Obstructive sleep apnea (OSA) is common but frequently underdiagnosed in patients with atrial fibrillation (AF), particularly in low- and middle-income countries, where access to polysomnography is limited. This study evaluated the diagnostic accuracy of the STOP-BANG, Berlin, and Epworth Sleepiness Scale (ESS) questionnaires for OSA screening in Vietnamese patients with non-valvular AF. This cross-sectional diagnostic accuracy study was conducted in a tertiary inpatient cardiology department in Ho Chi Minh City, Vietnam. Consecutive adults with chronic non-valvular AF completed the STOP-BANG, Berlin, and ESS questionnaires, followed by overnight unattended type III polysomnography (ApneaLink™, ResMed). OSA was defined as an apnea–hypopnea index ≥ 5 events/hour according to International Classification of Sleep Disorders (ICSD)-3 criteria. Diagnostic performance was assessed using sensitivity, specificity, area under the receiver operating characteristic curve (AUC), predictive values, and likelihood ratios. Among 126 participants (median age 61 years; 45.2% male vs. 54.8% female), OSA prevalence was 74.6%. STOP-BANG showed the highest diagnostic accuracy (AUC 0.903, 95% confidence interval 95% CI 0.852–0.954), with sensitivity of 81.9% and specificity of 96.9%. The Berlin questionnaire demonstrated moderate performance (AUC 0.824, 95% CI 0.735–0.914; sensitivity 89.4%; specificity 59.4%). ESS performed poorly in this population (AUC 0.499, 95% CI 0.386–0.613; sensitivity 6.4%). STOP-BANG demonstrated superior diagnostic accuracy and practicality for OSA screening in patients with non-valvular AF, supporting its use to guide referrals for confirmatory sleep testing in resource-limited settings.
Phuong et al. (Fri,) studied this question.