Abstract Background Mild cognitive impairment (MCI) is an early marker of dementia. REM sleep-related parameters—including %REM, duration, and latency—are potential contributors to MCI risk. Cognitive impairments are frequently observed in severe obstructive sleep apnea (OSA) patients. Objectives This study evaluates the impact of REM sleep disruption on MCI risk factors in severe OSA patients, classified by %REM thresholds. Methods We retrospectively analyzed 278 severe OSA patients, of whom 116 exhibited %REM 12%. Sleep parameters and continuous positive airway pressure (CPAP)-induced REM rebound effects were compared between low and high %REM groups. Results Patients with %REM 12% (n = 21 of 116) showed significantly lower %REM, reduced REM duration, fewer REM episodes, and increased N1 sleep compared to those with %REM ≥12% (n = 11 of 162) (p 0.01). The ratio of %REM increase per reduction in one apnea-hypopnea event following treatment was median: 0.195, interquartile range (IQR): 0.115-0.261 in the lower %REM group, and median: 0.042, IQR: -0.026-0.160 in the high %REM group (p = 0.0033, adjusted for age, apnea-hypopnea index, and body mass index) (Fig. 1). CPAP-induced REM rebounds occurred only in the low %REM group, with sustained improvements in sleep parameters observed at one week and one-month post-treatment. Conclusions A distinct phenotype characterized by REM sleep reduction was identified in severe OSA patients. Given the established association between REM sleep disruption and MCI, further investigations—including moderate and mild OSA cases—are warranted to elucidate the pathophysiological mechanisms underlying REM-specific impairment and to refine targeted intervention strategies. This abstract is funded by: Japan Agency for Medical Research and Development (AMED), Grant Numbers JP21wm0425018
(jin) et al. (Fri,) studied this question.