Rate Control versus Electrical Cardioversion for Persistent Atrial Fibrillation - RACE


Assessment of the effect of rate control versus rhythm control on morbidity and mortality in patients with recurrent persistent atrial fibrillation.


Rate control of persistent AF is not inferior to rhythm control in terms of morbidity and mortality.

Study Design

  • Randomized

Patients Enrolled: 522
NYHA Class: Class I=50%; class II=47%; class III=3%
Mean Follow Up: At least 2 years
Female: 37%

Patient Populations:

Recurrent persistent atrial fibrillation or flutter; no contraindications to oral anticoagulation.


Arrhythmia lasted >1 year; New York Heart Association class IV heart failure; current or previous treatment with amiodarone or a pacemaker.

Primary Endpoints:

A composite of cardiovascular death, hospital admissions for heart failure, thromboembolic complications, severe bleeding, pacemaker implantation, and severe adverse effects of therapy.

Secondary Endpoints:

Individual components of the primary endpoint.

Drug/Procedures Used:

The 522 patients included in RACE had persistent AF/atrial flutter; all had undergone one or two electrical cardioversions in the previous 2 years. They were randomized to a strategy of rate control (n=256) using beta blockers, digoxin, or calcium antagonists titrated to a heart rate <100 bpm, or rhythm control (n=266) with electrical cardioversion and sotalol prophylaxis.

Anticoagulation in the rate control arm was titrated to achieve an International Normalized Ratio of 2.0 to 3.5. In the rhythm control arm, anticoagulation was given for 1 month before cardioversion was attempted, and discontinued if chronic sinus rhythm was obtained.

Principal Findings:

The primary endpoint, a composite of cardiovascular death, hospital admissions for heart failure, thromboembolic complications, severe bleeding, pacemaker implantation, and severe adverse effects of therapy, occurred in 17.2% of the rate control arm compared with 22.6% of the rhythm control arm. This -5.4% absolute difference was within the 90 percent confidence interval of -11.0%-0.4% and thus rate control met the criterion for noninferiority. Among patients with hypertension, the primary endpoint was achieved by 17.3% of the patients randomized to rate control and 30.8% of the patients randomized to rhythm control (p<0.05).

None of the individual components of the composite endpoint were significantly different with the exception of severe adverse effects of therapy, which was more frequent in the rhythm control arm (4.5%) than the rate control arm (0.8%).


RACE showed that rate control is not inferior to rhythm control and that rate control is a viable alternative for patients with a high risk of AF recurrence. The subgroup of patients with hypertension tended to be at higher risk with a strategy of rhythm control with sotalol. It is unknown if the same finding would be observed among patients treated with amiodarone. This subgroup of patients requires further exploration of the most optimal therapy and validation of these findings. These findings are consistent with the larger AFFIRM trial. It should be acknowledged that in both trials, these results by definition do not apply to those patients who cannot tolerate rate control, where this strategy is not an option.


N Engl J Med 2002;347:1834-40.

Clinical Topics: Arrhythmias and Clinical EP, Heart Failure and Cardiomyopathies, Prevention, EP Basic Science, Atrial Fibrillation/Supraventricular Arrhythmias, Acute Heart Failure, Hypertension

Keywords: Digoxin, Electric Countershock, Heart Rate, Calcium Channel Blockers, International Normalized Ratio, Heart Failure, Sotalol, Confidence Intervals, Hypertension, Atrial Flutter

< Back to Listings