Tirzepatide for Sleep Apnea: What SURMOUNT-OSA Showed and the FDA's First-Ever OSA Drug Approval

Until 2024, the standard pharmacological treatment for obstructive sleep apnea (OSA) was nothing — there wasn't one. CPAP machines and oral appliances were the only options, and roughly two-thirds of patients prescribed CPAP don't tolerate it long-term. The SURMOUNT-OSA trial, published in the New England Journal of Medicine on October 3, 2024 [1], changed that. Tirzepatide cut the apnea-hypopnea index by 25–29 events per hour from a baseline of around 50, both in patients actively using CPAP and patients who weren't. The FDA followed with what its own press release describes as the first medication ever approved for obstructive sleep apnea [2]. Here's what the trial actually showed and what it changes for the millions of obese adults living with sleep apnea.

Why this trial mattered before it ran

Obstructive sleep apnea is the most common sleep-disordered breathing condition in adults. It's strongly associated with obesity — visceral fat around the upper airway and neck contributes to the airway collapse that defines the disease. For decades, clinicians have observed that significant weight loss (typically via bariatric surgery) reduces OSA severity, sometimes dramatically. But until SURMOUNT-OSA, no randomized trial had specifically tested whether a pharmacological weight-loss intervention could deliver the same benefit on a measurable apnea endpoint.

The clinical context made the question urgent. Standard OSA treatment is positive airway pressure (PAP) therapy — delivered via CPAP, BiPAP, or APAP machines that hold the airway open with pressurized air during sleep. PAP works when patients use it, but long-term adherence is a well-documented clinical problem in sleep medicine: a substantial fraction of patients prescribed CPAP either abandon the therapy or use it inconsistently within the first year, though published adherence rates vary widely by study population, definition of “adherent,” and follow-up duration. For non-adherent patients, the remaining options before SURMOUNT-OSA were oral appliances (modest efficacy), positional therapy (very limited), and bariatric surgery (effective but invasive and inappropriate for most patients).

A pharmacological option that worked even in patients who couldn't tolerate CPAP would change the standard of care. SURMOUNT-OSA was designed to find out whether tirzepatide could be that option.

SURMOUNT-OSA trial design

SURMOUNT-OSA was actually two parallel randomized trials, run simultaneously and reported together [1]:

• Trial 1 — PAP-naïve cohort: patients with moderate-to-severe OSA and obesity who were NOT using PAP therapy at baseline. Mean baseline AHI was 51.5 events per hour. Mean baseline BMI was 39.1 kg/m².
• Trial 2 — PAP-treated cohort: patients with moderate-to-severe OSA and obesity who WERE using PAP therapy at baseline and continued PAP throughout the trial. Mean baseline AHI (on PAP) was 49.5 events per hour. Mean baseline BMI was 38.7 kg/m².

Both trials were phase 3, double-blind, placebo-controlled, randomized, and ran for 52 weeks. Participants in the tirzepatide arms received the maximum tolerated dose (10 mg or 15 mg subcutaneously weekly) per the standard SURMOUNT titration schedule. The primary endpoint in both trials was absolute change in AHI (apnea-hypopnea events per hour of sleep) from baseline to week 52, measured by polysomnography in a sleep lab — the gold-standard objective measurement of OSA severity [1].

The primary results

Both trials hit their primary endpoint with comfortable statistical margins [1]:

From a baseline of approximately 50 events per hour (which meets the clinical definition of severe OSA), tirzepatide cut AHI by roughly half — to around 25 events per hour in the PAP-naïve group and around 20 events per hour in the PAP-treated group. For context, an AHI below 5 is considered normal, 5–15 is mild OSA, 15–30 is moderate, and 30+ is severe. SURMOUNT-OSA participants moved from severe OSA territory to the lower end of moderate OSA on average, and a meaningful subset moved into the mild range or below clinically meaningful thresholds entirely.

What this means clinically

The 20-event-per-hour reduction in the PAP-naïve cohort is the more clinically remarkable result. These patients started the trial with no other OSA treatment, and tirzepatide alone — no CPAP, no oral appliance, no positional therapy — moved them from severe OSA to moderate. For patients who refuse or cannot tolerate PAP therapy, this is the first pharmacological intervention that delivers a clinically meaningful effect on the underlying disease, not just a downstream symptom.

The PAP-treated cohort result is also clinically interesting. Patients on tirzepatide PLUS PAP achieved an even larger AHI reduction (29.3 events/hour) than tirzepatide alone — but the most editorially important interpretation is that tirzepatide is additive to PAP rather than competing with it. Patients who currently tolerate CPAP and want to keep using it can add tirzepatide for a deeper response, while patients who don't tolerate CPAP have an alternative that works on its own.

Secondary endpoints

The published abstract [1] reports that all prespecified key secondary endpoints favored tirzepatide. These included measures the trial team selected because they capture the downstream consequences of OSA — the things that actually matter to patients beyond a number on a sleep study report:

• Body weight reduction
• Systolic blood pressure reduction
• Hypoxic burden — a measure of how much time the patient spent with low blood oxygen during sleep, which is the mechanistic driver of most of OSA's long-term cardiovascular harm
• High-sensitivity C-reactive protein (hsCRP), a marker of systemic inflammation associated with OSA
• Patient-reported sleep disturbance measures

The published trial reports that all of these moved favorably with tirzepatide vs placebo. The exact magnitudes for individual secondary endpoints are detailed in the full NEJM publication; we're reporting at the level of statistical direction here because the abstract verifies the direction but not every magnitude [1].

The FDA approval

Following SURMOUNT-OSA's positive primary endpoint result, the FDA approved Zepbound (tirzepatide) for the treatment of moderate-to-severe obstructive sleep apnea in adults with obesity. The FDA's own press release describes this as the first medication ever approved by the agency for obstructive sleep apnea [2].

The approval matters as much for what it signals as for what it does directly. For decades, OSA has been treated as a mechanical problem — your airway closes, so we deliver pressurized air to hold it open. Tirzepatide approval marks the first time the FDA has accepted a pharmacological, weight-loss-mediated approach as a legitimate treatment for OSA itself. That precedent is likely to influence the regulatory and clinical conversation around adjacent pharmacological approaches over the coming years.

For the most current label language, dosing information, and contraindications, refer to the official Zepbound prescribing information at the FDA's drug information portal [4].

Insurance coverage implications

The OSA approval is editorially significant for one reason most patients won't see in the press coverage: it creates a new covered indication for tirzepatide that some insurers will reimburse where they currently deny weight loss as the primary indication. Many commercial plans explicitly exclude GLP-1 / GIP-GLP-1 medications for weight loss alone, but those same plans typically cover the same drugs for an FDA-approved disease indication. Sleep apnea is now such an indication.

Patients with documented moderate-to-severe OSA whose insurance currently denies tirzepatide for weight loss should discuss with their prescriber whether the OSA indication may now provide a coverage path. Documentation requirements vary by insurer, but typically include a sleep study report with AHI and a treating physician's letter of medical necessity. See our GLP-1 insurance coverage audit for the per-insurer breakdown of how the major commercial plans handle GLP-1 prior authorization.

Adverse events: same pattern as SURMOUNT-1

SURMOUNT-OSA reported that the most frequently observed adverse events with tirzepatide were gastrointestinal in nature, primarily mild to moderate in severity, and consistent with what was seen in the obesity-only trial SURMOUNT-1 [1, 3]. Translation: nausea during dose escalation, diarrhea, occasional vomiting and constipation, most resolving with continued use.

For the full GLP-1 / GIP-GLP-1 adverse event picture including the percentage breakdowns from SURMOUNT-1 and STEP-1, see our GLP-1 side effects investigation. The OSA population didn't expose any new safety signal — the patient cohort is essentially a subset of the obese adult population already studied at length in the SURMOUNT program.

What this changes for OSA patients

Most OSA patients are not going to switch to tirzepatide monotherapy and discard their CPAP machine. CPAP, when tolerated, is still highly effective and considerably cheaper. The clinical utility of tirzepatide for OSA is sharper for two specific patient populations:

1. Patients with moderate-to-severe OSA + obesity who cannot tolerate CPAP. This is the largest unaddressed-need population in sleep medicine. Tirzepatide gives them, for the first time, an evidence-based alternative that actually changes the underlying disease rather than just a symptom.
2. Patients on PAP therapy who also have obesity and want a deeper treatment response. The PAP-treated arm of SURMOUNT-OSA (Trial 2) showed that adding tirzepatide on top of CPAP produces a larger AHI reduction than CPAP alone, even in patients with good PAP adherence at baseline.

For both populations, the conversation with a sleep specialist now needs to include the tirzepatide option as a legitimate first-line consideration. That wasn't true 18 months ago.

What we don't yet know

SURMOUNT-OSA gave us 52 weeks of randomized data. The open questions for the next several years of clinical research:

• Long-term durability. Does the AHI reduction persist past one year of treatment? What happens if patients lose insurance coverage and stop the drug? Based on the SURMOUNT-4 weight maintenance data for tirzepatide, the working assumption is that discontinuation will likely produce regain of both weight and AHI, but this needs direct measurement in OSA- specific populations.
• Semaglutide for OSA. No semaglutide- specific OSA trial has been published. It's plausible that semaglutide produces comparable benefits given its effective weight loss profile, but absent direct trial data, the FDA-approved indication is currently limited to tirzepatide.
• Mechanism — weight loss alone, or more? The relationship between weight loss and AHI improvement has been clinically observed for decades, but the published quantitative literature doesn't yet decompose how much of the SURMOUNT-OSA benefit is purely weight loss vs how much is potential additional effects (upper airway fat, inflammatory tone, autonomic function). This is an active research question.
• Head-to-head with CPAP. SURMOUNT-OSA measured tirzepatide vs placebo, not tirzepatide vs CPAP. A direct comparison would tell clinicians how to choose between the two for patients who could realistically use either.

The bottom line

SURMOUNT-OSA is the first randomized trial to demonstrate that a pharmacological intervention can produce a clinically meaningful reduction in objective obstructive sleep apnea severity. Tirzepatide cut AHI by roughly half from a severe baseline, both with and without concurrent CPAP. The FDA has since approved Zepbound for OSA in adults with obesity, making it the first prescription medication ever approved for the condition [2]. For the millions of obese adults who don't tolerate CPAP, this is the first credible alternative that addresses the disease itself rather than just compensating for its symptoms.