心电图ST-T段异常对缺血性脑卒中的预测效能

Predicative effectiveness of electrocardiographic ST-T abnormalities for ischemic stroke

目的 探讨心电图ST-T段异常(STA)对缺血性脑卒中的预测价值。方法 采用前瞻性研究方法,对3 027名体检人员(受试者)进行心电图检查,并随访60个月~120个月(随访截止2022年6月,或研究对象发生缺血性脑卒中则停止随访)。根据随访结果进行分组,将发生缺血性脑卒中的94例受试者设为研究组,将其余2 933例受试者设为对照组,统计STA发生情况。对缺血性脑卒中的影响因素采用单因素分析及多因素非条件Logistic回归分析,绘制受试者工作特征(ROC)曲线分析STA对缺血性脑卒中的预测价值。结果 两组受试者性别、年龄、STA占比、左心室肥厚、脑卒中家族史比较,差异有统计学意义(P<0.01)。多因素非条件Logistic回归分析显示,年龄、STA、高血压、高血脂症、饮酒均是缺血性脑卒中的危险因素(P<0.05或0.01)。ROC曲线显示,STA预测缺血性脑卒中的曲线下面积为0.859,灵敏度为0.830,特异度为0.888,准确度为0.869;关联性检验显著(χ~2=138.706,P=0.000),优势性检验不显著(χ~2=0.432,P=0.511)。结论 STA可增加人群缺血性脑卒中的发病风险,对缺血性脑卒中的预测效能较高。

Objective To explore the predictive value of electrocardiographic(ECG) ST-T abnormalities(STA) for ischemic st roke(IS).Methods ECG exa minations were conducted in 3027 subjects in a prospective study a nd followed up for 60-120 mont hs( follow-up ended in Ju n.2022,or follow-up discontinued when subjects developed ischemic stroke).Ninety-four IS subjects were set as study group and the other 2 933 non-IS ones as control group according to follow-up results,the occurrences of STA counted.Univariate analysis and multivariate unconditional Logistic regression analysis were conducted on the influencing factors of the IS and receiver operating characteristic(ROC) curves plotted to analyze the predictive value of the STA for the IS.Results Intergroup differences in such indexes were statistically significant as gender,age,the proportion of the STA,left ventricular hypertrophy,and family history of stroke(P<0.01).Multiva riate unconditional Logistic regression a nalysis showed that ages,STA,hypertension,hyperlipidemia,and alcohol drinking were all risk factors for the IS(P<0.05 or 0.01).The ROC curve shows that the area under the curve of STA for the IS was 0.859,sensitivity 0.830,specificity 0.888,and accuracy 0.869;correlation test was significant( χ~2=138.706,P=0.000) and dominance test wasn't( χ~2=0.432,P=0.511).Conclusion STA can increase the onset risk of the IS and has a higher predictive effectiveness for the IS.