Hydralazine-ISDN in Patients With Chronic Heart Failure — Hydralazine Heart Failure Trial (H-HeFT), component of the DANish randomised, double-blind, placebo-controlled trial in patients with chronic HEART failure (DANHEART)

H-HeFT

Patient / Population Intervention / Exposition Comparaison Critère de jugement

In 592 patients with HFrEF on guideline-directed therapy, hydralazine–isosorbide dinitrate did not reduce death or worsening heart failure compared with placebo.

N

592patients

Schéma

investigator-initiated, multicentre (23 centres in Denmark), randomised 1:1, double-blind, placebo-controlled, event-driven; part of the 2 × 2 factorial DANHEART programme with Met-HeFT (patients may enter one or both substudies); hydralazine 37.5 mg/isosorbide dinitrate 20 mg twice daily, uptitrated, vs matching placebo; symptomatic chronic heart failure, LVEF ≤40%, systolic BP ≥100 mmHg, NT-proBNP >350 pg/mL or BNP >80 pg/mL; follow-up up to 7 years

Critère

Time to first occurrence of the composite of death or hospitalisation with worsening heart failure (including an urgent visit requiring intravenous therapy or metolazone for heart failure, heart transplantation or LVAD implantation)

Pertinence

2Important — l’un des piliers parmi d’autres.

RésultatPrimary composite 8.6 vs 9.3 events per 100 patient-years (HR 0.90, 95% CI 0.67-1.21) over follow-up up to 7 years — no significant difference. All-cause death 19.4% vs 24.2% (HR 0.72, 95% CI 0.51-1.02). Treatment discontinuation was high. A companion meta-analysis of 3 trials (A-HeFT, H-HeFT and one other, 2101 patients) showed lower all-cause death (HR 0.71, 0.58-0.87) and cardiovascular death (HR 0.65, 0.47-0.90), with inconsistent effects on heart failure hospitalisation.

Køber L, et al. H-HeFT (DANHEART). Presented at ESC Congress 2026, Hot Line 6, 30 August 2026, Munich.

Discussion et critique

The first test of hydralazine-isosorbide dinitrate on top of four-pillar therapy, in a population defined by ejection fraction and natriuretic peptide rather than by race, and it was not powered for mortality. The mortality curves separate late and the intervals cross one; hospitalisations do not move at all, which is the harder finding to explain away for a vasodilator. Discontinuation was frequent enough that the investigators name tolerability as the practical limit — an old, cheap, six-tablets-a-day regimen has to be taken to work.