Quantification of retinal nerve fiber layer thickness reduction associated with a relative afferent pupillary defect.

Background: A relative afferent pupillary defect (RAPD) is known to develop only when more than 25% of retinal ganglion cells are ablated in monkeys' eyes. However, there was no prior study to estimate biometrically the degree of retinal nerve fiber layer (RNFL) thickness reduction leading to the development of RAPD in live human eyes. The purpose of this study was to examine the correlation between the amount of RNFL thickness reduction and the depth of a clinically detectable RAPD in patients with unilateral optic atrophy.

Methods: Enrolled were 20 patients with optic atrophy of various etiologies. We quantified RAPD by performing the swinging flashlight test with log-scaled neutral density filters placed over the unaffected eye. Average RNFL thickness was measured by OCT3000 with the average RNFL thickness program. Linear regression analysis was used in assessing the relationship between RAPD and the ratio of affected to unaffected average RNFL thickness.

Results: The mean of average RNFL thickness was 95.6+/-17.3 microm in the unaffected eyes and 50.7+/-19.3 microm in the affected eyes (P<0.001). Regression analysis between RAPD and the ratio of affected to unaffected average RNFL thickness revealed a correlation coefficient R2=0.48 (P=0.0007). The regression line intersected the y-axis at 0.77.

Conclusions: RAPD was not clinically detected until at least approximately 25% of the retinal nerve fibers were lost when compared with the unaffected eyes. Substantial retinal ganglion cell damage is required for the development of RAPD.