Comparison of three questionnaires for screening obstructive sleep apnea syndrome

Abstract

Background: Obstructive sleep apnea syndrome (OSAS) is one of the most common sleep disordered breathing and is still under-diagnosed. Screening patients at high risk of OSAS is extremely important. This study aimed at comparing the predictive probabilities of Epworth sleepiness scale (ESS), STOP-Bang questionnaire (SBQ) and Berlin questionnaire (BQ) in screening OSAS. Methods: A cross sectional study was carried out from Mai to December 2017 including 90 patients suspected for OSAS. All the participants answered the ESS, SBQ and BQ. Overnight respiratory polygraphy was done for all patients. The threshold for diagnosis of OSA was set at an AHI ≥5. Sensitivity, specificity and area under receiver operating characteristics curve (AUC) for the three questionnaires were calculated. Results: OSAS was diagnosed in 94% of patients. It was mild in 14%, moderate in 20% and severe in 66% of patients.ESS had the lowest sensitivities for screening OSAS, moderate-to-severe OSAS and severe OSAS (42%;42% and 41%).SBQ and BQ had proximate sensitivities for screening patients with AHI≥5 (95% and 96%) and patients with AHI≥15 (95% and 97%). However SBQ was superior to BQ to screen patients with AHI≥30 (100% and 96%). For screening OSAS, SBQ had the largest AUC (0.73; 95% CI:0.53-0.94) followed by BQ (0.71; 95% CI:0.45-0.95) and ESS (0.56; 95% CI:0.38-0.75) (p>0.05), none of the three scores was significant (p>0.05). Applying SBQ was significant to predict OSAS in patients with AHI ≥15 (0.66; 95% CI:0.51-0.81; p=0.03) and with AHI ≥30 (0.66; 95% CI: 0.54-0.78; p=0.01). Conclusions: SBQ and BQ seem to be the best tools in screening patients with high risk of OSAS. However, SBQ is superior to BQ in detecting severe OSA.