Preoperative and Perioperative Factors Effect on Adolescent Idiopathic Scoliosis Surgical Outcomes

Abstract
Prospective multicenter database. To identify factors associated with outcomes from adolescent idiopathic scoliosis (AIS) surgery outcomes and especially poor results. Because AIS is rarely symptomatic during adolescence, excellent surgical results are expected. However, some patients have poor outcomes. This study seeks to identify factors correlating with results and especially those making poor outcomes more likely. Demographic, surgical, and radiographic parameters were compared to 2-year postoperative Scoliosis Research Society (SRS) scores in 477 AIS surgical patients using stepwise linear regression to identify factors predictive of 2-year domain and total scores. Poor postoperative score patients (>2 SD below mean) were compared using t tests to those with better results. The SRS instrument exhibited a strong ceiling effect. Two-year scores showed more improvement with greater curve correction (self-image, pain, and total), and were worse with larger body mass index (pain, mental, total), larger preoperative trunk shift (mental and total), larger preoperative Cobb (self-image), and preoperative symptoms (function). Poor results were more common in those with Lenke 3 curve pattern (pain), less preoperative coronal imbalance, trunk shift and rib prominence (function), preoperative bracing (self-image), and anterior procedures (mental). Poor results also had slightly less average curve correction (50% vs. 60%) and larger curve residuals (31° vs. 23°). Complications, postoperative curve magnitude, and instrumentation type did not significantly contribute to postoperative scores, and no identifiable factors contributed to satisfaction. Curve correction improves patient's self-image whereas pain and poor function before surgery carry over after surgery. Patients with less spinal appearance issues (higher body mass index, Lenke 3 curves) are less happy with their results. Except in surgical patient selection, many of these factors are beyond physician control.

This publication has 26 references indexed in Scilit: