Impact of Dapagliflozin on MRI-Derived Epicardial Adipose Tissue: Secondary Imaging Analysis of a Randomized Trial

Why clinicians should know about it

• Picked for Anatomy (top studies of the week, 30 August 2026): Focus not on gross anatomy or surgical landmarks
• Picked for Radiology, Radiation Oncology, Nuclear Medicine, Medical Physics and Imaging (top studies of the week, 30 August 2026): Ranked by evidence level and journal quartile

Abstract

BACKGROUND: Epicardial adipose tissue (EAT) is a metabolically active fat depot associated with cardiovascular risk and represents a modifiable imaging biomarker. Studies have shown that sodium-glucose cotransporter-2 (SGLT2) inhibitors reduce EAT thickness in patients with type 2 diabetes mellitus (T2DM). However, it is unclear whether this is a direct effect on EAT or is secondary to concomitant weight loss. PURPOSE: To evaluate whether the SGLT2 inhibitor dapagliflozin reduces EAT independently of weight change in patients with T2DM without overt cardiovascular disease. STUDY TYPE: Secondary longitudinal imaging analysis of a prospective, randomized, placebo-controlled trial. POPULATION: Fifty-six adults with T2DM without overt cardiovascular disease were randomized to dapagliflozin 10 mg daily (N = 27) or placebo (N = 29) for 12 months. FIELD STRENGTH/SEQUENCE: Balanced steady-state free precession cine at 3T. ASSESSMENT: EAT thickness was quantified at eight predefined anatomical locations on MR images acquired at baseline and 12 months. Body weight, body mass index (BMI), cardiovascular medication history, and C-reactive protein were also assessed. The primary endpoint was change in average EAT thickness. STATISTICAL TESTS: Between-group comparisons were performed using the Mann-Whitney U test. Mediation and multivariable regression analyses assessed whether EAT change was independent of weight change. A p value < 0.05 was considered significant. RESULTS: Dapagliflozin significantly reduced average EAT thickness compared with placebo (median change -0.8 mm [interquartile range (IQR): -1.3 to -0.4] vs. 0.1 mm [IQR: -0.1 to 0.3]). Mediation analysis demonstrated a significant direct treatment effect not mediated by weight change [direct effect coefficient 0.81, 95% confidence interval (CI), 0.61-1.04] and a non-significant indirect effect (indirect effect coefficient 0.15, 95% CI, -0.17 to 0.42, p = 0.29). The association remained significant after adjustment for baseline EAT, baseline and change in body weight or BMI, age, sex, baseline use of any cardiovascular medication, and C-reactive protein. DATA CONCLUSION: Dapagliflozin reduced EAT in patients with T2DM without overt cardiovascular disease, predominantly independent of weight change, supporting MRI-derived EAT as a treatment-responsive imaging biomarker. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: Stage 4. TRIAL REGISTRATION: https://clinicaltrials.gov; Unique identifier: NCT03782259.

Abstract as published, via PubMed.