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, Boston Scientific, and Medtronic.

SYNOPSIS: For patients presenting to an ED with recent-onset atrial fibrillation, using rate control and outpatient cardioversion only as needed was associated with a high rate of spontaneous conversion within 48 hours of arrhythmia onset and noninferior short-term outcomes compared to immediate cardioversion in the ED.

SOURCE: Pluymaekers NAHA, Dudink EAMP, Luermans JGLM, et al. Early or delayed cardioversion in recent-onset atrial fibrillation. N Engl J Med 2019;380:1499-1508.

Symptomatic atrial fibrillation (AF) is a common reason for ED visits and referrals, both for first-time and recurrent episodes. Often, the treatment of choice is immediate pharmacologic or electrical cardioversion. Investigators sought to determine whether such early intervention is superior to a more conservative approach of outpatient observation and delayed cardioversion only as needed.

Pluymaekers et al randomized 437 adult patients from 15 hospitals in the Netherlands at the time of their ED visit to early or delayed cardioversion. Patients had to have hemodynamically stable, symptomatic AF with onset < 36 hours before, with no signs of myocardial ischemia or history of AF lasting > 48 hours. For early cardioversion, a pharmacologic approach was used initially in most patients, with electrical cardioversion as a backup. At enrollment, 40% of patients already were taking oral anticoagulant drugs; otherwise, anticoagulation was initiated before or immediately after cardioversion in patients with high stroke risk. Transesophageal echocardiography was not performed in any patients. For delayed cardioversion, patients were treated in the ED with beta-blockers, calcium channel blockers, or digoxin to achieve symptom relief and a heart rate of ≤ 110 beats per minute. Patients were discharged when stable and returned for an outpatient clinic visit as close as possible to 48 hours after onset of symptoms. If AF still was present on ECG, those patients were referred back to the ED for cardioversion.

The mean age of patients was 65 ± 11 years; 60% of patients were male. Hypertension was common, and most registered a CHA2DS2-VASc score of ≥ 2. The predominant symptom of AF was palpitations, although many patients also reported exercise-induced fatigue, dyspnea, and/or chest pain. About 23% of patients already were on an antiarrhythmic drug.

The median duration of the index ED visit was 158 minutes in the early cardioversion group and 120 minutes in the delayed cardioversion group. In the early cardioversion group, conversion to sinus rhythm occurred spontaneously in 16% of patients before the actual cardioversion, and 94% left the ED in sinus rhythm. In the delayed cardioversion group, conversion to sinus rhythm within 48 hours occurred spontaneously in 69% of patients; an additional 28% of patients underwent actual cardioversion.

During four weeks of follow-up, recurrence of AF documented via intermittent monitoring in a subset of patients occurred in 29% of the early cardioversion group and 30% of the delayed cardioversion group, with similar time to recurrence (median, 8 days vs. 12 days). At a four-week visit, 94% of patients in the early cardioversion group and 91% of patients in the delayed cardioversion group were in sinus rhythm, a noninferior outcome for delayed cardioversion per prespecified criteria. One patient in the early cardioversion group suffered a transient ischemic attack (TIA) within four weeks of randomization, and one patient in the delayed cardioversion group experienced an ischemic stroke. Other cardiovascular complications were uncommon and not significantly different between groups. Quality of life (QOL) scores via survey at the four-week visit also were not significantly different between groups. The authors concluded that in patients with recent onset AF, delayed cardioversion was not inferior to early cardioversion in restoring sinus rhythm at four weeks.


Several conclusions can be drawn from this study. First, episodes of AF starting < 36 hours before evaluation often terminate spontaneously. In this cohort, 16% of patients randomized to the early cardioversion group converted to sinus rhythm before cardioversion, and 69% of patients randomized to the delayed cardioversion group converted to sinus rhythm by 48 hours.

Second, focusing on anticoagulation as well as symptom and rate control at the time of the index ED visit (while deferring cardioversion to the outpatient setting) shortened the time spent in the ED. This obviated the need for cardioversion in more than half of patients, with no rhythm or quality of life disadvantages four weeks later.

Third, regardless of cardioversion approach, AF recurrence in this population is common — at least 30% over four weeks, undoubtedly an underestimate given the use of intermittent monitoring.

Finally, the risk of stroke or TIA, known to be higher after both active and spontaneous cardioversion from AF, is not completely eliminated even when therapeutic anticoagulation is used appropriately. One patient experienced a TIA 10 days after early electrical cardioversion despite rivaroxaban therapy, and one suffered a stroke five days after spontaneous conversion despite dabigatran therapy.

The authors noted another pragmatic difference between the two approaches: Patients in the delayed cardioversion group who went on to convert spontaneously would be classified with “paroxysmal” AF rather than “persistent” AF. This semantic distinction has no particular bearing on an individual patient’s symptoms or outcome at four weeks, but it could have important consequences for how future treatment options are considered. For example, catheter ablation is a class IIa recommendation for symptomatic persistent AF refractory to at least one antiarrhythmic drug, but class I for paroxysmal AF.

It is difficult to know what psychological effect an early vs. delayed approach might have on individual patients. On one hand, discharge from the ED in AF could reassure some patients that the arrhythmia is not acutely dangerous and does not necessarily require urgent medical attention. On the other hand, symptoms that originally were mild could be exacerbated after discharge by the new diagnostic knowledge for some patients. Perhaps QOL questionnaires administered after the index ED visit may have revealed a difference that faded by four weeks. Ultimately, the study revealed no clear disadvantage of a wait-and-see approach to recent onset of AF. Whether such an approach has advantages in overall resource use or long-term outcomes remains to be seen.