基于森林图构建中青年急性心肌梗死伴心律失常患者的早期风险预警模型

Construction of early risk warning model for young and middle-aged patients with acute myocardial infarction and arrhythmia based on forest graph

目的探讨基于森林图构建中青年急性心肌梗死(AMI)伴心律失常患者的早期风险预警模型。方法选取秦皇岛市第一医院2020年11月—2021年12月中青年AMI患者242例,根据是否伴有心律失常分为观察组(68例,伴有心律失常)与对照组(174例,未伴有心律失常)。收集两组临床资料,进行单因素分析,采用Logistic回归进行多因素分析,绘制森林图,构建风险预警模型,分析风险预警模型的预测价值。结果单因素分析显示,心肌梗死部位、冠状动脉病变位置、冠状动脉狭窄程度、Killip分级>Ⅱ级、病程、LVEF下降是中青年AMI伴心律失常的影响因素(P<0.05);多因素Logistic回归分析显示,前壁心肌梗死、左侧冠状动脉病变、冠状动脉重度狭窄/闭塞、病程、Killip分级>Ⅱ级、LVEF下降是中青年AMI伴心律失常的独立危险因素(P<0.05);森林图显示,前壁心肌梗死、左侧冠状动脉病变、冠状动脉重度狭窄/闭塞、病程、Killip分级>Ⅱ级、左心室射血分数(LVEF)下降均为正相关危险因素,与无效线不相交,均P<0.05;以森林图所示6个独立危险因素构建风险预警模型,似然比卡方、Wald卡方、Hosmer-Lemeshow拟合优度检验显示模型建立具有统计学意义、回归方程的系数差异有统计学意义、模型拟合效果较好;ROC曲线显示,AUC为0.914(95%CI:0.871~0.946),敏感度为82.35%,特异度为86.21%。结论基于森林图构建中青年AMI伴心律失常患者的早期风险预警模型具有良好的预测价值,能为临床及早采取针对性防治措施提供可靠依据。

Objective To explore the establishment of an early risk warning model for young and middle-aged acute myocardial infarction(AMI)patients with arrhythmia based on forest graph.Methods A total of 242 young and middleaged acute myocardial infarction(AMI)patients in our hospital from November 2020 to December 2021 were selected and divided into observation group(68 cases,with arrhythmia)and control group(174 cases,without arrhythmia).Two groups of clinical data were collected,univariate analysis was performed,logistic regression was used for multivariate analysis,forest plots were drawn,risk early warning models were constructed,and receiver operating characteristic(ROC)curves were used to analyze the predictive value of risk early warning models.Results Univariate analysis showed that myocardial infarction site,coronary artery lesion location,degree of coronary stenosis,Killip grade>Ⅱ,disease duration,and decreased LVEF were the influencing factors for young and middle-aged AMI with arrhythmia(P<0.05).Anter;or myocardial infarction,left coronary artery disease,severe coronary artery stenosis/occlusion,disease course,Killip grade>Ⅱ,and decreased LVEF were independent risk factors for AMI with arrhythmia in young and middleaged people(P<0.05).Forest plot showed that anterior myocardial infarction,left coronary artery disease,severe coronary stenosis/occlusion,disease duration,Killip grade>Ⅱ,and LVEF decline were all positively correlated risk factors,which did not intersect with the invalid line,all P<0.05.The risk early warning model was constructed with 6 independent risk factors shown in the forest plot.The likelihood ratio chi-square,Wald chi-square,and Hosmer-Lemeshow goodness-of-fit test showed that the model establishment was statistically significant,and the coefficients of the regression equation had statistically significant differences,the model fitting effect was good;ROC curve,AUC was 0.914(95%CI:0.871-0.946),the sensitivity was 82.35%,and the specificity was 86.21%.Conclusion The establishment of an early risk early warning model for young and middle-aged AMI patients with arrhythmia based on forest map has good predictive value,and can provide a reliable basis for early clinical application of targeted preventive measures.