Reviews/Heart Failure Management/Sacubitril/valsartan does not reduce mortality in heart failure with preserved ejection fraction
HFPEF · HFM-HFPEF-02 · v2
The claim says that in people with heart failure and a preserved pumping function, adding sacubitril/valsartan did not clearly lower the combined rate of cardiovascular death or heart failure hospitalisation. This matches exactly what the one large dedicated trial in this group reported: a modest reduction that fell just short of statistical significance [10]. Later analyses of the same trial patients restate that the main combined outcome was not significantly reduced [2], and a related renal-subgroup report shows the same borderline effect [6]. A broader analysis pooling patients across the whole range of pumping function suggests the benefit is concentrated in those with weaker hearts and fades as function improves [1]. The panel treated the claim as a faithful statement of the trial's directly measured result, while noting that nearly all the evidence comes from this single trial.
The pivotal randomised trial [10] enrolled 4822 patients with ejection fraction 45% or higher and reported the primary composite of total heart failure hospitalisations and cardiovascular death at a rate ratio of 0.87 (95% CI 0.75 to 1.01, P=0.06), concluding the drug 'did not result in a significantly lower rate' of that composite. A secondary analysis of the same trial states the drug 'did not significantly reduce total HF hospitalizations and cardiovascular death' in this population [2], and a renal-function subgroup analysis reports the same primary effect (RR 0.87, 95% CI 0.75 to 1.02) in the no-decline group [6]. The pooled EF-spectrum analysis [1] found an overall first-event HR of 0.84 (95% CI 0.78 to 0.90) that mixes reduced and preserved ejection fraction and was modified by LVEF (interaction P=0.02), with benefit diminishing toward higher ejection fraction. The corpus is internally consistent on the HFpEF-specific primary result. The contested point: The division is not over whether the trial result is non-significant but over how to read it: the effect was borderline (P=0.06, upper CI 1.01) with prespecified subgroup suggestions of benefit in women and at lower ejection fraction [10], and the pooled analysis [1] shows benefit extending toward mildly reduced ejection fraction. The panel held that these points bound interpretation without contradicting the claim as worded.
All 10 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.
Essentially all HFpEF primary-endpoint evidence derives from a single trial (PARAGON-HF, NCT01920711); papers [2] and [6] are re-analyses of the same patients, not independent replications, so evidential independence is low.
The result was borderline (P=0.06, upper CI 1.01) with prespecified subgroup suggestions of benefit in women and at lower EF; a reader could mis-frame 'non-significant' as 'no effect', though the claim as worded ('does not significantly reduce') remains exactly correct.
Where it remains uncertain. The result was borderline with subgroup suggestions of benefit in women and at lower ejection fraction [10], and the pooled EF-spectrum analysis [1] shows benefit concentrated below the preserved range, but these bound interpretation rather than overturn the claim.
Aleatory
The primary result sat at P=0.06 with the upper confidence limit at 1.01 [10], so random variation alone leaves the true effect direction near the no-difference boundary.
Epistemic
Nearly all HFpEF primary-endpoint evidence comes from one trial [10]; [2] and [6] are re-analyses of the same patients, so no independent replication exists in the corpus.
Inferential
The pooled HR 0.84 [1] combines reduced and preserved ejection fraction and is modified by LVEF (interaction P=0.02), so it cannot be inferred as an HFpEF-specific benefit.
The load-bearing papers from a corpus of 10, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Independent restatement (same trial) confirming the non-significant primary-endpoint finding
“did not significantly reduce total HF hospitalizations and cardiovascular death compared with valsartan in patients with HF with preserved EF (HFpEF)”
Panel auditInternal validity: high · scope of inference: Sacubitril/valsartan vs valsartan effect on the composite endpoint in elevated-NT-proBNP HFpEF; this paper itself assesses NT-proBNP prognosis and treatment effect.Key limitations: primary-endpoint statement is background, not this paper's own analysis · analysis centred on biomarker not endpoint
Reports PARAGON-HF primary-outcome rate ratios whose confidence intervals cross unity, consistent with non-significance
“PARAGON-HF (eGFR decline, rate ratio [RR]: 0.84; 95% CI: 0.52-1.36 and no eGFR decline, RR: 0.87; 95% CI: 0.75-1.02, Pinteraction = 0.92)”
Panel auditInternal validity: high · scope of inference: Whether early eGFR decline modifies sacubitril/valsartan's effect on primary outcomes in HFpEF (and HFrEF); estimates are subgroup-level.Key limitations: subgroup estimates, not overall effect · CIs wide in eGFR-decline subgroup · pools HFrEF and HFpEF trials
Directly establishes that the primary composite endpoint was not significantly reduced (CI crosses 1.0, P=0.06)
“Sacubitril-valsartan did not result in a significantly lower rate of total hospitalizations for heart failure and death from cardiovascular causes among patients with heart failure and an ejection fraction of 45% or higher.”
Panel auditInternal validity: high · scope of inference: Sacubitril-valsartan vs valsartan effect on total HF hospitalisations and cardiovascular death in NYHA II-IV HFpEF with EF ≥45% and elevated natriuretic peptides.Key limitations: near-significant P=0.06 with CI upper bound at 1.01 · prespecified subgroup heterogeneity suggested (lower EF, women) · comparator is valsartan not placebo
Contextual: benefit concentrates below the preserved range, supporting the absence of a significant effect in HFpEF
“The effect of sacubitril/valsartan was modified by LVEF (treatment-by-continuous LVEF interaction P =0.02), and benefit appeared to be present for individuals with EF primarily below the normal range”
Panel auditInternal validity: high · scope of inference: Treatment effect of sacubitril/valsartan on CV death or HF hospitalisation across the whole LVEF spectrum, with EF-dependent modification.Key limitations: pools HFrEF and HFpEF · no isolated HFpEF primary-endpoint estimate in abstract · post-hoc EF categorisation
4 of 10 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.
The claim restates the exact primary-endpoint result of a large, well-conducted pivotal RCT [10] that directly measured the named composite in the named population and found non-significance.
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 HFpEF confirmatory trial. All primary-endpoint evidence traces to one trial [10]; [2] and [6] are not independent replications, so the non-significant finding has no second-trial confirmation. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A second adequately powered randomised trial of sacubitril/valsartan versus comparator in heart failure with preserved ejection fraction, reporting the same composite primary endpoint.
Open question: Would an independent, adequately powered HFpEF trial reproduce the non-significant primary composite result reported in PARAGON-HF [10]? Why it is still open: The corpus contains only one dedicated HFpEF trial; [2] and [6] re-analyse the same patients rather than replicate the finding.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19