老年急性ST段抬高型心肌梗死患者左心室血栓形成的影响因素及其Nomogram风险预测模型构建  

Influencing factors of left ventricular thrombosis in elderly patients with acute ST segment elevation myocardial infarction and the construction of a Nomogram risk prediction model

作　　者：桂元[1] 詹继东[1] 蒋毅[1] 肖婷[1] 张静[1] 余晶[1] GUI Yuan;ZHAN Ji-dong;JIANG Yi;XIAO Ting;ZHANG Jing;YU Jing(Department of Internal Medicine,Huazhong University of Science and Technology Hospital,Wuhan 430074,China)

机构地区：[1]华中科技大学医院内全科,湖北省武汉市430074

出　　处：《中国心血管病研究》2024年第8期719-724,共6页Chinese Journal of Cardiovascular Research

基　　金：湖北省卫生计生委科研项目(WJ2017M053)。

摘　　要：目的探讨老年急性ST段抬高型心肌梗死(STEMI)患者左心室血栓(LVT)形成的影响因素,并构建Nomogram风险预测模型。方法将华中科技大学同济医学院附属同济医院2020年1月至2023年12月收治的873例急性STEMI患者以7∶3分为训练集(n=611)和验证集(n=262)。根据训练集患者LVT发生情况分为非LVT组(n=549)和LVT组(n=62)。多因素logistic回归分析急性STEMI患者发生LVT的影响因素;利用R软件rms程序包构建预测急性STEMI患者LVT发生风险的Nomogram模型,并通过ROC及Hosmer-Lemeshow检验验证其预测效能。结果与非LVT组比较,LVT组血红蛋白(Hb)[(131.15±15.99)g/L比(143.28±16.71)g/L]和左心室射血分数(LVEF)[(39.99±5.01)%比(45.24±6.69)%]明显较低(t=5.441、5.991,P<0.05),合并室壁瘤(37.10%比20.40%)、基线心肌梗死溶栓治疗(TIMI)血流分级0级(59.68%比44.81%)、未形成冠状动脉侧支循环(88.71%比72.13%)以及未进行急诊经皮冠状动脉介入治疗(PCI)或药物溶栓的占比(69.35%比48.82%)明显较高(χ^(2)=9.022、4.953、7.921、9.403,P<0.05)。多因素Logistic回归显示,Hb(OR=0.948)和LVEF(OR=0.868)为急性STEMI患者发生LVT的保护因素(P<0.05),合并室壁瘤(OR=6.812)、基线TIMI血流分级0级(OR=6.884)、未形成冠状动脉侧支循环(OR=7.273)、未进行急诊PCI或药物溶栓(OR=4.173)为其独立危险因素(P<0.05)。ROC分析显示,训练集AUC为0.871(95%CI 0.830~0.912),验证集为0.838(95%CI 0.731~0.945);Hosmer-Lemeshow检验中训练集χ^(2)=11.379,P=0.181,验证集χ^(2)=11.261,P=0.187。结论急性STEMI患者LVT的发生与Hb、LVEF、室壁瘤、基线TIMI分级、冠状动脉侧支循环形成及是否接受急诊PCI或药物溶栓密切相关,基于这些因素构建的Nomogram预测模型区分度及一致性良好,有助于临床预测。Objective To explore the influencing factors of left ventricular thrombosis(LVT)formation in elderly patients with acute ST segment elevation myocardial infarction(STEMI)and to construct the Nomogram risk prediction model.Methods A total of 873 elderly acute STEMI patients admitted to Tongji Hospital from January 2020 to December 2023 were separated into the training set(n=611)and the validation set(n=262)at a ratio of 7∶3.According to the occurrence of LVT in the training set,the patients were separated into the non LVT group(n=549)and LVT group(n=62).Multivariate logistic regression was applied to analyze the influencing factors of LVT in the patients.RMS package of R software was applied to construct the Nomogram model for predicting the risk of LVT in the patients;and ROC and Hosmer-Lemeshow tests were applied to verify its predictive performance.Results Compared with the non LVT group,the LVT group had maredly lower hemoglobin(Hb)[(131.15±15.99)g/L vs.(143.28±16.71)g/L]and left ventricular ejection fraction(LVEF)[(39.99±5.01)%vs.(45.24±6.69)%](t=5.441,5.991,P<0.05);the proportions of patients with concomitant ventricular aneurysm(37.10%vs.20.40%),baseline thrombolysis in myocardial infarction(TIMI)blood flow grade 0(59.68%vs.44.81%),no formation of coronary collateral circulation(88.71%vs.72.13%),and no emergency percutaneous coronary intervention(PCI)or drug thrombolysis(69.35%vs.48.82%)were markedly higher(χ^(2)=9.022,4.953,7.921,9.403,P<0.05).Multivariate logistic regression showed that Hb(OR=0.948)and LVEF(OR=0.868)were the protective factors for LVT in the patients(P<0.05),while concomitant ventricular aneurysm(OR=6.812),baseline TIMI blood flow grade 0(OR=6.884),no formation of coronary collateral circulation(OR=7.273),and no emergency PCI or drug thrombolysis(OR=4.173)were the independent risk factors(P<0.05).ROC analysis showed that the AUC of the training set was 0.871(95%CI 0.830-0.912),and the AUC of the validation set was 0.838(95%CI 0.731-0.945);in the Hosmer-Lemeshow test,the training se

关 键 词：急性ST段抬高型心肌梗死 左心室血栓 影响因素 Nomogram模型 

分 类 号：R542.22[医药卫生—心血管疾病]