Adults with type 2 diabetes and peripheral artery disease, the plaque buildup that narrows the arteries feeding the legs, were about half as likely to lose part of a limb to amputation over five years if they took a GLP-1 receptor agonist instead of metformin. That is the headline number from a retrospective study of more than 4,200 patients published July 1 in the Journal of the American Heart Association ↗.
The same comparison found lower rates of death (10.3 percent versus 14.5 percent, a hazard ratio of 0.74, meaning roughly a quarter fewer deaths), fewer procedures to reopen blocked arteries (revascularization, down 36 percent), and fewer hospital stays (down 13 percent). Major amputation, removal above the ankle, fell from 4.4 percent to 2.3 percent, a hazard ratio of 0.52. Minor amputations dropped by a similar margin.
Here is the part the drug's reputation would not predict. The events GLP-1 drugs are best known for preventing, the big cardiovascular ones, heart attack and stroke, plus serious kidney events, were no different between the two groups. In this population, against this comparator, the benefit landed on the legs and on staying alive, not on the ledger that cardiovascular outcome trials like LEADER, SELECT, and the kidney trial FLOW were built to move.
The design decides how much weight to put on it. This was not a randomized trial. The researchers pulled electronic health records from the TriNetX network and matched 2,133 GLP-1 users against 2,133 metformin users on measured characteristics, a method that balances what is recorded and cannot balance what is not. Metformin is an active comparator, not a placebo, so the numbers describe one reasonable diabetes drug against another, not a drug against nothing. People who start a GLP-1 drug can differ from metformin starters in ways a database never sees, and a five-year gap of four percentage points in death is exactly the kind of number that unmeasured differences can manufacture.
Peripheral artery disease affects more than 236 million people worldwide, and diabetes is one of its biggest drivers. Amputation is the outcome patients fear most and the one that reshapes a life around it. A signal that a widely prescribed class might cut major amputations by half, even a retrospective one, is the kind of finding that earns a randomized trial rather than settling the question. Limb outcomes have been an afterthought in the GLP-1 cardiovascular program, measured in passing rather than as the point.
The drugs in the study act on the GLP-1 receptor ↗, the same fullness-signaling switch that semaglutide ↗, tirzepatide, and liraglutide all hit. Most of the GLP-1 outcome story, from the Novo Nordisk and Eli Lilly programs down, has been written about the heart and the kidney. This one points somewhere the class effect has barely been mapped, and where the stakes are a leg.