急性ST段抬高型心肌梗死介入治疗前并发心室颤动的预测模型构建  

Development of a predictive model for ventricular fibrillation in acute ST-segment elevation myocardial infarction

作　　者：陈红珍[1] 刘幸娟 郝琦 周强[1] 赵巧燕[1] CHEN Hongzhen;LIU Xingjuan;HAO Qi;ZHOU Qiang;ZHAO Qiaoyan(Department of Heart Transplantation Centre,Zhengzhou 7^(th)People's Hospital,Zhengzhou,450016,China)

机构地区：[1]郑州市第七人民医院心脏移植中心,郑州450016

出　　处：《临床心血管病杂志》2023年第11期870-875,共6页Journal of Clinical Cardiology

基　　金：河南省医学科技攻关计划项目(No:LHGJ20220843)。

摘　　要：目的:构建急性ST抬高型心肌梗死并发心室颤动的预测模型。方法:回顾性纳入2017年—2022年716例急性ST段抬高型心肌梗死患者的临床资料,根据介入治疗前有无心室颤动分成两组。应用多因素分析筛选出危险因素,然后使用Lasso回归筛选全部变量,合并共同变量,并与实际相结合,绘制列线图对心室颤动发生概率进行预测。利用ROC曲线下面积(AUC)、Harrell的C指数和校准曲线用于评估和内部验证该模型的性能。应用决策曲线分析评估该模型的临床效用。结果:ST段抬高型急性心肌梗死介入治疗前并发心室颤动的预测因素有年龄>57岁、Killip分级、左室射血分数≤41%、心率>90次/min、心绞痛、心房颤动、吸烟史、降脂药物使用。同时根据实际情况和差异变量,构造可信度比较高的列线图预测心室颤动的发生概率。结论:基于年龄、Killip分级、左室射血分数、心率、心绞痛、心房颤动、吸烟史、降脂药物等因素,构建了具有一定可信度的预测急性ST段抬高型心肌梗死介入治疗前并发心室颤动的模型。Objective:To construct a prediction model for forecasting the occurrence of ventricular fibrillation in cases of acute ST-segment elevation myocardial infarction.Methods:Clinical data of 716 patients with acute ST-segment elevation myocardial infarction from 2017 to 2022 were retrospectively included.Patients were divided into two groups according to the presence or absence of ventricular fibrillation before interventional therapy.Risk factors were selected by applying multivariate analysis,then all variables were further screened by using Lasso regression.Common variables were integrated with clinical context to formulate a nomogram predicting the likelihood of ventricular fibrillation.The area under the ROC curve(AUC),Harrell's C-index,and calibration curves were used to assess and internally validate the performance of the model.Decision curve analysis was applied to assess the model's clinical utility.Results:Predictors of ventricular fibrillation complicating acute ST-segment elevation myocardial infarction before intervention were age >57 years,Killip classification,left ventricular ejection fraction ≤41%,heart rate >90 beats/min,angina pectoris,atrial fibrillation,history of smoking,and antilipidemic medication use.Based on the actual situation and difference variables,we developed a nomogram with relatively high reliability for predicting the probability of ventricular fibrillation.Conclusion:We have constructed a nomogram with moderate reliability for predicting ventricular fibrillation before interventional treatment in acute ST-segment elevation myocardial infarction.This nomogram incorporates age,Killip classification,left ventricular ejection fraction,heart rate,angina pectoris,atrial fibrillation,history of smoking,and antilipid medication use.

关 键 词：ST抬高型急性心肌梗死 心室颤动 列线图 预测模型 

分 类 号：R541.4[医药卫生—心血管疾病]