Abstract
BACKGROUND: The incidence of obstructive sleep apnea (OSA) in children has increased in recent years. Many primary hospitals have not polysomnography which can diagnose OSA. To investigate the lowest blood oxygen saturation at night (LSaO(2)) combine with obstructive sleep apnea 18 items survey (OSA-18) scale to initially screen the sleep status of children with OSA. METHODS: A retrospective study of 189 children aged 4-12 years with sleep-disordered breathing was conducted. All children were monitored using polysomnography and divided into the simple snoring and OSA groups (mild, moderate, and severe). Their parents completed the OSA-18 scale. Correlation among three indices [LSaO2, OSA-18 scale, and obstructive apnea hypoventilation index (OAHI)] was assessed. Subsequently, series and parallel tests were used to understand the sensitivity and specificity of diagnosis. RESULTS: There was no statistical differences in sex (P = 0.909) and age (P = 0.894), and a significant difference in OSA-18 scores between the simple snoring and OSA groups (P = 0.014) but not in the LSaO2 (P = 0.409). OSA-18 and LSaO2 scores of the mild, moderate, and severe groups were significantly different (P < 0.05), and LSaO2 was correlated with the OAHI. Use of the OSA-18 scale combined with LSaO2 to assess the boundary value of mild and moderate-to-severe OSA was better than that of each index independently. The sensitivity and specificity the parallel test were 85.7% and 62.7%, respectively, can better predict OSA than series test. CONCLUSION: The OSA-18 scale combined with LSaO2 has diagnostic value for the diagnosis of OSA, and can be used as a prediction tool for OSA.