Duel Over Digoxin: Is It Safe for HF?

Digoxin has long been a mainstay in the treatment of systolic heart failure (HF). Recently, however, some studies have called its safety into question after finding the use of the drug in contemporary practice was associated with higher mortality rates.

One recent study found incident use of digoxin in patients with newly diagnosed systolic HF was associated with a 72 percent increased risk of death. Risk was higher even among patients who took other heart-related medications (Circ Cardiovasc Qual Outcomes 2013; 6:1-8).

“Current guidelines recommend digoxin with beta-blockers, but we found that even if people took digoxin with them, they still have a higher risk of dying,” says study co-author Alan S. Go, MD, of Kaiser Permanente Northern California.

These findings are similar to results of a follow-up analysis of data from AFFIRM (AF Follow-Up Investigation of Rhythm Management) (European Heart Journal 2013; 34:1481-1488). In this study, all-cause mortality was 41 percent higher in patients with atrial fibrillation (AF) on digoxin whether they had HF or not.

The original AFFIRM study randomized 4,060 patients with AF with a high risk for stroke to receive either rate control drugs or rhythm control drugs over a four-year period. After following patients for an average of 3.5 years, the researchers found a rhythm control strategy did not improve survival compared with a rate control strategy.

In the follow-up analysis, the authors re-analyzed the AFFIRM data to determine whether digoxin use was a predictor of all-cause, cardiovascular and arrhythmic cardiovascular mortality in all AF patients as well as patients with and without congestive HF and/or an ejection fraction of less than 40 percent.

In addition to the higher rate of all-cause mortality, they found that patients taking digoxin had a 35 percent higher rate of cardiovascular mortality and a 61 percent higher rate in cardiovascular mortality related to arrhythmia. “These findings call into question the widespread use of digoxin in patients with AF,” the authors wrote.

Samy Claude Elayi, MD, of the University of Kentucky College of Medicine and the senior author of the AFFIRM study, acknowledges that patients were not randomized to receive or not receive digoxin. “Patients in general, whether they had heart failure or not, had increased mortality on digoxin, but it’s not clear whether it was digoxin or because the patients who took the drug were sicker,” he says.

Although these trials made some experts wary about the safety of the drug, others say the final verdict is not in yet, given methodological limitations of both trials.

“Since the DIG [Digitalis Investigation Group] trial, we know that digoxin has been reserved for patients with more severe heart failure often not used until other therapies are failing. So selective use introduces a tremendous bias,” says Kirkwood Adams, MD, of the University of North Carolina in Chapel Hill. DIG is a large randomized controlled trial that compared digoxin to placebo in patients with chronic HF. “Naturally if patients treated with digoxin are sicker, they’re more likely to die than patients not given the drug.”

Go says patients taking digoxin in his trial were healthier in some ways. They tended to be older, hypertensive, diabetic and on beta-blockers, but were less likely to have had a heart attack and on other heart-related medications.

Clinical use

Elayi argues that their findings inform practice. “In heart failure patients, this study reminds us that if we use digoxin, we should be extremely careful,” he says. “We need to use a low dose and carefully monitor patients. The dose that is toxic is close to the dose that is therapeutic.”

Findings from other research, however, are quite different.

DIG researchers recently found in an additional analysis of data acquired in the 1990s that digoxin significantly reduced HF hospitalizations and 30-day all-cause hospital admissions among chronic HF patients, although there was no reduction in all-cause mortality (American Journal of Medicine, 2013; 126:701-708).

“It’s a neutral drug when it comes to mortality,” says co-author Ali Ahmed, MD, of the University of Alabama at Birmingham.

Ahmed argues that digoxin may be a fiscally desirable option for hospitals because of the association between use of the drug and a lower rate of hospitalizations. Medicare penalizes some 30-day HF readmissions. While DIG did not assess readmission rates, Ahmed presented data at the European Society of Cardiology Heart Failure Association meeting this year that suggested digoxin use may also lower 30-day all-cause readmission. 

“Heart failure is a leading cause of readmissions and we have a drug that can potentially reduce readmissions without increasing mortality or delaying hospitalizations,” he says.

Patient selection

There is also debate over the use of the drug among patients with AF. Ahmed says the drug is still prescribed for people with AF because it can control heart rate. “Most people with AF are older adults, and in that subset of patients, rate control is probably better than rhythm control,” he explains.

Go counters that it has been shown that digoxin is not a very good way to control heart rate in people with AF. “If a person has AF but not HF, why use digoxin?” he says. “There are no benefits and only risks.”

Adams, who was not involved in the recent digoxin research, calls himself a proponent of the drug’s use, but not for everyone. “Our practice, consistent with current guidelines, is to use digoxin targeting serum concentrations less than 1 ng/ml, in patients with persistent symptomatic heart failure due to systolic dysfunction,” he says.

He believes the drug should be used in people with class III and IV HF with an ejection fraction (EF) of less than 25 percent. Class II patients also may benefit from the drug as long as the concentration is low.

“Many of us also feel it would be safer if given with a beta-blocker,” he adds. Spironolactone and digoxin also may be a good combination. The risks would be reduced even further, he argues, by using spironolactone, digoxin and an implantable cardioverter-defibrillator.

Ahmed supports a dose of 0.125 mg in HF patients, which is associated with lower serum digoxin concentrations and better outcomes. This dose is lower than the 0.25 mg dose used in the DIG trial. The higher dose, however, was standard in the early 1990s when the original DIG data were gathered.

Adams argues that the methodological limitations of the recent digoxin research leave him unconvinced of the need to change the use of digoxin in patients with HF due to reduced EF. But Elayi says there are other drugs available that are safer and cheaper than digoxin. 

Go adds that other medications, such as ACE inhibitors, beta-blockers and angiotensin-receptor blockers, are better choices for patients with systolic HF. Supplementation with diuretics is also an option. “The original DIG trial was not performed in the era of contemporary treatments for HF,” he explains. “We can optimize the treatment of HF with other medications now.”

Providers see digoxin as an option for HF treatment and although other medications have emerged, the value of digoxin is ingrained in the minds of many practitioners, he says. “Our study suggests that we need to re-evaluate who really will benefit from digoxin. Our concern is that maybe we are overusing it in patients with HF.”

Adjusted Hazard Ratios for Digoxin Use and Outcomes
TherapyDeath from any cause Hospitalization for heart failure
Overall cohort1.721.05
Concurrent beta-blockers*1.551.08
No concurrent beta-blockers*2.490.88
*Digoxin vs. nondigoxin users with nondigoxin users as reference
Source: Circ Cardiovasc Qual Outcomes 2013; 6: 1-8