Surgical risk stratification based on preoperative risk factors in severe pediatric spinal deformity surgery

Study design Retrospective review. Objective The purpose of this study is to review the postoperative complications in pediatric patients undergoing spine surgery and to establish a preoperative classification that stratifies surgical risk and case difficulty. Summary of Background Data Pediatric spinal deformity (PSD) surgery can be challenging technically as well as economically. Often, a multidisciplinary approach to managing these patients is necessary. In an environment where resources are limited, such as in global outreach efforts, a method for stratifying PSD surgical cases can be useful for allocating appropriate resources and assigning appropriate skill sets in order to optimize patient outcomes and to streamline efforts. Materials and Methods A total of 145 consecutive PSD patients who underwent instrumented spinal fusion were reviewed. Radiographic measurements and demographic data were reviewed. A classification was established based on the curve magnitude, etiology, ASA grade, number of levels fused, the preoperative neurologic status, body mass index and type of osteotomies. Multiple regression analysis (MRA) and logistic regression analysis (LRA) were applied to indicate risk factors for complications. Results The average age was 14.3 years (10-20 years). The etiology was idiopathic scoliosis (n = 71), congenital scoliosis (n = 38), infectious (n = 11), and others. 23 patients had neurologic deficits preoperatively. Twenty-three patients had a posterior vertebral column resection. Patients were classified as Level 1 (n = 5), Level 2 (n = 19), Level 3 (n = 24), Level 4 (n = 58), and Level 5 (n = 39). Intraoperative neuro-monitoring changes were observed in 46 cases. Major complications were seen in 45 cases. A major complication consisted of implant related (n = 13), deep wound infection (n = 8), neurologic deficit (n = 7), death (n = 2), and others (n = 9). MRA demonstrated a significant correlation between classified level and %EBL/TBV, operative time, and complication rate. The risk level predicted the occurrence of general (odds ratio [OR] = 1.54; 95% confidence interval [CI] = 1.08-2.21; p =.019) and neurologic (OR = 3.34; 95% CI = 1.06-17.70; p =.036) complications. Osteotomy and resection procedures were independent predictors for postoperative neurologic complications (OR = 1.7, 95% CI = 1.11-2.85; p =.015). Conclusion Corrective spine surgery for complex pediatric deformity is challenging and carries a substantial risk. No single parameter appears to independently predict postoperative complications. However, when all risk factors are considered, there is a trend toward increased intraoperative electromonitoring change and postoperative neurologic risk with the higher level score in our classification. The newly established surgical risk stratification based on patient-specific clinical and radiographic factors can guide surgeons in their preoperative planning and surgical management of severe spine deformity in order to achieve optimal outcomes.