By Joshua Moss, MD

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

Dr. Moss reports he is a consultant for Abbott and Biosense Webster. The relevant financial relationships listed have been mitigated.

SYNOPSIS: In two randomized trials published simultaneously, cryoballoon ablation proved superior to drug therapy for prevention of arrhythmia recurrence in patients with paroxysmal atrial fibrillation.

SOURCES: Wazni OM, Dandamudi G, Sood N, et al. Cryoballoon ablation as initial therapy for atrial fibrillation. N Engl J Med 2020; Nov 16. doi: 10.1056/NEJMoa2029554. [Online ahead of print].

Andrade JG, Wells GA, Deyell MW, et al. Cryoablation or drug therapy for initial treatment of atrial fibrillation. N Engl J Med 2020; Nov 16. doi: 10.1056/NEJMoa2029980. [Online ahead of print].

Mounting evidence suggests there are better outcomes with a rhythm control strategy than a rate control strategy for many patients with atrial fibrillation (AF). An important question remains: Once selected, should clinicians attempt the rhythm control strategy first with antiarrhythmic drug (AAD) therapy, or with catheter ablation? Guidelines recommended ablation after a failed trial of at least one AAD, while gradually accepting there are circumstances in which starting with ablation might be preferred.

In STOP-AF (Wazni et al) and EARLY-AF (Andrade et al), investigators sought to compare cryoballoon ablation to AAD therapy as first-line treatment for paroxysmal AF. Wazni et al randomized 203 patients 1:1 to cryoablation or AAD (class I or III agent). Mean age was 61 years, mean ejection fraction (EF) was 61%, and mean left atrial (LA) diameter was 39 mm. Most patients recorded a CHA2DS2-VASc score of 1 or 2, and 27% had undergone electrical or pharmacologic cardioversion in the prior 12 months. In the ablation group, 101 of 104 patients underwent a “successful” procedure. In the AAD group, about half were treated with flecainide (100-200 mg daily), and 12 of 99 crossed over to ablation before a documented arrhythmia recurrence. The primary endpoint was freedom from atrial arrhythmia recurrence after a 90-day blanking period. Patients were monitored via ECG every three months, via patient-activated phone monitoring for symptoms weekly, and via 24-hour ambulatory monitoring at six and 12 months.

Andrade et al randomized 303 patients 1:1 to cryoablation or AAD. Patient profiles were similar to those in STOP-AF: mean age was 58 years, mean EF was 60%, and mean LA diameter was 39 mm. The mean CHA2DS2-VASc score was 1.9, and 39% had undergone prior cardioversion. In the ablation group, complete pulmonary vein isolation was confirmed in 152 of 154 patients (two patients did not undergo the procedure). In the AAD group, a prespecified protocol was used to up-titrate medications to the maximum dose associated with an “acceptable” side effect profile. Most were treated with flecainide (median dose of 200 mg daily). No patients crossed over to ablation before the occurrence of a primary endpoint event. The primary endpoint was freedom from atrial arrhythmia recurrence after a 90-day blanking period. Notably, all patients received a LINQ implantable loop recorder (ILR) for continuous cardiac monitoring, inserted no more than 24 hours after catheter ablation or initiation of AAD.

Although patient populations and treatment strategies were similar, EARLY-AF was designed with more sensitive arrhythmia monitoring via ILR. Andrade et al also used a more regimented protocol for drug titration. Flecainide was most commonly used in both trials, although at somewhat lower doses in STOP-AF.

The results of both trials also were similar. In STOP-AF, freedom from procedural failure or recurrent arrhythmia at 12 months was 75% in the ablation group and 45% in the AAD group (P < 0.001). There were two pericardial effusions, two phrenic nerve injuries, and a TIA in the ablation group. Adverse events in the AAD group included four episodes of syncope in three patients, bradycardia in two patients, and a variety of drug side effects. Quality of life in the ablation group as assessed with two different measures improved significantly from baseline to 12 months. Scores were not reported for patients in the AAD group.

In EARLY-AF, freedom from recurrent arrhythmia at 12 months was 57% in the ablation group and 32% in the AAD group (P < 0.001). Symptomatic recurrence was documented in 11% of the ablation group and 26% of the AAD group. Overall arrhythmia burden was low. Median time in AF was 0% in the ablation group vs. 0.13% in the AAD group. There were two patients who required a pacemaker, one episode of syncope, and three persistent phrenic nerve palsies in the ablation group. Adverse events in the AAD group included one episode of tamponade (in a patient who underwent ablation after arrhythmia recurrence), two episodes of syncope and five episodes of presyncope, two wide complex tachycardia or proarrhythmic events, two patients who required a pacemaker, and one TIA event. Quality of life scores improved significantly in both groups.

The authors of both studies concluded that in patients with paroxysmal AF, cryoablation was superior to AAD therapy for preventing recurrent AF.


To cardiologists and electrophysiologists treating paroxysmal AF, the top line results of these two trials come as little surprise: Ablation is superior to AAD therapy in preventing recurrence of atrial arrhythmias. However, there are several additional important findings, with a few caveats to remember.

First, the benefit of intensive arrhythmia monitoring for accurate reporting in AF trials is demonstrated in EARLY-AF. There were higher overall rates of documented arrhythmia recurrence compared with STOP-AF, but with outcomes that were similarly superior for catheter ablation. In other words, more sensitive monitoring detected more asymptomatic episodes but did not “falsely” improve outcomes in one arm over the other.

The restricted crossover in EARLY-AF also was notable and helps increase confidence in the results. Patients with paroxysmal AF clearly experience asymptomatic arrhythmia episodes despite ablation and/or AAD therapy, affirming the need for continuous anticoagulation when risk factors are present.

On the other hand, these trial authors used cryoballoon ablation. Although similar outcomes might be expected with radiofrequency ablation (preferred by many operators), this was not part of either trial. Procedural complications were infrequent but not absent, particularly phrenic nerve palsy, an injury that often heals with time but can be associated with uncomfortable shortness of breath. The trials likely were underpowered and follow-up too brief to detect differences in “hard” outcomes, such as stroke or death. Critics also point to the lack of a “true control” observation (and presumably risk factor modification). Nevertheless, for many patients who feel unwell with AF, in whom AF is likely to progress, and for whom recent data support earlier rhythm control to improve cardiovascular outcomes, ablation by an experienced operator after thorough discussion of risks is a reasonable first-line strategy.