Objective and Subjective Refractive Error Measurements in Monkeys

PURPOSE: To better understand the functional significance of refractive-error measures obtained using common objective methods in laboratory animals, we compared objective and subjective measures of refractive error in adolescent rhesus monkeys. METHODS: The subjects were 20 adolescent monkeys. Spherical-equivalent spectacle-plane refractive corrections were measured by retinoscopy and autorefraction while the animals were cyclopleged and anesthetized. The eye's axial dimensions were measured by A-Scan ultrasonography. Subjective measures of the eye's refractive state, with and without cycloplegia, were obtained using psychophysical methods. Specifically, we measured spatial contrast sensitivity as a function of spectacle lens power for relatively high spatial frequency gratings. The lens power that produced the highest contrast sensitivity was taken as the subjective refraction. RESULTS: Retinoscopy and autorefraction consistently yielded higher amounts of hyperopia relative to subjective measurements obtained with or without cycloplegia. The subjective refractions were not affected by cycloplegia and on average were 1.42 ± 0.61 D and 1.24 ± 0.62 D less hyperopic than the retinoscopy and autorefraction measurements, respectively. Repeating the retinoscopy and subjective measurements through 3 mm artificial pupils produced similar differences. CONCLUSIONS: The results show that commonly used objective methods for assessing refractive errors in monkeys significantly overestimate the degree of hyperopia. It is likely that multiple factors contributed to the hyperopic bias associated with these objective measurements. However, the magnitude of the hyperopic bias was in general agreement with the "small-eye artifact" of retinoscopy.