Combining OCTA Metrics from Retinal, Choroidal Microvasculature May Improve DR Detection

Published on: 

A recent analysis suggests that incorporating retinal and choroidal microvascular metrics may improve the identification of eyes with no diabetic retinopathy from those with DR.

The evaluation of optical coherence tomography angiography (OCTA) parameters from both the retinal and choroidal microvasculature could improve the discrimination of early-stage diabetic retinopathy (DR), according to new research.1

Investigators, led by Gavin Tan, MBBS, PhD, Singapore Eye Research Institute, indicated a classification model including both retinal and choroidal microvasculature 3×3 mm scans significantly improved the discrimination between DR and no DR compared with each parameter separately (P = .029).

“We found that flow deficit density (FDD) with a size threshold better stratified DR severity, and our findings suggest that incorporating retinal and choroidal microvascular metrics improves the discriminative power of our models identifying eyes with no DR from those with DR,” wrote the investigative team.

Current classification and staging of DR center around examining changes in the retinal vasculatures, due to being easily observable on ophthalmoscopy and color fundus photography. But a new understanding of the pathophysiology of DR and new imaging technology may require an update and revision of these classification systems. OCTA has been used to detect microvascular abnormalities associated with DR and may serve as an aid for the classification of DR, particularly being a non-invasive tool.2

On the other hand, choroidal vascular changes have been described in eyes with DR but are infrequently used to determine the severity of DR. Newer advances in swept-source OCTA can enable choriocapillaris visualization, where the choriocapillaris can be characterized by a dense capillary network interspace by flow density. Recent literature has shown choriocapillaris flow impairment in patients with diabetes with and without commercial OCTA systems.

Nevertheless, it is unknown whether combining measures of the retinal and choroidal microvasculature would increase the discriminative ability of OCTA for diabetes and DR. In this cross-sectional case–control study, Tan and colleagues compared SS-OCTA measures in 3 groups to evaluate the discriminative power of OCTA metrics on the presence of diabetes, the presence of DR, and the need for referral: group 1 (no diabetes vs. diabetes and no DR), group 2 (no DR vs. any DR), and group 3 (non-proliferative DR [NPDR] vs. proliferative DR [PDR]).

The study was performed from April 2018 - July 2019 and included patients aged ≥21 years old with type 2 diabetes (T2D) for a duration under 5 years, while the non-diabetes population included patients with no known diabetes. The severity of DR was assessed using 2 field fundus photography and Early Treatment Diabetic Retinopathy Score (ETDRS) grading scale. OCTA parameters from retinal vasculature, fovea avascular zone (FAZ), and choriocapillaris were evaluated from 3 x 3 mm2 fovea-centered scans.

A total of 35 eyes from 27 participants with no diabetes and 132 eyes from 75 participants with diabetes were included in the analysis. The severity of DR was classified into no DR (n = 62 eyes), NPDR (n = 51 eyes), and PDR (n = 19 eyes), and higher HbA1c levels were associated with the severity of DR (P = .001), but other variables saw no significant differences among groups.

However, investigators found systolic blood pressure was higher in the PDR group than the no PDR group (P = .001) and the cholesterol ratio (total/high-density lipoprotein) was higher in the PDR group than the no DR or NPDR group (P = .008). As anticipated, all retinal vascular parameters, FAZ parameters, and choriocapillaris parameters were associated with DR stages (P <.01), except for deep plexus FAZ area (P = .619).

Moreover, FDD increased with DR severities and setting a size selectivity on FDD could better stratify the severity of DR, according to Tan and colleagues. The analysis showed choriocapillaris parameters allowed for better discrimination between no diabetes versus diabetes and no DR group, compared with retinal parameters (area under the ROC curve, 0.954 vs. 0.821; P = .006).

“In summary, evaluating SS-OCTA parameters from retinal and choroidal microvasculature in 3×3 mm FOV improves the discrimination in early-stage DR, where the predominant changes happen in the choriocapillaris, but not in late-stage DR,” investigators wrote. “It might open on differential therapeutic target sites and potential mechanisms depending on the stage of severity.”


  1. Tan B, Lim NA, Tan R, et al. Combining retinal and choroidal microvascular metrics improves discriminative power for diabetic retinopathy [published online ahead of print, 2022 Feb 9]. Br J Ophthalmol. 2022;bjophthalmol-2021-319739. doi:10.1136/bjophthalmol-2021-319739
  2. Sun JK, Aiello LP, Abràmoff MD, et al. Updating the Staging System for Diabetic Retinal Disease. Ophthalmology. 2021;128(4):490-493. doi:10.1016/j.ophtha.2020.10.008