Abstract
Study DesignRetrospective comparative analysis.ObjectiveTo evaluate the impact of the spinal cord deformation rate (SCDR) on postoperative neurological deficits (PNDs) in spinal deformity surgery and to determine the optimal SCDR threshold for risk stratification.MethodsPatients undergoing corrective surgery between October 2010 and March 2022 were included. Spinal cord morphology at the apex was assessed using axial T2-weighted magnetic resonance imaging, categorizing spinal cords into type-A and type-B. Differences between the two groups were analyzed and the odds ratio (OR) for PNDs in patients with type-B was calculated. For patients classified as type-B, univariate analysis was conducted to identify variables, including SCDR, that may influence PNDs risk. Statistically significant variables were further analyzed using multivariate analysis. The receiver operating characteristic (ROC) curve and area under the curve (AUC) analyses were performed to evaluate the discriminative ability of risk factors and determine the optimal cut-off value.ResultsAmong 200 patients included, PNDs occurred in 17 cases (8.5%), predominantly in type-B (15/82 in type-B vs 2/118 in type-A, P < 0.001). Multivariate analysis revealed SCDR as an independent predictor of PNDs (OR = 1.076, P = 0.001). ROC analysis determined an optimal SCDR cutoff of 42.15% (AUC: 0.867), with 82.4% sensitivity, 88.5% specificity, and 98.2% negative predictive value (NPV). SCDR demonstrated superior predictive performance compared to the major Cobb angle (AUC: 0.776) and T-DAR (AUC: 0.769).ConclusionsFor patients undergoing corrective surgery, an SCDR of ≥42.15% serves as a strong predictor of PNDs. Its high NPV effectively excludes low-risk patients, offering improved risk stratification.