A practical ECG-based model for early identification of acute heart failure following acute myocardial infarction

Abstract

Ischemic infarction of the acute myocardium (AMI) is a major cause of acute heart failure (AHF), which markedly increases mortality and readmission rates. One purpose of this study was to establish an ECG-based model for the early prediction of AHF risk in AMI patients. In a retrospective analysis, 301 patients admitted to the hospital (October 2022–March 2025) for AMI met the study criteria. With LASSO, we identified six predictors—prolonged QTc interval, abnormal Q wave, heart rate > 100 bpm, reduced left ventricular ejection fraction (LVEF) from routine echocardiography, male sex, and age 60–75 years—and entered them into a logistic regression nomogram. The model showed good discrimination (AUC = 0.84; internal validation AUC = 0.823) and was validated internally with 2000 bootstrapping iterations; external multicenter validation is planned to confirm its generalizability. The QTc interval and heart rate were positively correlated with Killip grade, and age and QRS duration were negatively correlated. However, P-wave duration and dispersion were negatively correlated with Killip grade. This ECG-based nomogram offers a simple, low-cost and practical AHF early identification tool for patients with a history of AMI, especially those in primary healthcare settings and those who live away from cities.