Semaglutide and Tirzepatide Show Real-World Promise in Cutting Heart Failure Hospitalizations and Deaths in Diabetic HFpEF Patients
Research Articles
Semaglutide and Tirzepatide Show Real-World Promise in Cutting Heart Failure Hospitalizations and Deaths in Diabetic HFpEF Patients
Research Articles

Semaglutide and Tirzepatide Show Real-World Promise in Cutting Heart Failure Hospitalizations and Deaths in Diabetic HFpEF Patients

New Real-World Evidence Links GLP-1 / GIP Therapy With Lower Risk of HF Hospitalization and Mortality in HFpEF Patients With Diabetes

Boston / Munich, August 31, 2025 — In a large observational analysis published online today in JAMA, researchers report that initiation of semaglutide or tirzepatide in patients with obesity‐associated heart failure with preserved ejection fraction (HFpEF) and type 2 diabetes was associated with a notably lower risk of hospitalization for heart failure or death over the first year, compared with sitagliptin. However, in a head-to-head comparison, tirzepatide did not show a statistically meaningful advantage over semaglutide.

Background & Rationale

HFpEF (heart failure with preserved ejection fraction) has become the dominant phenotype in heart failure, especially in populations burdened by obesity, diabetes, and metabolic dysfunction. Despite its prevalence, effective therapies that reduce hard outcomes (hospitalization or death) remain limited.

Recently, randomized trials have suggested that semaglutide (a GLP-1 receptor agonist) and tirzepatide (a dual GIP/GLP-1 agonist) may improve symptoms and quality of life in HFpEF populations, but these trials were small in size and had limited cardiovascular event data, making it difficult to draw definitive conclusions about mortality or hospitalizations. As such, the JAMA authors aimed to generate real-world evidence about the effectiveness and safety of these drugs in broader patient populations.

To do so, they used U.S. claims data to emulate the trial conditions (benchmarking against STEP-HFpEF DM and SUMMIT) and then expanded eligibility criteria to reflect typical clinical practice.

Study Design & Methods

  • The analysis used three large U.S. claims databases (Medicare, Optum Clinformatics, MarketScan) spanning 2018 to 2024.
  • Five active‐comparator, new‐user cohort studies were constructed:
    1. Semaglutide vs sitagliptin (benchmarking to STEP-HFpEF DM),
    2. Tirzepatide vs sitagliptin (benchmarking to SUMMIT),
    3. Expanded versions of the above two, and
    4. A direct head-to-head of tirzepatide vs semaglutide.
  • The primary end point was a composite of hospitalization for heart failure or all-cause mortality over up to 52 weeks of follow-up.
  • Propensity score overlap weighting was used to balance baseline covariates (demographics, comorbidities, medication use, health care utilization, etc.) between comparison groups.
  • The authors also used negative control outcomes (lumbar radiculopathy, abdominal hernia) to assess residual confounding.
  • Subgroup and sensitivity analyses were prespecified (by age, sex, BMI thresholds, alternate definitions of outcomes, etc.).

Benchmarking of the trial‐emulated cohorts showed reasonable agreement in effect estimates versus the original randomized trials, bolstering confidence in the analytic approach.

Major Findings

Semaglutide vs Sitagliptin (Expanded Population)

  • In the expanded eligibility cohort (reflecting real-world patients), pooled 1-year risk of the composite end point was 5.5% for semaglutide vs 8.6% for sitagliptin, yielding an absolute risk difference of –3.2% (95% CI, –4.1% to –2.3%).
  • The hazard ratio (HR) for semaglutide vs sitagliptin was 0.58 (95% CI, 0.51–0.65), corresponding to ~42% relative risk reduction.
  • The number needed to treat (NNT) over one year was approximately 31.

Tirzepatide vs Sitagliptin (Expanded Population)

  • The pooled 1-year risk was 3.6% with tirzepatide vs 7.5% with sitagliptin; absolute risk difference –3.9% (95% CI, –5.7% to –2.1%), NNT ~26.
  • The HR for tirzepatide vs sitagliptin was 0.42 (95% CI, 0.31–0.57), indicating ~58% relative risk reduction.

Tirzepatide vs Semaglutide (Head-to-Head)

  • Direct comparison yielded an HR of 0.86 (95% CI, 0.70–1.06), a non-statistically significant difference, suggesting no meaningful superiority of tirzepatide over semaglutide in this setting.
  • Secondary endpoints, subgroup analyses, and sensitivity checks generally supported the consistency of the results across populations, and no large safety signals emerged on the evaluated endpoints (gastrointestinal, infections, urinary tract events).

Interpretation & Implications

These findings suggest that, among patients with obesity-related HFpEF and type 2 diabetes in clinical practice, initiating semaglutide or tirzepatide may substantially reduce the risk of HF hospitalization or death compared with a comparator glucose-lowering drug (sitagliptin) that is relatively neutral on heart failure risk. The magnitude of relative risk reduction exceeded 40%.

The lack of a clear advantage of tirzepatide over semaglutide in the head-to-head analysis suggests that the incremental benefits of the dual GIP/GLP-1 mechanism may not translate into major differences in hard outcomes—at least within the one-year follow-up observed.

Because this is an observational (non-randomized) study, residual confounding cannot be excluded, despite robust propensity weighting and negative control analyses. The authors themselves emphasize that these data should serve as complementary evidence to ongoing and future randomized trials, rather than as conclusive proof.

If borne out in prospective trials, these results might support a role for GLP-1 / GIP therapies in the therapeutic arsenal for HFpEF patients with metabolic disease—expanding the paradigm from purely symptomatic benefit toward prevention of hard cardiovascular events.

Strengths & Limitations

Strengths:

  • Use of large, real-world data across multiple insurance databases increases generalizability.
  • Emulation of randomized trial design (STEP-HFpEF DM, SUMMIT) offered internal validation (benchmarking) of the approach.
  • Extensive adjustment for confounders, and use of negative control outcomes, reinforce the credibility of the associations.
  • Preplanned subgroup and sensitivity analyses help assess robustness.

Limitations:

  • Residual confounding and bias (confounding by indication, measurement error) remain possible in claims-based observational data.
  • Relatively short on-treatment follow-up (median ~4–5 months) for many patients, due largely to drug discontinuation or switching.
  • The study population was limited to U.S. insured individuals, possibly limiting generalizability to other health systems or populations (e.g. low/middle income settings).
  • Outcomes (e.g. hospitalization for HF) were identified via claims algorithms, which may misclassify events, though validations suggest high positive predictive value.
  • The study design cannot replace randomized controlled trials; causality cannot be definitively established.

Future Directions

  • Randomized clinical trials testing semaglutide, tirzepatide, and similar agents specifically in HFpEF populations with cardiometabolic disease remain essential.
  • Longer follow-up is needed to assess durability of benefit, long-term safety, and effects on quality of life, functional capacity, and other cardiovascular outcomes (e.g. stroke, MI).
  • Subgroup analyses by degree of obesity, baseline LVEF, kidney function, and ethnicity may help tailor patient selection.
  • Real-world implementation studies will be needed to understand adherence, cost-effectiveness, and integration into standard heart failure care pathways.