O037 Diagnostic Accuracy of OSA Screening Tools for Acute Coronary Syndrome

Obstructive sleep apnoea(OSA) is associated with cardiovascular diseases including acute coronary syndrome(ACS). Current estimates suggest some 70% of patients with ACS have OSA. OSA in this population remains under-recognised. This high prevalence calls for effective OSA screening tools for this high-risk population. We aimed to assess the diagnostic accuracy of the Epworth Sleepiness Scale (ESS), Berlin and STOP-BANG questionnaire in screening for OSA in patients with ACS. 70 patients with ACS (age IQR: 58[52,63] years, BMI: 27[24,30]kg/m2) were recruited from the cardiology department following an admission for ACS. Patients were administered with the ESS, Berlin and STOP-BANG questionnaires within 6 months of ACS admission. A level 2 polysomnogram was conducted to confirm the presence and severity of OSA (AHI≥5). The diagnostic accuracy of ESS>10, Berlin questionnaire (high-risk category), STOP-BANG score ≥3 (intermediate-high risk) and STOP-BANG score ≥5 (high-risk) was compared to a level 2 polysomnogram for different OSA severity. OSA (AHI≥5 events/h) was diagnosed in 94% of patients with ACS. 67% of patients had moderate-severe OSA (AHI≥15 events/h). ESS, Berlin and STOP-BANG questionnaires have poor diagnostic accuracy across all severity of OSA (ESS>10, AUC:0.60 (95%CI: 0.42-0.79); Berlin, high risk, AUC:0.54 (95%CI: 0.39-0.68); STOP-BANG ≥3, AUC:0.69 (95%CI: 0.56-0.82); STOP-BANG ≥5, AUC: 0.66 (95%CI: 0.53-0.79)). This study confirms a very high prevalence of OSA in patients with ACS. The Berlin and STOP-BANG questionnaires alone are inadequate to identify OSA in this high-risk cohort. Simplified objective measures to screen for OSA may be better suited for this patient population.