Abstract
BACKGROUND AND PURPOSE: Epicardial adipose tissue (EAT) comprises three distinct lipid components, each exerting differential effects on cardiovascular diseases. During disease progression, dynamic alterations in lipid composition and spatial distribution contribute to the inherent heterogeneity of EAT. The excessive activation of inflammatory cells may contribute to chronic inflammation, promoting atherosclerosis and cardiac diseases. However, the role of EAT in patients with myocardial infarction (MI) who develop heart failure with preserved ejection fraction (HFpEF) remains unclear. This study aims to quantify the overall and perivascular volumes of EAT using cardiac magnetic resonance (CMR) imaging and assess its heterogeneity, exploring the predictive value of EAT heterogeneity and different EAT volumes combined with inflammatory cells for the occurrence of HFpEF in MI patients with normal left ventricular ejection fraction (LVEF). METHODS: This retrospective cohort study enrolled patients diagnosed with MI with preserved LVEF via clinical assessment and CMR at the Second Affiliated Hospital of Kunming Medical University between January 2015 and July 2023. Patients who did not undergo percutaneous coronary intervention (PCI) were followed, with the incidence of HFpEF serving as the primary endpoint. The cohort was stratified into two groups: those without HFpEF and those who developed HFpEF.Cardiac structure, function, EAT volume, and infarct volume parameters were obtained using the CMR post-processing software CVI-42, while EAT heterogeneity parameters entropy were derived using Python software. Independent sample t-tests, non-parametric tests, and chi-square tests were employed to analyze the differences in clinical baseline data and CMR metrics between the two groups. Spearman's rank correlation was utilized to analyze the associations between EAT parameters and inflammatory cells, inflammatory markers, and diastolic dysfunction indicators. Furthermore, we conducted univariate and multivariate Cox regression analyses to determine the predictive value of each parameter for the development of HFpEF in MI patients. Time-dependent ROC curves were generated to evaluate the efficacy of each parameter in predicting HFpEF, the AIC values of each parameter and the final model were calculated to evaluate the predictive performance. The optimal cut-off values were identified using time-dependent ROC curves in R software, and Kaplan-Meier event-survival curves were plotted to illustrate the event-free rates based on these optimal thresholds.The median follow-up time was calculated using the reverse Kaplan-Meier method. RESULTS: A total of 203 MI patients with normal LVEF were included, with 74 in the HFpEF group and 129 in the non-HFpEF group. No significant differences were observed between the two groups regarding age, sex, and infarct volume; however, significant statistical differences were noted in BMI, diabetes, renal failure, leukocytes, neutrophils, monocytes, total EAT, EAT entropy, left ventricular EAT (LV EAT), left atrial end-systolic volume (LAESV), triglycerides, NHR, MHR and LACI(Left atrioventricular coupling index) (P < 0.05). Both overall and local EAT volumes showed a positive correlation with leukocytes and monocytes,as well as with the inflammatory markers MHR and SIRI. Furthermore, EAT volume exhibited a positive correlation with the LACI, a marker of diastolic dysfunction. Univariate and multivariate Cox regression analyses indicated that BMI, diabetes, monocyte, LV EAT, and EAT entropy are independent risk factors for HFpEF. And the AIC value of the multivariate regression model was the smallest.Further time-dependent ROC analysis revealed that the maximum AUC for BMI was 0.67, while the AUC for LV EAT was 0.63, and EAT entropy was 0.60, the maximum AUC for monocyte was 0.70, and the combined prediction of LV EAT and EAT entropy had a maximum AUC of 0.70. After a median follow-up of 34 months, Kaplan-Meier survival curves demonstrated that LV EAT greater than 21.23 mL was associated with the occurrence of HFpEF, whereas EAT entropy was not. CONCLUSIONS: In patients with chronic MI, normal LVEF, and no prior PCI, the occurrence of HFpEF is not correlated with infarct volume; however, BMI, diabetes, monocyte, LV EAT, and EAT entropy are independent risk factors for HFpEF with significant predictive value, with the highest predictive efficacy observed monocyte and when combining EAT entropy and LV EAT. Additionally, both overall and local EAT volumes exhibit a moderate positive correlation with leukocytes,monocytes and inflammatory markers, and were also positively correlated with diastolic dysfunction. This suggests that, in clinical practice, beyond traditional indicators, there should be an increased focus on EAT heterogeneity and perivascular EAT in MI patients with normal LVEF who have not undergone PCI to to reduce the incidence of HFpEF.