Refraction under general anesthesia in children, using cycloplegic refraction only as a reference.

To compare the measurements of cycloplegic refraction and refraction (R1-1) under general anesthesia (GA) when using the same portable auto-refractometer (ARF) in pediatric patients. 36Thirty-six to 60-month-old patients who underwent refraction measurements using a portable ARF (Retinomax® K plus 3, Righton, Japan), who did not receive prior cycloplegics under this GA and who had cycloplegic refraction using 1% cyclopentolate and the same Retinomax® device < 3 months prior this GA, between 2015 and 2018, were included in this study. The agreement (Bland-Altman analysis) and correlation (Pearson correlation) between the mean values of the measurements were analyzed. Two-hundred-twenty-two right eyes of 222 patients (114 male and 108 female) were included in this study. The mean age was 45.04 ± 11.24 months. The mean spherical refractions (R1-1, R2-1) under GA and cycloplegic refraction were 1.08 ± 3.50 diopter (D) (-8.00 to +8.00) and 2.58 ± 3.28 D (-6.50 to +9.25), respectively. A strong positive correlation was detected between the two measurements (r = 0.95). When comparing measurements, the mean measurement under GA was -1.49 D (95% confidence interval: lower limit, -3.63; upper limit, +0.63) more myopic than the mean cycloplegic refraction (R1-1) value (Bland-Altman analysis test). The differences between the measurements were within ± 1 D in 92 eyes (41.44%) and within ± 2 D in 180 eyes (81.01%). No significant difference was detected when comparing the cylindrical refractive error values ( Refractive measurements under GA were more myopic than cycloplegic refraction (R1-1) measurements. It is important to consider that complete cycloplegia is not achieved under GA.