Effect of a Binocular iPad Game vs Part-time Patching in Children Aged 5 to 12 Years With Amblyopia: A Randomized Clinical Trial

Jonathan M. Holmes, Vivian M. Manh, Elizabeth L. Lazar, Roy W. Beck, Eileen E. Birch, Raymond T. Kraker, Eric R. Crouch, S. Ayse Erzurum, Nausheen Khuddus, Allison I. Summers, David K. Wallace

A binocular approach to treating anisometropic and strabismic amblyopia has recently been advocated. Initial studies have yielded promising results, suggesting that a larger randomized clinical trial is warranted. To compare visual acuity (VA) improvement in children with amblyopia treated with a binocular iPad game vs part-time patching. A multicenter, noninferiority randomized clinical trial was conducted in community and institutional practices from September 16, 2014, to August 28, 2015. Participants included 385 children aged 5 years to younger than 13 years with amblyopia (20/40 to 20/200, mean 20/63) resulting from strabismus, anisometropia, or both. Participants were randomly assigned to either 16 weeks of a binocular iPad game prescribed for 1 hour a day (190 participants; binocular group) or patching of the fellow eye prescribed for 2 hours a day (195 participants; patching group). Study follow-up visits were scheduled at 4, 8, 12, and 16 weeks. A modified intent-to-treat analysis was performed on participants who completed the 16-week trial. Binocular iPad game or patching of the fellow eye. Change in amblyopic-eye VA from baseline to 16 weeks. Of the 385 participants, 187 were female (48.6%); mean (SD) age was 8.5 (1.9) years. At 16 weeks, mean amblyopic-eye VA improved 1.05 lines (2-sided 95% CI, 0.85-1.24 lines) in the binocular group and 1.35 lines (2-sided 95% CI, 1.17-1.54 lines) in the patching group, with an adjusted treatment group difference of 0.31 lines favoring patching (upper limit of the 1-sided 95% CI, 0.53 lines). This upper limit exceeded the prespecified noninferiority limit of 0.5 lines. Only 39 of the 176 participants (22.2%) randomized to the binocular game and with log file data available performed more than 75% of the prescribed treatment (median, 46%; interquartile range, 20%-72%). In younger participants (aged 5 to <7 years) without prior amblyopia treatment, amblyopic-eye VA improved by a mean (SD) of 2.5 (1.5) lines in the binocular group and 2.8 (0.8) lines in the patching group. Adverse effects (including diplopia) were uncommon and of similar frequency between groups. In children aged 5 to younger than 13 years, amblyopic-eye VA improved with binocular game play and with patching, particularly in younger children (age 5 to <7 years) without prior amblyopia treatment. Although the primary noninferiority analysis was indeterminate, a post hoc analysis suggested that VA improvement with this particular binocular iPad treatment was not as good as with 2 hours of prescribed daily patching. http://www.clinicaltrials.gov Identifier: NCT02200211.