Neural constraints on visual acuity in proliferative diabetic retinopathy.

PURPOSE: Visual acuity (VA) in normally sighted individuals is highly correlated with equivalent intrinsic blur, a measure of the amount of blur within the visual system that is generated by optical and neural sources. This study assessed the extent to which VA, equivalent intrinsic blur, optical blur, and neural blur are abnormal in subjects with proliferative diabetic retinopathy (PDR) and characterized the relationships among these parameters.
METHODS: Best-corrected VA of 10 subjects with PDR (ages 25 to 68) and 10 normally sighted individuals (ages 46 to 63) was measured for tumbling E optotypes. The Es were either unblurred or blurred through convolution with Gaussian functions of different widths. Values of equivalent intrinsic blur (σ(int)) and unblurred VA (MAR0) were derived using a standard model. Optical blur (σ(opt)), a measure of blur generated by higher-order aberrations, was quantified using Shack-Hartmann aberrometry. An index of neural blur (η) was defined as 1--σ(opt)/σ(int), which represents the remaining blur once the contributions of σ(opt) to σ(int) have been accounted for.
RESULTS: Log MAR0 and log σ(int) were correlated significantly (r = 0.98, p < 0.05) for the PDR subjects and the values of these parameters ranged from normal to more than a factor of 2 above the upper limit of normal. In comparison, log MAR measured for the most blurred E was elevated by a relatively small amount for all PDR subjects and was not correlated significantly with log σ(int) (r = 0.40, p = 0.25). MAR0, σ(int), and η differed significantly between the PDR subjects and the controls (all p < 0.05) but σ(opt) did not (p = 0.50).
CONCLUSIONS: Subjects with PDR and VA loss had higher than normal equivalent intrinsic blur that resulted primarily from neural blur elevations, suggesting that neural blur is an important factor that limits VA in these patients.