A Pilot Randomized Clinical Trial of Base-in Relieving Prism Spectacle Treatment of Intermittent Exotropia

This pilot randomized trial, the first to evaluate a specific base-in relieving prism treatment strategy for childhood intermittent exotropia, did not support proceeding to a full-scale clinical trial. Defining and measuring prism adaptation in children with intermittent exotropia are challenging and need further study.This study aimed to determine whether to proceed to a full-scale trial of relieving base-in prism spectacles versus refractive correction alone for children with intermittent exotropia.Children 3 years old to those younger than 13 years with distance intermittent exotropia control score of ≥2 points on the Intermittent Exotropia Office Control Scale (Strabismus 2006;14:147-150; 0 [phoria] to 5 [constant]), ≥1 episode of spontaneous exotropia, and 16 to 35∆ by prism-and-alternate-cover test, who did not fully prism adapt on a 30-minute in-office prism-adaptation test were randomized to base-in relieving prism (40% of the larger of distance and near exodeviations) or nonprism spectacles for 8 weeks. A priori criteria to conduct a full-scale trial were defined for the adjusted treatment group difference in mean distance control: "proceed" (≥0.75 points favoring prism), "uncertain" (>0 to <0.75 points favoring prism), or "do not proceed" (≥0 points favoring nonprism).Fifty-seven children (mean age, 6.6 ± 2.2 years; mean baseline distance control, 3.5 points) received prism (n = 28) or nonprism (n = 29) spectacles. At 8 weeks, mean control values were 3.6 and 3.3 points in prism (n = 25) and nonprism (n = 25) groups, respectively, with an adjusted difference of 0.3 points (95% confidence interval, -0.5 to 1.1 points) favoring nonprism (meeting our a priori "do not proceed" criterion).Base-in prism spectacles, equal to 40% of the larger of the exodeviations at distance or near, worn for 8 weeks by 3- to 12-year-old children with intermittent exotropia did not yield better distance control than refractive correction alone, with the confidence interval indicating that a favorable effect of 0.75 points or larger is unlikely. There was insufficient evidence to warrant a full-scale randomized trial.