Meridional amblyopia in astigmatic students documented with computer generated stimuli on commercial displays

Abstract

Purpose: Determine if best-corrected grating acuity (GA) deficits in astigmats measured with computer generated stimuli are consistent with previous findings using printed stimuli.

Methods: Subjects were 2nd - 7th grade Tohono O’odham students who have a high prevalence of with-the-rule astigmatism. Students were classified as non-astigmat (NonA) (RE cyl<1.00,n = 261), myopic or mixed astigmat (M/MA) or hyperopic astigmat (HA) (RE cyl>/=1.00D, axis 90+/-15, M/MA n=154, HA n=76) based on right eye cycloplegic refraction. Right eye best corrected GA for vertical (V) and horizontal (H) stimuli was measured. Stimuli were V and H gratings (50.39 to 3.15 cy/deg, 0.1 log unit steps) presented at 2m on a Retina Display Screen (Apple, Inc) utilizing MatLAB (MathWorks, Inc). Testing order for V and H stimuli were counterbalanced across students and stimuli were presented from lowest to highest spatial frequency. Each trial consisted of two 10x10cm squares: one contained a grating and the other a gray distractor stimulus (1 pixel checkerboard). Students indicated which square contained a grating. Target location was randomized. GA for each stimulus orientation was scored as the highest spatial frequency (log cy/deg) on which the student could correctly locate the grating on at least 3 of 4 trials. Students unable to resolve largest grating were assigned acuity 0.1 log unit below. ANOVA compared mean V, mean H, and mean difference between V and H grating acuity across NonA, M/MA, and HA groups with age as covariate.

Results: Students who had ocular abnormalities, anisometropia > 1.5D, or did not meet criteria for NonA, M/MA or HA group were excluded. There was a marginally significant effect of group for V GA (p=0.05), and significant effects of group for H GA and for the difference between V and H GA. The M/MA group had significantly poorer mean H GA than the NonA group (p<0.008), the HA group had significantly poorer mean V GA than the NonA group (p < 0.04), and both significantly differed from NonA group on the mean difference between V and H GA (p< 0.01).

Conclusions: The pattern of effects is consistent with previous studies on this population, confirm previous findings of meridional amblyopia in astigmatic children, and demonstrate utility of computerized grating stimuli produced with currently commercially available displays.

Keywords: 754 visual acuity • 709 screening for ambylopia and strabismus • 465 clinical (human) or epidemiologic studies: systems/equipment/techniques