Diagnostic value of automated retinal microcirculation analysis in patients with type 1 diabetes mellitus

Relevance Automated OCT-A image analysis can significantly improve diagnostics due to objective quantitative assessment of vascular biomarkers, which ensures early detection of changes even before the onset of clinical symptoms [1]. Objective To evaluate the informativeness of clinical and visualized retinal biomarkers in patients with type 1 diabetes using a new algorithm for automated OCT-A image analysis. Material and methods The study included 188 patients, conducted using automated software for quantitative analysis of OCT-A images. A correlation analysis was carried out between the relationship of these biomarkers with age, diabetes duration, HbA1c level, DR stage and best-corrected visual acuity (BCVA) [2, 3]. Results An increase in the FAZ area was found with DR progression at the levels of the superficial (K = 0.788; p = 0) and deep (K = 0.764; p = 0.03) capillary plexuses. A decrease in vascular density (VAD) was shown at both levels with disease progression. An inverse correlation was found between the HbA1c level and vascular density in the SCP (K = –0.636; p = 0) and DCP (K = –0.619; p = 0.05). An important finding was the identification of a relationship between the deterioration of BCVA and an increase in the FAZ area (K = –0.728; p = 0), as well as a decrease in VSD (K = –0.754; p = 0) at the level of the deep capillary plexus. Conclusion Automated OCT-A analysis allows for an objective assessment of changes in retinal microcirculation in patients with type 1 diabetes. The identified biomarkers, such as FAZ area and vascular density, have high diagnostic value for early detection of DR. Keywords: biomarkers; diabetes mellitus; diabetic retinopathy; image analysis; OCT angiography