† Obstructive sleep apnoea (OSA) is associated with perioperative morbidity but is under-diagnosed in the community. † In this study of Canadian surgical patients, both anaesthetists and surgeons often failed to diagnose OSA. † Preoperative diagnosis was poor, even in patients with symptoms of moderate-to-severe OSA. Background. Obstructive sleep apnoea (OSA) affects 9-24% of the general population, and 90% remain undiagnosed. Those patients with undiagnosed moderate-to-severe OSA may be associated with an increased risk of perioperative complications. Our objective was to evaluate the proportion of surgical patients with undiagnosed moderate-to-severe OSA. Methods. After research ethics board approval, patients visiting preoperative clinics were recruited over 4 yr and screened with the STOP-BANG questionnaire. The 1085 patients, who consented, subsequently underwent polysomnography (PSG) (laboratory or portable) before operation. Chart review was conducted in this historical cohort to ascertain the clinical diagnosis of OSA by surgeons and anaesthetists, blinded to the PSG results. The PSG study-identified OSA patients were further classified based on severity using the apnoea-hypopnoea index (AHI) cutoffs. Results. Of 819 patients, 111 patients had pre-existing OSA and 58% (64/111) were not diagnosed by the surgeons and 15% (17/111) were not diagnosed by the anaesthetists. Among the 708 study patients, PSG showed that 233 (31%) had no OSA, 218 (31%) patients had mild OSA (AHI: 5-15); 148 (21%) had moderate OSA (AHI: 15-30), and 119 (17%) had severe OSA (AHI.30). Before operation, of the 267 patients with moderateto-severe OSA, 92% (n¼245) and 60% (n¼159) were not diagnosed by the surgeons and the anaesthetists, respectively. Conclusions. We found that anaesthetists and surgeons failed to identify a significant number of patients with pre-existing OSA and symptomatic undiagnosed OSA, before operation. This study may provide an impetus for more diligent case finding of OSA before operation.

Introduction:The diagnosis of obstructive sleep apnea (OSA) requires polysomnography (PSG). The screening of OSA utilizes questionnaires like the Epworth sleepiness scale (ESS) and the STOP-BANG questionnaire (SBQ) that have variable predictive value, as reported from the Western literature. Materials and methods: A retrospective cohort study was done in adult subjects (n = 80) presenting with symptoms of OSA. The demographic profile, ESS, and SBQ scores were evaluated. The level one recorded PSG was evaluated for apnea-hypopnea index (AHI) and positive airway pressure (PAP) therapy prescribed. Results: Of the 80 subjects, 75 were diagnosed as OSA on PSG. The mean age of the study group was 49 years, with 71% men and a mean BMI of 28.25 kg/m 2 . Epworth sleepiness scale ≥11 was in 62.66%, and SBQ ≥ 5 in 53% of the 75 subjects. Hence, both questionnaires failed to predict OSA in nearly half of the population. The mean AHI was 33.8/hour, and the mean continuous positive airway pressure (CPAP) was 10.05 cm H 2 O. The AHI had a significant correlation with BMI, ESS score, and CPAP. Epworth sleepiness scale had a 53% sensitivity and 60% specificity for diagnosing OSA using a cutoff of 11, whereas SBQ had a 68% sensitivity and 100% specificity using a cutoff of 5. Conclusion:The SBQ has a higher sensitivity and specificity to detect OSA than ESS as it envisages distinct clinical manifestations and risk factors of OSA. However, neither of the two can replace PSG.