Skip to content
What Causes What
MECHANISM+ INCREASES

Obstructive sleep apnea → Insulin resistance (HOMA-IR)

Higher Obstructive sleep apnea increases Insulin resistance (HOMA-IR).

In 80 adults with severe OSA and morbid obesity, 12 weeks of CPAP improved two-hour glucose versus conservative care. HOMA-IR did not change in that trial. Pooled CPAP trials later found a modest HOMA-IR drop versus sham or no machine.

Randomized trial · Moderate · N=80

Supported by an experiment or a tightly identified design. Read the caveats.

Mechanism

Salord, Fortuna, Monasterio and colleagues assigned 80 adults with severe obstructive sleep apnea and morbid obesity, none with diabetes, to 12 weeks of CPAP or conservative care (NCT 01029561). Weight loss and activity were similar. HOMA-IR did not move. Impaired glucose tolerance reversed in nine people on CPAP and appeared in none. On conservative care it reversed in five and appeared in five. Two-hour glucose after a load changed more on CPAP (-0.5 ± 1.5 versus +0.33 ± 1.9). The authors read that as a peripheral insulin-resistance gain that fasting HOMA missed. Iftikhar, Khan, Das and Magalang pooled five CPAP trials in 244 people without diabetes. HOMA-IR sat 0.44 units lower on CPAP than on sham or no machine (-0.82 to -0.06). I-squared was 0. Iftikhar, Hoyos, Phillips and Magalang updated the RCT pool: HOMA-IR -0.43 (-0.75 to -0.11). Visceral fat and adiponectin did not move. Sixteen case-control files sat 0.51 standardized units higher in OSA. That association does not raise the stamp. Yang, Liu and Yang compared HOMA-IR to each person's baseline. Eight to 24 weeks of CPAP used more than four hours a night cut HOMA-IR by 0.75. Irregular use did not. BMI did not move. Abud, Salgueiro, Drake and colleagues pooled nine trials in 443 people without diabetes. HOMA-IR fell 0.39 units (-0.69 to -0.08). Fasting glucose did not. GRADE was low. Shang, Zhang, Wang and Han stayed with type 2 diabetes and OSA: seven trials, 691 people. HOMA-IR standardized mean difference -1.05. HbA1c -0.32. Fasting glucose -0.39. Patruno randomized 31 people with severe OSA to fixed or autoadjusting CPAP for three months. Both machines cut events. HOMA-IR and blood pressure fell on fixed CPAP, not on the autoadjusting machine. The signed object is obstruction. A machine that holds the airway open is less OSA. When HOMA-IR or two-hour glucose falls on that machine, more apnea still points to more insulin resistance. Hours in bed sit elsewhere. This page is the blocked airway.

Caveats

Salord is 80 completers, 12 weeks, morbid obesity, severe OSA, no diabetes. Two university hospitals. The main paper is the numbers are from the published abstract. The abstract names 80 people. HOMA-IR did not move in the main trial. The stamp rests on assigned CPAP versus conservative care and a glucose-tolerance endpoint the authors treat as peripheral insulin resistance. The HOMA drop lives in the pooled trials, not in Salord. Conservative care is not sham CPAP. Patruno compares two machines. Everybody was treated. It cannot identify untreated OSA. Yang is change from baseline and hours of use, not sham CPAP. They asked for larger randomized trials. Iftikhar 2013 is 83 percent men. Iftikhar 2016 also pools case-control association; that half is not the assigned design. Abud's GRADE is low. Shang is people who already have type 2 diabetes. Weight is the open backdoor. RICCADSA found insulin resistance improved over a year with BMI change, not with OSA type or CPAP assignment. MIMOSA assigned a Mediterranean weight-loss diet on top of CPAP prescribed to every arm. Neither sits on this page. Dapagliflozin treats glucose and then looks at the apnea index. That is the inverted arrow. Tirzepatide and other GLP-1 or SGLT2 files are not on this page for the same reason. This page is not hours in bed. Spiegel and the sleep-restriction trials live on the sleep-and-insulin page. More papers of the same design do not settle the claim. This is a journal entry, not medical advice and not a reason to start a machine.

Effect

qualitative. Salord 2016: 2-h glucose -0.5 vs +0.33, HOMA-IR ns, n=80. Iftikhar RCT metas: HOMA-IR -0.44 (2013) and -0.43 (2016) vs sham or no CPAP.

Nodes

Biology & endocrinologyObstructive sleep apneaBiology & endocrinologyInsulin resistance (HOMA-IR)

Named confounders

Cite this page

What Causes What. “Obstructive sleep apnea → Insulin resistance (HOMA-IR).” https://whatcauseswhat.org/edges/e-osa-insulin (atlas updated 2026-09-02).

https://whatcauseswhat.org/edges/e-osa-insulin

Papers