Abstract
BACKGROUND: Postoperative bone nonunion is a critical complication following instrumented fusion for spinal tuberculosis. Preoperative prediction is essential for prevention. While clinical risk factors exist, current predictive tools lack validation in infected cohorts. PURPOSE: This study developed and validated a multivariate nomogram, provided an individualized preoperative estimate of nonunion risk in spinal tuberculosis patients, incorporating key clinical and radiological predictors to guide preventative strategies. METHOD: A retrospective cohort of 178 patients undergoing debridement and instrumented fusion for spinal tuberculosis (Shandong Public Health Clinical Center, January 2021-January 2024) was stratified by Bridwell classification into union (n = 120) and nonunion (n = 58) groups. Perioperative variables were compared between groups. Predictive features were selected via least absolute shrinkage and operator selection (LASSO) regression and incorporated into a multivariate logistic regression model. A nomogram was constructed based on the model. Calibration was assessed using the Hosmer-Lemeshow test with calibration curves, and discriminative ability was evaluated by the area under the ROC curve (AUC). Decision curve analysis (DCA)was performed to estimate the clinical usefulness of the prediction model by quantifying the net benefits at different threshold probabilities. RESULTS: The training cohort of this study comprised 178 patients, of which 120 presented with union and 58 with nonunion. Five predictor variables were screened by LASSO regression and plotted as a nomogram, including ALB, CRP normalization days, Bone graft materials, Psoas abscess, Jumping lesions. The nomogram showed strong discrimination and solid calibration, AUC = 0.947 (95% confidence 0.915-0.978). The calibration curves of the diagnostic models showed the optimal concordance between the predicted results and the actual observations. The DCA indicated that the substantial clinical net benefit across threshold probabilities. CONCLUSION: The study successfully developed a precise and effective nomogram for identifying postoperative bone nonunion in spinal tuberculosis patients. This nomogram aids early detection and prevention in postoperative bone nonunion, improving clinical decisions and treatment optimization.