A new view of diabetic retinopathy: a neurodegenerative disease of the eye

Abstract

Diabetic retinopathy (DR) is a common complication of diabetes and a leading cause of legal blindness in working-age adults. The clinical hallmarks of DR include increased vascular permeability, leading to edema, and endothelial cell proliferation. Much of the research effort has been focused on vascular changes, but it is becoming apparent that other degenerative changes occur beyond the vascular cells of the retina. These include increased apoptosis, glial cell reactivity, microglial activation, and altered glutamate metabolism. When occurring together, these changes may be considered as neurodegenerative and could explain some of the functional deficits in vision that begin soon after the onset of diabetes.

This review will present the current evidence that neurodegeneration of the retina is a critical component of DR. There are two basic hypotheses that account for loss of cells in the neural retina. First, the loss of blood–retinal barrier integrity, which initially manifests as an increase in vascular permeability, causes a failure to control the composition of the extracellular fluid in the retina, which in turn leads to edema and neuronal cell loss. Alternatively, diabetes has a direct effect on metabolism within the neural retina, leading to an increase in apoptosis, which in turn causes breakdown of the blood–retinal barrier. It is not clear which hypothesis will be found to be correct, and, in fact, it is likely that vascular permeability and neuronal apoptosis are closely linked components of DR. However, the gradual loss of neurons suggests that progress of the disease is ultimately irreversible, since these cells cannot usually be replaced. In light of this possibility, new treatments for DR should be preventive in nature, being implemented before overt clinical symptoms develop. While vascular permeability is the target that is primarily considered for new treatments of DR, evidence presented here suggests that apoptosis of neurons is also an essential target for pharmacological studies. The vision of people with diabetes will be protected only when we have discovered a means to prevent the gradual but constant loss of neurons within the inner retina.

Introduction

Neurodegeneration is recognized as a pivotal feature of many diseases of the central nervous system. It is also recognized as a component of some retinal diseases including glaucoma and retinitis pigmentosa Kerrigan et al., 1997, Baumgartner, 2000, but has been largely overlooked in diabetic retinopathy (DR) despite the degree of vision loss associated with this disease. DR is a common complication of diabetes that is present to some degree in nearly all individuals who have had diabetes for more than 15 years, regardless of whether the diabetes is due to loss of insulin secretion (Type I) or to insulin resistance (Type II) (Klein et al., 1984). It is the leading cause of new cases of legal blindness (defined as vision worse than 20/200) in Americans between the ages of 20 and 74 years (Klein and Klein, 1995).

One of the earliest clinically detectable signs of DR is increased vascular permeability, due to a breakdown in the blood–retinal barrier, which causes macular edema (Cunha-Vaz et al., 1975). This is followed later by the development of vascular microaneurysms, deposition of lipoprotein exudates called drusen, and finally vascular proliferation (Aiello et al., 1998). Of these clinical features, macular edema is most closely correlated with the degree of vision loss (Moss et al., 1988), but the actual relationship between vascular permeability and loss of neural function is not known.

The current treatment for DR is photocoagulation, in which laser flashes are used to burn the areas of retina containing leaking blood vessels (L'Esperance, 1968). This treatment arrests the course of the disease in about 50% of cases but cannot reverse vision loss and must be repeated in many individuals (Blankenship, 1991). In actuality, laser photocoagulation can be likened to amputation because it destroys the diseased parts of the retina in an attempt to rescue less affected regions. Since the development of laser surgery for DR, there have been no major advances in treatment for the disease, despite numerous clinical trials on a variety of drugs including calcium dobesylate, aspirin, aldose reductase inhibitors, and antihistamines Daubresse et al., 1977, Anonymous, 1991, Arauz-Pacheco et al., 1992, Gardner et al., 1995.

Much research has focused on vascular endothelial growth factor (VEGF) because this potent biogenic permeability factor is elevated in the vitreous of people with DR and in animal models of diabetes Aiello et al., 1994, Ambati et al., 1997, Sone et al., 1997, and is sufficient to increase both vascular proliferation and permeability (Aiello et al., 1997). VEGF is produced by the neurons and glial cells of the retina in response to hypoxia, but it is not clear if this is the signal that triggers its expression in diabetes Neufeld et al., 1999, Murata et al., 1996, Amin et al., 1997.

The vascular changes in DR have been given much attention but it is becoming clear that the retinal parenchyma is also altered by diabetes. These changes are reminiscent of chronic neurodegenerative diseases, and include increased apoptosis, a gradual loss of neurons, altered expression of glial fibrillary acidic protein (GFAP) in the astrocytes and Müller cells, and activation of microglia. The following review will attempt to summarize the current evidence supporting the hypothesis that chronic neurodegeneration is a component of DR.