Background: The polysomnography (PSG) is the gold-standard for obstructive sleep apnea (OSA) syndrome diagnosis and assessment under positive airway pressure (PAP) therapies in children. Recently, an innovative digital medicine solution, including a mandibular jaw movement (MJM) sensor coupled with automated analysis, has been validated as an alternative to PSG for pediatric application.
Objective: This study aimed to assess the reliability of MJM automated analysis for the assessment of residual apnea/hypopnea events during sleep in children with OSA treated with noninvasive ventilation (NIV) or continuous PAP (CPAP).
Methods: In this open-label prospective non-randomized multicentric trial, we included children aged from 5 to 18 years with a diagnosis of severe OSA. The children underwent in-laboratory PSG with simultaneous MJM monitoring and at-home recording with MJM monitoring 3 months later. Agreement between PSG and MJM analysis in measuring the residual apnea-hypopnea index (AHI) was evaluated by the Bland-Altman method. The treatment effect on residual AHI was estimated for both PSG and MJM analysis.
Results: Fifteen (60% males) children were included with a median age of 12 years [interquartile range 8-15]. Two (17%) were ventilated with NIV and 13 (83%) with CPAP. There was a good agreement between MJM-AHI and PSG-AHI with a median bias of -0.25 (95% CI: -3.40 to +2.04) events/h. The reduction in AHI under treatment was consistently significant across the three measurement methods: in-laboratory PSG and MJM recordings in the laboratory and at home.
Conclusion: Automated analysis of MJM is a highly reliable alternative method to assess residual events in a small population treated with PAP therapies.
Keywords: continuous positive airway pressure; machine learning; mandibular jaw movement; noninvasive ventilation; pediatric obstructive sleep apnea.
© 2024 The Authors. Pediatric Pulmonology published by Wiley Periodicals LLC.