115. Discriminative ability of commonly used contemporary risk indices to predict adverse outcomes following adult spinal deformity corrective surgery

BACKGROUND CONTEXT It is imperative to determine which factors have greater implications on postoperative outcomes, which can afford tailored treatment plans for adult spinal deformity (ASD) patients. PURPOSE To determine the discriminative ability of commonly used indices to predict adverse outcomes after corrective surgery for adult spinal deformity. STUDY DESIGN/SETTING Retrospective cohort. PATIENT SAMPLE This study included 9,763 patients. OUTCOME MEASURES Perioperative complications. METHODS ASD patients from the National Surgical Quality Improvement Program (NSQIP) 2005-2015 were included. Logistic regression analysis determined significant odds ratios among complications between the following indices and demographics: Passias et al modified frailty index score for ASD (mFI-ASD), the modified Charlson Comorbidity Index (mCCI), ASA classification score, age and body mass index (BMI). Using multivariate analysis, indices and demographics that demonstrated significance for predicting complications were identified. CIT run forest analysis generated an index threshold value for all complications tested. RESULTS Included: 9,763 ASD patients. At least 4 of the 5 risk indices were significant for the following complications: any type, major, cardiac, infection and death. None of the indices correlated with reoperation or readmission. The mFI-ASD demonstrated the highest odds ratio (OR) for all complications (p <.001), followed by ASA status. Modified CCI also correlated with a higher OR for all five complications, compared to age and BMI. An index threshold value for each complication was determined by CIT run forest analysis. Analysis of threshold values showed mFI had the highest ORs for any complication (OR: 3.50) as well as infection (OR: 2.53). ASA status, on the other hand, had the highest ORs for major complications (OR: 2.93), cardiac complications (OR: 4.09) and death (OR: 10.18). CONCLUSIONS The modified FI-ASD demonstrated superiority in predicting adverse postoperative outcomes, compared to various commonly used indices and patient characteristics. These findings are important as it allows spine surgeons to appropriately counsel their patients preoperatively. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs. It is imperative to determine which factors have greater implications on postoperative outcomes, which can afford tailored treatment plans for adult spinal deformity (ASD) patients. To determine the discriminative ability of commonly used indices to predict adverse outcomes after corrective surgery for adult spinal deformity. Retrospective cohort. This study included 9,763 patients. Perioperative complications. ASD patients from the National Surgical Quality Improvement Program (NSQIP) 2005-2015 were included. Logistic regression analysis determined significant odds ratios among complications between the following indices and demographics: Passias et al modified frailty index score for ASD (mFI-ASD), the modified Charlson Comorbidity Index (mCCI), ASA classification score, age and body mass index (BMI). Using multivariate analysis, indices and demographics that demonstrated significance for predicting complications were identified. CIT run forest analysis generated an index threshold value for all complications tested. Included: 9,763 ASD patients. At least 4 of the 5 risk indices were significant for the following complications: any type, major, cardiac, infection and death. None of the indices correlated with reoperation or readmission. The mFI-ASD demonstrated the highest odds ratio (OR) for all complications (p <.001), followed by ASA status. Modified CCI also correlated with a higher OR for all five complications, compared to age and BMI. An index threshold value for each complication was determined by CIT run forest analysis. Analysis of threshold values showed mFI had the highest ORs for any complication (OR: 3.50) as well as infection (OR: 2.53). ASA status, on the other hand, had the highest ORs for major complications (OR: 2.93), cardiac complications (OR: 4.09) and death (OR: 10.18). The modified FI-ASD demonstrated superiority in predicting adverse postoperative outcomes, compared to various commonly used indices and patient characteristics. These findings are important as it allows spine surgeons to appropriately counsel their patients preoperatively.