Atrial Fibrillation and Congestive Heart Failure - AF-CHF


Current evidence from clinical trials does not support the routine use of rhythm control strategies in patients with atrial fibrillation (AF). These trials, however, included only a small proportion of patients with congestive heart failure (CHF). Data from observational studies indicate that AF is associated with worse outcomes in patients with CHF.

Accordingly, the aim of the AF-CHF trial was to evaluate whether rhythm control in patients with systolic HF would be associated with decreased cardiovascular mortality, as compared with rate control.


Rhythm control will be associated with lower cardiovascular mortality compared with rate control in patients with AF and systolic HF.

Study Design

  • Randomized
  • Parallel

Patients Enrolled: 1,376
NYHA Class: III or IV 31.5% at baseline, 76% during the past 6 months
Mean Follow Up: 37 months
Mean Patient Age: 66.5 years
Female: 18
Mean Ejection Fraction: 27%

Patient Populations:

  • LVEF ≤35%, determined within the preceding 6 months
  • History of CHF with NYHA class III-IV symptoms within the preceding 6 months
  • An asymptomatic condition for which the patient had been hospitalized for HF during the previous 6 months, or LVEF ≤25%
  • Electrocardiographic documentation of AF, lasting for at least 6 hours, or requiring cardioversion within the preceding 6 months, or an episode lasting for at least 10 minutes within the previous 6 months and previous electrical cardioversion for AF
  • Eligibility for long-term therapy


  • Persistent AF >12 months
  • A reversible cause of AF or HF
  • Decompensated CHF within 48 hours of intended randomization
  • Use of antiarrhythmic drugs for other arrhythmias
  • Second- or third-degree atrioventricular (AV) block (bradycardia of <50 bpm
  • History of long QT syndrome
  • Previous AV node ablation
  • Anticipated cardiac transplantation within 6 months
  • Renal failure requiring dialysis
  • Lack of birth control in women of child-bearing potential
  • Life expectancy <1 year
  • Age <18 years

Primary Endpoints:

  • Death from cardiovascular causes

Secondary Endpoints:

  • All-cause mortality
  • Stroke
  • Worsening CHF
  • Hospitalization
  • Quality of life
  • Cost of therapy
  • Composite of death from cardiovascular causes, stroke, and worsening CHF

Drug/Procedures Used:

Rhythm control: Aggressive therapy to prevent AF was recommended. Electrical cardioversion was recommended within 6 weeks after randomization in patients who did not have conversion to sinus rhythm after antiarrhythmic drug therapy. If necessary, a second cardioversion was recommended within 3 months after enrollment. Additional cardioversions were recommended for subsequent recurrences of AF. Amiodarone was the drug of choice for the maintenance of sinus rhythm, and either sotalol or dofetilide was used if required. Patients who did not respond to medications could be referred for nonpharmacologic therapy.

Rate control: Rate control: Adjusted doses of metoprolol and digitalis were used to achieve a target resting heart rate of <80 bpm, and <110 bpm on the 6-minute walk test.

Concomitant Medications:

Digoxin (65%), beta-blockers (79%), angiotensin-converting enzyme inhibitors (86%), aldosterone antagonists (44.5%), oral anticoagulation (88%), and aspirin (38.5%)

Principal Findings:

A total of 1,376 patients were randomized, 682 to rhythm control, and 694 to rate control. Baseline characteristics were fairly similar between the two groups. The mean left ventricular ejection fraction (LVEF) was 27%, and about 32% of the patients had New York Heart Association (NYHA) class III or IV symptoms at baseline. About one-half of the patients had ischemic cardiomyopathy. AF was persistent in about 69% of the patients. Prior antiarrhythmic drugs had been used in about 43.5% of the patients. By 12 months, amiodarone was used in 82% of the patients in the rhythm control group versus 7% in the rate control group, whereas beta-blockers and oral anticoagulants were used more often in the rate control group compared with the rhythm control group.

During the study, 21% of the patients in the rhythm control group crossed over to the rate control group, mostly due to failure to maintain sinus rhythm, whereas about 10% of the patients in the rate control group crossed over to the rhythm control group, mostly due to worsening CHF. The prevalence of AF was lowest at about 4 months in the rhythm control group (17%), and was about 27% at 4 years of follow-up.

There was no difference in cardiovascular mortality between the rhythm control and rate control groups (27% vs. 25%, hazard ratio [HR] 1.06, 95% confidence interval [CI] 0.86-1.30, p = 0.59). Other outcomes, including overall mortality (32% vs. 33%, p = 0.68), CHF (11% vs. 8%, p = 0.11), and stroke (1% vs. 2%, p = 0.68) were similar between the two arms.

Hospitalization was more frequently required in the rhythm control arm (64% vs. 59%, p = 0.06), especially in the first year (46% vs. 39%, p = 0.001). The rhythm control group also had more hospitalizations for AF (14% vs. 9%, p = 0.001), and for bradyarrhythmias (6% vs. 3%, p = 0.02). Electrical cardioversion was also more frequent in the rhythm control arm (59% vs. 9%, p < 0.001).


Earlier trials had failed to demonstrate a significant benefit with rhythm control over rate control in unselected patients with AF. The results of the AF-CHF trial confirm these results in patients with AF and systolic HF as well. The high cross-over rates, and the fact that despite being on rhythm control, about 58% of the patients had at least one recurrence of AF, may have helped bias the results toward the null.

It is unknown if other methods of restoring sinus rhythm such as pulmonary vein isolation or newer drugs, such as dronedarone, would be more effective in maintaining sinus rhythm, and thus permit a more valid assessment of true rhythm versus rate control strategies for AF.


Roy D, Talajic M, Nattel S, et al., on behalf of the Atrial Fibrillation and Congestive Heart Failure Investigators. Rhythm control versus rate control for atrial fibrillation and heart failure. N Engl J Med 2008;358:2667-77.

Presented by Dr. Denis Roy at the American Heart Association Annual Scientific Session, Orlando, FL, November 2007.

Keywords: Stroke, Asymptomatic Diseases, Pulmonary Veins, Electric Countershock, Phenethylamines, Heart Rate, Prevalence, Recurrence, Cardiomyopathies, Heart Failure, Digitalis, Stroke Volume, Atrial Fibrillation, Bradycardia, Sotalol, Metoprolol, Hospitalization, Sulfonamides

< Back to Listings