颈动脉狭窄斑块易损性的计算机断层扫描评估对缺血性卒中的预测价值  

Predictive value of computed tomography evaluation of vulnerability of carotid artery stenosis plaques for ischemic stroke

作　　者：曾宇琪 郑进 曹俊杰 谭梓仪 姚志超 周大勇[1] 黄剑[1] Zeng Yuqi;Zheng Jin;Cao Junjie;Tan Ziyi;Yao Zhichao;Zhou Dayong;Huang Jian(Department of Vascular Surgery,Gusu School of Nanjing Medical University/Affiliated Suzhou Hospital of Nanjing Medical University/Suzhou Municipal Hospital(HQ),Suzhou 215002,Jiangsu,China)

机构地区：[1]南京医科大学姑苏学院/南京医科大学附属苏州医院/苏州市立医院(本部)血管外科,江苏苏州215002

出　　处：《血管与腔内血管外科杂志》2025年第3期342-347,共6页Journal of Vascular and Endovascular Surgery

基　　金：江苏省卫生健康委科研项目(H2023146)。

摘　　要：目的探讨颈动脉计算机断层扫描(CT)评估颈动脉狭窄患者发生缺血性卒中的风险预测模型。方法收集2012年1月至2022年5月于南京医科大学姑苏学院/南京医科大学附属苏州医院/苏州市立医院(本部)行CT检查的506例颈动脉狭窄患者的临床资料,按照缺血性卒中发生情况将其分为卒中组(n=185)和非卒中组(n=321)。比较两组患者的临床特征,使用最小绝对收缩和选择算子(LASSO)回归分析筛选预测变量,基于多因素逻辑回归构建颈动脉狭窄患者发生缺血性卒中风险的预测模型,通过曲线下面积(AUC)、校准曲线和决策曲线(DCA)评估构建模型的预测能力和临床应用价值。结果两组患者斑块性质、高血压、服用抗血小板药物、服用他汀类药物、体重指数(BMI)比较,差异均有统计学意义(P﹤0.05)。经过LASSO回归筛选出3个预测变量用于构建缺血性卒中预测模型,AUC为0.841(95%CI:0.807~0.876),模型拟合良好(P﹥0.05),DCA进一步证实模型在风险范围为7%~88%内的预测价值。结论基于颈动脉斑块性质、服用抗血小板药物、服用他汀类药物3个危险因素构建的缺血性卒中风险预测模型,能够有效预测颈动脉狭窄程度≥30%患者缺血性卒中的发生风险,特别是在7%~88%风险范围内的患者,这将为颈动脉狭窄患者提供更加个体化的风险评估并提供相应的治疗方案。Objective To investigate the risk prediction model of ischemic stroke in patients with carotid artery stenosis assessed by carotid computed tomography(CT).Method Clinical data of 506 patients with carotid artery stenosis who underwent CT examination in Gusu School of Nanjing Medical University/Affiliated Suzhou Hospital of Nanjing Medical University/Suzhou Municipal Hospital(HQ)from January 2012 to May 2022 were collected.According to the incidence of ischemic stroke,the patients were divided into stroke group(n=185)and non-stroke group(n=321).The clinical features of the two groups were compared,the least absolute shrinkage and selection operator(LASSO)was used to screen predictive variables,and the predictive model of ischemic stroke risk in patients with carotid artery stenosis was constructed based on multifactor Logistic regression.The predictive power and clinical application value of the constructed model were evaluated by area under curve(AUC),calibration curve,and decision curve analysis(DCA).Result There were significant differences in plaque nature,hypertension,taking antiplatelet drugs,taking statins and BMI between the two groups(P<0.05).Three predictors were selected by LASSO regression to construct the ischemic stroke prediction model,and the AUC was 0.841(95%CI:0.807-0.876),indicating a good model fit(P>0.05).DCA further confirmed the predictive value of the model in the risk range of 7%-88%.Conclusion The ischemic stroke risk prediction model constructed based on the three risk factors of carotid plaque nature,taking antiplatelet drugs and taking statins can effectively predict the risk of ischemic stroke in patients with carotid artery stenosis degree≥30%,especially in patients with the risk range of 7%to 88%.This will provide a more individualized risk assessment for patients with carotid artery stenosis and provide appropriate treatment options.

关 键 词：颈动脉狭窄 计算机断层扫描 卒中 预测模型 

分 类 号：R543[医药卫生—心血管疾病]