Zepbound for Sleep Apnea: What the Evidence Actually Shows

For decades, the only real treatment for obstructive sleep apnea was a machine. CPAP works, when people use it — and a lot of people do not. In December 2024 that changed: the FDA approved Zepbound (tirzepatide) to treat moderate-to-severe obstructive sleep apnea in adults with obesity, the first medication ever cleared for the condition [3].

This page explains what the trial found, what the approval does and does not say, how it affects insurance coverage, and the genuinely unresolved question of what it means for CPAP. Nothing here is medical advice and no dose is recommended.

What did the sleep apnea trial actually find?

The evidence comes from SURMOUNT-OSA, which was really two trials run side by side under one master protocol [1]. Trial 1 enrolled adults with moderate-to-severe obstructive sleep apnea and obesity who were not using positive airway pressure therapy. Trial 2 enrolled adults who were using it and planned to keep using it. Together, 469 people were randomized to the maximum tolerated dose of tirzepatide — 10 mg or 15 mg weekly — or to placebo for 52 weeks.

The main measurement is the apnea-hypopnea index, or AHI. It counts how many times an hour your breathing stops or becomes very shallow during sleep. Above 30 events an hour is severe; 15 to 30 is moderate. Participants started around 50 events an hour, which is well into severe territory. Average BMI was about 39.

After a year:

	Not using CPAP	Using CPAP
AHI change on tirzepatide	−25.3 events/hour	−29.3 events/hour
AHI change on placebo	−5.3 events/hour	−5.5 events/hour
Difference	−20.0 events/hour	−23.8 events/hour
Body weight change	−16.1%	−17.3%

All differences were statistically significant, with p-values below 0.001 [1].

The trial also measured things that matter beyond the number on the sleep report: oxygen dips during the night (the “hypoxic burden”), a marker of inflammation called hsCRP, blood pressure, and how people reported their own sleep quality and daytime function. All improved more on tirzepatide [1].

Then there is the number that gets quoted most often. Up to 50.2 percent of tirzepatide-treated participants ended the trial with either fewer than 5 events an hour, or 5 to 14 events an hour with minimal daytime sleepiness — thresholds at which the trial investigators noted CPAP therapy may not be recommended [2].

A later post hoc analysis sorted people by severity category. Most participants on tirzepatide moved to a milder category, and between 20 and 32 percent finished with no measurable sleep apnea at all. Benefit looked similar whether starting BMI was under 35, between 35 and 40, or above 40 [6].

Is this just weight loss, or is something else happening?

Fair question, and researchers went looking for the answer.

A secondary analysis published in Nature Medicine in January 2026 used a statistical technique called mediation analysis to split the credit. The improvements in sleep apnea measures had their own independent effect on inflammation, insulin resistance and triglycerides — separate from the weight loss. Blood pressure improvement, on the other hand, tracked mainly with weight [5].

The authors’ conclusion was practical rather than theoretical: to get the full benefit for heart and metabolic health, you probably need to treat both the sleep apnea and the obesity, not one or the other.

That said, it would be a mistake to read this as proof of a special airway effect. The AHI improvement followed the weight loss closely in time, and a drug producing 16 to 17 percent weight loss in people with severe obesity would be expected to improve sleep apnea substantially on that basis alone.

What exactly does the FDA approval cover?

The current Zepbound prescribing information, verified on September 14, 2026, lists two indications: reducing excess body weight and maintaining that reduction long term, and treating moderate-to-severe obstructive sleep apnea in adults with obesity [4]. Both are prescribed alongside a reduced-calorie diet and increased physical activity.

Three boundaries worth knowing:

• It is adults with obesity. The trial excluded anyone with a BMI below 30, so the label does not cover sleep apnea in people who are not obese.
• It is moderate-to-severe. People with mild sleep apnea, meaning an AHI between 5 and 15, were not enrolled.
• It is Zepbound, not Mounjaro. Same molecule, different label. And semaglutide has no sleep apnea indication at all — no randomized trial has ever tested semaglutide with AHI as the primary endpoint.

On dosing, the label states the maintenance dosage for obstructive sleep apnea is 10 mg or 15 mg once weekly, reached through the standard escalation starting at 2.5 mg with increases no sooner than every four weeks [4]. Those match the doses used in the trial. What dose is appropriate for any one person is a prescriber’s call.

Can Zepbound replace CPAP?

This is the question people actually have, and the honest answer is: sometimes, maybe, but not on your own initiative.

The American Academy of Sleep Medicine published prescribing guidance in September 2025 that is worth reading closely. Its core messages: tirzepatide is a tool, not a cure; it works best combined with lifestyle change and continued sleep apnea therapy; CPAP should not be discontinued without proper reassessment; and patients should understand this may be long-term therapy, much like blood pressure medication [7].

The AASM also points to its guidance that follow-up sleep testing may be used if a clinically significant weight change of roughly 10 to 20 percent has occurred since diagnosis or since treatment started [8]. In practical terms, that is the mechanism by which someone might find out their sleep apnea has improved enough to change the plan: lose the weight, get retested, discuss the result.

What is happening in practice is not quite that orderly. A claims analysis covering roughly 330 million US adults found that in the six months after the label change, new tirzepatide starts among people with sleep apnea rose 16 percent. Among people with sleep apnea and no prior CPAP use, only 4 percent of those who started tirzepatide went on to start CPAP within six months, against 22 percent of those who did not — an 83 percent relative difference. Among existing CPAP users, 26 percent of those who started tirzepatide had a six-month gap in CPAP-related claims, versus 15 percent of matched patients who did not [9].

A gap in claims is not proof anyone stopped for medical reasons. But the direction is unmistakable, and the same analysis found that primary care clinicians wrote about 84 percent of these prescriptions while sleep medicine wrote about 0.2 percent — meaning most of these decisions are being made outside sleep specialty care.

A small real-world Italian study adds a related data point. In 20 patients, adding tirzepatide to CPAP produced far more weight loss and better sleep apnea control than CPAP alone. But CPAP adherence at six months was 60 percent in the combination group versus 90 percent in the CPAP-only group — not statistically significant in such a small sample, but pointing the same way [14].

Does insurance cover Zepbound for sleep apnea?

This is where the sleep apnea indication has an outsized practical effect.

Medicare Part D has been barred since 2003 from covering drugs used for weight loss. But it can cover a drug used for a separate FDA-approved indication. The December 2024 sleep apnea approval created exactly that opening, and in January 2025 CMS confirmed that Part D plans may cover Zepbound when it is prescribed for obstructive sleep apnea [12]. Coverage still depends on the individual plan’s formulary and typically requires prior authorization, usually with documentation of a sleep study and a qualifying BMI.

There is a second, separate route that runs the other way. The Medicare GLP-1 Bridge, which started July 1, 2026 and runs through December 31, 2027, covers the Zepbound KwikPen, Wegovy and Foundayo for weight management at a $50 monthly copay. But CMS is explicit: people who have type 2 diabetes, moderate-to-severe obstructive sleep apnea or MASH are eligible for Part D coverage and are therefore not eligible for the Bridge, even if they otherwise meet the criteria [11]. Single-dose Zepbound vials and pens are also excluded; only the KwikPen is covered.

In other words, an OSA diagnosis moves you out of the Bridge and into ordinary Part D coverage. Which of those is better for any given person depends on their plan, and it is worth checking rather than assuming. Commercial insurance operates under entirely different rules.

What happens if you stop?

Nobody has run a trial that answers this directly for sleep apnea, but the pieces are not hard to assemble.

The AHI benefit followed the weight loss. Separate trial data on tirzepatide withdrawal show substantial weight regain after stopping. Reviews written by the SURMOUNT-OSA investigators themselves note that discontinuation leads to significant weight regain and that use for sleep apnea may therefore require prolonged or indefinite therapy — while also noting that the long-term safety profile of extended use is not fully known [13].

The AASM guidance makes the same point in plainer language: discontinuing tirzepatide may affect any reductions in AHI and any improvement in severity achieved while on the drug [7].

What are the limits of this evidence?

Worth listing plainly, because they are all real:

• Fifty-two weeks. Sleep apnea is a lifelong condition. The trial followed people for a year.
• No head-to-head against CPAP. Nobody has randomized people to tirzepatide versus CPAP and compared outcomes. CPAP works immediately; tirzepatide takes months of dose escalation and weight loss to do anything.
• No hard outcomes. The trials were not designed or powered to show whether tirzepatide prevents the heart attacks, strokes or deaths associated with untreated sleep apnea. That may eventually come from other trials.
• Narrow population. People with mild sleep apnea, people with type 2 diabetes and people with a BMI under 30 were all excluded.
• Cost and access. Sleep medicine commentators have raised the concern that an expensive drug could widen existing disparities in who gets treated well.

The bottom line

Zepbound is a genuine first: a pill-bottle answer, or rather an injection-pen answer, to a condition that has only ever had a machine. The effect size is large and the trial was well run. For people who cannot tolerate CPAP or will not use it, that matters a great deal.

What it is not is a replacement you can decide on yourself. The sleep medicine field’s position is that CPAP should not be stopped without repeat testing, that treatment is likely long-term, and that combining the two approaches may be better than either alone. If you have sleep apnea and are considering this, the useful next step is a conversation with the clinician managing your sleep apnea — not a switch made on the strength of a weight-loss result.

Sources

1. Tirzepatide for the Treatment of Obstructive Sleep Apnea and Obesity — New England Journal of Medicine, June 2024
2. Tirzepatide for the Treatment of Obstructive Sleep Apnea and Obesity (full text) — PubMed Central
3. Zepbound sNDA S-013 approval package (obstructive sleep apnea indication) — US Food and Drug Administration, December 2024
4. ZEPBOUND (tirzepatide) injection prescribing information — US Food and Drug Administration, 2026
5. Tirzepatide on obstructive sleep apnea-related cardiometabolic risk: secondary outcomes of the SURMOUNT-OSA randomized trial — Nature Medicine, January 2026
6. Participants with Improvement in OSA Severity Following Treatment: A Post Hoc Analysis of SURMOUNT-OSA Studies — SLEEP, May 2025 (abstract)
7. What clinicians should know about prescribing tirzepatide for sleep apnea — American Academy of Sleep Medicine, September 2025
8. Obesity Management Resources for Sleep Medicine — American Academy of Sleep Medicine, August 2026
9. FDA’s First GLP-1 Approval for Sleep Apnea May Reduce Reliance on CPAP — Komodo Health, February 2026
10. Tirzepatide Linked to Reduced CPAP Use and Mortality in Patients With OSA — Endocrinology Advisor, January 2026
11. Medicare GLP-1 Bridge: Information for Part D Plans — Centers for Medicare & Medicaid Services, August 2026
12. Medicare can now cover Eli Lilly’s Zepbound for sleep apnea, CMS says — CNBC, January 2025
13. The Role of Incretins in Obstructive Sleep Apnea — Journal of Diabetes, November 2025
14. Tirzepatide Combined With CPAP in Obstructive Sleep Apnoea Syndrome With Comorbid Obesity — Diabetes, Obesity and Metabolism, February 2026