Determinants of Survival in Atrial Fibrillation

Abstract & Commentary

Synopsis: Results of this study suggest that if an effective method for maintaining sinus rhythm with fewer adverse effects were available, it might improve survival.

Source: Corley SD, et al and the AFFIRM Investigators. Circulation.2004;109:1509-1513.

The AFFIRM study was a randomized comparison of 2 strategies for management of patients with atrial fibrillation. Patients with atrial fibrillation requiring therapy, who also had one or more risk factors for stroke or death, were randomized to either a rate control strategy of cardioversion and treatment with antiarrhythmic drugs or a rate control strategy using atrioventricular (AV) nodal blocking agents to control ventricular response. In this paper, Corley and colleagues report an "on-treatment" analysis of the relationship of survival to cardiac rhythm and treatment over time. In AFFIRM, 4060 patients were randomized over a 4-year period that ended on October 31, 1999. The mean duration of follow-up was 2.5 years with a maximum of 6 years. Patients in AFFIRM could be in sinus rhythm at the time of randomization. The initial results using an "intention to treat analysis" showed that there was no significant difference between rate control and rhythm control initial strategies. In this paper, an "on-treatment analysis" was used. For this purpose, a Cox proportional hazards regression was performed with evaluation of several time-dependent covariates. This allowed for drug therapies and heart rhythm at the time of follow-up visits to be evaluated and for multiple covariates to be included in a statistical model to assess each variable’s relationship to the primary endpoint after adjustment or control for the other covariates. The analysis involved 6 time-dependent covariates: sinus rhythm, and the use of the following drugs, warfarin, digoxin, beta-blockers, calcium channel blockers, and rhythm control drugs. In addition to the 6 time-dependent covariates, 12 baseline variables were included in the analysis. These included: age at time of enrollment in the study, sex, gender, history of coronary artery disease, congestive heart failure, hypertension, diabetes, stroke or transient ischemic attack, smoking, first vs recurrent episode of atrial fibrillation, and the presence of either left atrial enlargement, left ventricular dysfunction, or mitral valve regurgitation on a 2-dimensional echocardiography. Since echocardiographic data were only available in a subset of 2796 patients, the analysis was carried out both with and without exclusion of the echocardiographic data. With echocardiographic data included, 12 covariates were identified that influenced survival. The following baseline variables were significantly associated with an increased risk of death: increasing age (HR = 1.06), history of coronary artery disease, (HR = 1.56), history of CHF (HR = 1.57), history of diabetes (HR = 1.56), history of stroke or TIA (HR = 1.70), recent history of smoking (H = 1.78), left ventricular dysfunction (HR = 1.36), and mitral regurgitation (HR = 1.36). Among the time dependent variables, the presence of sinus rhythm (HR = 0.53), warfarin use (HR = 0.50), digoxin use (HR = 1.42), and rhythm control drug use (HR = 1.49) were significantly related to survival after adjustment for other covariates. Results were similar if the echocardiographic data were excluded.

Corley et al believe that the their time-dependent analysis clearly shows that warfarin use improves survival, and that the presence of sinus rhythm, but not antiarrhythmic drug use, is associated with a lower risk of death. They then conclude that these results "suggest that if an effective method for maintaining sinus rhythm with fewer adverse effects were available, it might improve survival."

Comment by John DiMarco, MD, PhD

The AFFIRM Study and several other trials on rate control vs rhythm control strategies have not shown any overall benefit with attempts to maintain sinus rhythm with antiarrhythmic drugs. This led the American College of Physicians to state in their new clinical guidelines that a rate control strategy was appropriate for most patients with atrial fibrillation (McNamara et al, Ann Intern Med. 2003;139:1018-1033). This report from Corley et al seems to counteract this position. However, the analyses in this study are quite complex. Although many baseline variables were included in the Cox regression model, any changes from baseline in a patient’s condition, other than the time dependent variables that were analyzed, use of selected drugs and presence of sinus rhythm were not controlled. Corley et al seem to imply that maintaining sinus rhythm was an important determinant of survival. However, an alternative hypothesis, which they mention in their discussion but which might be overlooked, is that progressive disease is what determines whether antiarrhythmic drugs are likely to maintain sinus rhythm. Patients who deteriorate for whatever reason are likely to have recurrence of atrial fibrillation even if they continue antiarrhythmic therapy. Clinicians know that patients who switch back and forth are often less stable, and they also realize that the occurrence of significant disease, either cardiac or extra cardiac, often precipitates recurrences of atrial fibrillation. Thus, recurrent atrial fibrillation may just be a marker for patients whose overall status is declining. Atrial fibrillation may further complicate this decline but it may not be the primary driver.

For these reasons, I think the question of whether or not maintaining sinus rhythm will improve survival remains open. No one should argue that atrial fibrillation is a better rhythm than sinus rhythm. However, in the absence of complicating factors, it may not make much difference what rhythm a person is in as long as symptoms can be controlled and anticoagulation be maintained.

John DiMarco, MD, PhD, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville, is a member of the Editorial Board of Clinical Cardiology Alert.