AF-CHF: Even in heart failure, rate-control strategy best for atrial fib

June 18, 2008

Boston, MA - Pharmacologic rate control in patients with both atrial fibrillation and systolic heart failure should be preferred over the drug- and defibrillation-based rhythm-control approach, concluded investigators from the Atrial Fibrillation and Congestive Heart Failure (AF-CHF) trial, now published in the June 19, 2008 New England Journal of Medicine[1].

Its finding of comparable clinical outcomes for the two methods suggests that clinicians can safely use the far less invasive and usually less costly rate-control approach in a particularly high-risk subset of AF patients for whom the best method was still an open question.

The issue had been essentially settled for the broad population with AF based on a number of randomized trials, the most cited of which is probably AFFIRM[2], but they left open the possibility that one method might prove superior in the many heart-failure patients who have atrial fibrillation.

AF-CHF, which followed more than 1300 patients for a mean of three years, had been presented at the American Heart Association 2007 Scientific Sessions and reported by heart wire at the time.

Many of us believed that you had to force the rhythm control, because you actually did believe it had a favorable impact on mortality. But that's not the case.

"Our results cannot be generalized to patients with heart failure and preserved left ventricular function," caution the authors, led by Dr Denis Roy (Montreal Heart Institute and Université de Montréal, QC). Atrial fibrillation is common in such patients, but little is known in that setting about its therapy or prognostic import, they write.

Still, AF-CHF was a complete wash for the primary end point of cardiovascular mortality as well as the secondary end points of all-cause mortality and worsening HF, in the setting of NYHA class 3-4 heart failure, an LVEF <35%, and ECG-documented AF among patients well-treated with standard HF medications.

In patients with both AF and heart failure, "if they're not terribly symptomatic from the atrial fibrillation, it's fine to then go ahead and use a rate-control strategy," Dr Michael E Cain (University at Buffalo, NY), the coauthor of an editorial accompanying the AF-CHF paper[3], told heart wire . "Many of us believed that you had to force the rhythm control, because you actually did believe it had a favorable impact on mortality. But that's not the case." The AF-CHF trial, he said, "I think has taken care of that question."

Cardiovascular mortality was 27% in the rhythm-control group (n=682) and 25% in the rate-control group (n=694), for a hazard ratio of 1.05 (p=0.67) after adjustment for baseline variables reflecting LV function, heart-failure class, comorbidities, use of implantable devices, duration of AF, renal function, and medical therapy.

Studies must also test the superiority of ablation therapy, as compared with rate control, since ample data show that rate control is an acceptable strategy and one that is almost certainly more cost-effective than any other approach.

On the other hand, hospitalizations were more frequent in the rhythm-control group, the difference reaching significance (46% vs 39%, p=0.001) during the first follow-up year. "This finding probably reflects the need for repeated cardioversion and adjustment of antiarrhythmic therapy [with rhythm control], which is consistent with the results of previous studies," write Roy et al.

An accompanying editorial, coauthored by Cain and Dr Anne B Curtis (University of South Florida, Tampa), makes a case for ablation therapy in patients with both heart failure and AF. Noting that trial after trial has shown no outcomes difference between treatments that aim to control ventricular rate and those that strive to maintain sinus rhythm, they write that "sufficient data are available to lay the blame for these setbacks on the drug-based rhythm-control strategy itself, rather than on the conclusion that sinus rhythm is no better for patients than atrial fibrillation."

Interviewed, Cain considered whether nature actually made a mistake, such that sinus rhythm isn't that important. "Or is it that the treatment doesn't work in some patients, or perhaps the treatment is worse than the disease?" AF ablation "is probably the best approach that we have" to answer those questions, because it achieves rhythm control more effectively than drugs and doesn't have the side effects that detract from the overall effectiveness of drugs.

In their editorial, Cain and Curtis—noting that AF ablation trials are ongoing—observe that "studies must also test the superiority of ablation therapy as compared with rate control, since ample data show that rate control is an acceptable strategy and one that is almost certainly more cost-effective than any other approach."

ANDROMEDA in print

Recent clinical trials have raised hopes that the investigational class 3 antiarrhythmic dronedarone (Multaq, Sanofi-Aventis) would prove a safer alternative to amiodarone in the treatment of atrial fibrillation, though not for patients who also have heart failure.

Dronedarone given as antiarrhythmic protection in such patients who were on standard HF medications apparently worsened outcomes in a trial called ANDROMEDA, which had been terminated prematurely five years ago but has finally made its appearance in the literature[4].

Still, the drug made something of a splash at last month's Heart Rhythm Society (HRS) 2008 Scientific Sessions when, in AF patients who received it for maintenance of sinus rhythm in a trial called ATHENA, it reduced the primary end point of CV hospitalizations or death by 24% over two years with a much better safety profile than amiodarone typically shows. As noted in heart wire coverage of the trial from the HRS sessions, the findings were consistent with similar but more tentative results from the earlier and smaller EURIDIS and ADONIS trials.

But at the recent HRS sessions, long after ANDROMEDA's early termination was announced (and reported by heart wire ) in January 2003, it was on the minds of many observers who tempered their enthusiasm over ATHENA with cautions about avoiding dronedarone in patients with heart failure—unless, perhaps, a more definitive trial suggests it's safe. They are also awaiting results from ongoing direct comparisons of dronedarone and amiodarone to get a better fix on how the apparently safer drug might be used.

As reported alongside the AF-CHF trial in the June 19, 2008 New England Journal of Medicine, ANDROMEDA randomized 627 patients (1000 had been the target) hospitalized with systolic heart failure to receive either 400 mg dronedarone twice daily or placebo. Over a median of two months, mortality was 8.1% and 3.8% (p=0.03). Events making up the primary composite end point of all-cause death or HF hospitalization occurred in 17.1% and 12.6%, respectively (p=0.12).

According to the authors, led by Dr Lars K ø ber (University of Copenhagen, Denmark), a post hoc analysis suggested that heart failure accounted for most of the mortality difference, risk was inversely related to LVEF, and dronedarone apparently increased HF hospitalizations. "These findings raise the possibility that dronedarone directly or indirectly causes worsening heart failure, particularly in patients who have poor systolic function," they write.

Referring to ANDROMEDA in their editorial that was primarily about AF-CHF 2], Cain and Curtis— acknowledging that the trial wasn't specifically about AF suppression—noted that "there is increasing evidence that ACE inhibitors, statins, and other nonantiarrhythmic drugs can favorably modulate the natural course of atrial fibrillation and appear to be protective in many settings, even in patients with heart failure."


AF-CHF was partially supported by Crystaal, Merck FrosstCanada, Sanofi-Aventis, and Medtronic Canada. The disclosures for AF-CHF coauthors are detailed in the report. Cain reports receiving consulting fees from Sanofi-Aventis and Cryo-Cor, and Curtis reports receiving consulting fees from Sanofi-Aventis, Biosense Webster, and St Jude Medical. ANDROMEDA was funded by Sanofi-Aventis; coauthor disclosures are included in the report.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.