H-ISDN Meta-Analysis Lowers Mortality in Heart Failure, With Jawad Butt, MD, PhD

A prespecified meta-analysis of three placebo-controlled trials found hydralazine plus isosorbide dinitrate (H-ISDN) reduced all-cause and cardiovascular death in patients with heart failure with reduced ejection fraction (HFrEF). Jawad Haider Butt, MD, PhD, presented the pooled findings, drawn from 2101 patients across V-HeFT I, A-HeFT, and the newly completed H-HeFT trial, at the European Society of Cardiology (ESC) Congress 2026 in Munich, Germany.¹

H-ISDN cut the risk of all-cause death by 29%, though the effect on heart failure hospitalization proved less consistent.³

Guidelines currently reserve H-ISDN for patients unable to tolerate an ACE inhibitor, ARB, or ARNI, and for patients who self-identify as Black with NYHA class III or IV symptoms on foundational therapy. This guidance traces to the 2004 A-HeFT trial, which showed a 43% mortality reduction with H-ISDN added to standard therapy at the time.² The H-HeFT trial set out to test whether similar benefit extended to a broader, contemporary population.

Guideline recommendations for H-ISDN stem from small, older trials individually underpowered for mortality. The meta-analysis sought to strengthen the evidence base by combining all available placebo-controlled data in HFrEF.

H-ISDN Meta-Analysis Design and Mortality Findings

Butt and colleagues combined trial-level data from V-HeFT I (1986), A-HeFT (2004), and H-HeFT to answer a question neither trial alone was powered to resolve: does H-ISDN reduce mortality? In the primary fixed-effect analysis, H-ISDN was associated with reduced all-cause death (HR, 0.71; 95% CI, 0.58-0.87) and cardiovascular death (HR, 0.65; 95% CI, 0.47-0.90) compared with placebo.³ A reduction in heart failure hospitalizations also emerged, but significant between-trial heterogeneity prompted a sensitivity analysis using a random-effects model.

Under the random-effects model, mortality estimates held steady, but the hospitalization benefit lost statistical significance.³ Butt noted 2 of the 3 trials were underpowered for mortality alone, part of the rationale for pooling the data. The meta-analysis was prespecified before H-HeFT's primary results were unblinded.

H-HeFT Trial Results and Clinical Implications

H-HeFT, presented separately by principal investigator Lars Køber, MD, enrolled 592 patients with HFrEF who did not self-identify as Black, all receiving foundational therapy. Over up to 7 years of follow-up, H-ISDN failed to reduce the composite primary endpoint of death, worsening heart failure, urgent intravenous or metolazone therapy, transplantation, or ventricular assist device implantation compared with placebo.¹

Event rates were 8.6 vs 9.3 per 100 patient-years (HR, 0.90; 95% CI, 0.67-1.21). All-cause death occurred in 19.4% of the H-ISDN group versus 24.2% of the placebo group (HR, 0.72; 95% CI, 0.51-1.02), a gap not statistically significant on its own.¹

Treatment discontinuation rates were high in H-HeFT, a factor that may have diluted the drug's apparent effect.¹ Because the primary endpoint in H-HeFT was neutral, Butt considers the meta-analysis findings hypothesis-generating rather than practice-changing.

Køber said a new randomized trial will be needed to establish whether H-ISDN truly reduces mortality.¹ Butt agreed such a trial remains justified, given the residual risk for death and hospitalization left by current four-pillar HFrEF therapy, though he noted the absence of commercial sponsorship makes investigator-initiated trials difficult to mount.

Editors’ Note: Butt reports disclosures with AstraZeneca, Bayer, and Novartis.

References

1. European Society of Cardiology. No demonstration of benefit with hydralazine plus isosorbide dinitrate in heart failure. Published August 30, 2026. Accessed September 4, 2026.

2. Taylor AL, Ziesche S, Yancy C, et al. Combination of isosorbide dinitrate and hydralazine in blacks with heart failure. N Engl J Med. 2004;351:2049-2057.

3. Butt JH. H-ISDN Meta-Analysis: Hydralazine-Isosorbide Dinitrate in HFrEF. Presented at: ESC Congress 2026; August 30, 2026; Munich, Germany.