Purpose: Polysomnography (PSG) is the gold standard in the diagnosis of obstructive sleep apnea (OSA). However, due to high cost and limited availability, this is difficult to access and often delayed. To evaluate the reliability of overnight oximetry as a screening tool for OSA diagnosis.
Method: All children suspected of OSA who underwent an overnight oximetry and subsequent PSG between January 2014 and April 2016 were studied retrospectively. The "McGill oximetry scoring" was compared with OSA diagnosis as per the American Association of Sleep Medicine.
Results: A total of 110 patients had both oximetry and PSG. Sixty-one children (56%) had normal oximetry, whereas 30 (27%) had McGill grade 2 and 19 (17%) had McGill 3 and 4. Sixty-four percent (64%) of children with normal oximetry had a normal PSG. Seventy percent (70%) of children with McGill 2 had either a normal or mild OSA on PSG. All the children with McGill 3 and 4 had moderate/severe OSA by PSG. The overall sensitivity and specificity of oximetry in identifying OSA were 63% and 78%, respectively, and the positive and negative predictive values (PPV and NPV) were 78% and 64%, respectively. The sensitivity and specificity of McGill 3 and 4 in diagnosing moderate/severe OSA on PSG were 59% and 100%, respectively, and the PPV and NPV were 100% and 78%, respectively.
Conclusion: Overnight oximetry provides satisfactory diagnostic performance in detecting moderate and severe OSA; however, a normal or McGill 2 score does not rule out OSA and a PSG is required for diagnosis.
Keywords: Children; Polysomnography; Pulse oximetry; Sleep disordered breathing.