CardioExchange Archive

The CardioExchange Editors interview Manesh R. Patel about his research group’s retrospective analysis of data on digoxin use among patients with atrial fibrillation in the ROCKET AF trial. The study is published in The Lancet.

CardioExchange Editors: Please recount your study design and main findings.

Patel: In a retrospective analysis of the ROCKET AF trial, we looked at the use of digoxin at baseline and outcomes. We stratified patients by the presence or absence of heart failure presence and did several types of analysis to account for possible bias, including multivariate adjustment and propensity analysis.
 After the adjustments, digoxin was associated with significantly increased risks for all-cause mortality, vascular death, and sudden death.

Editors: How does this add to the other recent articles on this topic?

Patel: Some recent reports have noted a possible hazard with digoxin. We found that a significant number of patients with atrial fibrillation (with or without heart failure) were treated with digoxin in the ROCKET AF trial. In fact, we found that some patients with atrial fibrillation were treated with digoxin before other rate-control agents. What we add to the available literature is that ROCKET AF had data on blindly adjudicated events, including CV death, sudden death, MI, stroke, and bleeding. Digoxin increased the risk for CV death and sudden death without the other events. Note that this was a retrospective, non-prespecified analysis with many possibly confounding factors.

Editors: In your article’s discussion section, you stop short of calling for a moratorium on the use of digoxin for patients with atrial fibrillation. Given your findings, the findings of others, and the absence of trial support, why do you think that digoxin use should not be suspended for these patients?

Patel: We believe, as we wrote in the discussion, that the question for clinicians should not be whether digoxin is harmful. That would be hard to answer absolutely without a large randomized trial of the drug in patients with atrial fibrillation. However, we say that even if you think our analysis contains bias from unmeasured factors for which the adjustments cannot account, it is still unlikely that digoxin yields a clinical improvement for patients. So for patients with atrial fibrillation (with or without heart failure) who require rate control, many other alternatives (such as beta-blockers or calcium-channel blockers) should be considered before digoxin. Furthermore, we know that digoxin is a poor rate-control agent.

Editors: What do you tell your patients who are on digoxin for atrial fibrillation?


Patel: I tell patients with atrial fibrillation without heart failure that is it not clear that digoxin is helpful, and we often stop it and use other rate-control therapies. For patients with heart failure and low blood pressure, where other agents cannot be used, we will consider digoxin. But aside from these caveats, I am not using digoxin or recommending it specifically for atrial fibrillation. My colleagues in heart failure may still use digoxin to help with long-term management, but that is a separate discussion.

JOIN THE DISCUSSION

How do the findings from Dr. Patel’s study affect your view of digoxin use in patients with atrial fibrillation?