By Joshua D. Moss, MD

Associate Professor of Clinical Medicine, Cardiac Electrophysiology, Division of Cardiology, University of California, San Francisco

Dr. Moss reports no financial relationships relevant to this field of study.

SYNOPSIS: In patients with persistent but rate-controlled atrial fibrillation and left ventricular systolic dysfunction of otherwise uncertain etiology, catheter ablation to restore sinus rhythm can result in significant improvement or normalization of ejection fraction.

SOURCE: Prabhu S, Taylor AJ, Costello BT, et al. Catheter Ablation Versus Medical Rate control in Atrial Fibrillation and Systolic Dysfunction (CAMERA-MRI). J Am Coll Cardiol 2017 Aug 22. pii: S0735-1097(17)39349-X. doi: 10.1016/j.jacc.2017.08.041. [Epub ahead of print].

Atrial fibrillation (AF) and heart failure frequently coexist, and cardiologists often debate which disease preceded the other in a given patient, or whether they developed in parallel. A trial of pharmacologic rhythm control (predominantly with amiodarone) in patients with heart failure did not show superiority over rate control, but antiarrhythmic drugs are known to produce limited efficacy and their own deleterious effects. Prior trials of catheter ablation for AF in patients with heart failure from a variety of etiologies have shown mixed results.

In this multicenter randomized trial, investigators sought to determine whether restoration of sinus rhythm would be beneficial in a more homogeneous population of patients with persistent AF and left ventricular systolic dysfunction; specifically, patients with cardiomyopathy unexplained by ischemic heart disease, valvular disease, or other known causes. Over three years, 68 patients with left ventricular ejection fraction (LVEF) ≤ 45%, persistent AF, and at least New York Heart Association (NYHA) class II symptoms were randomized to either catheter ablation or medical rate control, and 66 were analyzed. Cardiac MRI with late gadolinium enhancement (LGE) imaging was performed prior to randomization and after four weeks of initial rate control, targeting an average ventricular rate < 100 bpm on 24-hour Holter monitoring. Thereafter, catheter ablation (CA) was performed within one month of randomization, or medical rate control (MRC) was titrated to achieve resting heart rate < 80 bpm, average 24-hour ventricular rate < 100 bpm, and post-exercise rate < 110 bpm. Recurrence of AF after ablation was monitored via an implantable loop recorder.

In the CA group, single-procedure freedom from atrial arrhythmias was 75% (including patients on antiarrhythmic drugs), with an average AF burden at six months of 1.6 ± 5.0%. In the MRC group, average ventricular rates were well controlled throughout (80 ± 10 bpm at six months), although significantly higher than the average sinus rates achieved after CA. The CA group demonstrated improvement in LVEF by 18.3% vs. 4.4% for the MRC group.

By six months, 58% of patients undergoing CA had recovered LVEF to > 50%, while only 9% of the MRC group demonstrated similar results. Patients in the CA group also exhibited significant decreases in left ventricular end-systolic volume, left atrial volume, serum B-type natriuretic peptide (BNP), and NYHA class compared with the MRC group. Patients without significant LGE on cardiac MRI were much more likely to demonstrate normalization of LV function, and absence of LGE was the only predictor of normalization after CA in a multivariate analysis. Complications after CA were rare.

The authors concluded that AF is an underappreciated reversible cause of left ventricular systolic dysfunction in patients with unexplained systolic heart failure despite adequate rate control.


This study was the first randomized trial of CA for rhythm control vs. medical rate control of persistent AF in a population of patients with exclusively idiopathic cardiomyopathy. The remarkable results, highly significant despite the small study size, suggested an entity the authors referred to as “arrhythmia-mediated cardiomyopathy.” This was distinct from cardiomyopathy resulting from tachycardia — robust maintenance of sinus rhythm with CA (and antiarrhythmic drugs as needed) resulted in significant improvements in LVEF that were not approached by rate control alone. The absence of LGE on cardiac MRI predicted better response, with 73% of LGE-negative patients undergoing ablation, demonstrating normalization of LV systolic function (compared with 21% of LGE-positive patients).

Obvious strengths of the study included the randomization after four weeks of rate control and cardiac MRI, with apparently even distribution of baseline characteristics between groups (although somewhat surprisingly, statistical baseline differences were not reported explicitly). Additionally, continuous follow-up rhythm monitoring, as well as multiple other useful assessments, such as six-minute walk tests, BNP levels, symptom surveys, and repeat MRI, were performed.

Several limitations that do not call into doubt the plausibility of the results but may have enhanced the magnitude of the effect should be considered. First, the AF burden after CA was quite low, particularly for patients with previously persistent AF; real-life results may not be as favorable, especially over a longer period of follow-up. Second, rate control in the MRC group was assessed via 24-hour Holter monitors at baseline, three months, and six months, and during and after six-minute walk tests. These values may not accurately reflect true long-term average heart rates for a patient unencumbered by a monitor. Third, patients with MRI-incompatible devices were excluded from randomization, which may have added selection bias. Finally, the study was small and underpowered to detect differences in hospitalizations and mortality.

Nevertheless, cardiologists and electrophysiologists managing patients with AF and otherwise unexplained reduced LVEF should strongly consider the potential benefits and risks of CA for restoration of sinus rhythm, even if the heart rate in AF already is well controlled.

Especially for patients without LGE on cardiac MRI, the AF itself may indeed represent the sole cause of left ventricular systolic dysfunction. Maintenance of sinus rhythm, with minimization of potentially harmful antiarrhythmic drugs as feasible, may yield significant improvement in or even normalization of LVEF.