Predictive modeling of surgical outcomes in lumbar stenosis and degenerative scoliosis using 3D gait-based spine-pelvic compensation analysis.

OBJECTIVE: To explore the clinical value of a surgical effect prediction model for patients with lumbar spinal canal stenosis and degenerative scoliosis (LSS-DS). The model is based on the spine-pelvis compensation state measured by a three-dimensional gait system. METHODS: A total of 215 patients with LSS-DS who underwent surgery from January 2022 to December 2024 were enrolled. They were randomly divided into a training set (n = 151) and a validation set (n = 64) at a 7:3 ratio. Spine and pelvis parameters were measured using a three-dimensional gait system. Multivariate logistic regression analysis was used to screen independent predictors of surgical effect, and a nomogram model was constructed. RESULTS: In the training cohort, 35 cases (23.18%) had suboptimal surgical outcomes, while the validation cohort showed 15 cases (23.44%) with unsatisfactory results (P = 0.872, χ2 = 0.006). Multivariate analysis identified the Cobb angle of scoliosis, preoperative sagittal vertical axis, pelvic incidence-lumbar lordosis difference (PI-LL), pace, step size, affected lower extremity support time proportion, and preoperative VAS score as independent risk factors (P < 0.05). The nomogram model had a C-index of 0.852 and 0.849 in the training and validation sets, respectively. The AUC values were 0.860 (95% CI: 0.768-0.953) and 0.856 (95% CI: 0.712-0.980), with sensitivities/specificities of 0.759/0.896 and 0.572/0.500. CONCLUSION: The nomogram model based on spine-pelvis compensation can effectively predict surgical outcomes in LSS-DS patients. It provides a basis for individualized treatment.