Digoxin for AF: Mortality Risk Again Seen, Now Linked to Serum Levels

March 08, 2018

A new report has reheated debate on whether to use digoxin for rate control in patients with atrial fibrillation (AF), one of the drug's most common uses worldwide.

Does the drug raise mortality beyond its capacity for good in such patients, or is it safe enough to be routinely used in AF alongside other rate-control drugs? Those remain open questions, with proponents taking both views, but the new study adds a novel twist to the debate.

In a post hoc analysis of nearly 18,000 patients in the ARISTOTLE trial, about a third of whom were receiving digoxin when the trial started, the adjusted risk for cardiovascular death and death from any cause went up proportionally with digoxin serum concentrations.

The trial was primarily a test of the oral anticoagulant apixaban (Eliquis, Pfizer/Bristol-Myers Squibb) in nonvalvular AF.

Moreover, in the new analysis, the risks for sudden death and any-cause death were significantly increased in patients who started receiving digoxin after the trial was underway. The sudden death risk tended to start climbing soon after patients initiated the drug.

The associations between digoxin and mortality were independent of whether patients also had heart failure, according to the study, published March 5 in the Journal of the American College of Cardiology.

In contrast to prior looks at digoxin risk in AF, the report includes separate analyses for patients who were receiving the drug at baseline and for those prescribed it for the first time during the trial, according to lead author, Renato D Lopes, MD, PhD, Duke Clinical Research Institute, Durham, North Carolina.

Also novel, he told theheart.org | Medscape Cardiology, was that serum digoxin levels were available for the patients; the drug is well known to have a narrow safe, effective therapeutic range and to be toxic at high concentrations.

Up to a third of patients with AF worldwide are treated with digoxin, write Lopes and his coauthors, and current AF management guidelines look favorably on its use for rate control.

Although baseline digoxin use per se didn't predict mortality, the story was different when digoxin serum concentration was considered as a continuous variable. "The higher the level, the higher your risk of mortality," Lopes said. 

That, paired with the significant mortality risk seen in those who started on the drug later, suggests that "maybe digoxin is just toxic, no matter what."  A key message from the study is that, in general, digoxin should be avoided in AF "regardless of whether there's heart failure."

Some patients with AF, however, might have no other options. "In those rare cases where physicians are completely certain that the patient needs digoxin," he said, "I think it's very reasonable to measure digoxin more often than we typically do," to make sure serum levels are below 1.2 ng/mL and preferably below 1.0 ng/mL.

Avoid Temptation?

An editorial accompanying the report essentially agrees. "For the majority of patients with AF, there is no reason to believe that there is any benefit, but rather possible or probable harm, of digoxin compared with other atrioventricular-nodal agents," writes Mintu P Turakhia, MD, Stanford University School of Medicine and the Veterans Affairs Palo Alto Health Care System, Palo Alto, California.

"The temptation to preferentially reach for digoxin in heart failure is also not supported by these data. Cautious initiation (or discontinuation) guided by digoxin levels seems biologically rational and empirically supported. However, a more pragmatic solution is to simply stop using it."

The idea that patients taking digoxin for AF should have their serum levels monitored isn't new, but in general it isn't done much, Lopes said. Monitoring would be a good idea, and it wouldn't have to be as frequent as monitoring of INR for people receiving warfarin.

"I don't think we have the data to say it should be done every month."

The current analysis tries to extend the serum level–dependent risk for bad outcomes on digoxin seen in patients with heart failure to patients with AF, Paul Hauptman, MD, Saint Louis University School of Medicine, Missouri, said in an interview. And even with the study's limitations, "quite honestly they do about as good a job as you're ever going to see."

It clearly shows an association between digoxin serum levels and increased mortality in AF, but "it's not enough to say there's causality, based on this study."

Hauptman said he hopes the current analysis will at least lead to more patients with AF to have their digoxin levels checked. "If these data are correct about atrial fib," he said, "we would take a great stride forward if clinicians just checked it once."

Hauptman referred to digoxin as "the gift that keeps on giving," partially in reference to the seemingly unending controversy about whether the drug's hazards outweigh its benefits in AF.

Notably, a 2012 analysis of digoxin for AF in patients participating in the 2002 AFFIRM trial, an analysis not conducted by AFFIRM investigators, saw digoxin associated with a significant rise in adjusted all-cause mortality whether or not the patients had heart failure.

The next year, essentially in response to that report, AFFIRM researchers published their own analysis based on the trial, concluding that any significant mortality increase seen in patients taking digoxin was the result of confounding from other factors, especially age and comorbidities.

A coauthor of that latter report, Wilbert S Aronow, MD, Westchester Medical Center and New York Medical College, Valhalla, New York, told theheart.org | Medscape Cardiology that he stands by its findings.

That report read, in part, that AFFIRM provides "no evidence of an increased risk of mortality or hospitalization among those receiving digoxin for rate control, either as monotherapy or in combination with other rate-control drugs. These findings do not support the recent suggestion that the use of digoxin in AF should be questioned nor support that there is a need to reassess the role of digoxin in the management of AF in patients with and without HF."

Aronow pointed to the various limitations of the current analysis, such as its nonrandomized nature and the likelihood of residual confounding (as did Hauptman, the editorialist Turakhia, and indeed Lopes and his coauthors), which "limit the usefulness of this paper."

The appearance of greater mortality on digoxin in the 2012 analysis of AFFIRM data is likely due to its "use of digoxin as a time-dependent treatment variable," according to the responding 2013 report by the AFFIRM investigators.

"If treatment with digoxin was continued for sicker patients, many of whom had HF and those who developed HF during the follow-up, then the observed higher mortality associated with digoxin use is not a real treatment effect, but a confounded association due to a higher sickness burden," the report says.

"Patients with atrial fibrillation receiving digoxin may be at increased risk of mortality because of comorbidities than patients not prescribed digoxin," Aronow said when interviewed. "If the serum digoxin level is 1.2 ng/mL or higher, you may increase mortality because at that level digoxin exerts a positive inotropic effect."

Baseline Users and New Users

Lopes and his colleagues looked at relationships between outcomes, digoxin use, and digoxin serum levels in 17,897 patients in the ARISTOTLE trial. They did it by adjusting for propensity scores for those receiving the drug at baseline and by comparing those newly starting it during the trial with propensity-matched controls.

No increased risks for death from any cause or for clinical endpoints were seen in the 5824 patients receiving digoxin at baseline, which Lopes said is evidence of "survivor bias" — that is, patients taking digoxin at baseline tended to be those who fared well on the drug anyway. 

Table 1. Adjusted Hazard Ratio for Clinical Events for Patients Taking vs Not Taking Digoxin at Baseline in ARISTOTLE

Endpoints Hazard Ratio (95% CI) P Value
All-cause mortality 1.09 (0.96 - 1.23) .19
Cardiovascular death 1.11 (0.93 - 1.32) .24
Sudden cardiac death 1.27 (0.96 - 1.67) .09
Noncadiovascular death 1.14 (0.92 - 1.42) .24
Heart failure hospitalization 1.00 (0.85 -1.16) .95

 

However, there was a continuous, linear relationship between mortality and serum digoxin levels among the 76% of patients initially receiving digoxin for whom serum concentrations were available.

The adjusted risk for death among those receiving vs not receiving digoxin went up significantly for every 0.5-ng/mL increment in baseline serum digoxin levels (hazard ratio [HR], 1.19; 95% CI, 1.07 - 1.32) and for every such elevation of 0.1 ng/mL (HR, 1.04; 95% CI, 1.01 - 1.06).

The risk for death jumped 56% for patients with baseline levels of at least 1.2 ng/mL: HR of 1.56 (95% CI, 1.20 - 2.04).

Each of 779 new-user patients, those not receiving digoxin at baseline but who started taking it during the study, were propensity-matched to 3 patients who didn't receive digoxin. Mortality and the risk for sudden death in particular were significantly increased in the new digoxin users.

Table 2. Hazard Ratio for Clinical Events in Patients Newly Starting Digoxin vs Non–Digoxin Recipients in ARISTOTLE

Endpoints Hazard Ratio (95% CI) P Value
All-cause mortality 1.78 (1.37 -2.31) <.0001
Cardiovascular death 1.60 (1.07 - 2.38) .0218
Sudden cardiac death 2.14 (1.11 - 4.12) .0230
Noncadiovascular death 1.67 (1.12 - 2.49) .0121
Heart failure hospitalization 1.69 (1.15 -2.49) .0083

 

Similar results for most of the endpoints were seen for patients with and without heart failure, Lopes said.

"A randomized prospective trial is essential to determine the effect of digoxin on mortality in patients with atrial fibrillation," according to Aronow.

Lopes also welcomed such a trial but said, "I worry now about equipoise and ethical considerations, because I'm not sure how much physicians and patients would be willing to accept being randomized to digoxin, after so many important studies showing harm."

According to Hauptman, "Everyone acknowledges that there will never be a properly powered study of digoxin in atrial fibrillation, so we're going to have to rely on analyses like this. I think that this is about as comprehensive an analysis as you're going to get."

The ARISTOTLE study was funded by Bristol-Myers Squibb and Pfizer, which also partly supported the current analysis. Lopes discloses receiving research grants from Bristol-Myers Squibb, GlaxoSmithKline, Medtronic, and Pfizer and consulting fees or honoraria from Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi-Sankyo, GlaxoSmithKline, Medtronic, Merck, Pfizer, and Portola. Disclosures for the other authors are in the report. Turakhia discloses research or grant support from Janssen Pharmaceuticals, Medtronic, AstraZeneca, Cardiva Medical, AliveCor, Amazon, Zipline Medical, iBeat, and iRhythm Technologies and honoraria from Abbott, Medtronic, Boehringer Ingelheim, Precision Health Economics, iBeat, Akebia, Cardiva Medical, and theheart.org | Medscape Cardiology. Hauptman said he has no relevant conflicts. Aronow said he has no conflicts.

J Am Coll Cardiol. 2018;71:1063-1074, 1075-1077. Abstract, Editorial

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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.
Post as:

processing....