A multilevel meta-regression of 42 interventional studies (50 arms, 1,890 participants) found that BMI reduction explains only a modest fraction of epicardial adipose tissue (EAT) change (β = 0.19, p = 0.03). Intervention class itself was a stronger moderator (p < 0.001), with GLP-1 receptor agonists, dietary interventions, and bariatric surgery producing EAT reductions exceeding what their degree of BMI loss would predict.
Epicardial fat is not passive padding — it sits directly on the myocardium and coronary arteries, secreting pro-inflammatory cytokines and contributing to atrial fibrillation, coronary artery disease, and heart failure risk independent of general obesity metrics. The finding that GLP-1 agonists like semaglutide appear to preferentially mobilize this depot aligns with emerging evidence that these drugs exert direct cardioprotective effects beyond glycemic control and weight reduction — a mechanistic story that may partly explain the outsized cardiovascular event reductions seen in trials like LEADER and SELECT.
Critical caveats apply: the evidence certainty is rated low to very low, the intervention comparisons are exploratory rather than confirmatory, and the 1,890-participant pool is modest for meta-regression. Confounding by diet quality, baseline EAT volume, and imaging modality heterogeneity (echocardiography vs. CT vs. MRI) limits precision. Still, the paradigm shift here — that EAT remodeling is a partly intervention-specific biological process, not merely a surrogate of scale weight — is conceptually important and warrants dedicated randomized trials with EAT as a primary endpoint.
