O042 Does severity reflect burden? Equivalence of REM- and non-REM-predominant OSA

Abstract Introduction REM-predominant Obstructive Sleep Apnoea (REM-OSA) is often considered a distinct clinical phenotype, yet evidence for meaningful differences in symptoms, comorbidities, or prognosis remains inconsistent. Equivalence testing may clarify whether REM-OSA meaningfully differs from non-REM-predominant-OSA (nREM-OSA) across common clinical and patient-reported outcomes. Methods Participants with OSA per polysomnography from the Simplified Assessment and Monitoring of OSA cohort were included. REM-OSA was defined as a REM/non-REM apnoea-hypopnea index (AHI) ratio > 2 and non-REM AHI ≤ 15 events/h. All others were classified as nREM-OSA. Two one-sided tests of equivalence assessed similarity between REM- and nREM-OSA based on predefined minimally important differences (MIDs). Outcomes and MIDs were: Epworth Sleepiness Scale (ESS, 2 points), Insomnia Severity Index (ISI, 6 points), Functional Outcomes of Sleep Questionnaire-10 (FOSQ-10, 1.8 points), the EQ-5D visual analogue scale (EQ-5D-VAS, 5 points), systolic/diastolic blood pressure (SBP/DBP, 2mmHg), and self-reported depression/anxiety (2-fold greater odds). Results 68 participants were included in final analyses (age[mean ± SD] = 54 ± 14 years; 51/49% male/female). REM-OSA participants had significantly lower AHI (8 ± 6 events/h) vs. nREM-OSA (32 ± 29 events/h), p<.001. Despite this, equivalence within MIDs was found for FOSQ-10 and ISI scores (p<.001). Equivalence was not established for ESS, EQ-5D-VAS, blood pressure, anxiety, or depression. Notably, self-reported anxiety was significantly more prevalent in REM-OSA vs. nREM-OSA participants (48 vs.19%, p=.02). Discussion Despite milder OSA (lower AHI), those with REM-OSA reported symptoms of insomnia and daytime impairment equivalent to those with nREM-OSA, within MIDs. Anxiety was significantly more common in REM-OSA. Milder OSA may not equate to milder symptom burden in REM-predominant-OSA.