The American College of Chest Physicians published a clinical practice guideline on obstructive sleep apnea during pregnancy, concluding that all nine conditional recommendations were conditional and based on very low–certainty evidence because of limited pregnancy-specific data.
Obstructive sleep apnea (OSA) affects approximately 9% of the general pregnant population and is more prevalent in complicated pregnancies. OSA has been associated with hypertensive disorders of pregnancy, gestational diabetes, severe cardiovascular morbidity, intensive care unit (ICU) admission, preterm birth, adverse fetal growth, neonatal ICU admission, and congenital anomalies.
The American College of Chest Physicians (CHEST) panel conducted a systematic review addressing nine population, intervention, comparator, and outcome questions and used the Grading of Recommendations, Assessment, Development, and Evaluations framework to assess evidence certainty and formulate recommendations.
The panel suggested screening pregnant patients for OSA using either pregnancy-specific or standard screening instruments. However, the panel found no direct evidence that screening improves perinatal outcomes, and currently available tools have limitations. Pregnancy-specific instruments require further external validation, while standard questionnaires, such as the Berlin and Snoring, Tired, Observed, Pressure, BMI, Age, Neck, Gender (STOP-BANG), were developed and validated in nonpregnant populations and have shown limited performance in pregnancy.
Among patients undergoing diagnostic evaluation, the CHEST panel suggested home testing or in-laboratory polysomnography, which remains the diagnostic standard; however access, cost, and overnight testing requirements may limit its feasibility during pregnancy. Home testing may improve access, but it may underestimate the apnea-hypopnea index (AHI) compared with polysomnography, particularly in mild OSA. The panel advised that patients with negative home tests but persistent high clinical suspicion may require polysomnography.
Treatment recommendations incorporated AHI, symptoms, and comorbidities. Positive airway pressure (PAP) therapy was recommended when the AHI was greater than 15 events per hour regardless of symptoms or comorbidities. For an AHI of 5 to 15 events per hour, treatment was suggested when symptoms, sequelae, or additional risk factors such as hypertension or diabetes were present. The panel noted that this approach extrapolates from recommendations for nonpregnant patients because pregnancy-specific treatment data remain limited.
Among patients already receiving continuous PAP prior to pregnancy, they suggested autotitrating PAP when available without additional financial or access burden. The panel found no available evidence on maternal or fetal outcomes, symptom improvement, or treatment adherence to inform this recommendation.
Evidence that PAP improves perinatal outcomes remains limited. One randomized controlled trial included 340 pregnant patients with sleep-disordered breathing. Patients assigned to PAP had lower blood pressure measurements and fewer hypertensive disorders compared with those receiving usual care. In a separate randomized trial of 36 patients with OSA and gestational diabetes, PAP did not improve glycemic control.
The CHEST panel favored PAP over alternative therapies when treatment was indicated, although comparative evidence remains sparse. They highlighted low PAP adherence during pregnancy and recommended monitoring PAP use, adverse effects, sleep quality, and daytime functioning throughout treatment.
For residual sleepiness despite optimized PAP treatment, the panel recommended first addressing sleep duration, sleep hygiene, other sleep disorders, physical activity, and comorbid insomnia. Scheduled naps were suggested when feasible, with stimulants reserved for persistent symptoms after scheduled naps had failed, with treatment determined through shared decision-making because of potential maternal and fetal risks.
The CHEST panel also suggested postpartum reassessment among patients whose OSA was first diagnosed during pregnancy. In one small study, 67% (n = 16/24) of patients had persistent OSA 6 to 8 months postdelivery. Another study of 17 patients with mild to moderate OSA found that 12% (n = 2) patients met the investigators’ definition of normalization—an AHI of 10 events per hour or lower—approximately 4 months postpartum.
The panel identified research priorities and knowledge gaps, including development and validation of pregnancy-specific screening tools, optimal timing of screening and treatment, strategies to improve PAP adherence, effects of therapy on perinatal outcomes, and predictors and natural history of postpartum OSA.
"Because most of the guidance is based on a low level of evidence with regard to screening and treatment of OSA in pregnancy, many knowledge gaps remain and call for future research in this area," wrote lead guideline author Carolyn M. D'Ambrosio, MD, of Harvard Medical School, and colleagues.
Full disclosures can be found in the published guidelines.
Source: CHEST Journal
