Construction and validation of a nomogram risk prediction model for in-stent restenosis after carotid artery stenting
10.3969/j.issn.1002-1671.2025.08.026
- VernacularTitle:颈动脉支架植入术后支架内再狭窄的列线图风险预测模型的构建和验证
- Author:
Jialong CHEN
1
;
Yang YU
;
Yong YANG
;
Chunling ZHU
Author Information
1. 哈尔滨市第四医院超声科,黑龙江 哈尔滨 150026
- Publication Type:Journal Article
- Keywords:
nomogram;
prediction model;
carotid artery stenting;
carotid artery stenosis
- From:
Journal of Practical Radiology
2025;41(8):1375-1378
- CountryChina
- Language:Chinese
-
Abstract:
Objective To develop and validate a nomogram risk prediction model for in-stent restenosis(ISR)in patients undergoing carotid artery stenting(CAS).Methods A retrospective analysis was conducted on the clinical data from 267 CAS patients.The dataset was randomly divided into training set(187 cases,for modeling)and test set(80 cases,for validation)in a 7∶3 ratio.Based on whether ISR occurred within one year after CAS,patients were divided into ISR and non-ISR groups.Univariate and multivariate logistic regression analyses were conducted on the training set to identify independent influencing factors and to construct a model.The model was then validated using samples from the test set.The model's discrimination was assessed using the area under the curve(AUC)of the receiver operating characteristic(ROC),while the calibration was evaluated using the Hosmer-Lemeshow test and calibration plots.Results ISR occurred in 46 out of 267 CAS patients(17.23％)within one year after surgery.Multivariate logistic regression analysis of the training set identified hypertension[odds ratio(OR)=2.730,95％confidence interval(CI)1.124-6.630],hyperlipidemia(OR=3.135,95％CI 1.293-7.599),preoperative calcified plaque(OR=3.485,95％CI 1.422-8.542),and residual stenosis(OR=5.440,95％CI 1.570-18.854)as significant predictors of ISR in patients with CAS within one year after surgery.The predictive model achieved an AUC of 0.798(9 5％CI 0.715-0.881),with a sensitivity of 73.5％,specificity of 75.8％,and accuracy of 75.4％.The AUC of external validation of the test set was 0.764(95％CI 0.599-0.929).Hosmer-Lemeshow test yielded x2=14.76,P=0.12,while calibration plots indicated high concordance between the predicted and actual outcomes in both the training set and the test set.Conclusion The nomogram risk prediction model for ISR in CAS patients within one year after surgery demonstrates good discrimination and calibration,provi-ding valuable insights for clinical assessment of ISR risk,control of ISR incidence and precision treatment planning.
