Accuracy of the Retinomax K-plus3 in measuring refractive error in a pediatric population

Purpose To compare the Retinomax K-plus3 handheld autorefractor to a retinscopically based refraction performed by a pediatric ophthalmologist. Methods The medical records of 311 children aged 5 months to 17 years who underwent cycloplegic refraction by both autorefractor and a pediatric ophthalmologist on the same day were retrospectively reviewed. Spherical equivalent, vector dioptric distance (VDD) values, and t tests were performed for each metric with an accompanying age-group comparison (<5 years, 5-7 years, 8-10 years, and ≥11 years). Bland-Altman plots were generated to evaluate the agreement between methods. Results Overall, the refraction provided by the Retinomax autorefractor was not significantly different in regard to cylinder power and axis. An age-group comparison showed statistically similar readings for all measures of refractive error in children 8-10 years of age. Statistically similar readings were obtained for cylinder power and axis except in children aged 5-7 years. The VDD was statistically different between the two methods for every age group. Bland-Altman plots suggested agreement between the two approaches for the entire cohort with respect to sphere, cylinder, axis, and spherical equivalent. Conclusions In children <18 years of age, there was no clinical difference in the cycloplegic refraction determined by the Retinomax autorefractor when compared to retinoscopy by a pediatric ophthalmologist.