左房功能自动成像预测非瓣膜性阵发性心房颤动患者卒中风险  

Left atrial automated functional myocardial imaging to identify patients with paroxysmal atrial fibrillation at high risk of stroke

作　　者：刘海兰[1] 陈丽丽 宋嫣[1] 徐莹莹 章春泉[2] Liu Hailan;Chen Lili;Song Yan;Xu Yingying;Zhang Chunquan(Department of Ultrasound,Nanchang First Hospital,Nanchang 330006,China;Department of Ultrasound,the Second Affiliated Hospital of Nanchang University,Nanchang 330006,China)

机构地区：[1]南昌市第一医院超声科,南昌330006 [2]南昌大学第二附属医院超声科,南昌330006

出　　处：《中华超声影像学杂志》2023年第8期672-678,共7页Chinese Journal of Ultrasonography

基　　金：国家自然科学基金(82260349)。

摘　　要：目的应用左房功能自动成像(AFILA)技术分析非瓣膜性阵发性心房颤动(NVPAF)患者左房功能, 探讨NVPAF合并缺血性脑卒中(CIS)的危险因素, 并建立Logistic回归模型。方法回顾性纳入2019年8月至2022年8月于南昌大学第二附属医院治疗的NVPAF患者205例, 分为单纯NVPAF组(154例)和NVPAF合并CIS组(51例)。收集所有患者的临床基线资料、血常规及生化结果、AFILA超声检查数据, 比较两组组间差异, 并进行单因素回归分析。利用多因素Logistic回归分析, 得到独立危险因素。比较Logistic回归模型与CHA2DS2-VASc评分系统ROC曲线下面积(AUC)、敏感性、特异性。结果 NVPAF组与合并CIS组间年龄, CHA2DS2-VASc评分, 抗凝及抗血小板药物使用, 有无高血压、糖尿病及冠心病病史, 左房排空分数(LAEF), 左房储存期应变(SR), 左房导管期应变(SCT), 白细胞(WBC), 中性粒细胞(NEUT), 同型半胱氨酸(HCY), 尿素氮(UREA), D-二聚体(NDD), N末端B型利钠肽前体(NT-proBNP), 纤维蛋白原(Fib), 心肌肌钙蛋白I (cTnI), 中性粒细胞与淋巴细胞比值(NLR)差异有统计学意义(均P<0.05)。多因素回归分析结果显示, 年龄、高血压、SCT、UREA、NLR、Fib、cTnI是NVPAF合并CIS的独立相关的危险因素[OR值分别为1.608 (P=0.003)、3.821(P=0.019)、1.259(P=0.001)1.326(P=0.001)、1.352 (P=0.011)、1.502(P=0.042)、7.651(P=0.001)]。采用CHA2DS2-VASc评分中包含的项目年龄、性别、高血压史进行校正后, NVPAF患者左房收缩期应变SCT仍显著导致卒中风险[OR=1.259(1.095~1.447), P=0.001]。Logistic回归模型诊断效能优于CHA2DS2-VASc评分(AUC 0.931比0.717, 95%CI 0.896~0.967比0.634~0.799, 敏感性0.883比0.755, 特异性0.849比0.713, 均P<0.001)。结论年龄、高血压、SCT、UREA、NLR、Fib、cTnl是NVPAF合并CIS的独立危险因素;结合AFILA超声测量的左房应变指标建立的回归模型较CHA2DS2-VASc评分模型诊断效能更优, 且敏感性、特异性均较高。Objective To investigate the risk factors of non-valvular paroxysmal atrial fibrillation(NVPAF)with cerebral ischemic stroke(CIS)and analyze NVPAF by using left atrial automatic imaging(AFILA).Logistic regression model was established for left atrial(LA)function parameters.Methods A total of 205 patients with NVPAF were included in the study and divided into the NVPAF group without ischemic stroke(154 patients)and the CIS group(51 patients).The clinical baseline data,blood biochemical results and AFILA ultrasound data of all patients were collected.Univariate analysis was performed to compare the above data between the two groups of patients.The independent risk factors were obtained by multivariate logistic regression analysis.Logistic regression model was compared with CHA2DS2-VASc scoring system in terms of area under ROC curve,sensitivity and specificity.Results There were significant differences in age,CHA2DS2-VASc score,taking anticoagulant drugs,history of hypertension,diabetes and coronary heart disease,LAEF,S_R,S_CT,WBC,NEUT,HCY,UREA,NDD,NT-proBNP,Fibrinogen(Fib),Cardiac troponin I(cTnI)and NLR between the two groups(all P<0.05).The results of multifactor analysis showed that:age,hypertension,S_CT,UREA,NLR,Fib and cTnI were independent risk factors associated with CIS in patients with paroxysmal atrial fibrillation[OR value:1.608(P=0.003),3.821(P=0.019),1.259(P=0.001),1.326(P=0.001),1.352(P=0.011),1.502(P=0.042),7.651(P=0.001)].After adjusting for the age,sex and history of hypertension included in CHA2DS2-VASc score,S_CT significantly led to NVPAF complicated with stroke[OR value 1.259(1.095-1.447),P=0.001].The diagnostic efficacy of Logistic regression model is better than that of CHA2DS2-VASc scoring(AUC of 0.931 vs 0.717,95%CI:0.896-0.967 vs 0.634-0.799,sensitivity of 0.883 vs 0.755,specificity of 0.849 vs 0.713,all P<0.001).Conclusions Age,hypertension,S_CT,UREA,NLR,fibrinogen,cTnI are independently associated risk factors for patients with combined CIS;The diagnostic efficacy of Logistic regression

关 键 词：超声检查 非瓣膜性阵发性心房颤动 LOGISTIC回归模型 新技术 

分 类 号：R541.75[医药卫生—心血管疾病] R743.3[医药卫生—内科学] R540.45[医药卫生—临床医学]