Return to Presentations Return

Measuring visual acuity – high or low luminance, and high or low contrast?

Presented in Fort Lauderdale, Florida April 2004, sponsored by the Association for Research in Vision and Ophthalmology (ARVO). Published in Invest Ophthalmol Vis Sci 44: 2769.

K. Pesudovs1, J.D. Marsack1, L.N. Thibos2 and R.A. Applegate1.

1 Optometry, University of Houston, Houston, TX
2 Optometry, Indiana University, Bloomington, IN

Commercial Relationships: K. Pesudovs, None; J.D. Marsack, None, L.N. Thibos, Indiana University P, R.A. Applegate, University of Texas P, University of Houston P.

Grant Identification: NEI R01 08520 grant to RAA and NEI R01 05109 to LNT

Abstract

Purpose. Our long–term aim is to develop single–valued wavefront aberration metrics that explain a high level of variance in visual acuity. The aim of this study is to explore whether high luminance high contrast acuity is an appropriate reference and whether low luminance or low contrast acuity provides any advantage.

Methods. Subjects were recruited for a cross–sectional study of wave aberration in nuclear cataract. They (N=148) ranged in age from 21.6–83.8 years and from clear lens to dense nuclear cataract. Acuity was measured under four conditions: high luminance high (HLHC) and low (HLLC) contrast, low luminance high (LLHC) and low (LLLC) contrast (HL=280cd/m2, LL=0.75cd/m2). To ensure suitability for parametric analysis variables were tested for compliance with normality (–2.00<kurtosis<2.00, skewness<2.00). Linear regression and Bland–Altman limits of agreement were used to test relationships between acuity variables and between acuity and wave aberration metrics.

Results. The two high luminance measures HLHC (skew=2.57, kurtosis=12.2) and HLLC (1.67, 5.41) were non–Gaussian. Transformation was required before performing parametric analysis. The two low luminance measures LLHC (0.88, 1.67) and LLLC (0.29, –0.18) were naturally normally distributed. LLHC and LLLC were strongly related (linear regression R2=0.55). However, 95% limits of agreement ranged over 0.34 logMAR, whereas retest data suggested the range should be 0.16. Therefore the two tests are not entirely interchangeable.

Conclusions. Only low luminance acuities were suited to parametric testing of relationships with other variables. Although strongly correlated, high contrast and low contrast testing provide different information.

Key Words: visual acuity

 

Return