321. Evaluation of global alignment and proportion score in an independent adult spinal deformity database

BACKGROUND CONTEXT Sagittal spino-pelvic alignment has been associated with patient-reported outcome measures and mechanical complication rates. Linear numerical values of pelvic tilt and lumbar lordosis measurements with different pelvic incidences may be misleading. The use of “relative” measurements embedded in a weighted scoring of Global Alignment and Proportion (GAP) has been described. The purpose of our study was to evaluate the validity of the GAP score in an independent database. PURPOSE The purpose of our study was to evaluate the validity of the GAP score in an independent database. STUDY DESIGN/SETTING Retrospective review of an independent adult spinal deformity data base from a single center. PATIENT SAMPLE Adult spinal deformity patients from a single center who had >5 level fusion to the pelvis between 2004-2014. OUTCOME MEASURES Demographic, clinical, surgical and radiographic parameters. METHODS ASD patients who underwent ≥5 levels fusion to the pelvis between 2004 to 2014 were included. Demographic, clinical, surgical and radiographic patient characteristics were recorded. Cochran-Armitage tests were used to determine mechanical complication trends across GAP categories. Uni /multivariable logistic regression analyses were used to obtain crude and adjusted odds ratios of predictor (GAP categories) and the outcome (mechanical complication). The diagnostic performance of the GAP score was tested using the area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, negative predictive value and accuracy in predicting mechanical complications. RESULTS A total of 338 out of 657 patients (295F, 43M) with a mean age of 58±9.6 met the inclusion criteria. Mean follow-up was 55 months (24-138). Every mechanical complication was included from minor proximal compression fracture and iliac screw disengagement. Most common complications were rod failure in 25.4% (86/338) patients, 23 patients (6.8%) implant complication at the L-S junction, PJF in 45 patients (13.3%) and PJK in 20 pts (5.9%). Mechanical complication in proportioned (GAP-P), moderately (GAP-MD) and severely disproportioned (GAP-SD) patients were 23.8%, 55.7% and 66.1%, respectively. AUC for the GAP score was 0.653 (95% CI, 0.59 to 0.71, p CONCLUSIONS This study validates the efficacy of GAP Score in predicting mechanical complications in an independent database. A trend was observed in which lower GAP scores were associated with lower rates of mechanical complications. Both the crude and adjusted odds ratios were high, showing the extent of effectiveness of this predictive tool. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.