Predictive value of left atrial epicardial adipose tissue volume for atrial fibrillation recurrence after radiofrequency ablation
K Aytemir, A Keresteci, C Sezer, C Coteli, H YorgunAbstract
Background
Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia associated with increased morbidity and mortality. Catheter ablation, particularly pulmonary vein isolation (PVI) using radiofrequency (RF) energy, is an established treatment for rhythm control; however, recurrence remains a major limitation. Epicardial adipose tissue (EAT), a metabolically active and inflammatory depot surrounding the myocardium, has been implicated in atrial remodeling and fibrosis through paracrine mechanisms. Recent imaging studies using cardiac computed tomography (CT) have suggested an association between increased EAT and AF recurrence, but the prognostic role of left atrial (LA) EAT volume quantified by three-dimensional (3D) imaging remains unclear.
Objective
To investigate whether LA EAT volume and area quantified by ADAS 3D software on pre-procedural CT are associated with AF recurrence after RF ablation.
Methods
This study included patients with AF who underwent radiofrequency ablation between November 2019 and December 2023 at our center. Pre-procedural cardiac CT scans were retrospectively analyzed using ADAS 3D software to quantify LA EAT volume and area. Measurements were performed by semi-automated segmentation of the left atrium in multiplanar reconstructions. Clinical, echocardiographic, and procedural data were collected, and patients were followed for AF recurrence after ablation.
Results
A total of 101 patients (median age 60 years, 48% female) were included. During a mean follow-up of 32.9 ± 13.8 months, AF recurrence occurred in 28 patients (27.7%). Patients who experienced recurrence had significantly greater LA EAT volume (2.32 ± 0.86 vs 1.86 ± 0.59 cm³, p = 0.003), LA EAT area (17.67 ± 3.33 vs 13.93 ± 1.63 cm², p = 0.007), and LA area (202.99 ± 31.15 vs 189.87 ± 28.85 mm², p = 0.048). In univariate Cox regression analyses, both LA EAT volume (HR = 1.999, 95% CI 1.236–3.232, p = 0.005) and LA area (HR = 1.015, 95% CI 1.003–1.027, p = 0.014) were significantly associated with AF recurrence. In the multivariate model, LA EAT volume remained an independent predictor of recurrence (HR = 1.722, 95% CI 1.014–2.926, p = 0.044). All procedures were performed using radiofrequency ablation (100%), predominantly with the Carto 3D mapping system (90%). Additional ablation lines included anterior (22.7%), roof (7.9%), and posterior box (12.8%) isolation. Procedure-related complications were rare, consisting of one pericardial effusion, one transient phrenic nerve palsy, two vascular access-site complications, and one pulmonary vein stenosis.
Conclusion
In patients undergoing radiofrequency ablation for atrial fibrillation, increased LA EAT volume was independently associated with AF recurrence during long-term follow-up. Assessment of LA EAT by cardiac imaging may provide additional prognostic value in the preprocedural evaluation of AF ablation candidates.