The paradox of the association between epicardial adipose tissue, infarct size and left ventricular remodelling after ST-segment elevation myocardial infarction: a CMR study

Abstract Background Epicardial adipose tissue (EAT) is a biologically active fat deposit contained below the visceral pericardium. In latest years, EAT has emerged as a predictor of atherosclerosis, coronary artery disease and adverse cardiovascular events. Nevertheless, EAT significance after ST-segment elevation myocardial infarction (STEMI) in acute phase and its association with left ventricular remodelling is unknown. Purpose We aimed to assess the association of cardiac magnetic resonance (CMR)-derived EAT with the extent of infarct size (IS), microvascular obstruction (MVO) and left ventricular remodelling in a homogeneous cohort of STEMI patients. Methods We prospectively included 255 patients discharged after a first reperfused, anterior STEMI. An early CMR (1-week after STEMI) was performed in all patients. EAT volume (mL/m2), left ventricular (LV) end-diastolic and end-systolic indexed volumes (LVEDiV and LVESiV, mL/m2), left ventricular ejection fraction (LVEF, %), infarct size (IS, % of LV mass) and microvascular obstruction (MVO, % of LV mass) were quantified. A subgroup of 195 patients underwent a follow-up (6-month) CMR. Patients were categorized as extensive EAT (>35 mL/m2) or non-extensive EAT (≤35 mL/m2) according to the upper tertile of this parameter. A specific analysis was performed in patients with extensive IS (>30% of LV mass). Results Patients with extensive EAT were older (62±12 vs. 57±13 years, p=0.004), had an increased prevalence of diabetes mellitus (31% vs. 16%, p=0.007) and depicted a worse baseline risk profile (GRACE risk score 138±37 vs. 126±36 points, p=0.02). At 1-week CMR, patients with extensive EAT displayed larger IS (31±17% vs 24±16%, p=0.001), more MVO (1.4 [0-6.2]% vs 0 [0-2.5]% of LV mass, p=0.022) and more myocardial oedema (39±16% vs 32±17% of LV mass, p=0.003). After adjustment for baseline characteristics, EAT was independently associated with IS (standardized beta coefficient=0.30, p<0.001). However, patients with extensive EAT did not show neither larger LV volumes nor more depressed LVEF (p>0.1 for all comparisons). In the extensive IS subgroup (n=69, 27%), patients with extensive EAT depicted less dilated ventricular volumes (LVEDiV 90±26 vs 111±40 mL/m2, p=0.011; LVESiV 51±26 vs 72±36 mL/m2, p=0.009) and higher LVEF (46±13 vs. 38±10%, p=0.007) at 6-month CMR. At 6-month CMR, in a propensity score-matched population (n=170, 85 with extensive EAT and 85 without extensive EAT), EAT was independently associated with more preserved LVEF, and less dilated LVEDiV and LVESiV only in the subgroup of patients with extensive IS (p<0.05 for all comparisons) (Figure 1). Conclusions EAT seems to exert a bimodal effect in patients with anterior STEMI. More extensive EAT is associated with more extensive infarct size in acute phase but, in patients with large IS, those with extensive EAT display less adverse remodelling during follow-up. This "paradox" of EAT after STEMI merits further investigation.Figure 1.