Analysis of factors influencing the efficacy of vagus nerve stimulation for the treatment of drug-resistant epilepsy in children and prediction model for efficacy evaluation

Abstract

ObjectiveVagus nerve stimulation (VNS) has been widely used in the treatment of drug-resistant epilepsy (DRE) in children. We aimed to explore the efficacy and safety of VNS, focusing on factors that can influence the efficacy of VNS, and construct a prediction model for the efficacy of VNS in the treatment of DRE children.MethodsRetrospectively analyzed 45 DRE children who underwent VNS at Qilu Hospital of Shandong University from June 2016 to November 2022. A ≥50% reduction in seizure frequency was defined as responder, logistic regression analyses were performed to analyze factors affecting the efficacy of VNS, and a predictive model was constructed. The predictive model was evaluated by receiver operating characteristic curve (ROC), calibration curves, and decision curve analyses (DCA).ResultsA total of 45 DRE children were included in this study, and the frequency of seizures was significantly reduced after VNS treatment, with 25 responders (55.6%), of whom 6 (13.3%) achieved seizure freedom. There was a significant improvement in the Quality of Life in Childhood Epilepsy Questionnaire (15.5%) and Seizure Severity Score (46.2%). 16 potential factors affecting the efficacy of VNS were included, and three statistically significant positive predictors were ultimately screened: shorter seizure duration, focal seizure, and absence of intellectual disability. We developed a nomogram for predicting the efficacy of VNS in the treatment of DRE children. The ROC curve confirmed that the predictive model has good diagnostic performance (AUC = 0.864, P < 0.05), and the nomogram can be further validated by bootstrapping for 1,000 repetitions, with a C-index of 0.837. Besides, this model showed good fitting and calibration and positive net benefits in decision curve analysis.ConclusionVNS is a safe and effective treatment for DRE children. We developed a predictive nomogram for the efficacy of VNS, which provides a basis for more accurate selection of VNS patients.