Independent · Evidence-led · We don't sell peptides
EU / NordicsUpdated weeklyEN

Explore this article's sources with AI

ChatGPTClaudePerplexityGeminiGrokGoogle AI

Follow PeptideMethods on Google

First published

Semaglutide in HFpEF: Pooled Events

Three semaglutide studies favored fewer HFpEF events, but pooled mortality estimates remained inconclusive after 52 weeks.

Why we wrote this. The event estimate is easy to mistake for a mortality result or a class-wide finding. We separate the pooled event signal from mixed designs and underpowered death outcomes.

In this article (5 sections)
  1. Three studies, but only two randomized trials
  2. Heart-failure events produced the clearest pooled signal
  3. Symptoms and walking improved, but were not one pooled number
  4. The mortality analyses remain underpowered
  5. How much confidence should readers place in it?

A 2026 meta-analysis combined three 52-week studies of semaglutide in 1,463 people with obesity-related heart failure with preserved ejection fraction, or HFpEF. Adjudicated heart-failure events were less frequent with semaglutide than placebo, with a pooled odds ratio of 0.32. The review enrolled 0 new participants and generated 0 original patient outcomes[1].

The mortality estimates were much less certain. All-cause mortality had an odds ratio of 0.63 with a 95% confidence interval from 0.27 to 1.48. Cardiovascular mortality had an odds ratio of 0.27 with a confidence interval from 0.04 to 1.66. Both intervals crossed 1, so neither analysis demonstrated a mortality reduction[2].

Three studies, but only two randomized trials

The authors searched five databases, screened 276 records after duplicate removal and reviewed 29 full texts. They included STEP-HFpEF, STEP-HFpEF DM and a 318-person retrospective cohort, for 1,463 participants in total. Every study followed participants for 52 weeks. Two tested semaglutide against placebo under randomized protocols; the third used observational records[2].

Mixing those designs matters. Random assignment can balance known and unknown baseline factors on average. A retrospective comparison can adjust recorded differences but cannot remove unmeasured confounding or treatment-selection effects. The random-effects model acknowledges variation among studies, yet it does not turn the observational cohort into a randomized trial.

The populations were also narrower than the title may suggest. All three studies involved HFpEF with obesity, and STEP-HFpEF DM additionally required type 2 diabetes. All evaluated semaglutide rather than several different GLP-1 medicines. This result should not be generalized automatically to HFpEF without obesity, reduced-ejection-fraction heart failure or another medicine in the class[2].

Heart-failure events produced the clearest pooled signal

All three studies reported adjudicated heart-failure events, defined as hospitalization or an urgent emergency visit for heart failure. The pooled odds ratio was 0.32, with a 95% confidence interval from 0.15 to 0.67. Statistical heterogeneity was low at 5%[3]. This supports a lower event odds within the included evidence, not a universal individual forecast.

An odds ratio is not the same as an absolute risk reduction. The paper's pooled summary does not tell a reader how many events would be prevented among 100 similar people without the underlying event counts and baseline risk beside it. Follow-up lasted one year, so the estimate also cannot establish durability over several years.

The pooled event result is distinct from the symptom-centered STEP-HFpEF reports already covered elsewhere. It combines a specific adjudicated endpoint across three datasets. That focus also separates it from reviews comparing semaglutide and tirzepatide programs, because this analysis contained semaglutide evidence only.

Symptoms and walking improved, but were not one pooled number

The two randomized trials reported better Kansas City Cardiomyopathy Questionnaire clinical summary scores with semaglutide than placebo. Between-group differences were 7.8 points in STEP-HFpEF and 7.3 points in STEP-HFpEF DM. These were separate trial estimates, not one combined KCCQ effect[3].

Six-minute walk distance also improved in all three studies. Reported between-group changes were 20.3 meters and 14.3 meters in the two randomized trials, while the retrospective cohort reported 15.1 meters. Keeping the designs separate is essential. Similar direction does not make their methods interchangeable[2].

NT-proBNP, a heart-stress biomarker, declined significantly in both randomized trials but not in the retrospective cohort. Biomarker movement can support biological interpretation, but it is not equivalent to fewer symptoms, hospital care or death. The paper could not determine how much of any clinical improvement came from weight loss versus another cardiac or metabolic pathway[2].

The mortality analyses remain underpowered

Only two studies contributed mortality outcomes, and deaths were infrequent. For all-cause mortality, the pooled odds ratio was 0.63, but plausible values ranged from 0.27 to 1.48. For cardiovascular mortality, the estimate was 0.27 with an even wider range from 0.04 to 1.66[3]. These are inconclusive estimates, not proof of no effect and not evidence of a survival benefit.

One year is short for a mortality question in a chronic syndrome. A confidence interval can cross from a substantial reduction to possible harm when events are scarce. Longer and larger event-driven trials are needed before mortality can be estimated with useful precision.

How much confidence should readers place in it?

The review used a random-effects model and assessed the cohort with the Newcastle-Ottawa scale and the trials with the Cochrane risk-of-bias tool. The authors declared no competing interests. The full text does not report a dedicated funding statement[2]. Those details aid appraisal but do not erase the small evidence base.

Only three studies qualified. Design differences, the small number of deaths and 52-week follow-up limit precision. Low statistical heterogeneity for the event endpoint does not guarantee that the populations, methods or biases were identical. With three datasets, heterogeneity statistics themselves are not highly informative.

The practical conclusion is narrow. In three studies of people with obesity-related HFpEF, semaglutide evidence favored fewer adjudicated heart-failure events and better functional outcomes over 52 weeks. The analysis did not establish lower all-cause or cardiovascular mortality. It also cannot select treatment for an individual or replace the rest of guideline-directed heart-failure care.

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.

Frequently asked

Was this a new randomized trial?

No. It was a systematic review and meta-analysis of two randomized trials and one retrospective cohort, totaling 1,463 participants.

Did semaglutide reduce heart-failure events?

The pooled odds of adjudicated hospitalization or urgent heart-failure visits were lower versus placebo, with an odds ratio of 0.32. The evidence included mixed study designs.

Did the analysis show that semaglutide reduces mortality?

No statistically significant mortality reduction was demonstrated. Both all-cause and cardiovascular mortality estimates had wide confidence intervals that crossed 1.

Sources

  1. [1]Davis et al. GLP-1 receptor agonists versus placebo in HFpEF. PMID 43033336.Tier 1 · primary↩
  2. [2]PMC full text for the HFpEF systematic review and meta-analysis.Tier 1 · primary↩
  3. [3]Egyptian Heart Journal full text for the HFpEF meta-analysis.Tier 1 · primary↩

No revisions yet. First published .

About the editorial team

PeptideMethods is written and edited by the PeptideMethods Editorial Team and published by Digital Compass Group Ltd. The team is not made up of medical professionals; every health, regulatory or dosage claim on the site is tied to a primary source and is not a substitute for advice from a qualified clinician.

See our editorial policy and methodology for how we research, source and verify.

Read the pillars