Measuring visual acuity--mesopic or photopic conditions, and high or low contrast letters?

To develop single-valued wavefront aberration metrics that correlate strongly with visual performance. The purpose of this study is to explore whether photopic high contrast visual acuity (VA) is an appropriate visual performance reference and whether mesopic and/or low contrast testing provides any advantage.

Subjects from the Texas Investigation of Cataract Optics study (N = 148) ranged in age from 21.6 to 83.8 years and from clear lens to dense nuclear cataract. Visual acuity was measured under four conditions: photopic high (VA(PHC)) and low (VA(PLC)) contrast, mesopic high (VA(MHC)) and low (VA(MLC)) contrast (photopic = 280 cd/m(2), mesopic = 0.75 cd/m(2)). Variables were tested for compliance with normality (-2.00 < skew and kurtosis < 2.00) and transformed if required. Linear regression and Bland-Altman 95% limits of agreement (+/-1.96 SD) were used to examine relationships between VA variables and between VA and wavefront aberration metrics.

The two photopic measures VA(PHC) (skew = 2.57, kurtosis = 12.2) and VA(PLC) (1.67, 5.41) were poorly distributed, but the two mesopic measures VA(MHC) (0.88, 1.67) and VA(MLC) (0.29, -0.18) were normally distributed. Strong correlations existed between the (transformed) VA variables (R: 0.53 to 0.84). However, limits of agreement ranged over 0.30 to 0.42 logMAR, whereas retest data suggested a range of 0.15. All four were offered in stepwise multiple linear regression for 30 wavefront metrics: 20 included VA(MLC) alone, two included VA(PLC), two included VA(MHC) and six included both VA(MLC) and VA(MHC); R(2) averaged 25%.

Although strongly correlated, low contrast and/or mesopic VA testing provides different information. Wave aberration metrics correlates better with VA(MLC) making this the visual performance test of choice.