Reviews/Heart Failure Management/SGLT2 inhibitors reduce mortality in heart failure with reduced ejection fraction
Pharm · HFM-PHARM-01 · v2
Two large, well-designed trials gave either an SGLT2 inhibitor or a placebo to people whose hearts pump weakly, then counted deaths from heart causes and hospital stays for worsening heart failure. In DAPA-HF [9] the combined count of those events fell by about a quarter, and in EMPEROR-Reduced [10] it fell by a similar amount, each with results that were statistically clear. The hospitalisation part of the combined count was the stronger and most consistent driver in both trials. Death from heart causes on its own was reduced in [9] but by a smaller and less certain margin, and was not reported on its own in [10]. A pooled analysis [11] and a real-world cohort [12] pointed the same way. The panel treated the finding as firmly supported, while noting it rests on two specific drugs standing in for the whole class.
Two pivotal double-blind randomised trials in the exact population — reduced ejection fraction, EF 40% or less — each cut the composite of cardiovascular death or heart-failure hospitalisation: DAPA-HF [9] hazard ratio 0.74 (95% CI 0.65 to 0.85) and EMPEROR-Reduced [10] hazard ratio 0.75 (95% CI 0.65 to 0.86). The hospitalisation component was robust in both — first worsening-HF event HR 0.70 (0.59 to 0.83) in [9] and total HF hospitalisations HR 0.70 (0.58 to 0.85) in [10]. Benefit was consistent regardless of diabetes status in both trials. A prospectively designed pooled analysis [11] reported a consistent primary-outcome HR of 0.76 to 0.77, and a single-centre ischaemic-HF cohort [12] a composite HR of 0.24, pointing the same direction. The contested point: The claim conjoins cardiovascular death AND heart-failure hospitalisation, but both trials are powered on the composite and the hospitalisation component drives it more than death. Cardiovascular death alone is only modestly significant in [9] (HR 0.82, CI to 0.98) and is not reported as individually significant in [10]'s abstract.
All 12 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 claim conjoins CV death AND HF hospitalisation, but both pivotal trials are powered on the composite. CV-death-alone is only modestly significant in DAPA-HF [9] (HR 0.82, CI to 0.98) and is not reported as individually significant in EMPEROR-Reduced's abstract — the hospitalisation component drives the composite more than CV death.
[12] is a small single-centre observational cohort (N=222) comparing initiation timing (both arms receive SGLT2i), confounded by indication; its HR 0.24 is implausibly large and not a placebo comparison — supportive only.
Where it remains uncertain. The composite is driven more by hospitalisation than by cardiovascular death, which is only modestly significant in [9] (HR 0.82) and not reported separately in [10], and only two molecules stand in for the whole class.
Aleatory
Each trial's estimate rests on a bounded number of events, e.g. 386 vs 502 primary events in [9] and 361 vs 462 in [10], so the confidence intervals reflect irreducible sampling variation.
The load-bearing papers from a corpus of 12, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Pivotal HFrEF RCT: composite of worsening HF or CV death reduced 26% vs placebo.
“Death from cardiovascular causes occurred in 227 patients (9.6%) in the dapagliflozin group and in 273 patients (11.5%) in the placebo group (hazard ratio, 0.82; 95% CI, 0.69 to 0.98)”
Panel auditInternal validity: high · scope of inference: HFrEF patients (NYHA II-IV, EF ≤40%), with and without type 2 diabetes, on recommended therapy.Key limitations: primary HF component is a worsening-HF event (hospitalisation or urgent visit), not hospitalisation alone
Second independent pivotal HFrEF RCT, different molecule/sponsor: composite reduced 25%.
“The total number of hospitalizations for heart failure was lower in the empagliflozin group than in the placebo group (hazard ratio, 0.70; 95% CI, 0.58 to 0.85; P<0.001).”
Panel auditInternal validity: high · scope of inference: HFrEF patients (class II-IV, EF ≤40%) on recommended therapy, with and without diabetes.Key limitations: abstract reports the composite and total HHF, not a separate significant CV-death result
Pooled primary-data analysis showing consistent HHF-or-CV-death benefit (across EF strata).
“The efficacy and safety of empagliflozin were consistent irrespective of revascularization history for all major outcomes (primary outcome: REVASC: 0.76 [0.66-0.89], no REVASC: 0.77 [0.68-0.87], Pinteraction: 0.94).”
Panel auditInternal validity: high · scope of inference: Empagliflozin efficacy across the pooled HF spectrum (EF below and above 40%) by prior coronary revascularisation.Key limitations: pools HFrEF with HFpEF · not stratified by ejection fraction · no HFrEF-only estimate in abstract
This is a design paper that states the trial 'will determine' efficacy; no outcome data are reported.
Panel auditInternal validity: not assessable from abstract · scope of inference: Design for chronic HFrEF patients (LVEF ≤40%, elevated NT-proBNP), with and without diabetes, testing dapagliflozin vs placebo.Key limitations: no results reported · protocol/design paper only
4 of 12 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.
Two pivotal double-blind RCTs in HFrEF deliver consistent, large, directly measured reductions in the hard composite outcome, meeting the Established bar despite a weaker CV-death-alone component and a molecule-vs-class caveat.
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.
Cardiovascular-death effect in HFrEF. The claim conjoins CV death, yet death alone is only modestly significant in [9] and not reported separately in [10], so the death-specific benefit is weakly characterised. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A pre-specified, adequately powered meta-analysis of individual HFrEF trial data reporting cardiovascular death as a standalone endpoint.
Open question: Does the SGLT2-inhibitor class reduce cardiovascular death on its own in HFrEF, independent of the hospitalisation component? Why it is still open: CV death alone is only modestly significant in [9] (HR 0.82, 0.69-0.98) and not reported individually in [10], so the composite is carried largely by hospitalisation.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19