Abstract
Coronary artery disease (CAD) is a leading cause of cardiovascular mortality worldwide, and its early diagnosis is essential for prevention and treatment. Emerging evidence from ocular imaging suggests that structural and functional alterations in the retinal vasculature may mirror systemic vascular changes, offering a promising avenue for the early identification of cardiovascular conditions such as CAD. Among these techniques, OCTA stands out as a non-invasive, high-resolution modality capable of capturing detailed microvascular architecture and quantifying retinal blood flow dynamics. In this study, we analyzed OCTA images from 747 participants including 332 patients with CAD and 415 controls to extract retinal microvascular parameters and evaluate their associations with disease status. The results revealed that patients with CAD exhibited significantly lower retinal vessel density, vessel length density, tortuosity, and vascular bifurcation complexity on OCTA compared to controls, particularly in the left eye. This suggests possible lateral asymmetry in microvascular responses associated with CAD. Overall, we highlight the potential of OCTA-derived retinal biomarkers in supporting the early diagnosis of CAD, providing a standard tool for future clinical services and research. Biomarkers in retinal OCTA images can provide useful information for clinical decision making and diagnosis of CAD.