Skip to content
GLP Loss
← Research
Evidence

GLP-1 and Sleep Apnea: What SURMOUNT-OSA Measured

Tirzepatide lowered the apnea-hypopnea index by 20.0 events per hour more than placebo in one trial and 23.8 in the other, from a baseline near 50 events per hour.

Tessa Whitfield8 min read
SURMOUNT-OSA: apnea-hypopnea index at 52 weeksMean fall in events per hour, tirzepatide against placeboTrial 1, tirzepatide25.3Trial 1, placebo5.3Trial 2, tirzepatide29.3Trial 2, placebo5.5Treatment difference: 20.0 events per hour in trial 1.Baseline mean AHI was 51.5 in trial 1 and 49.5 in trial 2.Mean body-mass index was 39.1 and 38.7 at entry.

Obstructive sleep apnea and obesity travel together, and tirzepatide now carries a regulatory indication for moderate-to-severe apnea in adults with obesity. That makes it the first drug licensed to treat the breathing disorder itself rather than the sleepiness it causes. The trial behind the indication is worth reading before the marketing built on it, including the marketing that surrounds the tirzepatide sellers.

What SURMOUNT-OSA ran

Two phase 3 double-blind randomized trials enrolled adults with moderate-to-severe obstructive sleep apnea and obesity. Trial 1 took participants who were not using positive airway pressure at baseline. Trial 2 took participants who were already on it and stayed on it. Assignment was 1:1 to the maximum tolerated dose of tirzepatide, either 10 mg or 15 mg, or to placebo for 52 weeks.[1]

The primary endpoint was the change in the apnea-hypopnea index, the number of apneas and hypopneas recorded during an hour of sleep. At entry the mean index was 51.5 events per hour in trial 1 and 49.5 in trial 2, and mean body-mass index was 39.1 and 38.7.[1] Those are severely affected people, well above the threshold most sleep laboratories use for severe disease.

The result

In trial 1 the index fell by a mean of 25.3 events per hour on tirzepatide (95% CI, 21.2 to 29.3) against 5.3 on placebo, an estimated treatment difference of 20.0 events per hour (95% CI, 14.2 to 25.8). In trial 2 it fell by 29.3 against 5.5, a difference of 23.8 events per hour (95% CI, 17.9 to 29.6). Both results carried P values below 0.001.[1]

Every prespecified key secondary endpoint also moved in tirzepatide's favor: body weight, hypoxic burden, patient-reported sleep impairment, high-sensitivity C-reactive protein and systolic blood pressure. The most frequent adverse events were gastrointestinal and mostly mild to moderate, the same profile described in the side-effect article.[1]

Subtract the larger of those differences from the larger of those baselines and the arithmetic is sobering. A participant starting near 50 events per hour and improving by 25 lands near 25, still well above the index of 15 that the prevalence literature counts as moderate-to-severe disease.[4] This is a substantial improvement in a severe condition rather than a cure for it, and the trial reported a mean, so individual results sat on both sides of it.

An index is not an outcome

A falling event count is a measurement of breathing, and it is tempting to read it as a promise about hearts. SAVE is the reason to resist that. It randomized 2,717 adults aged 45 to 75 who had moderate-to-severe apnea plus coronary or cerebrovascular disease to CPAP with usual care or usual care alone.[2]

CPAP drove the mean index from 29.0 events per hour at baseline to 3.7 during follow-up, a larger fall than any drug has produced. After a mean 3.7 years, a primary cardiovascular event had occurred in 17.0% of the CPAP group and 15.4% of the usual-care group, a hazard ratio of 1.10 (95% CI, 0.91 to 1.32).[2] CPAP did reduce snoring and daytime sleepiness and improved quality of life and mood.

Adherence in that trial averaged 3.3 hours a night, which is part of the explanation and not all of it. The honest reading is narrow: lowering the index is worth having for symptoms and for sleep, and nobody has shown that lowering it prevents heart attacks. Cardiovascular evidence in this drug class rests on a different trial entirely.

Weight loss already moved this number

The mechanism is old news. Sleep AHEAD enrolled 264 participants with type 2 diabetes, a mean age of 61.2 years and a mean index of 23.2 events per hour, then randomized them to an intensive behavioral weight-loss program or to group diabetes education.[3]

The lifestyle arm lost 10.8 kg at one year against 0.6 kg, and the adjusted index fell by 9.7 events per hour relative to the control arm. More than three times as many lifestyle participants reached total remission of their apnea, and starting index and weight lost were the strongest predictors of who improved.[3] Tirzepatide's contribution is magnitude, plus a trial designed around the sleep endpoint rather than reporting it as an aside.

Who is carrying this condition

Prevalence estimates from the Wisconsin Sleep Cohort, extrapolated to United States body-mass distributions, put moderate-to-severe sleep-disordered breathing at 10% of men aged 30 to 49 (95% CI, 7 to 12) and 17% of men aged 50 to 70. Among women the same estimates were 3% at 30 to 49 and 9% at 50 to 70.[4] Those figures rose by between 14% and 55% across two decades, tracking the weight distribution underneath them.

Apnea is diagnosed by a sleep study, not by snoring and not by a checkbox on an intake form. A telehealth questionnaire cannot produce an index, and no seller listed on the review index is running polysomnography. A reader who suspects apnea needs a diagnosis before a prescription is the relevant question at all.

What the trial does not settle

Trial 2 added tirzepatide to positive airway pressure. It did not test replacing it, and the decision to stop a machine belongs to a prescriber holding a repeat sleep study, not to a drop in weight. Nothing in these results licenses anyone to put a CPAP unit in a closet.

The indication also attaches to a branded product at labeled doses. A compounded vial shares a molecule name with the drug tested here and shares none of the review that produced the indication, which is the distinction drawn throughout the compounding article. Sellers quoting a sleep-apnea claim beside a compounded product are borrowing an indication their vial does not hold.

The doses studied were 10 mg and 15 mg of branded product over 52 weeks under trial supervision. What happens to the index after treatment stops was not measured here, and the weight trajectory after stopping is described in the stopping article. How a figure on this site gets checked before publication is set out in the methodology.

Frequently asked

Does tirzepatide treat obstructive sleep apnea?
In SURMOUNT-OSA it lowered the apnea-hypopnea index by an estimated 20.0 events per hour more than placebo in participants not using positive airway pressure, and by 23.8 events per hour more in those who were. Baseline indexes averaged 51.5 and 49.5 events per hour.
Does a lower apnea-hypopnea index mean a lower heart risk?
That has not been shown. In SAVE, CPAP cut the mean index from 29.0 to 3.7 events per hour and cardiovascular events still occurred in 17.0% of the CPAP group against 15.4% on usual care, a hazard ratio of 1.10. The index measures breathing, not outcomes.
Can someone stop CPAP after starting tirzepatide?
SURMOUNT-OSA does not answer that. Its second trial added tirzepatide to positive airway pressure and kept participants on the machine throughout. Stopping is a prescriber decision informed by a repeat sleep study, not by weight change alone.
Does losing weight by any method improve sleep apnea?
Sleep AHEAD suggests it helps. An intensive behavioral program produced 10.8 kg of weight loss against 0.6 kg at one year, with the index falling 9.7 events per hour relative to the control arm, and starting severity and weight lost predicted who improved most.

Sources

  1. [1] Malhotra A, Grunstein RR, Fietze I, et al. (2024). Tirzepatide for the Treatment of Obstructive Sleep Apnea and Obesity. N Engl J Med. PMID 38912654
  2. [2] McEvoy RD, Antic NA, Heeley E, et al. (2016). CPAP for Prevention of Cardiovascular Events in Obstructive Sleep Apnea. N Engl J Med. PMID 27571048
  3. [3] Foster GD, Borradaile KE, Sanders MH, et al. (2009). A randomized study on the effect of weight loss on obstructive sleep apnea among obese patients with type 2 diabetes: the Sleep AHEAD study. Arch Intern Med. PMID 19786682
  4. [4] Peppard PE, Young T, Barnet JH, et al. (2013). Increased prevalence of sleep-disordered breathing in adults. Am J Epidemiol. PMID 23589584

Where to get it

Best GLP-1 injections

Every injectable seller we can verify, with the price each one publishes and an honest read of what the trials measured.

Compare providers →

More in Evidence