A retrospective cohort study posted as a preprint and indexed by Europe PMC reports that patients with peripheral artery disease and diabetes taking both a GLP-1 receptor agonist and an SGLT2 inhibitor had lower hazards of major adverse limb events, major amputation and death than patients on neither drug. The work has not yet been peer reviewed.

The two drug classes involved are GLP-1 receptor agonists, the family that includes semaglutide and tirzepatide, and SGLT2 inhibitors, which lower blood glucose by pushing it out through the kidneys. Current guidelines already recommend both for type 2 diabetes with atherosclerotic cardiovascular disease, but the authors say real-world evidence on their combined use in peripheral artery disease, specifically for limb outcomes, remains limited.

Key results

The cohort drew 13,471 patients with peripheral artery disease and diabetes from the Clalit Health Services electronic medical record database between 2015 and 2026, split into four groups: neither drug, GLP-1 alone, SGLT2 inhibitor alone, and both. Median follow-up was 5.75 years for limb events and 4.42 years for mortality.

Compared with neither treatment, combined therapy was associated with a hazard ratio of 0.50 (95% CI, 0.33 to 0.75) for major adverse limb events, 0.30 (95% CI, 0.17 to 0.53) for major amputation, and 0.23 (95% CI, 0.19 to 0.28) for all-cause mortality. SGLT2 inhibitor monotherapy was associated with reductions across the same three outcomes.

Why the limb numbers matter

Amputation is the outcome this population actually fears, and it is the one most cardiovascular trials are not powered to answer. A study that reports limb-specific hazards alongside mortality in a cohort this size is doing something the field has mostly left to secondary analyses.

GLP-1 monotherapy is the weaker arm here. It was associated with lower mortality, but the authors say its limb-specific associations should be interpreted with caution given the smaller subgroup and event numbers.

Limitations

This is a retrospective observational study, not a randomized trial, so the findings are associations rather than demonstrated effects. The authors used multinomial propensity scores and stabilized inverse probability of treatment weighting to balance baseline characteristics, which addresses measured confounding but not unmeasured confounding.

The analysis also draws on a single national health service database, and the preprint has not completed peer review. Limb estimates in the GLP-1 monotherapy group rest on small subgroup and event counts.

Bottom line

The authors conclude that among patients with peripheral artery disease and diabetes, combined GLP-1 and SGLT2 inhibitor treatment was associated with lower limb events and mortality, with SGLT2 inhibitors showing the same trends. Whether that association survives a randomized test, and whether GLP-1 agonists contribute anything to the limb side of it, is still open.