Repeatability and Agreement of a New Scheimpflug Device and a Hartmann-Shack Aberrometer With a Ray-Tracing Aberrometer in Normal, Keratoconus, and CXL Groups

PURPOSE: To assess repeatability and agreement of a Scheimpflug topographer and a Hartmann-Shack aberrometer with a ray-tracing aberrometer in normal, keratoconus, and corneal cross-linking (CXL) groups. METHODS: In this cross-sectional, observational study, normal, keratoconus, and CXL groups with 125 eyes in each of the three groups were included. All eyes underwent three sets of measurements, at a single visit, using the Pentacam AXL Wave (Oculus Optikgerate GmbH) and iTrace (Tracey Technologies). Keratometry, pachymetry, objective refraction, and total ocular aberrations including root mean square (RMS) lower order aberrations (LOAs), RMS higher order aberrations (HOAs), RMS coma, and spherical aberrations (SA) were analyzed. Objective assessment of haze after CXL was done using the Pentacam AXL Wave. Repeatability was assessed using within-subject standard deviation (Sw), test-retest variability, within-subject coefficient of variation (COV), and intraclass correlation coefficient (ICC). Bland-Altman analysis assessed 95% limits of agreement and correlation coefficient (r) determined the strength of the relationship between measurements. RESULTS: The Pentacam AXL Wave had Sw for keratometry of 0.12 in the normal group and 0.15 in the keratoconus group and lower (poorer) Sw of 0.17 in the CXL group. For pachymetry, Sw was 9.18, 9.53, and 10.11 in the normal, keratoconus, and CXL groups, respectively. Total aberrations had ICCs ranging from 0.88 for RMS HOAs to 0.95 for SA in the normal group, 0.86 for RMS HOAs to 0.92 for SA in the keratoconus group, and 0.72 for RMS HOAs to 0.82 for SA (poorer) in the CXL group. In the normal group, mean differences between the two devices were nonsignificant for all parameters except SA (0.011 +/- 0.038 mu m, P < .01; limits of agreement =-0.09 to 0.07; r = 0.87). In the keratoconus group, mean differences were significant in all aberrations except RMS LOAs (-0.27 +/- 0.85 mu m, P = .10; limits of agreement =-3.3 to 3.8; r = 0.92). In the CXL group, all parameters varied significantly (P < .006). CONCLUSIONS: The Pentacam AXL Wave showed comparable repeatability in the normal and keratoconus groups, but was poorer in the CXL group, more so with increasing corneal haze. Both devices can be used interchangeably in normal eyes, except for SA, but not in eyes with keratoconus or CXL for aberration measurements. [J Refract Surg. 2022;38(3):201-208.