Subclinical Coronary Atherosclerosis and Retinal Optical Coherence Tomography Angiography

Importance Systemic disease burden is well reflected in the retinal microvasculature, but it is not established whether optical coherence tomography angiography (OCTA) parameters can reliably reflect coronary atherosclerosis. Objective To assess the association of reduced retinal vascular density and subclinical coronary atherosclerosis in asymptomatic individuals. Design, Setting, and Participants This cross-sectional cohort study included asymptomatic individuals with elevated cardiovascular risk who self-referred for a health screening program involving coronary computed tomography angiography (CTA) at Asan Medical Center in Seoul, South Korea, and subsequently underwent OCTA as part of ophthalmic evaluations. This study was conducted from October 2015 to December 2020, and data analysis was conducted from January 2021 to May 2025. Main Outcomes and Measures The primary outcome was the association between OCTA parameters and coronary CTA parameters. Results A total of 1286 eyes from 1286 participants were analyzed (mean [SD] age, 64.2 [9.9] years; 482 female participants [37.5%]). Coronary artery calcium score (CACS), presence of plaque and its subtypes, obstructive coronary artery disease (CAD), severe CAD, segment stenosis score (SSS), and segment involvement score (SIS) were significantly increased across the quartiles of superficial and deep parafoveal vascular density (PFVD). When modeled as continuous variables, superficial capillary plexus (SCP) and deep capillary plexus (DCP) PFVD were the most significant metrics and correlated with CACS, number of coronary vessels involved, SSS, and SIS. The lowest quartile of SCP PFVD was associated with significantly higher odds of obstructive CAD (adjusted odds ratio [aOR], 2.91; 95% CI, 1.83-4.73), severe CAD (aOR, 3.30; 95% CI, 1.55-7.91), and elevated SSS and SIS scores compared to the highest quartile. Similar but attenuated associations were observed for DCP PFVD. Continuous variable analysis for ORs for unit increase supported a linear inverse association between PFVD and CAD burden. Incorporating PFVD into models with cardiovascular risk factors improved area under the curve (AUC) for identifying severe CAD (AUC, 0.79; 95% CI, 0.75-0.82), obstructive CAD (AUC, 0.78; 95% CI, 0.76-0.81), and SSS of 10 or higher (AUC, 0.77; 95% CI, 0.74-0.80), with SCP yielding superior performance over DCP. Conclusions and Relevance In this cross-sectional cohort study, reduced retinal PFVD was independently associated with subclinical coronary atherosclerosis in a population with elevated vascular risk. In this context, decreased PFVD may reflect greater subclinical coronary atherosclerotic burden and help identify individuals who could benefit from further coronary evaluation, beyond traditional risk factors.