Evidence for photoreceptor changes in patients with diabetic retinopathy.

Purpose: To determine whether the rod and cone photoreceptors are affected in patients with diabetic retinopathy.
Methods: Twelve patients with diabetes and varying levels of retinopathy and nine age-similar control observers participated in this study. Two-color (500 versus 650 nm) dark-adapted thresholds were measured as a function of retinal eccentricity. Full-field flash electroretinograms were obtained using brief, high-intensity flashes. Dark-adapted rod-isolated (Wratten 47B filter) and light-adapted cone-isolated (Wratten 26 filter) electroretinographic responses were measured as a function of flash intensity. The a-wave data were fitted with a model based on photopigment transduction to obtain values for the parameters of Rmax (the maximal response) and log S (sensitivity). Standard clinical 30-Hz flicker electroretinographic responses were also measured.
Results: Psychophysically measured dark-adapted thresholds were elevated primarily at eccentricities of 5 degrees and 10 degrees from the fovea. Analysis of rod and cone a-wave data showed that Rmax was normal in most of the patients, but log S was reduced. Analysis of b-wave and oscillatory potential parameters showed rod and cone postreceptoral abnormalities, including changes in the rod-isolated semisaturation constant (log k), cone-mediated 30-Hz flicker, and cone-isolated oscillatory potentials. The electrophysiological results were not significantly correlated with blood glucose or glycosylated hemoglobin level.
Conclusions: The results provide evidence for rod and cone receptoral and postreceptoral deficits in patients with diabetic retinopathy. The photoreceptor changes are primarily in the log S (sensitivity) parameter and are attributed to transduction abnormalities.