Noehren A, Brockmann PE, Urschitz MS, Sokollik C, Schlaud M, Poets CF (2010): Detection of respiratory events using pulse rate in children with and without obstructive sleep apnea
Pediatr. Pulmonol. 45 (5): 459-468. Epub Apr 13. doi: 10.1002/ppul.21196.

Oximetry is a simple but insensitive diagnostic test modality for obstructive sleep apnea (OSA) in children. Sensitivity for OSA may be enhanced if pulse rate is analyzed in conjunction with oxygen saturation. We analyzed 25 ambulatory polysomnographic recordings obtained from children with (n = 5) and without (n = 20) OSA. To assess sensitivity and specificity, pulse rate increases were determined during respiratory (i.e., apneas and hypopneas; n = 965) and non-respiratory sleep events (i.e., body movements; n = 1,197), and contrasted to baseline fluctuations (n = 209). The absolute pulse rate increase (APRI) was the parameter that differentiated best between baseline fluctuations and sleep events (area under the receiver operating characteristic curve [AUC]: 0.88). At a cutoff value of 12 beats per minute, APRI showed a sensitivity and specificity for detecting sleep events of 0.81 each. However, AUC was lower for the differentiation between respiratory and non-respiratory sleep events (0.77) and between central and obstructive respiratory events (0.68). In OSA cases, influencing factors for APRI following respiratory events were age, occurrence of a body movement, and severity of concomitant oxygen desaturation. We conclude that pulse rate analysis may be used to detect respiratory events in oximetry recordings in children.