Can a GLP-1 cause low blood sugar if you do not have diabetes?
Reviewed 13 sourcesPublished by PepVault
Rarely. In weight-loss trials, readings under 54 mg/dL hit 0.3% on Zepbound and 0.6% on Foundayo, none on placebo. When the risk rises and what to do about it.

- Last reviewed
- Sources
- 13
- Questions
- 6
- Compounds
- 5
The short answer
Rarely, but it can. In the weight-loss trials of adults without diabetes, a reading under 54 mg/dL showed up in 0.3% on Zepbound and 0.6% on Foundayo, against none on placebo. These drugs push insulin out only while blood sugar is normal or high. Without diabetes, lows cluster around days of barely eating or vomiting, past bariatric surgery, alcohol or too large a dose.
Evidence: FDA prescribing information for Wegovy, Zepbound, Saxenda, Foundayo, Trulicity, Ozempic and Mounjaro (DailyMed, versions current in November 2026) with their placebo-controlled trial data; two small human clamp studies (2002, 2018), one small human study after gastric bypass (2014) and two case reports (2024, 2026); NIDDK's patient guidance on treating a low.

How often do GLP-1s cause low blood sugar without diabetes?
Seldom, on every label that reports it. Low blood sugar (hypoglycemia) means glucose under 70 mg/dL; the Wegovy label calls a reading under 54 mg/dL, about 3.0 mmol/L, clinically significant113. Here is what the weight-loss labels report for adults and teens without type 2 diabetes, each against its own placebo (a dummy injection or pill), as of November 2026:
| Drug and trial | What was counted | Placebo | On the drug |
|---|---|---|---|
| Zepbound (tirzepatide), Study 1, adults, 72 weeks | Glucose under 54 mg/dL | 0% | 0.3%2 |
| Foundayo (orforglipron, a daily pill), Trial 1, adults | Glucose under 54 mg/dL | 0% | 0.6%; no severe lows4 |
| Saxenda (liraglutide), adult trials | Symptoms reported as a low, not confirmed by a reading | 1.1% (19 of 1,729) | 1.6% (46 of 2,962)3 |
| Saxenda, adult trials | Fasting glucose under 54 mg/dL at a clinic visit | 0.1% | 0.1%3 |
| Wegovy (semaglutide), SELECT heart trial, about 3 years | Serious lows | 1 episode in 8,801 people | 3 episodes in 8,803 people1 |
| Saxenda, teens aged 12 to 17, 56 weeks, meters handed out | Under 70 mg/dL with symptoms | 4% | 15.2%3 |
| Saxenda, the same teens | Under 54 mg/dL | 0.8% | 1.6%; no severe lows3 |
Read the adult rows as a floor. The Wegovy, Zepbound, Foundayo and Saxenda labels all say their adult trials without diabetes did not systematically capture lows1234; the Saxenda label adds that nobody was given a meter or a diary to look for them3. The one trial in these labels that did hand out meters, Saxenda's in teenagers, found mild lows nearly four times as often as placebo, while readings under 54 mg/dL stayed under 2% and nobody needed help from another person3. The pattern across all of them: mild dips are more common than the adult tables show, and serious lows are rare.
In PepVault, a shaky spell goes in the side-effect log with its date, a severity from minimal to very severe and a note for the reading, filed under the compound you take. Log it in PepVault

Why do GLP-1s rarely push blood sugar too low on their own?
Because their push on insulin switches off as sugar falls. A GLP-1 drug copies GLP-1, a gut hormone that asks the pancreas for insulin only while glucose is normal or high (our guide on whether Ozempic is insulin covers that part). Two experiments show where the switch sits.
In 2002, nine healthy volunteers had their blood sugar stepped down on purpose with an insulin drip, once with a GLP-1 drip and once with salt water. GLP-1's extra insulin was negligible below 4.3 mmol/L, about 77 mg/dL. It lowered glucagon, the hormone that raises sugar, only at normal sugar; at 3.7 mmol/L (about 67 mg/dL) and below, glucagon climbed just as it did without GLP-1. The other alarm hormones and the warning symptoms rose normally too; only the growth hormone response was smaller8.
In 2018, 38 people with type 2 diabetes took weekly semaglutide or placebo for 12 weeks each before the same kind of test. Their glucagon response to a low was the same on both, and the drip of sugar needed to hold the low steady was too, though noradrenaline and cortisol rose a little less on semaglutide. Symptom scores at the lowest point were lower on semaglutide, though, and fewer people recognized the low9. That is a reason to trust a reading over a feeling.
Why does the risk rise with insulin or a sulfonylurea?
Insulin lowers sugar whatever the level, and a sulfonylurea (glipizide, glimepiride, glyburide: older diabetes pills) squeezes insulin out of the pancreas whatever the level too. The GLP-1's brake no longer protects you, and every label warns that the pairing raises the risk of lows, severe ones included1235. Taken alone in diabetes trials, the serious lows vanish: Mounjaro had no reading under 54 mg/dL against 1% on placebo7, and Ozempic had no severe or confirmed low at or under 56 mg/dL against 1.6% on placebo, though 1.6% and 3.8% on its two doses had a symptomatic reading at or under 70 mg/dL, against none on placebo6. Trulicity, the dulaglutide pen, is approved only for diabetes, and its label shows the gradient clearly:
| Trulicity (dulaglutide) taken with | Weeks | Placebo | Trulicity, under 54 mg/dL |
|---|---|---|---|
| Metformin | 26 | 0% | 0.3% at 0.75 mg, 0.7% at 1.5 mg5 |
| Glimepiride, a sulfonylurea | 24 | 0% | 3.3% at 1.5 mg5 |
| Basal insulin (glargine) | 28 | 9.3% | 14.7% at 1.5 mg5 |
| Mealtime insulin | 52 | Not reported | 77% at 0.75 mg, 69% at 1.5 mg5 |
The weight-loss labels show the same split in their diabetes trials. On Zepbound, 10.3% of people also taking a sulfonylurea had a low under 54 mg/dL, against 2.1% of those not taking one2. On Foundayo the figures were 7% and 0.5%4. In the Saxenda diabetes trial, 28.2% of people on a sulfonylurea had one, against 12.7% on placebo, even though their sulfonylurea dose had been halved at the start3. The labels leave the adjustment to the prescriber, who may lower the insulin or sulfonylurea when the GLP-1 starts125.
What makes a low more likely if you don't have diabetes?
- Days of barely eating or vomiting. Case reports describe lows with ketoacidosis (acid build-up from burning fat for fuel) in people without diabetes on tirzepatide: four female patients aged 17 to 34 in Kuwait with vomiting or diarrhea, three of them around week 5 or right after a step-up to 5 mg, with a median glucose under 3.89 mmol/L (70 mg/dL)11; and a 21-year-old woman in Japan on a low-carb diet whose nausea and vomiting after her fifth injection cut her eating sharply, with a glucose of 68 mg/dL. Her acid build-up cleared within 12 hours on a glucose drip12. Case reports are the weakest kind of evidence: they show it can happen, not how often.
- Past bariatric surgery. The Wegovy label names it a risk factor: in SELECT, 2.3% (2 of 87) of people with that history had a serious low on Wegovy, against none of 97 on placebo1. After a gastric bypass the gut already releases surges of GLP-1 after meals, and blocking the GLP-1 receptor corrected post-meal lows in all nine patients with them in one study10. Our guide to a GLP-1 after bariatric surgery covers the rest.
- Alcohol. Our guide to alcohol on a GLP-1 covers how drinking adds to the risk of a low.
- Too large a dose. The Wegovy and Saxenda labels list severe low blood sugar among the effects of overdoses with GLP-1 drugs13.
Is feeling shaky on a GLP-1 a sign of low blood sugar?
Sometimes. The Zepbound label's signs of a low are dizziness or light-headedness, sweating, confusion or drowsiness, headache, blurred vision, slurred speech, shakiness, a fast heartbeat, anxiety or irritability, hunger, weakness and feeling jittery2. Several of those also come with eating far less and with dehydration from nausea, so the feeling alone does not settle it. A finger-stick meter or a glucose sensor does: under 70 mg/dL with symptoms is a low, under 54 mg/dL is a clinically significant one113.
No label gives the week in which lows start, because the trials without diabetes did not look for them124. The case reports above cluster in the first weeks and around a step-up, the same stretch in which the Zepbound label places most nausea and vomiting21112.
Commonly reported: shaky, sweaty or light-headed spells on a day of eating very little, most often in the first weeks or just after a dose step-up, easing once the person eats. On our reports page, 2 of the 77 tirzepatide write-ups name low blood sugar, and some of those posters have diabetes.
PepVault's food log sets one daily target, protein first, so a day when appetite barely lets you eat shows up on the plate instead of only in how you feel. Track your food in PepVault
What do you do if your blood sugar runs low?
The NIH's diabetes institute (NIDDK) gives the standard steps for a reading under 70 mg/dL, or symptoms with a reading below your target13:
- Eat or drink 15 to 20 grams of fast sugar right away: four glucose tablets, one tube of glucose gel, half a cup (4 ounces) of regular fruit juice, 4 to 6 ounces of regular soda, or a tablespoon of sugar or honey13.
- Wait 15 minutes, then check again13.
- Still low? Take another 15 to 20 grams and check after another 15 minutes, until you are back in range13.
- If your next meal is more than an hour away, have a snack such as crackers or a piece of fruit13.
- Write down the time, the reading and what you had eaten that day, and tell your prescriber; whether a dose changes is their decision.
When to get help A low with confusion, fainting or a seizure, or one you cannot treat because you cannot eat or keep food down, needs emergency care right away13.
Sources
- WEGOVY (semaglutide) injection and tablets prescribing information, Novo Nordisk, DailyMed (SPL version 19), June 2026 dailymed.nlm.nih.gov
- ZEPBOUND (tirzepatide) injection prescribing information, Eli Lilly, DailyMed (SPL version 40), August 2026 dailymed.nlm.nih.gov
- SAXENDA (liraglutide) injection prescribing information, Novo Nordisk, DailyMed (SPL version 22), 2026 dailymed.nlm.nih.gov
- FOUNDAYO (orforglipron) tablets prescribing information, Eli Lilly, DailyMed (SPL version 10), July 2026 dailymed.nlm.nih.gov
- TRULICITY (dulaglutide) injection prescribing information, Eli Lilly, DailyMed (SPL version 62), June 2026 dailymed.nlm.nih.gov
- OZEMPIC (semaglutide) injection prescribing information, Novo Nordisk, DailyMed (SPL version 20), June 2026 dailymed.nlm.nih.gov
- MOUNJARO (tirzepatide) injection prescribing information, Eli Lilly, DailyMed (SPL version 40), August 2026 dailymed.nlm.nih.gov
- Nauck MA et al. Effects of glucagon-like peptide 1 on counterregulatory hormone responses, cognitive functions, and insulin secretion during hyperinsulinemic, stepped hypoglycemic clamp experiments in healthy volunteers. J Clin Endocrinol Metab. 2002;87(3):1239-46. PubMed 11889194
- Korsatko S et al. Effect of once-weekly semaglutide on the counterregulatory response to hypoglycaemia in people with type 2 diabetes: a randomized, placebo-controlled, double-blind, crossover trial. Diabetes Obes Metab. 2018;20(11):2565-2573. PubMed 29893488
- Salehi M, Gastaldelli A, D'Alessio DA. Blockade of glucagon-like peptide 1 receptor corrects postprandial hypoglycemia after gastric bypass. Gastroenterology. 2014;146(3):669-680. PubMed 24315990
- Bitar Z et al. Exploring hypoglycemic ketoacidosis in nondiabetic patients on tirzepatide: is starvation the culprit? Am J Case Rep. 2024;25:e946133. PubMed 39686534
- Minoda Y et al. Starvation ketoacidosis associated with tirzepatide use for weight loss in a non-diabetic East Asian woman: a case report. Int J Emerg Med. 2026;19(1):119. PubMed 41840480
- National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). Low blood glucose (hypoglycemia). Last reviewed July 2021. niddk.nih.gov
Newsletter · Weekly
The weekly peptide briefing.
New guides, the week’s peptide news, one tool. Free, one email a week, unsubscribe anytime.