Reviews/Heart Failure Management/Mineralocorticoid receptor antagonist reduces mortality in heart failure with reduced ejection fraction
Pharm · HFM-PHARM-04 · v2
Borderline: within 4 points of the Established/Likely boundary, so a re-grade could land either side.
In patients with symptomatic but mild heart failure with a weak pump, a large high-quality trial of the drug eplerenone found that patients taking it were less likely to die and less likely to be admitted to hospital for heart failure than those taking a dummy pill [10]. A second analysis that pooled this trial with one other drug trial found the same lower risk of death [7], and a further look at the same eplerenone trial confirmed the benefit on the combined death-or-hospitalisation measure [5]. One real-world study that simply watched who was prescribed these drugs did not see a benefit, but its authors said their data could not settle the question [6]. Because most of the strong supporting evidence comes back to the single eplerenone trial, the panel treated limited independence, rather than weak quality, as the main reason not to place even more weight on the claim.
A large double-blind placebo-controlled randomised trial of eplerenone in NYHA class II HFrEF reduced both all-cause death (hazard ratio 0.76; P=0.008) and the composite of cardiovascular death or heart-failure hospitalisation (HR 0.63), with heart-failure hospitalisations also reduced [10]. A pooled patient-level analysis of two MRA trials (N=4,396) reported all-cause mortality HR 0.72 (95% CI 0.64-0.82) [7], and an EMPHASIS-HF post-hoc confirmed the benefit on the composite endpoint in both subgroups examined [5]. The one real-world outcome test in the corpus found MRA use not associated with death or heart-failure hospitalisation (adjusted HR 1.02, 95% CI 0.66-1.58) but called its inference inconclusive [6]. The finerenone trials [1][8] used the surrogate NT-proBNP against an active comparator, and the eplerenone-versus-spironolactone cohort [11] addressed agent choice, not MRA versus no MRA. The contested point: The division is over independence rather than direction: the directly positive placebo-controlled signal largely traces to one trial, since [5] is a post-hoc of the same trial as [10] and [10] is one of the two trials pooled in [7]. The only real-world outcome test [6] was flat, though its authors describe their inference as inconclusive.
All 11 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Heart Failure Management, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The directly positive placebo-controlled evidence is not independent: [5] is a post-hoc of the same trial as [10], and [10] is one of the two trials pooled in [7], so the hard support rests substantially on a single RCT (EMPHASIS-HF).
The only real-world outcome test in the corpus, [6] BIOSTAT-CHF, found MRA use not associated with death or HF hospitalisation (adjusted HR 1.02, 0.66-1.58), a flat signal the authors call inconclusive.
Where it remains uncertain. The hard placebo-controlled support largely overlaps one trial ([10], [5], and half of [7]), and the sole real-world outcome test was null and self-described as inconclusive [6].
Aleatory
The only real-world outcome estimate carries a wide confidence interval spanning benefit and harm (adjusted HR 1.02, 0.66-1.58) [6].
Epistemic
Direct placebo-controlled support derives chiefly from one trial reused across [10], [5], and [7], limiting independent confirmation.
Inferential
The finerenone trials [1][8] report NT-proBNP change, not death or HF hospitalisation, so no mortality inference follows from them.
The load-bearing papers from a corpus of 11, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
EMPHASIS-HF post-hoc reinforcing reduction of the composite of CV death or HF hospitalisation
“a significant benefit of eplerenone for the primary outcome was noted in both normal [hazard ratio (HR) 0.77, 95% confidence interval (CI) 0.61-0.98, P = 0.03] and increased (HR 0.48, 95% CI 0.37-0.63, P < 0.0001) WC subgroups”
Panel auditInternal validity: medium · scope of inference: Mildly symptomatic (NYHA II) HFrEF patients, effect of eplerenone overall and by abdominal obesity.Key limitations: post hoc subgroup analysis · composite is CV death not all-cause death · hypothesis-generating per authors
Pooled RALES + EMPHASIS-HF patient-level analysis corroborating all-cause mortality reduction across the MRA class
“All outcomes were reduced by MRA therapy overall, with consistent effects across SBP categories (e.g., all-cause mortality, overall hazard ratio [HR] of 0.72; 95% CI: 0.64 to 0.82; p < 0.001”
Panel auditInternal validity: high · scope of inference: HFrEF patients randomized in RALES and EMPHASIS-HF, effect of MRA on outcomes across baseline SBP.Key limitations: secondary analysis focused on blood pressure · HF hospitalisation not quantified separately in abstract
Direct placebo-controlled RCT evidence of all-cause mortality reduction with an MRA in symptomatic (NYHA II) HFrEF
“Hospitalizations for heart failure and for any cause were also reduced with eplerenone.”
Panel auditInternal validity: high · scope of inference: Patients with systolic heart failure (EF ≤35%) and mild (NYHA II) symptoms, effect of eplerenone vs placebo.Key limitations: trial stopped early, which can inflate effect estimates · single MRA (eplerenone), mildly symptomatic population only
Both arms receive an MRA and the primary endpoint is NT-proBNP change; with no non-MRA comparator the abstract cannot say whether MRAs reduce mortality or HF hospitalisation.
Panel auditInternal validity: medium · scope of inference: Patients with worsening HFrEF plus CKD and/or diabetes, comparing one MRA with another on a biomarker endpoint.Key limitations: head-to-head of two MRAs, no placebo · surrogate primary endpoint · phase 2, underpowered for clinical events
4 of 11 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Direct placebo-controlled RCT plus pooled RCT data carry both endpoints of the claim, with independence rather than quality the main limiter.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Independent placebo-controlled confirmation. Nearly all direct positive evidence traces to EMPHASIS-HF, so a separate trial would test whether the effect is reproducible rather than trial-specific. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A contemporary placebo-controlled MRA outcomes trial in symptomatic HFrEF on current background therapy, powered for death and HF hospitalisation.
Open question: Does the MRA mortality and HF-hospitalisation benefit reproduce in a placebo-controlled trial independent of EMPHASIS-HF? Why it is still open: Direct positive evidence overlaps across [10], [5], and [7], all tracing to one trial, so independence is unestablished.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19