Abstract
Purpose: To evaluate layer-specific relationships between retinal architecture, measured by optical coherence tomography (OCT), and cardiovascular health (CVH).
Methods: We analyzed data from 26,230 UK Biobank participants with high-quality spectral-domain OCT images and complete CVH assessments. CVH was quantified using the American Heart Association’s Life’s Essential 8 (LE8) score. Multivariable linear regression models assessed associations between thickness of eight retinal layers and CVH. The diagnostic performance of retinal layer thickness for identifying poor CVH (LE8 score <50) was evaluated using the area under the receiver operating characteristic curve (AUC).
Results: Each 5 µm increase in ganglion cell complex (β = 0.141; 95% confidence interval [CI], 0.068-0.213), retinal nerve fiber layer (β = 0.225; 95% CI, 0.093-0.357), ganglion cell-inner plexiform layer (β = 0.175; 95% CI, 0.062-0.288), photoreceptor segments (PS; β = 0.753; 95% CI, 0.590-0.917), and macular thickness (β = 0.133; 95% CI, 0.086-0.179) was independently associated with better CVH (all P ≤ 0.002). The addition of retinal layers to a baseline model of age and sex slightly improved the diagnostic performance for poor CVH. Among individual retinal layers, the PS showed the strongest incremental discriminative ability (AUC = 0.632; 95% CI, 0.608-0.657). The highest overall performance was achieved by the model incorporating the total retinal thickness in the macular region (AUC = 0.662; 95% CI, 0.638-0.687).
Conclusions: Reduced thickness in both neuroretinal layers and PS demonstrates a significant but weak association with impaired CVH. Although these findings reinforce a structural connection between retinal integrity and cardiovascular status, the modest discriminative performance indicates that retinal structure alone may have limited value for diagnostics or screening.