A nomogram for predicting adverse neurovascular events after carotid artery stenting in patients with symptomatic carotid stenosis

Defang LuoHailan ZengCheng XieQianliang HuangMingang Zou^*Qiuhua Jiang^*

• Ganzhou People's Hospital, Ganzhou, China

Background: Carotid artery stenting (CAS) is considered a crucial treatment option for patients with symptomatic carotid artery stenosis. Nevertheless, adverse neurovascular events (ANEs) following this procedure remain a significant challenge. This study aimed to identify risk factors for ANEs and to construct a predictive nomogram to assist in perioperative risk stratification. Methods: This retrospective study (January 2020 - January 2025) enrolled consecutive symptomatic carotid stenosis patients undergoing CAS from two centers: 209 in the training cohort from Ganzhou People's Hospital and 148 in the external validation cohort from The First Affiliated Hospital of Nanchang University. Patients were categorized into ANE and non-ANE groups based on postoperative outcomes within 30 days. Within the training cohort, independent predictors were identified through a three-step approach: (1) univariate screening, (2) LASSO regression for variable selection, and (3) multivariable logistic regression for final risk factor determination. The nomogram was constructed using R. Internal validation was performed via 1,000 bootstrap resamples. The model's predictive accuracy and clinical utility were assessed using the C-index, receiver operating characteristic (ROC) curve, and decision curve analysis (DCA). Results: Age, ulcerated plaque, hemodynamic suppression, and balloon dilation were found to be independent risk factors for the occurrence of ANEs. The Hosmer‒Lemeshow test confirmed a good model fit (training: p=0.845; validation: p=0.356), and the calibration curve showed no significant deviation of the predicted probabilities from the actual probabilities. The bootstrap-corrected C-index for internal validation was 0.773. Discriminatory performance was robust, with C-index of 0.802 (training) and 0.816 (validation), and AUCs of 0.798 (95% CI: 0.707-0.889, training) and 0.819 (95% CI: 0.724-0.913, validation). DCA confirmed the substantial clinical value of the nomogram. Furthermore, stratified analyses further revealed different but consistent risk profiles for ischemic and hemorrhagic ANEs, while the composite nomogram maintained robust predictive performance across both subgroups. Conclusions: The nomogram demonstrated good predictive performance for assessing the risk of ANEs in symptomatic carotid stenosis patients undergoing CAS. Its use aids in optimizing clinical decision-making and reducing postoperative ANEs.