Development and validation of a nomogram for predicting postoperative recurrent lumbar disc herniation after unilateral biportal endoscopic discectomy

Background No study has investigated the incidence or risk factors of recurrent lumbar disc herniation (rLDH) after unilateral biportal endoscopic discectomy (UBED) in patients with lumbar disc herniation. Therefore, we aimed to construct and verify a nomogram that can predict the risk of recurrence after UBED for lumbar disc herniation and therefore help clinicians understand the recurrence rate of lumbar disc herniation before surgery and formulate corresponding interventions to improve the prognosis of patients. Methods This is a retrospective study based on the data from all patients who underwent UBED for lumbar disc herniation in our hospital between September 2021 and December 2023. The risk factors for recurrence after UBED in patients with lumbar disc herniation were determined via stepwise regression. Finally, a nomogram was constructed on the basis of these selected variables to predict the recurrence rate of lumbar disc herniation after UBED. The receiver operating characteristic (ROC) curves and the areas under the curve (AUC) were used to evaluate the discrimination ability of the model. In addition, calibration curve and Hosmer-Lemeshow goodness of fit test (HL test) were constructed to evaluate the consistency between the predicted risk and the actual risk. Decision curve analysis (DCA) was used to evaluate the clinical net income and the clinical applicability of the nomogram. Results The stepwise regression analysis showed that Age, disease duration, segmental range of motion (SROM), global range of motion (GROM), Modic I changes, and Modic II changes are risk factors for the recurrence of lumbar disc herniation after UBED. The AUCs of the model on the training set and the validation set are 0.763 (95% CI: 0.648-0.879) and 0.721 (95% CI: 0.573-0.868), respectively. The results showed that our nomogram has better recognition ability. The calibration plot and goodness-of-fit test results indicated that the prediction results of our model were consistent with the actual results (training set HL test, P = 0.087; HL test of the validation set, P = 0.193). DCA showed that our model has good net clinical benefit. Conclusion Our study revealed that age, disease duration, SROM, GROM, and Modic changes are risk factors for the recurrence of lumbar disc herniation after UBED. On the basis of these variables, we successfully constructed and verified a nomogram that can predict the risk of recurrence of lumbar disc herniation after UBED, which can help clinicians predict the possibility of rLDH in patients in advance and develop interventions to reduce the risk of recurrence.