Redistribution of adipose tissue is associated with left atrial remodeling and dysfunction in patients with atrial fibrillation.

Objective: Adipose tissue is recognized as a crucial regulator of atrial fibrillation (AF). However, the effect of epicardial adipose tissue (EAT) on the pathophysiology of AF might be different from that of other adipose tissues. The purpose of this study was to explore the distribution features of different adipose tissues in AF patients and their relationships with left atrial (LA) remodeling and function.

Methods: A total of 205 participants (including 112 AF and 93 non-AF patients) were recruited. Color doppler ultrasound was used to measure the thickness of subcutaneous, extraperitoneal, and intra-abdominal adipose tissue. Cardiac CT scan was performed to measure the mean thickness of EAT surrounding the whole heart (total-EAT) and specific regions, including left atrium (LA-EAT), left ventricle, right ventricle, interventricular groove, and atrioventricular groove. LA anatomical remodeling and function were measured by echocardiography, while electrical remodeling was evaluated by P-wave duration and dispersion using Electrocardiography (obtained after cardioversion or ablation in AF patients). Relationship between the thickness of different adipose tissues and LA remodeling and function was analyzed.

Results: The thickness of subcutaneous, extraperitoneal, and intra-abdominal adipose tissue was similar between AF and non-AF patients, and had no or only weak association with LA remodeling and dysfunction. However, compared to non-AF participants, total-EAT thickness significantly increased in both paroxysmal and persistent AF patients (non-AF vs. paroxysmal AF vs. persistent AF: 6.31 ± 0.63 mm vs. 6.76 ± 0.79 mm vs. 7.01 ± 1.18 mm, < 0.001), which was positively correlated with the LA size and P-wave duration and dispersion, and negatively correlated with LA ejection fraction and peak strain rate. More interestingly, EAT thickness in AF patients did not increase uniformly in different regions of the heart. Compared to EAT surrounding the other regions, LA-EAT was found to accumulate more greatly, and had a closer relationship to LA remodeling and dysfunction. Multivariate logistic regression analysis also showed that LA-EAT was significantly correlated with the presence of AF (OR = 4.781; 95% CI 2.589-8.831, < 0.001).

Conclusion: Rather than other adipose tissues, accumulation and redistribution of EAT, especially surrounding the LA, is associated with LA remodeling and dysfunction in AF patients.