Relationship Between Foveal Avascular Zone Measured by Optical Coherence Tomography Angiography and Visual Acuity in...

Diabetic retinopathy (DR) is a major cause of vision loss, with early microvascular changes often preceding visual deterioration. Optical coherence tomography angiography (OCTA) allows non-invasive imaging of the foveal avascular zone (FAZ), a potential biomarker for visual function. However, the association between FAZ metrics and best-corrected visual acuity (BCVA) remains unclear. Prior studies have reported variable findings regarding the diagnostic relevance of FAZ parameters, partly attributable to differences in OCTA imaging techniques, segmentation strategies, and analytic methodologies.

To systematically review the association between OCTA-derived FAZ parameters (area, perimeter, and circularity) and BCVA in patients with DR.

This review followed PRISMA 2020 guidelines. Four databases (PubMed, Scopus, EbscoHost, and ScienceDirect) were systematically searched through April 2025. Studies were eligible if they examined the relationship between FAZ metrics and BCVA in diabetic patients using quantitative analysis. Risk of bias was assessed using the Joanna Briggs Institute (JBI) tools. A narrative synthesis was performed.

Six studies (1 prospective cohort and 5 cross-sectional) involving 651 diabetic and 100 control eyes were included. FAZ area, particularly in the deep vascular plexus (DVP), showed the strongest and most consistent negative correlation with BCVA. In contrast, FAZ circularity and perimeter demonstrated weak and inconsistent associations with visual acuity. Heterogeneity in OCTA devices, scan protocols, study design, and outcome measurements limited meta-analysis.

FAZ area, particularly in the DVP, shows the most consistent association with visual acuity in DR, while circularity and perimeter demonstrate variable and limited standalone value. However, current evidence is limited by methodological variability and high heterogeneity across studies. Standardization of study methods, OCTA acquisition, and segmentation are essential to enable reliable translational and clinical application of FAZ-based biomarkers. Hybrid, multimodal, and artificial intelligence-based approaches may enhance future longitudinal prediction models.