Abstract
Abstract: :
Purpose: The purpose of this research is to test the validity of the hypothesis that measures of optical quality in normal, healthy eyes (20/15 acuity) become increasingly predictive of visual performance as the difficulty of the visual performance task increases. Methods: Forty–nine individuals in good health aged 21.8 to 62.6 years served as subjects. Subjects were required to be in good health and have normal acuity of 20/17 or better (mean 20/15; range 20/17 to 20/12) through pupils dilated to at least 7mm. Dilated Shack–Hartmann wavefront error for a 7mm pupil and four types logMAR visual acuity were measured: 1) photopic (270 cd/m2) high contrast; 2) photopic low (11%) contrast; 3) mesopic (0.7cd/m2) high contrast; and 4) mesopic low contrast. Acuity was defined by giving credit for all letters read correctly up to the 5th miss. Thirty different single value metrics of optical performance were calculated for the high order aberration coefficients (3rd to 10th radial orders) of each subject (Thibos et al 2004 www.journalofvision.org/4/4/3). Coefficients of determination were calculated for each of the four visual acuity tasks by regressing visual acuity against each metric of optical performance. The 10 most predictive metrics of optical quality across all 4 measures of acuity were determined by both weighting each metric by the amount of variance in acuity for which it was able to account and its rank in its ability to account for the acuity measure of interest. Results: The metric most predictive of visual performance was the log of the area under the optical transfer function. Log AreaOTF accounted of 4.5%, 14.6%, 26.4% and 37.4% of the variance in the four acuity tasks respectively. All of the top 10 metrics followed the same trend, with predictive power increasing with increasing difficulty of the acuity task. All metrics in the top 10 were similar to the top metric accounting for 4% ±2%, 13% ±2%, 23% ±4%, and 33% ±4% (mean and 1 SD of top 10 metrics) of the variance in the acuity tasks, respectively. Conclusions: 1. Optical quality metrics increase in predictive value with increasing difficulty of the visual task. 2. In normal healthy eyes the information content of photopic high contrast acuity is redundant and higher order aberrations only account for 4% of the variance. 3. Mesopic low contrast acuity is 8 times more sensitive than photopic high contrast acuity to variation in optical quality metrics based on higher order wavefront aberration. 4. To measure the gains or losses in acuity resulting from wavefront guided corrections in normal eyes with normal acuity assessment should to be made with a more sensitive test of visual acuity than photopic high contrast acuity.
Keywords: refractive surgery: optical quality • visual acuity • optical properties