Predicting the Presence and Severity of Obstructive Sleep Apnea with Optical Coherence Tomography.

Abstract

Rationale: The impact of obstructive sleep apnea (OSA) on microvascular system suggests that spectral domain optical coherence tomography (SD-OCT) evaluation of the choroidal vasculature could provide clinically relevant insights into disease presence and severity.

Objective: To investigate the value of choroidal vascular imaging with SD-OCT in diagnosing the presence and predicting the severity of OSA.

Methods: SD-OCT images of 120 patients with OSA were analyzed to extract choroidal biomarkers. Patients were categorized into four levels of OSA severity according to their apnea-hypopnea index. ImageJ/FIJI (National Institutes of Health, Bethesda, Maryland, USA) was used to measure choroidal thickness and vascular indices in Haller's and non-Haller's layers across regions on the nasal and temporal side of the fovea. Thickness ratios of choroidal layers, total choroidal area, choroidal vascularity index, and luminal-to-stromal ratios were compared between individuals without OSA and with different severity of OSA. ANOVA, ROC analysis, and logistic regression were employed to evaluate inter-group differences and the predictive value of choroidal parameters.

Results: OSA's presence and increasing severity significantly impacted the non-Haller's layer thickness and Haller's/non-Haller's layer thickness ratios, particularly in the nasal region (1000-2500 µm). The nasal 2500 µm region showed the highest discriminative power for severe OSA (AUC = 0.733, p < 0.001). Logistic regression analysis identified the Haller's/non-Haller's layer thickness ratio at nasal 2500 µm as the most significant predictor of severe OSA (Odds ratio = 2.147, p = 0.002), adjusted for age, gender, and comorbidities.

Conclusions: OSA is associated with choroidal microvascular remodeling, especially nasal to the fovea. This remodeling increases the ratio of Haller's/non-Haller's layer thickness ratio, which may be a potential biomarker for OSA severity. These findings highlight the utility of SD-OCT in non-invasively detecting systemic vascular alterations linked to OSA, supporting its role in early diagnosis and monitoring of disease progression.