The forgotten chamber, epicardial fat thickness as an independent predictor of ventricular arrythmias; a retrospective single center imaging-based analysis
M Abdelfattah, M Wanees Ahmed El Husseny, E Ahmed, M Abdelfadil, M S Riad, A Omar, M Abdelfattah, P S Karam, A Abdelsamed, T Abd-Elsalam Ashraf Taha, O KamelAbstract
Introduction
Epicardial adipose tissue (EAT) is a biologically active, pro-inflammatory fat depot located between the myocardium and the visceral pericardium. Unlike other visceral fats, EAT is in direct anatomical contact with the myocardium, often penetrating it as fatty infiltrates, and shares a common blood supply (1 ). EAT secretes numerous adipocytokines and pro-inflammatory mediators that can promote cardiac fibrosis and arrhythmogenesis. While the relationship between EAT and atrial fibrillation is well-studied, its specific association with ventricular arrhythmia (VA) in patients with non-ischemic cardiomyopathy (NICM) remains less clear (2).
Purpose
We aimed to assess the association between EAT thickness, measured by cardiovascular magnetic resonance imaging (CMR), and the incidence of ventricular arrhythmia in patients with non-ischemic cardiomyopathy.
Methods
This was a retrospective cohort study utilizing our database for non-ischemic cardiomyopathy. We enrolled patients with NICM (LVEF ≤ 40%) and compared those with documented VA (n=27) to a control group without VA (n=50). EAT thickness was measured on end-diastolic CMR images at the right and left atrioventricular grooves (AVGs), interventricular grooves (IVGs) (anterior, inferior and superior), and the right ventricular free wall (Figure 1). Univariate and multivariate logistic regression analyses were performed to identify predictors of VA. Multivariate models were adjusted for age, body mass index (BMI), left ventricular ejection fraction (LVEF), presence of late gadolinium enhancement (LGE), and use of mineralocorticoid receptor antagonists (MRA).
Results
The VA group had a significantly higher prevalence of syncope of arrhythmogenic origin (25.9% vs. 0%, p<0.001) and LGE presence (65.4% vs. 40%, p=0.04). The VA group also had significantly lower use of MRAs (77.8% vs. 98%, p=0.007) and SGLT2 inhibitors (74.1% vs. 100%, p<0.001). Patients with VA had significantly greater EAT thickness in the Right AVG (11.2 mm vs. 9 mm, p=0.019), Total AVG Score (22 vs. 18.3, p=0.011), Total IVG Score (13.6 vs. 11.5, p=0.031), and Total EAT Score (39.3 vs. 33, p=0.003). In multivariate analysis, Total EAT Score (aOR 1.019, 95% CI 1.017–1.17, p=0.014) and Total AVG Score (aOR 1.13, 95% CI 1.02–1.25, p=0.026) remained independent predictors of VA. LGE presence and MRA non-use were also consistent independent predictors (Table 1).
Conclusion
Increased epicardial adipose tissue, particularly in the atrioventricular grooves as quantified by the Total EAT and Total AVG Scores, is independently associated with ventricular arrhythmia in patients with non-ischemic cardiomyopathy. This finding suggests a synergistic interplay between EAT-driven pro-fibrotic processes and the underlying arrhythmogenic substrate, evidenced by the co-association with LGE. CMR-based EAT assessment could serve as a novel imaging tool for improved arrhythmic risk stratification in such patients.Epicardial Fat measuremnet on CMRMultivariate Logistic Regression Models