CMR—Quantified Epicardial Adipose Tissue Heterogeneity and Its Predictive Value for Ventricular and Atrial Arrhythmias After Myocardial Infarction

Background Epicardial adipose tissue (EAT) mediate both electrophysiological disturbances and structural remodeling within substrates. Fibrotic remodeling within EAT under pathological conditions revealed the presence of heterogeneity. Methods and Results The cohort study included 241 consecutive post-myocardial infarction (MI) patients, 49 experienced ventricular arrhythmias (VAs) VAs and 30 experienced atrial tachyarrhythmias (AAs) during the follow-up period. EAT volume, myocardial scar, functional and strain parameters were obtained using CVI42 workstation. EAT heterogeneity was calculated using the entropy formula in Python. Patients in the VAs(+) group showed impaired cardiac pumping function, reduced left ventricular (LV) strain, and a greater extent of myocardial fibrosis. Similarly, patients with elevated left atrial (LA) strain, left atrioventricular coupling index (LACI), total EAT volume, right ventricular (RV) EAT volume, and EAT entropy were more likely to develop AAs. Myocardial fibrosis exhibited modest correlations with EAT entropy. Multivariable stepwise regression models identified EAT entropy, LA storage period strain (Es), infarct core (IC) percentage, and global longitudinal strain (GLS) as independent predictors of VAs. EAT entropy, Es, and EAT thickness were predictors of AAs. Time-dependent receiver operating characteristic (ROC) curves demonstrated that the predictive performance for VAs improved progressively with longer follow-up durations. Conclusion CMR-quantified EAT entropy is a significant indicator for predicting VAs and AAs after MI and shows a linear correlation with myocardial fibrosis.