急性心肌梗死患者经皮冠状动脉介入术后心力衰竭的预测模型及效能评价

A predictive model of heart failure and performance evaluation in acute myocardial infarction patients after percutaneous coro-nary intervention

目的基于心肌做功超声显像技术建立急性心肌梗死(AMI)患者行急诊经皮冠状动脉介入(PCI)术后新发心力衰竭的预测模型。方法回顾性纳入2021年1月至2023年3月金华市中心医院确诊AMI并接受急诊PCI患者193例,出院后常规随访6个月,根据左心室射血分数(LVEF)<50%分为心力衰竭组31例和无心力衰竭组162例。比较两组患者临床资料、常规超声检测指标[左心室舒张末内径(LVEDd)和左心室舒张末容积(LVEDV)、左心室收缩末内径(LVESd)和左心室收缩末容积(LVESV)]、二维斑点追踪参数整体纵向应变(GLS)及心肌做功参数[整体做功指数(GWI)、整体有用功(GCW)、整体无用功(GWW)和整体做功效率(GWE)]。采用最小绝对收缩和选择算法(LASSO)和多因素logistic回归筛选影响因素,构建列线图预测模型,Bootstrap法计算一致性指数,绘制ROC曲线计算AUC,采用校准曲线和决策曲线进行预测效能验证。结果与无心力衰竭组比较,心力衰竭组年龄、峰值肌钙蛋白I(cTnI)、肌酸激酶同工酶(CK-MB)、GLS和GWW均增加,而完全血运重建率、GWI、GCW、GWE均降低(均P<0.05)。LASSO回归和多因素logistic回归分析显示,峰值cTnI、GLS和GWW是AMI患者急诊PCI术后6个月新发心力衰竭的危险因素,而GWI是保护因素(均P<0.05)。Bootstrap法计算列线图一致性指数为0.856;ROC曲线分析列线图模型的AUC为0.893(95%CI:0.842~0.935,P<0.001),提示列线图模型的预测效能较好;校准曲线显示有较好的吻合度,决策曲线显示有较好的临床净获益比。结论AMI患者急诊PCI术后仍有一定的新发心力衰竭风险,二维超声斑点追踪和无创心肌做功超声显像技术能够较好地评估心肌舒缩功能;列线图模型对指导临床早期识别新发心力衰竭的高危人群具有重要的实践价值。

Objective To establish a predictive model and evaluate its diagnostic performance for new onset heart failure after emergent percutaneous coronary intervention(PCI)in patients with acute myocardial infarction(AMI)based on myocardial work ultrasound imaging technology.Methods A total of 193 patients diagnosed with AMI and underwent emergent PCI in Jinhua Municipal Central Hospital were included from January 2021 to March 2023 in this retrospective study.They were followed up for 6 months after discharge,and divided into new onset heart failure group(n=31)and non-heart failure group(n=162)based on left ventricular ejection fraction(LVEF)<50%.The two groups were compared in general clinical data,conventional ultrasound indicators[left ventricular end diastolic diameter(LVEDd)and left ventricular end diastolic volume(LVEDV),left ventricular end systolic diameter(LVESd)and left ventricular end systolic volume(LVESV)],two-dimensional speckle tracking parameters[global longitudinal strain(GLS)],and myocardial work parameters[global work index(GWI),global constructive work(GCW),global waste work(GWW),and global work efficiency(GWE)].Then least absolute shrinkage and selection operator(LASSO)regression model and multivariate logistic regression were applied to screen risk factors so as to construct the nomogram prediction model.Bootstrap method was used to calculate consistency index,ROC curve was drawn to calculate AUC,and calibration curve and decision curve were used for predictive performance validation.Results The age,peak cardiac troponin I(cTnI),creatine kinase isoenzyme(CK-MB),GLS,and GWW in heart failure group were significantly higher than those of the non-heart failure group(all P<0.05),while complete revascularization rate,and GWI,GCW,and GWE values were lower(all P<0.05).LASSO regression and multivariate logistic regression showed that peak cTnI,GLS,and GWW were independent risk factors for new onset heart failure in AMI patients 6 months after emergent PCI,while GWI was a protective factor(all P<0.05).The consistency index calculated by Bootstrap method was 0.856,and AUC calculated by ROC was 0.893(95%CI:0.842-0.935,P<0.001),indicating good predic-tive performance of the nomogram.Calibration curve showed the model was in good agreement with the actual occurrence rate;and decision curve showed the model had good clinical net benefit ratio.Conclusion Patients with AMI still have a certain risk of new onset heart failure after emergency PCI,and two-dimensional ultrasound speckle tracking and non-invasive myocardial work imaging technique could effectively evaluate myocardial systolic and diastolic functions;the nomogram model has important practical values in guiding clinical early identification of high-risk populations for new onset heart failure.