Accuracy of portable devices in sleep apnea using oximetry-derived heart rate increases as a surrogate arousal marker.

PURPOSE: Type 3 home study (HS) monitors do not detect cortical arousal-related hypopneas and may therefore underestimate the polysomnography (PSG)-based apnea-hypopnea index (AHI). Our aim was to test the hypothesis that scoring hypopneas using heart rate accelerations as a surrogate marker for cortical arousal (autonomic hypopnea; AnH) improves the accuracy of HS for OSA diagnosis, using PSG AHI as the diagnostic gold standard.

METHODS: We retrospectively identified patients referred for OSA who underwent complete PSG following an initial inconclusive HS. Respiratory events were scored using AASM research (Chicago) criteria with additional HS scoring for AnH, defined as hypopneas based on flow criteria associated with an increase in pulse oximetry-derived heart rate ≥ 6 beats/min.

RESULTS: A total of 178 patients met inclusion criteria, with mean (±SD) HS AHI = 4.4 ± 4.2/h, which increased to 8.5 ± 5.3/h with AnH scoring. The hypopnea arousal index on subsequent PSG was 7.6 ± 7.7/h, with total AHI 15.6 ± 11.9/h. Bland-Altman analysis showed improved agreement between HS and PSG AHI (mean difference 11.2/h (95%CI 33.6, - 11.1) without vs. 7.2/h (95%CI 29.6, - 15.4) with AnH scoring). Overall diagnostic accuracy was improved with AnH scoring as reflected by an increased area under the receiver-operating characteristic curve for AHI thresholds of 10 and 15 events/h.

CONCLUSIONS: In this retrospective analysis, the diagnostic accuracy of type 3 HS was improved by the inclusion of hypopnea-associated heart rate increases as a surrogate marker of arousal. Prospective studies are warranted to evaluate the impact of AnH scoring on clinical decision-making and patient outcomes.