Purpose
To understand the retinal non-neuronal cells that may contribute to diabetic macular edema (DME), and to determine possible cellular targets for better management of this condition.
Methods
A retrospective review of consecutive identified patients with DME who had been followed at Drexel University College of Medicine Retina Clinic from July 2012 to June 2013 (45 eyes) was conducted. These patients were further classified into two groups based on the type of diabetic retinopathy (DR) including nonproliferative DR (NPDR 25 eyes) and proliferative DR (PDR 20 eyes). For each patient, a dilated fundus examination, fundus photography, SD-OCT (Cirrus HD-OCT Model 4000), and fluorescein angiography (FA) were performed. Two independent raters experienced in analyzing retina images evaluated the OCT and FA.
Results
The imaging analysis demonstrates changes of multiple non-neuronal cells that may contribute to the formation of DME. This includes dysfunction of endothelial cells and pericytes (E & P), displayed by the breakdown of the inner blood-retinal barrier (BRB), Muller cells, presented as vertical elongation of the cell body on OCT but without significant angiographic leakage on FA, RPE cells demonstrated by breakdown of the outer BRB and microglial cells signified by multiple microcysts (in size of 20 to 70 µm) in the inner nuclear layer (INL). In this study, microcystic edema in INL was seen in 26 of 45 (57.8%) eyes. The contribution of the above-mentioned cells in DME is listed as follows: E & P in 44/45 (97.8%) eyes, Muller cells in 10/45 (22%) eyes and RPE cells in 8/45 (17%) eyes. For the non-neuronal cellular contribution to DME, there is no significant difference between NPDR and PDR groups.
Conclusions
A significantly high prevalence of microcystic pattern of edema in INL, indicating microglial dysfunction has prompted us to pay attention to microglial cells as a pathogenic contribution to DME. Microglial cells are known as the primary resident immune cell of the retina. The evolution of the OCT has made it possible to detect microglial cell dysfunction as a pathognomonic sign in early stages of DME. Therefore, new therapies for targeting dysfunctional microglial cells in early DME merit further study.
Keywords: 499 diabetic retinopathy •
550 imaging/image analysis: clinical •
595 microglia