Eur Spine J (2015) 24 (Suppl 6):S669–S710 DOI 10.1007/s00586-015-4129-1

ABSTRACTS

ORAL PRESENTATIONS EUROSPINE 2015 THORACO-LUMBAR SPINE, DEGENERATIVE I

1 ANNULAR REPAIR USING FIBROCHONDROCYTES SEEDED IN A CROSS-LINKED COLLAGE GEL: IN VIVO OUTCOME IN A RODENT MODEL Yu Moriguchi, Brandon Borde, Peter Grunert, Thamina Khair, Lawrence Bonassar, Roger Ha¨rtl Weill Cornell Medical College, New York, USA Background: Despite alleviating associated neurological symptoms, discectomy of herniated intervertebral discs (IVDs) fails to repair the underlying degenerative process. Persistent annular defect post discectomy is associated with increased risk of reherniation, progressive IVD degeneration, and chronic low back pain. Although many studies have previously employed cell transplantation to regenerate nucleus pulposus (NP), few have focused on annulus fibrosus (AF) repair. We recently demonstrated that riboflavin crosslinked high-density collagen gels (HDC) can facilitate annular repair in vivo. In this study, we assessed the in vivo efficacy of fibrochondrocytes seeded in the gel scaffold in enhancing the reparative process at the site of annular defect to prevent further degenerative changes in a post-puncture rattail model. Methods: 24 athymic rats, punctured with an 18-gauge needle in the tail disc, were divided into 3 groups: untreated (n = 4), injected with crosslinked HDC (n = 10), and injected with fibrochondrocyte-laden crosslinked HDC (n = 10). AF fibrochondrocytes, harvested from sheep spine, were labeled with DiI and mixed with HDC gels prior to injection at a concentration of 106 cells/ml. At 1, 2, and 5 weeks postoperatively, a series of in vivo imaging with X-ray and 7T MRI were conducted to determine disc height index and nucleus pulposus (NP) size and hydration, respectively [Grunert P, Spine 2014]. Histological assessments were performed to evaluate the viability of implanted cells, degree of annular repair and secondary disc degeneration. Results: At 1 week, both HDC gel groups showed significantly higher retention of NP size and hydration, while only the cell-laden group better maintained NP retention and disc height up to 5 weeks. The untreated discs showed substantial NP herniation by 2 weeks and NP absence with signs of degeneration by 5 weeks. Histological assessments indicate that while gels influence the sealing of the defect, the addition of cells generates abundant tissue

growth and extracellular matrices at the site of annular defect, accelerating the reparative process by as early as 2 weeks. DiI detected implanted fibrochondrocytes in the process of reorganization within gels. Conclusions: We are the first to demonstrate that AF fibrochondrocyte-laden HDC gels have the ability to better repair annular defects than acellular gels, thereby suggesting their potential feasibility in cell-based therapy for annular defects. Disclosures: author 1: none; author 2: none; author 3: grants/research support; Company = AO-Spine; author 4: none; author 5: stock/ shareholder; Company = 3D BioCorp; author 6: grants/research support; Company = baxter, consultant; Company = Depuy Synthes.

2 PESSIMISTIC BACK BELIEFS AND LACK OF EXERCISE: A RISKY COMBINATION FOR FUTURE SHOULDER, NECK, AND BACK PAIN Anne F. Mannion, Urs Mu¨ller, C. Rolli, O. Tamcan, Achim Elfering Spine Center, Schulthess Klinik, Zurich; Institute for Psychology, University of Bern; Institute for Evaluative Research in Orthopaedic Surgery, University of Bern, Switzerland Introduction: Beliefs regarding the inevitability of the future as a consequence of having back trouble are known to be associated with pain maintenance and pain-related absenteeism, presenteeism and use of the healthcare system. The aim of this cross-lagged-panel study was to assess the longitudinal validity of the back beliefs questionnaire (BBQ) in predicting future pain in the shoulders, neck, and back. Methods: A questionnaire booklet containing questions on demographics, lifestyle, pain, and the BBQ was completed by 2507 randomly selected individuals who were otherwise participating in a population-based cross-sectional survey of musculoskeletal health. One year later, 1833 (73 %) returned a follow-up questionnaire. Structural equation modelling was used to evaluate associations between baseline BBQ scores and the intensity of pain in the shoulders, neck, and back (in the preceding 4 weeks) at 1-year follow-up, while accounting for pain in these regions at baseline, and to determine whether the models were influenced by the presence of back pain and exercise habits at baseline. Results: Baseline BBQ scores significantly predicted the intensity of 1-year shoulder, neck, and back pain in individuals who reported having back pain at baseline (b = 0.11, p \ .05) but not in those who didn’t (b = 0.02, p = .259). Baseline BBQ also predicted 1-year shoulder, neck, and back pain in those who didn’t exercise (b = 0.15, p \ .05) but not in those who did (b = 0.04, ns). The predictive strength of back beliefs for future shoulder, neck, and back pain

123

S670 intensity was greatest in individuals who both had back pain and did not exercise at baseline (b = 0.29, p \ .05). Discussion: The findings confirm the longitudinal validity of the BBQ. Cognitive behavioural interventions should address pessimistic back beliefs in high-risk groups such as those experiencing back pain who are not regular exercisers. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 4 MINI-OPEN VERSUS OPEN TRANSFORAMINAL INTERBODY FUSION FOR LUMBAR DEGENERATIVE DISORDERS: A COMPARATIVE EFFECTIVENESS STUDY Frank S. Kleinstu¨ck, Daniel Haschtmann, Tamas F. Fekete, Dezso¨ Jeszenszky, Anne F. Mannion Schulthess Klinik, Zu¨rich, Switzerland

3 THE PROGNOSIS OF L5 RADICULOPATHY AFTER REDUCTION AND INSTRUMENTED FUSION OF ADULT ISTHMIC HIGH-GRADE LUMBOSACRAL SPONDYLOLISTHESIS AND THE ROLE OF MULTIMODAL INTRAOPERATIVE NEUROMONITORING Ralph T. Scha¨r, Martin Sutter, Anne F. Mannion, Andreas Eggspu¨hler, Dezso¨ Jeszenszky, Tamas Fekete, Frank Kleinstu¨ck, Daniel Haschtmann Department of Neurosurgery, Inselspital, University of Bern, Bern, Switzerland; Spine Center, Schulthess Clinic, Zurich, Switzerland Objective: Iatrogenic L5 nerve root injury is a common complication of reduction of isthmic high-grade spondylolisthesis (HGS). However, the clinical presentation and prognosis as well as its impact on patient-rated outcome have never been analysed systematically and the role of multimodal intraoperative neuromonitoring (MIOM) is largely unknown. Methods: Clinical and radiographic data from consecutive patients who underwent reduction and instrumented fusion of HGS between 2005–2013 in a single center were analysed. The multidimensional Core Outcome Measures Index (COMI) was completed before and after surgery. MIOM with transcranial MEP and continuous EMG of index muscles was carried out. Results: 17 patients (13f, 4m) with a mean radiographic follow-up of 19 (range 3–48) months were included. Mean (±SD) age was 26.5 (±9.2) years. The mean L5–S1 slip was 72 % (±21 %), which was reduced to 19 % (±13 %). After 1 year, COMI improved from 6.7 ± 1.7 to 3.7 ± 3.1, leg pain from 5.2 ± 3.1 to 2.3 ± 3 and back pain from 6.2 ± 1.9 to 3.4 ± 2.6. In 5 patients (29 %) an incomplete L5 motor deficit occurred: 2 with 3/5 paresis or worse, and 3 with 4/5. 4 of these patients fully recovered after 3 months, and 1 (5.9 %) was lost to follow-up (FU). In 15 out of 17 procedures, 25 intraoperative MIOM alerts were recorded. Based on intraoperative recovery of the signals, MIOM predicted 1 new neurological deficit and 16 patients without deficits. The patient with the MIOM true positive deficit was the one lost to FU. Related to the long-term outcome of patients with FU, MIOM has a sensitivity and specificity of 100 %. However, in predicting the early postoperative and mostly transient L5 motor deficits [occurring in 5/17 (29 %)], the sensitivity was only 20 %. Conclusions: Reduction and instrumented fusion of HGS showed a satisfactory outcome. The rate of transient L5 palsy was relatively high. However, the prognosis for this deficit is favorable and full recovery can be expected after 3 months in most cases. MIOM alerts occurred in 88 % of the procedures and thereby possibly limited the prevalence of neurological deficits to 5.9 %. Intraoperative recovery of deteriorating MIOM signals cannot predict transient neurological deficits. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: consultant; Company = DePuy Synthes, royalties; Company = DePuy Synthes; author 6: none; author 7: none; author 8: none.

123

Introduction: Mini-open transforaminal lumbar interbody fusion (mTLIF) through a Wiltse approach is a less invasive modification of the conventional open TLIF (cTLIF). Previous comparative studies of mTLIF and cTLIF have been beset with design flaws, such as low sample size, non-systematic data collection, selection bias, and confounding. We addressed the issue by analysing prospectively collected data from consecutive patients in our local database within the framework of Eurospine’s Spine Tango Registry. Methods: Patients undergoing mTLIF for 1–2 level degenerative disorders from 2011–2014 were compared with historical controls (cTLIF, same surgeons) from 2005–2011. Medical history, surgical procedures, and complication rates were recorded systematically using Spine Tango surgery forms (98 % completion rate). Patients completed the Core Outcome Measures Index (COMI) before and at 3 mo (98 % follow-up) and 12 mo (95 %) after surgery. Multiple regression was used to evaluate the effect of treatment group, whilst adjusting for age, gender, ASA, number of levels, prior surgery and decompression. Results: 154 patients were identified: 101 cTLIF (age 59 ± 13 y; 41 % M) and 53 mTLIF (age 57 ± 13 y; 45 % M). mTLIF was associated with lower blood loss (by mean 380 ml; p \ 0.0001) and a shorter stay (by mean 3.5 days; p \ 0.0001). Operation time tended to be shorter (by mean 0.14 h; p = 0.28) for mTLIF. General medical complications tended to be lower in the mTLIF than the cTLIF group (4 and 13 %, respectively; p = 0.13). There were no significant group differences in COMI scores at either 3mo (p = 0.11) or 12mo followup (p = 0.23). Discussion: Our study confirms the previously reported surgical benefits of mTLIF in a consecutive series with systematic, prospective, and almost complete data collection. mTLIF enjoys the advantages of a less invasive approach, complemented by comparable patient outcomes to cTLIF, suggesting it is a valid alternative to the conventional open approach. Disclosures: author 1: none; author 2: none; author 3: none; author 4: consultant; Company = DePuy Synthes, royalties; Company = DePuy Synthes; author 5: none.

5 GLOBAL TILT AND LUMBAR LORDOSIS INDEX: TWO PARAMETERS TO UNDERSTAND POSITIVE BALANCE ANALYSIS Louis Boissiere, Jean-Marc Vital, Vincent Challier, Ahmet Alanay, Ferrran Pellise´ Urquiza, Francisco Javier Sanchez Perez-Grueso, Emre Acaroglu, Frank Kleinstu¨ck, Ibrahim Obeid, ESSG European Spine Study Group Spine Unit 1, Bordeaux University Hospital, Bordeaux, France; ESSG, Hospital University Vall Hebron, Barcelona, Spain Introduction: In Adult Spinal Deformity (ASD), symptoms severity increases in a linear fashion following positive balance. Many radiologic parameters were reported including Sagittal Vertical Axis

Eur Spine J (2015) 24 (Suppl 6):S669–S710 (SVA), Pelvic Tilt (PT) and PI-LL. European literature, reports other parameters such as Lumbar Lordosis Index (LLI) and Global Tilt (GT). LLI (LL/PI ratio) demonstrated a therapeutic impact in ASD management, and a 0.5 value correlated with major sagittal malalignment. GT combines the C7 vertical tilt and PT, and is less affected by patient positioning than SVA or PT. It offers the possibility to understand spinal and pelvic mismatches when SVA and PT seem opposite. The purpose of this study was to validate these parameters by investigating their correlation with health related quality of life (HRQL) scores and other sagittal radiographic parameters in patients with ASD. We will also investigate the relationship between these parameters and HRQL scores when patients are stratified by SVA to understand the disability induced by each parameter. Methods: In this, multicenter cross-sectional analysis study, inclusion criteria were operated adults presenting any of the following radiographic parameter: Cobb C20°, SVA C5 cm, Thoracic kyphosis C60° or PT C25°. HRQL scores and radiographic parameters were collected at baseline. A p \ 0.05 value was considered significant. Results: 471 preoperative patients were included. LLI correlated with SVA (-0.65), GT (-0.77), PT (-0.73), PI-LL (-0.98). GT correlated with SVA (0.80), PT (0.91), PI-LL (0.80). LLI and GT correlated with HRQL questionnaires (-0.34 and 0.41 with ODI respectively). When patients were divided in SVA subgroups following Schwab’s classification correlations with HRQL scores significantly decreased for all parameters. Conclusion: GT correlates the most with HRQL-scores and LLI correlates better than PI-LL in this series. These results also underline the high correlation between all these radiologic sagittal parameters. However, when patients are stratified by SVA, correlation tends to disappear demonstrating that these parameters (GT, LLI, PI-LL, PT, SVA) describe the same pathology representing only part of disability. Their importance relies more on the purpose they describe to understand the malalignment mechanisms. Disclosures: author 1: none; author 2: consultant; Company = Depuy, Euros, royalties; Company = LDR; author 3: none; author 4: grants/research support; Company = Depuy Synthes, consultant; Company = Stryker; author 5: grants/research support; Company = dePuy Synthes, K2M; author 6: grants/research support; Company = DePuy Synthes, consultant; Company = De puy synthes; author 7: grants/research support; Company = Medtronic, Depuy Synthes, stock/shareholder; Company = IncredX; author 8: none; author 9: grants/research support; Company = Depuy Synthes, consultant; Company = Depuy Synthes Medtronic; author 10: grants/ research support; Company = DePuy-synthes.

6 THORACIC CURVE IN ADULT IDIOPATHIC SCOLIOSIS: A DIMINISHING PROBLEM WITH INCREASING AGE Javier Pizones, Mar Pe´rez Martı´n-Buitrago, Francisco Javier Sa´nchez Pe´rez-Grueso, Ferra´n Pellise´, Ahmet Alanay, Ibrahim Obeid, Frank Kleinstu¨ck, Emre Acaroglue, ESSG European Spine Study Group

S671 Methods: Retrospective analysis of data collected prospectively in an adult deformity database containing 1035 patients with coronal deformity. The database was searched to identify patients with main thoracic idiopathic curves, defined as: Schwab type T curves and Schwab type D in which TL/L curves were \40° and the difference between MT-TL/L Cobb was C15°. Demographic data, and eleven different coronal and sagittal radiological parameters were assessed. HQRL questionnaires at minimum 2 years’ follow-up were evaluated. Results: The consultation rate was 4.1 % (42 patients). The group showed the following characteristics: age, 30.9 ± 12.5 years; thoracic Cobb, 55.6° ± 10.8°; Lumbar Cobb, 28.1° ± 7.3°; SVA, 2.9 ± 19.3 cm; Global tilt, 12.2° ± 7.6°; COMI, 4.0 ± 2.5; ODI, 20.4 ± 17.4; SRS-22, 3.6 ± 0.7; SF-36 mental-health, 46.1 ± 10.1; SF-36 physical-health, 47.3 ± 11.1. The incidence of surgery for thoracic deformity was 1.26 % (13 p). Only patients under 50 years underwent surgery, and compared with non-operated patients they were younger (24.3 ± 7.3 vs 33.8 ± 13.4 years; P = 0.009), had larger MT curves (58.7° ± 9.6° vs 50.6° ± 8.3°; P = 0.012), and had worse SRS-22 self-image scores (2.9 ± 0.8 vs 3.5 ± 0.8; P = 0.042). No patients over 50 years underwent surgery, despite having worse SRS-22 function (3.0 ± 0.9 vs 4.1 ± 0.9, P = 0.032), worse ODI scores (42.4 ± 19.9 vs 18.7 ± 18.0, P = 0.026), and a trend towards worse Cobb angle (67.8° ± 10.1° vs 58.7° ± 9.6°; P = 0.075) compared with operated patients under 50 years of age. Conclusions: Very few adult patients (4.1 %) consulted for treatment because of a thoracic scoliotic curve. The probability of undergoing surgery was very low (1.26 %), and it was associated with youth, curve magnitude and poor self-image. The need for surgical treatment of the thoracic curve appears to diminish with age, despite it being associated with poorer function and greater disability in the older patient. Disclosures: author 1: grants/research support; Company = Johnson and Johnson (Depuy Spine); author 2: grants/research support; Company = Depuy Synthes; author 3: grants/research support; Company = DePuy Synthes, consultant; Company = De Puy Synthes; author 4: grants/research support; Company = DePuy Synthes, K2M, consultant; Company = DePuy Synthes, Biomet; author 5: grants/research support; Company = Depuy Synthes, consultant; Company = Stryker; author 6: grants/research support; Company = Depuy Synthes, consultant; Company = Depuy Synthesmedtronic; author 7: none; author 8: grants/research support; Company = Medtronic, Depuy Synthes, stock/shareholder; Company = IncredX; author 9: grants/research support; Company = DePuySynthes.

7 IDENTIFYING THE BEST TREATMENT IN ADULT SPINAL DEFORMITY (ASD): A DECISION ANALYSIS (DA) APPROACH Emre Acaroglu, Aysun Cetinyurek-Yavuz, Umit Ozgur Guler, Selcen Pehlivan, Yasemin Yavuz, Montse Domingo-Sabat, Ferran Pellise, Francesco Sanchez Perez Grueso, Ahmet Alanay, Frank Kleinstu¨ck, Ibrahim Obeid, ESSG

Hospital Universitario de Getafe, Madrid, Spain Introduction: Idiopathic thoracic curves are a big concern during adolescence. If left untreated, Cobb angles over 50° progress, produce functional impairment and chronic back pain, and influence self-image even 50 years after initial diagnosis. However, our hypothesis is that only few patients consult because of thoracic curves in adulthood. Our aim was to report the rate of consultation for untreated thoracic curve in adult idiopathic scoliosis, analyze patient characteristics, and establish the rate and risk factors for surgical intervention; facts that are still unknown.

Ankara Spine Center, Ankara TR; Clinistats EU, Zurich, CH; Hospital Universitari Vall d’Hebron, Barcelona, E; Acibadem Maslak Hospital, Istanbul, TR; Hospital Universitari La Paz, Madrid, E; Schultess Clinic, Zurich, CH; Bordeaux University Hospital, Bordeaux, France Background: Adult spinal deformity (ASD) is a major public health problem. There are pros and cons of the available treatment alternatives (surgical or non-surgical) and it had been difficult to identify the best treatment modality.

123

S672 Aim: To construct a statistical DA model to identify the optimum overall treatment in ASD. Methods: From an international multicentre database of ASD patients (968 pts), 535 who had completed 1 year follow-up (371 non-surgical -NS, 164 surgical -S), constitute the population of this study. DA was structured in two main steps of: (1) Baseline analysis (Assessing the probabilities of outcomes, Assessing the values of preference -utilities-, Combining information on probability and utility and assigning the quality adjusted life expectancy (QALE) for each treatment) and (2) Sensitivity analysis. Results: Four hundred and thirty-two patients (309 NS, 123 S) had baseline and 1 year follow-up ODI measurements. Overall, 104 (24.1 %) were found to be improved (a decrease in ODI[8 points), 225 (52.1 %) unchanged (-8 [ ODI [ 8) and 65 deteriorated. Surgery presented with a higher chance of improvement (54.2 %) vs. NS (9.7 %). The overall QALE ranged from 56 to 69 (of 100 years) and demonstrated better final QALE in the NS group (60 vs 65, p = 0.0038), this group having started with higher QALE as well (56 vs 65 years, p \ 0.0001). There were improvements in overall QALE in both groups but this was significant only in the surgical group (S from 56 to 60 years, p \ 0.0001; NS from 65 to 65 years, p = 0.27). In addition, in the subgroup of patients with significant baseline disability (ODI [ 25) surgery appeared to yield marginally better final QALE (58 vs 56 years, p = 0.1) despite very a similar baseline (54 vs 54 years, p = 0.93). Discussion and conclusions: This study demonstrated that a single best treatment modality for ASD may not exist. Conservative treatment appears to yield higher (up to 6 %) QALE compared to surgery, most probably secondary to a higher baseline QALE. On the other hand, surgery provides a significantly higher increase in QALE. Especially in patients with significant disability at baseline, the final QALE tended higher in the S group (although not significant). Finally, chances of a relevant improvement at 1st year turned out to be significantly lower with NS treatment. Disclosures: author 1: grants/research support; Company = Medtronic, Depuy Synthes, stock/shareholder; Company = IncredX; author 2: none; author 3: grants/research support; Company = Eurospine; author 4: consultant; Company = Yildirim Beyazit University; author 5: none; author 6: grants/research support; Company = Depuy-synthes; author 7: grants/research support; Company = DePuy Synthes, K2M, consultant; Company = DePuy Synthes, Biomet; author 8: grants/research support; Company = DePuy Synthes, consultant; Company = DePuy Synthes; author 9: grants/research support; Company = Depuy Synthes, consultant; Company = Stryker; author 10: grants/research support; Company = DePuy-synthes;

8 POSTERIOR VERTEBRAL COLUMN RESECTION (PVCR) FOR THE MANAGEMENT OF SHARP ANGULAR KYPHOTIC DEFORMITY IN ADULT POPULATION Meric Enercan, Sinan Kahraman, Bahadir Gokcen, Sinan Yilar, Tunay Sanli, Erden Erturer, Cagatay Ozturk, Ufuk Talu, Azmi Hamzaoglu Istanbul Spine Center, Istanbul Florence Nightingale Hospital, Istanbul, Turkey Summary: PVCR performed for the treatment of sharp angular kyphosis provides spinal cord decompression, corrects angular sharp deformity and restores the sagittal balance. Correction technique of the sharp angular kyphosis should include anterior column elongation with gradual posterior compression sequentially and placement of an expandable cage anteriorly for safe and ideal correction.

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Design: Retrospective. Introduction: The management of sharp angular kyphosis (SAK) can be a challenging since correction of rigid deformity is technically difficult and often require combined approaches or major spinal osteotomy. Pain, progressive deformity causing sagittal imbalance and deterioating neurological deficit are major problems. The purpose of this study is to evaluate the results of PVCR in the treatment of SAK in adult population. Methods: 35 pts (21 M,14F), av age 42.1 years (19–74) who underwent PVCR for sharp angular kyphotic deformity were included. Following PVCR, correction technique included anterior column elongation with gradual posterior compression sequentially and placement of an expandable cage anteriorly to prevent any dural buckling. Preop, postop and f/up X-rays were evaluated for radiological data including local kyphosis angle (LKA), sagittal and pelvic parameters. Functional status of the patients were assessed by Oswestry score. Results: Ave f/up was 47 months (24–120). Etiologies were posttraumatic kyphosis for 24 pts and neglected congenital kyphosis in 11 pts. Osteotomies were grade 5 in 27 pts and grade 6 resections in 8 pts according to Schwab’s Classification. Preoop ave LKA of 49.52° improved to 7.35° with a correction rate of 89 %. Preop SSA of av 118.3° was restored to 132.7°. 18 pts who had preop neurologic deficit(11 ASIA D, 5 ASIA C, 2 ASIA B) had at least one grade improvement at the final f/up. Minor complications included 6 (17 %) dural tears, 4 (11 %) superficial wound problem which responded well to debridement. Oswestry functional scores decreased from a mean of 56–16. Solid fusion was achieved in all patients without significant loss of correction in the sagittal plane at the final f/up. Conclusion: Correction of kyphosis and restoring sagittal balance is very challenging in SAK. PVCR provides spinal cord decompression, improves neurological deficit and quality of life. Anterior column elongation with gradual posterior compression sequentially and placement of expandable cage corrects local angular kyphosis and restores sagittal balance 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: consultant; Company = Medtronic. CERVICAL SPINE

9 CERVICAL KYPHOSIS DOES NOT IMPLY CERVICAL DEFORMITY: PREDICTING CERVICAL CURVATURE REQUIRED FOR HORIZONTAL GAZE BASED ON SPINAL GLOBAL ALIGNMENT AND THORACIC KYPHOSIS Vincent Challier, Bassel Diebo, Jonathan Oren, Shaleen Vira, Barthelemy Liabaud, Renaud Lafage, Jensen Henry, Themistocles Protopsaltis, Frank Schwab, Virginie Lafage NYU Hospital for Joint Diseases, New York, USA Hypothesis: Cervical kyphosis may be a physiologic alignment necessary for the maintenance of horizontal gaze depending on underlying thoracolumbar (TL) alignment. Design: Retrospective review. Introduction: Cervical curvature (CC) is affected by thoracic and global alignment. Recent studies suggest large variability in normative CC ranging from lordotic to kyphotic alignment. This study investigates the effect of thoracic and global alignment on CC in maintenance of horizontal gaze.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Methods: For formula development, full body xrays of 744 patients (pts) without presenting cervical complaints or existing fusions higher than T3 were studied. Only pts who maintained their horizontal gaze (CBVA -5° and 17° or McGregor’s slope between -6° and 14°) were included. Pts were stratified based on thoracic kyphosis (TK) into ([50, 40–50, 30–40 and \30). Pts were sub-stratified by SRSSchwab sagittal vertical axis (SVA) modifier into (posterior alignment SVA \0, aligned 0–50 and malaligned [50 mm). C2–C7 cervical curvature was assessed among SVA grade in every TK group. Stepwise linear regression analysis was applied. A simplified formula was validated on random selection of 1905 patient visits from same database. Results: In each TK group (n = 265, 172, 163, 144), CC was significantly more lordotic by increased Schwab SVA grade. In SVA\0, CC was neutral for TK 40–50°, and kyphotic for TK \40°. All pts with SVA\50 mm, and TK\30° were kyphotic. Regression analysis revealed LL minus TK (LL-TK) as an independent predictor (r = 0.653, r2 = 0.426) with formula: CC = 10 - (LL-TK)/2. Validation of the formula revealed error of 1.2° between predicted CC and real CC (r = 617, r2 = 381). Conclusions: Kyphotic cervical alignment is necessary in the maintenance of horizontal gaze in some well aligned and some sagittal backward pts depending on thoracic curvature. CC can be predicted from underlying TK and lumbar lordosis, which can be clinically relevant in thoracolumbar deformity correction. 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 = Zimmer, consultant; Company = Medicrea, Biomet, AlphaSpine; 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.

10 CERVICAL INTERBODY FUSION WITH STANDALONE LORDOTIC TANTALUM CAGES. CHANGES IN SAGITTAL ALIGNMENT. DO THE SUBSIDENCE HAVE CLINICAL IMPLICATIONS? A PROSPECTIVE LONG-TERM STUDY Fe´lix Tome´-Bermejo, Julia´n A MORALES-VALENCIA, Javier MORENO-PE´REZ, Angel R. Pin˜era, Luis Alvarez Spine Department, Fundacio´n Jime´nez Dı´az University Hospital, Madrid, Spain Background context: Cage subsidence is the most frequently reported complication after anterior cervical discectomy-and-fusion (ACDF) with interbody fusion cages (IFC). However, most reports fail to correlate cage subsidence with lower fusion rates, or with unsatisfactory clinical results. Purpose: Retrospective, observational study of prospectively collected outcomes to investigate the long-term clinical course of ACDFIFC with lordotic tantalum implants and to correlate the radiological findings with the clinical outcomes, with special emphasis in the significance and influence of implant subsidence. Mean follow-up of 4.91 years Patient sample: Forty-one consecutive patients with symptomatic degenerative cervical disc disease with failure of conservative treatment. All patients underwent one or two level ACDF-IFC with lordotic tantalum implants. Outcome measures: Visual Analogue Scale (VAS) and Neck Disability Index questionnaire (NDI). Radiological assessment of interspace height (IH) at the affected levels, C1–C7 sagittal alignment

S673 and segmental lordosis measured using Cobb angle, bone fusion and implant subsidence. Results: A total of thirty-three patients (41 discs) were finally included. Final follow-up fusion rate was 96.96 % (32/33). IH was significantly incremented after insertion of the implant, and despite a gradual loss in the height occurred over time, final follow-up IH was significantly higher than preoperatively (P \ 0.0001). Anterior-IH and posterior-IH lost 55.8 and 76.2 % of the initially incremented height respectively, with a final increase of 72 % in the AIH-PIH height differential. Implant subsidence ([3 mm) occurred in 11 disc spaces (26.82 %). Pre and postop-IH was significantly higher in the subsidence patients, however, there was no difference in final followup IH (P [ 0.05). Patients with C3 years of follow-up (n = 29) did not demonstrate further significant subsidence beyond the second year. Regarding C1-C7 lordosis, segmental Cobb angle, cervical VAS and NDI, no difference between patients with or without final followup endplate subsidence was encountered. Conclusions: Until fusion occurs, tantalum cage settlement into the vertebral body is to be expected. Further subsidence could be the result of segmental adaptative changes. Graft subsidence did not affect the clinical outcome in any of our patients during long-term follow-up, dynamical implant subsidence occurrence, especially at the posterior-IH, positively affected cervical lordosis. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: grants/research support; Company = Biomet, consultant; Company = Spineart.

11 PROSPECTIVE MULTICENTER ASSESSMENT OF EARLY COMPLICATION RATES ASSOCIATED WITH ADULT CERVICAL DEFORMITY (ACD) SURGERY IN 78 PATIENTS Justin Smith, Christopher Shaffrey, Themistocles Protopsaltis, Peter Passias, Gregory Mundis, Robert Hart, Han Jo Kim, Christopher Ames International Spine Study Group, Chicago, USA Hypothesis: Surgery for ACD is associated with high early complication rates. Design: Prospective multicenter cohort study. Introduction: Although ACD can have profound impact, few reports have focused on the treatment of these patients. We present early complication rates associated with surgical treatment for ACD based on a prospective multicenter cohort. Methods: Prospective multicenter database of consecutive operative ACD patients was reviewed for early (\30 days from surgery) complications. Enrollment required at least one of the following: cervical kyphosis[10°, cervical scoliosis,[10°, C2–7 SVA[4 cm or chin-brown vertical angle [25°. Results: 78 patients (59 % women) underwent surgical treatment for ACD and had a mean age of 60.7 years and previous surgery in 52 %. Surgical approaches included anterior-only (A, 14 %), posterior-only (P, 49 %), anterior-posterior (AP, 35 %) and posterior-anterior-posterior (PAP, 3 %). Mean numbers of fused anterior and posterior vertebral levels were 4.7 and 9.4, respectively. A total of 52 early complications were reported, including 26 minor and 26 major for an overall complication rate of 66.7 %. 22 (28.2 %) patients had at least one minor complication, and 19 (24.4 %) had at least one major complication. Overall, 45 (57.7 %) patients had at least one complication. The most common complications included dysphagia (11.5 %), deep wound infection (6.4 %), new C5 motor deficit (6.4 %) and respiratory failure (5.1 %). One mortality (1.3 %) was

123

S674 reported. Overall early complication rates were significantly different based on approach: A (27.3 %), P (68.4 %) and AP/PAP (79.3 %) (p = 0.007). Conclusion: Among 78 patients treated for ACD and prospectively followed, a total of 52 early complications were reported (26 minor, 26 major). Overall, 45 (57.7 %) patients had at least one complication and 24.4 % of patients had at least one major complication. Significantly higher rates of complications were associated with combined and posterior- only approaches compared with anterior-only approaches. These findings may prove useful in treatment planning and patient counseling. Disclosures: author 1: grants/research support; Company = DePuySynthes, ISSGF, consultant; Company = Biomet, Nuvasive, Cerapedics, Medtronic, royalties; Company = Biomet; author 2: grants/ research support; Company = Depuy-Synthes, consultant; Company = Biomet, Medtronic, Nuvasive, Stryker, stock/shareholder; Company = Nuvasive, royalties; Company = Biomet, Medtronic, Nuvasive; author 3: grants/research support; Company = Zimmer, consultant; Company = Medicrea, Biomet, AlphaSpine; author 4: none; author 5: grants/research support; Company = dePuy Synthes, consultant; Company = nuvasive, K2M, Medicrea, Misonix, royalties; Company = nuvasive, K2M; author 6: grants/research support; Company = Medtronic, ISSGF, consultant; Company = Depuy Synthes, Globus, Medtronic, stock/shareholder; Company = Spine Connect, royalties; Company = Seaspine, Depuy Synthes, other financial report; Company = Depuy Synthes; author 7: consultant; Company = K2M, Biomet, Medtronic, other financial report; Company = Depuy, Stryker; author 8: grants/research support; Company = Depuy Synthes Spine, consultant; Company = Depuy, Medtronic, Stryker, stock/shareholder; Company = Baxano Surgical, Doctor’s Research Group, royalties; Company = Aesculap, Biomet Spine, employee; Company = University California San Francisco.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 area (FCSA) measurements of the cervical paraspinal muscles were performed at the C4–5 level. FCSA was measured by grayscale thresholding in a standardized region-of-interest within the posterior cervical extensor mass to isolate lean muscle. Student t tests were used to determine significance at p \ 0.05. Results: Cervical IVD heights increased 0.35 ± 0.42 mm immediately after 6-month exposure to microgravity. The heights decreased an average 0.08 ± 0.37 mm at 30 days recovery. The posterior aspect of the IVD heights was significantly decreased during recovery (0.30 ± 0.36 mm). Anterior and middle segments remained swollen. Lean muscle mass, as indicated by the PSM FCSA, decreased 9.9 ± 17.8 % immediately post-flight. The PSM FCSA recovered 2.2 ± 3.8 %, after 30 days post-flight recovery. Conclusion: This early data shows cervical spine IVD swelling and PSM atrophy after long-duration spaceflight. Some recovery was seen with 30 days post-flight in a terrestrial environment, but it was incomplete compared to pre-flight levels. Additional information is expected this Spring/Summer, with ongoing work involving 4 crewmembers. This project was supported by NASA grants NNX13AM89G and NAG 9-1425. Disclosures: author 1: grants/research support; Company = NASA; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

13 IS THE UNILATERAL STABILIZATION IN THE SUBAXIAL CERVICAL SPINE SUFFICIENT? A BIOMECHANICAL STUDY Gregor Schmeiser, Christoph Schilling, T. M. Grupp, K. Pu¨schel, Ralph Kothe

12 CERVICAL SPINE INTERVERTEBRAL DISC AND PARASPINAL MUSCLE MORPHOLOGY IN HUMANS AFTER 6-MONTH MICROGRAVITY EXPOSURE AND 30-DAY TERRESTRIAL RECOVERY Douglas Chang, Robert Healey, Jacquelyn Holt, Brandon Macias, Alex Snyder, Jeffrey Lotz, Alan Hargens University of California, San Diego and San Francisco, USA Microgravity exposure during long-duration spaceflights results in *5 cm spinal lengthening, damage to intervertebral discs (IVD), paraspinal muscle (PSM) atrophy, and back pain. The process is not well characterized as of yet, but is most likely related to supraphysiological IVD swelling, and the loss of crewmember muscle mass and flexibility, in the absence of gravity. NASA crewmembers have a high risk of herniated IVDs, compared to the general and military aviator populations. This risk is highest in the cervical spine, and in the first year after a mission. Purpose: To evaluate cervical IVD height changes and PSM crosssectional area before and after (immediate and 30-days post-flight) a 6-month space mission on the International Space Station. The longterm objective of this project is to promote IVD health and prevent injury during space missions as well as here on Earth. Methods: This prospective case series studied 3 NASA crewmembers, using supine sagittal MRI (3T) imaging from C2-T1. The imaging was conducted pre-flight, immediately post-flight and 30–45 days later post-flight. IVD heights were measured at the anterior, middle and posterior sections. Functional cross-sectional

123

Scho¨n Klinik Hamburg-Eilbek, Spinale Chirurgie, Hamburg, Germany Introduction: Multisegmental spondylogenic cervical myelopathy (mSCM) is the most frequent progressive disease in the cervical spine in the elderly population. The current standard operative procedures are laminoplasty (motion preserving) or laminectomy with screw-rodstabilization (fusion). To reduce invasivity of surgery an unilateral approach for laminoplasty was developed which could be used in combination with a unilateral lateral mass screw fixation. The goal of the study was to evaluate the stabilization behavior of this new treatment option in comparison to the current standard procedures. Material and method: Ten human cervical specimens C2–C7 were tested in the segment conditions: (1) Native (NAT); (2) Laminoplasty (LP); (3) LP with unilateral stabilization (LPU); (4) LP with bilateral stabilization (LPB); (5) Laminectomy with bilateral stabilization (LCB); (6) L without stabilization (LC). The instrumented level was from C3–C6. For each segment condition in vitro flexibility tests were performed using a spinal simulator. The 3D-kinematics of the entire cervical spine segment in the 3 main loading directions flexion–extension (FE), lateral bending (LB) and axial rotation (AR) was measured with an ultrasonic motion analysis system (ZEBRIS). The applied load was ±2.5 Nm [Wilke et al. 1998]. Statistical analysis on the parameters ROM and neutral zone (NZ) to determine differences of the specific segment conditions was assessed using an ANOVA followed by a post hoc test (Scheffe), with a significance level of p = 0.05. Results: In FE and LB the ROM and NZ of LP and LC stayed at the level of NAT (p [ 0.05), whereas the instrumentations LPU, LPB, LCB lead to a significant reduction (p \ 0.01) without significant

Eur Spine J (2015) 24 (Suppl 6):S669–S710 difference of LPU to LPB and LCB (p [ 0.05). In AR the stabilization of all instrumentations were less pronounced but with the same tendency seen in FE and LB. Discussion: For the two standard instrumentations (LP and LCB) the degree of stabilization was found as expected, namely maximal stabilization for LCB and preservation of motion for LP. Interestingly the unilateral lateral mass fixation (LPU) exhibited almost the same degree of stabilization as the bilateral instrumentation (LCB). Therefore from a biomechanical point of view an unilateral approach for laminoplasty and stabilization seems to be a minimal invasive treatment option for the surgical therapy of mSCM. The clinical results are currently under evaluation. Disclosures: author 1: consultant; Company = Aesculap AG, Tuttlingen, Germany; author 2: employee; Company = Aesculap AG; author 3: employee; Company = Aesculap AG; author 4: no indication; author 5: consultant; Company = Stryker, Aesculap, Silony.

14 ASSESSMENT OF SWALLOWING FUNCTION USING EAT10: A SWALLOWING ASSESSMENT TOOL Chikato Mannoji, Masaaki Aramomi, Masao Koda, Takeo Furuya, Masazumi Murakami Chiba Aoba Municipal Hosiptal, Chiba, Japan Introduction: Dysphagia is one of the problematic complications after anterior cervical spine and occipitocervical fixation surgeries. Therefore, it is important to assess swallowing function before these surgeries. However, there has not been a standard method of assessing swallowing function preoperatively, and the definition of postoperative dysphagia is ambiguous. Purpose: The purpose of this study is to assess the swallowing function of patients without previous surgeries of the cervical region using the Eating Assessment Tool (EAT-10), which is composed of 10 questions regarding swallowing function. Each question is allocated a maximum of 4 points, and the maximum total is 40 points. Patients who score C3 points may have problems in swallowing. Methods: Patients were included if they visited our department, consented to participate in this study, and completed the EAT-10 questionnaire between September 2013 and March 2014. Results: Of 2032 patients who gave consent and completed the EAT10 questionnaire, 125 patients with previous cervical spine, esophagus, trachea, thyroid gland or oral surgeries were excluded. Thus, the data from 1907 patients (906 men and 1001 women) were analyzed. The average age was 57.2 years (range 9–103 years), and 166 patients (8.7 %) scored C3 points on the EAT-10. Multiple regression analysis showed that age and sex influence the EAT-10 score, which means older people and women showed higher EAT-10 scores. The medical records of 27 patients who scored C10 points were reviewed in detail. Six patients had past histories that might influence their swallowing function. Among the remaining 21 patients, there were 4 trauma patients, who had been injured in traffic accidents, and 6 patients, who had been receiving treatment for psychiatric problems, indicating that the EAT-10 score may be influenced by mental conditions. Conclusions: Age and sex influenced the EAT-10 score; older people and women had higher scores, indicating that swallowing dysfunction may exist in these patients without any apparent causes. Before performing cervical surgeries, surgeons should assess the patients swallowing function to assess and compare dysphagia before and after the surgeries. EAT-10 can be used as a concise assessment tool. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

S675 15 4-YEAR RESULTS OF A PROSPECTIVE MULTI-CENTRE STUDY ON 200 SINGLE-LEVEL CERVICAL DISC PROSTHESES: CLINICAL AND RADIOGRAPHIC OUTCOME Petr Suchomel, Hans Joerg Meisel, Ian Shackleford, Jussi Antinheimo, Steffen Sola, Bernhard Bruchmann, Jan Stulik, Fabrizio Caroli, Stefan Kroppenstedt, Christian Woiciechowsky Neurocenter, Department of Neurosurgery, Regional Hospital Liberec, Czech Republic Introduction/aim: Despite encouraging outcomes after cervical total disc replacement (cTDR), critical debates regarding heterotopic ossification (HO) and the prevalence of adjacent level degeneration (ALD) are ongoing. Materials and methods: In this prospective multi-centre study 200 subjects underwent single-level cTDR for the treatment of symptomatic cervical degenerative disc disease. Assessments were performed preoperatively, at discharge and again at 6 weeks, 6 months, and at 1, 2 and 4 years. Qualitative and quantitative radiographic analyses were done by an independent core lab. Results: NDI and VAS neck and arm pain decreased significantly from baseline to the four-year follow-up. Mean improvement for NDI was 20, for VAS severity and frequency of neck pain 26.4 and 28 and for VAS arm pain 30.7 and 35.1, respectively. Neurologically most patients (86.4 %) improved, with 76.1 % non-symptomatic cases. Subsequent reoperations were reported for 7 %, including 3 % device removals. In 2.5 % a subsidence [3 mm was recorded, one case with a migration [3 mm. No further adverse events were observed. Segmental lordotic alignment changed from -2.4° (preop) to -6.2° at 4 years. Postoperative height was maintained during the four-year follow-up. HO (grade III–IV) was present in 27.1 % of cases and 82.4 % showed segmental mobility. ALD occurred in 28.2 %, 4.5 % thereof required surgery. Conclusions: These mid-term results confirm the effectiveness of this semi-constrained cTDR. It preserves motion for most subjects; but further long-term observations are required to confirm these findings. Disclosures: author 1: consultant; Company = Aesculap AG, Germany, royalties; Company = Aesculap AG Germany; author 2: none; author 3: grants/research support; Company = Aesculap AG; author 4: none; author 5: none; author 6: no indication; author 7: none; author 8: grants/research support; Company = author 9: consultant; Company = BBraun/Aesculap; author 10: grants/research support; Company.

16 CERVICAL TDR WITH MOBILE PROSTHESIS: MID-TERM CLINICAL ASSESSMENT AND RADIOLOGICAL ANALYSIS OF OPERATED AND ADJACENT SEGMENTS Jacques Beaurain, Thierry Dufour, Jean Huppert, Pierre Bernard, Istvan Hovorka, Jean-Paul Steib University Hospital Bocage central; Neurosurgery Department, 14 Rue Paul Gaffarel, Dijon, France Introduction: Cervical arthroplasty could restore motion, relieve pain by however preserving adjacent segments condition. Our study assessed the clinical and radiological performances of a device with controlled mobility 3 years after surgery. Adjacent discs were analyzed regarding ossification and degenerative status.

123

S676 Methods: A longitudinal observational prospective study was led on 384 patients and involved 10 centers. Data were analyzed preoperatively and at 1, 3, 6, 12, 24, 36 months postop. Clinical outcomes investigated NDI, VAS, SF-36, complications and re-operations rates. Radiological assessment explored Range of Motion (ROM) both at index and adjacent levels; heterotopic ossification (HO) and radiographic evidence of discs degeneration were analyzed according to McAfee and Walraevens classification respectively. Results: Patients mean age was 44.8 ± 8.1 years with 44 % men and 56 % women. All clinical outcomes (NDI, VAS for arm and neck pain, SF-36 PCS and MCS) were significantly improved at 3 years FU compared to baseline. Medication consumption was significantly reduced after surgery: 83.2 % patients were under medication preop vs 28.0 % at 3 years FU. At 3 years, 92.5 % patients were satisfied regarding the overall outcome, 82.4 % regarding cervical pain and 79.9 % regarding arm pain. Professional status was changed with 60.3 % of patients working at 3 years vs 35.6 % preop. Radiological outcomes showed a significant increase of motion at 3 year-FU for the index level (mean ROM = 8.5° vs 6.0° at baseline), 76.9 % of the segments were mobile (ROM [ 3°) at 3 years. Ossification was responsible of the fusion of 9.6 % of the implanted segments at 3 years. Superior and inferior adjacent discs showed no significant change of motion at 3 year-FU compared to preop. At 3 year-FU, ossification was moderate and no fusion of adjacent discs was reported. At 3 years, adjacent discs showed mainly no or mild degeneration (grade 0–3). Complications rate was 8.3 % and re-operation at the index level represented 0.37 % of the implanted prostheses. Reoperation rate of adjacent discs was 1.3 % at 3 years. Conclusion: This 3 year-outcomes sustain the clinical efficacy of the device with a low complication rate. Regarding adjacent levels, motion was not impaired with time, and the occurrence of ossification or of disc degeneration was low. Reoperation rate for adjacent disc disease was low compared to literature. Disclosures: author 1: consultant; Company = LDR Medical, royalties; Company = LDR Medical; author 2: royalties; Company = LDR medical; author 3: stock/shareholder; Company = LDRH, royalties; Company = LDRH; author 4: stock/ shareholder; Company = LDR, royalties; Company = LDR, SpineArt, OSD; author 5: royalties; Company = LDR; author 6: grants/ research support; Company = Clariance, LDR, Medtronic, Integra, Stryker, consultant; Company = Clariance, royalties; Company = Clariance, LDR, Medtronic.

17 TOTAL DISC ARTHROPLASTY VERSUS ANTERIOR CERVICAL INTERBODY FUSION: USE OF THE SPINE TANGO REGISTRY TO SUPPLEMENT THE EVIDENCE FROM RCTS Emin Aghayev, Anne Mannion, Christoph Ro¨der, Lukas Staub Institute for Evaluative Research in Medicine of the University of Bern, Switzerland Background: Several randomized controlled trials (RCTs) have compared patient outcomes of anterior cervical interbody fusion (AIF) with those of arthroplasty (TDA). RCTs have known limitations with regard to their external validity, and the comparative effectiveness of the two therapies in daily clinical practice remains unknown. Purpose: To compare patient-centered outcomes after TDA vs. AIF based on data from an international spine registry. Study design: Retrospective analysis of registry data.

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Patient sample: Inclusion criteria were degenerative disc or disc herniation of the cervical spine treated by single-level TDA or AIF, no previous surgery and a Core Outcome Measures Index (COMI; including 0–10 pain scales) completed at baseline and at least 3 months’ follow-up. Overall, 987 patients were identified. Outcome measures: Neck and arm pain relief and COMI score improvement. Methods: Three separate analyses were performed to compare TDA and AIF surgical outcomes. 1. Mimicking an RCT setting, with exclusion criteria typical of those in published RCTs, a 1:1 matched analysis was carried out in 739 patients. 2. An analysis was performed on 248 patients outside the classic RCT spectrum i.e. with one or more typical RCT exclusion criteria. 3. A subgroup analysis of 149 patients with additional follow-up longer than 2 years. Results: Matching resulted in 190 pairs with an average follow-up of 17 months. Significant group differences in favor of TDA were seen for postop neck pain (mean difference [MD] 0.6 points; 95 % confidence intervals [95 % CI] 0–1.2; p = 0.037), arm pain (MD 0.7; 95 % CI 0.1–1.3; p = 0.020), and COMI score (MD 0.8; 95 % CI 0.2–1.4; p = 0.010). Furthermore, the change-score (pre to postop) was significantly different in favor of TDA for COMI (MD 1.0; 95 % CI 0.4–1.6; p = 0.002), but not for neck pain or arm pain (p = 0.16 and 0.06, respectively). Subgroup analyses of atypical patients and of patients with longerterm follow-up showed no significant differences in outcome between the treatments (range of MDs TDA-AIF 0–0.2, p [ 0.57, and -0.2 to 1.1, p [ 0.08, respectively). Conclusions: The results of the observational study were in accordance with those of the published RCTs, suggesting significant pain reduction both after AIF and TDA, with slightly greater benefit after arthroplasty. The analysis of atypical patients suggested that, in patients outside the spectrum of clinical trials, the two surgical interventions were associated with similar outcomes. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

18 CLINICAL AND RADIOLOGICAL OUTCOME AT THE 10 YEAR FOLLOW UP OF TOTAL CERVICAL DISC REPLACEMENT Christoph Mehren, Franziska Heider, Bernhard Zillner, Daniel Sauer, Christoph Siepe, Andreas Korge, H. Michael Mayer Scho¨n Klinik Mu¨nchen Harlaching, Germany Purpose: There is still a lack of facts about long term impacts after implantation of a total cervical disc replacement (cTDR). The aim of this current analysis was to evaluate the long- term clinical results and efficacy, the radiographic findings of the index level as well as the adjacent level in a cohort of cTDR patients treated with ProDisc C (Synthes, Paoli, USA) 10 years after surgery. Methods: Clinical outcome scores and patient satisfaction rates were acquired within the framework of an ongoing single centre prospective non randomised clinical trial. The patient’s employment status, complications, reoperations and further treatments due to symptomatic adjacent segment disease (ASD) were recorded. X-ray evaluation included the incidence of heterotopic ossifications (HO)

Eur Spine J (2015) 24 (Suppl 6):S669–S710 according as well as the rates and timing of ASD. Follow up was recorded 1, 5 und 10 years after surgery. Results: 38 patients were available for final FU, resembling a 80.8 % FU rate after a mean FU of 122.9 months. The overall results revealed a highly significant improvement from baseline VAS and NDI levels at all postoperative FU stages (p \ 0.0001). VAS scores demonstrated a statistically significant improvement (VAS neck 6.4 to 1.9; VAS arm 6.3 to 2.1 at the last FU). The NDI baseline values showed a stable improvement (21/50 to 6/50). Patient satisfaction rates remained stable throughout the postoperative course, with 78.9 % of patients reporting a ‘highly satisfactory’ or a ‘satisfactory’ outcome in 21.1 %. The radiographic results demonstrated a noteworthy incidence of HO as well as a progression of the HO in process of time. HO Grade I appeared in 10 %, Grade II in 10 %, Grade III in 32 %, and Grade IV in 26 %. The overall complication rate (excluding HO) was 13.3 %. The incidence of revision surgery for device related complications was 2.2 %. The incidence of ASD was 32.3 %. 70 % of ASD were asymptomatic, none of the ASD patients need further surgery. Within this group of patients with ASD 80 % of the patients presented with a high-grade HO with loss or reduced function of the prosthesis in the 5 year FU. Conclusion: The results demonstrate that cTDR is a viable treatment option resulting in improved clinical outcomes and low rates of subsequent surgical procedures. The progression of HO with a high amount of non functional TDR may leads to the high figures of ASD. On the other hand, the current data reveal that a functional cTDR may significantly reduce the incidence of ASD. Disclosures: author 1: consultant; Company = Silony, Depuy Synthes; author 2: none; author 3: none; author 4: none; author 5: consultant; Company = DePuy Synthes, Silony Medical; author 6: consultant; Company = Silony; author 7: none.

TRAUMA, TUMOR

19 FLOWCYTOMETRIC EVALUATION OF INTRAOPERATIVE SALVAGED BLOOD FILTERED WITH LEUCOCYTE DEPLETION FILTER IN METASTATIC SPINE TUMOUR SURGERY Naresh Kumar, Aye Sandar Zaw, Rishi Malhotra National University Health System, Singapore Background: Intra-operative cell salvage (IOCS) has not been widely adopted in oncological surgery due to hypothetical concern of reinfusion of malignant cells. We conducted the study to evaluate the feasibility of IOCS in combination with leucocyte depletion filter (LDF) in metastatic spine tumour surgery (MSTS) using quantitative flow cytometry tumour detection technique. Methods: Patients with known primary epithelial tumour, operated for metastatic spinal disease, were recruited. Blood samples were collected at five different stages during surgery: 2 stages from patient vein during induction and at the time of maximum tumour manipulation, 3 stages from the operative blood prior to IOCS processing, after IOCS processing and after IOCS-LDF processing. Of the samples taken at each stage, 5 ml were analyzed for tumour cells using flow cytometry.

S677 Results: Of twelve patients recruited, only 11 could be finally analyzed. Flow cytometry analysis of their samples showed that 8/11 patients had tumour cells in the unfiltered salvaged blood. In filtered salvaged blood, the tumour cell count was zero in the majority of samples (8/11 patients) while 3 patients’ samples had a few tumour cells. The difference between the mean tumour cell quantity in the samples taken from stage A and E was significant. (P = 0.04) Similarly, the difference between the mean quantities of tumour cells in the samples in stage B and E was significant. (P = 0.01) However, there were no significant differences between the mean quantities of tumour cells when comparing the samples from either stage A and B, stage D and E or stage C and E. Conclusion: IOCS-LDF was shown to be effective in removing tumour cells from blood salvaged during MSTS. If there were any tumour cells found, the quantity was significantly less than that in patient’s circulation. The results of this study supported that fact that IOCS-LDF treated blood in MSTS is safe for transfusion. Disclosures: author 1: none; author 2: none; author 3: none.

20 IMMUNOGENICITY OF HUMAN INDUCED PLURIPOTENT STEM CELLS-DERIVED NEURAL STEM CELLS AS A CELL SOURCE OF CELL TRANSPLANTATION THERAPY FOR SPINAL CORD INJURY Masahiro Ozaki, Go Itakura, Hiroki Iwai, Jun Kohyama, Akio Iwanami, Morio Matsumoto, Hideyuki Okano, Masaya Nakamura Department of Orthopaedic Surgery, Keio University School of Medicine, Tokyo, Japan Background: Recent studies demonstrated the effectiveness of transplantation of human induced pluripotent stem cells-derived neural stem cells (hiPSC-NSCs) for rodent as well as non-human primate spinal cord injury (SCI) model. Since it is time-consuming and expensive to establish neural stem cells derived from autologous iPS cells, allogeneic transplantation would be a realistic plan of firstin-human trial in taking advantage of iPS bank in Japan. However, immunological investigation of hiPSC-NSCs has not been reported in detail. The purpose of this study is to determine immunogenicity and immune response of hiPSC-NSCs. Materials and methods: Integration-free human iPS clone derived NSCs were used in this study. The expression of immune-related surface antigens on hiPSC-NSCs was analyzed using flow cytometry. To mimic an inflammatory environment of the injured spinal cord, hiPSC-NSCs were stimulated with IFN-c. Effect of passage number of hiPSC-NSCs on immunogenicity was also evaluated simultaneously. Furthermore, hiPSCs and human fetal neural stem cells (hfNSCs) were analyzed in the same way to compare immunogenicity of hiPSC-NSCs. Mixed lymphocyte reaction (MLR) was performed to evaluate the degree of immune response in vitro. hiPSC-NSCs inactivated by irradiation were co-cultured with autologous and allogeneic lymphocytes, and then proliferation of lymphocyte was quantitatively measured, and compared between the autologous and allogenic conditions. Results: Less than 0.5 % of hiPSC-NSCs expressed HLA class II or co-stimulatory molecules which play an important role in antigen presentation. IFN-c administration significantly increased the expression of HLA class I and CD54, but did not increase the expression of HLA class II and co-stimulatory molecules. Small percentage of hfNSCs stimulated by IFN-c expressed HLA class II.

123

S678 Compared to the hiPSCs, hiPSC-NSCs showed significantly higher expression of HLA class I and CD54. Multiple passages less than 7 times did not affect the expression of surface antigen. In MLR, proliferation of lymphocyte co-cultured with allogeneic hiPSC-NSCs was significantly stimulated compared to that co-cultured with autologous lymphocytes. Conclusions: Immunological examination revealed that hiPSC-NSCs exhibited comparatively low immunogenicity. It was also suggested that allogeneic transplantation of hiPSC-NSCs indicated higher immune response than autologous transplantation of those. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

21 EVALUATION OF TLICS FOR THORACOLUMBAR FRACTURES Galateia Katzouraki, Spyridon Koufos, Giuseppe Lambros Morassi, Panayiotis Karampinas, Dimitrios Stergios Evangelopoulos, Spiros George Pneumaticos 3rd Department of Orthopaedic Surgery, University of Athens, KAT Hospital, Athens, Greece Introduction: Thoracolumbar spine trauma are the most common site of spinal cord injuries, with clinical and epidemiological importance. Classification systems for thoracolumbar injuries can be simplified or complex in clinical use. The most appropriate classification system remained controversial until today. The posterior ligament integrity, the morphology of the injury and the neurological status were introduced together from Spine Trauma Group in Thoracolumbar Injury Classification and Severity Score (TLICS) facilitating the communication between physicians, education of residents and fellows and guiding to treatment decision with better outcome. Objectives: This study aimed to research the effectiveness of TLICS scoring for thoracolumbar vertebral fractures without neurological deficits. In general, patients with a total score of B3 can be treated nonoperatively, depending on the type of injury. In contrast, patients with a total score of C5 usually require surgical treatment. Patients with a total score of 4 are in an intermediate zone where either operative or nonoperative treatment might be equally appropriate. Methods: The study included 58 patients with TLICS = 4 (37 males, 21 females; mean age: 46.03 years; range 18–64 years) were conservatively treated for thoracolumbar (T8–L5) vertebral fractures according to TLICS classification between 2008 and 2012. Etiology were falls from various heights (49 %), motor vehicle accidents (25 %) and simple falls (26 %). None of the patients had any neurologic deficit. VAS pain and functional scales were used in the clinical evaluation. Local kyphosis angle and height loss percentage were measured in the radiologic evaluation. Post-fracture and follow-up values were compared. Mean follow-up period was 32.0 (range 9–60) months. Results: Fifty patients returned to their pre-trauma work and activities completely and eight patients with small limitations. Mean period for returning to work was 3.8 (range 2.5–6) months. Local kyphosis angle and height loss percentage values did not demonstrate any significant increase at follow-up (p \ 0.005) Conclusions: The study results demonstrate that conservative treatment for TLICS = 4 thoracolumbar fractures can be safely applied. Early mobilization with functional thoracolumbar brace produces effective functional results without significant loss of vertebral body height. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 22 TOWARDS THE DEVELOPMENT OF INTERNATIONAL DISEASE SPECIFIC OUTCOME INSTRUMENTS FOR SPINE TRAUMA: SURGEON REPORTED OUTCOME Said Sadiqi, A. Mechteld Lehr, F. Cumhur Oner, AOSpine Knowledge Forum Trauma Department of Orthopaedics, University Medical Center Utrecht, The Netherlands; AOSpine Knowledge Forum Trauma Aim: In the absence of a universally accepted and validated tool to measure outcome of spine trauma, it is difficult to compare outcomes of different treatments within observational or randomized cohort studies. Therefore, AOSpine Knowledge Forum Trauma initiated a project to develop and validate such instruments, which include both the patients’ and health professionals’ perspective on functioning, disability and health. Beside a patient reported outcome, the aim is to develop and validate a surgeon reported outcome, which aims to predict the functional outcome of patients with traumatic spine fractures on the short term (3 months–2 years) and long term (C2 years), both for nonsurgically and surgically treated patients. Materials and methods: An international cross-sectional web-based survey was conducted among spinal surgeons from the five AOSpine International world regions. They were asked to give their opinion on the relevance of a compilation of 16 clinical and radiological parameters on a five-point scale, based on a systematic review and experts interviews. The target group of patients was defined as adults (C18 years) who have suffered acute traumatic injury to the thoracic, thoracolumbar or lumbar spine. Furthermore, patients’ age that substantially influences their outcome was asked, as well as the measurement techniques for assessing fracture kyphosis and vertebral body height loss. The responses were analysed using frequency analysis and descriptive statistics. Results: Out of the 279 invited members of AOSpine International and International Spinal Cord Society (ISCoS), 118 (42.3 %) participated in this study. Out of the 16 parameters included in the survey, 5 were identified as relevant by at least 70 % of the participants for all defined spine regions, and both for short term and long term. In contrast, 5 parameters were neither identified as relevant for a specific spine region nor for the time points, except for ‘‘comorbidity’’. The mean age that substantially influences patients’ outcome did not differ among the spine regions and time points. The Cobb method was most frequently used for assessing fracture kyphosis among all defined spine regions, and anterior/middle compression ratio for assessing vertebral body height loss. Discussion: In future studies, relevant parameters will be identified for patients with traumatic cervical fractures, and the surgeon reported outcome will be developed and subjected to further validation. Disclosures: author 1: none; author 2: consultant; Company = AOSpine International; author 3: grants/research support; Company = DePuy-Synthes; author 4: none.

23 OPEN VERSUS PERCUTANEOUS INTRUMENTATION IN THORACOLUMBAR FRACTURES: MRI COMPARISON OF PARVERTEBRAL MUSCLES AFTER IMPLANT REMOVAL Yann Philippe Charles, Yves Ntilikina, David Bahlau, Julien Garnon, Se´bastien Schuller, Axel Walter, Erik Andre´ Sauleau, Mickael Schaeffer, Steib Jean-Paul Service de Chirurgie du Rachis, Hoˆpitaux Universitaires de Strasbourg, France

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Introduction: Percutaneous instrumentation in thoracolumbar fractures is intended to decrease paravertebral muscle damage by avoiding dissection. The aim of this study was to compare muscles at instrumented levels in patients who had been treated by open or percutaneous surgery. Materials and methods: 27 patients were instrumented open and 65 percutaneously. A standardized MRI protocol using axial T1-weighted has been performed at minimum 1-year follow-up after implant removal. Cross section areas CSA (cm2) and region of interest ROI signal intensity (pixel) of paravertebral muscles were measured using Osirix by two independent observers at fracture level, cranial and caudal instrumented pedicle levels. An inter-observer comparison was made using the Bland–Altman method. The reference ROI of muscle was assessed in the psoas and the ROI of fat subcutaneously. The ratio ROI-CSA/ROI-fat was compared for open versus percutaneous patients using a linear mixed model. A linear regression analyzed additional factors: age, gender, BMI, Pfirrmann grade of adjacent discs and duration of instrumentation in situ. Results: The inter-observer agreement was good for all CSA. Average ROI-fat and ROI-muscle were comparable in open versus percutaneous groups: 146.7 versus 145.8 and 538.3 versus 526.2 respectively. The ROI-CSA varied between 154.7 and 226.2 for open, 154.4 and 195.5 for percutaneous depending on instrumented levels. A significant difference of ROI-CSA/ROI-fat (0.4 versus 0.3) was present at fracture levels T12-L1 (p = 0.0329), at adjacent cranial (p = 0.0139) and caudal (p = 0.0100) instrumented levels. Differences were not significant at thoracic and lumbar levels. Age had a significant influence on this ratio (p \ 0.0001). Gender, BMI, Pfirrmann grade and duration of instrumentation had no significant influence. Conclusion: Percutaneous instrumentation is aimed to decrease muscle damage and the amount of muscle atrophy compared to conventional open surgery. This seems only valid for instrumented levels around the thoracolumbar junction. The present results evidenced an MRI signal difference for T12 and L1 fractures, indicating less muscle atrophy within the CSA in patients who received percutaneous treatment. Differences were not evidenced in the thoracic spine, where the global paravertebral muscle mass is smaller. Muscle atrophy was not significantly different in the lumbar spine with either procedure, but age influenced the amount of muscle atrophy. Disclosures: author 1: consultant; Company = Stryker, Medtronic, DePuy Synthes, Ceraver, LDR Me´dical, Clariance, royalties; Company = Clariance; author 2: none; author 3: none; author 4: consultant; Company = Galil Medical; author 5: consultant; Company = Clariance; author 6: none; author 7: none; author 8: none; author 9: grants/research support; Company = Clariance, LDR, Medtronic, Integra, Stryker, consultant; Company = Clariance, royalties; Company = Clariance, LDR, Medtronic.

S679 in 1990s. Recently, there are some reports that additional vertebroplasty (VP) can prevent this failure. Meanwhile, we presented satisfactory results of temporary instrumentation for thoracolumbar burst fractures without augmentation at annual meeting of Eurospine 2013. The purpose of this study was to compare the treatment outcomes in patients with thoracolumbar burst fractures with or without VP. Methods: A total of 51 consecutive patients with thoracolumbar burst fracture (T11–L3) who underwent surgery by ligamentotaxis procedure using Schanz screws in three institutions were included in this study. There were 35 men and 16 women, with an average age of 39 years (range 16–66). 27 patients were operated without augmentation and 24 patients with VP. Their implants were removed around 1 year after operation. We have measured local vertebral body angle (VBA) and superior-inferior endplate angle (SIEA) before and just after operation, approximately 1 and 2 year after initial operation. We also evaluated fracture severity according to load sharing classification. Results: Operation was performed 0 to 9 (mean 3.5) days after injury. Mean operating time was 101 min without VP and 165 min with VP. Mean blood loss was 131 ml without VP and 654 ml with VP. Load sharing classification score was 7.1 points without VP and 6.5 points with VP. Without VP, VBA was corrected 11.2° after surgery and loss of correction was 1° 1 year after removal. SIEA was corrected 12° after surgery and loss of correction was 9.8° 1 year after removal. Meanwhile, with VP, VBA was corrected 12.8° after surgery and loss of correction was 1° 1 year after removal. SIEA was corrected 10° after surgery and loss of correction was 9.3° 1 year after removal. Taken together, these results indicate that postoperative kyphotic change was related to disc level not fractured vertebra with or without VP. Conclusion: Fractured vertebra was reduced and maintained after surgery and kyphotic change was observed due to loss of disc height mostly after removal of implants with or without VP. Such change might be inevitable, as adjacent discs can be injured at the onset. Kyphotic change may thus be a limitation of this procedure. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

25 IMPLANT REMOVAL RESTORES MOBILITY AFTER FRACTURE OF THE THORACOLUMBAR SEGMENT: A RADIOSTEREOMETRIC STUDY Paul Axelsson, Bjo¨rn Stro¨mqvist Department of Orthopaedics, Ska˚ne University Hospital, 205 01 Malmo¨, Sweden

24 TEMPORARY PEDICLE SCREW FIXATION FOR THORACOLUMBAR BURST FRACTURES— COMPARATIVE STUDY WITH OR WITHOUT VERTEBROPLASTY Hiroyuki Aono, Keisuke Ishii, Hidekazu Tobimatsu, Yukitaka Nagamoto, Chiaki Horii, Motoki Iwasaki Osaka National Hospital, Osaka, Japan Introduction: Short-segment posterior instrumentation for thoracolumbar burst fracture has its merit based on superior correction of kyphosis by indirect reduction technique. However, failure of this procedure in loss of kyphosis correction has been frequently reported

Introduction: Randomized trials have stated that treating spinal burst fractures with reduction and posterior fixation is adequate without the use of bone grafting for definitive fusion. To restitute intervertebral mobility of such an unfused segment after fracture healing would theoretically unload the adjacent parts of the spine and reduce the risk for degeneration of these segments. Objectives: To study if late implant removal restores intervertebral mobility of a thoracolumbar segment treated with posterior instrumentation for unstable spinal fracture. Patients and methods: Seven patients were identified with implant related back pain minimum 18 months after a thoracolumbar fracture (Th12 or L1) treated with reduction and posterior instrumentation. The patients were all accepted for a late implant removal and the procedure also included percutaneous application of tantalum indicators for radiostereometric analysis (RSA). Three months after this

123

S680 surgery, each patient was examined with spinal RSA in flexion and extension. Measurement error of spinal RSA is 0.5–0.7 mm. The intervertebral translations and rotations of the thoracolumbar segment, induced by the change in position from flexion to extension, were measured. Progressive deformity was registered by conventional radiography and the overall clinical outcome was assessed by the patients. Results: According to RSA, all seven patients regained mobility of the fractured thoracolumbar segment expressed as significant intervertebral translation and/or rotation. Intervertebral mobility was also regained in the non-fractured lumbar segments included in 4 patients who had multilevel fixation to assure stability during the healing period. Thoracic segments used for the same purpose showed no mobility. In one patient, conventional radiography revealed a progressive kyphotic deformity 3 months after implant removal and the clinical outcome was poor. One patient graded a fair clinical outcome and 5 patients were good. Conclusion: Late implant removal may restore segmental mobility after posterior fracture fixation of the thoracolumbar segment if bone grafting is not applied. Arthrodesis of the facet joints is not induced by the trauma itself, by the surgical procedure placing pedicular screws in close relation to the joints or by the long period of postoperative fixation. Implant removal can be contemplated at least for younger patients in order to regain mobility and unload the adjacent segments. Disclosures: author 1: none; author 2: none.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Conclusions: Morphologically, the osteoporotic thoracolumbar junctional vertebral body fracture with a SPF represents the highenergy injury mechanism with the flexion forces of the anterior column and the tensile forces on the posterior column. The tensile forces on the posterior column is expected to be concentrated not on the posterior ligament but on the spinous process one level above the fractured body due to more caudally tilted spinous process in osteoporotic thoracolumbar spine. We suggest that the osteoporotic thoracolumbar junctional vertebral body fracture with a SPF is related with a poor prognosis. Disclosures: author 1: none.

27 THE VALUE OF BONE BIOPSY DURING PERCUTANEOUS VERTEBROPLASTY IN TREATMENT OF PRESUMED OSTEOPOROTIC VERTEBRAL COMPRESSION FRACTURES Bahadir Gokcen, Meric Enercan, Emel Kaya, Sinan Kahraman, Sinan Yilar, Mutlu Cobanoglu, Tunay Sanli, Erden Erturer, Cagatay Ozturk, Azmi Hamzaoglu Istanbul Spine Center, Istanbul, Turkey

26 OSTEOPOROTIC THORACOLUMBAR JUNCTIONAL VERTEBRAL BODY FRACTURE WITH A SPINOUS PROCESS FRACTURE: ITS CLINICAL AND RADIOLOGIC SIGNIFICANCES Tae-Hwan Kim Hallym University Sacred Heart Hospital, Anyang City, Korea Osteoporotic thoracolumbar junctional vertebral body fracture with a spinous process fracture: its clinical and radiologic significances Background: Several characteristic features of the osteoporotic thoracolumbar junctional vertebral body fracture with a spinous process fracture (SPF) were reported. In all cases, the SPFs were located just one level above the fractured vertebral body, and the direction of the SPFs was vertical or oblique vertical. However, there have been no reports about the mechanism or the prognosis. Retrospective comparative study was planned to assess the clinical and radiologic outcome of the patients with the osteoporotic thoracolumbar junctional vertebral body fracture with a SPF. Methods: A total of 391 patients with the osteoporotic thoracolumbar junctional (T10–L2) fragility fracture were analyzed. The patients were divided into the two groups. The group I comprised 332 patients with only vertebral body fracture, and the group II comprised 59 patients with vertebral body and a SPF. The clinical and radiologic parameters were compared between the two groups. Results: In all cases of the group II, the SPF were located just one level above the fractured vertebral body and the injury of the posterior ligament was not found. The direction of the SPF was vertical or oblique vertical (relatively perpendicular to the long axis of the spinous process). At the time of the injury, the group II patients showed worse outcomes in the anterior vertebral body compression percentage, cranial disc injury, kyphotic Cobb’s angle, and rate of the initial neurologic injury. There was no difference in the degree of canal encroachment between the two groups. At 12 months after diagnosis, the group II patients showed more severe posttraumatic kyphotic changes despite the similar changes in the anterior body compression percentage.

123

Summary: Bone biopsy in presumed osteoporotic vertebral compression fractures (VCF) treated via percutaneous vertebroplasty (PV) plays a significant role in the diagnosis of etiology. Design: Retrospective. Introduction: The most common cause of VCF is osteoporosis. Malignant conditions (metastasis, multiple myeloma (MM), lymphoma) also may be responsible for vertebral fractures. We have reviewed the biopsy results of patients treated via PV. The aim of this study is to determine the value of performing a routinely applied bone biopsy during PV. Methods: Between 2009–2013, 136 patients older than 50 y/o were included. Biopsies were performed during PV procedure. Pre-operative imagings were evaluated second time by a radiologist with the pathological results of the biopsies. Six patients with diagnosis of osteoporotic VCF presenting with abnormal blood tests were consulted with hematologist and the biopsy specimens of these patients were re-analyzed with CD-138 marker by the same pathologist. Results: 187 biopsies were obtained from 136 patients (85F, 51M). The mean age was 70.1 (50–96). In 17 patients (12.5 %) pathologic process underlying the fracture was MM, metastasis and lymphoma. MM was diagnosed in 13 patients (9.5 %). In 6 of 13 (46 %) patients with osteoporotic biopsy results, MM was diagnosed by re-analyzing the specimens with CD-138 marker. Metastasis was found in 3 patients (2.2 %). Lymphoma was found in 1 patient (0.7 %). Conclusion: This study found a 12.5 % incidence of malignancy (mostly MM) in patients with presumed osteoporotic VCF. Even the pathologic result is normal in 46 % of MM pts (6 of 13) at initial evaluation, consulting patients with abnormal blood test with hematologist and re-analyzing the pathology specimens with CD-138 marker diagnosed MM. We believe that routine vertebral body bone biopsy can play a significant role to assist in initiating concurrent medical treatment especially patients with multiple myeloma and metastasis. As a result, we recommend routine obtainment of bone biopsy during every PV procedure and also analyzing the biopsy specimens with CD-138 marker for MM. 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: consultant; Company = Medtronic.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 28 THE ROLE OF SINS (SPINAL INSTABILITY NEOPLASTIC SCORE) SYSTEM TO PREDICT THE INSTABILITY IN METASTATIC SPINAL TUMORS Sejun Park, ChongSuh Lee, Junyoung Lee, Sangsoo Kang, Wansseok Kim, Seongkee Shin, Kyungchung Kang Samsung Medical Center, Seoul, Korea Introduction: SINS (Spinal Instability Neoplastic Score) system has been introduced recently by Spinal Oncology Study Group to suggest widely-accepted definition of instability in spinal neoplasm. It was developed using the Delphi technique, which produces consensus by round table discussion not based on real patients’ data. However, there have been few studies about the clinical application of this system. In this study, we evaluated the predictive value of SINS system using patient cohort who undertook the radiotherapy as the initial treatment. Materials and methods: We evaluated 46 patients prospectively who had spinal oligo-metastases and underwent radiotherapy as the initial treatment between Oct 2012 and Mar 2014. Total 95 segments of 46 patients were evaluated. For all treated segments, 6 elements as well as total sum of SINS were scored. After 6 months since the radiotherapy, each treated segment was divided into success and failure group. Success group was defined as \50 % collapse and no need for other treatment modalities, and failure group as C50 % collapse or need for further treatments such as additional radiotherapy, surgery, or vertebroplasty. SINS score was compared in success and failure group respectively. Results: The 46 patients consisted of 32 men and 14 women with a mean age of 56.9. The sites of primary tumor were lung in 16 patients, liver in 8, kidney in 7, colorectal in 10, breast in 2. The average number of metastases per patient was 2.1 segments. There were 60 segments which was defined as success group and 35 segments as failure group. The mean SINS score was 6.2 point in success group, and 9.3 in failure group (p \ 0.001, independent t test). Regarding each element, the score of pain, collapse, and posterolateral involvement differed significantly between two groups (all p \ 0.001, respectively, independent t test). On univariate analysis, pain, collapse, and posterolateral involvement were shown as significant risk factors for instability as well (p = 0.002, p = 0.047, p = 0.008, respectively). Conclusion: SINS system could be utilized usefully in the decision making such as ‘‘radiotherapy versus surgery’’ for metastatic spinal tumor patients. Total sum of SINS score with less than 6 point can be regarded as stable segment. However, total sum of SINS score exceeding 9 points should be regarded as possible unstable segment, for which the surgical treatment considered as the initial treatment. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

29 CLINICAL AND RADIOLOGICAL OUTCOME AFTER TOTAL EN BLOC SPONDYLECTOMY AT THREE OR MORE LEVELS Katsuhito Yoshioka, Hideki Murakami, Satoru Demura, Satoshi Kato, Moriyuki Fujii, Takashi Igarashi, Noritaka Yonezawa, Hiroyuki Tsuchiya Department of Orthopaedic Surgery, Kanazawa University, Kanazawa, Japan

S681 Introduction: Multi-level total en bloc spondylectomy (TES) are sometimes required to achieve oncologically sufficient resection margins for spinal tumor, while it is still one of the most challenging surgery. This is the first case series on the clinical and radiological outcome after three or more levels of TES. Patients and methods: Between April 2006 and April 2012, 26 cases underwent three or more levels of TES at our institute. Excluding 4 cases with deteriorating general condition, radiological evaluation was performed more than 1 year in 22 cases, with an average age of 49.6 years. The mean follow up time was 37.4 months. Six cases had primary tumors, and the other 16 cases had metastatic tumors. In all cases reconstruction was performed with pedicle screw fixation and a titanium mesh cage implant filled with autograft. We assessed clinical results, perioperative complication, and radiological outcome of reconstruction on X-ray and CT with multiplanar reconstructions (MPRs). Results: The median operative time was 565 min, with the median intraoperative blood loss of 820 mL. Eight of the 22 patients did not need blood transfusions. Complications showed in 8 cases including 3 cases with CSF leakage. Although 18 cases developed distant metastasis, all 22 cases had no evidence of local recurrence at the surgical site of the spine during the follow-up period. Cage subsidence ([2 mm) was observed in 11 cases (50 %). Eight of the 11 patients showed the appearance of subsidence 1 month after the surgery. One case with 7 mm cage subsidence 1 month after surgery developed instrument failure with rod breakage. Revision surgery was performed with rod replacement by the use of four rods 8 months after first surgery. Radiolucent lines around the pedicle screws were observed in only one patient at the distal end, 40 months after surgery, but without clinical symptoms. The mean length of the resection vertebra was 84.6 mm (52–124 mm), while the length of the shortening was 10.4 mm (3–22 mm). Conclusion: No local recurrence was observed after TES at three or more levels. In the present study, 21 cases achieved good spinal stability in our reconstruction method. However, there was one instrument failure that required revision surgery. Although our TES surgery might not be all round, we consider that it was effective when applied at three or more levels of TES. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

30 RESECTION OF VERTEBRAL OSTEOID OSTEOMAS IN FULL-ENDOSCOPIC TECHNIQUE IN ADOLESCENTS ¨ zdemir, Patrick Hahn, Martin Komp, Harry Merk, Semih O Georgios Godolias, Sebastian Ruetten Center for Spine Surgery and Pain Therapy; Center for Orthopaedics and Traumatology of the St. Elisabeth Group, Catholic Hospitals Rhein-Ruhr, St. Anna Hospital Herne, Marienhospital Herne University Hospital, Marien Hospital Witten, Germany The osteoid osteoma is a benign, primary osteoblastic bone tumor without tendency to spread. It frequently presents in adolescence. A typical symptom is diffuse pain. The diameter is \1.5 cm. Findings [2 cm are classified as osteoblastomas. 20 % of osteoid osteomas affect the spine and dorsal structures are affected in 70 % of all cases. Radiological evaluation characteristically shows the nidus with a surrounding sclerotic zone and soft-tissue reaction. The natural healing process takes several years. Drug therapy involves NSAR drugs while as an interventional procedure CT-guided percutaneous techniques are used. Disadvantages of these procedures are recurrences, because of incomplete resection, and lack of

123

S682 histological analysis. Location of the osteoid osteoma at a distance of \1 cm to neural structures is discussed as a contraindication. En-bloc resection is the standard surgical procedure. The object of this study was to investigate the full-endoscopic resection technique used for osteoid osteomas in adolescent patients. 19 patients [(13m, 5f); age 12–19 (Ø = 16)] with an osteoid osteoma located at the lumbar and thoracic spine underwent surgery using the full-endoscopic technique. Dorsal structures were affected in 16 cases (9 9 lamina, 4 9 spinous process, 3 9 pedicle), ventral structures were affected in 3 cases. Preoperative pain was on average at level 6 on the VAS 0–10. All patients were treated conservatively in advance for [3 months. The follow-up period was 24 months. Follow-up examinations could be carried out in 17 patients. The average operation time was 28 min. A measurable blood loss was not identified. In each case, material for histological examination was harvested in the operation. The histological result was positive in each case. No complication occurred during or after the operation. 3 patients with secondary scoliosis before the operation subsequently demonstrated normal growth. There were no clinical or radiological signs of recurrence. All patients were pain free in the period following surgery. The full-endoscopic resection of osteoid osteomas on the spine is appropriate and rarely leads to complications. The procedure can be applied in immediate proximity to neural structures. There were no recurrences in this study. Trauma resulting from the operation was low compared with open en-bloc resection. Clear histological diagnosis is possible. The fullendoscopic technique is an alternative to open and percutaneous procedures. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Results: In PSM group, idiopathic-type scoliosis was developed in 7 pigs. The mean coronal Cobb angle was 27.8° immediately postoperatively and progressed to 59.3° after 8 week tethering period. The average scoliosis was 55.2° on average 8 weeks after removal of the tethering. No scoliosis was observed in the control group. In PSM group, the number of chondrocytes of hypertrophic zones in the concave side was less than that of the convex side, and the irregular arrangement of chondrocytes was observed in the hypertrophic zones in PSM group. In PSM group, the thickness of the hypertrophic zone was significantly different between the concave and convex side (P \ 0.05). Of the convex side, the average area of cell-nest and the average number of cells in cell-nest were larger than those of the concave side in PSM group (P \ 0.05). However, no difference was observed between the concave and convex side of the number of proliferative chondrocytes (P [ 0.05). Conclusions: In porcine scoliosis model, it is asymmetrical loading that contributes to unbalanced growth pattern on both sides, resulting in vertebral wedging. Growth activity on the growth plate on the concave side is highly inhibited, characterized by significantly decreased cell number and zone height of the hypertrophic zone. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

32 PROXIMAL JUNCTIONAL KYPHOSIS: EFFECT OF MULTILEVEL POSTERIOR LIGAMENT DISSECTION AFTER POSTERIOR INSTRUMENTATION ON ADJACENT SEGMENT BIOMECHANICS Tobias Lange, Raul Mayr, Georg Gosheger, Christian Heinrichs, Werner Schmo¨lz, Tobias L. Schulte

DEGENERATIVE, SCOLIOSIS Department of Orthopedics, University Hospital Mu¨nster, Germany 31 THE HISTOMORPHOLOGICAL CHANGE OF THE GROWTH PLATE UNDER ASYMMETRICAL LOADING IN PORCINE SCOLIOSIS MODEL Xin Zheng, Yong Qiu, Xu Sun, Bangping Qian, Zezhang Zhu, Yang Yu, Bin Wang From the Spine Surgery, Drum Tower Hospital, Nanjing University Medical School, Nanjing, China Background: It is thought that the vertebrae become increasingly wedged as a result of asymmetrical loading and asymmetrical growth. Purpose: To investigate the histological changes of the vertebral growth plates under asymmetrical loading in the porcine scoliosis model (PSM). Methods: A total of 13 female Yorkshire pigs (age, 6 weeks; weight, 6–8 kg) were enrolled in this study. In PSM group, 8 pigs were instrumented and tethered ipsilaterally through three separate incisions. After 8 weeks, the posterior instrumentations were removed and the pigs were observed with an additional 8-week period. In the control group, the spine of pigs were implanted internal fixation but not tethered. Then, all the pigs were euthanized, and the spines were harvested for histologic study. The specimens were embedded in paraffin, followed by decalcification and then stained by hematoxylin and eosin (HE) for histomorphometric analysis. The thickness of the hypertrophic zone in iliac cartilage, the area and number of chondrocytes in cell-nest and the number of proliferative chondrocytes were analyzed using the Image-Pro Plus soft.

123

Several factors increasing the risk for Proximal Junctional Kyphosis (PJK) after posterior instrumentation which is widely performed treating spinal pathologies are discussed in literature. Posterior elements including spinous processes and ligaments are often sacrificed on purpose (for neural decompression and spondylodesis) or accidentally. The aim of this biomechanical study was to evaluate the potential destabilizing effect of an incremental dissection of posterior ligaments causing PJK. Twelve calf lumbar L1–L6 spines were instrumented from L3 to L6 using a titanium pedicle screw rod system (5.5 mm rod diameter, pedicle screws 7.0 9 45 mm and 6.0 9 45 mm). Specimens were randomly assigned to one of the following study groups. In the treatment group 6 specimens underwent an incremental (from caudal to cranial) posterior ligament dissection (supra- and interspinous ligaments) followed by a cyclic flexion motion (250 cycles, moment: +2.5 to +20.0 Nm) and a flexibility testing with pure moments of ±15.0 Nm after each stepwise dissection. Segmental range of motion (ROM) of each segment (L1–L2, L2–L3, L3–L4, L4–L6) in flexion/ extension was measured using an ultra-sound based motion analysis system. In the control group 6 specimens underwent the same test protocol but without posterior ligament dissection. Regarding the adjacent not instrumented segment L2/3 the normalized mean ROM increased to 104.7, 107.3 and 118.9 % after dissection of the posterior ligaments L4–L6, L3–L4 and L2–L3 in the treatment group. In the control group, mean ROM increased only to 103.2, 106.7 and 108.7 % after cyclic flexion motion without dissections for each state, respectively. In both groups the incremental increase of ROM after each surgical step was significant. The differences between the two groups in change of ROM over testing states were also significant.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 These results show that an dissection of the posterior ligaments in the adjacent segment of the posterior instrumentation leads to a significant ROM increase of the adjacent segment itself and therefore is a potential risk factor for PJK. Nevertheless, it could be shown that there is also a slight trend towards an increase of ROM in the adjacent segment even after dissection of the posterior ligaments of the lower instrumented segments. Considering that calf lumbar spines were used in this study, it may be hypothesized that an even major effect could be expected in human thoracic spines, especially in elderly patients. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

33 3D MOTION ANALYSIS IN LUMBAR SPINAL STENOS WITH DEGENERATIVE SPONDYLOLISTHESIS BEFORE AND AFTER DECOMPRESSIVE SURGERY USING PROVOCATION COMPUTED TOMOGRAPHY. AN RCT COMPARING LAMINECTOMY TO BILATERAL FORAMINOTOMY Peter Fo¨rsth, Per Svedmark Uppsala University/Stockholm Spine Center, Uppsala, Sweden Introduction: In this randomised pilot study we used the 3DCT technique to study in vivo if preservation of the segmental midline structures gives less postoperative instability when compared to traditional central decompression with laminectomy in patients with lumbar spinal stenosis (LSS) with concomitant degenerative spondylolisthesis (DS). Methods: 22 patients (mean age 68 years) with typical symptoms of LSS with claudication and back pain were included in the study. All had MRI findings of central spinal stenosis with concomitant DS ([3 mm) in one or two adjacent lumbar levels. The patients were randomised to either laminectomy, (LE, central decompression with removal of the midline structures) or bilateral laminotomies (BL) with preservation of these structures. Each examination comprised two CT scans, one in provoked flexion, and one in provoked extension and was made preoperatively and 6 months postop. Analyses of the segmental movements were done with an in-house developed software tool, which creates rigid body transformation that merges the CT volumes into spatial alignment. These volumes were then merged or fused for evaluation and further analysis. The accuracy for this method is 0.6 mm in translation and 1 degree in rotation. Patient reported outcome measures (PROM) were collected from the register Swespine preop and 1 year postop. Results: The mean preop values for 3D rotation and translation were 6.2° and 1.8 mm. The mean increase in 3D rotation 6 months postop was 0.4° after BL and 0.7° after LE (p = 0.79). The mean increase in 3D translation was 0.2 mm after BL and 0.8 mm after LE (p = 0.42). Both groups had a significant improvement in PROM at follow up compared to preoperatively with no differences between the two treatments. The improvement in ODI was 20 after LE and 21 after BF (p = 0.90). Two patients in the BL group were excluded due to conversion to LE as a result of dural tears in need of suture. Conclusion: The results from this pilot study suggest that there is negligible increase in segmental instability after decompressive surgery for LSS with DS. Traditional LE with removal of the midline structures did not create a greater instability compared to if these structures were preserved with BL. The latter technique was associated with more surgical complications. The results suggest that central decompression with laminectomy is a favorable technique for decompression in patients with LSS with DS. Disclosures: author 1: none; author 2: none.

S683 34 DOES THE FACET JOINT VIOLATION FOLLOWING TRANSPEDICULAR INSTRUMENTATION OF THE LUMBAR SPINE CAUSE ADJACENT SEGMENT PATHOLOGY OR NOT? : A MATCHED CONTROL STUDY Chang-Hoon Jeon, Nam-Su Chung, Han-Dong Lee Ajou University School of Medicine, Suwon, Korea Purpose: The aim of this study was to evaluate whether the facet joint violation following transpedicular instrumentation of the lumbar spine is a risk factor for adjacent segment pathology (ASP) or not. Introduction: Facet joint violation by pedicle screws can potentially result in a higher rate of adjacent segment degeneration and facet joint arthritis. Although numerous studies have investigated risk factors in ASP, there was no study to compare the effect of facet joint violation on ASP with lumbar spine surgery without facet joint violation. Methods: Forty-five consecutive patients who showed facet joint pedicle screw violation after lumbar fusion were identified as a study group (V group). As controls groups, there was an age- and gendermatched cohort of 45 patients with no facet joint pedicle screw violation after lumbar fusion (NV group) and a cohort of 45 patients who had undergone a lumbar decompression (D group). The diagnosis of radiological ASP was made using plain radiographs and follow-up MRI. The Pfirrmann’s classification was used to grade the adjacent disc degeneration (ADD). Facet joint violation was evaluated on the postoperative CT images, using the Shah’s classification for facet joint violation. Results: The average age was 57.3 ± 17.5 years in the V group, 59.1 ± 18.1 years in the NV group, and 59.9 ± 16.5 years in the D group (P = 0.843). There were 21 males in the V group, 25 in the NV group, and 29 in the D group (P = 0.756). The mean follow-up duration was 65.0 ± 35.6 months (range 24–192 months). In the V group, there was superior facet joint violation in 40 cases (88.9 %) and inferior facet joint violation in 5 cases (11.1 %) and there were ASP in upper segment (15 cases, 33.3 %) and in lower segment (2 cases, 4.4 %). In NV group, there was ASP in upper segment (8 cases, 20 %) and in lower segment (1 case, 2.2 %). In D group, there was ASP in upper segment (3 cases, 6.7 %) and in lower segment (1 case, 2.2 %). The prevalence of radiological ASP was significantly higher in the V group (37.7 %) than those of NV group (20 %) and D group (8.9 %) (P = 0.032). ADD was also significantly higher in the V group (53.3 %) than those of NV group (31.1 %) and D group (17.7 %) (P = 0.002). CONCLUSIONS: The facet joint violation following transpedicular instrumentation of the lumbar spine is a risk factor for ASP and ADD. We suggest that the careful pedicle instrumentation can reduce the incidence of facet joint violation and ASP. Disclosures: author 1: none; author 2: none; author 3: none.

35 PROTECTIVE EFFECTS OF ADJACENT SEGMENT DEGENERATION AFTER LUMBAR FUSION WITH POSTERIOR LIGAMENTS COMPLEX PRESERVED IN EIGHT-YEAR MINIMUM FOLLOW-UP Liang Yan, Baorong He, Dingjun Hao Hong Hui Hospital, Xi’an Jiaotong University College of Medicine, Xi’an, China Purpose: To assess the protective effects of ASD after lumbar fusion with posterior ligaments complex preserved.

123

S684 Methods: This prospective study included 190 patients with L4–5 degenerative spinal disorders. From January 2004 to January 2007, patients were randomly allocated into 2 groups who were statistically similar with respect to demographic and clinical data. Patients in Group A were allocated for total laminectomy and L4–5 fusion, Group B for semilaminectomy and fusion. The patients were followed up postoperatively and were assessed with regard to radiological and clinical outcomes. Radiologic outcomes were assessed mainly on the basis of disc degeneration, facet joint degeneration and bone fusion. Clinical outcomes were evaluated mainly with use of visual analog scale (VAS) for pain and the Oswestry Disability Index (ODI). Results: 172 of 190 (90.5 %) patients were available for at least 8-year radiological and clinical follow-up data: 91.4 % (84/92) patients in Group A and 89.8 % (88/98) patients in Group B. At the last follow-up, bone fusion was achieved in all patients. ASD was proven in 103 (59.9 %) of 172 patients. The incidences of radiographic and symptomatic ASD were 56.9 % (98/172) and 2.9 % (5/ 172), respectively. The incidence of ASD in Group B was significantly lower than Group A. Results of clinical outcomes showed lower VAS and ODI scores in two groups compared with those preoperative, but Group B had greater improvement on the ODI scores compared with Group A in patients with ASD. Conclusions: The incidence of ASD was significantly higher after total laminectomy than semilaminectomy. Preserving the posterior complex as much as possible during lumbar fusion surgery plays an important role in preventing ASD and in reducing the reoperation rate. Disclosures: author 1: none; author 2: none; author 3: none.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 (5 %), had disc tissue infected by p. acnes. Two of these four patients had no Modic changes. Discussion: In the present study we found a low (5 %) incidence of p. acnes. These findings are contrary to a recent study review, which detected nine studies examining the presence of bacteria in disc tissue. Of these, a pooled estimate of the proportion with positive samples showed 34 %. Furthermore, that p. acne was the most prevalent bacteria. The low incidence found in this study may be due to a low contamination-rate i.e. the strict perioperative sample handlingtechnique; am. Kamme, and/or open surgical technique versus scopic techniques. Conclusion: A low incidence of p. acnes in herniated discs (5 %) was found using an open MIS approach combined with a tissue-handling technique am. Kamme. No correlation was found between Modic changes and p. acnes. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none.

37 THE LONG-TERM OUTCOME OF LUMBAR FUSION IN THE SWEDISH LUMBAR SPINE STUDY Rune Hedlund, Christer Johansson, Olle Ha¨gg, Peter Fritzell, Tycho Tullberg Sahlgrenska University Hospital, Department of Orthopedics, Gothenburg, Sweden

36 PROPIONIBACTERIUM ACNES, HERNIATED DISC AND MODIC CHANGES. NO CORRELATION FOUND IN 80 PATIENTS Laura Mathiesen, Anne Kirstine Hansen, Mikala Wang, Thomas Bender, Peter Lemche, Soeren Fruensgaard, Malene Laursen Regional Hospital Silkeborg; Center of Elective Surgery, Viborg, Denmark Aim: The aim of this study was to investigate the presence of propionibacterium acnes in patients operated due to disc herniation. Furthermore, to test a hypothesised correlation between the presence of Modic changes and bacteria. Background: A recent systematic review has found moderate evidence for correlation between the presence of p. acnes and back pain and/or Modic changes in patients with disc herniation. However, further studies are needed to determine whether these findings are results of contamination or represent low grade infection. Method: Eighty patients undertaking an open MIS approach due to disc herniation. A total of six disc-tissue samples, using sterile single use instruments, were harvested for microbiological analysis and PCR. Five samples were inoculated into thioglycolate agar tubes, incubated long term (14 days) and observed for growth. Positive samples were interpreted according to the Kamme and Lindberg definition: the ‘‘standard’’ used for discrimination between contamination and true infection: Growth in 0–1 of 5 samples: Contamination. Growth in 2 samples: Possible infection. Growth of the same agent in 3–5 samples: Infection. The first of the six samples was stored at -80 °C and later subjected to a bacteria molecular analysis, PCR. The preoperative MRI scans were interpreted for Modic changes. Results: Eighty patients, mean age of 44.5 years. Modic changes was seen in 47 % of patients, with a mean age 49 years. Four patients

123

Background context: Current literature suggest that fusion of the lumbar spine in chronic low back pain (CLBP) in the long term does not result in an outcome clearly better than structured conservative treatment modes. Purpose: To assess the long-term outcome of lumbar fusion in CLBP, and also to assess methodological problems in long-term RCTs. Study design: A prospective randomized study. Patient sample: 294 patients, 144 women and 150 men, with a mean age of 47 years (range 28–72) with CLBP of at least 2 years duration were randomized to lumbar fusion or non-specific physiotherapy. The mean follow-up time was 12.8 years (9–22 years). The follow-up rate was 92 %. Outcome measures: Global Assessment (GA) by the patient was the primary outcome measure. Oswestry Disability Index (ODI), VAS back and leg pain, Zung depression scale, work status, pain medication and pain frequency was also documented. Methods: Standardized outcome questionnaires were obtained before treatment, at 2 years and at long-term follow up. To optimize control for group changers four models of data analysis were used, according to: 1/intention to treat (ITT), 2/’’as treated’’ (AT), 3/per protocol (PP), and 4/in the conservative group automatically classify group changers as unchanged/worse in GA (GCAC). Results: Except for the ITT model the GA was significantly better for fusion. The proportion of patients ‘‘much better’’ or ‘‘better’’ in the fusion group was 66, 65, and 65 % in the AT, PP, and GCAC models, respectively. In the conservative group the same proportions were 31, 37, and 22 %, respectively. However, the ODI, VAS back pain, work status, pain medication and pain frequency was similar between the two groups. The level of disability as assessed by ODI at long term follow-up (range 35.8–40.1) was moderate, approaching severe levels, for both treatments using all models of analysis. Initially 8 % crossed over from the surgical to the conservative group, whereas in the conservative group 36 % eventually were fused.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Conclusions: From the patient’s own perspective lumbar fusion can be considered in CLBP, as defined in the present study. On the other hand, the level of disability showed a limited effect on disability and pain, best assessed by the PP analysis, which is not biased by group changers. An important finding of the study is that the crossover problem often transforms a RCTs to an observational study, representing only Level 2 evidence. Disclosures: author 1: grants/research support; Company = Zimmer, De Puy, consultant; Company = Zimmer, K2M, Globus; author 2: none; author 3: consultant; Company = Spine Center Go¨teborg, stock/shareholder; Company = Spine Center Go¨teborg; author 4: none; author 5: stock/shareholder; Company = Stockholm Spine Center, Global Health Partner.

38 REVISION SURGERY FOR RESOLVING LUMBAR INTERSPINOUS DEVICE FAILURE Dong-Hwa Heo, Seung-Hwan Yoon The Leon Wiltse Memorial Hospital, College of Medicine, Inha University, Gyeonggi-do, Korea Purpose: Lumbar interspinous devices were developed to treat degenerative lumbar disease. Previous studies have reported on the foraminal widening effects and effectiveness of segmental stabilization of interspinous devices. We recently experienced cases with poor postoperative clinical outcomes and delayed complications of interspinous devices. In the present study, we aimed to analyze the clinical and radiologic characteristics of patients who required revision surgery following implantation of interspinous devices. Methods: In total, 36 patients were enrolled in the present study. All enrolled patients underwent revision surgeries following operations involving interspinous devices such as interspinous U and DIAM. We analyzed the clinical parameters such as visual analog scale (VAS) of pain, Oswestry disability index (ODI), history of pain treatment following interspinous device placement including administration of opioid medication and epidural injections, time from initial surgery involving interspinous devices to revision surgery, and methods of revision surgery. Moreover, we examined the radiologic parameters such as foraminal stenosis, grade of spondylolisthesis, erosion of the spinous process, migration of the interspinous device, and extent of disc degeneration of the operated segment. Results: The mean time from initial surgery to revision surgery was 3.9 ± 2.9 years. Of 36 patients, 29 had a history of opioid administration or epidural injection following the initial surgery. Foraminal re-stenosis was noted in 16 patients, recurrent disc herniation was noted in 8 patients, and aggravation of spondylolisthesis was observed in 12 patients after the placement of interspinous devices. We performed device removal and fusion surgery in 31 patients and device removal with discectomy in 5 patients. Recurrent disc herniation was significantly associated with pull-out of the interspinous device (P \ 0.05). Furthermore, foraminal re-stenosis was significantly associated with erosion of the spinous process by the interspinous device (P \ 0.05). ODI and VAS were significantly improved after revision surgery Conclusions: Based on the results of our study, the long-term outcomes of interspinous devices may be poor due to pull-out of the device and erosion of the spinous process by the device. Interspinous devices may not provide sufficient stabilization for unstable lumbar segments. Disclosures: author 1: none; author 2: none.

S685 39 ASSOCIATED LUMBAR SCOLIOSIS DOES NOT AFFECT OUTCOMES IN PATIENTS UNDERGOING FOCAL MINIMALLY INVASIVE TRANSFORAMINAL LUMBAR INTERBODY FUSION (MISTLIF) FOR NEUROGENIC SYMPTOMS: A MINIMUM 2-YEAR FOLLOW-UP STUDY Anupreet Bassi, Kae Sian Tay, William Yeo, Wai Mun Yue Singapore General Hospital, Singapore Rationale: To determine whether associated lumbar scoliosis will result in different outcomes in patients undergoing focal MISTLIF for neurogenic symptoms, without specific correction of the scoliosis. Methods: Between 2005 and 2012, 230 patients who underwent MISTLIF for lumbar pathologies with neurogenic symptoms with minimum 2-year follow-up were included. Prospectively collected data were retrospectively analysed. Patients with previous lumbar operations, tumour, infections and acute trauma were excluded. Based on the lumbar spine Cobb angle 2 groups were formed—Scoliosis Group (SG n = 57) with Cobb angle [10° and Non-scoliosis Group (NSG n = 173). Both groups were compared for demographics, peri-operative parameters, clinical and radiological outcomes. Clinical outcomes were assessed using the 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. Radiological findings included Cobb angle, fusion, implant failure/ loosening/migration and Adjacent Segment Disease (ASD). Patients were evaluated pre-operatively, 2 years and 5 years after surgery. Results: The average follow-up was 2.88 years (2–5.9) and 2.71 years (2–6.5) for SG and NSG respectively (p \ 0.05). Differences in mean age (SG-62.9 ± 10.9 years and NSG-57.6 ± 13.1 years) and sex (SG87.7 % female, NSG-64.2 % female) were significant (p \ 0.05) while race distribution and BMI were similar (p [ 0.05). Levels operated, type of graft used, blood loss, fluoroscopic time, morphine used, time to ambulation, length of operation and length of hospitalization were similar (p [ 0.05). Pre-operative and post-operative clinical outcomes were similar in both groups (p [ 0.05) except SF36 Physical Function Score which was found to be significantly higher in NSG at final followup (p \ 0.05). Clinical outcomes scores improved significantly postoperatively (p \ 0.05) and grade 1 fusion (Bridwell classification) was achieved in 93 % in both groups (p [ 0.05). Complications were found in 10 (17.5 %—1 major requiring reoperation and 9 minor) patients in SG and in 18 (10.4 %—3 major requiring reoperation and 15 minor) patients in NSG p [ 0.05. There were 15 new ASD cases (14 NSG, 1 SG p [ 0.05), only 1 (NSG) required surgery. Conclusion: Associated lumbar scoliosis does not result in significantly different peri-operative, clinical and radiological outcomes in patients undergoing MISTLIF for neurogenic symptoms. Disclosures: author 1: none; author 2: none; author 3: none; author 4: grants/research support; Company = Medtronics, Depuy-Synthes, consultant; Company = Medtronics, Depuy-Synthes, other financial report; Company = Nuvasive.

40 ROLE OF LUMBAR INTERBODY FUSION IN ADULT SPINAL DEFORMITY SURGERIES: DOES THE NUMBER OF CAGES MATTER? Vincent Fie`re, Evalina L. Burger, Michael S. Chang, Sean Molloy Centre Orthope´dique Santy France; University of Colorado SOM, US; Sonoran Spine, US; NHS Royal National Orthopaedic Hospital, UK

123

S686 Introduction: While lumbar interbody fusion (IBF) is a commonly surgical technique used to achieve anterior fusion, there is a significant variability in possible approaches and number of IBFs in the setting of adult spinal deformity (ASD) surgery. This study investigates the impact of number of IBFs on radiographic and clinical outcomes of ASD. Materials and methods: Retrospective review of multicenter prospective database. 60 out of 72 consecutive patients operated between 2012 and 2014 for an adult degenerative scoliosis or kyphosis have reached their 6 months follow-up and are eligible for the study. Demographics, surgical metrics, complications, as well as baseline and 3–6 month post-operative radiographs and Health related quality of life (HRQL) were collected. The cohort was subdivided in two groups based on the number of IBFs: 3 or more IBFs (C3IBFs; n = 33), 2 or less IBFs (B2IBFs; n = 27). Group comparisons were carried out using Chi square and unpaired t test with a 0.05 level of significance. Results: From a surgical perspective, there were no significant differences in rod material/diameters, ALIF vs. DLIF distribution, number of levels fused posteriorly, and laminectomy between groups. Patients with C3IBFs were less likely to receive BMP (45.5 vs 77.7 %, p = 0.011) and more likely to undergo fusion extended to the pelvis (94 vs 67 %, p = 0.007), with the use of SPO (87 vs 59 %, p = 0.011). These patients also had significantly less peri-operative complications (6.06 vs. 29.6 %, p = 0.015), including dural tears (0 vs. 11 %, p = 0.049). Both groups had similar post-operative complications 15.1 vs. 18.5 %. Post-operatively, patients with C3IBFs had larger PI-LL correction (9° vs. 1.8°, p = 0.025), and coronal lumbar curve correction (15.4° vs. 4.9°, p = 0.003), without significant difference in PJK incidence (33 vs 26 %). Patients with C 3IBFs had significantly larger improvement in SRS (p [ 0.05). Conclusions: The role played by lumbar interbody cages in fusion is described since decades. We now demonstrate that the number of cage is significantly associated with the restoration of a balanced spine as well as clinical improvement. Disclosures: author 1: consultant; Company = Medicrea Clariance, royalties; Company = Medicrea Clariance; author 2: grants/research support; Company = Aesculap,consultant;Company = Medicrea, Paradigm, Signus; author 3: grants/research support; Company = Medicrea, consultant; Company = Medtronic, Stryker; author 4: grants/research support; Company = Metronic, consultant; Company = Zimmer, K2M, royalties; Company = Medicrea.

41 CLINICAL DISCRIMINATION PROPERTIES OF THE SRSSCHWAB CLASSIFICATION IN 292 NON-US ADULT SPINAL DEFORMITY PATIENTS 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 three 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. Until now the clinical impact of the classification has only been evaluated in patients from the American International Spine Study Group database, and the Pelvic Tilt

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 modifier was not shown to discriminate HRQOL scores between patients with grade + and ++. Objective: We evaluated the discriminative properties of the SRSSchwab Classification with the most common HRQOL measures in a non-US population of adult spinal deformity (ASD) patients. Methods: Between March 2013 and May 2014 all patients at least 18 years of age with sufficient standing antero-posterior and lateral radiographic images taken at one outpatient clinic to evaluate a spinal deformity were prospectively enrolled. Exclusion criteria were spine deformity surgery within 6 months or missing questionnaires. Analysis of variance across modifier grades and post hoc individual grade comparisons were performed for the following outcomes: ODI, SRS22r, SF36, EQ 5D scores including the SRS22r domains and visual analogue scales (VAS) for back and leg pain. All P-values were corrected for multiple testing using the sequential rejective Bonferroni method. Results: A total of 460 patients were eligible and 292 were included. 30 had deformity surgery within 6 months and 138 did not return questionnaires. All HRQOL measures except mental scores and VAS for back pain were significantly different between patients with normal and abnormal modifiers, P-values ranged \0.001 to 0.027. The significance of differences in HRQOL scores between individual grades were equivocal for all modifiers, however the PT modifier did not discriminate between grade + and ++ according to any score or scale, P-values [0.110. Conclusion: The SRS-Schwab ASD Classification reliably discriminates between patients with abnormal and normal sagittal modifiers according to HRQOL, but the PT modifier is unable to discriminate HRQOL scores between patients with grade + and ++. 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 = Medtronic, Globus Medical, K2M, consultant; Company = Medtronic, Globus Medical, K2M; author 6: grants/research support; Company = Medtronic, Globus Medical, K2M

THORACO-LUMBAR SPINE, DEGENERATIVE II

42 COMBATING OXIDATIVE STRESS-INDUCED PREMATURE SENESCENCE (SIPS) IN THE INTERVERTEBRAL DISC (IVD) Olga Krupkova, Sonia Rossi, Marian Hlavna, Juergen Klasen, Stephen J. Ferguson, Karin Wuertz-Kozak D-HEST, ETH Zurich, Switzerland During progressive IVD degeneration, nutrition supply is compromised. Together with a degradation of aggrecan, which inhibits infiltration, low nutrition level leads to enhanced vascular ingrowth into the disc. However, this causes oxygen accumulation in an originally avascular tissue, with a subsequent increase in oxidative stress, SIPS and apoptosis. Recently, cellular senescence has been shown to be positively correlated with the degree of IVD degeneration. Our previous research has shown that the polyphenols epigallocatechin 3-gallate (EGCG) or resveratrol significantly inhibit inflammatory responses and IVD-related radiculopathy via modulation of major stress signaling pathways. However, their effect on SIPS in IVD cells has not yet been studied. The aim of this project was to first develop in vitro model system of SIPS and to secondly test the potential of the above mentioned substances as anti-senescence compounds, including their mechanism of action.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 IVD cells were isolated from patients undergoing surgeries due to disc-related pathologies and cultured in vitro up to passage 3 (n = 15). Senescence was induced using sub-lethal H2O2 (250 lM, 2 h) with subsequent recovery period up to 15 days with media exchange at day 5 and 10. Intracellular reactive oxygen species generation, indicated by mitochondrial depolarization, was verified by JC-1 staining/microscopy at day 1. Percentage of senescent cells was measured by b-galactosidase assay at day 1, 5, 10 and 15. Proliferation and metabolic activity of cells was determined at day 8 and 15. Activity of the p53/p21 pathway was measured at day 15 by Western blotting. Short-term (24 h) and long-term (10 days) effects of EGCG on SIPS were investigated by analyzing its antioxidant activity (=direct ROS trapping) and its antioxidant-independent action (=interaction with cellular components). Sub-lethal H2O2 induces senescence of IVD cells, which gradually increases from day 1 (&10 %) up to day 15 (&45 %) compared to control cells (&8 %). SIPS is characterized by the inhibition of proliferation and activation of p53/p21, showing cell cycle arrest. Addition of EGCG did not influence cell cycle arrest markers, but diversely decreased the rate of senescence, indicating ongoing DNA repair process. An in vitro IVD SIPS model was successfully developed, validated and used to test the effect of EGCG on SIPS. EGCG exhibited moderate anti-senescent activity, hence specific mechanism of its action is currently being studied. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: grants/research support; Company = Herzog Egli Foundation.

43 REPEATED 3.0 TESLA MAGNETIC RESONANCE IMAGING AFTER CLINICALLY SUCCESSFUL LUMBAR DISC SURGERY Clemens Weber, Øystein Petter Nygaard, Ole Solheim, Sasha Gulati Department of Neurosurgery, Stavanger University Hospital, Stavanger, Norway; Department of Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway Methods: In the present study of 30 patients, that underwent 3.0 Tesla MRI within 24 h, at 6 weeks and 3 months postoperatively, we aimed to describe the natural course of postoperative MRI findings after lumbar disc surgery with favorable clinical outcomes and the occurring changes of theses imaging findings over time. The interrater agreements of these MRI findings were assessed. Results: In the immediate postoperative phase, compression or dislocation of the nerve root at the operated level is common without having a clinical correlate. Contrast enhancement of the ipsilateral nerve root in the operated level was observed in almost 50 % of the early postoperative images, but over time contrast enhancement vanished. At day 1, 80 % of the successfully operated patients have a residual mass that typically shrinks or disappears over the next 12 weeks. If the residual mass has a light signal as compared to muscle on the same T2-weighted image it seems to have a greater tendency to resolve spontaneously. Conclusion: When using high field MRI in the early postoperative phase after lumbar disc surgery several imaging changes can be found. Even if the patient is free of radicular symptoms residual masses with displacement or compression of the nerve root can be seen. Due to the higher image quality of 3.0 Tesla MRI nerve root changes and residual masses in the spinal canal can be evaluated more accurately. Early reoperations in patients with residual masses with

S687 lighter signals on T2-weighted images at the operated side and level should be done with caution. Disclosures: author 1: none; author 2: none; author 3: grants/research support; Company = St. Olavs University Hospital, Trondhrim, Norway; author 4: none.

44 EFFICACY AND SAFETY OF CONDOLIASE IN PATIENTS WITH LUMBAR DISC HERNIATION: A PROSPECTIVE RANDOMIZED CONTROLLED CLINICAL TRIAL Kazuhiro Chiba, Yukihiro Matsuyama Department of Orthopaedic Surgery, National Defense Medical College; Department of Orthopaedic Surgery, Hamamatsu University School of Medicine, The Japanese SI-6603 Study Group, Saitama, Japan Condoliase, an enzyme that specifically degrades glycosaminoglycans, the main constituents of the nucleus pulposus, and reduces the compressions on nerves, is expected to serve as a less-invasive curative treatment drug for lumbar disc herniation (LDH). We have conducted a multicenter, double-blind, randomized, placebo-controlled trial to evaluate the efficacy and safety of condoliase in patients with LDH. The patients were 20–70 years of age, with a contained LDH at L4–L5 or L5–S1 and a positive straight leg raising test, who showed no improvement of their symptoms after at least 6 weeks of conservative treatment. One milliliter of solution containing either 1.25U of condoliase or saline was injected into the intervertebral disc via posterolateral approach. The primary endpoint was the changes of mean visual analog scale (VAS) score of the worst leg pain over the past 24 h from the baseline to week 13. Magnetic resonance image (MRI) and X-ray were evaluated at 6 weeks, 3, 6, 9 and 12 months after the injection and were compared to the pre-administration imaging data. A total of 163 subjects received either condoliase (n = 82) or placebo injection (n = 81). The difference in the VAS changes in worst leg pain from baseline to Week 13 between the two groups was -15.2 mm (95 % confidence interval, -24.2 to -6.2) and the condoliase group showed significantly better (p = 0.001) improvement. Imaging analyses revealed that the decrease in the volume of the herniated mass was greater in the condoliase than that in the placebo group. Patients who showed more than 20 % decrease in the hernia volume on post-injection MRI tend to demonstrate better improvement in their leg pain. The most common adverse event (AE) was back pain, which was observed in 36.6 % in the condoliase and 33.3 % in the placebo group, respectively. Investigators rated most events as ‘‘mild’’ or ‘‘moderate’’ and those events were generally observed within 1 day after the administration. Modic Type1 change and decrease in disc height were observed more frequently in the condoliase than that in the control group. However these changes were not associated with clinical symptoms. Condoliase was well tolerated and did not cause clinically important AEs. Condoliase is considered to be a novel and potent chemonucleolytic drug and expected to provide a new treatment option for patients with LDH. Disclosures: author 1: grants/research support; Company = Seikagaku Corporation, consultant; Company = Seikagaku Corporation,

123

S688 Dentsu Sudler & Hennessey Inc., General Insurance Rating Organization, royalties;Company = Sumitomo Dainippon Pharma Co., Ltd., Japan Broadcasting Corporation, Medical View C, Ltd., other financial report; Company = Seikagaku Corporation, Astellas Pharma Inc., Showa Yakuhin Kako Co., Ltd.; author 2: grants/research support; Company = Seikagaku kogyo Company, consultant; Company = Seikagaku kogyo Company.

45 INCIDENTAL DURA LESIONS AND THE EFFECT ON OUTCOME AFTER DECOMPRESSION FOR LUMBAR SPINAL STENOSIS: RESULTS OF A MULTI-CENTER STUDY WITH 980 PATIENTS Ralph Kothe, Markus Quante, Nils Engler, Franziska Heider, Christoph Siepe Scho¨n Klinik Hamburg Eilbek, Germany Introduction: Incidental durotomy is one of the most common complications in lumbar spine surgery. The reported incidence of dura lesions in patients with lumbar spinal stenosis in the literature is between 4 and 17 %, with higher incidence associated with previous surgery and increased patient age. In contrast to this well documented data there are conflicting reports wether or not a dural lesion is a predictor of inferior outcome after lumbar decompression. The purpose of this prospective spine registry study was to analyse the effect of dura lesions in this specific group of patients. Materials and methods: 980 patients (mean age 70 years, 53 % male and 47 % female) with lumbar spinal stenosis and exclusive decompression surgery were included in this prospective study. All procedure were performed in 2012 in three different spine clinics in Germany. Outcome measures (ODI, EQ 5D, VAS back pain and leg pain) were obtained preoperatively, at discharge, 3 and 12 month after surgery. The effect of incidental durotomy on the outcome was statistically analyzed under SAS 9.3. format. Absolute and relative changes in both groups (Dura + and Dura-) were calculated and compared using the independent sample t test. Results: In 67 patients (6.8 %) an incidental intraoperative dura lesion occurred (Dura+). Both groups (Dura+ and Dura-) showed no differences regarding risk factors and co-morbidities (previous surgery, age, BMI, Charlson-Index, ASA). The follow-up for the outcome measures was 71 % after three months and 76 % after 1 year. Patients without dura lesions (Dura-) showed a significant better improvement for the VAS back pain after 12 months (D21.4 versus D7.2 points; p \ 0.05). The differences for the remaining outcome measures were not significant (improvement ODI 16.1 versus 12.6; p [ 0.05). Discussion: In the presented study we could analyse the effect of an incidental durotomy on the clinical outcome after lumbar decompression on the basis of a homogenous patient group in a defined period of time. In contrast to the current opinion, that a dura lesions does not affect the outcome after lumbar surgery, we found a significant better improvement of the VAS back pain in the patient group without durotomy (Dura-). A follow-up for the outcome measures of 76 % after 1 year is an excellent result for a spine registry, so that our findings should be judged as valid. Disclosures: author 1: consultant; Company = Stryker, Aesculap, Silony; author 2: consultant; Company = Nuvasive, Silony; author 3: employee; Company = Scho¨n Klinik Hamburg Eilbek, Clinic for spinal surgery; author 4: none; author 5: consultant; Company = DePuy Synthes, Silony Medical.

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 46 DOES DAILY TOBACCO SMOKING AFFECT OUTCOMES AFTER MICRODECOMPRESSION FOR DEGENERATIVE CENTRAL LUMBAR SPINAL STENOSIS? A MULTICENTER OBSERVATIONAL REGISTRY-BASED STUDY Charalampis Giannadakis, Sasha Gulati, Ulf Nerland, Ole Solheim Department of Neurosurgery, St. Olavs University Hospital, Trondheim, Norway Objective: The aim of this multicenter observational study was to examine the relationship between daily smoking and patient-reported outcome at 1 year using the Oswestry Disability Index (ODI) after microdecompression for single and two-level central lumbar spinal stenosis (LSS). Secondary outcomes were length of hospital stays, perioperative and postoperative complications. Materials and methods: Data were collected through the Norwegian Registry for Spine Surgery. Results: 825 patients were included (619 nonsmokers and 206 smokers). For the whole patient population there was a significant difference between preoperative ODI and ODI at 1 year (17.3 points, 95 % CI 15.93–18.67, p \ 0.001). There was a significant difference in ODI change at 1 year between nonsmokers and smokers (4.2 points, 95 % CI 0.98–7.34, p = 0.010). At 1 year 73.7 % of nonsmokers had achieved a minimal clinically important difference (C8 points ODI improvement), compared to 62.7 % of smokers (p = 0.008). There was no difference between nonsmokers and smokers in the overall complication rate (11.6 vs 9.2 %, p = 0.34). There was no difference between nonsmokers and smokers in length of hospital stays for either single-level (2.3 vs 2.2 days, p = 0.99) or two-level (3.1 vs 2.3 days, p = 0.175) microdecompression. Smoking was identified as a negative predictor for ODI change in a multiple regression analysis (p = 0.001). Conclusion: Nonsmokers experienced a significantly larger improvement at 1 year following microdecompression for LSS compared to smokers. Smokers were less likely to achieve a minimal clinically important difference. However, it should be emphasized that considerable improvement also was found among smokers. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

47 COMPARISON OF DIBOTERMIN ALFA WITH AUTOGRAFT FOR SINGLE-LEVEL LUMBAR INTERBODY ARTHRODESIS USING POOLED INDIVIDUAL PATIENT DATA AND META-ANALYSIS Kenneth Burkus, David Donnell, Koen van der Heijden, Fujian Song The Hughston Clinic; Medtronic BioPharma B.V.; UEA Consulting LimitedLimited, Columbus, GA, USA Reviews of the efficacy and safety of dibotermin alfa may be confounded by heterogeneity of study designs. Considering only the use of dibotermin alfa/absorbable collagen implantation matrix (ACIM) as licensed in Europe, this efficacy meta-analysis used individual patient data (IPD, n = 536) from 6 company-sponsored randomised controlled trials (RCTs). Pooled safety data (n = 700) included additional IPD from 2 company-sponsored single-arm trials. Trials used various lumbar interbody devices and surgical approaches. At 24 months, the fusion success rate with dibotermin alfa/ACIM (94.5 %, 241/255) was significantly higher than with iliac crest

Eur Spine J (2015) 24 (Suppl 6):S669–S710 autograft (84.7 %, 177/209): odds ratio 3.26 (95 % CI 1.172, 9.075; P = 0.024). The estimated absolute difference in fusion success rate was 11.7 % (95 % CI 8.0, 22.5 %; P = 0.035), with number needed to treat corresponding to the estimated risk difference of 8.5 (95 % CI 4.4, 125). For patient-reported outcomes, the mean difference (95 % CI) in 24-month change from baseline for dibotermin alfa/ACIM versus autograft was -4.252 (-8.727, 0.223; P = 0.063) for Oswestry Disability Index, -2.267 (-4.295, -0.239; P = 0.028) for back pain, and 2.649 (0.316, 4.982; P = 0.026) for Short Form 36 physical component summary. In the pooled safety data no new safety concerns were identified. Pseudarthrosis was the most frequent individual treatment-related adverse reaction, with a frequency that was approximately 2-fold lower for dibotermin alfa/ACIM (4.8 %, 22/456) compared with autograft (12.7 %, 31/244). The proportion of patients who experienced both pseudarthrosis adverse events and failure-related secondary surgery was significantly lower at 24 months for dibotermin alfa (3.73 %, 17/456) compared with autograft (7.79 %, 19/244) (relative risk 0.4788, 95 % CI 0.2506, 0.9147, P = 0.026, n = 700). There was a potential causal link between dibotermin alfa/ACIM and radiculitis/radiculopathy, which occurred commonly with a frequency of 6.6 %. The interaction of smoking on the difference in fusion success rate between dibotermin alfa/ACIM and autograft at 24 months was not significant (P = 0.096, 8 clinical trials). The data provide convincing evidence of a favourable benefit/risk balance for dibotermin alfa/ACIM as a substitute for autograft in single-level lumbar interbody fusion using CE-marked interbody devices or allograft spacers in adults with degenerative disc disease who have had at least 6 months of nonoperative treatment. Disclosures: author 1: grants/research support; Company = Medtronic,consultant;Company = Medtronic; author 2: employee; Company = Medtronic BioPharma B.V.; author 3: employee; Company = Medtronic BioPharma; author 4: grants/research support; Company = Medtronic Biopharma B.V., consultant; Company = Medtronic BioPharma B.V.

48 DOES LORDOTIC ANGLE OF CAGE DETERMINE LUMBAR LORDOSIS IN POSTERIOR LUMBAR INTERBODY FUSION? Young-Tae Kim, Kyu-Jung Cho Inha University Hospital, Incheon, Korea Introduction: To compare the radiological results of posterior lumbar interbody fusion (PLIF) using 4° cages, 8° cages, and 15° cages in degenerative lumbar spine diseases. Materials and methods: We evaluated 177 patients after PLIF and pedicle screw instrumentation at one or two levels with at least 1 year follow-up. PLIF using 4° lordotic angle cages (4° group) was performed in 65 patients, 8° cages (8° group) in 49 patients, and 15° cages (15° group) in 63 patients. Lumbar lordosis angles, segmental lordosis angles, disc height, and bony union rate were evaluated in the radiographs. Results: The lumbar lordosis was 36.7° before surgery, corrected to 42.4° after surgery, and changed to 34.2° at the final follow-up in the 4° group. In the 8° group, it was 33.3° preoperatively, improved to 39.7° postoperatively and decreased to 35.1° at the final follow-up. In the 15° group, it was 31.1° preoperatively, improved to 42.9° postoperatively and decreased to 36.4° at the final follow-up. These changes showed statistical significances (p \ 0.001). The segmental lordosis was 6.9° before surgery, 9.5° after surgery, and 6.2° at the

S689 final follow-up in the 4° group. It was 6.7°, 9.8°, and 8.1° in the 8° group, and 5.9°, 10.3°, and 9.4° in the 15° group respectively (p \ 0.001). The disc height was 10.9 mm-13.9 mm-11.7 mm preoppostop-at the final follow-up in 4° group, 9.2 mm-12.3 mm-9.8 mm in the 8° group, and 7.3 mm-15.3 mm-11.4 mm in the 15° cage group respectively (p \ 0.001). The bony union was achieved in 90.7 % of patients in the 4° group, 93.9 % in the 8° group, and 93.7 % in the 15° group without significant differences (p = 0.087). Conclusion: The lordotic angle of cage has determined the degree of restoration of lumbar lordosis after instrumented posterior lumbar interbody fusion. Cages with sufficient lordotic angle showed better restoration of lumbar lordosis as well as preventing loss of correction. Disclosures: author 1: none; author 2: none.

49 PLIF SURGERY WITH EITHER PEEK OR TITANIUMCOATED PEEK-CAGES. A RANDOMISED CLINICAL AND RADIOLOGICAL TRIAL Klaus Schnake, Nikolai Fleiter, Andreas Pingel, Christoph Hoffmann, Matti Scholz, Frank Kandziora Center for Spinal Therapy, Scho¨n Klinik Nu¨rnberg Fu¨rth, Germany Introduction: PLIF surgery commonly involves usage of PEEK or titanium cages for interbody fusion. Common fusion rates range from 77 to 100 % for both materials. However, comparative trials examining the bony incorporation of these materials for PLIF surgery have not been published yet. The aim of this study was to compare fusion rates and clinical results of a pure PEEK cage with a titanium-coated PEEK cage. Material/methods: We randomised 60 consecutive patients with lumbar degenerative disease into group A (titanium-coated PEEK cage) or group B (non-coated PEEK cage). Patients with osteoporosis or systemic diseases were excluded from the study. PLIF surgery included pedicle screw fixation and 2 cages per level. Three independent examiners evaluated plain X-ray and thin-sliced CT scans using various scores. Clinical results were evaluated using the Oswestry and Visual Analogue Scores. Follow-up time was 12 months. Results: 55 patients (92 %) (36 female, 19 male) with an average age of 51 years (31–70) had a complete follow-up. Demographic data in both groups did not differ statistically. Ten patients were treated with two-level fusion. The following segments were operated: L2/3 (2 times), L3/4 (5 times), L4/5 (29 times) and L5/S1 (29 times) (group A: 29 segments, group B: 36 segments). Perioperative and intraoperative data did not differ significantly between the two groups. Both groups equally took benefit from surgery: Oswestry-score preoperatively (A: 43, B: 45), postoperatively (A: 20, B: 20); VAS low back pain preoperatively (A: 6.1, B: 5.2) and postoperatively (A: 2.6, B: 2.6). Four (6.2 %) intraoperative complications occurred (dural tears). Titanium-coated cages were indistinguishable from PEEK cages in radiographic studies. At 12 months follow-up there was one cage sintering visible in each group. No radiolucency or dislocation could be detected. Fusion was present in CT scans as follows: a) bone growth through cage pores (A: 100 %, B: 100 %); b) bone growth outside the cages (A: 48 %, B: 61 %). Fusion rate was 100 % in both groups. No statistically significant difference was noted for the clinical and radiological results of the two groups. Conclusion: Pure PEEK and titanium-coated PEEK cages for PLIF produce equally favourable clinical and radiological results 12 months post surgery. The fusion rate was 100 % in both groups. Disclosures: author 1: grants/research support; Company = Medtronic, consultant; Company = OttoBock, Expanding Orthopedics;

123

S690 author 2: grants/research support; Company = Medtronic Deutschland GmbH, Earl-Bakken-Platz 1, 40670 Meerbusch; author 3: none; author 4: grants/research support; Company = Medtronic; author 5: grants/research support; Company = Medtronic, consultant; Company = Depuy Synthes, royalties; Company = Globus Medical, employee; Company = BGU Frankfurt am Main; author 6: grants/ research support; Company = AMT, Medtronic.

50 LEG PAIN IS A POSSIBLE SAGITTAL MALALIGNMENT CLINICAL PRESENTATION IN ADULT SPINAL DEFORMITY SURGERY Mitsuru TAKEMOTO, Louis Boissiere, Ibrahim Obeid, Jean-Marc Vital, Ferran Pellise´, Francisco Javier Sanchez Perez-Grueso, Frank Kleinstuck, Emre R. Acaroglu, Ahmet Alanay, ESSG European Spine Study Group

Eur Spine J (2015) 24 (Suppl 6):S669–S710 DePuy Synthes, K2M, consultant; Company = DePuy Synthes, Biomet; author 6: grants/research support; Company = DePuy Synthes, consultant; Company = De Puy Synthes; author 7: none; author 8: grants/research support; Company = Medtronic, Depuy Synthes, stock/shareholder; Company = IncredX; author 9: grants/research support; Company = Depuy Synthes, consultant; Company = Stryker; author 10: grants/research support; Company = DePuy Synthes spine.

51 RISK FACTORS ASSOCIATED WITH MECHANICAL COMPLICATIONS INCLUDING PROXIMAL JUNCTIONAL FAILURE AFTER OSTEOTOMY FOR ADULT SPINAL DEFORMITY Evalina Burger, Cameron Barton, Andriy Noshchenko, Vikas Patel, Christopher Cain, Christopher Kleck

Spine Unit 1, Bordeaux University Hospital, Bordeaux, France

University of Colorado, Anschutz Medical Campus, Aurora, USA

Introduction: Leg pain is one of the most common adult spine deformity (ASD) patient complaints. During the surgical procedure, adequate decompression provides pain relief. However, some patients suffer from unexpected residual leg pain after surgery. The aim of this study is to clarify the clinical presentation of leg pain after ASD surgery and discuss the possible origin. Methods: This study is a retrospective review of a multicenter prospective database of patients surgically treated for ASD. The inclusion criteria were age C50 years with one of the following preoperative radiographic parameters (coronal curvature C20°, SVA C5 cm, PT C25° or thoracic kyphosis C60°). Revision cases were excluded. Preoperative (Preop) and six-month postoperative (Postop) radiographic parameters, HRQL scores and numerical rating scale (NRS) of leg and back pain were analyzed. Results: Eighty-two patients (69 female and 13 male, average age 66.3 years, 50–85) were included. The average fusion length was 10.4 vertebra (2–18). They included 9 cases of simple posterior fusion, 43 decompression procedures, 37 TLIF/PLIFs, 36 SPOs, 6 PSOs, and 1 VCR. Sixty-nine patients presented a Preop leg pain (NRS C2). Rate of decompression procedure (58.0 vs. 20.0 %, p \ 0.05) and coronal Cobb angle (35.0° vs. 47.9°, p \ 0.05) were significantly different between the Preop leg and no leg pain groups. At 6 months after surgery, leg pain was significantly improved in both groups and no difference was found between them (NRS 1.8 vs. 2.1, p = 0.83). No correlation was found between Postop leg pain and Preop leg pain, fusion length, decompression and other open canal procedures (e.g. TLIF/PLIF, SPO, PSO and VCR). These results indicate that Preop leg pain was adequately treated with decompression procedure. However, 18 patients (22 %) still presented a Postop leg pain (NRS C5). For the Postop leg pain (NRS C5) and no leg pain (NRS B4) groups, SVA was 64 mm and 23 mm (p = 0.002), Global Tilt was 34.8° and 27.4° (p = 0.04), and Pelvic Tilt was 31° and 26° (p = 0.09), respectively. Discussion and conclusion: A review of a prospective multicenter database revealed a positive correlation between postoperative leg pain and sagittal malalignment after ASD surgery. These patients received a proper decompression and fusion surgery and the residual leg pain could be a sagittal malalignment clinical presentation rather than a radicular pain. Disclosures: author 1: none; author 2: none; author 3: grants/research support; Company = Depuy Synthes, consultant; Company = Depuy Synthes Medtronic; author 4: consultant; Company = LDR, DepuySynthes, and Euros; author 5: grants/research support; Company =

Introduction: Osteotomies including pedicle subtraction (PSO) and Smith-Peterson (SPO) are widely used to facilitate correction in the treatment of adult spinal deformity (ASD), but are associated with mechanical complications (MC). Purpose: To determine incidence and risk factors for mechanical complications after osteotomy for adult spinal deformity. Study design: Retrospective review of adult deformity database (COMIRB #14-1258). Methods: Eighty-three consecutive ASD patients (55 female, avg. age at initial operation 58) treated with 90 separate osteotomy operations met strict inclusion/exclusion criteria and follow-up of at least 1 year or incidence of MC. Data was extracted for each operation including patient variables, surgical variables, instrumentation variables, and postoperative variables. Operations were divided into groups based on MC (e.g. proximal junctional failure (PJF) vs. nonPJF) and odds ratios were calculated to assess risk factors for specific MCs. Results: Incidence of MC occurred in 32/83 (39 %) of patients following 37/90 (41 %) of osteotomy operations and trended higher in PSO 49 % (24/49) vs. SPO 32 % (13/41) (p = .13). Number of operations with specific MCs: 13 with loose screws, 11 PJFs, 10 pseudarthrosis, 7 subsequent adjacent decompressions, 5 broken rods, 5 compression fractures, 4 isolated pedicle fractures, 4 uncoupling of instrumentation, 1 sacral fx, 1 screw fx, and 1 osteotomy displacement. Risk factor meeting statistical significance for entire MC group was postop lumbar lordosis (LL) \49° (OR = 2.53, p = .04). Statistically significant risk factors for MCs: Loose screws risk factor of fusion crossing lumbosacral junction (OR = 7.66, p = .05); PJF risk factors were ever smoker (OR = 8.76, p = .04) and presence of 2 or more cross-links (OR = 3.87, p = .04); subsequent decompression surgery risk factor was fusion stopping above sacrum (OR = 13.92, p = .02). Conclusions: Current study shows incidence of mechanical complications occurred in 39 % of patients following osteotomy and trends higher in PSO vs. SPO. Postop LL \49° was found to be a risk factor for all mechanical complications. Fusion across the lumbosacral junction was found to be a risk factor for loose screws. Ever smoking and presence of 2 or more cross-links were found to be risk factors for PJF. Fusion stopping above the sacrum was found to be a risk factor for subsequent decompression surgery. Disclosures: author 1: grants/research support; Company = Aesculpa, consultant; Company = Medicrea, Paradigm, Signus; author 2: grants/research support; Company = Medicrea; author 3: employee;

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Company = Colorado University Denver, USA; author 4: grants/research support; Company = Medtronic, Synthes, SI-Bone, MTF, OMeG, Aesculap, vertiflex, consultant; Company = Aesculap, Allosource, royalties; Company = Springer, SLACK, Aesculap, Biomet; author 5: grants/research support; Company = Medicrea, Medtronic, SI Bone, Vertiflex, Aesculap, consultant; Company = DePuy Synthes, royalties; Company = DePuy Synthes; author 6: grants/research support; Company = Medtronic Sofamor-Danek, Integra, Orthofix, Aesculap, Medicrea, Vertiflex.

THORACO-LUMBAR SPINE, DEGENERATIVE III

52 COMPARISON OF AN ANTI-INFLAMMATORY DRUG DELIVERY SYSTEM VERSUS MSC THERAPY TO CONTROL INFLAMMATION IN A PRO-INFLAMMATORY INTERVERTEBRAL DISC ORGAN CULTURE SYSTEM Graciosa P.Q. Teixeira, Catarina L. Pereira, Karin Benz, Hans-Joachim Wilke, Anita Ignatius, Mario A. Barbosa, Raquel Goncalves, Cornelia Neidlinger-Wilke Institute of Orthopaedic Research and Biomechanics, Center for Musculoskeletal Research, University of Ulm, Germany, Institute of Biomedical Engineering (INEB), Universidade do Porto, Portugal, Instituto de Cieˆncias Biome´dicas Abel Salazar (ICBAS), Universidade do Porto, Portugal, NMI Natural and Medical Sciences Institute at the University of Tuebingen, Reutlingen, Germany Introduction: Disc degeneration is often associated with inflammation. Resolution of inflammation is a pre-requisite for regeneration and could possibly be achieved by pharmacological and/or cell-based strategies. We have established a pro-inflammatory disc organ culture model in which two intradiscal therapeutic approaches were tested: (a) a biomaterial-based strategy using Chitosan/Poly-(c-glutamic acid) nanoparticles with diclofenac, a non-steroidal anti-inflammatory and analgesic drug (Df-NPs) and (b) a cell-based therapy using MSCs. Methods: Bovine caudal discs (4–10/condition) were needle-punctured and stimulated with IL-1b (10 or 100 ng/mL). Discs were treated by injection of (a) Df-NPs or (b) human MSCs embedded in albumin/hyaluronan hydrogel (MSCs-Alb), 3 h after stimulation. Untreated discs were used as controls. Gene expression of pro-inflammatory factors (TNF-a, IL-6, IL-8), matrix metalloproteinases (MMPs 1 and 3), and ECM proteins (Coll II, Aggrecan) was analyzed by qRT-PCR 48 h after treatment. Prostaglandin E2 (PGE2) was quantified in medium supernatants by ELISA. Cell viability was analyzed by LIVE/DEAD assay and metabolic profile was traced by Glucose/Lactic acid monitoring. Mann–Whitney and Kruskal–Wallis statistical tests (p B 0.05) were used to compare two or several groups, respectively. Results: Df-NPs injection induced significant down-regulation of IL-6 and MMP1 expression, and up-regulation of ColI II and Aggrecan compared to IL-1b-stimulated groups, as well as decreased PGE2 production (p \ 0.05). MSCs-treated groups showed slight down-regulation of IL-6 and -8 expression compared to IL-1b stimulus alone. PGE2 results revealed variability between MSCs donors. In all test conditions, cells remained viable and showed similar metabolism. Discussion: Df-NPs were able to significantly reduce the pro-inflammatory response induced by IL-1b treatment and may have the potential to delay/decrease ECM degradation. Current long-term

S691 experiments will show if increase of ColI II and Aggrecan can be confirmed at protein level. MSCs short-term treatment had no conclusive influence on the disc inflammatory response, since the results showed high donor variability. Overall, this pro-inflammatory model is a suitable model for in vitro testing of anti-inflammatory strategies for disc degeneration. Acknowledgement: Supported by PEST founds, FCT, AO Foundation, Luso-German Integrated Actions 2012: CRUP/DAAD and by the German Spine Foundation. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: no indication; author 8: none.

53 DOES A PREOPERATIVE COGNITIVE-BEHAVIOURAL INTERVENTION AFFECT DISABILITY, PAIN BEHAVIOUR, PAIN AND RETURN TO WORK AFTER LUMBAR SPINAL FUSION SURGERY? Nanna Rolving, Claus Vinther Nielsen, Finn Bj arke Christensen, Randi Holm, Cody Erik Bu¨nger, Lisa Gregersen Oestergaard Aarhus University Hospital, Aarhus, Denmark Background: In the last decades the Western World has seen a significant increase in lumbar spinal fusion (LSF) rates due to degenerative spinal disorders. Few published studies have looked at the potential of rehabilitation to improve the surgical outcome, but rehabilitation programmes using cognitive-behavioural therapy (CBT) are recommended. Further, initiating interventions preoperatively have been found beneficial in terms of reducing bed days and hospital costs. Only limited data exists in the field of LSF regarding the use of interventions using a CBT approach, initiated already prior to surgery. Objective: To examine the effect of a preoperative CBT intervention for patients undergoing LSF. Methods: In a randomized controlled study design 90 patients with degenerative disc disease or spondylolisthesis undergoing LSF were allocated to either usual care (control group) or preoperative CBT and usual care (CBT group). Primary outcome was change in Oswestry Disability Index (ODI) from baseline to 1-year follow-up. Secondary outcomes were catastrophizing, fear-avoidance belief, work status and back and leg pain. Results: The CBT group had achieved a significant ODI reduction of -15 points (-26; -4) already at 3 months (between group difference P = 0.003) and this reduction was maintained throughout the year. At 1-year follow-up the difference was no longer statistically significant due to improvements in the control group (P = 0.082). There were no differences between groups at 1-year follow-up with regards to any of the secondary outcomes. Conclusions: Participating in a preoperative CBT intervention in addition to usual care resulted in a faster recovery of function after LSF surgery. However, the reduced disability seen in the CBT group already after 3 months did not translate into a faster return to work. Our findings additionally support the need for further research into the use of targeted rehabilitation interventions among patients with elevated levels of catastrophizing and fear avoidance beliefs. Disclosures: author 1: grants/research support; Company = Danish Reumatism Association; author 2: none; author 3: none; author 4: none; author 5: none; author 6: grants/research support; Company = The Danish Council for Strategic Research.

123

S692 54 PREOPERATIVE COGNITIVE-BEHAVIOURAL PATIENT EDUCATION VERSUS STANDARD CARE AFTER LUMBAR SPINAL FUSION: ECONOMIC EVALUATION ALONGSIDE A RANDOMIZED CONTROLLED TRIAL Nanna Rolving, Rikke Soegaard, Claus Vinther Nielsen, Finn Bjarke Christensen, Cody Erik Bu¨nger, Lisa Gregersen Oestergaard Aarhus University Hospital, Aarhus, Denmark Background: The clinical effectiveness of a preoperative cognitivebehavioural therapy (CBT) intervention for patients undergoing lumbar spinal fusion surgery (LSF) has been investigated in a randomised control trial. Economic evaluation is however essential for decision makers to make informed choices regarding allocation of scarce resources. Objective: To assess the cost-effectiveness of a preoperative CBT intervention compared to usual care for patients undergoing LSF. Methods: A cost–utility and cost-effectiveness analysis were conducted alongside the clinical trial. 90 patients undergoing LSF due to a primary diagnosis of degenerative disc disease, spinal stenosis or spondylolisthesis were randomly allocated to usual care (control group) or usual care plus a preoperative CBT intervention (CBT group). Outcome parameters included quality-adjusted life years (QALY), based on the EQ-5D, and pain-related disability, based on the Oswestry Disability Index (ODI). Health care use and productivity loss were estimated from national registries. The time frame was 1 year from date of surgery. Results: The estimated QALY was significantly better for the CBT group (0.710 [95 % CI 0.670; 0.749) versus 0.636 (95 % CI 0.5573; 0.687)]. For the ODI, the CBT group reported significantly larger disability reductions at 3 and 6 months, but this difference had evened out at the 1-year follow-up (between-group difference of 5 points (95 % CI -1.7; 11.6) in favour of CBT group).There was no difference in the overall costs of the two groups [-€89 (95 % CI -12,080; 11,902)] despite the extra costs of the CBT group related to the CBT intervention. The probability of the CBT intervention being costeffective at a willingness-to-pay of €40,000 for 1 additional QALY was 70 %. For an additional gain of 15 ODI points this probability was 90 % at thresholds as low as €10,000. These results remained largely unaffected by relevant sensitivity analyses, confirming the robustness of findings. Conclusions: Preoperative CBT appears to be more effective and cost neutral when considering the overall health care sector and labour market perspective. Disclosures: author 1: grants/research support; Company = Danish Reumatism Association; author 2: none; author 3: none; author 4: none; author 5: grants/research support; Company = Danish strategic research Council; author 6: grants/research support; Company = The Danish Council for Strategic Research.

55 QUALITY OF LIFE AND DISABILITY: CAN THEY BE IMPROVED BY ACTIVE POSTOPERATIVE REHABILITATION AFTER SPINAL FUSION SURGERY? A RANDOMISED CONTROLLED TRIAL WITH 12-MONTH FOLLOW-UP

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Department of Orthopaedics and Trauma, Tampere University Hospital, Tampere, Finland; Department of Health Sciences, University of Jyvaskyla, Jyvaskyla, Finland; Department of Physical Medicine and Rehabilitation, Central Finland Health Care District, Jyvaskyla, Finland; Department of Rehabilitation Medicine and Department of Psychiatry, VU University Medical Center, Amsterdam, the Netherlands; Department of Orthopaedics and Trauma, Central Finland Health Care District, Jyvaskyla, Finland Study design: Randomised controlled trial (RCT). Background: Although lumbar spine fusion surgery (LSF) in patients with isthmic or degenerative spondylolisthesis reduces disability and improves health-related quality of life, there is still a gap between these patients and the normal population at the 1-year follow-up. To restore functional abilities, the effectiveness of postoperative rehabilitation and exercise interventions needs to be evaluated. Objective: To study the effectiveness of the postoperative 12-month exercise program compared to usual care on disability and healthrelated quality of life (HRQoL) and the maintenance of results 1-year post-intervention. Methods: Altogether 98 patients (mean age 59 years, 75 % females) with isthmic (31) or degenerative (67) spondylolisthesis were randomized either to exercise therapy group (EG)(n = 48) or control group (CG) (n = 50) 3 months after LSF. EG patients had progressive combined home-based strength and aerobic training program for 12 months. They were seen and encouraged every 2 months by the physiotherapist. CG patients received only oral and written instructions of exercises given during one session by physiotherapist. All patients fulfilled the questionnaires at the time of randomization (3 months postoperatively) at the end of the intervention (15 months postoperatively) and 1 year after intervention. The primary outcome measures were the Oswestry Disability Index (ODI) and the RAND36 questionnaire. Results: The mean ODI score decreased from 24 (12) to 18 (14) in the EG (p = 0.003) and from 18 (12) to 13 (11) in the CG (p = 0.012) during the intervention (between groups p = 0.69). At 1-year follow-up, 25 % of the EG and 28 % of the CG had an ODI score C20. No between group differences in HRQoL change scores were found at any time point. The mean (95 % CI) Physical Functioning dimension of the HRQoL improved by 10.0 (4.6–15.3) in the EG and by 7.8 (2.5–13.0) in the CG. In addition, the Role Physical score improved by 20.0 (7.7–32.3) in the EG and by 16.4 (4.4–28.4) in the CG during the intervention. Conclusions: Disability and functional performance -related dimensions of HRQoL improved in both groups, and thus, the 12-month home-based exercise intervention did not have an impact on these outcomes beyond the improvement achieved by usual care. However, a considerable proportion of participants still had at least moderate disability indicating need for more individual rehabilitation. Disclosures: author 1: other financial report; Company = Medtronic, Depuys Synthes; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none; author 7: none; author 8: none; author 9: none.

56 INDIVIDUAL COURSES OF LOW BACK PAIN IN ADULT DANES—A COHORT STUDY WITH 4- AND 8-YEAR FOLLOW-UPS Tom Bendix, Lars Korsholm, Lise Hestbaek, Per Kjaer

Marko Neva, Outi Ilves, Joost Dekker, Liisa Pekkanen, Ilkka Marttinen, Kimmo Vihtonen, Kirsi Piitulainen, Salme Jarvenpaa, Arja Hakkinen

123

Rigshospitalet, Glostrup Hospital, University of Copenhagen, Denmark

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Background: Previous studies have reported the prevalence rates of low back pain (LBP) in many different cohorts, countries and age groups. However, fewer longitudinal studies have described the variation of LBP over time and at individual levels. The aim of this study was to describe individual courses of various definitions of selfreported LBP over an eight-year period in a middle-aged Danish population. Methods: A cohort of 625 men and women aged 40 was sampled from a general population. Questions about LBP were asked at ages 41, 45, and 49. Five definitions of LBP were investigated: LBP during the past ‘year; LBP during the past ‘month; ‘[30 days’ past year; ‘seeking care’ for LBP; and ‘non-trivial’ LBP ([30 days past year with either seeking care or other activity restrictions). Prevalence estimates at the three times were reported, individual patterns of reporting LBP were mapped and their associated risks of future LBP were estimated as odds ratios (OR). Results: Questionnaires were completed by 412 (66 %), 348 (56 %), and 293 (47 %) at the three surveys, respectively. Their LBP prevalences did not change significantly over time: ‘year’: 69, 68, 70 %; ‘month’: 42, 48, 41 %; ‘[30 days’: 25, 27, 24 %; ‘seeking care’: 28, 30, 36 %; and ‘non-trivial’: 18, 20, 20 %. The 293 people who participated at all three times were included in the longitudinal analysis. More than 50 % reported LBP within the same definitions after 4 years, whereas between 27 and 54 % reported in a fluctuating pattern over 8 years. Reporting any definition of LBP significantly increased the odds of the same type of LBP 4 years later and most markedly so for ‘[30 days’ and ‘non-trivial’ LBP (OR: 5.9/8.8). Reporting the same definition of LBP at both ages 41 and 45 significantly increased the risk of also reporting the same at 49 for all definitions, but most strongly for ‘seeking care’ and ‘non-trivial’ LBP (OR 17.6/18.4). Conclusion: Total prevalence rates of different definitions of LBP were constant over time, but many individuals reported in varying patterns. Two thirds report LBP ‘year’ but only about one fifth seems to have serious LBP with consequences. General and less specific definitions month/year carry the lowest risk of future LBP. LBP in more severe categories indicated a higher risk of also reporting future LBP within the same category. Particularly the LBP definition ‘nontrivial’ seems to represent the most consistent sub-group with regards to reporting any later pain. Disclosures: author 1: none; author 2: no indication; author 3: none; author 4: grants/research support; Company = Industrial Insurance (baseline study only), none.

57 ADDRESSING LUMBAR DEGENERATIVE DISC DISEASE, A PROSPECTIVE 5-YEAR-FOLLOW-UP STUDY WITH AN UNCONSTRAINED TDR Joel Delecrin, Jacques Beaurain, Jerome Allain, Herve Chataigner, Thierry Dufour, Jean Huppert, Alexandre Poignard, Marc Ameil, Jean-Paul Steib Service de Chirurgie Orthope´dique, 1 place Alexis Ricordeau, Nantes, France Introduction: Lumbar TDR is an emerging technique for DDD treatment. A device with a mobile core could allow pain relief, with however preserving segmental mobility and function of the intervertebral joint. This study aims to clinically and radiologically assess a lumbar TDR device with controlled mobility at 5 years follow-up. Methods: A total of 411 consecutive patients (455 prostheses) were enrolled in a 10-year ongoing prospective study involving 8 centers.

S693 Data were analyzed preoperatively and 1.5, 3, 6, 12, 24, 36, 60 months postoperatively. Clinical outcomes included ODI, VAS (Back or leg pain), SF-36 (PCS and MCS) scores, patient satisfaction, medication consumption, complication and reoperation rates. Radiological assessment explored implanted segments regarding ROM, lordosis, HO occurrence (MacAfee classification); and adjacent segments regarding ROM, HO and disc degeneration (Lane classification). Results: 31.9 % of the patients were women and 68.1 % were men; 89.5 % were treated at a single level and 10.4 % were multi-level implanted. Clinical outcomes showed a significant improvement for ODI, VAS and SF-36 at 5 years compared to preop baseline. ODI decreased from 49.3 % (preop) to 20.1 % at 5 years FU, back VAS from 6.3 cm to 2.5 cm, leg VAS from 5.6 cm to 2.8 cm; SF-36 (PCS) increased from 33.1 to 44.2 and SF-36 (MCS) from 24.8 to 33.6. Medication consumption was significantly reduced at 5 years FU: 37 % patients were under medication vs 93 % preop. Patients overall satisfaction was 90 % 5 years after surgery. ROM and lordosis were significantly improved at 5 years postop. Low ossification occurrence in operated discs was noted, 1/191 disc was grade IV (fusion). Regarding adjacent discs, ROM was significantly increased in both superior disc and inferior disc compared to preop. Ossification was minor in adjacent discs with 100 % discs grade 0-I for n + 1 discs and 100 % grade 0-II for n - 1 discs at 5 years FU. Occurrence of disc degeneration was reported for 20.5 % of n + 1 disc and 9.1 % of n - 1 discs at 5 years FU compared to baseline. The overall reoperation rate (whatever the cause, all lumbar segments included) was 14.4 % (59/411 patients) at 5 years FU. Surgery for removal and revision of the device involved 5/455 implanted prostheses (1.1 %). Conclusion: This 5-year clinical and radiological outcomes showed an overall improvement in comparison to baseline, with acceptable complication and reoperation rates following lumbar TDR. Disclosures: author 1: royalties; Company = LDR Medical; author 2: consultant; Company = LDR Medical, royalties; Company = LDR Medical; author 3: consultant; Company = Medtronic, royalties; Company = LDR; author 4: royalties; Company = LDR, Euros, Medicrea; author 5: royalties; Company = LDR Medical; author 6: stock/shareholder; Company = LDRH, royalties; Company = LDRH; author 7: none; author 8: royalties; Company = LDR; author 9: grants/research support; Company = Clariance, LDR, Medtronic, Integra, Stryker, consultant; Company = Clariance, royalties; Company = Clariance, LDR, Medtronic.

58 FINAL RESULTS OF AN INTERNATIONAL MULTICENTER, PROSPECTIVE, OBSERVATIONAL TRIAL ON LUMBAR ARTHROPLASTY FOR SYMPTOMATIC DISC DISEASE Richard Assaker, Karsten Ritter-Lang, Dominique Vardon, Stephane Litrico, Stephane Fuentes, Michael Putzier, Joerg Franke, Peter Jarzem, Pierre Guigui, Gerard Nakach, Jean Charles Le Huec CHRU de Lille, France Objective: The aim of this study was to show the effectiveness of total lumbar disc replacement (TLDR) using a prosthesis for the treatment of discogenic back pain (DBP). We present the final results (24 months) of the first international multicenter prospective observational real-world study on TDLR (NCT01338493). Patients and methods: 11 international centers implanted 134 patients (pts) with a TLDR to treat back pain with or without radiculopathy linked to a degenerative disc disease (DDD). Available

123

S694 for follow up: 113 patients at 12 months (mo) and 104 at 24 mo. Patient demographics, surgical parameters, Oswestry Disability Index (ODI), Visual Analog Scale (VAS) scores for back and leg pain, QoL Short-Form Health Survey (SF-36) and adverse events (AEs) were assessed according to center’s routine practice, follow up 6, 12 and 24 mo. Results: Patients mean age 43 years, 52 % female, mean BMI 25. DBP presented for median time of 15 mo, 31 % of the patients reported previous lumbar surgeries, 63 % of which at the implanted level. 92 % of the patients were implanted at 1-level = 1L (46 % L5– S1); 7 % at 2-levels = 2L (5 % L4–L5 + L5–S1), 1 % at 3 levels = 3L (L3–S1). Operative time (in min): 101 at 1L, 149 at 1L + fusion hybrid procedure (1L), 90 at 2L and 98 at 3L. Blood loss in mL: 150 (1L), 273 (1L, 270 (2L) and 510 (3L). Patient outcomes improved significantly (p \ 0.0001) between baseline, 6, 12 and 24 mo: ODI score 50.1 to 24.2, 21.3 and 20.3 points; VAS score for back pain 7.0 to 3.0, 2.9 and 2.8; VAS leg pain from 5.2 to 2.3, 2.4 and 2.8, respectively. SF-36 PCS survey 30.6 to 42.6, 45.0 and 45.5. AEs reported: Up to 6 mo: 46 (1 death, ischemic stroke 3 weeks postop, unrelated to implantation); from 6 to 24 mo: 38 (13 radiculopathies, 6 non-specific low back pain, 1 foreign body reaction (patient explanted after 15 mo) and 18 others). Conclusion: Our prospective, multicenter, observational study shows statistically significant improvement in physical function with reduced pain scores at 24 mo. The rate of complications in this study is comparable to those from controlled trials. This is the first study to show the effectiveness of a lumbar motion segment replacement in an international real-world clinical setting. Disclosures: author 1: consultant; Company = Medtronic, depuysynthes, royalties; Company = Medtronic; author 2: none; author 3: other financial report; Company = Medtronic; author 4: consultant; Company = LDR Spine, Medtronic, SafeOrthopedics; author 5: consultant; Company = Medtronic, Medicrea; author 6: grants/research support; Company = Medtronic; author 7: grants/research support; Company = Baxter, Medtronic, Zimmer, Relievant, consultant; Company = Medtronic, Silony, Medacta, Exp Orthopedics, royalties; Company = Ohst AG; author 8: grants/research support; Company = Medtronic; author 9: none; author 10: grants/ research support; Company = Medtronic, consultant; Company = Medtronic.

59 AN 11-YEAR MINIMUM FOLLOW-UP OF THE CHARITE III LUMBAR DISC REPLACEMENT FOR THE TREATMENT OF SYMPTOMATIC DEGENERATIVE DISC DISEASE Yong Hai, Shibao Lu, Chao Kong Orthopedic Surgery, Beijing Chaoyang Hospital, Capital Medical University of China, Beijing, China Purpose: To report our 11-year minimum clinical and radiological outcomes, as well as complications of the Charite III total disc replacement (TDR). Methods: A total of 35 patients indicated for total disc replacement were implanted with the Charite III prosthesis. Clinical evaluation included visual analog scale (VAS) for back pain and the Oswestry disability index (ODI). Radiological parameters of intervertebral disc height (IDH), range of motion (ROM), lumbar lordosis, lumbar scoliosis and prosthesis position were evaluated. Complications and reoperation rates were also assessed. Results: 32 patients had a minimum 11-year follow-up, and 33 prostheses were implanted. The mean follow-up time was 11.8 years

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 (11.3–13.8). Twenty-eight patients (87.5 %) had a successful outcome. Reoperation was performed in 2 patients for ASD and pedicle fracture (1 case each). Both VAS and ODI scores showed significant improvement compared to baseline. At the final follow-up, the ROM of both the index- and adjacent-level showed an obvious decrease. The IDH of the index level showed a tendency to decrease but the difference was not significant. The IDH of adjacent levels were not significantly affected by the surgery. Mean lumbar lordosis was increased at the final follow-up, and lumbar scoliosis over 3° was observed in 12 patients, with a mean angle of 5.6° (3–12°). Of all 35 prostheses, 15 were left-shifted, 3 right-shifted and 14 in the middle. In the coronal plane, 25 were rated as ideally placed, 5 discretely shifted, 4 slightly shifted and 1 markedly shifted. In the sagittal plane, only 12 prostheses were rated as ideally placed, 14 discretely shifted and 9 suboptimally placed. Prosthesis subsidence was noted in 3 (9.4 %) patients. Heterotopic ossification was detected in 25 segments (71.4 %), consisting of Class-I heterotopic ossification in 7 segments (20.0 %), Class-II in 9 segments (25.7 %), and Class-III in 9 segments (25.7 %). Class-IV heterotopic ossification was not observed. Conclusion: The cumulative survival was 100 % at a mean follow-up of 11.8 years. Clinical and radiological results were satisfactory and long-term clinical results were maintained. Reoperation and complication rates are acceptable, and our study does not substantiate the fear of reoperation or late complications. The results of our long-term follow-up indicate that, with strict indication, TDR is a safe and effective procedure as an alternative to lumbar fusion. Disclosures: author 1: none; author 2: none; author 3: none.

60 GLOBAL SAGITTAL ANGLE (GSA): A STEP TOWARD FULL BODY ASSESSMENT FOR SPINAL DEFORMITY Vincent Challier, Bassel Diebo, Shaleen Vira, Matthew Spiegel, Bradley Harris, Renaud Lafage, Barthelemy Liabaud, Jensen Henry, Frank Schwab, Virginie Lafage NYU Hospital for Joint Diseases, New York, USA Hypothesis: Global sagittal angle is clinically relevant. Design: Single center retrospective review. Introduction: According to Dubousset’s ‘‘conus of economy’’ theory, deterioration of sagittal alignment requires higher energy expenditure to maintain erect posture. Since the clinical impact of sagittal alignment is affected both by the severity of the deformity and a patient’s inability to recruit compensatory mechanisms, it is important to investigate new parameters that reflect both disability level and compensatory mechanisms for all patients. This study investigates the clinical relevance of the GSA. Methods: Retrospective review of patients who underwent full body X-rays and completed ODI and SRS-22. GSA was defined as the angle subtended by a line from the midpoint of the femoral condyles to the center of C7, and a line from the midpoint between the femoral condyles to the posterior superior corner of the S1 sacral endplate. After evaluating the correlation of GSA/ODI with classic sagittal parameters, linear regression models were generated to investigate how ODI related to radiographic parameters (TPA, PT). TPA and PT’s relation to GSA was then investigated. Results: 143 patients (mean 44y) were identified. GSA correlated significantly with ODI, PT, SVA, TPA and lower limb alignment. Regression between ODI and classic parameters only retained GSA as independent predictor (r = 0.517, r2 = 0.267, p \ 0.001). Analysis of standardized coefficients revealed that GSA increases when TPA increases (beta: 1.991) with concurrent decrease in PT (-1.323).

Eur Spine J (2015) 24 (Suppl 6):S669–S710 These findings echoed those of ODI, which increased with increased TPA (beta: 1.038) and decreased PT (-0.696). Conclusions: GSA goes further than classic parameters and quantifies the clinical impact of all patients’ compensation mechanisms in addition to their malalignment. Analyzing the relationship between GSA and PT revealed that compensation is the body’s defense against malalignment, and patients who lack compensation have both higher GSA and higher ODI scores. GSA is a significant and clinically relevant metric that quantifies both spinal deformity and disability even among those with alternative compensation mechanisms. 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: stock/shareholder; Company = Nemaris Inc; 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.

61 DEVELOPMENT OF VALIDATED COMPUTER BASED PREOPERATIVE PREDICTIVE MODEL FOR PROXIMAL JUNCTION FAILURE (PJF) OR CLINICALLY SIGNIFICANT PJK WITH 86 % ACCURACY BASED ON 510 ASD PATIENTS WITH 2-YEAR FOLLOW-UP Christopher Ames, Justin Smith, Frank Schwab, Amit Jain, Tamir Ailon, Eric Klineberg, Justin Scheer

S695 stratification, and need for prophylactic strategies for patients undergoing ASD surgery. Disclosures: author 1: grants/research support; Company = depuy Synthes Spine issg, consultant; Company = stryker medtronic depuy synthes, royalties; Company = biomet, stryker, employee; Company = ucsf author 2: grants/research support; Company = DepuySynthes, ISSGF, consultant; Company = Biomet, Nuvasive, Cerapedics, Medtronic,royalties;Company = Biomet; author 3: grants/ research support; Company = DePuy, AO, MSD, consultant; Company = K2M, Medicrea, MSD, stock/shareholder; Company = Nemaris INC, royalties; Company = MSD, K2M, other financial report; Company = Biomet, Nuvasive; author 4: none; author 5: none; author 6: grants/research support; Company = AO Spine, Depuy, OREF, other financial report; Company = Speakers Fees Depuy AO Spine; author 7: none.

62 PROSPECTIVE MULTICENTER ASSESSMENT OF COMPLICATION RATES ASSOCIATED WITH ADULT SPINAL DEFORMITY (ASD) SURGERY IN 291 PATIENTS WITH MINIMUM 2-YEAR FOLLOW-UP Justin Smith, Eric Klineberg, Frank Schwab, Christopher Shaffrey, Shay Bess, Gregory Mundis, Han Jo Kim, Justin Scheer, Christopher Ames International Spine Study Group, Chicago, USA

International Spine Study Group, Chicago, USA Hypothesis: A model based on baseline demographic, radiographic, and surgical factors can predict PJF or PJK. Design: Retrospective review of prospective multicenter ASD database. Introduction: PJF and PJK are significant complications. It remains unclear what are the specific drivers behind the development of either. This study attempts to develop a preoperative predictive model to identify patients at risk to develop PJF or PJK. Methods: Inclusion criteria: age C18, ASD, C4 levels fused. Variables included in the model were: demographics, primary/revision, use of 3-column osteotomy, UIV/LIV levels and anchor (screw, hooks), number of levels fused, and baseline sagittal radiographs (PT, PI, PI-LL, TK, and SVA). PJF was defined as requiring revision for PJK and PJK was defined as an increase from baseline of PJK [20° and with deterioration by at least 1 SRS-Schwab sagittal modifier grade from 6 weeks postop. An ensemble of decision trees were constructed using the C5.0 algorithm with 5 different bootstrapped models, and internally validated via a 70:30 data split for training and testing. Accuracy and the area under a receiver operator characteristic curve (AUC) were calculated. Final model utilized 13 preop variables. Results: 510 patients were included, with 357 for model training and 153 as testing targets (PJF: 37, PJK: 102). The overall model accuracy was 86.3 % with an AUC of 0.89 indicating a good model fit. The 6 strongest (importance C0.95) predictors were (% target): age ([64 years, 41.4 %), PI-LL ([48.7 deg, 35.6 %), UIV (T10–L3, 35.1 %), SVA ([ 13.5 cm, 32.5 %), LIV (sacroiliac, 31.6 %), and UIV Type (screws, 29.8 %). If a patient met these criteria, they had a 66.7 % chance of developing PJF or PJK with deterioration of sagittal alignment. Conclusion: A successful model (86 % accuracy, 0.89 AUC) was built predicting either PJF or clinically significant PJK. This model can set the groundwork for preop point of care decision making, risk

Hypothesis: Complication rates associated with ASD surgery are more accurately assessed with rigorous prospective documentation. Design: Prospective multicenter cohort study. Introduction: Previously reported rates of ASD surgery complications are limited by retrospective design, single-center cohorts, lack of rigorous complication collection or limited follow-up. Documenting these rates may prove useful for treatment planning and patient counseling. Methods: As part of a prospective ASD database, standardized collection forms, on-site coordinators and auditing helped ensure complete capture of complications. Intraop, periop (\6 weeks) and postop (6 weeks to last follow-up) complications were collected for patients with minimum 2 years follow-up. Results: Of 345 eligible patients, 291 (84 %) achieved 2 year followup (mean 2.1 years) and had a mean age of 56 years and previous surgery in 46 %. The majority (98 %) had treatment including posterior instrumentation and 20 % had a 3-column osteotomy. 82 (28.2 %) patients required at least one revision. A total of 102 periop complications (43 major, 59 minor) occurred in 70 (24.1 %) patients, including 84 from index surgery and 18 from revisions, and a total of 168 periop complications (82 major, 86 minor) were reported in 108 (37.1 %) patients, including 148 from index surgery and 20 from revisions. A total of 199 postop complications (137 major, 62 minor) occurred in 124 (42.6 %) patients. Collectively, 469 complications were reported in 204 (70.1 %) patients. Overall mean numbers of intraop, periop, postop and total complications per patient were 0.35, 0.58, 0.68 and 1.61, respectively. Conclusion: Although multiple reports have documented significant benefit from surgical treatment of ASD, these procedures are associated with high rates of complications, many of which may have minimal or no impact on outcome. In the present multicenter prospective study, 469 complications occurred in 204 (70.1 %) of 291 patients with minimum 2 years follow-up. These finding may prove useful in treatment planning and patient counseling.

123

S696 Disclosures: author 1: grants/research support; Company = DepuySynthes, ISSGF, consultant; Company = Biomet, Nuvasive, Cerapedics, Medtronic, royalties; Company = Biomet; author 2: grants/ research support; Company = AO Spine, Depuy, OREF, other financial report; Company = Speakers Fees Depuy AO Spine; author 3: grants/research support; Company = DePuy, AO, MSD, consultant; Company = K2M, Medicrea, MSD, stock/shareholder; Company = Nemaris INC, royalties; Company = MSD, K2M, other financial report; Company = Biomet, Nuvasive; author 4: grants/research support; Company = Depuy-Synthes, consultant; Company = Biomet, Medtronic, Nuvasive, Stryker, stock/shareholder; Company = Nuvasive, royalties; Company = Biomet, Medtronic, Nuvasive; author 5: grants/research support; Company = depuy, medtronic, k2, innovasis, stryker, consultant; Company = k2, allosource, nuvasive, royalties; Company = pioneer; author 6: grants/research support; Company = dePuy Synthes, consultant; Company = nuvasive, K2M, Medicrea, Misonix, royalties; Company = nuvasive, K2M; author 7: consultant; Company = Biomet, K2M, Medtronic, other financial report; Company = Stryker, Depuy; author 8: none; author 9: grants/research support; Company = Depuy Synthes Spine, consultant; Company = Depuy, Medtronic, Stryker, stock/shareholder; Company = Baxano Surgical, Doctor’s Research Group, royalties; Company = Aesculap, Biomet Spine, employee; Company = University California San Francisco.

63 DEVELOPMENT OF A PREOPERATIVE PREDICTIVE MODEL FOR INTRA- OR PERI-OPERATIVE MAJOR COMPLICATIONS WITH HIGH ACCURACY VALIDATED WITH 558 ASD PATIENTS

Eur Spine J (2015) 24 (Suppl 6):S669–S710 C0.90) and included (in decreasing importance): age, leg pain, ODI, #decompression levels, #interbody fusion levels, PCS, SRS-Schwab coronal curve type, Charlson comorbidity index, SRS Activity, T1PA, ASA grade, presence of osteoporosis, PT, SVA, primary vs revision, SRS pain, SRS total, use of BMP, use of iliac crest graft, and PI-LL. Conclusion: A successful model (87 % accuracy, 0.89 AUC) was built predicting major intra- or periop complications. This model can provide the foundation toward improved education and point-of-care decision making for patients undergoing ASD surgery. Disclosures: author 1: grants/research support; Company = Depuy Synthes Spine, consultant; Company = Depuy, Medtronic, Stryker, stock/shareholder; Company = Baxano Surgical, Doctor’s Research Group, royalties; Company = Aesculap, Biomet Spine, employee; Company = University California San Francisco; author 2: grants/ research support; Company = Depuy-Synthes, ISSGF, consultant; Company = Biomet, Nuvasive, Cerapedics, Medtronic, royalties; Company = Biomet; author 3: grants/research support; Company = DePuy, AO, MSD, consultant; Company = K2M, Medicrea, MSD, stock/shareholder; Company = Nemaris INC, royalties; Company = MSD, K2M, other financial report; Company = Biomet, Nuvasive; author 4; author 5: grants/research support; Company = Depuy-Synthes, consultant; Company = Biomet, Medtronic, Nuvasive, Stryker, stock/shareholder; Company = Nuvasive, royalties; Company = Biomet, Medtronic, Nuvasive; author 6: grants/ research support; Company = dePuy Synthes, consultant; Company = nuvasive, K2M, Medicrea, Misonix, royalties; Company = nuvasive, K2M; author 7: none; author 8; author 9: grants/research support; Company = AO Spine, Depuy, OREF, other financial report; Company = Speakers Fees Depuy AO Spine; author 10: none.

GROWING SPINE Christopher Ames, Justin Smith, Frank Schwab, Christopher Shaffrey, Gregory Mundis, Tamir Ailon, Eric Klineberg, Justin Scheer International Spine Study Group, Chicago, USA Hypothesis: A model based on baseline demographic, radiographic, and surgical factors can predict patients sustaining an intra- or periop major complication. Design: Retrospective review of prospective multicenter ASD database Introduction: The operative management of patients with ASD has a high complication rate and it remains unknown whether baseline patient characteristics and surgical plans may predict early complications (intraop and periop \ 6 weeks). The development of an accurate preoperative predictive model can aid in patient counseling, shared decision making, and improved surgical planning. Methods: Inclusion criteria: age C18, ASD. 45 variables were included in the initial training of the model and included demographic data, comorbidities, modifiable surgical variables, baseline HRQOL, and coronal and sagittal radiographic parameters. Patients were grouped as either having at least 1 major intra- or periop complication (COMP) or not (NOCOMP). An ensemble of decision trees was constructed using the C5.0 algorithm with 5 different bootstrapped models. Internal validation was accomplished via a 70:30 data split for training and testing each model, respectively. Overall accuracy, and the area under a receiver operator characteristic curve (AUC) were calculated. Results: 558 patients were included, NOCOMP: 410 (73.5 %), COMP: 148 (26.5 %). The overall model accuracy was 87.6 % correct with an AUC of 0.89 indicating a very good model fit. 20 variables were determined to be the top predictors (importance

123

64 RISK FACTORS OF SEVERE ADOLESCENT SCOLIOSIS TIED TO HTN Dingjun Hao, En Xie Department of Spine Surgery, Hong Hui Hospital, Xi’an Jiaotong University College of Medicine, Xi’an, China Background: Children with severe spinal curving from scoliosis were more likely to have worsening risk indicators for pulmonary hypertension, researchers reported here. Objective: To test for associations between degree of scoliosis curvature and cardiac function and to measure the incidence of pulmonary hypertension with adolescent idiopathic scoliosis, the study conducted a chart review of 1000 patients (ages 11–21) who underwent spinal fusion therapy from 2010 to 2014 Methods: Participants were compared with 300 age matched healthy control patients who visited the center for 2D-echocardiogram for innocent flow murmurs. In addition to echocardiogram outcomes, participants received measurement of scoliosis Cobb angle, aortic coordinations, and tricuspid regurgitation jet velocity. And also reviewed patient demographics. Right ventricular systolic pressure was a marker for pulmonary hypertension in cases where pressure was at or exceeded 25 mm/Hg. Spearman correlation was done to assess associations between spine curvature and systolic pressure, while logistic regressions were used to measure tricuspid regurgitation jet velocity betweenidiopathic scoliosis and control groups.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Result: Patients had an average age of 13.7, and an average Cobb angle of 57.7°. Also, 1000 patients with idiopathic curvature had preoperative echoes. In those with idiopathic scoliosis, 57.3 % adverbial mitral regurgitation, 21.7 % had mild tricuspid regurgitation, 2.77 % had mild mitral regurgitation, 2.73 % had mitral valve prolapse, 1.67 % had atrial septal defect, and 1.77 % had ventricular septal defect. There was one control with an abnormal tricuspid regurgitate jet velocity. Right ventricular systolic pressure was significantly lower in control participants compared with the patients (20.7 mm/Hg versus 27.7 mm/Hg, P \ 0.05). Conclusion: Children with severe spinal curving from scoliosis were more likely to have worsening risk indicators for pulmonary hypertension, researchers reported here. Follow-up studies would include postoperative echocardiograms of patients with elevated systolic pressure. The study recommended patients with scoliosis receive screening via echocardiogram to identify structural heart defects and pulmonary hypertension, and that those with elevated right ventricular systolic pressure ‘‘be considered for early surgery to prevent progression of pulmonary hypertension and even possible mortality. Disclosures: author 1: none; author 2: none.

65 THE ASSOCIATION BETWEEN SEVERITY OF SCOLIOSIS AND LUNG CLEARANCE INDEX (LCI) IN PATIENTS WITH ADOLESCENT IDIOPATHIC SCOLIOSIS (AIS) Anne Katrine Blyme, Birgitte Hanel, Martin Gehrchen, Kim G. Nielsen, Benny Dahl Rigshospitalet, Copenhagen, Denmark Introduction: AIS causes progressive distortion of intrathoracic contents commonly reflected in reduction of various dynamic and static lung volumes. The effect on the ventilation inhomogeneity of the lungs measured by nitrogen multiple breath wash-out (N2 MBW) and the lung clearance index (LCI) as the outcome is scarcely reported. The hypothesis of the presents study was that there is a significant correlation between LCI and Cobb angle in patients with AIS. Methods: In a prospective, single-centre study, a consecutive series of patients referred for surgical assessment of AIS or planned for surgery was included. In relation to their first outpatient visit or 1 week before surgery, spirometry, wholebody-plethysmography and N2 MBW was performed obtaining: forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC), total lung capacity (TLC), and LCI. An independent observer evaluated long-standing radiographs and registered largest Cobb angle and apical vertebrae. Correlation analysis using Spearman´s correlation test was applied due to the index nature of the LCI. Comparison of LCI-values with reference values was done using unpaired t test. Results: A total of 41 patients (85 % female) were included. The age was 14.7 (10–18) years and the Cobb angle was 49 (11–99) degrees (average with range) and 88 % of the curves were thoracic. LCI was significantly increased in AIS-patients compared to the age matched reference population; 7.35 vs. 6.54 (p = 0.001). There was a significant correlation between largest Cobb angle and LCI (Spearman’s rho 0.477, p = 0.002), whereas the correlation between TLC and largest Cobb angle was small and not statistical significant (Spearmans’s rho 0.273, p = 0.084). Additionally, no significant correlation was found between Cobb angle and spirometric values.

S697 Conclusion: LCI was significantly increased in patients with AIS, indicating that the thoracic deformity and resulting intrathoracic distortion may cause increased pulmonary ventilation inhomogeneity, thereby reflecting a more complex affection of lung function than previously assessed with conventional lung variables. Future studies will reveal if surgical treatment improves LCI or can be used as an additional variable in the assessment of curve progression. Disclosures: author 1: none; author 2: none; author 3: grants/research support; Company = Medtronic, Globus Medical, K2 M,consultant; Company = Medtronic, Globus Medical, K2M; author 4: none; author 5: grants/research support; Company = Medtronic, Globus Medical, K2M.

66 PRE-OPERATIVE MRI STUDY ON A CONTINUOUS SERIES OF ADOLESCENT IDIOPATHIC SCOLIOSIS: NEUROAXIAL ABNORMALITIES INCIDENCE AND ITS INFLUENCE ON SURGICAL TREATMENT Laura Scaramuzzo, Fabrizio Giudici, Marino Archetti, Leone Minoia, Antonino Zagra I.R.C.C.S. Galeazzi Orthopedic Institute, Milan, Italy Introduction: Aim of the study was to evaluate the neuroaxial abnormalities incidence in patients affected by adolescent idiopathic scoliosis (AIS) underwent to fusion, its association with curve type and influence on the scheduled surgical treatment and intraoperative neuromonitoring. Method: From May 2010 to December 2014 a continuous series of 140 patients affected by AIS with normal neurologic exam underwent pre-operative total MRI study. Patients were stratified considering the presence and type of neuroaxial abnormalities, age, sex, curve type according to Lenke classification, presence and type of alterations at intraoperative neuromonitoring. Patients were divided in 2 groups: group A patients with neuroaxial abnormalities; group B patients with normal MRI. Presence and type of abnormalities were registered and correlated to every considered factor to obtain a statistical significance and evaluate their influence on surgical treatment. Results: 140 patients, 116 female and 24 male satisfied the inclusion criteria. Male:female ratio was of 1:4. Mean age was 14.5 years (11–18). In 28 patients neuroaxial abnormalities were registered: 18 central spinal cord ectasia less than 3 mm, 8 syringomyelia, 1 ArnoldChiari 1, 1 ependymoma. Total incidence was 17 %, female group incidence 22.8 %, male group incidence 8.6 %. 22 patients, 16 %, showed an intraoperative neuromonitoring alteration, in 12 of the 22 a concomitant neuroaxial abnormality was registered. Group A showed an intraoperative neuromonitoring alteration incidence of 46 % versus a 9.5 % incidence in group B. A prevalence of neuroaxial abnormalities was registered in Lenke 1 C curve, with a no statistical significance p = 0.09. In no case the presence of a neuroaxial abnormalities required a change in the surgical program, only in one case (ependymoma) an adjunctive neurosurgical lesion excision was necessary. No neurological sequaele, nor post-operative complications were registered. Clinical outcome was optimal in 132 patients (96 %), satisfactory in 8 (4 %). Discussion: Neuroaxial abnormalities incidence in AIS is relatively high. Its identification doesn’t change the scheduled surgical treatment and doesn’t increase the intra and post-operative incidence of neurological sequaele. The presence of a neuroaxial abnormality seems to increase the detection of false positive at intraoperative neuromonitoring, especially for motor evocated potential.

123

S698 Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

67 EFFECTS OF FREQUENCY OF DISTRACTION IN MAGNETICALLY-CONTROLLED GROWING ROD LENGTHENING ON OUTCOMES AND COMPLICATIONS Caglar Yilgor, Kenneth Cheung, Kenny Kwan, Dino Samartzis, Ahmet Alanay, John Ferguson, Colin Nnadi, Ilkka Helenius, Muharrem Yazici, Gokhan Demirkiran, Behrooz Akbarnia Acibadem University School of Medicine; The University of Hong Kong; Starship Children’s Hospital/Auckland Bone and Joint Surgery; Oxford University Hospitals; Turku University Central Hospital, Department of Paediatric Orthopaedic Surgery; Hacettepe University; Orthopaedic Surgery, University of California, San Diego, USA Summary: This is a retrospective review of prospectively collected data from a multicentre study of early-onset scoliosis treated by the magnetically-controlled growing rod with a minimum of 2-year follow-up. Higher distraction frequency was associated with an increased incidence of re-operations due to failure of rod distraction but lower rate of implant-related complications. Hypothesis: More frequent distractions were less likely to be associated with re-operations and implant-related complications. Design: Retrospective review of prospectively collected data. Introduction: Magnetically-controlled growing rods (MCGR) are an alternative to traditional growing rods in skeletally immature patients by providing non-invasive, outpatient distractions mimicking a patient’s physiological growth. However, the ideal frequency of MCGR distraction is currently not known. This study aimed to determine the effects of distraction frequencies on implant-related complications and re-operations. Methods: Consecutive patients undergoing MCGR treatment with a minimum of 2-year follow-up from 6 centres were included. Clinical and radiographic data were collected prospectively. Results: Thirty patients were included in this study. The mean age at the time of surgery was 7.3 years (range 4–14 years) and the mean follow-up period was 35 months (range 24–61 months). Patients were divided into 2 groups according to their distraction frequency: Group 1 (every 1 week–2 months), and Group 2 (every 3–6 months). There were 14 patients in Group 1, and 16 in Group 2. Patients in Group 1 had more re-operations due to failure of rod distraction (71 vs 25 %) and a higher incidence of PJK (21 vs 13 %) than Group 2. However, there were fewer incidences of implant-related complications including rod breakage and proximal foundation failure (14 vs 31 %) in Group 1 compared with Group 2. Conclusion: This is the largest series with the longest follow-up to date that examines the effect of distraction frequency in MCGR lengthening. Our study showed more frequent distractions were associated with increased incidence of rod distraction failure and PJK but lower incidence of implant-related complication. Clinicians should be aware of a potential higher risk for re-operation if the interval between each distraction is less than 3 months. Further studies with a larger cohort are required to determine the critical threshold for distraction frequency and reoperations. Disclosures: author 1: none; author 2: grants/research support; Company = Ellipse Technologies; author 3: none; author 4: none; author 5: grants/research support; Company = Depuy Synthes, consultant; Company = Stryker; author 6: consultant; Com-

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 pany = K2M, Ellipse, stock/shareholder; Company = Nuvasive, K2M; author 7: royalties; Company = Ellipse Technologies; author 8: grants/research support; Company = Medtronic, Baxter, consultant; Company = Medtronic, Depuy Synthes; author 9: consultant; Company = DePuy Synthes; author 10: none.

68 OUTCOME OF SCOLIOSIS SURGERY IN SEVERE CEREBRAL PALSY PATIENTS Tobias Schulte, Klaas Beckmann, Georg Gosheger, Albert Schulze Bo¨vingloh, Matthias Borowski, Ulf Liljenqvist, Viola Bullmann, Tobias Lange University Hosptial Mu¨nster, Department of Orthopedics, Mu¨nster, Germany Scoliosis surgery in patients with serious cerebral palsy (CP) is a challenge for patients, families and surgeons. The patients’ general condition is often restricted and it needs to be discussed individually how much surgery is justifiable and reasonable. In this context the question is: How much can surgery improve the patients’ quality of life. A cohort of consecutive CP scoliosis patients GMFCS 4 and 5 who underwent either posterior-only or combined anterior-posterior instrumentation in a single institution between 1997 and 2012 was studied. Only patients with a min. follow-up of 2 years and a complete set of data were included. Patient records and radiographs were investigated. Quality of life (QoL) was assessed using an established questionnaire for neuromuscular scoliosis patients and families (Bridwell et al., Spine 1999) concluding in a sum score from -34 worst to +34 best (20 questions in total: 3 questions on general assessment of surgical success, 17 detailed questions on need for care, function, respiratory symptoms, pain etc.). 46 patients met the inclusion criteria, averaged follow-up 49 months: 29 patients with posterior-only (Group P; age 15.1 years; BMI 14.3; 11 male, 18 female) and 17 patients with anterior-posterior instrumentation (Group AP; age 16.2 years; BMI 14.3; 9 male, 8 female). The preoperative Cobb angle of the main curve was 86° in Group P, 40° after surgery and 44° at follow-up; in Group AP the curves were corrected from 107° to 54° resulting at 55° at follow-up. The major predictor for relative curve correction was flexibility before surgery (Group P 33 %; Group AP 28 %). Type of surgery (Group P vs. AP), preoperative Cobb angle, gender, BMI and pelvic obliquity did not statistically noticeable influence relative curve correction. Perioperative complications according to Master et al. (Spine 2011) occurred in 21 % of patients in Group P and 47 % in Group AP. Blood loss 1800 ml vs. 2750 ml. Overall, in 96 % surgery was assessed as success. The averaged total QoL score was +11.7 points without significant differences between Group P (+11.2) and AP (+12.7). Interestingly, patients with more severe preoperative radiologic and clinical pathologies rated surgery as more beneficial than those with less severe. In conclusion, radiologic outcomes were comparable to the literature. QoL at follow-up was noticeable good. Posterior-only instrumentation showed less morbidity, however achieving similar results as combined instrumentation. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: consultant; Company = DePuy Synthes Medtronic ullrich; author 7: none; author 8: none.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 69 COMPARISON OF SEGMENTAL PEDICLE SCREWS VERSUS HYBRID CONSTRUCTS USING SUBLAMINAR WIRES FOR DEFORMITY CORRECTION IN CEREBRAL PALSY Hossein Mehdian, Luigi Aurelio Nasto, Ana Belen Perez Romera, Oliver Stokes The Centre for Spinal Studies and Surgery, Queen’s Medical Centre University Hospital, Nottingham, NG7 2UH, UK Background context: Correction of spinal deformity in cerebral palsy patients is challenging. Sublaminar wires have been used in an attempt to overcome the difficultly of placing pedicle screws in small pedicles and in an attempt to reduce pullout of instrumentation. Matched cohorts were followed-up after deformity correction using segmental pedicle screw constructs versus hybrid constructs using sublaminar wires. The degree and maintenance of the correction and the complication rates were compared. Study design: Prospectively enrolled consecutive series of spastic quadriplegic cerebral palsy patients requiring deformity correction Objectives: Correction of spinal deformity in cerebral palsy patients is challenging. Sublaminar wires have been used in an attempt to overcome the difficultly of placing pedicle screws in small pedicles and in an attempt to reduce pullout of instrumentation. The adequacy and maintenance of correction and complication rates were compared in matched cohorts of 48-patients treated with segmental pedicle screw constructs and hybrids using sublaminar wires. Segmental pedicle screw constructs provided better maintenance of correction and were associated with fewer complications. Methods: 28 male and 20 female spastic quadriplegic cerebral palsy patients, age of 16-years (11–25) underwent correction for spinal deformity. Group I (25 patients) were treated with segmental pedicle screw constructs and in Group II (23 patients) received hybrid constructs. There was a mean follow up of 4.8 years (2–13). Results: Instrumentation was from T2 to pelvis in all. In group I, preoperatively the coronal plane deformity measured 79°, thoracic kyphosis 68°, lumbar lordosis 43°, pelvic obliquity 19° and there was 32 mm list. Corresponding measures in group 2: 75°, 65°, 40°, 18° and 33 mm. Postoperatively these measures of deformity in group 1 were: 10°, 40°, 49°, 2° and 2 mm. In group 2: 24°, 44°, 55°, 7°, and 10 mm. The surgical time was 40 min longer in group 2 and there was a 10 % increase in implant failure and junctional kyphosis and 16 % failure of maintenance of correction. Conclusions: The magnitude and maintenance of correction of spinal deformity of in spastic quadriplegic cerebral palsy patients is better with segmental pedicle screw constructs than hybrid constructs, with reduced complication rates. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none.

70 WHICH RADIOLOGIC PARAMETERS ARE ASSOCIATED WITH DISC AND FACET DEGENERATION IN THE LUMBAR CURVE AFTER SELECTIVE THORACIC FUSION IN AIS: AN MRI STUDY WITH MINIMUM 10 YEARS FOLLOW UP Sinan Kahraman, Meric Enercan, Bahadir Gokcen, Levent Ulusoy, Ayhan Mutlu, Tunay Sanli, Erden Erturer, Cagatay Ozturk, Ahmet Alanay, Azmi Hamzaoglu Istanbul Spine Center, Istanbul, Turkey

S699 Summary: 31 AIS patients whom underwent selective thoracic fusion with minimum 10 years of f/up; compared with their preoperative and final f/up X-rays and MRI’s in terms of disc (DD) and facet joint degeneration (FJD). All radiologic parameters were measured by two radiologists.We found that residual curve more than 10° and sacral oblique take off angle more than 5° correlated with FJD of L1–L2 and L5–S1 DD respectively. Design: Retrospective. Introduction: The purpose of this study was to evaluate the DD and FJD of mobile lumbar levels in Selective Thoracic Fusion (STF) pts with MRI and find out which radiological parameters affected DD and FJD at least 10 years after the operation. Methods: We reviewed 31 (29f, 2m) pts with AIS who underwent STF. There were 5 pts with type T1A, 11 pts with T1B, 14 pts with T1C and 1 pts with T2B. All pts had complete radiographic data with a mean 12.1 years follow-up (11–18). Mean age was 14 (11–16). They were analyzed in terms of the difference between lumbar DD and FJD grades preop and at f/up; and correlation with residual curve, LIV tilt, disc angulation of LIV, sacral oblique take off angle, leg length discrepancy (LLD) and the difference between all coronal and sagittal parameters were assessed. All statistical analyses were made with Spearman correlation test. Results: All FJD grades were significantly different between preop (mean 1.8) and post-op (mean 2.0) except in the T11–T12 and T12– L1 facet joints (p \ 0.001). DDs in L4–L5 and L5–S1 were significantly (p \ 0.001) different at preop and f/up, and insignificant in other levels. Statistical analyses showed that increased residual curve more than 10° correlated with L1–L2 concave FJD (p = 0.03) and sacral oblique take-off angle more than 5° correlated with L5–S1 DD. Also residual curve more than 10° correlated with sacral oblique take off angle more than 5° (80 %).The other parameters including disc angulation of LIV, LLD, LIV tilt angle, all sagittal and coronal X-ray parameters were not correlated with any DD or FJD (p [ 0.05). At the f/up SRS score was 4.34 (3.74–4.92) and ODI scores was 5.3 (0–13.3). Conclusion: Despite a mild difference in radiologic DD and FJD, SRS (4.34) and ODI (5.3) scores indicated that there were no patients with back pain and clinical complaints after STF at minimum 10 years f/up. Residual curve more than 10°, and sacral oblique takeoff angle more than 5° were found as risk factors for degeneration and were associated with L1–L2 FJD and L5–S1 DD respectively. 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: grants/research support; Company = Depuy Synthes, consultant; Company = Stryker; author 10: consultant; Company = Medtronic.

71 DEEP PARASPINAL MUSCLES IN IDIOPATHIC SCOLIOSIS: AN ELECTROPHYSIOLOGICAL AND HISTOCHEMICAL STUDY Martin Krbec1, Josef Za´menı´k2, Ivana Stetkarova3 1 Department of Orthopedics and Traumatology, 3rd Faculty of Medicine, Charles University in Prague and Faculty Hospital Kra´lovske´ Vinohrady, Prague; 2Department of Pathology and Molecular Medicine, 2nd Faculty of Medicine, Charles University in Prague and Motol University Hospital, Prague; 3 Department of Neurology, Charles University in Prague, Third Faculty of Medicine and Faculty Hospital Kralovske Vinohrady, Prague, Czech Republic

Introduction: The pathogenesis of idiopathic scoliosis has been poorly understood. The potentially involved local changes in deep

123

S700 paraspinal muscles and/or dysfunction of spinal inhibitory circuits have remained unknown. Objectives: To characterize the electrophysiological and histological findings in deep paraspinal muscles (on the concave and the convex side of the curve) in patients with idiopathic scoliosis. To compare the morphological and functional changes of paraspinal muscles on both sides of the main scoliotic curve. Methods: 18 patients with idiopathic scoliosis (14 females, 4 males, 12–29 years), with Lenke 1, 2 or 3-type were enrolled into this study. Lower thoracic curves were focused. Needle EMG of the paraspinal muscles at convexity and concavity of the maine curve was performed 1 day before corrective surgery. Cutaneous silent period (CSP) induced by noxious digit II stimulation in thenar muscles was recorded in IS patients and in healthy volunteers. The study also included 56 biopsy samples from different deep paraspinal muscles obtained from all the patients during corrective spinal surgery. The muscle tissue samples were frozen in isopentane pre-cooled in liquid nitrogen. Serial cross-sections were stained for myofibrillar ATPase for the classification of fibers types and the fibers were counted in each biopsy to establish the distribution of fiber types. Results: All patients examined by needle EMG showed increased amplitude of motor unit action potentials (MUP) on the convex side of the curve. One subject presented double curve and had higher amplitude of MUP on the left side. 14 subjects have increased MUP amplitude on this side (range 10–62 %). CSP did not significantly differ between concavity and convexity but shortening of CSP duration was observed in more pronounced scoliotic curve. All of them presented changes in muscle fiber distribution with numerical predominance of type I on the curve convexity and occasional fiber type grouping on the convex side of the curve. Conclusions: Our data demonstrate a significant asymmetry in fiber type distribution corresponding with an altered function in paraspinal muscles with predominance on the convexity of the curve. Spinal inhibitory reflex is preserved; however, it could be modulated in more progressive IS. Thus, local neurogenic changes in the paraspinal muscles might play an important role in the development of idiopathic scoliosis. Disclosures: author 1: grants/research support; Company = IGA NT/ 13693-4, CZ; author 2: none; author 3: none.

72 VARIANTS OF POC5 GENE IN ADOLESCENT IDIOPATHIC SCOLIOSIS IN FRENCH CANADIAN FAMILIES Florina Moldovan, Shunmoogum A. Patten, Soraya Barchi, Julie Couillard, Patrick Edery, Stefan Parent CHU Sainte Justine and Faculty of Dentistry University de Montreal, Montreal, Canada; CRCHUM and Department of Neuroscience, Universite´ de Montreal, Montreal; Hospices Civils de Lyon, Lyon, France; Inserm U1028, CNRS UMR5292, University Lyon, France Summary: We investigated centrosomal protein-coding gene POC5 by exome sequencing and Sanger analysis. We previously discovered that POC5 is a causative gene in a large multiplex family with Idiopathic Scoliosis (IS). This study reveals variant c.C1286T (p.A429V; rs146984380 of POC5 gene as a cause of AIS in French Canadian Families. In the context of this highly heterogeneous disorder, the first AIS-causing gene (POC5) is a major step toward deciphering the genetic component of AIS. Hypothesis: New variants in POC5 gene are the cause of familial AIS. Design: Causal genetic variants were investigated in French Canadian Families with AIS and ethnically matched controls.

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Introduction: The etiological heterogeneity of AIS disorders is very complex, and a more complete picture of genetic risk is needed to well understand the genetic events leading to this diseases. Our recent work identified the centrosomal protein-coding gene POC5 as a disease-causing gene in French Families with IS. We sought to expand on this study and to investigate for functional variants in POC5 gene in French Canadian Families with AIS. Methods: 48 French Canadian families (88 patients) from Quebec were included in this study. The inclusion criteria were Family history of at least 3 members with progressive or non-progressive AIS, Cobb’s angle [15. DNA was extracted from peripheral blood or saliva and analysed by Sanger and Exome Sequencing. The frequency of variants was compared to the controls of 1000 Genomes, Exac Browser database and 96 ethnically matched controls. Results: Variant c.C1286T (p.A429V; rs146984380; allele frequency: 0,007583), in POC5 gene was identified in 9 AIS patients out of 48 Families (88 AIS patients). We also identified the variant c.G1336A (p.A446T; rs34678567; allele frequency: 0.01101) in 2 AIS Family and 4 new missense variants located on the coding regions of POC5 (all absent from the genome variants databases and 96 controls), and two rare missense variants (Exome10) (allelic frequency of *0.6 % in the 1000 Genomes). Conclusion: This is the first study to identify a rare and novel variants in POC5 gene in French Canadian AIS families. This represents an important step towards decoding genetic causes of AIS. Supported by Yves Cotrel Foundation. Disclosures: author 1: grants/research support; Company = Yves Cotrel Foundation; author 2: none; author 3: none; author 4: none; author 5: none; author 6: grants/research support; Company = K2M, DePuy Synthes Spine, Medtronic, EOS-Imaging, consultant; Company = DePuy Synthes Spine, Medtronic, EOS-Imaging, stock/ shareholder; Company = Spinologics.

73 BENDING VS FULCRUM VS TRACTION X-RAY UNDER GENERAL ANESTHESIA (TRUGA) FOR EVALUATION OF FLEXIBILITY OF CURVES AND PREDICTION OF CORRECTION IN PATIENTS WITH AIS: WHICH IS BETTER? Meric Enercan, Sinan Kahraman, Bahadir Gokcen, Mutlu Cobanoglu, Sinan Yilar, Tunay Sanli, Erden Erturer, Cagatay Ozturk, Ahmet Alanay, Azmi Hamzaoglu Istanbul Spine Center, Istanbul, Turkey Summary: Traction X-ray under general anesthesia (TrUGA) reflects flexibility in main thoracic curves more than 60° better than Fulcrum and Bending X-rays. TrUGA also has higher agreement with postop results. Design: Retrospective. Introduction: Bending X-rays (BXR) are traditionally used for evaluation of flexibility AIS pts. Fulcrum X-rays (F) and Traction xray under general anesthesia (TrUGA) have been promoted as alternatives to BXR. The aim of this study is to compare all 3 methods in pts with AIS. Methods: 80 pts with AIS who were operated by pedicle screw posterior instrumentation and have had BXR,F and TrUGA were included in this study. Curves in standing and flexibility X-rays were measured and flexibility rates obtained by each method at proximal thoracic (PT), Main thoracic (MT) and Thoracolumbar/lumbar (TL/L) curves was compared. A subgroup of patients with more than 60° of MT curve was also evaluated. To evaluate the best method to predict

Eur Spine J (2015) 24 (Suppl 6):S669–S710 the postsurgical result, the agreement between the methods and postop results were examined by using Bland–Altman method. Results: Ave age of pts (70F, 10M) was 14 (11–17) years and ave f/up was 32.7 (24–104) months. There was no significant difference for PT curves between BXR and TrUGA. Highest flexibility was demonstrated by TrUGA for MT curves and the difference was significant when compared to BXR (p \ 0.001) but not significant compared to F (p = 0.56). However, TrUGA demonstrated significantly higher flexibility than BXR and F for curves over 60° (p \ 0.05). In TL/L curves, BXR demonstrated more flexibility rates than TrUGA (p [ 0.05) and F (p = 0.003). Each of the methods showed significantly less flexibility than postop correction for PT and MT curves (p \ 0.001). Regarding TL/L curve, the difference between F and postop correction was significant (p \ 0.01) while it was not significant for BXR and TrUGA (p [ 0.05). TrUGA demonstrated better 95 % limits of agreement with postop results for MT and TL/L curves when compared to BXR and F. Conclusion: TrUGA offers better flexibility for severe MT curves compared to BXR and F. BXR offers better flexibility rates for PT and TL/L curves however, overestimates postop correction in TL/L curves.TrUGA demonstrates higher agreement with post op correction for MT and TL/L curves. 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: grants/research support; Company = Depuy Synthes, consultant; Company = Stryker; author 10: consultant; Company = Medtronic.

74 PLASTICITY OF WEDGE DEFORMITIES OF VERTEBRAL BODIES IN SKELETAL IMMATURE PATIENTS WITH ADOLESCENT IDIOPATHIC SCOLIOSIS WHO TREATED CORRECTIVE SPINAL SURGERY Takahiro Makino, Takashi Kaito, Masafumi Kashii, Tsuyoshi Sugiura, Kazuomi Sugamoto, Hideki Yoshikawa Department of Orthopaedic Surgery, Osaka University Graduate School of Medicine, Osaka, Japan Purpose: Vertebral body deformities including wedge of vertebral bodies in patients with adolescent idiopathic scoliosis (AIS) is one of the morphometric characteristics; however, whether the plasticity of this change in skeletal immature patients remains unclear. The purpose of this study was to investigate the plasticity of wedge deformity of vertebral bodies in patients with AIS using in vivo three-dimensional (3D) analysis. Methods: Ten female patients with AIS who underwent posterior fusion in our hospital between 2008 and 2013 and whose Risser grade was B3 at the time of surgery were included. The mean age of the patients was 12.2 years at the time of surgery. From the DICOM data obtained by CT scans (slice thickness, 0.625 mm) taken immediately after surgery (within a week) and 1 year after surgery, 3D bone models of each vertebra in major structural curve were constructed. The height of vertebral body of concave and convex sides in anterior, middle, and posterior portion was measured in each vertebra and vertebral height ratio (VHR: concave side/convex side) was calculated. The approximate planes of vertebral endplates were made semiautomatically, and the angle between the upper and lower endplates in each vertebra (VA: vertebral angle) was calculated. Results: The mean VHR in end vertebrae (anterior/middle/posterior) was (1.017/1.018/1.018) immediately after surgery and (1.012/1.006/ 1.007) 1 year after surgery; in apex vertebrae, (0.859/0.802/0.852) and (0.892/0.836/0.880); in vertebrae except end and apex vertebrae,

S701 (0.918/0.881/0.916) and (0.938/0.904/0.928).The mean VA in end vertebrae was 3.66° immediately after surgery and 3.72° 1 year after surgery; in apex vertebrae, 8.90° and 8.01°; in vertebrae except end and apex vertebrae, 6.26° and 5.44°. In vertebrae except end and apex vertebrae, the VHR in the anterior and middle portion increased and the VA decreased significantly postoperatively. In contrast, neither the VHR nor VA in apex vertebrae changed postoperatively. Conclusions: This study revealed that the wedge deformities of the vertebrae except end and apex vertebrae were reversible after the distraction force had been applied in the concave side by the corrective surgery. This result can be extrapolated to the prognostic prediction of brace treatments, because the correction of asymmetric loading to vertebrae under brace treatments can induce vertebral remodeling in skeletal immature patients. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

THE ‘BEST OF SHOW’ PAPERS

75 GELATINE MATRIX WITH HUMAN THROMBIN DECREASES BLOOD LOSS IN ADOLESCENTS UNDERGOING POSTERIOR SPINAL FUSION FOR IDIOPATHIC SCOLIOSIS. A RANDOMIZED CLINICAL TRIAL Ilkka Helenius, Heli Keskinen, Johanna Syva¨nen, Olli Pajulo Department of Paediatric Orthopaedic Surgery, Turku University Central Hospital and University of Turku, Finland Gelatine matrix with human thrombin has been widely used in adolescents undergoing surgery for idiopathic scoliosis, but no randomized clinical trials exist on its effectiveness in children. A prospective, randomized comparison of gelatine matrix with human thrombin (Floseal Matrix 5 mL, Baxter) in addition to traditional surgical methods to address blood loss in adolescents undergoing posterior spinal fusion for idiopathic scoliosis was performed. Patients were randomized (1:1) to receive a minimum two and maximum four units of gelatine matrix with human thrombin (intervention, treatment group [T]) in addition to traditional surgical haemostatic methods, which were solely used in the control group (C). Sixty adolescents operated for idiopathic scoliosis using posterior pedicle screw instrumentation were enrolled (out of 90 eligible) and randomized into C (15.4 years, MC 55°) and T groups (16 years, MC 53°). The groups were similar at baseline with one exception: C group had significantly less flexible main thoracic curves on bending radiographs than T group (p = 0.033). Every operative hour increased the intraoperative blood loss by 357 mL (95 % CI 273–440 mL, p \ 0.001) and total blood loss by 431 mL (95 % CI 343–518 mL, p \ 0.001). After adjusting for operative time and number of pedicles instrumented, intervention significantly decreased intraoperative (171 mL, 29 %, 95 % CI 22–320 mL, p = 0.025) and total blood loss (177 mL, 16 %, 95 % CI 21–333 mL, p = 0.027) as compared with C group. Haemoglobin decrease from a day before the operation to the second postoperative day was significantly smaller in the T group (6 g/L, 95 % CI 1.3–11 g/L, p = 0.013). Use of gelatine matrix with human thrombin reduces intraoperative blood loss and postoperative haemoglobin decrease in adolescents undergoing posterior spinal fusion for idiopathic scoliosis.

123

S702

Eur Spine J (2015) 24 (Suppl 6):S669–S710

Disclosures: author 1: grants/research support; Company = Medtronic, Baxter, consultant; Company = Medtronic, Depuy Synthes; author 2: none; author 3: none; author 4: none.

77 STUDY OF LUMBAR MULTIFIDUS MUSCLES IN PATIENTS WITH CHRONIC LOW BACK PAIN AND IN DEGENERATIVE SCOLIOSIS HUMAN—A BLINDED RANDOMIZED CONTROLLED TRIAL (RCT)

76 HEALTH-RELATED QUALITY-OF-LIFE IN ADOLESCENT IDIOPATHIC SCOLIOSIS PATIENTS 25 YEARS AFTER TREATMENT

Dingjun Hao, En Xie

Ane Simony, Steen Bach Christensen, Leah Y. Carreon, Mikkel O. Andersen

Objective: To evaluate the para vertebral muscle (PVM) degeneration in patients with degenerative scoliosis human, using 3.0T magnetic resonance imaging. Summary of background data: Numerous studies have described the histological and morphological changes to the PVM in patients with chronic low back pain and lumbar radiculopathy. Methods: 297 patients with low back pain (LBP group) and 300 control patients without low back pain (healthy volunteers, HV group) were examined. The purpose of the double-blind was to prevent any bias generating from the researchers and participants who know the adopted treatment. The method we use is: Patients were randomly assigned to be taken to different treatments (conservative treatment and surgery). Another doctor of the clinic (who doesn’t know the patient’s treatment) is a researcher for the evaluation and follow-up. Doctors confirm that patients could not see the separate treatment from the wound. Because researchers and patients were masked, this study is a double blind clinical study. The cross-sectional area (CSA) and percentage of fat infiltration area (%FIA) of the bilateral multitudes and longitudinal muscles at the L1–S1 levels were measured using T2-weighted axial magnetic resonance imaging and computer software. A multitude muscle biopsy and histological evaluation were performed in some patients. Results: In the LBP group, the CSA of the multitudes muscle was significantly smaller and the % of FIA of both muscles was significantly higher on the concave side than on the convex side at all levels (P \ 0.0001 for each). These differences were also found in the longitudinal muscles at the L3–L4, L4–L5 and L5–S1 levels (P \ 0.05 for each). Histologically, the multitudes muscle exhibited reductions in the muscle fiber size and number of nuclei on the concave side. In the LSS group, the total CSA and %FIA did not differ substantially between the left and right sides. However, in patients with unilateral radiculopathy, the CSA of the multitudes muscle was significantly smaller (P \ 0.05) and the % of FIA of both muscles was significantly higher (P \ 0.05) on the symptomatic side, especially at 1 level below. Conclusion: This observational study of 3.0T magnetic resonance imaging and histology showed that muscle degeneration was more common on the concave side in patients with LBP. Radiculopathy and spinal deformity may contribute to the PVM degeneration. Disclosures: author 1: none; author 2: none.

Middelfart Hospital, Middelfart, Denmark Introduction: The purpose of this study was to evaluate the long term clinical outcomes using validated measures of health-related qualityof-life (HRQOL), 25 years after termination of scoliosis treatment, in a cohort of Adolescent Idiopathic Scoliosis Patients. Method: 219 consecutive patients treated with Boston brace (Brace) or posterior spinal fusion (PSF) using Harrington- DDT instrumentation between 1983 and 1990 at Rigshospitalet Copenhagen, were invited to participate in a long-term evaluation study. A validated Danish version of the Scoliosis Research Society 22R (SRS22R) and Short Form-36 (SF36v1) were administrated to the patients 2 weeks before the clinical and radiological examination. Results: 159 (72.6 %) patients participated in the clinical follow up and questionnaires, 11 patients participated only in the questionnaires, 8 emigrated, 4 were excluded due to progressive neurological disease and 2 were deceased. The total follow up was 170 patients (83 %), and the average follow up was 24.5 years (range 22–30 years). In the Brace group, the Cobb angel prior to treatment was 37.5° (35.1°–40.0°), after treatment 34.7° (31.9°–37.5°). Cobb angel after 20 years 40.2° (36.7°–43.6°). In the surgical group the Cobb angel prior to treatment was 54.5° (50.4°–58.8°), 1 year postoperative 29.5° (25.7°–33.9°).Cobb angel after 20 years 32.35° (27.9°–39.5°). 26 patients had distal segment degeneration in X-rays (16.5 %), 4 patients treated with Brace and 22 patients with posterior spinal fusion. 8 patients had proximal segment degeneration (5 %), 2 treated with brace and 6 patients with posterior spinal fusion. 4 patients were treated with posterior fusion of the distal adjacent segment (2.6 %), 1 brace pt and 3 Harrington-DDT pt. SRS22R domain scores were within the range described as normal for the general population with no statistical difference between the groups except in the Satisfaction domain, where the PSF group had better scores than the braced group. The SF36 PCS and MCS scores in both AIS cohorts were similar to the scores for the general population. Conclusion: HRQOLs, as measured by the SRS22R and SF-36, of adult AIS patients treated with Boston brace or PSF during adolescence were similar to the general population. No clinical progression of the deformity has been detected during the 25-year follow up period. The PSF group had a small but statistically significant higher score in the Satisfaction domain compared to the braced group. Disclosures: author 1: grants/research support; Company = Bevica Foundation; author 2: grants/research support; Company = Danish Rheumatic Association; author 3: 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; author 4: none.

123

Department of Spine Surgery, Hong Hui Hospital, Xi’an Jiaotong University College of Medicine

78 EFFICACY AND SAFETY OF PROPHYLACTIC LARGE DOSE OF TRANEXAMIC ACID IN ADOLESCENT WITH IDIOPATHIC SCOLIOSIS SURGERY: A PROSPECTIVE, RANDOMIZED, DOUBLE-BLIND, PLACEBO-CONTROLLED STUDY. Dingjun Hao, En Xie Department of Spine Surgery, Hong Hui Hospital, Xi’an Jiaotong University College of Medicine

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Objective: To evaluate efficacy and safety of large doses for TA on blood loss during spinal operations. Summary of background data: Blood loss associated with spinal operations is a common potential cause of morbidity and often require blood transfusion which subject patients to the known risks of blood transfusion including transmission of diseases. TA is used routinely to reduce bleeding in cardiac, orthopaedic, and hepatic surgery, however, it is used in neurosurgery is uncommon and few studies reported the use of antifibrinolytic drugs in spine surgery. Methods: 391 consecutive patients undergoing adolescent with idiopathic scoliosis Surgery with expecting significant blood loss at our Hospital between June 2010 and December 2013 were assigned to 2 groups randomly, TA and placebo. After taking short term the induction of anesthesia, patients received either TA or placebo as a loading dose of 2 g (for teenages) or 30 mg/kg (for children), followed immediately by continuous infusion of 100 mg/h (for adults) or 1 mg/kg/h (for children) during surgery and for 5 h after the operation. Outcoming measures included total (i.e., intraoperative and postoperative) blood loss, amount of blood transfusion, as well as postoperative hemoglobin, and hematocrite levels. The data were analyzed by means of Statistical Package for the Social Science Version 12.0. The results were presented as mean ± SD. Independent Student t test was used to compare the 2 groups and the differences were considered significant if the P-value was \0.05. Results: There were 201 males and 190 females, ranging in age from 4 to 14 years with a mean of 10 and median of 9 years. Statistical analysis showed no significant differences between the 2 study groups with regard to age, sex, weight, preoperative hemoglobin, and hematocrite levels, type of surgery, as well as operative time. In other way, patients who received TA had 49 % reduction of blood loss (P \ 0.007) and required 80 % less blood transfusion (P \ 0.008) than patients who received placebo. The hospital stay was shorter in the TA group, but it did not achieve statistical significance. There were no complications related to the use of large doses of TA in this study. Conclusions: Prophylactic use of large doses of TA provides an effective, safe, and cheap method for reducing blood loss during and after spinal operations. Hence, TA may help in reducing not only transfusion related complications but also operative expenses. Disclosures: author 1: none; author 2: none.

79 TLIF VERSUS INSTRUMENTED POSTEROLATERAL FUSION (PLF). A 5–10 YEARS FOLLOW UP—LONG TERM RESULTS FROM AN RCT Kristian Høy, Kamilla Truong, Thomas Andersen, Cody Bu¨nger Aarhus University Hospital, Aarhus, Denmark Introduction: TLIF has become a major approach to achieve anterior support and is used worldwide. Long-term result of this method is lacking. Former Long term results of anterior fusion has in some studies shown significant improvement in functional outcome. We conducted a long term follow up of our RCT at 5–10 years. Materials and methods: During 01.11.2003–01.11.2008 100 pat’s were prospectively randomized to TLIF or PLF. The TLIF-group was operated using TSRH (Medtronic) and Implex (Zimmer) and allograft. The PLF- group was operated using TSRH (Medtronic) and allograft. Inclusions criteria: segmental instability due to disc degeneration, former disc herniation, spondylolisthesis\2. Functional outcome was registered prospectively, and after 5–10 years, using Dallas Pain questionnaire (DPQ), SF-36, Low Back pain

S703 questionnaire, and ODI. All follow up patient had a new MRI, and new X-ray at 5–10 years follow up. Results: The overall follow up rate was 85 %, 90 %, (44) in the PLF group and 81 %, (41) in the TLIF group no difference between groups. The mean follow up time was 8.6 years (5–9 years). The mean age at follow up was 59 years of age ranging from (34–76 years) The reoperation rate in a long term perspective was 37.5 % in the TLIF group and 26 % in the PLF group. According to LBPRS: Back was pain was 3.8 (Mean), TLIF (3.65) PLF (3.97), Leg pain was 2.68 (Mean) 2.90 (TLIF) and 2.48 (PLF) no difference between groups. At the time of follow up, no difference in functional outcome (DPQ) concerning, daily activities, work leisure, Anxiety/ depression and social interest was found. According to ODI no difference between intervention groups was found. This was also the case using SF-36 scores. When asked after 8.6 years, if they would go through the operation, with now a days knowledge 76 % overall answered yes 75 % (TLIF) and 77 % (PLF). Conclusions: In a long term perspective TLIF does not seem to improve functional outcome in ODI, SF-36, Dallas pain Questionnaire, Low back pain rating scale compared to an Instrumented posterolateral Lumbar fusion (PLF). Disclosures: author 1: none; author 2: none; author 3: none; author 4: grants/research support; Company = Danish Strategic Research Council.

80 GUIDELINES FOR NON-SURGICAL AND SURGICAL MANAGEMENT OF OSTEOPOROTIC VERTEBRAL BODY FRACTURES Thomas Roger Blattert, Klaus John Schnake, Oliver Gonschorek, and the Working Group Osteoporotic Fractures of the German Trauma Society Schwarzach Orthopedic Clinic; Scho¨n Klinik Nu¨rnberg Fu¨rth; BGU Murnau, Germany Introduction: Treatment options for surgical and non-surgical management of osteoporotic vertebral body fractures are widely differing. Based on their classification for osteoporotic fractures (OF classification), the spine section of the German Society for Orthopaedic and Trauma Surgery has now introduced general treatment guidelines. Methods: 707 clinical cases from 16 hospitals were evaluated. An OF classification-based score was developed to guide in the option of surgical vs. non-surgical management. For either choice, differentiated treatment guidelines were deduced. Diagnostic prerequisites for reproducible treatment recommendations were defined: Conventional X-rays with consecutive follow-up images (standing position whenever possible); MRI-; and CT-scan. OF-classification allows for upgrading of fracture severity during the course of radiographic follow-up. The actual classification type is decisive for the therapeutic recommendation. Results: A score of less than 6 points advocates non-surgical management; more than 6 points recommend surgical management. The primary goal of treatment is fast and painless mobilization. Due to expected comorbidities in this age group, minimally invasive procedures are being preferred. As a general rule, stability is more important than motion preservation. It is mandatory to restore the physiological loading capacity of the spine. If the patient was in a compensated unbalanced state at the time of fracture, reconstruction of the individual pre-fracture sagittal profile is sufficient. Instrumentation technique has to account for compromised bone quality. We recommend the use of cement augmentation or high purchase screws.

123

S704 The particular situations of injuries with neurological impairment; necessity to fuse; multiple level fractures; consecutive and adjacent fractures; fractures in ankylosing spondylitis are being addressed separately. Discussion: The therapeutic guidelines presented here provide a reliable and reproducible basis to decide for treatment choices available. However, intermediate clinical situations remain with a score of 6 points allowing for both non-surgical and surgical options. Because of that, individualised treatment decisions may still be necessary. In a next step, the guidelines presented will be further evaluated in a multicenter controlled clinical trial. Disclosures: author 1: consultant; Company = Aesculap, Medtronic, Spontech, other financial report; Company = AOSpine; author 2: consultant; Company = Expanding Orthopedics, Otto Bock; author 3: consultant; Company = Aesculap, Medtronic.

81 SELECTIVE APPLICATION OF INTRAWOUND VANCOMYCIN POWDER WITH USE OF FIBLIN GLUE AND/OR INTRAVENOUS DAPTMYCIN FOR OPEN POSTERIOR THORACIC/LUMBAR ARTHRODESIS Kotaro Satake, Tokumi Kanemura, Hidetoshi Yamaguchi, Naoki Segi Konan Kosei Hospital, Konan, Japan Introduction: We have applied intrawound vancomycin powder with use of fiblin glue (IW-VPF) and/or intravenous daptomycin (IV-DAP) only for surgical site infection (SSI) high risk patients since 2012. This prospective cohort at a single institute was reviewed retrospectively. Methods: We investigated 446 consecutive patients who underwent an instrumented open posterior arthrodesis of the thoracic/lumbar spine. For prophylaxis, in Group C (2011.7–2012.8, n = 148), all cases received only intravenous cefazoline (IV-CEZ). In Group D (2012.9–2013.10, n = 180), patients with severe diabetes (HbA1c [7.1 % or possessing retinopathy/nephropathy), obesity (BMI [30 kg/m2), steroid use, and with a previous same site surgery were selected as the candidates for IV-DAP (n = 29). Furthermore, the cases with operation time longer than 4 h received IW-VPF (n = 24). In Group V (2013.11–2014.12, n = 118), all patients met the aforementioned criteria received IW-VPF (n = 59). IV-CEZ was applied for others. Postoperative deep SSI was determined based on CDC definition. The prevalence, incidence of microorganisms, and clinical outcome of the deep SSI cases were compared among these three groups. Results: There was no significant difference among the three groups in terms of patient demographic. The postoperative deep infection rate was 6.1 % in Group C, 1.6 % in Group D, and 0 % in Group V. In Group D and V, the SSI incidence was reduced significantly compared with Group C (p = 0.012). There were 7 incidences of MRSA and 2 of MRCNS in Group C, 1 of MRSA (IV-CEZ applied case) and 2 gram-negative rods (IV-DAP and IW-VPF applied cases) in Group D. One case required instrument removal and one patient died in the postoperative course in Group C. On the other hand in Group D infections were controlled with irrigation. Conclusion: Selective application of IW-VCM or IV-DAP significantly reduced deep SSI incidence. Because there were no deep SSI cases in Group V, IW-VCM is likely to have better prophylactic effect than IV-DAP.

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Disclosures: author 1: none; author 2: consultant; Company = Madtronic, DePuy Synthes, AOSpine, NuVasive; author 3: none; author 4: none.

82 TWO-LEVEL CERVICAL DISC ARTHROPLASTY VS. ACDF: A PROSPECTIVE, RANDOMIZED, CONTROLLED MULTICENTER CLINICAL TRIAL WITH 5-YEAR RESULTS Matthew F. Gornet, Todd Lanman, Jeffrey McConnell, J. Kenneth Burkus, Randall Dryer, Scott Hodges The Orthopedic Center of St. Louis, USA Introduction: Cervical disc arthroplasty (CDA) is intended to treat symptomatic cervical degenerative disc disease while maintaining motion at the treated level. ACDF has been the standard treatment for patients with cervical radiculopathy and/or myelopathy who remain unresponsive to nonoperative measures, but at the expense of lost motion. Few studies have focused on the safety and efficacy of CDA for treatment at 2 levels. At 24 months, the low-profile, titanium ceramic composite ball-in-trough cervical disc showed superior overall success results vs. 2-level ACDF. This study compared patient outcomes 5 years after surgery. Methods: A prospective, randomized, controlled multicenter FDA IDE clinical trial was conducted at 30 US centers. Patients with cervical degenerative disease at 2 contiguous levels from C3–C7 were randomized and treated with CDA (n = 209) or ACDF (n = 188). Patient outcome measures included NDI, numerical pain scales for neck pain and arm pain, SF-36, and radiographic outcomes. Patients were followed at regular intervals for 24 months, and again 36 and 60 months after surgery. Overall success was defined as: Neck Disability Index (NDI) Success (improvement C15 points); maintenance or improvement of neurologic status; no implant- or surgical-related adverse events; and no additional surgical procedures at an index level. Analysis will be confirmed by an independent team of biostatisticians. Results: CDA patients demonstrated statistical superiority over ACDF patients at 5 years based on overall success (79.2 vs 65.7 %, posterior probability of superiority = 99.6 %), NDI success (88.7 vs 77.4 %, posterior probability of superiority = 99.6 %), and SF-36 PCS Success (88.3 vs 79.7 %, posterior probability of superiority = 98.0 %). Implant and surgical adverse event rates up to 5 years after surgery were similar for CDA and ACDF patients (23.2 vs 24.7 %). Subsequent surgeries at an index level were significantly lower in the CDA group (4.2 %) than ACDF (11.5 %), with incidence of grade 3 or 4 heterotopic ossification of 3.6 and 8.4 %, respectively. Conclusion: The results of this long-term study represent Level I clinical evidence in support of cervical disc arthroplasty at two contiguous levels as a motion-sparing alternative to ACDF in appropriately selected patients. Disclosures: author 1: consultant; Company = K2M, Medtronic, stock/shareholder; Company = Bonovo; International Spine & Orthopedic Institute, LLC; Nocimed; OuroBoros; Viscogliosi Bros Venture Partners LLC, royalties; Company = Medtronic; author 2: consultant; Company = Stryker, Medtronic, royalties; Company = Medtronic; author 3: consultant; Company = stock/ shareholder; Company = royalties; Company=; author 4: grants/research support; Company = Medtronic, consultant; Company = Medtronic; author 5: royalties; Company=; author 6: grants/research support; Company = medtronic.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 NEW TECHNIQUES, IMAGING, PATIENT SAFETY, INFECTION, COMPLICATIONS

83 MULTICENTER, RANDOMIZED PLACEBO-CONTROLLED CLINICAL TRIAL TO EVALUATE THE EFFECT OF PERIOPERATIVE USE OF TRANEXAMIC ACID ON TRANSFUSION REQUIREMENTS AND SURGICAL BLEEDING IN MAJOR SPINE SURGERY Juan Bago, MJose Colomina, Frederic Font, Javier Pizones, Salvador Fuster, Ferran Pellise Hospital Vall d’Hebron, Barcelona, Spain Introduction: The effectiveness of TXA in major spine surgery has been suggested but never proven in a RCT. (Level I evidence proving the effectiveness of TXA in major spine surgery is still lacking). Hypothesis: Intraoperative tranexamic acid (TXA) is effective and safe in reducing perioperative blood loss in adult patients submitted to major spinal surgery. Design: Prospective, randomized, double-blind, placebo control study to evaluate the effects of TXA in complex spinal surgery. Primary outcomes were intraoperative and postoperative blood loss. Secondary outcomes included transfusion rate, length of hospital stay and the safety profile of TXA in the postoperative period and at mid-term follow-up (6 weeks). Main comparisons were performed using mixed models with logtransformed values of intraoperative and total bleeding, adjusted by age, BMI and number of fused levels. Results: Ninety five patients scheduled for major spinal surgery were randomized: 44 received TXA (10-mg/kg as a bolus or as a fast, 20-min intravenous infusion before the start of surgery, followed by perfusion of 2 mg/kg/h up to completion of surgery) and 51 received placebo. Groups did not differ in duration of surgery (TXA/placebo) (297 vs 300 min), levels fused (5.5 vs 6) and length of hospital stay (10 days vs 11 days). TXA reduced intraoperative (869.1 vs 1228.8 cc, p = 0.02) and total blood loss (1591.7 vs 2052.3 cc, p = 0.03) compared with placebo. 48 % patients in the TXA group didn’t require any transfusion compared to 33 % in the placebocontrol group (p = 0.05). The average number of units transfused were 1.14 in TXA group and 2.05 units in control group (p = 0.05). The occurrence of adverse events was similar in both groups: 2 patients in TXA group and 1 patient in control group presented an episode of deep venous thrombosis (DVT). Conclusions: For the first time, this study shows in a RCT that TXA significantly reduces blood loss and blood transfusion requirements in patients undergoing major spinal surgery, with a safety profile. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: grants/research support; Company = DePuy Synthes, K2M.

84 PATIENT REPORTED VALUE 1 YEAR AFTER SURGERY FOR LUMBAR DISC HERNIATION: PREDICTORS OF OUTCOME USING THE SWEDISH NATIONAL SPINE REGISTER; SWESPINE Peter Fritzell, Olle Ha¨gg, Bjo¨rn Stro¨mqvist, Bjo¨rn Knutsson Neuro-orthopedic clinic, Jo¨nko¨ping, Sweden; Spine Center Go¨teborg, Sweden; Orthopedic clinic, Sundsvalls sjukhus, Sweden; Lund University Hospital, Sweden

S705 Background: Patient reported value has been used in the national Swedish register, Swespine, since two decades. Using registries, it should be possible to predict outcome after different surgical procedures with respect to specific ‘‘patient profiles’’ at baseline. The predicted outcome could be used as one decision tool whether to perform surgery or not, and surgical units could be compared also using case-mix ‘‘adjusted outcome’’. Methods: We used the Swedish national registry, Swespine, with [22,000 patients reporting 1-year outcome after surgery for LDH. Value was defined from the patient perspective using PROM (patient reported outcome measure), here a five-graded scale ‘‘Global Assessment’’ (GA), where the patient reports change in pain. Outcome was assessed by statistical case-mix analysis using a computer model where different baseline variables were included, and the predicted approximate outcome for each patient was estimated. The same analyses were used on a surgical unit level, enabling adjusted comparisons on a population level. The base-line variables used were: age, gender, no of operated levels, education, working, smoking, previous back surgery, pain intensity (VAS) and duration, function (ODI), walking distance and comorbidity. Results: 1. Example 1. Female, 45, high education, non-smoker, no previous surgery, no comorbidities, pain 75/100 for 3–12 months; there is 90 % chance of reporting ‘‘Pain free/Much better’’, a 9 % chance of reporting ‘‘Somewhat Better/Unchanged’’, and a 1 % risk of reporting ‘‘Worse’’. Example 2. Female, 45, low education, smoking, no previous surgery, some co-morbidity (Elixhauser 2), pain 75/100 for 3–12 months; there is a 68 % chance of reporting ‘‘Pain free/Much better’’, 26 % of reporting ‘‘Somewhat better/Unchanged’’, and 6 % of reporting ‘‘Worse’’. 2. Surgical unit. Those units reporting significantly better results compared with the average mean using unadjusted values, in most cases fell within the average mean CI if consideration were taken to the patient profiles at baseline. Conclusions: 1. Using case-mix adjusted register data analysis, the predicted outcome could be used as decisions support whether to operate or not. 2. In comparing results from different surgical units it is important to adjust for documented baseline predictors. Disclosures: author 1: none; author 2: stock/shareholder; Company = Spine Centre Go¨tebrog AB, employee; Company = Spine Centre Go¨teborg AB; author 3: none; author 4: none.

85 PEDICLE SCREW MALPOSITION IN REVISION SPINE SURGERY: EFFICACY OF INTRA-OPERATIVE CT BASED NAVIGATION John Street, Harvey Wu, Andrew Pennington, Juliet Batke, Kaiyun Yang, Charles Fisher, Marcel Dvorak University of British Columbia, Vancouver, Canada Hypothesis: Intra-operative navigation would result in increased pedicle screw accuracy compared to non-navigated techniques. Design: Ambispective matched cohort of patients undergoing revision spinal instrumentation. Introduction: The improved screw accuracy of intra-operative conebeam CT navigation has been primarily reported in cadaveric studies and spinal deformity surgeries. Revision surgery poses unique technical challenges for pedicle screw instrumentation because of an established fusion mass, lack of reliable anatomical landmarks and limited fixation options.

123

S706 Methods: We identified patients who underwent revision instrumented thoracolumbar spine surgery using intra-operative cone-beam CT navigation (NAV) between January 1, 2008 and December 31, 2012. A historical control cohort was matched from patients who underwent revision surgery between January 1, 2006 and December 31, 2008 using non-navigated techniques (nonNAV). Results: 52 patients underwent revision surgery with intra-operative navigation and 30 without navigation. A total of 1074 pedicle screws were instrumented: 752 NAV and 322 nonNAV. A significant difference existed in the number of misplaced screws 28 and 49 respectively (P \ 0.001). Mean number of misplaced screws per case was 0.54 (SD = 0.92) NAV and 1.69 (SD = 2.44) nonNAV (P = 0.02). Number of screws revised intra-operatively was not significant (10 vs. 7, P = 0.50). One NAV patient and one nonNAV patient required early post-operative screw revision during the same admission (P = 1.00). No difference was observed in grade of screw malposition (P = 0.34), anatomical location of malposition (P = 0.26), duration of surgery (P = 0.12), incidence of intra-operative dural tear (P = 0.70), wound infection (P = 0.30) or length of stay (P = 0.73). A significant difference in intra-operative massive ([2 L in 3 h) blood loss existed; 5.8 % of NAV cases compared to 23.33 % of nonNAV cases. Conclusion: This early analysis of revision surgery demonstrates an increased accuracy of pedicle screw placement utilizing intra-operative cone-beam CT navigation without an increase in operative time. Clinical outcomes between NAV and nonNAV cases were similar. Disclosures: author 1: grants/research support; Company = MEDTRONIC; author 2: none; author 3: none; author 4: none; author 5: grants/research support; Company = MEDTRONIC; author 6: grants/research support; Company = OREF, consultant; Company = Medtronic, Nuvasive, royalties; Company = Medtronic; author 7: grants/research support; Company = Medtronic, consultant; Company = Medtronic, royalties; Company = Medtronic.

86 DEVELOPMENT, CHARACTERIZATION, AND IN VITRO ASSESSMENT OF ADDITIVE MANUFACTURED BIOMIMETIC IMPLANT SURFACES Regina MacBarb, Derek Lindsey, Chelsea Bahney, Shane Woods, Mark Wolfe, Scott Yerby SI-BONE, Inc., San Jose´, USA Introduction: Chronic low back pain is a public health epidemic that, when conservative treatments fail, may require surgical fusion to relieve pain. Successful fusion depends on implant osseointegration, which requires appropriate implant surfaces to promote biological fixation. Despite recent advances, host bone typically incorporates only 15–40 % of the available implant surface area. Additive manufacturing allows for fabrication of complex implant geometries and surfaces mimicking bone morphology that cannot be achieved via traditional machining. This study developed additively manufactured (AM) trabecular-like titanium implant surfaces and assessed how bone cells responded to them; results were compared to a commonly used titanium plasma spray (TPS) coating. Methods: Ti6Al4V ELI discs were designed with a 0.75 mm trabecular topology, and 3 physiological bone porosities (60, 65, 70 %) were targeted by altering pore size (ideally C300 lm to enhance osseointegration). The 3 designs were manufactured using electron beam melting technology. For comparison, machined Ti6Al4V ELI discs were coated with 0.75 mm thick TPS. Human osteoblasts (hFOB 1.19) were seeded onto discs and maintained in base medium

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 plus L-ascorbic acid and beta-glycerol phosphate. Initial cell attachment (t = 4 h), proliferation (t = 2, 7, 14, 21d), and intracellular alkaline phosphatase (ALP) activity, an early osteogenic marker, (t = 7, 14d) were measured. Results: The mean porosities of the AM groups were 60, 63, and 71 %, respectively, while TPS discs measured at 59 %. Mean pore sizes for the AM groups were 291, 281, and 357 lm, respectively; mean pores in the TPS coating were 142 lm. The AM 65 % group had significantly more cells attached at t = 4 h compared to TPS discs, with no difference among AM groups. At t = 2d and 7d, significantly higher cell numbers were found on all AM groups compared to TPS discs, with no significant differences among groups at t = 14d. By t = 21d, TPS discs had significantly more cells compared to all AM groups. AM discs collectively had 20–30 % (t = 7d) and 10–20 % (t = 14d) greater ALP levels compared to TPS discs, although results were not significant. Conclusions: Results demonstrate that additive manufacturing allows for controlled trabecular-like surfaces that allow for earlier cell proliferation and trend towards higher ALP activity than TPS coating. These initial findings suggest the AM surfaces may induce a more osteogenic environment that could induce accelerated bone formation. Disclosures: author 1: stock/shareholder; Company = SI-BONE, Inc., employee; Company = SI-BONE, Inc.; author 2: stock/shareholder; Company = SI-BONE, Inc., employee; Company = SIBONE, Inc.; author 3: consultant; Company = SI-Bone; author 4: consultant; Company = MPI Research; author 5: consultant; Company = MPI Research Inc.; author 6: stock/shareholder; Company = SI-BONE, employee; Company = SI-BONE.

87 GLOBAL TILT: A SINGLE PARAMETER INCORPORATING THE SPINAL AND PELVIC PARAMETERS CORRELATES WITH HEALTH-RELATED QUALITY OF LIFE PARAMETERS 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; Spine Unit, Hospital Valle Hebron; Hospital De La Paz, Madrid; Orthopedic Spine, Ankara Spine Center; CHU Bordeaux Pellegrin Hospital; Department of Orthopaedics and, Schulthess Klinik; Fundacio´ Institut de Recerca Vall Hebron, Barcelona, Spain Summary: This study demonstrates that Global Tilt, a novel radiographic measurement, correlates with health-related quality of life(HRQoL) parameters. Hypothesis: Global tilt correlates with HRQoL parameters. Design: Retrospective analysis of a multicenter, prospective, consecutive patient series. Introduction: Surgical goals and alignment objectives mostly depend on sagittal plane parameters. SVA, C7 tilt and T1 tilt account for spinal balance and PI, PT, SS, PI-LL account for pelvic compensatory mechanism, which are all affected by patient positioning. GT is the angle between the line drawn from the center of C7 to the center of the sacral endplate and a line drawn from the center of the sacral endplate to the center of the femoral head. GT is a single parameter that takes both balance and pelvic compensation into account as a single parameter. GT has been shown to be less affected by patient positioning compared to SVA and PT. Aim was to analyze the correlation of GT with HRQoL parameters and compare to other sagittal plane parameters.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Methods: A retrospective analysis of a multicenter, prospective, consecutive patient series. 337 patients (285F, 52M) with adult spinal deformity were included, (mean age: 58.2 ± 15.7). Pelvic parameters (PT, PI, SS), sagittal balance parameters (SVA, T1 tilt), GT and lordosis gap(LGap) were measured by using Surgimap. HRQoL instruments included: Oswestry Disability Index(ODI), Short Form36, SRS-22. Correlation analysis between radiographic pelvic and sagittal balance parameters, GT and LGap was pursued. Kruskal– Wallis test was used to compare GT, SVA and PT for ODI \20, 20–40 and [40 disability groups. Results: Global tilt significantly correlated with ODI, SRS-22 pain, function, self image, and subtotal scores and SF-36 PCS (p \ 0.01) similar to SVA, PT and LGap. None of the radiographic parameters correlated with SRS-22 mental health or SF-36 MCS. GT was more sensitive in detecting ODI disability groups when compared to SVA and PT (p \ 0.01). Conclusion: GT has a similar or better correlation with HRQoL domains compared to PT and SVA. This information adds another important feature of this parameter in addition to being a single parameter taking into account both the spinal balance and pelvic compensation and being less affected by patient positioning. 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: no indication; author 6: grants/research support; Company = Medronic, 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.

88 REDUCING ROD BREAKAGE AND NONUNION IN PEDICLE SUBTRACTION OSTEOTOMY: THE IMPORTANCE OF ROD NUMBER AND CONFIGURATION IN 264 PATIENTS WITH 2-YEAR FOLLOW-UP Virginie Lafage, Munish Gupta, Jensen Henry, Frank Schwab, Christopher Ames, Eric Klineberg, Justin Smith, Christopher Shaffrey, Douglas Burton, International Spine Study Group International Spine Study Group Introduction: Pedicle subtraction osteotomies (PSO) can provide substantial realignment, but are associated with nonunion and rod failure. The addition of supplementary rods and interbody fusion (IBF) may decrease the failure rate. Methods: Adult spinal deformity patients from a consecutive retrospective database with at least 1 lumbar PSO and 2-year follow-up were included. Demographic, operative, and radiographic data were collected. The number of rods across the PSO site was reviewed; more than 2 rods were categorized as accessory (A: connected to primary rods) or satellite (S: independently anchored). Potential risk factors were evaluated for PSO site failure (rod breakage or revision for nonunion). Results: From 264 patients included, there were 190 with 2 rods (2R), 36 with 3R, and 38 with 4R. There were no differences in demographics or sagittal alignment across groups. 2–3R had a trend of higher rates of failure (28, 29 %) than 4R (18 %; P = 0.128). The 3–4R patients had significantly fewer revisions for instrumentation failure and/or pseudarthrosis than the 2R group (15 vs. 26 %; P = 0.035). In 3–4R constructs, there were 45 A rods (61 %) and 29

S707 S rods (39 %). S rods had significantly lower failure rates than A rods (10 vs 31 %; P = 0.034); accessory rods were similar to 2R (31 vs 29 %; P = 0.452). In 3–4R constructs, larger rod diameter (C6.0 mm) failed significantly less than 4.5–5.5 mm rods (5 vs 33 %; P = 0.011). In 2R constructs, titanium rods had a significantly higher rate of breakage/nonunion (44 %) than cobalt chrome and stainless steel (25, 24 %; P = 0.037). There was no difference in failure by material in the 3–4R constructs (P = 0.127). Interbody fusion (IBF) at the PSO level resulted in fewer failures overall (21 vs 33 %; P = 0.046), and for 2R (22 vs 34 %; P = 0.112), 3–4R (16 vs 31 %; P = 0.164), and S (0 vs 27 %; P = 0.050) constructs. Conclusion: This study confirms a high rate of nonunion and rod breakage in the 2 years following lumbar PSO surgery. The lowest rates of rod failure/nonunion were found in constructs with satellite rods, IBF adjacent to the PSO, and larger diameter rods. 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: grants/research support; Company = Depuy medicrea, consultant; Company = Depuy Orthofix, royalties; Company = Depuy; author 3: none; author 4: none; author 5: grants/ research support; Company = depuy synthes, issg, consultant; Company = stryker medtronic depuy synthes, stock/shareholder; Company = Baxano Surgical, Doctor’s Research Group, royalties; Company = stryker biomet, employee; Company = ucsf; author 6: grants/research support; Company = AO Spine, Depuy, OREF, other financial report; Company = Speakers Fees Depuy AO Spine; author 7: grants/research support; Company = Depuy-Synthes, ISSGF, consultant; Company = Biomet, Nuvasive, Cerapedics, Medtronic, royalties; Company = Biomet; author 8: grants/research support; Company = Depuy-Synthes, consultant; Company = Biomet, Medtronic, Nuvasive, Stryker, stock/shareholder; Company = Nuvasive, royalties; Company = Biomet, Medtronic, Nuvasive; author 9: grants/research support; Company = Depuy, consultant; Company = Depuy, royalties; Company = Depuy; author 10: grants/ research support; Company = J&J, MSD, Inovasis, Stryker, Biomet.

89 IS THE C7 SAGITTAL VERTICAL AXIS (SVA) THE BEST RADIOGRAPHIC MEASURE TO PREDICT CLINICAL OUTCOMES IN ADULT SPINAL DEFORMITY? Yong-Chan Kim, Lawrence G. Lenke, Young-Woo Kim, Sirichai Wilartratsami, Cheol-Jung Yang Department of Orthopaedic Surgery, Hallym University Sacred Heart Hospital, Hallym University School of Medicine, Anyang-si, Korea Purpose: The goal of the study is to analyze the clinical correlation of the head sagittal vertical axis (hSVA), as measured from the head center of gravity, to health related quality of life (HRQOL) outcomes in adult spinal deformity (ASD). Secondarily, we sought to compare the correlation of hSVA and C7 SVA. We hypothesize that the hSVA will better predict HRQOL outcomes in ASD than C7 SVA. Methods: 108 consecutive ASD patients were evaluated using the EOSÒ 2D/3D radio-imaging device. A vertical plumb line from the head center of gravity was utilized to measure the distance to the posterior corner of S1 (hSVA-S), to the hip center (hSVA-H), to the knee center (hSVA-K), and to the ankle center (hSVA-A), as well as measuring the standard C7 SVA. We analyzed the correlation between each hSVA parameter with the Oswestry Disability Index (ODI) & Scoliosis Research Society form (SRS-22r) with Pearson’s correlation coefficients. A univariate linear regression was utilized to

123

S708 determine the effect of each parameter on HRQOL scores. We also compared the relationship of each hSVA measure and C7 SVA with the ODI & SRS-22r using multivariate linear regression. Results: All 4 hSVA measures demonstrated strong correlation with the ODI (r = 0.331–0.514) and SRS-22 domains (pain, r = -0.331 to -0.436; self-image, r = -0.290 to -0.363; function, r = -0.299 to -0.550) and total score (r = -0.382 to -0.499). Of note, hSVA-A (Global SVA) also strongly correlated with the SRS satisfaction subscore (r = -0.429). Univariate linear regression showed similar results. The strongest predictor of outcomes was hSVA, not C7 SVA; (hSVA-H for SRS total score & the pain & self-image domains; hSVA-S for function; Global SVA for satisfaction). Conclusion: The clinical correlation and effect on outcome scores of hSVA measures are validated. Global SVA has an especially strong correlation with ODI & all the SRS subscores except for mental health. Our study confirms that hSVA is a stronger predictor of clinical outcomes than the C7 SVA in adult deformity patients. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 Conclusion: Predicting medical complications has proven to be realistic and clinical helpful in daily routine. In conclusion it is very applicable in clinical practice and assists in the preoperative decision making process as well as in counselling patients about their medical complications risk. Disclosures: author 1: none; author 2: none.

91 INCIDENCE OF COMPLICATIONS AND ADVERSE EVENTS AFTER VIDEO ASSISTED THORACIC SURGERIES (VATS) AT THE SPINAL COLUMN: RESULTS OF A SYSTEMATIC RETROSPECTIVE ANALYSIS OF 282 CASES Daniel P. J. Sauer, Franziska Heider, Raphael D’Agostini, Imme Haubitz, Andreas Korge, Christoph Mehren, Michael Mayer, Christoph Siepe Schoen Clinic Munich Harlaching, Munich, Germany

90 PREDICTING MEDICAL COMPLICATIONS TO ENHANCE SAFETY OF SPINE SURGERY Maximilian Kasparek, Josef Grohs Department of Orthopaedic Surgery, University of Vienna, Austria Introduction: The preoperative evaluation of medical complications before spine surgery is gaining in importance nowadays. Lee et al. (Spine 2014;14:291–299) created the first validated model for predicting medical complications after spine surgery. This new tool was tested within the daily clinical practice in order to improve patient safety and to assess whether the forecasted evaluation would prove to be correct. Materials and methods: All medical diagnoses were assessed prospectively from each patient who underwent spine surgery. The risk of medical complications was estimated with this new online calculator, which is named SpineSageTM, prior to surgery. Within this process certain criteria are regarded: COPD, hypertension, renal conditions, neoplasm, anaemia, diabetes, congestive heart failure and previous cardiac complications. The Surgical Invasiveness Index counts the number of levels, that are decompressed, fused, or instrumented, posteriorly, and anteriorly. This score ranges from 0 to 48 with a higher score indicating higher invasiveness. Given all this information SpineSage then predicts any major and minor possible medical complication, the risk of infection and dural leak. Results: 100 patients have been assessed since April 2014, with an average age of 59.3 years (range 15–86) and a BMI of 27.5 (range 18.6–40.1). The mean Surgical Invasiveness Score was 7.9 points (range 1–31). The mean risk for all patients for major complications was 7.2 %; for minor complications 20.8 %; for infection 7.0 % and for dural leak 9.1 %. In total medical complications were reported in 23 % of the patients and in 4 % occurred a major complication. Infection was seen in 4 % and dural leak in 1 %. Furthermore, patient groups were compared with different preoperative risks. In the patient group with a medical risk under 15 %, the mean risk was 10.7 % and in 5/50 patients (10 %) a complication was reported. In the risk group between 15 and 30 % the mean medical risk was 20.8 % and 5/27 patients (18 %) suffered from complications. In the high risk group with more than 30 % preoperative risk the mean risk was 42.5 % and in 10/23 patients (44 %) complications were reported.

123

Introduction: In the early 1990s, the popularity of minimally invasive procedures such as the video assisted thoracic surgery (VATS) has increased immensely. VATS is a procedure with minimal tissue trauma and a broad spectrum of indications (e.g.: fractures, deformity, degeneration, tumor). The use of a validated classification system of complications has not been published for VATS yet. The study aim is the retrospective analysis of the incidence and degree of perioperative complications after VATS based on the Dindo (1) and Rampersaud Severity of Adverse Events (2) score systems and the utilization of this data in further clinical studies. Methods: 282 cases (average age 62 {15–86} years, n = 170 female, n = 112 male) with thoracoscopic procedures from 2008–2013 were included. The analysis of perioperative (60d) complications were based on the classification score 1 and 2. The present data were correlated with the kind, scope and degree of intervention and the patients’ demographic. Results: The overall complication rate of anterior procedures utilizing both scoring systems was 71 % (201/282). The complication rates Grade I–II in Dindo anterior procedures were 78 and 85 % in the Rampersaud classification system. Evaluation of anterior procedures for Grade III–IVb, requiring surgical intervention, shows a complication rate of 15.2 % using (1) scoring and 14.5 % using (2) grade II-III scoring. In 6 % of the cases (n = 18) a Re-Operation took place in the first 5 years postoperative. The mortality rate was 0.7 % (n = 2/282) grade V in Dindo and grade IV in Rampersauds classification system. Discussion: The data of this study represents an analysis of complications after VATS. Using this scoring system, 78 % of complications that occurred could be treated without surgical interventions. And 28.8 % did not have a complication at all in both scoring systems. The described rate of Grade III-V complications and postoperative ReOperations compared to open procedures shows that the VATS is a safe and valuable operative technique. The study data shows that the surgical complications should be collected prospectively with validated scores for all complex spine surgeries. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: consultant; Company = Silony; author 6: consultant; Company = Silony, Depuy Synthes; author 7: none; author 8: consultant; Company = DePuy Synthes, Silony Medical.

Eur Spine J (2015) 24 (Suppl 6):S669–S710 92 DIFFUSION TENSOR IMAGING IN CERVICAL SYRINGOMYELIA SECONDARY TO CHIARI I MALFORMATION: PRELIMINARY RESULTS Huang Yan, Zezhang Zhu, Zhen Liu, Yong Qiu, Bangping Qian, Xu Sun The Spine Surgery, the Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing, China Background: Syringomyelia is a common finding in patients with Chiari I malformation (CMI). Conventional imaging techniques frequently fail to assist clinicians in quantitatively assessing the neural damage in these patients. The purpose of this study was to perform a quantitative analysis of syringomyelia secondary to CMI using diffusion tensor imaging (DTI) and to further analyze the correlations between DTI parameters and the severity of syringomyelia. Materials and methods: Twenty patients suffering from syringomyelia secondary to CMI were prospectively enrolled (Male:Female = 11:9, mean age, 14.7 years; range 8–25 years) from April 2012 to August 2013. Another cohort of 8 age- and gendermatched normal adolescents was enrolled to serve as the control group. Sensitivity encoding single-shot echo-planar imaging (EPI) was used for the sagittal DTI. Average Fractional anisotropy (FA) values in the spinal cord were compared between the vertebral levels at the syrinx and the levels surrounding the visible syrinx cavity on T2-weighted images. To investigate the correlation between the FA values and the size of syrinx, ten patients with a distended syringomyelia constituted the D-syrinx group, whereas 10 patients with a non-distended syrinx were assigned into the ND-syrinx group. The FA values in relation to the length of syrinx were also investigated. Results: Compared to the normal controls, The FA values were significantly decreased within the syrinx (0.423 ± 0.011 vs. 0.533 ± 0.007; p \ 0.001), while no significant decreased FA value was measured in the area surrounding the visible syrinx. Concerning CMI patients with different size of the syrinx, significantly decreased FA values within the syrinx were observed in patients with a NDsyrinx in comparison to those with a D-syrinx (0.397 ± 0.013 vs. 0.480 ± 0.018; p \ 0.001), whereas the FA values were similar in the tissue at levels surrounding the syrinx in the two groups. In addition, the FA value within the syrinx was both significantly decreased in the symptomatic and non-symptomatic groups when compared with the control groups and there was also a significant difference between the two CMI-S groups (0.371 ± 0.018 vs. 0.479 ± 0.011; P = 0.003). Conclusions: Decreased FA values within and caudal to the area of syringomyelia may provide evidence of increased microstructural damage within the spinal cord at these sites. FA values are significantly related to syrinx size and neurological signs/symptoms. Disclosures: author 1: none; author 2: none; author 3: none; author 4: none; author 5: none; author 6: none.

S709 Cervical spondylotic myelopathy (CSM) is the leading cause of the spinal cord (SC) dysfunction worldwide, a major healthcare concern and socioeconomic burden. There are many studies attempting to detect the best surgical approach, but not any of them considered dynamic compression effect. The study was designed as comparison of 2 prospective cohorts operated on in one hospital by a single surgeon. Inclusion criterion was CSM classified as Nurick 3, 4, and 5. Exclusion criteria were MS, ALS, and toxic-metabolic myelopathies. Patients were allocated randomly. In the trial (dorsal) group patients underwent open door laminoplasty (ECL) and in the control group (ventral) corpectomy/corpectomies with fusion (ACCF). The differences between preoperative and one-year follow up mJOA and walking test (WT) constitute clinical outcome. The 2-D spinal cord silhouettes from C2 to T1 were taken from dynamic MRs and transformed into 3-D reconstructions. Differences between preoperative and follow-up SCenvelope areas (EA) were assessed. For statistical analyses chisquere, student-t, Fischer Exact, Mann–Whitney, and Bhapkar’s tests and Pearson correlation coefficient were used. From September 1, 2008 to December 31, 2014, 61 patients met inclusion criterion. Four patients were excluded and 57 were allocated in the trial [N = 31] and the control [N = 26] group. There were no statistically significant differences (p [ .05) in male gender (67.7 %/ 84.6 %), age [years] (59.1 ± 11.7/55.6 ± 11.1), comorbidity (0: 15/14, 1: 8/13, 2: 3/4), baseline mJOA (10.7 ± 1.8/11.6 ± 1.6), WT steps (57.84 ± 31.53/62.34 ± 12.95), WT time (58.00 ± 31.56/ 62.43 ± 9.61), and SC-EA [mm2] (2949 ± 287.68/3042.85 ± 320.63) between ECL and ACCF groups, respectively. Postoperatively both groups improved significantly to all analyzed variables without differences between groups according to mJOA scale (3.48 ± 1.46/2.92 ± 1.13) and WT time (-18.84 ± 37.11/-21.00 ± 11.33). Decreases in number of WT steps (-13.35 ± 34.96/ -12.81 ± 9.89; z = 2.15, p = .031) and SC-EAs (-99.32 ± 37.56/-21.27 ± 56.79; z = 5.1, p \ .001) were significantly greater in the ECL group. This study is the first one to measure the difference between preoperative and postoperative SC-EAs. The decrease in SC-EAs served for assessment of the SC visco-elastic relaxation. The greater shrinkage of the SC-EAs alongside better improvement in walking ability in the ECL group suggest that surgical strategy can influence release of tension of the long tracks. Disclosures: author 1: none; author 2: none; author 3: none.

94 PATTERNS OF LUMBAR PAIN: A COST ANALYSIS APPROACH TO LUMBAR PAIN IN PRIMARY CARE IN OCCUPATIONAL MEDICINE. A COMPARATIVE STUDY OF TWO RETROSPECTIVE COHORTS OF 2375 PATIENTS DURING 2014 (I) Oscar Zabalza Mantilla2, Mikel Ayala Garcı´a2, Jaime Diaz de Atauri1

93 SURGICAL APPROACH CAN INFLUENCE POSTDECOMPRESSION RELEASE OF TENSION ON THE MYELOPATHIC SPINAL CORD LONG TRACKS: A DYNAMIC CERVICAL MRI 3-D RECONSTRUCTION ANALYSIS Marin Stancic, Ivana Stancic, Zvonimir Ivan Lubina Clinical Hospital Center Zagreb, Zagreb, Croatia; TIA, Copenhagen, Denmark; Clinical Hospital Merkur, Zagreb, Croatia

Clı´nica Ercilla Mutualia, Bilbao; 2Mutualia Alava, Vitoria-Gasteiz, Spain Materials and methods: We performed a retrospective study of two cohorts of patients treated for lumbar pain at our Worker’s compensation insurance company in 2014. The first group were managed according to pain patterns (G1),while the second group were not (G2) and they received conventional treatment. Diagnosis, number of sick days and mean duration, number of complementary tests, pharmacy cost, hospital admissions, surgical interventions, referrals to physical therapy (duration and type of therapy) were studied. A statistical

123

S710 analysis was performed using SPSS software, a Kolmogorov–Smirnov test was performed and subsequently a Chi squared and a Mann– Whitney’s U test were performed with a sensitivity of 95 %. Results: We reviewed 2375 cases: 1252 (G1) and 1123 (G2). Average sick leave was 14.84 days (G1) and 19.16 (G2) days. Inpatient bed days were 0.0112 (G1) and 0.0285 (G2), the inpatient bed/day cost was 9863.84€ (G1) and 26068.72€ (G2), the surgical procedures were 13 (G1) and 44 (G2), with a cost of 15480.38€ (G1) and 46828.21€ (G2), the complementary tests mean was 1.2875 (G1) and 1.1505 (G2) and a cost of 69322.24€ (G1) and 63442.42€ (G2), the patients that not received any test were 17.01 % (G1) and 26.27 % (G2) with a cost 527807.24€ (G1) and 491794.13€ (G2). All of them had a significant statistical difference (P \ 0.01). We found no significant differences in: Physical therapy referral and treatment (20.33 vs 21.97

123

Eur Spine J (2015) 24 (Suppl 6):S669–S710 mean days of treatment and 47.88 vs 55.02 mean therapy sessions) with a total cost of 62945.01€ (G1) and 62887.86€ (G2) and the pharmacy cost 34705.2€ (G1) and 37304.24€ (G2). The total cost in both groups were 1357491,19€(G1) and 1461560,29€(G2) with a patient cost 1084.26€ (G1) and 1301.48€ (G2). Conclusions: A marked difference in cost, sick leave duration, inpatient bed days, number of surgical procedures and complementary test was observed after patients were grouped according to pain patterns. In primary care, this is an effective way to indicate the duration of sick leave and treatment. These results have already modified treatment in G2, given the resulting financial and health care repercussions for our institution. Disclosures: author 1: none; author 2: none; author 3: none.