P012 Nocturnal-pulse-oximetry is unreliable for screening moderate-to-severe OSA in middle-aged females

Abstract Background Obstructive sleep apnoea (OSA) exhibits distinct sex differences in epidemiology, clinical presentation and pathophysiology. Nocturnal pulse oximetry (NPO) has been proposed as an inexpensive, accessible screening tool compared to polysomnography (PSG), the gold standard for diagnosing OSA. This study aims to determine the accuracy of NPO in diagnosing moderate-to-severe OSA in males and females. Methods We assessed the diagnostic accuracy of NPO in 80 people aged 40–65 years (37 males and 43 females) who underwent at-home oximetry (WristOx2™, Nonin) for three nights prior to in-laboratory PSG. Patients were predicted to have moderate-to-severe OSA on NPO if the 3% oxygen desaturation index was ≥15 events/hour, averaged across three nights. Moderate-to-severe OSA was confirmed if apnoea-hypopnoea index was ≥15 events/hour on PSG. We calculated sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), stratified by sex. Results Among males (52.6 ± 1.3 years old), NPO had a sensitivity 87%, specificity 100%, PPV 100% and NPV 82% at diagnosing moderate-to-severe OSA. Among females (53.1 ± 1.1 years old), NPO had a sensitivity 33%, specificity 100%, PPV 100% and NPV 74% at diagnosing moderate-to-severe OSA. In females, those misclassified as not having OSA by NPO had a higher BMI compared to those correctly identified (29.2 ± 3.6 vs 24.1 ± 3.8 kg/m2, p=.0016). Conclusion NPO demonstrates poor sensitivity in diagnosing moderate-to-severe OSA in middle-aged females, but high diagnostic accuracy in middle-aged males. Larger studies are needed to validate the role of NPO as a screening tool for moderate-to-severe OSA.