The Effect of Hypopnea Scoring on the Arousal Threshold in Patients with Obstructive Sleep Apnea

A low respiratory arousal threshold (ArTHResp) is one of several endotypes that contribute to the pathogenesis of obstructive sleep apnea (OSA). Accordingly, it has emerged as a potential ‘drug-able’ target to treat OSA(1). Notably, Edwards et al.(2)developed a clinical screening tool to identify OSA patients with low ArTHResp based on three predictive variables obtained from standard overnight sleep studies [i.e. polysomnograms: PSGs]: 1) nadir oxygen saturation>82.5%; 2) Apnea-hypopnea-index (AHI)<30 events/hr; and 3) percentage of respiratory events that are hypopneas (%hypopnea)>58.3% that correctly predicted the presence of a low ArTHResp in 84% of patients. However, a key limitation of this tool is that it was developed using hypopnea scoring rules from the older“Chicago” (AASM1999) criteria(3) which have since been updated to the current AASM “Recommended” (AASM2012Rec) and “Acceptable” (AASM2012Acc) criteria. The AASM2012Rec defined hypopneas are associated with a ≥3% oxygen desaturation or EEG arousal, whereas AASM2012Acc requires a stricter ≥4% oxygen desaturation and does not consider arousals(4). While it is established that differing scoring criteria impact the AHI and %hypopneas(5-7), we recently demonstrated that scoring criteria also influence the measurement of another key OSA endotype(8), the sensitivity of the ventilatory control system. However, the extent to which changes in scoring criteria impact the predictive utility of the ArTHResp score is unknown. Accordingly, we aimed to determine the influence that the 2012 scoring criteria has on the tool’s performance.