Low-dose digoxin may not reduce the combined rate of worsening heart failure events and cardiovascular mortality compared with placebo among patients with heart failure and reduced or mildly reduced ejection fraction.
In the double-blind, placebo-controlled DECISION trial, researchers recruited 1,001 adult patients with symptomatic chronic heart failure, New York Heart Association functional class II to ambulatory IV symptoms, elevated natriuretic peptide concentrations, and a left ventricular ejection fraction (LVEF) of 50% or less despite guideline-directed treatment. Patients were randomly assigned to placebo or low-dose digoxin initiated at 0.1 or 0.2 mg daily according to age, renal function, and concomitant medications, with a target serum concentration of 0.5 to 0.9 ng/mL. Mean LVEF was 33%, and most participants had class II heart failure. Mean estimated glomerular filtration rate was 56 mL/min/1.73 m². Background treatment included beta-blockers, mineralocorticoid receptor antagonists, and sodium-glucose cotransporter-2 (SGLT2) inhibitors. About 53% received an angiotensin-receptor–neprilysin inhibitor.
The primary endpoint combined total hospitalizations and urgent hospital visits for worsening heart failure with cardiovascular mortality. Recurrent events were included in the primary analysis, which used a modified intention-to-treat approach excluding one randomized patient who did not receive study medication. Concentrations were measured 4 weeks following randomization and every 6 months thereafter, with dose adjustments guided by a central laboratory.
Over a median follow-up of 36.5 months, 238 primary outcome events occurred among 131 of 500 patients assigned to digoxin vs. 291 events among 152 of 501 patients receiving placebo, with event rates of 15.7 and 19.3 per 100 patient-years, respectively. The researchers recorded 155 worsening heart failure events among patients receiving digoxin vs. 203 among those receiving placebo. Cardiovascular mortality occurred in 83 patients (17%) receiving digoxin and 88 (18%) receiving placebo. In a separate time-to-first-event analysis,
All-cause mortality was similar between groups, occurring in 118 patients (24%) receiving digoxin and 126 (25%) receiving placebo. Total cardiovascular hospitalizations and urgent cardiovascular visits numbered 191 in the digoxin group and 169 in the placebo group. Total hospitalizations did not differ between groups. Because the primary endpoint was not statistically significant, the researchers considered subsequent outcome analyses exploratory.
The researchers reported no statistically significant differences in treatment effects across 11 prespecified subgroups, including those defined by sex, heart rhythm, and ejection fraction.
Among 284 women, the estimated primary outcome event rate was lower with digoxin, with a rate ratio of 0.71 vs. 0.85 among 717 men. Corresponding rate ratios were 0.76 among patients with atrial fibrillation or flutter and 0.82 among those in sinus rhythm.
The researchers reported no differing treatment effects between patients with mildly reduced ejection fraction and those with reduced ejection fraction. The subgroup findings were exploratory and did not establish efficacy in any specific population.
Low-dose digoxin was generally well tolerated. Serious adverse events, excluding primary and secondary endpoints, occurred at rates of 19.9 events per 100 patient-years with digoxin and 18.3 with placebo.
Treatment-related serious adverse events were reported in 40 patients (8%) receiving digoxin and 25 (5%) receiving placebo, a difference that was not statistically significant. No statistically significant differences were observed in patient-reported adverse effects or device implantation rates.
The median serum digoxin concentration was 0.6 ng/mL. At the 4-week assessment, 64% of patients with available measurements had concentrations within the target range. Following repeat measurement among patients initially outside the range, 84% with available samples were within the target range.
The researchers identified several limitations, including a predominantly male study population in which most patients had class II heart failure, limiting generalizability to populations with different clinical characteristics.
Treatment discontinuation was greater than anticipated. At the end-of-study visit, 24% of assessed patients in the digoxin group and 21% in the placebo group discontinued medications for reasons other than mortality.
The researchers identified disruptions associated with the COVID-19 pandemic, patient age, and changes in background heart failure treatment as possible contributors to the findings.
At the end of follow-up, SGLT2 inhibitor use was higher in the placebo group than in the digoxin group (80% vs. 72%), a treatment difference that could have influenced the overall results.
In an exploratory on-treatment sensitivity analysis that censored patients following study-drug discontinuation, the researchers recorded 135 primary outcome events in the digoxin group vs. 212 in the placebo group. The corresponding event rates were 10.60 and 16.12 per 100 patient-years, an estimated 34% lower rate with digoxin. This analysis did not change the nonsignificant result of the primary randomized comparison.
The researchers noted that the observed primary outcome difference was smaller than the 22% reduction assumed in the trial's sample size calculation.
The researchers compared DECISION with previous randomized studies of digitalis glycosides. The earlier DIG trial did not demonstrate a reduction in all-cause mortality with digoxin, although worsening heart failure hospitalizations were reduced. The more recent DIGIT-HF trial reported a reduction in its composite outcome with low-dose digitoxin in patients with more advanced heart failure.
They concluded that low-dose digoxin did not significantly reduce the primary composite endpoint when added to contemporary guideline-directed treatment. However, they suggested that low-dose digoxin "may be an easy, simple, and cheap treatment option" for patients with reduced or mildly reduced ejection fraction.
Full disclosures are provided in the published study.
Source: Nature Medicine
