Very little about the management of atrial fibrillation is an emergency. And yet we in the Emergency Medicine community have embraced it as such. We gleefully bring to bear the full technological powers available to an Emergency Physician in our attempts to tame its irregular irregularity. Using this aggressive approach, we have proven ourselves quite adept at controlling atrial fibrillation, discharging the majority of patients home from the Emergency Department in sinus rhythm1. But to what end? The recent publication if the RACE-7 trial by Pluymaekers et al in the NEJM, highlights just how absurd and wasted our struggles towards sinus normalcy are2.
The authors enrolled adult patients presenting to 15 Emergency Departments throughout the Netherlands with recent onset stable symptomatic atrial fibrillation (defined as starting within 36-hours of presentation). Patients were randomized to either immediate cardioversion in the Emergency Department or a delayed approach. In the delayed strategy patients received rate controlling medication in the ED and were then discharged with the plan to follow up in 48-hours. At 48-hours patients were assessed as to whether they remained in atrial fibrillation, and if so, were sent back to the Emergency Department for immediate cardioversion.
Over a 4-year period the authors randomized 437 patients, 218 to the delayed-cardioversion group and 219 to the early-cardioversion group. Overall the cohort was fairly representative of patients that typically present to the ED for symptomatic atrial fibrillation. The mean age was 65 years old, 40% of the patients were female, 44% presented with their first episode of atrial fibrillation, 64 had a CHA2 DS2 -VASc score of 2 or higher, and 40% were taking oral anticoagulant drugs at time of presentations.
Of the patients randomized to the immediate cardioversion strategy, reversion to sinus rhythm occurred in 94% of the patients, 16% experiencing spontaneous cardioversion and 78% undergoing either chemical or electrical cardioversion (37.9% and 40.2% respectively). In the delayed group, 98% of the patients achieved rate control in the Emergency Department, using a beta-blocker in 71.1%, a non-dihydropyridine calcium-channel blocker in 2.3% of the patients and digoxin in 6% of the patients. In 19% rate control was achieved without the assistance of any medication.
Despite their almost universal success in the early-cardioversion group at achieving sinus rhythm prior to discharge, the authors observed no difference in their primary endpoint, the presence of sinus rhythm on the ECG recorded at the 4-week visit. 91% in the delayed-cardioversion group and 94% in the early-cardioversion group. This 2.9% difference met the authors non-inferiority margin of 10% (95% CI −8.2 to 2.2; P=0.005 for noninferiority). Nor did the authors find any difference in any of their secondary endpoints. Almost all the patients in both groups were discharged following the index ED visit. ED revisits for atrial fibrillations were 7% in both groups. No differences were observed in the rate of cardiovascular complications, 10 occurring in the delayed-cardioversion group (including 1 patient with ischemic stroke and 3 with acute coronary syndrome or unstable angina), and 8 in the early-cardioversion group (including 1 patient with transient ischemic attack and 3 with acute coronary syndrome or unstable angina). Total median duration of the index visit was 120 minutes in the delayed-cardioversion group and 158 minutes in the early-cardioversion group.
This trial is far from ideal. It compared an immediate approach to a slightly less emergent one, leaving us all wondering what would they have found if they had not attempted to control the rhythm at all. In addition, the primary endpoint, number of patients in sinus rhythm in one month holds no clinical meaning, as we have known for sometime that a rhythm control strategy does not lead to better outcomes. Despite its flaws, this trial does give us some insight into the true futility of our emergency endeavors for early rhythm control. At the 48-hour follow up visit, 69% of the patients had spontaneously converted without active cardioversion. This number would likely have been even higher if the authors had foregone delayed cardioversion in favor of a simple rate control strategy. In addition, the number of patients who experienced repeat bouts of atrial fibrillation over the first 28-days was similar between the two groups, 30% and 29% in the delayed and immediate strategies, respectively. Nor was there any difference in patients’ quality of life scores, often cited as one of the major advantages of an immediate defibrillation strategy. Simply put, the majority of patients evaluated in the ED for stable symptomatic atrial fibrillation will spontaneous revert to sinus rhythm by 48-hours. Immediate cardioversion had no effect on how often patients experience bouts of atrial fibrillation in the following 28-days, the number of patients in sinus rhythm at 1-month, or any patient important outcomes.
Atrial fibrillation is a chronic disease with outcomes measured in years. Constraining its irregularities for the fleeting moments the patient is before us in the Emergency Department has minimal effect on patient outcomes or wellbeing. An immediate cardioversion strategy is a quixotic attempt to check a symptom that is not only incredibly difficult to control, but whose regulation has never been found to improve patient important outcomes. When viewed from this perspective, the futility of ED cardioversion seems obvious. It serves only as a distractor, diverting our attention from rate control and the appropriate use of anticoagulants, interventions that have proven benefits on patients’ downstream health and wellbeing.