1 %, p = 0.001).

123

P140 STENOTIC RATIO IS THE MOST USEFUL PARAMETER TO IDENTIFY PATIENTS WITH LUMBAR FORAMINAL STENOSIS NEEDING SURGERY OR NOT -MRI STUDY USING 3D T1 SPACE SEQUENCE Kentaro Yamada, Yuichiro Abe, Yasushi Yanagibashi, Takahiko Hyakumachi, Shigenobu Satoh

Objective: Lumbar foraminal stenosis (LFS) is still difficult to diagnose, especially in cases of comorbid LFS and superior adjacent lumbar canal stenosis, so-called double crush of lumbar nerve root. Several imaging techniques have been proposed to diagnose LFS. However, the differences between LFS patients needing surgery and those who did not need surgery have not been discussed. We focused on the stenotic length in the neural foramen. The purpose of this study was to compare among patients needing surgery, those with successful conservative treatment, and asymptomatic patients for LFS at L5-S using 3D MRI. Materials and methods: Patients who were taken MRI of 3D T1 SPACE sequence with a 3.0-T scanner to diagnose of radicular leg pain were enrolled in this study. Patients under 40 years old and those with prior lumbar surgery, acute trauma or inflammatory disease, or suspicion of double crush of L5 root by Schizas classification were excluded. A total of 45 patients, 90 foramina at L5-S, were investigated. LFS was defined positive at a slice of perineural fat obliteration on the vertical plane to the oblique images along L5-S foramen. Stenotic ratio (SR) was defined as a ratio of stenostic length to the length of the foramen on the oblique image. The minimum diameters of the nerve root (d) and the maximum transverse angles of nerve root (a) were also investigated on the reconstructed images. Each foramen at L5-S was divided into 3 groups: LFS needing surgery (Group 1) 15 foramina, LFS with successful conservative treatment (Group 2) 10 foramina, and asymptomatic foramen (Group 3) 65 foramina, to compare SR, d and a. The reliability was evaluated by intraclass correlation coefficient (ICC) among randomly chosen 10 samples. Results: Significant difference between Group 1 and 2 were only found in SR. SR showed a stepwise increase from Group 3 to Group 1 (Group 1: 49.8 ± 23.7 %, Group 2: 30.5 ± 15.0 %, Group 3: 7.0 ± 10.4 %). There were significant differences between Group 1 and 3, and Group 2 and 3 in all parameters: SR, d, and a. The ICC in SR for intraobserver error and interobserver error were almost perfect (0.97 and 0.95, respectively). Conclusions: The result suggested the all investigated parameters were useful to detected symptomatic LFS. However, only SR could identify LFS needing surgery from symptomatic LFS. The SR might be useful to identify which lesion is responsible for the patient’s symptom in case of double crush. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.