ABSTRACT
BACKGROUND: Ambulatory exams were preferred in children during the COVID-19 pandemic. Polysomnography (PSG), the gold standard for obstructive sleep apnea (OSA) diagnosis, requires several leads and sensors to be attached to the child's body. Children are more comfortable with respiratory polygraphic (RP) recording, which needs fewer sensors. OBJECTIVE: To compare respiratory parameters obtained by home RP with those obtained by home PSG with the device installed at the child's home by a trained sleep nurse from a national health care provider. METHODS: Data from home PSGs performed in children aged 2-19 years were retrospectively included. The obstructive apnea-hypopnea index (OAHI) was computed in PSG and then in RP after removing the sleep signals. The two indexes were compared using non-parametric paired Wilcoxon rank test, Bland-Altman analysis and sensitivity-specificity analysis. RESULTS: 44 PSGs of 44 children were included with only 34 (77%) PSGs interpretable. Median (min-max) OAHI was significantly underestimated in RP than in PSG (2.2 (0-25) vs 4.0 (0.4-28), p < 0.0001), confirmed also by the Bland-Altman diagram, the magnitude of the difference being mean ± standard deviation -1.7 ± 1.7. The sensitivity and specificity of OAHI in RP to identify an OAHI ≥2/h in PSG was 0.91 for both. CONCLUSION: Unattended ambulatory RP performed at child's house and installed under carefully controlled conditions is a useful exam for diagnosing OSA in children with or without comorbidities. However, RP must be installed in a supervised environment and interpreted with caution as it tends to underestimate OSA severity.