急性心肌梗死患者发生恶性室性心律失常预测模型的建立与评估  

Development and Evaluation of a predictive model for malignant ventricular arrhythmias in acute myocardial infarction patients in the intensive care unit

作　　者：李惠[1] 马凤锦[2] 郭梦真 LI Hui;MA Fengjin;GUO Mengzhen(Department of Cardiovascular Medicine,The Third People's Hospital of Zhengzhou,Zhengzhou 450000,China)

机构地区：[1]郑州市第三人民医院心血管内科,450000 [2]郑州市第三人民医院重症医学科,450000

出　　处：《心肺血管病杂志》2025年第3期234-239,共6页Journal of Cardiovascular and Pulmonary Diseases

基　　金：2020年度河南省医学科技攻关计划(LHGJ20200719)。

摘　　要：目的:探讨建立重症监护病房(intensive care unit,ICU)内急性心肌梗死(acute myocardial infarction,AMI)患者发生恶性室性心律失常(malignant ventricular arrhythmias,MVAs)的预测模型,并评估其有效性。方法:回顾性纳入2016年1月至2024年1月期间,在郑州市第三人民医院确诊为AMI并在诊疗过程中曾入住ICU的患者共纳入1294例患者,按8:2比例将患者随机分配入训练集和验证集。采集临床变量。验证集中,通过Lasso回归和多因素Logistic回归筛选与MVAs相关的变量,构建列线图模型。在训练集和验证集中,采用受试者工作特征曲线中曲线下面积(Area under curve)和校准曲线评估模型的预测效能。结果:本研究共纳入1294例AMI患者,其中160例患者发生了MVAs。多因素Logistic回归分析显示,女性(OR=0.478,P<0.001)、血钾水平(OR=0.774,P=0.049)、舒张压(OR=1.013,P=0.010)、SOFA评分(OR=1.072,P=0.002)、肌钙蛋白(OR=1.233,P=0.004)、充血性心力衰竭(OR=2.188,P<0.001)和房室传导阻滞(OR=2.078,P=0.017)是MVAs的预测因子。列线图模型在训练集和验证集中的AUC分别为0.769和0.754,校准曲线显示模型预测结果与实际结果较为吻合(Brier得分=0.103)。结论:本研究建立的预测模型能够有效预测ICU内AMI患者发生MVAs的风险,为临床医生提供了一个简便的工具,帮助提前识别高危患者并采取预防措施。Objective:This study aims to develop a model to predict the occurrence of malignant ventricular arrhythmias(MVAs)in patients with acute myocardial infarction(AMI)in the intensive care unit(ICU)and to evaluate its effectiveness.Methods:This single-center cross-sectional study retrospectively included 1294 patients diagnosed with AMI and admitted to the ICU at The Third People's Hospital of Zhengzhou from January 2016 to January 2024.Patients were randomly assigned to the training set and the validation set in an 8:2 ratio.Clinical characteristics were collected.Lasso regression and multivariate logistic regression were used to screen variables associated with MVAs,followed by the construction of nomogram model.The predictive performance of the model was evaluated using the area under the curve(AUC)of the receiver operating characteristic curve and calibration curves in the training cohort and validation cohort.Results:A total of 1294 AMI patients were included in the study,among whom 160 experienced MVAs.Multivariate logistic regression analysis showed that female gender(OR=0.478,P<0.001),blood potassium level(OR=0.774,P=0.049),diastolic blood pressure(OR=1.013,P=0.010),SOFA score(OR=1.072,P=0.002),troponin(OR=1.233,P=0.004),congestive heart failure(OR=2.188,P<0.001),and atrioventricular block(OR=2.078,P=0.017)were independent predictors of MVAs.The AUC of the nomogram model was 0.769 in the training set and 0.754 in the validation set.Calibration curves indicated that the model's predictions were in good agreement with the actual outcomes(Brier score=0.103).Conclusions:The predictive model developed in this study effectively forecasts the risk of MVAs in AMI patients in the ICU.This tool provides clinicians with a simple method to identify high-risk patients.

关 键 词：急性心肌梗死 重症监护病房 恶性室性心律失常 预测模型 LOGISTIC回归分析 

分 类 号：R54[医药卫生—心血管疾病]