[Preretinopathic changes in the oscillatory potential in diabetic retina: interpretation and significance]

1) The earliest electroretinographic manifestation of diabetic retina is a selective change in the oscillatory potential (OP), which originates in postsynaptic retinal neuronal circuits. "Diabetic intraretinal neuropathy" precedes angiopathic retinopathy. 2) The peak latency and the amplitude of the OP are significantly correlated with the psychophysical contrast threshold for motion perception (CTMP) measured with our original device. 3) The CTMP is not correlated with the conventional static contrast threshold measured with a commercially available chart (VCTS-6000). 4) The CTMP is beyond the upper limit in nondiabetic control eyes in 135 out of 162 diabetic eyes at stage 0 (no ophthalmoscopic diabetic retinopathy) and all retinopathic eyes at stages AI (11 eyes) and AII (21 eyes). The CTMP test, which is completely non-invasive and easy to perform, could be useful for mass-screening of early diabetic retinal dysfuction. 5) The amplitude of the OP is enhanced by dopamine and nomifensine (a potent dopamine-uptake blocker), and diminished by haloperidol (a potent antagonist to dopamine D1-D2 receptors). Intrinsic dopamine release in the retina would regulate the amplitude of the OP. 6) The intraretinal content of dopamine decreases in streptozotocin (STZ)-induced diabetic rats where the OP amplitude is diminished, and increases in eyes where the OP amplitude is enhanced at the early stage in spontaneously diabetic (OLETF) rats. The changes in the OP amplitude in STZ-induced and OLETF diabetic rats can not be accounted for by the intraretinal content of gamma-aminobutyric acid (GABA), glycine, glutamate, aspartate or taurine. The changes in the OP amplitude in the diabetic retina are at least partly due to dysfunction of dopaminergic retinal neurons. 7) The peak latency prolongation and the amplitude diminution of the OP in STZ-induced diabetic rats are normalized after insulin treatment, while vitreous fluorescein concentration by vitreous fluorophotometry is not restored by insulin. Thus, the OP changes in STZ-induced diabetic rats are not due to STZ toxicity but to diabetes per se, and are not attributable to blood-ocular barrier disruption as revealed by vitreous fluorophotometry. Retinal neuronal dysfunction revealed by the OP changes at the early preretinopathic stage in STZ-induced diabetic rats is reversible after insulin treatment. 8) The OP changes in OLETF rats are prevented by decreasing food intake by 30% to maintain their blood glucose level and body weight normal. Even in highly genetically diabetic animals, diet therapy starting at an early stage of life inhibits the diabetic OP changes. 9) The amplitude of the bicarbonate response from the retinal pigment epithelium diminishes even at the preretinopathic stage, and is significanly correlated with the peak latency and the amplitude of the OP. Diabetic retinal pigment epitheliopathy as well as retinal neuropathy takes place prior to angiopathic retinopathy in the diabetic retina.