The Early Changes in Visual Acuity Predict the Endpoint Improvement of Visual Function in Amblyopia after Perceptual Learning

Abstract

Purpose : Monocular perceptual learning with cut-off spatial frequency tasks has been implemented as supplemental therapy in amblyopes. The laboratory training performance is satisfactory but lacks systematic investigation in the clinical setting. We performed an observational study to evaluate the efficacy and characteristics of spatial vision and binocular function improvement and further explore the individualized effect in amblyopia after perceptual learning.

Methods : The study recruited forty-nine candidates (mean age 17±7.29 years). The patients were trained following a perceptual learning protocol for 10 to 15 days and completed an identification task near the cut-off spatial frequency in the amblyopic eye. Visual acuity (VA) was assessed every day during training. Contrast sensitivity function (qCSF method) and stereoacuity were measured before and after training.

Results : For the whole group, the average improvement in VA was 1.4 ETDRS lines（P<0.001）, and the improvement of VA correlated with baseline BCVA. The improvements in all CSF metrics were also observed after training (all P<0.001). The improvement in low-SF-AULCSF after training was more pronounced in the effective group with significant VA improvement (≥ 1 line). The spatial frequency of the maximum contrast improvement (SF of max CS) in the effective group was also shown to be in the lower SF area. However, stereopsis did not improve significantly before and after training. There was a significant correlation between the dynamic changes in VA in the first three days of training and the VA at the end of training.

Conclusions : Monocular perceptual learning improves spatial and binocular function in amblyopia. The broader activation of spatial vision contributes to improving clinical VA. The change of VA in the early stage can be used to predict the endpoint treatment outcomes.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.