Impact of Positive Airway Pressure and Mask Leakage on Dry Eye and Glaucoma Risk in Obstructive Sleep Apnea: A Cross-Sectional Analysis.

Purpose: This study investigates the association between obstructive sleep apnea (OSA), dry eye disease (DED), and glaucoma, focusing on the impact of positive airway pressure (PAP) usage and air leakage. Methods: This retrospective cross-sectional study included 57 adults with polysomnography-confirmed OSA between 2010 and 2023. Participants were grouped into PAP users (PAP+, n = 40) and non-users (PAP-, n = 17). Ocular assessments included tear film break-up time, Schirmer's test, Oxford staining, meibomian gland evaluation, intraocular pressure, cup-to-disc (C/D) ratio, and retinal nerve fiber layer thickness. PAP device data (usage duration and air leak rate) and OSA severity metrics were recorded. Group comparisons used chi-square and Student's t-test, and regression analyses identified associations between PAP leakage and ocular parameters. Results: Among the 57 OSA patients, PAP users showed a trend toward a higher risk of glaucoma (OR = 0.83) and DED (OR = 0.69) compared to non-users, but neither trend was statistically significant. PAP users had significantly more severe OSA, including longer N1 sleep stage (p = 0.0005), higher apnea-hypopnea index (AHI, p = 0.0001), and poorer oxygenation. PAP leakage: 95% (mean = 25.84 L/min) exceeded the 24 L/min threshold specified in ResMed's clinical guidelines, suggesting suboptimal therapy. Higher PAP leak was significantly associated with a lower Schirmer's test value (p = 0.031) and a higher C/D ratio (p = 0.040) on regression analysis. However, no significant differences were found in ophthalmic parameters between PAP+ and PAP- groups. Conclusions: Suboptimal PAP therapy as mask leakage or nocturnal hemodynamic changes may worsen evaporative dry eye and affect intraocular pressure. Our findings highlight the association between PAP mask leakage and reduced tear production, and suggest that OSA-related optic nerve stress may persist unless both hypoxia and nocturnal IOP fluctuations are properly managed. However, due to the relatively small sample size and retrospective cross-sectional design, future prospective studies with larger cohorts are needed to confirm these associations.