Respiratory polygraphy combined with pulse wave amplitude drops for the diagnosis of obstructive sleep apnea in pediatric patients

Respiratory polygraphy (RP) underestimates the obstructive apnea-hypopnea index (OAHI) compared to polysomnography (PSG) because it does not account for respiratory events associated with cortical arousals. To address this issue, RP can incorporate respiratory events associated with autonomic arousals, identified by drops in pulse wave amplitude (PWAD), obtained from the oximetry signal. This study aimed to compare the OAHI measured by RP alone, RP incorporating respiratory events associated with PWAD, and PSG. PSG recordings from 91 children (median age [1st; 3rd quartile]: 10 [8; 14]years) were analyzed. A few weeks later, the initial analysis was deleted, and the recordings were reanalyzed as RP alone and as RP incorporating respiratory events associated with PWAD. Three PWAD thresholds were tested: drops greater than 30 % (RP-PWAD30), 40 % (RP-PWAD40), and 50 % (RP-PWAD50) from baseline. The median [1st; 3rd quartile] OAHI measured by RP alone was 2.8 [1.7; 5.7]/h, underestimating OAHI measured by PSG (4.8 [3.0; 6.8]/h; Bland-Altman analysis: mean ± SD = -2.11 ± 2.51). In contrast, incorporating respiratory events associated with PWAD led to OAHI overestimation: 10.2 [7.0; 13.3]/h (Bland-Altman = 4.96 ± 4.18) for RP-PWAD30; 8.7 [6.0; 12.7]/h (Bland-Altman = 3.73 ± 3.76) for RP-PWAD40; 7.9 [5.1; 11.3]/h (Bland-Altman = 2.89 ± 3.69) for RP-PWAD50), in comparison to PSG. Since cortical arousals cannot be measured during RP, incorporating respiratory events associated with PWAD can improve OAHI estimation. Among the tested thresholds, the 50 % PWAD threshold resulted in the least overestimation in comparison to PSG.