Healio: GLP‑1s show no added DKA risk

GLP-1 drugs are now common enough that people with type 1 diabetes keep asking the obvious question: can these drugs help with weight and glucose control without pushing them into diabetic ketoacidosis? That fear has hung over the whole category for years. Now a real-world study presented at the AACE annual meeting in Las Vegas adds one more piece of reassurance — at least in one center’s experience, GLP-1 use in adults with type 1 diabetes did not come with more DKA or pancreatitis, and hospitalizations were actually lower. ### Why is this even controversial? Type 1 diabetes is the hard version of the GLP-1 story because these patients depend on insulin full stop. GLP-1 drugs suppress appetite and can reduce insulin needs, which sounds useful, but the catch is that insulin cuts that go too far can let ketones rise and trigger DKA. That is why these drugs are still not FDA-approved for type 1 diabetes, even as off-label use has grown. ### What did the new study actually look at? This was a retrospective, single-center cohort of 7,377 adults with type 1 diabetes. Of those, 255 had used a GLP-1 receptor agonist. The most common drugs were semaglutide in 167 patients and tirzepatide in 60. Researchers compared hospital admissions and safety events over one year between GLP-1 users and non-users. ### What happened to DKA and pancreatitis? In the GLP-1 group, there were no hospital admissions for DKA and no admissions for pancreatitis during the study window. In the non-user group, those events were rare but not zero — DKA admissions were 0.39% and pancreatitis admissions were 0.55%. Statistically, the study did not show a higher risk from GLP-1 treatment. ### Was there any upside beyond “not worse”? Yes — but this is where you have to stay careful. Overall hospitalization was lower in GLP-1 users, 7.45% versus 13.11% in non-users, with a reported P value of.01. That sounds impressive. But it does not prove the drug caused the difference, because the people prescribed GLP-1s may have been healthier, more closely followed, or selected in ways the dataset cannot fully capture. ### Why doesn’t this settle the question? Rare bad outcomes are hard to study when only 255 people were exposed. Zero events can mean “safe,” but it can also mean “too small to detect an uncommon problem.” This analysis was also retrospective, based on one center, and relied on coded hospital outcomes rather than a tightly controlled trial with dosing, adherence, and insulin-adjustment details. ### What about the pancreatitis worry? That concern has followed GLP-1 drugs for a long time, but broader evidence outside type 1 diabetes has been mixed and increasingly less alarming than early signals suggested. The new study does not erase the question, but it fits with a larger drift in the field: clinicians are becoming less convinced that pancreatitis is a class-wide dealbreaker. ### So what should clinicians take from this? Basically, this is a reassurance signal, not a green light. It suggests that carefully selected adults with type 1 diabetes did not show an obvious safety penalty from GLP-1 therapy in this cohort. But because the drugs remain off-label in type 1 diabetes, the real issue is still patient selection, insulin education, and watching for ketones when appetite drops fast. ### Bottom line? The news is not that GLP-1s are suddenly standard treatment for type 1 diabetes. The news is narrower, but still useful: one 2026 real-world cohort did not find the DKA or pancreatitis signal many clinicians feared. That makes the conversation less about blanket avoidance and more about who might benefit, who needs close monitoring, and what a proper prospective trial still has to prove.