Selected Retinal and Choroidal Parameters Assessed by Optical Coherence Tomography and Optical Coherence Tomography Angiography in Patients With Obstructive Sleep Apnea

Introduction

Obstructive sleep apnea (OSA) is a disorder characterized by repeated collapse of the upper airway during sleep, which may partially limit or completely block airflow [1]. Early diagnosis of OSA is crucial to improve sleep quality and reduce associated health risks. One of the primary diagnostic tools is polysomnography (PSG), a comprehensive sleep study that records various functions of the respiratory system [2]. Whereas PSG remains the gold standard in the diagnostic process, respiratory polygraphy is a less complex alternative. However, it may underestimate apnea-hypopnea index (AHI) values, which can potentially affect clinical decisions, particularly in mild to moderate cases [3].

OSA severity is commonly assessed using the AHI, which represents the number of apneas and hypopneas per hour of sleep. According to the American Academy of Sleep Medicine guidelines, as summarized by Epstein et al, an AHI of fewer than 5 events per hour is considered within the normal range; 5 to 14.9 events per hour corresponds to mild OSA, 15 to 29.9 represents moderate OSA, and 30 or more events per hour indicates severe OSA [4].

The prevalence of moderate and severe OSA ranges from 1.2% to 23.4% in women and from 3.9% to 49.7% in men [5–7]. According to the HypnoLaus study, the overall prevalence of sleep-disordered breathing (defined as AHI >5) reached 60.8% in women and 83.8% in men [7]. Major risk factors for OSA include overweight, obesity, male sex, older age, hypertension, and a family history of the condition [8].

OSA treatment methods include continuous positive airway pressure therapy, the use of oral appliances, weight reduction, and various types of upper airway surgery [9]. If left untreated, the disease may lead to serious long-term health complications and substantially increase the risk of cardiovascular disease, stroke, and metabolic disorders [10,11]. Additionally, patients frequently experience excessive daytime sleepiness, memory and concentration difficulties, and mood disturbances [8,10]. Therefore, early diagnosis and timely implementation of treatment are crucial to reduce the adverse effects of OSA [8,10,11].

Repetitive episodes of breathing cessation during sleep can lead to oxidative stress, increased inflammatory responses, and autonomic dysfunction, which may result in ocular complications [12]. Previous studies have highlighted associations between OSA and several disorders, including glaucoma, floppy eyelid syndrome, nonarteritic anterior ischemic optic neuropathy, and dry eye disease [13,14]. One study showed that 66.67% of patients with OSA using continuous positive airway pressure therapy had been diagnosed with dry eye disease, and 54.17% of these patients also exhibited floppy eyelid syndrome [15]. Treatment options for OSA-related ocular conditions include artificial tear supplements, high-viscosity topical gels, bandage contact lenses, and, in severe cases, eyelid surgery. Early detection and appropriate treatment are important for improving quality of life [15].

Efforts to identify economical and noninvasive screening methods are crucial for the detection and prediction of OSA complications. The increasing availability of ophthalmic imaging techniques has facilitated ocular condition monitoring. Optical coherence tomography (OCT) is a valuable tool for investigating alterations in the retina and choroid; it enables the detection of abnormalities in thickness and structure. Spectral domain OCT is a particularly noninvasive imaging technique that can identify even subtle alterations in the layers of ocular structures [16].

Optical coherence tomography angiography (OCTA) is a noninvasive technique used to image and quantify retinal vasculature, including superficial and deep vessel density (SVD and DVD), as well as nonperfusion areas such as the foveal avascular zone (FAZ). The use of standardized sectors for vessel density assessment in the macular and peripapillary regions enables consistent application of this method across various ophthalmologic conditions [17]. Given the capabilities of OCTA, visualization of specific vascular layers (eg, SVD and DVD) is possible without the need for intravenous dye injection [18,19]. Previous OCTA studies of OSA populations have shown reductions in SVD and DVD [17–19], particularly within the peripapillary and parafoveal regions; however, findings remain inconsistent, highlighting the need for further investigation.

The aim of the study was to assess the effect of OSA on the retina, including FAZ and vessel density parameters, as well as choroidal thickness (CT).

Material and Methods

ETHICS COMMITTEE APPROVAL:

The study protocol was approved by the Bioethics Committee of the Military Institute of Aviation Medicine (approval no. 10/2020; date of approval: January 13, 2021); it was conducted in accordance with the Declaration of Helsinki. The results of the present study are part of the project entitled “Assessment of the visual system in patients with diagnosed obstructive sleep apnea syndrome.” This observational study did not require registration in a clinical trial registry.

DECLARATION OF PATIENT CONSENT:

The authors explained to participating patients that the clinical findings obtained from the study would be reported in a scientific journal to advance medical knowledge; patient identification and data would remain confidential. All patients provided informed consent to take part in the study.

STUDY DESIGN:

This prospective cross-sectional study was conducted between January 1, 2023, and January 1, 2025, at the Department of Ophthalmology, Military Institute of Aviation Medicine, Warsaw, Poland.

Consecutive patients from the Department of Otorhinolaryngology, Faculty of Medicine and Dentistry, Medical University of Warsaw, were prospectively enrolled. The control group included individuals with no or mild OSA, whereas the study group comprised patients with moderate to severe OSA. OSA presence and severity were defined based on the AHI. According to the American Academy of Sleep Medicine guidelines, a result of fewer than 5 events per hour indicated a nonpathological range; 5 to 14.9 events per hour indicated mild OSA; 15 to 29.9 indicated moderate OSA; and 30 or more events per hour indicated severe OSA. The following patient data were verified: age, sex, ophthalmologic diseases, general health status, previous treatments and surgeries, and prior ophthalmologic treatments and surgeries. Inclusion and exclusion criteria were applied as detailed in Table 1.

All patients underwent a comprehensive ophthalmologic examination, including mydriasis, within 1 month of PSG. OCT, en face OCT, and OCTA were also performed. No OSA treatment was initiated before the ophthalmologic examination.

Best-corrected visual acuity was measured monocularly using logarithm of the minimum angle of resolution (logMAR) charts (Lighthouse International, New York, NY, USA) at a distance of 5 m. Only participants with normal visual acuity (best-corrected visual acuity ≤0.9 logMAR) were included (Table 1). The ophthalmologic examination was performed using a Topcon SLD701 slit lamp and Volk SuperField lens.

OCT and OCTA examinations were performed using the AngioVue Imaging System (Optovue Inc., Fremont, CA, USA). Macular parameters were assessed in the foveal (0–1 mm) and parafoveal (1–3 mm) regions across the nasal (N), temporal (T), superior (S), and inferior (I) quadrants. The HD Angio Retina protocol was used, covering a 6×6 mm scanning area for OCT, en face OCT, and OCTA imaging. The FAZ area was automatically measured using the built-in AngioVue software. Measurement depth for FAZ assessment was automatically determined by the device software, and investigators did not influence the segmentation settings. Retinal blood flow was quantified within a circular region (1-mm radius; 3.14 mm²) centered on the fovea at the level of the outer retina and choriocapillaris using an automated analysis algorithm. All OCT and OCTA scans were acquired by 2 qualified ophthalmologists with dedicated training in these imaging techniques. Image assessors were masked to group allocation during analysis. Quality control was performed before analysis by evaluating signal strength and the presence of artifacts. Scans with poor image quality, motion artifacts, segmentation errors, or signal strength of 6 or below were excluded to ensure the reliability of quantitative measurements.

Choroidal measurements were manually performed with tools provided in the AngioVue Imaging System software. Using deep choroidal imaging technology, the hyperreflective retinal pigment epithelium band and uveoscleral junction were identified on OCT scans. The distance from the outer border of the retinal pigment epithelium to the uveoscleral junction was then measured perpendicularly at the point of maximal CT. Measurements were performed 3 times by 2 independent observers, and the mean value was used for analysis. This approach was utilized to minimize measurement variability and ensure reliable CT assessment. To avoid diurnal variation in the measured parameters, all examinations were performed between 7: 00 AM and 10: 00 AM.

STATISTICAL ANALYSIS:

Numerical variables were described using the mean (M), standard deviation (SD), median (Me), and lower and upper quartile (Q) values. Discrete variables were presented as counts and percentages. The normality of distribution was assessed using the Shapiro-Wilk test. Levene’s test was used to evaluate homogeneity of variance. Multifactor analysis of variance without replication was performed for normally distributed numerical variables with homogeneous variance to estimate the significance of differences according to OSA status. In the analysis of variance without replication model, the following explanatory variables were included: age, sex, and body mass index. Generalized linear models were utilized when variables were not normally distributed or when the variance was heterogeneous. A logistic regression model was used for dichotomous dependent variables. Error correction procedures for intra-participant correlation due to binocular measurements were applied in the above-mentioned statistical analyses. P-values <0.05 were considered statistically significant. All statistical analyses were performed using Statistica 13.3 (TIBCO Software Inc., Palo Alto, CA, USA).

Results

Forty-nine patients (98 eyes) were included in the study: 33 patients with moderate or severe OSA (66 eyes) and 16 controls (32 eyes). Table 2 presents the baseline characteristics of the study cohort. Statistically significant differences between groups were observed in age, mean heart rate, and CT. The remaining parameters were examined with respect to sex and did not show statistically significant differences between the groups. Pearson correlation analysis revealed no significant relationships between age and most of the investigated ocular parameters (Table 3).

A statistically significant negative correlation was observed between age and parafoveal SVD in the temporal quadrant (TQ) (r=−0.22, P=0.0263). A stronger negative correlation was found between age and CT (r=−0.36, P=0.0002), indicating that CT decreases with age (Figure 1). Other correlations between age and clinical parameters, including FAZ, foveal vessel density within 300 μm (FD-300), SVD, and DVD in various regions, were not statistically significant (P>0.05).

PSG data for the control and study groups are presented in Table 4. Mean AHI in the study group was 46.41 events per hour, compared with 8.78 events per hour in the control group. Oxygen saturation metrics also significantly differed. The OSA group had lower mean SpO2 (91.26% vs 93.35%), a greater mean SpO2 drop (6.10% vs 3.63%), lower minimum SpO2 (77.27% vs 86.81%), and a longer duration with SpO2 below 90% (71.53 vs 9.38 minutes).

FAZ parameters were analyzed between the study and control groups (Table 5). The mean FAZ area was significantly larger in the study group than in controls (0.2609 mm2 vs 0.2207 mm2; P=0.0163) (Figure 2). Similarly, FAZ perimeter was greater in patients with moderate or severe OSA than in controls (1.98 mm vs 1.81 mm; P=0.0236) (Figure 3).

No difference was found in FD-300, with comparable values between the groups (50.35% vs 49.73%; P=0.2852). Furthermore, no differences were observed in SVD between the study and control groups across all evaluated retinal regions (Table 6). Similarly, no differences were observed in DVD between the groups across all evaluated retinal regions (Table 7).

Discussion

LIMITATIONS:

This study is limited by its cross-sectional design, which precludes causal inferences, a relatively small and unequal sample size, and the absence of longitudinal follow-up to assess long-term outcomes. All participants were from a Western population, and the lack of ethnic diversity may limit the generalizability of the findings to other populations. Although formal reliability metrics were not calculated, all measurements were performed by 2 trained ophthalmologists using a standardized protocol to minimize variability.

STRENGTHS:

The groups were selected with consideration of parameters that could affect the interpretation of results, such as patient age, refractive error, axial length, and visual acuity. Automated measurements were used to reduce potential errors associated with manual methods. For CT assessment, 3 repeated measurements were performed and averaged. Evaluation of the retina and choroid was conducted via multimodal imaging, including OCT and OCTA, allowing comprehensive assessment of both structural and vascular changes within the eye.

Conclusions

Our study demonstrates that patients with OSA exhibit subtle microvascular alterations in the retina, including FAZ enlargement, as well as a trend toward increased CT. Retinal vessel density remained relatively preserved, suggesting that some vascular parameters are resilient in moderate to severe OSA. These findings contribute to the understanding of the ocular effects of chronic intermittent hypoxia and highlight the potential of OCT and OCTA as noninvasive tools for detecting early retinal and choroidal changes. Further longitudinal studies are warranted to explore mechanisms underlying these alterations and evaluate their potential roles in early screening, risk stratification, and progression monitoring of OSA.