GLP-1 Drugs and Sleep Apnea

A 2026 meta-analysis pooled six randomized controlled trials in 1,069 adults with moderate-to-severe obstructive sleep apnea. This was a synthesis of existing trials, not one new trial. Across four liraglutide trials and two tirzepatide trials, treatment reduced the apnea-hypopnea index by a pooled 9.99 events per hour versus control. The 95% confidence interval ranged from 3.50 to 16.47 fewer events per hour, but heterogeneity was high at I² = 87%.^[1]

The apnea-hypopnea index, or AHI, counts breathing pauses and shallow-breathing episodes per hour of sleep. Pooled results point toward fewer events with these medicines, yet they do not mean every drug, trial population or patient had the same response. The review itself says the variation between trials limits interpretation of the effect size and does not establish these medicines as substitutes for continuous positive airway pressure, or CPAP.^[1]

What the pooled result found

Carvalhal and colleagues searched databases and trial registries through February 10, 2025. Their main AHI estimate combined the six randomized comparisons. The same analysis estimated 9.68 kg greater weight reduction with GLP-1 receptor agonist-based treatment than with control, with a 95% confidence interval from 3.51 to 15.84 kg. Weight heterogeneity was even higher at I² = 98%.^[1]

Blood-pressure estimates were smaller and more consistent in direction. Pooled systolic pressure was 4.77 mmHg lower and diastolic pressure was 1.41 mmHg lower. The review found no statistically significant difference in oxygen desaturation index or minimum oxygen saturation. It also reported no significant pooled increase in adverse-event risk.^[1] That last finding should not be read as proof that adverse effects were absent. The included tirzepatide trials reported gastrointestinal events as the most frequent adverse events.^[2]

High I² values matter because they measure how much trial estimates differ beyond random sampling error. An I² of 87% for AHI and 98% for weight means the pooled averages hide substantial disagreement. Drug, dose, treatment duration, diabetes status, CPAP use and control treatment all varied across the evidence base.^[1]

The two tirzepatide trials

The two tirzepatide comparisons came from SURMOUNT-OSA, which reported two separate phase 3 randomized trials in one publication. Trial 1 enrolled adults who were not using positive airway pressure at baseline. Trial 2 enrolled adults already using it. Both trials compared the maximum tolerated dose of tirzepatide with placebo for 52 weeks.^[2]

In trial 1, AHI changed by 25.3 fewer events per hour with tirzepatide and 5.3 fewer with placebo. The estimated treatment difference was 20.0 fewer events per hour. In trial 2, the corresponding changes were 29.3 and 5.5 fewer events per hour, for a treatment difference of 23.8 fewer events per hour.^[2]

Those are drug-specific trial results, not the meta-analysis result. SURMOUNT-OSA also found improvements in body weight, hypoxic burden, systolic blood pressure and sleep-related patient reports with tirzepatide. Its participants had obesity as well as moderate-to-severe sleep apnea, and the study was funded by the medicine's manufacturer.^[2]

The four liraglutide trials

The largest liraglutide study was the 359-participant SCALE Sleep Apnea trial. Adults with obesity and moderate or severe sleep apnea who were unwilling or unable to use CPAP received liraglutide or placebo alongside diet and exercise for 32 weeks. The treatment difference in AHI was 6.1 fewer events per hour, while the treatment difference in body weight was 4.2 percentage points.^[3]

A 90-participant trial in people with type 2 diabetes and severe sleep apnea added liraglutide to CPAP and other drug treatment. After three months, the liraglutide group had lower AHI, body-mass index and mean systolic blood pressure than the control group. Minimum oxygen saturation was higher, and one participant stopped liraglutide because of gastrointestinal symptoms.^[4]

A 30-participant proof-of-concept trial compared CPAP, liraglutide-mediated weight loss and the combination for 24 weeks. AHI fell by 45 events per hour with CPAP, 43 with combined treatment and 12 with liraglutide alone. Only CPAP alone significantly reduced the study's measure of vascular inflammation.^[5] This small trial tested an early cardiovascular marker as well as sleep outcomes, so it should not be folded into the larger tirzepatide evidence without keeping its design separate.

The fourth liraglutide record was the 26-week ROMANCE trial in people with type 2 diabetes and sleep apnea. Its European trial-results record enrolled 132 participants across four groups and reported results for 121 in the main baseline table.^[6] Registry data can contribute to a meta-analysis, but it offers less narrative context than a full peer-reviewed report. Together, these liraglutide studies differed sharply in size, CPAP use and comparison group from the tirzepatide program.

Why this is not a class-wide verdict

The label 'GLP-1 receptor agonist-based therapies' groups liraglutide with tirzepatide, even though tirzepatide activates both GIP and GLP-1 receptors. Four of the six comparisons studied liraglutide, while the two larger and more recent sleep-apnea trials studied tirzepatide. No semaglutide sleep-apnea trial appeared in this six-trial pool.^[1]

Pooling can answer whether the collected trials lean in one direction. It cannot show that all medicines in a broad class have the same effect. The 9.99-event estimate also should not replace the separate SURMOUNT-OSA treatment differences or the SCALE result. Each number answers a different question under different trial conditions.^[1][2][3]

What this means for sleep-apnea care

Weight loss can reduce sleep-apnea severity in some people with obesity, and the randomized evidence here supports a treatment effect on AHI. The size of that effect remains uncertain because the pooled trials varied so much. The findings apply to the studied populations, mainly adults with obesity or type 2 diabetes and moderate-to-severe sleep apnea, rather than to every person with the condition.^[1]

The meta-analysis does not support stopping CPAP or changing medicine without clinical assessment. CPAP directly keeps the airway open during sleep, while medicines such as tirzepatide act through different pathways and have their own contraindications and adverse effects. One included trial found a much larger AHI reduction with CPAP than with liraglutide alone.^[5]

What we do not know yet

The synthesis cannot separate how much of the sleep benefit came from weight loss and how much, if any, came from another drug effect. It also cannot settle long-term cardiovascular outcomes, treatment durability or comparative effects among individual medicines. More head-to-head trials with consistent CPAP rules and oxygen outcomes would reduce the current uncertainty.^[1]

Six randomized comparisons favored GLP-1 receptor agonist-based treatment for AHI and weight, while high heterogeneity makes the average effect hard to generalize. Readers considering tirzepatide for obesity-related sleep apnea should discuss both sleep treatment and weight management with a clinician rather than treating the pooled number as an individual forecast.^[1]

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.