In this column, Dr. Paul Thompson offers his perspective on the use of digoxin and explains why he has maintained use in certain patient populations despite its waning popularity in recent years.
A new clinical trial (1) and the accompanying clinical commentary (2) suggest that digoxin may be coming back from the dead.
The Rate control Therapy Evaluation in permanent Atrial Fibrillation trial (RATE–AF) was an open-label, but endpoint-blinded study that assigned 160 patients, aged 60 or older with permanent atrial fibrillation (AF) to treatment with either digoxin or bisoprolol.1 Patients had New York Heart Association Class II or greater heart failure.
The primary outcome was quality of life (QOL) assessed by questionnaire at 6 and 12 months; there were 17-20 secondary endpoints. Neither QOL nor 16 of 17 secondary parameters were different at 6 months, although more digoxin-treated patients improved their European Heart Rhythm Association symptom classification score 2 or more classes (53 vs 9%, P <.001).
At 12 months, 8 of 20 secondary outcomes improved more with digoxin than with bisoprolol, including NTproBNP (960 vs 1250 pg/mL in the two groups, respectively). Serious adverse events occurred in 37 of the bisoprolol, but in only 15 of the digoxin patients. Heart rates did not differ.
Frankly, I was relieved with these results! I had never totally abandoned digoxin, partly because when I was trained in the 1970s, digoxin was THE drug for AF rate control. But I constantly doubted my continued use. That’s because a series of cross-sectional studies showed worse clinical outcomes in acute MI patients treated with digoxin.
These studies and the accompanying editorials left the drug for dead. I vaguely remember one academic electrophysiologist editorializing that use of digoxin was essentially malpractice. I dismissed these results and editorials by arguing that patients who received digoxin did so because they were sicker and developed AF. The TREAT-AF study, however, was a controlled study that showed increased mortality in digoxin treated patients, but this was not a true randomized controlled clinical trial because it used propensity matching to create the control group.2
I stuck with digoxin over the years—not for everybody, but for those AF patients whose rates were difficult to control. Most patients with atrial arrhythmias can be controlled with beta-blockers or calcium channel blockers alone, but some with AF, and many with atrial flutter, require either high drug doses or an additional agent.
Beta-blockers block adrenergic stimulation, which is actually not great at rest and increases with exertion. In contrast, digoxin increases vagal tone, meaning that its greatest effect is on resting heart rate—when vagal tone is greatest. Consequently, combining digoxin with moderate doses of beta-blockers provides both resting and exercise AF rate control.
Also, one of the biggest hassles I get with beta-blockers is that the nurses hold the drug because the patient is hypotensive. Actually, most often the patient is hypotensive because their AF rate is too fast. Fast heart rates decrease filling time, leading to low stroke volumes and low systolic pressures.
Slowing the heart rate often cures the hypotension, but I have become hoarse trying to change nurses’ minds on this issue. Digoxin has saved my voice because it does not get held anywhere near as frequently as beta- or calcium channel blockers, but I always get pushback from the new cardiology fellows when I suggest using digoxin. This resistance usually dissipates after they see how well it works.
So, here are my suggestions for using digoxin effectively:
I think that many inexpensive, useful drugs get abandoned because there is no industry push to use them. Others are victims of poor study designs, often industry-sponsored trials designed to make the generic drug look bad. None of the AF guidelines list digoxin as first-line therapy because there are no true RCCTs of digoxin.2 RATE-AF is not a perfect RCCT because it was open-labeled but hopefully it will prompt a closer look at this 235-year-old drug.