Purchase this article with an account.
K.R. Seger, M.A. Crandall, B.–C. Jiang; Comparison of Refractions Derived From an Aberrometer and Subjective Refraction . Invest. Ophthalmol. Vis. Sci. 2006;47(13):1172.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
It has been suggested that objective refractions derived from wavefront aberrometry could replace subjective refractions (SR). We looked at the accuracy of wavefront aberrometry derived refractions (ADR) analyzed for various pupil diameters and compared the results to SR performed by an experienced clinician.
A COAS aberrometer was used to measure aberrations on 62 subjects (aged 24–30) in dim illumination. A manifest refraction was generated set for the spectacle plane (12mm vertex distance) using the Seidel sphere function (includes primary spherical aberration in the refraction calculation). The results were generated for pupil diameters of 4.0mm, 5.0mm and the natural undilated pupil (up to 7.25mm). These were compared to the undilated SR. We used the power vector (PV) method for comparison of the subjects' right eye only. We looked at the correlation between measurements and also looked at the absolute differences to find what percent of ADR could actually be used by a typical patient.
Correlation between the SR and the ADR was good for all pupil sizes (r=0.98, p=0.0001 for 4.0mm, 5.0mm and >5mm pupil diameters). However, only 29%, 34% or 35% (for 4mm, 5mm and >5mm pupils, respectively) of the ADR based PVs were within or equal to 0.25D of the PV for the SR. For all pupil diameters at least 90% of the errors were from over–minusing (consistent with previous results for auto–refractors). If the acceptable range is expanded to within or equal to 0.50D there is improvement to 53%, 55% and 56% with increasing pupil diameters. The corrections between the ADR PVs and the SR PV range from 0.36D to 0.32D as the pupil diameter increases. If these are applied the results for ADR PVs within 0.25D of the SR PV are 44%, 48% and 42% for increasing pupil sizes. Adjusted values for findings within 0.50D are 79%, 85% and 76% for the 4mm, 5mm and >5mm pupil diameters, respectively.
ADRs tend to be over–minused by a constant amount. Even when that is factored in, the results don't agree with SR closely enough to suggest clinical usefulness. Further studies on subjective patient comfort with the different prescriptions could be illuminating. The specific pupil diameter used for analysis makes only a small difference in accuracy.
This PDF is available to Subscribers Only