New England Journal of Medicine DECEMBER 05, 2002

A Comparison of Rate-Control and Rhythm-Control Strategies in Atrial Fibrillation

The AFFIRM Investigators (D. George Wyse, Brian B. Waldo, et al.)

Bottom Line

The AFFIRM trial demonstrated that a rhythm-control strategy does not provide a survival benefit compared to a rate-control strategy in patients with atrial fibrillation and risk factors for stroke or death.

Key Findings

1. The primary endpoint of all-cause mortality showed no statistically significant difference between the rate-control and rhythm-control strategies (25.9% vs. 26.7%; hazard ratio, 1.15; 95% confidence interval, 0.99 to 1.34; p=0.08).
2. There was no statistically significant difference between groups regarding the secondary composite endpoint of death, disabling stroke, anoxic encephalopathy, major bleeding, or cardiac arrest (32.7% vs. 32.0%; p=0.33).
3. The rate-control strategy was associated with significantly higher rates of successful symptom management and lower utilization of hospital resources, while the rhythm-control strategy resulted in a significantly higher incidence of adverse drug effects and hospitalizations.

Study Design

Design
RCT
Open-Label
Sample
4,060
Patients
Duration
3.5 yr
Median
Setting
Multicenter, US and Canada
Population Adults age ≥65 years (or <65 with risk factors for stroke/death) with atrial fibrillation likely to be recurrent and eligible for long-term anticoagulation.
Intervention Rhythm-control strategy using antiarrhythmic drugs and/or cardioversion to maintain sinus rhythm.
Comparator Rate-control strategy using beta-blockers, calcium channel blockers, and/or digoxin to manage heart rate.
Outcome All-cause mortality.

Study Limitations

The trial population was predominantly elderly (mean age 69.7 years), limiting generalizability to younger, healthier cohorts.
The rhythm-control strategy was heavily reliant on older antiarrhythmic agents, which were associated with significant non-cardiovascular adverse effects.
A high rate of cross-over between the two treatment strategies occurred during the study, complicating the primary intention-to-treat analysis.
Modern therapeutic options, specifically atrial fibrillation ablation, were not evaluated as part of the rhythm-control protocol.

Clinical Significance

The results of the AFFIRM trial shifted the clinical paradigm for managing atrial fibrillation, establishing rate control as a safe and acceptable primary strategy for many elderly patients, thereby reducing the exposure to the toxicities of long-term antiarrhythmic drug therapy.

Historical Context

Prior to the publication of the AFFIRM trial, the management of atrial fibrillation was heavily biased toward a rhythm-control strategy in an effort to restore sinus rhythm. The AFFIRM trial was one of the landmark studies that fundamentally challenged this assumption, proving that maintaining rhythm at the cost of antiarrhythmic medication exposure did not confer mortality benefits.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Based on the physiological principles of the cardiac cycle, why was it initially hypothesized that maintaining sinus rhythm (rhythm control) would lead to better clinical outcomes than simply controlling the heart rate in patients with atrial fibrillation?

Key Response

Sinus rhythm preserves the 'atrial kick,' which contributes approximately 20-30% of ventricular filling (preload). Theoretically, this improves cardiac output, prevents structural remodeling of the atria, and reduces the stasis of blood in the left atrial appendage, thereby potentially lowering the risk of heart failure and thromboembolic stroke.

Resident
Resident

A 72-year-old patient with atrial fibrillation is successfully cardioverted to sinus rhythm and started on amiodarone. According to the findings and subsequent analyses of the AFFIRM trial, why is it still necessary to continue long-term anticoagulation even if the patient appears to be in stable sinus rhythm?

Key Response

The AFFIRM trial found that the majority of strokes in the rhythm-control group occurred after warfarin had been stopped or when the INR was subtherapeutic. Because atrial fibrillation can recur asymptomatically and 'sinus rhythm' on a periodic EKG does not guarantee the absence of paroxysmal episodes, anticoagulation must be continued based on the patient's underlying stroke risk factors (e.g., CHA2DS2-VASc score) rather than their perceived rhythm.

Fellow
Fellow

The 'AFFIRM Paradox' suggests that while sinus rhythm is statistically associated with improved survival, the rhythm-control strategy failed to show a benefit. How do the adverse effects of antiarrhythmic drugs (AADs) explain this discrepancy in the trial results?

Key Response

Post-hoc analyses of AFFIRM demonstrated that the survival advantage of being in sinus rhythm was offset by the increased mortality associated with AADs. The toxicity of these drugs (including proarrhythmia, pulmonary toxicity, and organ damage) negated the hemodynamic benefits of sinus rhythm, leading to the conclusion that currently available AADs are a major limiting factor in a rhythm-control strategy.

Attending
Attending

How did the AFFIRM trial fundamentally shift the 'burden of proof' when choosing between rate and rhythm control for an asymptomatic elderly patient with atrial fibrillation and multiple comorbidities?

Key Response

AFFIRM established rate control as an acceptable first-line strategy, shifting the paradigm from 'rhythm control for everyone' to 'rate control as the default.' It taught clinicians that unless a patient is symptomatic despite adequate rate control, the toxicity and hospitalization risks associated with rhythm control (cardioversions, drug loading, AAD monitoring) do not justify its use for the sole purpose of mortality reduction.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

AFFIRM utilized an intention-to-treat (ITT) analysis. Given that 37.5% of patients in the rhythm-control arm and 14.9% in the rate-control arm crossed over or discontinued their assigned therapy by five years, how might a per-protocol analysis have changed the interpretation of the results regarding drug efficacy versus strategy efficacy?

Key Response

An ITT analysis preserves the benefits of randomization and reflects real-world clinical practice where drugs are stopped due to side effects or failure. However, the high crossover rate 'dilutes' the treatment effect, potentially masking the true efficacy of rhythm control. A per-protocol analysis might show higher efficacy for rhythm control in those who can tolerate it, but it would be subject to significant selection bias (healthy-user effect).

Journal Editor
Journal Editor

Considering the AFFIRM study population had a mean age of 69.7 years and significant risk factors for stroke, what are the primary threats to the external validity of these findings when applying them to younger, lower-risk 'lone AF' patients or those with highly symptomatic paroxysmal AF?

Key Response

The trial specifically recruited patients at high risk for stroke or death, meaning the results may not generalize to younger patients where long-term remodeling and lifetime drug toxicity are different. Furthermore, since the primary endpoint was mortality, the trial was not powered to detect quality-of-life improvements which are the primary driver for rhythm control in highly symptomatic paroxysmal AF populations.

Guideline Committee
Guideline Committee

How does the evidence from the AFFIRM trial reconcile with current AHA/ACC/HRS guidelines which provide a Class I recommendation for rhythm control in certain populations, and how does this impact the 'Level of Evidence' for anticoagulation in rhythm-controlled patients?

Key Response

Guidelines (e.g., 2023 ACC/AHA/ACCP/HRS) recommend rhythm control (Class I) primarily for symptom improvement or in patients with heart failure (HFrEF) where it may improve EF. However, based on AFFIRM, the guidelines maintain that the decision to anticoagulate should be based on thromboembolic risk scores (Class I, LOE B-R), not on the success of a rhythm-control strategy, because the trial showed rhythm control does not eliminate stroke risk.

Clinical Landscape

Noteworthy Related Trials

2002

RACE Trial

n = 522 · NEJM

Tested

Rhythm-control strategy (electrical cardioversion + antiarrhythmic drugs)

Population

Patients with persistent atrial fibrillation

Comparator

Rate-control strategy (digitalis, beta-blockers, or calcium-channel blockers)

Endpoint

Composite of cardiovascular death, heart failure, thromboembolism, bleeding, pacemaker implantation, or severe adverse drug effects

Key result: The study found that a rate-control strategy was noninferior to a rhythm-control strategy for the prevention of cardiovascular events in patients with persistent atrial fibrillation.
2009

ATHENA Trial

n = 4,628 · NEJM

Tested

Dronedarone

Population

Patients with atrial fibrillation or flutter and additional cardiovascular risk factors

Comparator

Placebo

Endpoint

First hospitalization due to cardiovascular events or death

Key result: Dronedarone significantly reduced the primary endpoint of hospitalization for cardiovascular events or death in patients with atrial fibrillation.
2020

EAST-AFNET 4 Trial

n = 2,789 · NEJM

Tested

Early rhythm-control therapy

Population

Patients with early atrial fibrillation and cardiovascular conditions

Comparator

Usual care (primarily rate control)

Endpoint

Composite of cardiovascular death, stroke, or hospitalization for worsening heart failure or acute coronary syndrome

Key result: Early rhythm-control therapy was associated with a lower risk of cardiovascular death, stroke, or hospitalization for heart failure compared with usual care.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis