By itself, rarely. A GLP-1 drives insulin release in proportion to how high glucose already is, so the mechanism largely switches itself off once blood sugar is normal. That is why hypoglycemia sits low on the adverse event lists for the class. The risk arrives from what is taken alongside it, and from one surgical complication that works the other way round entirely.
Where the risk actually comes from
Insulin and sulfonylureas lower glucose whether or not it needs lowering, and a GLP-1 added on top of either of them can turn a previously correct dose into an excessive one. The adjustment belongs to whoever manages the insulin, not to the person filling a weight-loss prescription. Which other medications change when one of these is added is set out in what a GLP-1 replaces.
The size of that adjustment is visible in the trial data. A phase 3 study added cagrilintide-semaglutide to basal insulin in 274 adults with type 2 diabetes whose mean baseline HbA1c was 8.8% [3]. HbA1c fell by 2.33 and 2.10 percentage points at the two dose levels, against 0.66 on dose-matched placebo at week 40. A drop of that size in somebody already injecting insulin is the kind that changes what the insulin dose should be.
A separate 26-week trial randomized 72 adults with type 1 diabetes and a BMI of 30 or higher to semaglutide up to 1 mg or placebo. Every participant was already on an automated insulin delivery system [2]. Two severe hypoglycemia events occurred in each group, and no diabetic ketoacidosis was reported in either. Seventy-two people over 26 weeks cannot rule out an event that happens once in a few hundred patient-years, so neither figure is a clearance.
The version that has nothing to do with the drug
The severe form of this problem turns up in people who have had bariatric or other upper gastrointestinal surgery, and it is caused by their own hormone rather than by any prescription. After a gastric bypass, food reaches the small intestine faster than it used to and the gut releases far more GLP-1 than it should. That drives an insulin surge an hour or two after eating, and a crash behind it [1].
It complicates up to 30% of Roux-en-Y procedures and up to 10% of sleeve gastrectomies, and the same picture follows total gastrectomy, esophagectomy and Nissen fundoplication. People describe confusion, sweating and blackouts after ordinary meals. Nobody in that position is a candidate for a GLP-1 agonist. It is the mirror image of the regain that sends people back to these drugs after surgery, described in the order in which the two are used.
What blocking the receptor does
An investigational drug called avexitide blocks the receptor these drugs activate, and a phase 2b crossover study tested it in that surgical population. Sixteen people completed it, each spending fourteen days on 45 mg twice daily and fourteen on 90 mg once daily after a fourteen day baseline. Daytime severe lows measured by continuous glucose monitoring fell 47.7% and 59.0%, centrally adjudicated severe events fell 67.5% and 66.1%, and time below 54 mg/dL fell 45% and 64%.
What to do with a low reading
Somebody taking a GLP-1 alone, with no insulin and no sulfonylurea, who is getting repeated low readings has a reason to ask what else is going on. Assuming the drug explains it is the weaker inference. Reduced intake, alcohol, and the surgical mechanism above are all commoner explanations than the agonist itself. Which laboratory values are worth watching during treatment is covered in the monitoring nobody quotes you a price for.
Somebody on insulin is in a different position and should have had the dose reviewed before starting, not after the first low. What a GLP-1 takes over from an insulin regimen is set out in what swapping mealtime insulin actually buys. The equivalent question for the other major diabetes class is in taking a GLP-1 alongside an SGLT2 inhibitor.