Diabetic vascular hyperpermeability: optical coherence tomography angiography and functional loss assessments of relationships among retinal vasculature changes

Our study assessed the influence of vascular permeability on vascular flow density (FD)-correlated retinal sensitivity (RS) in DR. In this cross-sectional, prospective, consecutive study, RS

in the extrafoveal macula of DR patient was measured by microperimetry. FD was measured in the total, superficial, and deep capillary plexus layers (TCP, SCP, and DCP) by optical coherence

tomography angiography. All measurement points were classified into four categories according to intensity of fluorescein leakage and FD, and the RS reduction was compared. A stratified

analysis by retinal thickness (RT) was also performed. Fourteen eyes (14 patients) were enrolled. FDs at 207 RS measurement points were analyzable. For TCP, SCP and DCP, the leakage did not

decrease RS at points where FD was maintained. The greater the leakage, the smaller the RS reduction at points with low FD in TCP (P = .020). Points with high leakage showed a significant 

smaller RS reduction than points with low leakage (P = .001 for TCP, P = .040 for SCP, and P = .046 for DCP) only in areas with low RT and low FD. Our results suggested that vascular

hyperpermeability may inhibit the RS reduction in the non-edematous ischemic diabetic retina.

Diabetic retinopathy (DR) is a major microvascular complication of diabetes mellitus and a leading cause of vision loss in working-age adults1,2. Vascular abnormalities, including

hyperpermeability-associated retinal edema and ischemia due to capillary loss, are involved in the pathogenesis of DR3, but how these abnormalities affect visual function in patients is

unclear.

Retinal sensitivity (RS) is a common measure of visual function4. Fluorescein angiography (FA) studies have demonstrated significant reduction in RS in non-perfused areas5,6, suggesting that

perfusion status affects RS. Unlike FA, optical coherence tomography angiography (OCTA) allows quantification of flow density (FD)7: several studies have shown that FD is associated with RS

in DR8,9.

Enhanced vascular permeability can cause retinal edema, followed by retinal neurodegeneration10. OCTA is a non-invasive alternative to FA; however, OCTA cannot identify vascular leakage,

which can occur regardless of perfusion11. Following a report of RS reduction in areas with greater leakage12, we hypothesized that the greatest RS reduction occurs in areas with reduced

blood perfusion and enhanced vascular permeability. To test our hypothesis, we used microperimetry to measure RS in patients with DR. We divided the measurement points into four groups

according to fluorescein leakage intensity and FD value, then compared RS reduction values among groups. Contrary to our expectations, the area with leakage showed significantly less RS

reduction in the area with reduced FD. Stratified analysis by retinal thickness (RT) showed that leakage could maintain RS only in areas where RT and FD were both low. Although prolonged

edema due to enhanced vascular permeability promotes impairment of visual function13, our results suggest that leakage may inhibit the RS reduction in the non-edematous ischemic retina in

patients with DR.

This study was conducted in accordance with the principles of the Declaration of Helsinki and approved by the Institutional Review Board of Kyushu University Hospital (28-473, UMIN

000028656); written informed consent was obtained from all healthy volunteers as well as patients prior to participation in the study.

This cross-sectional, prospective, comparative, consecutive case series included 14 eyes of 14 patients with DR who visited the Department of Ophthalmology, Kyushu University Hospital,

between February 2017 and January 2020. All patients underwent microperimetry, FA, OCT and OCTA on the same day. Given the poor fixation in patients with macular edema14, patients with

center-involving macular edema (central macular thickness ≥ 300 μm on OCT) were excluded. Anti-vascular endothelial growth factor (VEGF) therapy (intravitreal injection of ranibizumab 0.5 

mg) is administered to treat diabetic macular edema in our department. Because anti-VEGF agents influence retinal vascular permeability15, patients who had received anti-VEGF treatment in

the previous 34 days were excluded16. Other exclusion criteria were: active proliferative DR (PDR), other macular diseases, glaucoma, or ocular surgery within the previous 6 months. Six

healthy volunteers were included for determination of reference RS values.

An MP-3 microperimeter (Nidek Co. Ltd, Aichi, Japan) was used for RS measurements. The stimulus dynamic range was 0–34 dB. The background luminance was 31.4 asb, maximum luminance was 10,000

asb, and the stimulus was Goldmann size III. RS was measured in the upper or lower temporal area of the macula (a 3 × 3-mm square with 25 points/patient; see Step 1 in Figure, Supplemental

Figure 1, which demonstrates FD calculation and leakage evaluation). Only reliable data (false-positive and false-negative rates