Preliminary assessment of portable sleep monitoring for diagnosis of obstructive sleep apnea in children.

Abstract

By observing the differences in sleep parameters between portable sleep monitoring (PM) and polysomnography (PSG) in children, we aimed to investigate the diagnostic value and feasibility of PM in children withsuspected obstructive sleep apnea (OSA). This prospective study enrolled consecutive children (aged 3-14years) with suspected OSA in Shenzhen Children's Hospital. They had PSG and PM in the sleep laboratory. Clinical parameters of the two sleep monitoring methods were compared. A total of 58 children participated. They were classified into two groups according to age: 28 children aged 3 to 5years and 30 children aged 6 to 14years. No significant differences were observed in apnea-hypopnea index (AHI), lowest oxygen saturation (LSaO2), andmean oxygen saturation (MSaO2) between PM and PSG, but the sleep efficiency with PMwas significantly higher (3-5years age: 92.2 ± 11.3% vs 85.2 ± 14.3%, 6-14years age: 93.2 ± 14.5% vs 84.8 ± 16.3%, both P < 0.05) than the sleep efficiencywith PSG. Pearson correlation analysis indicated a strong correlation between AHI, LSaO2, MSaO2, and sleep efficiency measured by PSG and PM. Receiver operating characteristic curve (ROC) analysis showed that PM was a reliable diagnostic tool for OSA. PM has high sensitivity (3-5years age: 95.8%, 6-14years age: 96.3%) and low specificity (3-5years age: 25.0%, 6-14years age: 33.3%) for OSA in children. Thus, there is a low rate of missed diagnoses, but there is some inaccuracy in excluding childrenwho do not have OSA. The results showed that PM has a good correlation with the various parameters of PSG. PM may be a reliable tool for diagnosing moderate and severe OSA in children, especially those who cannot cooperate with PSG or who have limited access to PSG.