GLP-1s and Sleep Apnea: What 5 Randomized Trials Found

On August 18, 2026, the journal Otolaryngology-Head and Neck Surgery, published on behalf of the American Academy of Otolaryngology-Head and Neck Surgery Foundation, released a pooled analysis of five randomized controlled trials covering 948 adults (479 assigned to a GLP-1 receptor agonist, 469 to a control group). Across those trials, GLP-1 receptor agonists reduced the apnea-hypopnea index by 12.4 events per hour more than controls (95% CI -17.2 to -7.7, P < .00001), and the authors report the effect held whether or not participants were using CPAP.

That is the headline number. The useful part sits underneath it.

What the apnea-hypopnea index actually counts

Obstructive sleep apnea is a mechanical problem before it is anything else. When you fall asleep, the muscles holding your throat open relax. In some people the airway narrows or closes outright. Breathing stops for a stretch (an apnea) or gets shallow enough to drop blood oxygen (a hypopnea). The brain partly rouses, muscle tone returns, the airway reopens, and the cycle starts again.

The apnea-hypopnea index, or AHI, counts those events per hour of sleep during a sleep study. Broadly, 5 to 15 events per hour is mild, 15 to 30 is moderate, and above 30 is severe.

Body fat enters that mechanism in two specific places. Fat in the neck and tongue presses inward on an airway that, once you are asleep, is held open by muscle tone alone. Abdominal fat reduces lung volume, and lower lung volume means less downward tension on the trachea to keep the upper airway stiff. Take weight off and you relieve both. That is why a metabolic drug ends up in a sleep journal: the GLP-1 is not doing anything to the airway itself, it is reducing the load the airway is carrying.

What the pooled analysis reported

Alongside the AHI result, the same five trials showed greater weight loss in the GLP-1 groups (12.5% more than controls, 95% CI -22.0% to -3.0%), waist circumference smaller by 3.3 cm, systolic blood pressure lower by 4.5 mmHg and diastolic lower by 1.30 mmHg. Gastrointestinal adverse events were more common with the GLP-1s. A separate analysis of the SURMOUNT trials found a similar pattern of weight and blood pressure moving together in the same person, not just as separate averages.

Look at the width of that weight-loss interval. The pooled trials agree that participants on a GLP-1 lost more weight than controls, and disagree considerably about how much. The diastolic blood pressure interval reaches almost to zero. Only two drugs were tested across the five trials, liraglutide and tirzepatide, so the pooled figure says nothing about other GLP-1 medications.

The two largest contributors

SURMOUNT-OSA, published in the New England Journal of Medicine in 2024, was two phase 3 double-blind randomized placebo-controlled trials in adults with moderate-to-severe obstructive sleep apnea and obesity. One enrolled people not using positive airway pressure, the other enrolled people already on it. Participants received the maximum tolerated dose of tirzepatide or placebo for 52 weeks. Mean baseline AHI was 51.5 and 49.5 events per hour. At week 52 the estimated treatment difference was 20.0 events per hour in the group not using PAP and 23.8 events per hour in the group already using it, both favoring tirzepatide. The trial was funded by Eli Lilly, which makes the drug.

SCALE Sleep Apnea, published in the International Journal of Obesity in 2016, randomized 359 non-diabetic adults with obesity and moderate or severe sleep apnea who were unwilling or unable to use CPAP. Over 32 weeks, liraglutide reduced AHI by an estimated 6.1 events per hour more than placebo (P = 0.0150) alongside 4.2% greater weight loss, with larger reductions in HbA1c and systolic blood pressure. Novo Nordisk employees were among the authors.

Where the regulators are

As of the label revision dated April 22, 2026, the FDA-approved label for Zepbound (tirzepatide) lists two indications, both in combination with a reduced-calorie diet and increased physical activity: reducing excess body weight and maintaining that reduction long term in adults with obesity or overweight plus a weight-related condition, and treating moderate to severe obstructive sleep apnea in adults with obesity.

The same label carries a boxed warning: in rodent studies, tirzepatide caused thyroid C-cell tumors, and whether that applies to humans has not been determined. It is contraindicated in people with a personal or family history of medullary thyroid carcinoma or MEN 2 syndrome. Listed adverse reactions include nausea, diarrhea, vomiting, constipation, abdominal pain, and injection site reactions, with acute pancreatitis, gallbladder disease, and acute kidney injury from dehydration among the more serious documented concerns.

The honest limits

"Independent of CPAP use" means the benefit showed up in both groups. It does not mean a medication substitutes for CPAP. No trial in this analysis randomized anyone to stop using it.

A few more things the data does not cover:

• Five trials and 948 people is a small evidence base for a condition this common, and it covers two drugs.
• AHI is a count of breathing events, not a measure of how you feel. SURMOUNT-OSA did also measure patient-reported sleep outcomes, but a lower index does not automatically translate into feeling rested.
• Follow-up ran 32 to 52 weeks. None of these trials followed participants after stopping the medication, so what happens to AHI on discontinuation was not tested.
• Both of the largest trials were run with funding or staffing from the manufacturers.
• These are group averages describing what happened to trial participants. They are not a forecast for any individual.

If you are weighing this

If you have diagnosed sleep apnea and obesity and you are curious whether a GLP-1 belongs in your plan, the conversation starts with numbers you may already have: your AHI from your most recent sleep study, your weight and waist, your blood pressure, and your thyroid and pancreatic history, which determine whether this class is an option for you at all. Bring those to a licensed provider who can look at them together.

If you currently use CPAP, keep using it as prescribed unless the provider managing your care tells you otherwise. Nothing in this analysis tested going without.