Diabetics on GLP-1 drugs were slightly less likely to end up needing surgery for a trigger finger. The gap did not show up at all until a year had passed.
Trigger finger is one of the most common hand complaints in people with type 2 diabetes. The tendon that bends the finger catches in its sheath, so the finger locks and snaps painfully when it moves. The first-line fix is a corticosteroid injection. When that fails, surgeons cut the A1 pulley, the ring of tissue the tendon keeps snagging on, to free it. Diabetics are both more likely to get trigger finger and more likely to need that operation.
A retrospective study published September 1 in Hand ↗ asked whether GLP-1 drugs change that path. Using the TriNetX research network, the authors pulled 69,222 adults with type 2 diabetes and trigger finger who got an initial steroid injection, split them by whether they were also taking a GLP-1 receptor agonist, and matched 5,863 in each group on age, other conditions, and other diabetes medications. These are drugs like Ozempic (semaglutide ↗) and Mounjaro (tirzepatide ↗) that act on the GLP-1 receptor ↗ to cut appetite and drive weight loss.
At six months there was no difference: 6.1 percent of GLP-1 users had gone to surgery versus 6.8 percent of non-users. The gap opened later. At twelve months, 10.6 percent of GLP-1 users had needed the operation against 11.9 percent of non-users (relative risk 0.89, an 11 percent lower rate). At twenty-four months it was 14.2 versus 15.7 percent (relative risk 0.90). Both later differences cleared statistical significance.
Read the absolute size before reading anything into it. The two-year gap is 1.5 percentage points. The authors put it in plain surgical terms: you would have to treat 59 patients with a GLP-1 drug to prevent one A1 pulley release at two years, and 71 to prevent one at a year. This is a small effect that a large database can detect precisely, not a large effect that changes how any single patient is managed.
The delayed pattern is the interesting part, and the trap. Nothing at six months and a widening gap after that is what you would expect from a slow tissue effect. High blood sugar stiffens and thickens connective tissue by cross-linking collagen, and weight and inflammation load the tendon sheath. A drug that lowers glucose and sheds weight could plausibly ease that over a year in a way it cannot in a few weeks. But the same time course is also what confounding looks like: people who stay on a GLP-1 drug for two years are, on average, healthier and more engaged with their care than those who do not, and that alone bends outcomes their way. This is a matched database, not a randomized trial, and the study is graded Level III evidence. It can show the association. It cannot show the drug caused it.
The finding also does not stand alone. We covered a companion signal earlier this year, when tirzepatide users developed less carpal tunnel syndrome, with weight loss the likely explanation ↗. Two of the most common hand problems in diabetes, trigger finger and carpal tunnel, both nudging down among GLP-1 users, both by small margins, both in databases that cannot rule out that the healthier patients were simply the ones on the drug. It points somewhere worth a real trial. It does not close the question, and the honest version says so.