Pediatric obstructive sleep apnea diagnosis by respiratory polygraphy including hypopneas associated with autonomic microarousals

Respiratory polygraphy (RP) can be used to detect OSA in children, but it underestimates the obstructive apnea-hypopnea index (OAHI) because hypopneas associated with cortical arousals are not evaluated. To compensate for this disadvantage, hypopneas linked with autonomic micro-arousals (AMA) can be calculated from the drop in pulse wave amplitude, as increased respiratory efforts are related to autonomic nervous system activation, tachycardia, and elevated blood pressure. The aim of this study was to compare OAHI obtained by RP alone or by RP considering hypopneas associated with AMA with OAHI by polysomnography (PSG). 95 PSG of children with mean±SD age of 10±4y were analyzed. Few weeks later, the initial analysis was deleted, and recordings were analyzed as RP alone and as RP by considering hypopneas associated with AMA (computed as drops of 30%, 40% or 50% in pulse wave amplitude (PWA) obtained by pulse oximetry and reported to baseline). Bland-Altman analysis was performed to determine the agreements. OAHI by RP alone, 5.5±7.6/h, was underestimated vs OAHI by PSG, 7.4±8.2/h, and overestimated when considering hypopneas related to AMA (OAHI 9.7±8.5/h, for 30% drop in PWA; OAHI 8.9±8.2/h, for 40% drop in PWA; OAHI 8.4±8.1/h, for 50% drop in PWA) vs OAHI by PSG (p<0.0001 for all). Bland-Altman analysis confirmed the overestimation, the approach accounting for a 50% reduction in PWA being the most accurate. Because cortical micro-arousals cannot be obtained during an RP, hypopneas associated with AMA may be utilized to compute OAHI more precisely, with the approach accounting for a 50% reduction in PWA less overestimating OAHI compared to PSG.