CAS评分对非瓣膜性心房颤动患者血栓栓塞风险的预测价值

Predictive value of the CAS score model for thromboembolism risk in patients with non-valvular atrial fibrillation

目的评价CAS评分模型对中国急诊非瓣膜性心房颤动患者1年内血栓栓塞风险的预测价值。方法本研究是对一项前瞻性、多中心、观察性研究数据的事后分析。利用中国急诊心房颤动注册研究在2008年至2011年在全国20家医院纳入的2023例患者的数据,对其中符合条件的1327例患者分别根据CAS评分模型和CHA_(2)DS_(2)-VASc评分模型进行风险度分层和分组,分析两种不同模型的风险分层与患者1年内血栓栓塞(体循环栓塞和/或脑卒中)事件的相关性,比较两种模型风险分层的一致性和对终点事件的预测效力。结果患者年龄(71.0±12.7)岁,体重指数为(23.8±3.6)kg/m^(2),其中女678例(51.1%,678/1327),CHA_(2)DS_(2)-VASc评分(3.7±2.1)分;CAS评分(1.4±1.1)分。CAS评分模型不同风险度分层间的血栓栓塞(χ^(2)=9.119,P=0.010)和脑卒中(χ^(2)=8.632,P=0.013)发生率均差异有统计学意义,体循环栓塞(χ^(2)=1.221,P=0.543)发生率差异无统计学意义。CHA_(2)DS_(2)-VASc评分模型不同风险度分层间的血栓栓塞(χ^(2)=7.032,P=0.030)和脑卒中(χ^(2)=8.632,P=0.013)发生率差异均有统计学意义,体循环栓塞(χ^(2)=2.202,P=0.332)发生率均差异无统计学意义。CAS评分≥1分患者的脑卒中、血栓栓塞风险显著增高,与CHA_(2)DS_(2)-VASc评分高危组相似。两种评分模型对脑卒中和血栓栓塞风险的分层一致性较差。接收者操作特征曲线(ROC)分析显示:CAS评分与CHA_(2)DS_(2)-VASc评分的ROC曲线下面积(AUC)值差异有统计学意义(预测脑卒中的差异为0.046,预测血栓栓塞的差异为0.043,P=0.030),CAS评分在预测脑卒中和血栓栓塞能力方面较CHA_(2)DS_(2)-VASc评分稍弱。结论CAS评分模型较CHA_(2)DS_(2)-VASc评分相比在预测中国急诊非瓣膜性心房颤动患者1年内脑卒中和血栓栓塞风险方面效力稍差,但前者更加简便且易于实施。CAS评分模型在筛选脑卒中/血栓栓塞低危患者方面可能具有优势。

Objective To evaluate the predictive value of the CAS score model for the risk of thromboembolism within 1 year in Chinese emergency patients with non-valvular atrial fibrillation.Methods This study is a post hoc analysis of data from a prospective,multicenter,observational study.Data from 2023 patients enrolled in the Chinese Emergency Atrial Fibrillation Registry Study from 2008 to 2011 in 20 hospitals across China were analyzed.Data from 1327 of these eligible patients were stratified and grouped according to the CAS scoring model and the CHA_(2)DS_(2)-VASc scoring model respectively,and the correlation between risk stratification of the two different models and patients′thromboembolic events(stroke or systemic embolism)within 1 year was analyzed.The consistency of risk stratification and the predictive validity of the two models for endpoint events were compared.Results The mean age of the patients was(71.0±12.7)years,the mean body mass index was(23.8±3.6)kg/m^(2),of which 678 were females(678/1327,51.1%),the mean CHA_(2)DS_(2)-VASc score was 3.7±2.1;the mean CAS score was 1.4±1.1.The risk of thromboembolism(χ^(2)=9.119,P=0.010)and stroke(χ^(2)=8.632,P=0.013)differed significantly between different groups according to CAS scores,and there was no significant difference in the risk of systemic embolism(χ^(2)=1.221,P=0.543).The risk of thromboembolism(χ^(2)=7.032,P=0.030)and stroke(χ^(2)=8.632,P=0.013)differed significantly between different groups according to CHA_(2)DS_(2)-VASc scores,and there was no significant difference in the risk of systemic embolism(χ^(2)=2.202,P=0.332).Patients with a CAS score≥1 had a significantly higher risk of stroke and thromboembolism,similar to the CHA_(2)DS_(2)-VASc score high-risk group.The stratification of stroke and thromboembolism risk was less consistent between the 2 models,with a statistically significant difference.Receiver operating characteristic curve(ROC)analysis showed that the AUC values of the CAS score were significantly different from those of the CHA_(2)DS_(2)-VASc score(P<0.05)and that the CAS score was slightly weaker than the CHA_(2)DS_(2)-VASc score in predicting stroke and thromboembolism.Conclusion The CAS scoring model is slightly less effective than the CHA_(2)DS_(2)-VASc score in predicting the risk of stroke and thromboembolism within 1 year in Chinese emergency patients with non-valvular atrial fibrillation,but the former is simpler and easier to be applied.The CAS scoring model may be more advantageous in screening patients at low risk of stroke/thromboembolism.