Editorial: Applications of optical coherence tomography angiography (OCTA) in ocular complications of diabetes mellitus

A subset of evidence indicate that the imaging parameters of macular OCTA correlate with the severity of DR. As the disease progress, higher grade of capillary density comprise occurs in retinal capillary layers. This is also observed in the peripapillary retinal nerve fiber layer (pRNFL), where the average pRNFL thickness and vascular density is reduced in diabetic patients, especially in eyes with non-proliferative DR (NPDR), compared to age-matched healthy objects. pRNFL capillary density had relatively high distinguishing ability to detect DR from healthy eyes.(5) (https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2024.1404157/full) Recent studies investigating the progression of DR, has highlighted the vital predicting role of peripheral non perfusion in future complications. Ultrawide-field fluorescein angiography is the diagnostic tool that provides valuable insights to the degree of retinal ischemia. However, it is invasive with limited access in routine ophthalmology clinics. Guo et al. showed that ultrawidefield OCTA reveal the non-perfusion areas within its field of view, underscoring its potential role as a surrogate for an invasive diagnostic procedure.(6) (https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2025.1485311/full) More extensive capillary nonperfusion in both superficial and deep retinal capillary layers has been suggested in eyes with DME. The current finding of reduced capillary vascular density, and altered choriocapillaris flow area accentuates the importance of retinal ischemia in the development of DME. (7) (https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2023.1275200/full) The future longitudinal investigation, utilizing the OCTA parameters, can help predict eyes at risk for DME, and providing guide to devise prophylactic plans.The status of retinal vascular density could expand our understanding with some disease processes. Corneal microdots are highly reflective spherical structures found throughout the entire depth of cornea. OCTA could shed light on the mechanisms of microdots development. Low grade inflammation and relative ischemia could be the initiating factors, as these microdots appear more in corneas with diabetic retinopathy, and their prevalence had correlation with OCTA parameters.( 8) (https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2025.1481018/full)The advent of artificial intelligence heralds future advancements in the care of patients with DR. Herein, we included a meta-analysis reporting the accuracy deep learning models on OCT and retinal imaging in the diagnosis of DR.( 6) (https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2025.1485311/full) These models showed promising performance; however significant challenges are still remained to be addressed. Generating standardized datasets of imaging are a crucial infrastructural issue. Attention to include high quality OCTA images, with accepted image processing techniques, and developing a platform to uniformly compare data from various devices are of paramount importance that ensures high quality feed for deep learnings model. Subtle ocular characteristics should be considered in accurate reporting of OCTA images, including axial length, refractive error, or even mydrias.(9) (https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2024.1292255/full) Although presence of mydrias did not affect vascular density measurement in images with comparable quality score, this simple intervention can significantly affect the overall image quality in retinal clinic.The current topic, that explores diverse aspects of OCTA application in DR, adds an insight into the vast applicability of this imaging. Future investigations will complement the promising utility of OCTA as a non-invasive and accessible imaging modality in DR patients.