Adults with diabetic nerve damage who took a GLP-1 drug had fewer foot ulcers than those on an older diabetes pill, but nearly twice the rate of a rarer complication that quietly destroys the bones of the foot. Both signals come from the same records study published September 7 in the Journal of Neurology ↗.
The comparison is narrower than the usual GLP-1 story. The patients all had type 2 diabetes and already had diabetic neuropathy, the nerve damage that strips feeling from the feet and legs. And the drug they were measured against was not a placebo but another diabetes medicine, a DPP-4 inhibitor. Semaglutide ↗ and tirzepatide ↗, the drugs behind Ozempic, Wegovy, Mounjaro, and Zepbound, flood the GLP-1 receptor ↗ directly; tirzepatide also hits the GIP receptor ↗. DPP-4 inhibitors like sitagliptin work more gently, blocking the enzyme that breaks down the body's own GLP-1 so a little more of it lingers. That enzyme is dipeptidyl peptidase-4, the "DPP-4" in the name. So this was not GLP-1 against nothing. It was a strong dose of GLP-1 signaling against a mild one.
What the records showed
The researchers used TriNetX, a network of electronic health records, to find adults with type 2 diabetes and neuropathy who started either drug class. After matching each GLP-1 user against a DPP-4 user with a similar profile of age, sex, other conditions, and medications, they had 19,770 people in each group and followed them for one and two years.
On the most common problem, diabetic foot ulcers, the GLP-1 group did better: 2.2 percent developed one in the first year against 2.7 percent on the DPP-4 drug, about a fifth lower in relative terms (hazard ratio 0.81). The GLP-1 group also died less often over the window, though the authors flagged that mortality result as a hypothesis to test rather than a conclusion.
Two of the worst outcomes showed no difference at all. Amputations happened to 0.5 percent of each group. Bone infection in the foot or ankle, called osteomyelitis, hit about 0.4 percent either way. Neither gap was statistically meaningful.
Then the number that runs the other direction. Charcot neuroarthropathy, a condition where a numb foot fractures and collapses without the person feeling it because the warning of pain is gone, was almost twice as common in the GLP-1 group: 0.3 percent versus 0.2 percent, a hazard ratio of 1.99. In a foot that cannot register damage, the bones can crumble under ordinary walking and the joint deforms before anyone notices.
Reading a split verdict
A drug that beats its comparator on one outcome, ties on two, and loses on a fourth is harder to explain away than a drug that wins on everything. When a records study shows an across-the-board benefit, the usual culprit is the patients, not the pill. Healthier, better-resourced people tend to get the newer, pricier injectable, and that alone manufactures a glow on every measure. This result does not glow evenly. The ulcer benefit is real but small, half a percentage point. The amputation and osteomyelitis lines sit flat. And the Charcot signal points the wrong way.
That last number needs its caveats stated plainly. The absolute jump is from roughly two in a thousand to three in a thousand, so the near-doubling rests on a small count of extra cases. Charcot is also notoriously underdiagnosed and inconsistently coded. A patient losing weight and moving more on a GLP-1 drug may simply walk enough to stress a vulnerable foot into a diagnosis that a more sedentary DPP-4 patient never receives. The study is retrospective and matched rather than randomized, which means it can flag an association and cannot prove the drug caused either the benefit or the harm.
The read the authors land on is foot monitoring, not a verdict on the drug class. Earlier this year a snapshot study found GLP-1 users had less painful diabetic nerve damage ↗, a hint the drugs might do the nerve some good. This one looks past the nerve to the foot it protects, and finds the ledger mixed: fewer ulcers, a possible cost in collapsed joints, and a clear reason to keep checking the feet of anyone on these drugs who has already lost the feeling in them.