DECISION Trial Misses Primary Endpoint as Pooled Meta-Analysis of 9,013 Patients Finds Digitalis Cuts Heart Failure Events 15 Percent

Overview

The Dutch-led DECISION trial of low-dose digoxin in chronic heart failure missed its primary endpoint when results were presented on May 10 at the Heart Failure Association of the European Society of Cardiology’s annual congress in Barcelona, according to a press release from the European Society of Cardiology. On the same day, a study-level meta-analysis combining DECISION with the 2025 DIGIT-HF digitoxin trial and the 1997 DIG trial — pooling 9,013 patients — was published in JAMA and found that digitalis glycosides reduced the combined risk of cardiovascular death or first worsening heart-failure event by 15 percent versus placebo, according to JAMA.

Taken together, the two readouts argue that low-dose digitalis — a class of drugs derived from the foxglove plant, according to Medical Xpress — still has a defined role in heart-failure care, even after a contemporary trial designed to revisit the question came up short on its own terms.

DECISION: a clean test, a null primary result

DECISION (NCT03783429) randomized 1,001 patients with symptomatic chronic heart failure and a left ventricular ejection fraction of 50 percent or less to low-dose digoxin or placebo at 43 sites in the Netherlands, according to the ESC. The trial targeted a serum digoxin concentration of 0.5 to 0.9 ng/ml — the low end of the therapeutic range — with periodic dose adjustments based on serum measurements, as described in the trial’s design paper. The mean age of participants was 73 years, 28 percent were women, and 29 percent had atrial fibrillation, according to the ESC.

Over a median follow-up of 36.5 months, 238 primary-outcome events — total worsening heart-failure events plus cardiovascular mortality — occurred in 131 of 500 patients in the digoxin group, compared with 291 events in 152 of 501 placebo patients, yielding a rate ratio of 0.81 (95% CI, 0.61 to 1.07; p=0.133), according to the ESC. The point estimate favored digoxin but did not reach statistical significance. The worsening heart-failure component trended in the same direction (rate ratio 0.76; 95% CI, 0.54 to 1.05), while cardiovascular mortality was nearly identical between groups (hazard ratio 0.93; 95% CI, 0.69 to 1.26).

Low-dose digoxin was generally well tolerated and safe, the ESC release reported. Principal investigator Professor Dirk van Veldhuisen of the University Medical Center Groningen framed the result in the press release: “Digoxin is the oldest drug in cardiovascular medicine, but there has been uncertainty about its value in HF(m)rEF management.”

A 15 percent reduction across three trials

The second presentation of the session was a study-level meta-analysis combining DECISION with the 2025 DIGIT-HF digitoxin trial and the 1997 DIG trial. Across 9,013 patients, digitalis glycoside treatment reduced the primary endpoint of cardiovascular death or first worsening heart-failure event with a hazard ratio of 0.85 (95% CI, 0.80 to 0.90; p<0.001), according to JAMA. The effect was driven mainly by worsening heart-failure events (HR 0.75; 95% CI, 0.69 to 0.81; p<0.001) rather than mortality: cardiovascular death and all-cause death were essentially unchanged in the pooled analysis.

The meta-analysis, published in JAMA on May 10 by lead author Kevin Damman alongside Dirk J. van Veldhuisen and Johann Bauersachs, drew on the three trials’ aggregate counts: 1,852 of 4,510 digitalis-treated patients (41 percent) hit the composite primary endpoint, versus 2,037 of 4,503 placebo patients (45 percent). The pooled population had a weighted mean age of 64.5 years and was 22 percent female. The included trials covered three eras of heart-failure care — the 1997 DIG trial enrolled 6,800 patients on largely diuretic-and-ACE-inhibitor backgrounds, while DIGIT-HF and DECISION enrolled patients on contemporary guideline-directed therapy.

Veldhuisen’s conclusion in the ESC release: “In patients with HF(m)rEF, low-dose digitalis glycosides seem to be an effective additional medical treatment option, which are cheap, safe and easy to use.”

Withdrawal signal

A third presentation by Professor Peter van der Meer examined 587 patients who had digoxin withdrawn, finding the post-withdrawal event rate ratio was 7.37 (95% CI, 1.56 to 34.88; p=0.012) compared with placebo, according to the ESC. The wide confidence interval reflects the small subgroup and rare-event nature of the analysis, but the direction is consistent with patients who had been deriving clinical benefit from digoxin losing it on discontinuation.

DIGIT-HF context

The digitoxin half of the meta-analysis comes from the DIGIT-HF trial, presented at the European Society of Cardiology Congress in Madrid in August 2025 and summarized by the American College of Cardiology. DIGIT-HF randomized 1,240 patients with advanced heart failure and reduced ejection fraction at 55 sites in Germany, Austria, and Serbia to digitoxin 0.07 mg daily plus guideline-directed therapy or placebo. The primary composite of all-cause death or hospital admission for worsening heart failure occurred in 39.5 percent of digitoxin-treated patients versus 44.1 percent on placebo over a median 36-month follow-up.

A published summary in European Heart Journal Supplements reports DIGIT-HF’s primary-outcome hazard ratio of 0.82 (95% CI, 0.69 to 0.98; p=0.03), corresponding to an absolute risk reduction of approximately 5 percent and a number-needed-to-treat of 22. Serious adverse events occurred in 4.7 percent of digitoxin patients versus 2.8 percent on placebo, according to the ACC, with the difference predominantly in cardiac disorders. Principal investigators included Johann Bauersachs and Udo Bavendiek at Hannover Medical School, as covered by Medical Xpress.

Why this matters

The 1997 DIG trial — the previous large outcome study, and the largest single contributor to the pooled meta-analysis at 6,800 patients according to JAMA — found digoxin reduced heart-failure hospitalizations without affecting all-cause mortality. Use has declined since as newer drug classes with proven mortality benefits have established themselves in the heart-failure treatment regimen.

DECISION was designed to test whether tight serum-level control could revive the drug’s profile in a modern background-therapy era. The trial cleared its safety bar — a meaningful result given digoxin’s narrow therapeutic index and historical concerns about proarrhythmic risk — but did not produce a statistically significant signal on its composite primary endpoint. The point estimates are uniformly directional and the worsening-heart-failure subcomponent trends consistently with the larger pooled analysis, but in regulatory terms a p of 0.133 does not unlock new labelling.

The meta-analysis is therefore the more consequential publication for clinical practice. By aggregating across three trials covering 9,013 patients and three decades of background therapy, it produces a tight confidence interval around a 15 percent relative reduction in the composite of cardiovascular death or first worsening event — small in absolute terms, but stable across trial subgroups, treatment periods, and the two specific digitalis glycosides studied. The mortality signal remains absent, which means digitalis is, on present evidence, a hospitalization-reducing add-on rather than a life-extending one.

What we do not know

DECISION’s null result on a composite primary endpoint will complicate any attempt to bring digoxin formally onto a contemporary heart-failure label. It remains unclear from the pooled analysis which patients benefit most, whether the benefit holds in patients already on the full four-pillar regimen of contemporary heart-failure therapy, or how the digitalis effect interacts with SGLT2 inhibitors specifically.

The DECISION trial was funded by the Netherlands Heart Foundation, according to the ESC, and remains the most rigorous test of low-dose digoxin in the contemporary era. Its negative primary result will sit alongside the positive meta-analysis as the next set of heart-failure guidelines decides what weight to give the totality of the evidence.