Is the Fit Error Associated With the Zernike Polynomial Visually Important in Keratoconus?

Abstract

Abstract: : Purpose: To determine the visual benefit of correcting Zernike orders and fit error for high and low contrast visual acuity (VA) in keratoconus. Methods: Eight keratoconus corneal topographies ranging in maximum corneal power from 44D to 78D in 5D steps were identified. For each topography, aberrations over a 4mm pupil were decomposed into a 10^th order Zernike expansion and residual aberrations (fit error). Nine conditions were considered. For each condition n, Zernike orders 0 to n+1 were removed from the Zernike representation. This gave simulations representing second to tenth order Zernike aberration corrections. PSFs (calculated from the uncorrected Zernike terms and the residual aberrations) were convolved with high (100%) and low (11%) contrast logMAR VA letter charts. These charts were read under photopic conditions (120 ^cd/^_m2) by three cyclopleged healthy normal observers through a 3 mm artificial pupil. The retinal image formed simulated that of the keratoconus eye represented by the topography. Each observer also read three perfect charts to establish baseline acuity. VA was scored by– letter giving credit for correct responses to the fifth miss. Results:For both high and low contrast conditions, VA for eyes with less than 60D maximum corneal power plateaued at a 6^th order Zernike correction with no improvement for higher order corrections. For high contrast conditions, a 6^th order correction gave the < 60D cohort VA comparable to baseline, while a 6^th order correction in low contrast conditions resulted in a loss of 0.1 logMAR from baseline. For both high and low contrast conditions, eyes with more than 60D maximum corneal power were not optimally characterized with a Zernike fit; residual error limited visual performance. Conclusions: A 6^th order Zernike expansion representation of a keratoconus wavefront provides equivalent high contrast VA to complete aberration correction for the majority of keratoconus cases. Visual performance for all Zernike corrections is reduced for the low contrast VA task, suggesting fit error may impair visual performance for more difficult tasks. A minimalist alternative for custom soft contact lens design could be a 6^th order Zernike correction, but a non–Zernike fit with low fit error may be required for optimal visual performance. This conclusion may be confounded by residual fit error being affected by noise characteristics of the topographies.

Keywords: optical properties • keratoconus • visual acuity