Purchase this article with an account.
Jack Benjamin Margines, Fei Yu, Shiva Mehravaran, Anne L Coleman; Agreement between non-cycloplegic refraction with Retinomax and cycloplegic refraction in pre-school children informs screening criteria.. Invest. Ophthalmol. Vis. Sci. 2018;59(9):4129.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To evaluate the agreement between non-cycloplegic and cycloplegic refraction in an ethnically diverse captive population of preschool-aged children.
This study included preschoolers who were screened by UCLA Preschool Vision Program (UPVP) using a Retinomax autorefractor (Righton, Japan). Those students who were measured as having a significant refractive error or noticed to have other ocular pathology (strabismus, nystagmus, etc.) were referred for a full exam, including cycloplegic refraction. We used right eye data of sphere and cylinder to compare non-cycloplegic refraction (NCR) performed on the screening day with cycloplegic refraction (CR) data from the exam day.
There were 5,612 preschoolers who met the inclusion criteria (48.5% female, 88.0% Latino, mean age 4.3 ± 0.67 years). The mean values of sphere as measured in NCR was -1.74 ± 2.23 diopters (D), while for CR it was 0.24 ± 2.24 D. There was a mean difference of 1.98±1.52 D with 95% limits of agreement (LoA) of -1.0 to 4.97 D. This difference is statistically and clinically significant. Less error was found in myopic eyes compared to emmetropic or hyperopic eyes. The mean values of cylinder as measured in NCR was 1.99 ± 1.11 D, while for CR it was 1.92 ± 1.08 D. This represents a mean difference of -0.07 ± 0.42 D with a 95% LoA of -0.89 to 0.75 D. While these values were significantly different (p<0.05), this difference was not clinically significant.
Non-cycloplegic refraction with Retinomax in 3- to 5-year-old children is not sufficiently accurate to discern actual refractive error on cycloplegic examination as performed by UPVP ophthalmologists. The levels of agreement can be relatively poor for sphere while cylinder measurements are reasonably accurate. Screening programs should evaluate thresholds for recommending cycloplegic examination, as many hyperopic children may be accommodating 2.0 D or more and miss being prescirbed refractive correction. Future research should identify trends in astigmatism within this population. A preliminary analysis of cylinder data from 67,866 preschoolers demonstrated that average levels of astigmatism differ among various ethnicities. Future research should identify trends in astigmatism within this population.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
Agreement between spherical measurements from CR and NCR
Agreement between cylindrical measurements from CR and NCR
This PDF is available to Subscribers Only