The Effect of Spironolactone on Morbidity and Mortality in Patients with Severe Heart Failure (RALES)

Authors, Journal, Affiliations, Type, DOI

• Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky J, Wittes J, for the Randomized Aldactone Evaluation Study Investigators
• New England Journal of Medicine, 1999;341:709–717
• University of Michigan (Pitt, Cody); Centre d'Investigation Clinique de Nancy, France (Zannad); STICARES Rotterdam (Remme); Hôpital Henri Mondor, France (Castaigne); Searle Global Medical Operations (Perez); Statistics Collaborative Washington DC (Palensky, Wittes); 195 centres, 15 countries
• Double-blind, placebo-controlled, randomised controlled trial
• DOI: https://doi.org/10.1056/NEJM199909023411001
• Funding: Searle (manufacturer of spironolactone/Aldactone); no independent sponsor

Overview

RALES was a landmark double-blind RCT enrolling 1,663 patients with severe HFrEF (NYHA III/IV with required recent NYHA IV; LVEF ≤35%) at 195 centres in 15 countries. Background therapy was ACEi + loop diuretic ± digoxin; beta-blockers were not required — the trial was conducted at the dawn of the BB era (CIBIS-II and carvedilol were published concurrently in 1999). Spironolactone 25 mg OD vs placebo. Stopped early at mean 24 months when the interim analysis crossed the prespecified efficacy boundary. Spironolactone reduced all-cause mortality by 30% (RR 0.70; P<0.001) — one of the largest mortality reductions ever seen in an HF RCT at the time — driven by reductions in both progressive HF death and SCD. RALES established spironolactone as a Class I therapy for NYHA III/IV HFrEF and is the foundational MRA trial; benefit in NYHA II was later established by EMPHASIS-HF (eplerenone).

Keywords

spironolactone, aldosterone antagonist, mineralocorticoid receptor antagonist, heart failure, LVEF, NYHA III/IV, mortality, sudden cardiac death, hyperkalemia, RALES

Key Takeaways

Study Design and Population

• Double-blind two-arm RCT: spironolactone 25 mg OD (n=822) vs placebo (n=841)
• Eligibility: NYHA III or IV at randomisation with required recent NYHA IV within 6 months; LVEF ≤35% within 6 months; ACEi + loop diuretic (if tolerated); HF diagnosis ≥6 weeks prior
• Exclusions: creatinine >2.5 mg/dL (221 µmol/L); potassium >5.0 mmol/L; potassium-sparing diuretics; primary operable valvular disease; recent transplant; active cancer
• Background therapy: ACEi (if tolerated; ~95% of patients) + loop diuretic + digitalis (~74%) — no beta-blocker requirement
• Recruitment: March 24, 1995 – December 31, 1996; 195 centres, 15 countries
• Stopped early: August 24, 1998 (mean follow-up 24 months) on recommendation of DSMB after fifth planned interim analysis exceeded prespecified critical z=2.02
• Dose flexibility: could increase to 50 mg OD at 8 weeks for progressive HF without hyperkalemia; could reduce to 25 mg every other day for hyperkalemia
• Mean dose achieved: 26 mg OD (essentially no net diuretic effect at this dose — no change in weight, sodium retention, or urinary sodium excretion)
• Funded by Searle

Primary Endpoint — All-Cause Mortality

• Placebo: 386 deaths (46%); spironolactone: 284 deaths (35%)
• RR 0.70 (95% CI 0.60–0.82; P<0.001) — 30% RRR; absolute −11% at mean 24 months

Secondary Endpoints

• Death from cardiac causes: RR 0.69 (95% CI 0.58–0.82; P<0.001)
• Death from progressive HF: RR 0.64 (95% CI 0.51–0.80; P<0.001)
• Sudden cardiac death: RR 0.71 (95% CI 0.54–0.95; P=0.02)
• HF hospitalisation: RR 0.65 (95% CI 0.54–0.77; P<0.001) — 35% lower
• NYHA class improvement: P<0.001 (Wilcoxon); improved in 41% spironolactone vs 33% placebo
• Effect onset: 2–3 months after initiation; persisted throughout follow-up

Subgroup Analyses

• Consistent across all 6 prespecified subgroups: age (median split), LVEF (<26% vs ≥26%), aetiology (ischaemic vs non-ischaemic), creatinine (median split), ACEi use, digitalis use
• Also consistent across sex, NYHA class, beta-blocker use (subgroup), and baseline potassium
• No prespecified subgroup showed attenuation

Safety

• Serious hyperkalemia (K ≥6.0 mmol/L): 2% spironolactone vs 1% placebo (P=0.42 — non-significant)
• Gynecomastia or breast pain in men: 10% vs 1% (P<0.001) — 10 vs 1 patient discontinued
• Median creatinine increased ~0.05–0.10 mg/dL in spironolactone arm; significant (P<0.001) but clinically minor per investigators
• Median potassium increased +0.30 mmol/L; significant (P<0.001) but clinically minor per investigators
• No significant differences in BP, HR, or serum sodium

Proposed Mechanism

• Spironolactone at 25 mg OD had no net diuretic effect in the prior dose-finding study and in RALES (no weight change, no natriuresis) — benefit is cardioprotective, not primarily diuretic
• Aldosterone at this dose drives: myocardial and vascular fibrosis; baroreceptor dysfunction; sympathetic activation; parasympathetic inhibition; impaired norepinephrine reuptake by myocardium
• Benefit attributable to anti-fibrotic and anti-adrenergic cardioprotection; explains reduction in both progressive HF death and SCD

Limitations of the document

• Stopped early (mean 24 months): Prespecified efficacy boundary crossed at fifth interim analysis — early termination typically overestimates magnitude of effect; the 30% RRR may be inflated vs a full-duration trial
• Pre-beta-blocker era: No BB requirement; CIBIS-II and carvedilol published simultaneously (1999). Incremental MRA benefit on top of modern ARNi + BB + SGLT2i quadruple GDMT is not directly established by RALES
• Severe HF only: NYHA III/IV (required recent class IV) — RALES explicitly notes "effectiveness in patients at lower risk...will require further prospective study"; EMPHASIS-HF (2011) extended MRA evidence to NYHA II with eplerenone
• Funded by Searle: Spironolactone manufacturer; no independent funder
• Real-world hyperkalemia divergence: RALES's 2% serious hyperkalemia rate required strict screening (creatinine ≤2.5, K ≤5.0). An observational study (Juurlink et al., JAMA 2004) found a 20-fold increase in spironolactone prescriptions after RALES, a 6.8-fold increase in hyperkalemia hospitalisation, and associated excess mortality in elderly patients — reflecting routine clinical failure to apply RALES exclusion criteria; this limits direct generalisability without careful patient selection

Key Concepts Mentioned

• entities/HFrEF — foundational MRA mortality trial; Class I NYHA III/IV basis; RALES data
• entities/Heart-Failure — MRA in HFrEF management

Key Entities Mentioned

• entities/HFrEF — RALES as the foundational MRA RCT establishing Class I recommendation; pre-BB era caveat
• entities/Heart-Failure — MRA RALES headline data for HFrEF pharmacotherapy summary

Wiki Pages Updated

• wiki/sources/mra-hfref-rales-nejm-1999.md — created
• wiki/entities/HFrEF.md — MRA bullet enriched with full RALES data; RALES contradiction added; source added; source_count 20→21
• wiki/entities/Heart-Failure.md — MRA/BB/SGLT2i line updated with RALES MRA data; source added; source_count 24→25
• wiki/sourceindex.md — new entry added at top
• wiki/wikiindex.md — HFrEF entry updated
• wiki/log.md — new entry added at top