0481 USING STOP-BANG AS A SCREENING TOOL FOR SUSPECTED OBSTRUCTIVE SLEEP APNEA AMONG PATIENTS REFERRED TO A COMMUNITY SLEEP MEDICINE CENTER

• Published in: Sleep (New York, N.Y.) Vol. 40; no. suppl_1; p. A179
• Main Authors: Mou, J, Pflugeisen, B, Crick, B, Amoroso, P, Harmon, K, Tarnoczy, S, Ho, S, Mebust, K
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 28-04-2017
• Subjects: Body mass index; Gender; Patients; Sleep apnea
• ISSN: 0161-8105; 1550-9109
• DOI: 10.1093/sleepj/zsx050.480

Abstract
Introduction:
Effective screening is essential for the diagnosis and treatment of chronic obstructive sleep apnea (OSA). Early diagnosis also helps address and prevent related adverse health outcomes. The STOP-Bang is a validated screening tool used internationally for diverse patient subgroups, particularly preoperative patients and community populations. The predictive accuracy and discriminatory power among clinically referred patients is not well described. Its relative performance by gender is also poorly understood among patients with higher OSA risks.
Methods:
935 clinically referred patients suspicious of OSA were evaluated at MultiCare Sleep Medicine Center during 2015–2016. The STOP-Bang questionnaire was administered to each patient. In-center polysomnography (PSG) or home testing with portable monitors (PM) was performed for all patients. Apnea-hypopnea index (AHI) or respiratory disturbance index (RDI) was used as the gold standard for OSA diagnosis. Predictive performance of STOP-Bang was described and compared using optimal operating points (OOP) based on the Youden Index (J: maximum (sensitivity + specificity −1)) by gender. Utility of adjusted STOP-Bang was tested using alternative thresholds for body mass index (BMI) and/or neck circumference (NC) by receiver-operating characteristic (ROC) analysis. BMI at 25/30/40 and NS at 15/17 inches were taken as cutoffs for the adjustment.
Results:
Of the 403 female and 532 male patients, mean age, BMI and NC were 51.1 years, 35.7, and 41.8 cm, respectively. Mean AHI/RDI score was 25.4 with gender difference (F:18.9, M: 30.3, p <.001). 22.3% of the patients were free of OSA (AHI/RDI < 5). Overall, 30% and 20.2% were identified as having severe (AHI/RDI ≥ 30) or moderate (AHI/RDI ≥ 15 but < 30) OSA, respectively. Performance utility for moderate-or-severe OSA was shown at ≥ 4 for females (J = 0.30) and 5 for male patients (J = 0.28). STOP-Bang score at ≥ 5 reflects OOP in predicting severe OSA for both genders. At all three OSA severity levels, the combination of BMI≥30/NS≥17 inches outperformed conventional STOP-Bang in predictive utility.
Conclusion:
Use of STOP-Bang in clinically referred high-risk patients is appropriate but thresholds for total score and item cutoffs may need modification.
Support (If Any):