Pattern-reversal visual-evoked potentials in the diagnosis of amblyopia in children.

We assessed the potential clinical usefulness of pattern-reversal visual-evoked potentials in the diagnosis of amblyopia. Twenty-seven children with anisometropic amblyopia and four children without amblyopia participated. Estimates of visual acuity for each eye (Snellen visual acuity) were obtained by conventional psychometric methods. Visual-evoked potentials to reversing checks subtending 15 minutes of visual arc were also obtained. Visual-evoked potential testing and interpretation were done in a masked fashion. Ten of the 31 children were retested seven to 21 days after the first test to estimate reliability of the procedures. Of the 27 amblyopic children, 22 were correctly identified by the visual-evoked potential test alone. In four patients initial visual-evoked potential tests failed to identify the disparity in visual acuity between the eyes and retests in two of the four again had false-negative results. In one child initial visual-evoked potential testing incorrectly identified the amblyopic eye but repeat testing did identify it. Of the four children with symmetrically good vision, three were correctly identified as normal by the initial visual-evoked potential test. The other normal child was incorrectly identified by the visual-evoked potential test as having amblyopia.