• Xin Li, Bin Pan, Lin Cheng, Gen Li, Jian Liu, and Feng Yuan.
• Department of Orthopedics, the First People's Hospital of Lianyungang, Lianyungang, Jiangsu, China.
• Pain Physician. 2023 Jan 1; 26 (1): 819081-90.

BackgroundRecurrence of lumbar disc herniation (LDH) is an adverse event after percutaneous endoscopic transforaminal discectomy (PETD). Accurate prediction of the risk of recurrent LDH (rLDH) after surgery remains a major challenge for spine surgeons.ObjectivesTo develop and validate a prognostic model based on risk factors for rLDH after PETD.Study DesignRetrospective study.SettingInpatient surgery center.MethodsClinical data were retrospectively collected from 645 patients with LDH who underwent PETD at the Affiliated Hospital of Xuzhou Medical University from January 1, 2017 to January 1, 2021. Predictors significantly associated with rLBH were screened according to least absolute shrinkage and selection operator (LASSO) regression, and a prognostic model was established, followed by internal model validation using the enhanced bootstrap method. The performance of the model was assessed using receiver operating characteristic (ROC) curves and calibration curves. Finally, the clinical usefulness of the model was analyzed using decision curve analysis (DCA) and clinical impact curves (CICs).ResultsAmong the 645 patients included in this study, 56 experienced recurrence of LDH after PETD (8.7%). Seven factors significantly associated with rLDH were selected by LASSO regression, including age, type of herniation, level of herniation, Modic changes, Pfirrmann classification, smoking, and history of high-intensity physical work. The bias-corrected curve of the model fit well with the apparent curve, and the area under the ROC curve was 0.822 (95% confidence interval, 0.76-0.88). The DCA and CIC confirmed that the prognostic model had good clinical utility.LimitationsThis is a single-center study, and we used internal validation only.ConclusionsThe prognostic model developed in this study had excellent comprehensive performance and could well predict the risk of rLDH after PETD. This model could be used to identify patients at high risk for rLDH at an early stage to individualize the patient's treatment modality and postoperative rehabilitation plan.

20 keywords...

hide…