左房及左心耳应变参数预测非瓣膜性心房颤动患者脑卒中的临床价值

Clinical value of left atrium and left atrial appendage strain parameters in predicting stroke in patients with non-valvular atrial fibrillation

目的 应用二维斑点追踪技术测量非瓣膜性心房颤动(以下简称房颤)患者左房及左心耳应变参数,探讨其预测脑卒中的临床价值。方法 收集80例非瓣膜性房颤患者,根据脑卒中病史分为卒中组38例和非卒中组42例,均行经食管及经胸超声心动图检查获得左房内径、左房容积指数、左房射血分数(LAEF)、左心耳排空速度(LAAEV)和充盈速度(LAAFV)、左心耳排空分数(LAAEF),二维斑点追踪技术获得左房及左心耳应变参数,比较两组上述参数的差异。采用多元Logistic回归分析非瓣膜性房颤患者发生脑卒中的影响因素;绘制受试者工作特征(ROC)曲线分析左房及左心耳应变参数对脑卒中的预测效能。结果 卒中组LAEF、左房储备期峰值应变(LASr)、LAAEF、LAAEV、LAAFV、左心耳整体峰值纵向应变(LAAS)均较非卒中组减低,差异均有统计学意义(均P<0.05)。Logistic回归分析显示,LASr、LAAS均为非瓣膜性房颤患者发生脑卒中的独立预测因子[OR=0.790(95%可信区间:0.678~0.920)、0.727(95%可信区间:0.575~0.919),均P<0.05]。ROC曲线分析显示,LASr预测非瓣膜性房颤患者发生脑卒中的截断值为16.96%,曲线下面积、灵敏度、特异度分别为0.766、72.5%、75.0%;LAAS预测非瓣膜性房颤患者发生脑卒中的截断值为-6.88%,曲线下面积、灵敏度、特异度分别为0.753、52.5%、90.0%。结论 LASr、LAAS均为非瓣膜性房颤患者发生脑卒中的独立预测因子,能早期、有效地预测脑卒中的发生。

Objective To measure the strain parameters of left atrium and left atrial appendage in patients with non-valvular atrial fibrillation by two-dimensional speckle tracking technique,and to investigate its clinical value in predicting stroke.Methods A total of 80 patients with non-valvular atrial fibrillation were collected and divided into stroke group(38 cases)and non-stroke group(42 cases)according to stroke history.Transesophageal and transthoracic echocardiography were performed to obtain left atrial diameter,left atrial volume index,left atrial ejection fraction(LAEF),left atrial atrial emptying velocity(LAAEV),filling velocity(LAAFV),and left atrial atrial emptying fraction(LAAEF).Two-dimensional speckle tracking technique was used to obtain left atrium and left atrial appendage strain parameters.The differences of above parameters were compared between the two groups.Multivariate Logistic regression was used to analyze the influencing factors of stroke in patients with non-valvular atrial fibrillation.Receiver operating characteristic(ROC)curve was drawn to analyze the predictive efficacy of left atrial and left atrial appendage strain parameters in predicting stroke.Results LAEF,left atrial strain reservoir function(LASr),LAAEF,LAAEV,LAAFV and left atrial appendage strain(LAAS)in stroke group were lower than those in non-stroke group(all P<0.05).Logistic regression analysis showed that LASr and LAAS were independent predictors of stroke in patients with non-valvular atrial fibrillation[OR=0.790(95%confidence interval:0.678~0.90),0.727(95%confidence interval:0.575~0.919),both P<0.05].ROC curve analysis showed that the cut-off value of LASr for predicting stroke in patients with non-valvular atrial fibrillation was 16.96%,the area under the curve(AUC),sensitivity,specificity were 0.766,72.5%,75.0%,respectively.The cut-off value of LAAS for predicting stroke in patients with non-valvular atrial fibrillation was-6.88%,the AUC,sensitivity,specificity were 0.753,52.5%,90.0%,respectively.Conclusion LASr and LAAS are independent predictors of stroke in patients with non-valvular atrial fibrillation,which can predict the occurrence of stroke early and effectively.