Abstract
PURPOSE: To provide insight into rod pathway dysfunction in early-stage diabetic retinopathy (DR) by measuring dark-adapted ERGs and pupillary light reflexes (PLRs) across a broad range of stimulus luminance. METHODS: Seventeen diabetics with no clinically apparent DR (NDR), 17 with mild nonproliferative DR (MDR), and 15 nondiabetic controls participated. Dark-adapted, full-field ERGs and PLRs were obtained. Achromatic (-4 to 1 log cd-s-m-2) and long-wavelength (-4.0 to 2.6 log cd/m2) flashes were used for ERG and pupillometry, respectively. The b-wave amplitudes and pupil diameters were fit with Naka-Rushton functions to obtain (1) maximum b-wave amplitude (Vmax), (2) maximum PLR (Pmax), (3) b-wave sensitivity (kb), and (4) PLR sensitivity (kp). RESULTS: ERG a-wave amplitude was reduced (0.13 log µV averaged across stimulus luminance) in DR compared with the controls, but this was not statistically significant (F = 1.41; P = 0.25). ERG Vmax did not significantly differ among groups (F = 2.20, P = 0.12), whereas kb was elevated (reduced sensitivity) for both groups (both t > 2.40; P < 0.02). Pupil Pmax was reduced in MDR (t = 2.83, P = 0.01), but not NDR (t = 0.99, P = 0.33). Pupil kb was significantly elevated in NDR and MDR (both t > 2.1; P < 0.04). CONCLUSIONS: Reduced b-wave amplitude may largely be accounted for by the reduced a-wave amplitude. By contrast, pupil sensitivity loss greatly exceeded the b-wave sensitivity loss, suggesting sites of abnormality beyond the bipolar cells contribute to pupil response deficits in diabetics.