急性心肌梗死后主要不良心血管事件预测模型的构建与验证  

Construction and validation of a prediction model for major adverse cardiovascular events after acute myocardial infarction

作　　者：王艳玲 王彦 郭春宏 李林 WANG Yan-ling;WANG Yan;GUO Chun-hong;LI Lin(Department of Electrocardiography,Henan Veterans Hospital,Xinxiang 453002,Henan,CHINA;Department of Electrocardiography,Xinxiang Central Hospital,Xinxiang 453000,Henan,CHINA;Department of Electrocardiography,Xinxiang First People's Hospital,Xinxiang 453000,Henan,CHINA)

机构地区：[1]河南省荣军医院心电图室,河南新乡453002 [2]新乡市中心医院心电图科,河南新乡453000 [3]新乡市第一人民医院心电图室,河南新乡453000

出　　处：《海南医学》2025年第6期784-789,共6页Hainan Medical Journal

基　　金：河南省科技攻关项目(编号:222102312694)。

摘　　要：目的基于24 h心电图心率变异性联合游离甲状腺激素(FT4)/游离三碘甲状腺原氨酸(FT3)、肌钙蛋白I(cTnI)构建急性心肌梗死后主要不良心血管事件(MACE)的预测模型,并对所构建的模型进行验证。方法前瞻性选取2021年9月至2024年8月河南省荣军医院收治的249例急性心肌梗死患者作为研究对象,根据经皮冠状动脉介入术(PCI)术后住院期间是否发生MACE分为MACE组62例和无MACE组187例。比较两组患者PCI术后即刻24 h心电图心率变异性[24 h正常R-R间期标准差(SDNN)、24 hQT间期标准差(24 h-QTV)、低频/高频(LF/HF)、相邻R-R间期差值在50 ms以上占总窦性心搏的百分数(PNN50)]、FT4/FT3、cTnI水平,采用多因素Logistic回归分析MACE的独立预警因素,运用R语言构建急性心肌梗死后MACE的预测模型,并对模型进行验证。结果MACE组患者的SDNN、24 h-QTV、LF/HF、PNN50分别为(80.79±16.84)ms、(18.26±5.19)ms、1.48±0.17、(4.29±1.30)%,明显低于无AMCE组的(105.65±19.30)ms、(24.15±4.62)ms、1.76±0.23、(11.06±3.51)%,差异均有统计学意义(P<0.05);MACE组患者的FT4/FT3、cTnI分别为4.72±1.55、(0.25±0.07)μg/L,明显高于无MACE组的3.61±1.08、(0.16±0.04)μg/L,差异均有统计学意义(P<0.05);多因素Logistic回归分析结果显示,SDNN升高、24 h-QTV升高、LF/HF升高、PNN50升高、FT4/FT3升高、cTnI升高均是MACE的独立相关影响因素(P<0.05);基于MACE的独立相关影响因素构建急性心肌梗死后MACE的预测模型,其C-index为0.871;绘制ROC曲线显示,所构建列线图预测模型的AUC为0.871(95%CI:0.818~0.924);绘制校准曲线显示,模型与实际观测结果贴合度良好;绘制临床决策分析曲线显示,在阈概率0~0.8时模型具有正向的净获益(P<0.05)。结论24 h心电图心率变异性、FT4/FT3、cTnI均是急性心肌梗死后MACE的影响因素,基于以上因素构建的MACE预测模型具有良好的价值,能为临床预警MACE的发生提供参考。Objective To construct and validate a predictive model for major adverse cardiovascular events(MACE)after acute myocardial infarction based on 24-hour electrocardiogram heart rate variability(HRV)combined with free thyroxine(FT4)/free triiodothyronine(FT3)and cardiac troponin I(cTnI).Methods A prospective study was conducted on 249 patients with acute myocardial infarction admitted to Henan Veterans Hospital from September 2021 to August 2024.Patients were divided into a MACE group(n=62)and a non-MACE group(n=187)based on the occurrence of MACE during hospitalization after percutaneous coronary intervention(PCI).HRV parameters(24-hour standard deviation of normal R-R intervals[SDNN],24-hour QT variability[24 h-QTV],low-frequency/high-frequency ratio[LF/HF],percentage of adjacent R-R interval differences exceeding 50 ms in total sinus beats[PNN50]),FT4/FT3 ratio,and cTnI levels were compared between the two groups.Multivariate logistic regression was used to identify independent predictors of MACE,and a prediction model was constructed using R software and validated.Results The SDNN,24-hour QTV,LF/HF,and PNN50 in the MACE group were(80.79±16.84)ms,(18.26±5.19)ms,1.48±0.17,and(4.29±1.30)%,respectively,which were significantly lower than(105.65±19.30)ms,(24.15±4.62)ms,1.76±0.23,and(11.06±3.51)%in the non-MACE group(P<0.05).The FT4/FT3 and cTnI levels in the MACE group were 4.72±1.55 and(0.25±0.07)μg/L,respectively,which were significantly higher than 3.61±1.08 and(0.16±0.04)μg/L in the non-MACE group(P<0.05).Multivariate logistic regression analysis showed that increased SDNN,24-hour QTV,LF/HF,PNN50,FT4/FT3,and cTnI were all independently associated with MACE(P<0.05).Based on the independent related factors of MACE,a predictive model for MACE after acute myocardial infarction was constructed,with a C-index of 0.871.ROC curve showed that the AUC of the constructed nomogram prediction model was 0.871(95%CI:0.818-0.924).Calibration curve showed good agreement between predicted and observed outcomes,and de

关 键 词：急性心肌梗死 游离甲状腺激素 24 h心电图 心率变异性 游离三碘甲状腺原氨酸 肌钙蛋白I 预测模型 主要不良心血管事件 

分 类 号：R542.22[医药卫生—心血管疾病]