Validation of the Sonomat Against PSG and Quantitative Measurement of Partial Upper Airway Obstruction in Children With Sleep-Disordered Breathing

Abstract

To validate the Sonomat against polysomnography (PSG) metrics in children and to objectively measure snoring and stertor to produce a quantitative indicator of partial upper airway obstruction that accurately reflects the pathology of pediatric sleep-disordered breathing (SDB). Simultaneous PSG and Sonomat recordings were performed in 76 children (46 male, age 5.8 ± 2.8, BMI = 18.5 ± 3.8 kg/m2). Sleep time, individual respiratory events and the apnea/hypopnea index (AHI) were compared. Obstructed breathing sounds were measured from the unobtrusive non-contact experimental device. There was no significant difference in total sleep time (TST), respiratory events or AHI values, the latter over-estimated by 0.3 events hr-1 by the Sonomat. Poor signal quality was minimal and gender, BMI, and body position did not adversely influence event detection. Obstructive and central events were classified correctly. The number of runs and duration of snoring (13 399 events, 20% TST) and stertor (5748 events, 24% TST) were an order of magnitude greater than respiratory events (1367 events, 1% TST). Many children defined as normal by PSG had just as many or more runs of snoring and stertor as those with mild, moderate and severe obstructive sleep apnea (OSA). The Sonomat accurately diagnoses SDB in children using current metrics. In addition, it permits quantification of partial airway obstruction that can be used to better describe pediatric SDB. Its non-contact design makes it ideal for use in children.