Abstract Introduction Obstructive Sleep Apnea (OSA) can occur either during Non-Rapid Eye Movement (NREM) and Rapid Eye Movement (REM) sleep and as such constitutes two different phenotypic aspects of OSA constellation. Previously published articles on OSA and their response to bariatric surgery considers OSA as a singular entity and the effect of surgery on these two different phenotypes are less studied. Our study compared baseline polysomnographic characteristics of both NREM and REM OSA, the effect of bariatric surgery on these characteristics and the predictors of surgical outcomes. Methods We retrospectively analyzed 31 patients who underwent polysomnography before and after bariatric surgery in a single center that serves urban, racially and socio-economically diverse community. Patients were classified as REM OSA if AHIREM was more than two times the AHINREM with a REM duration of at least 30 minutes in total. Demographic, anthropometric, and sleep metrics were compared using t-tests and χ² tests. Multivariable linear regression identified independent predictors of AHI improvement (ΔAHI = postoperative − preoperative AHI), adjusting for phenotype, baseline AHI, weight loss, and time since surgery. Results Baseline characteristics were comparable between NREM and REM groups (mean age = 47.56 vs 48.93 years, BMI=41.83 vs 42.53 kg/m²). REM OSA had significant female preponderance 86.7% vs 68.8% and longer total sleep time (442 vs 332 min, p = 0.003), higher REM percentage (19.9 vs 12.0%, p = 0.010), and markedly lower AHI (12.9 vs 59.2 events/h, p 0.001). Following bariatric surgery, mean weight decreased by 52 lb (ΔBMI −8.7 ± 4.8 kg/m²), and overall, AHI decreased by a mean of 45.6 in NREM and 7.6 in REM, p 0.001. On multivariate regression, baseline AHI (β = −0.87, 95 % CI − 0.96 to − 0.79,p 0.001) and time to repeat study since surgery (β = +0.13, 95 % CI 0.04-0.22,p=0.005) independently predicted ΔAHI, whereas REM OSA phenotype (β = +3.73, p = 0.13) and weight loss (β = +0.04, p = 0.36) were not significant predictors. Conclusion Bariatric surgery markedly improves OSA severity regardless of sleep phenotype. Higher baseline AHI predicts larger postoperative improvement, whereas longer intervals after surgery are associated with attenuated benefits suggesting a time-dependent decline, possibly due to partial weight regain. REM OSA represents a milder, less fragmented phenotype but doesn’t affect the overall weight loss. This abstract is funded by: None
Dasan et al. (Fri,) studied this question.