Catheter Ablation Versus Medication in Atrial Fibrillation and Systolic Dysfunction

Objectives This study sought to determine the long-term outcomes of restoring sinus rhythm with catheter ablation (CA). Background The CAMERA-MRI (Catheter Ablation Versus Medical Rate Control in Atrial Fibrillation and Heart Failure—An MRI-Guided Multicenter Randomized Controlled Trial) study demonstrated that restoration of sinus rhythm with CA significantly improved left ventricular ejection fraction (LVEF) compared with medical rate control (MRC) at 6 months in persistent atrial fibrillation and otherwise unexplained systolic heart failure. However, the long-term outcomes have not been reported. Methods Patients enrolled in the CAMERA-MRI study were followed for 4 years with echocardiogram and cardiac magnetic resonance. CA involved pulmonary vein isolation and posterior left atrial wall isolation in 94%. Patients crossed over to CA after 6-month study duration. Arrhythmia burden was determined with implanted cardiac monitors or cardiac devices. Results Sixty-six patients (age 62 ± 10 years, atrial fibrillation duration of 22 ± 16 months, and LVEF 33 ± 9%) were randomized 1:1 to CA versus MRC. Eighteen of 33 patients crossed over from MRC group to CA group. At 4.0 ± 0.9 years, atrial fibrillation recurred in 27 patients (57%) in the CA group with a mean burden of 10.6 ± 21.2% after 1.4 ± 0.6 procedures. There was an absolute increase in LVEF with CA of 16.4 ± 13.3% compared with 8.6 ± 7.6% in MRC (p = 0.001). In the CA group, the absence of ventricular late gadolinium enhancement was associated with a greater improvement in absolute LVEF (19 ± 13% vs. 10 ± 11% in the late gadolinium enhancement–positive group; p = 0.04) and LVEF normalization in 19 patients (58%) versus 4 patients (18%) in the late gadolinium enhancement–positive group (p = 0.008) at 4.0 ± 0.9 years follow-up. Conclusions CA is superior to MRC in improving LVEF in the long term in patients with atrial fibrillation and systolic heart failure. The greatest recovery in systolic function was demonstrated in the absence of ventricular fibrosis on cardiac magnetic resonance.