Abstract
Purpose :
Refractive status and the optical quality of the eye impact visual acuity (VA) in a direct manner. Variability of VA measurements is often not provided, so that the confidence limits of the results are unknown. We examined the means and variability in VA measurements by inducing a positive defocus among normal subjects, similar to the defocus with retinal elevation in exudative eye disease.
Methods :
Normally sighted adults (n=44, age = 32 +/- 13.01 yr, 21-71 yr) participated in a visual acuity task with refractive correction and three test conditions with the addition of plus lenses (0.75 D, 1.50 D, and 2.25 D). All subjects had best-corrected visual acuity of at least 20/20 and mean spherical error of 1.35 +/- 2.25 D. All subjects were consented and tested in a manner approved by the Indiana University Institutional Review Board, which adhered to the Declaration of Helsinki. Distance VA was measured and analyzed using a 4 alternative forced-choice task with tumbling E’s displayed using custom Matlab software on a monitor calibrated for luminance (85cd/m2) and distance (11.2 feet), with whole pixel increments for the letter size changes. The E was presented in a 1-up-1-down staircase method starting from MAR of 2 arc min, in logMAR steps, providing VA mean and SD both from staircases data and fits to the psychometric function with a cumulative Gaussian distribution.
Results :
As expected the subjects had increasingly reduced VA in conditions with higher plus lenses. Mean VA (76% fitted threshold) for refractive correction, +0.75 D, +1.50 D, and +2.25 D were 0.99 +/- 0.45 MAR, 1.75 +/- 0.9 MAR, 3.51 +/- 1.69 MAR, and 5.91 +/- 2.84 MAR, respectively, (paired t-test, p<0.00331 for all with Bonferroni correction). The coefficient of variation of all the sizes after the first reversal and the standard deviation of the fitted curve normalized by mean was significantly different across the test conditions (ANOVA, F= 16.86; df=41; P<0.0001 and F F=73.27, df=41; P<0.0001). The confidence limits of VA increased with higher plus lenses, e.g., a subject with no refractive error had a CI for MAR of 0.60 to 0.78 for 0 D but MAR of 3.39 to 4.15 for +2.25 D.
Conclusions :
VA measurements in individuals with poor refractive/optical quality of eye have higher variability, leading to inaccurate assessment of visual function in clinical and research environments.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.