We compared near and distance visual acuities of subjects 2 to 6 years of age with moderate amblyopia from strabismus, anisometropia or both combined, by using standardized visual acuity testing protocols at distance and near with single surrounded HOTV optotypes. Near visual acuity of amblyopic eyes did not differ from distance visual acuity. The fact that the distance and near testing protocols were not identical could be considered a limitation of the study. The near test included no reinforcement phase or second threshold phase, which were part of the distance testing protocol. Nevertheless, we did not find any difference comparing near and distance visual acuity. The difference in variability between distance and near visual acuity was not examined since the distribution of distance scores was constrained by the eligibility criteria.
Our findings differ from earlier reports regarding near visual acuity in amblyopic eyes. Catford1
found that 18 of 35 hypermetropic amblyopic subjects (51%) had worse visual acuity of the amblyopic eye at near, whereas 5 (14%) were better at near. Von Noorden and Helveston2
found that 9 of 46 esotropic amblyopic subjects (19%) had worse visual acuity of the amblyopic eye at near, whereas 17 (37%) were better at near. Lennarson and colleagues3
studied 70 patients with amblyopia from strabismus, anisometropia, or both, as in our study, and found that 33 of 70 subjects (48%) had worse visual acuity of the amblyopic eye at near, whereas 8 (11%) were better at near. All three studies of amblyopic children found a substantial proportion of subjects with no difference between distance and near visual acuity.1–3
We calculated mean values for the distance minus near visual acuity difference for each of these studies: Catford,1
−0.07 (−0.14, −0.004); Von Noorden and Helveston,2
0.03 (−0.03, 0.09); Lennarson and colleagues,3
−0.08 (−0.11, −0.05); the first and last of these studies noted slightly poorer near visual acuity in the amblyopic eye.
We do not include analyses of fellow eyes because the near-testing protocol created a ceiling effect at 20/20 since near visual acuity better than 20/20 could not be measured. Therefore, subjects testing at or near 20/20 at distance in the fellow eye were constrained with respect to how much better they could test at near, introducing a bias into any observed differences, and in fact 55 (43%) of fellow eyes tested 20/20 at near. Also, the limitations on the range in distance visual acuity imposed by eligibility criteria precluded use of standard Bland-Altman methods10
for comparing distance and near acuities, as these methods assume no constraints on either measurement. However, we believe these issues introduced minimal bias in our evaluation of amblyopic eyes, because we observed minimal ceiling effect with only 2 amblyopic eyes with 20/20 (or possibly better) near visual acuity (), and minimal regression to the mean with relatively equal numbers of subjects testing better and worse at near within each level of distance visual acuity. Testing distance visual acuity first may introduce two offsetting biases. Some subjects may become fatigued and test more poorly on the second test, while others may experience a learning effect and test better.
We found no systematic difference between distance and near visual acuity in 2- to 6-year-old children with moderate anisometropic, strabismic, and combined amblyopia. Individual differences between distance and near visual acuity are likely due to test–retest variability.