The epicardial adipose tissue confined in the atrioventricular groove can be used to assess atrial adipose tissue and atrial dysfunction in cardiac magnetic resonance imaging

Aims The growing interest in epicardial adipose tissue (EAT) as a biomarker of atrial fibrillation is limited by the difficulties in isolating EAT from other paracardial adipose tissues. We tested the feasibility and value of measuring the pure EAT contained in the atrioventricular groove (GEAT) using cardiovascular magnetic resonance (CMR) imaging in patients with distinct metabolic disorders. Methods and results CMR was performed on 100 patients from the MetaCardis cohort: obese (n = 18), metabolic syndrome (MSD) (n = 25), type-2 diabetes (T2D) (n = 42), and age- and gender-matched healthy controls (n = 15). GEAT volume measured from long-axis views was obtained in all patients with a strong correlation between GEAT and atrial EAT (r = 0.95; P < 0.0001). GEAT volume was higher in the three groups of patients with metabolic disorders and highest in the MSD group compared with controls. GEAT volume, as well as body mass and body fat, allowed obese, T2D, and MSD patients to be distinguished from controls. GEAT T1 relaxation and peak longitudinal left atrial (LA) strain in CMR were decreased in T2D patients. Logistic regression and random forest machine learning methods were used to create an algorithm combining GEAT volume, GEAT T1, and peak LA strain to identify T2D patients from other groups with an area under curve (AUC) of 0.81 (Se: 77%, Spe: 80%; 95% confidence interval 0.72–0.91, P < 0.0001). Conclusion Atrioventricular groove adipose tissue characteristics measured during routine CMR can be used as a proxy of atrial EAT and integrated in a multi-parametric CMR biomarker for early identification of atrial cardiomyopathy.