Screening Tools for Obstructive Sleep Apnea in Pregnant Women: An Extended and Updated Systematic Review and Meta-analysis.

Abstract

The prevalence of obstructive sleep apnea syndrome (OSA) increases in women during pregnancy and negatively affects maternal and fetal outcomes. The updated systematic review and meta-analysis aimed to evaluate the validity of the Berlin, STOP-Bang, and Epworth sleepiness scale (ESS) questionnaires in detecting OSA in pregnant women. PubMed, Embase, and Web of Science were searched systematically up to March 2022. After eligible studies inclusion, two independent reviewers extracted demographic and clinical data. Bivariate random effects models were used to estimate the pooled accuracy measures including sensitivity and specificity, positive (PPV) and negative predictive values (NPVs), diagnostic odds ratio (DOR), and receiver operating characteristic curve (ROC) curve. We included 8 studies including 710 pregnant women with suspected OSA. The performance values of Berlin, STOP-Bang, and ESS questionnaires were as follows: the pooled sensitivity were 61% (95% confidence interval (CI): 40%-80%), 59% (95% CI: 49%-69%), and 29%, (95% CI: 10%-60%); pooled specificity were 61% (95% CI: 42%-78%), 80% (95% CI: 55%-93%), and 80% (95% CI: 50%-94%); pooled PPVs were 60% (95% CI: 0.49-0.72), 73% (95% CI: 61%-85%), and 59% (95% CI: 31%-87%); pooled NPVs were 60% (95% CI: 0.49-0.71), 65% (95% CI: 54%-76%), and 53% (95% CI: 41%-64%); and pooled DORs were 3 (95% CI: 1-5), 6 (95% CI: 2-19), and 2 (95% CI: 1-3), respectively. It seems that the Berlin, STOP-Bang, and ESS questionnaires had poor to moderate sensitivity and specificity in pregnancy, with the ESS showing the worst characteristics. Further studies are required to evaluate the performance of alternative screening methods for OSA in pregnancy.