Abstract
BACKGROUND: This study aimed to estimate the prevalence of rapid eye movement-related obstructive sleep apnea (REMrOSA) using common definitions. METHODS: This was a retrospective cohort study that used three sets of criteria to identify patients with REMrOSA. These criteria were defined as strict, intermediate, and lenient depending on the apnea-hypopnea index (AHI), AHI during REM sleep/(AHI during non-REM sleep [NREM-AHI]), NREM-AHI and REM duration. RESULTS: The study included 609 patients with OSA and full sleep study. The prevalence of REMrOSA was 26%, 33%, and 52% using the strict, intermediate, and lenient criteria, respectively. There were no differences in the patients' general and demographic characteristics between the different groups of the three definitions. REMrOSA patients were more likely to be younger females than non-REMrOSA (NREMrOSA) patients. Comorbidities were more frequent in the REMrOSA group compared to NREMrOSA when using strict and intermediate definitions. In contrast, AHI, mean O2 saturation, and time spent <90% O2 saturation were significantly worse during NREMrOSA compared to REMrOSA, regardless of the criteria used. Our study reported higher AHI, lower mean oxygen saturation, lower minimum oxygen saturation, and longer time of desaturation during REMrOSA when lenient definition was used compared to when strict and intermediate definitions were used. CONCLUSIONS: REMrOSA is a common condition with a prevalence ranging between 26% and 52% depending on what definition is applied. Although OSA tends to be more severe with lenient definition, however, the clinical and polysomnographic features were similar among REMrOSA groups regardless of the definition used.