Digitoxin in Heart Failure with Reduced Ejection Fraction (DIGIT-HF)

Authors, Journal, Affiliations, Type, DOI

• Bavendiek U, Bauersachs J, Berliner D, et al.
• New England Journal of Medicine, 2025;393:1155–1165
• Hannover Medical School and multiple German/Austrian/Serbian centres; 65 sites across 3 countries
• Phase 4 double-blind, randomised, placebo-controlled trial
• DOI: https://doi.org/10.1056/NEJMoa2415471
• Funding: German Federal Ministry of Education and Research (BMBF); non-industry-funded

Overview

DIGIT-HF was a phase 4, double-blind, randomised controlled trial evaluating digitoxin (0.07 mg once daily) versus placebo in 1,212 patients with HFrEF on contemporary guideline-directed medical therapy. The trial enrolled patients between May 2015 and September 2023 across 65 sites in Austria, Germany, and Serbia. The primary composite endpoint (all-cause death or first HF hospitalisation) was significantly reduced with digitoxin (HR 0.82; 95% CI 0.69–0.98; P=0.03), with an NNT of 22. All-cause death noninferiority was confirmed (P<0.001 vs threshold 1.303), providing the first evidence that a cardiac glycoside does not increase mortality in the modern GDMT era. Notably, 39.5% of patients received ARNi and 19.3% received SGLT2i — making this the first cardiac glycoside RCT on contemporary quadruple GDMT. The trial was substantially underpowered (enrolled 1,212 of a planned 2,190), and individual components did not reach statistical significance individually.

Keywords

Digitoxin, heart failure with reduced ejection fraction, cardiac glycosides, HFrEF, GDMT, digoxin, randomized controlled trial, cardiovascular outcomes, HF hospitalisation, all-cause mortality, noninferiority

Key Takeaways

Study Design and Population

• Phase 4 double-blind placebo-controlled RCT; 65 sites; Austria, Germany, Serbia (May 2015 – September 2023)
• Modified ITT population: n=1,212 (613 digitoxin, 599 placebo); planned sample 2,190 — significantly underpowered
• Median follow-up 36 months; median treatment duration 18 months (58.9% discontinued before study end)
• Digitoxin 0.07 mg OD — fixed dose adjusted at 6 weeks to target serum concentration 8–18 ng/mL
• Eligibility: LVEF ≤40% + NYHA III/IV, OR LVEF ≤30% + NYHA II; on evidence-based therapy ≥6 months
• Baseline demographics: Mean age 66; 20.4% women; mean LVEF 29%; 27.2% AF; mean eGFR 65 mL/min/1.73m²
• Background GDMT: Beta-blocker ≥93%; ARNi 39.5%; MRA 76.2%; SGLT2i 19.3% (partial data); ICD 64.3%; CRT 25.2%

Primary Endpoint

• All-cause death or first HF hospitalisation (composite): HR 0.82 (95% CI 0.69–0.98; P=0.03)
• Event rates: 39.5% (digitoxin) vs 44.1% (placebo); NNT=22
• First cardiac glycoside RCT to demonstrate significant primary composite benefit on contemporary GDMT

Secondary Endpoints

• All-cause death: HR 0.86 (95% CI 0.69–1.07) — not significant individually
• Noninferiority of all-cause death confirmed: P<0.001 (threshold 1.303) — no mortality detriment
• First HF hospitalisation: HR 0.85 (95% CI 0.69–1.05) — NS individually
• Total events (recurring events) rate ratio: 0.85 (P=0.20) — NS
• Benefit consistent across all prespecified subgroups including ARNi users, SGLT2i users, triple and quadruple GDMT users

Safety

• Serious adverse events: 4.7% (digitoxin) vs 2.8% (placebo)
• Cardiac serious adverse events: 3.4% vs 1.8%
• Higher SAE rate consistent with digitalis pharmacology (narrow therapeutic window); no specific toxidrome characterised separately

Context — Modern GDMT Era

• DIGIT-HF is the only cardiac glycoside trial conducted on modern ARNi + SGLT2i-containing GDMT
• Prior DIG trial (digoxin; 1997): enrolled on ACEi and diuretics only — no ARNi, no SGLT2i, no MRA in most; 28% hospitalization reduction with no mortality benefit
• DIGIT-HF used digitoxin (hepatically metabolised; less renal accumulation) rather than digoxin (renally cleared; toxicity risk in CKD)

Limitations of the document

• Substantially underpowered: 1,212 enrolled vs planned 2,190; primary composite significant but individual components both NS
• Median treatment only 18 months despite 36-month follow-up; 58.9% discontinued — intention-to-treat dilution, not per-protocol efficacy
• Digitoxin ≠ digoxin: Hepatic vs renal metabolism; different serum concentration targets; cannot directly extrapolate to digoxin (the agent with existing guideline recommendations)
• SGLT2i uptake partial (19.3%): Not all patients were on full quadruple GDMT; generalisability to complete modern regimens requires caution
• Three European countries only: Austria, Germany, Serbia — geographic and demographic limitations; external validity to other populations uncertain
• Competing background therapies: High CRT (25.2%) and ICD (64.3%) use may attenuate HF hospitalisation reduction — sicker, device-treated patients may have lower residual drug-preventable risk
• High SAE rate: 4.7% vs 2.8% — narrow therapeutic window of digitalis requires careful monitoring

Key Concepts Mentioned

• entities/HFrEF — primary disease context; quadruple GDMT background
• entities/Heart-Failure — broader syndrome context; cardiac glycoside positioning

Key Entities Mentioned

• entities/HFrEF — population studied; impact on digoxin/digitoxin line in HFrEF pharmacotherapy
• entities/Heart-Failure — master HF page updated with DIGIT-HF data

Wiki Pages Updated

• wiki/sources/digitoxin-hfref-digithf-nejm-2025.md — created
• wiki/entities/HFrEF.md — digoxin bullet enriched with DIGIT-HF data; new contradiction added; source_count 18→19
• wiki/entities/Heart-Failure.md — additional agents line updated with DIGIT-HF data
• wiki/sourceindex.md — new entry added
• wiki/wikiindex.md — HFrEF entry updated