Repeatability of Internal Aberrometry with a New Simultaneous Capture Aberrometer/Corneal Topographer

Abstract

To compare repeatability of internal eye aberrations derived from aberrometry and corneal topography (CT) measured simultaneously by the Innovative Visual Systems Discovery, against time displaced but same instrument measurement (Nidek OPD-Scan 3), and time displaced different instrument measurement (Medmont E300 and Imagine Eyes irx3). Three aberrometry and CT measurements were captured with each instrument, except for the OPD-Scan 3 where three aberrometry scans were followed by a single CT. Measurements were repeated across 2 days. Corneal surface Zernike coefficients were derived from CT and subtracted from aberrometry to establish internal aberration coefficients. For the OPD-Scan 3, internal Zernike coefficients were derived by the instrument's software. Repeatability for second-order root mean square (RMS), spherical aberration, coma RMS, trefoil RMS, and the refraction components M, J0, and J45 were assessed using intraclass correlation coefficient (ICC). Intrasession repeatability was similar between the Discovery and E300/irx3, with the E300/irx3 slightly more repeatable for second-order RMS. Across days, the Discovery was most repeatable for second-order RMS (ICC 0.98) followed by the E300/irx3 (ICC 0.96) and OPD-Scan 3 (ICC 0.88). All instruments were less repeatable for higher order aberrations with only the Discovery moderately repeatable for spherical aberration and trefoil RMS (both ICC ≥ 0.75). The Discovery was highly repeatable for all derived refractive components (ICC ≥ 0.96). The E300/irx3 was highly repeatable for M (ICC 0.98) and moderately repeatable for J0 (ICC 0.89). The OPD-Scan 3 was highly repeatable for the M component (ICC 0.98) but not repeatable for the cylindrical components. The Discovery was highly repeatable for second-order RMS and derived refractive components. The lower repeatability for internal higher order aberrations measured with all instruments suggests caution in their use until further work is carried out to investigate sources of error and to develop methods to improve repeatability.