Sleep disorders
Mild OSA is hidden in most pregnant women with obesity — and tracks with glucose
At a glance
| Study type | Cross-sectional study |
|---|---|
| Year | 2019 |
| Journal | Journal of Clinical Endocrinology & Metabolism 104(7):2569–2579 |
| Sample | n = 18 |
| Effect size | r=0.5–0.74 across glycemic and metabolic markers |
| Population | Pregnant |
Background
A prospective study of 18 normal-glucose-tolerant pregnant women with obesity (BMI 30–40) at 32–34 weeks gestation. 67% had mild obstructive sleep apnea (AHI ≥ 5) — none diagnosed before the study. AHI and oxygen-desaturation index correlated strongly with 24-hour glucose, hepatic insulin resistance, and waking cortisol.
What did the study actually find?
Eighteen pregnant women with obesity (BMI 30–40), all with normal glucose tolerance, were monitored at 32–34 weeks gestation. They wore a continuous glucose monitor for 72 hours and had three nights of home sleep testing (WatchPAT) on a controlled eucaloric diet. 12 of 18 (67%) had AHI ≥ 5, the threshold for mild OSA. None of them had been diagnosed with OSA before the study.
Why does that matter if their glucose tolerance was 'normal'?
Because the severity of their breathing disturbance correlated with metabolic dysfunction across the day — even though they didn't yet meet the bar for a gestational-diabetes diagnosis. AHI and oxygen-desaturation index (ODI) correlated with 24-hour glucose AUC (r = 0.50–0.54), mean 24-hour glucose (r = 0.55–0.59), estimated hepatic insulin resistance (r = 0.59–0.74), free fatty acids (r = 0.53–0.56), and waking cortisol (r = 0.49–0.64). The metabolic load of OSA appears to start before clinical diabetes shows up.
How would I know if I'm in this group?
Pre-pregnancy BMI ≥ 30 plus loud snoring, witnessed apneas, gasping, or daytime sleepiness — especially in the third trimester — should prompt a conversation with your OB. The 2023 SASM/SOAP guideline recommends pregnancy-specific screening tools (Facco, Louis, BATE) over Berlin/Epworth/STOP-BANG, and home sleep testing is acceptable for elevated-risk patients (covered on our Dominguez 2023 page).
Why is this study small but still informative?
n = 18 is small in absolute terms but the design is unusually clean: every participant had identical caloric intake on a controlled diet, continuous glucose monitoring around the clock, and home sleep testing on three separate nights. The within-subject correlations between sleep-disordered breathing and metabolic markers are tight (r > 0.5 with p < 0.05) precisely because the design controls noise that bigger studies can't.
How does Solas apply this?
Solas does not diagnose sleep apnea. If onboarding signals match elevated-risk criteria from the SASM/SOAP guideline (obesity, chronic hypertension, preeclampsia history, witnessed apneas), the app surfaces the screening recommendation and links to this study so users understand why the metabolic case for screening is concrete, not hypothetical.
Source
Farabi SS, Barbour LA, Heiss K, Hirsch NM, Dunn E, Hernandez TL
(2019).
Obstructive Sleep Apnea Is Associated with Altered Glycemic Patterns in Pregnant Women with Obesity Journal of Clinical Endocrinology & Metabolism 104(7):2569–2579.
DOI: 10.1210/jc.2019-00159
Read the original paper →
Solas does not own or distribute the underlying paper. Follow the link to the publisher for the full text. See our methodology.