Abstract
Purpose: :
To investigate whether normal eyes (NEs) viewing through keratoconic eyes aberrations can achieve similar visual performances to keratoconic eyes (KEs).
Methods: :
We measured the high-contrast visual acuity (VA) (computer-generated Landolt-C optotypes) and the 10-c/deg, 15-c/deg and 20-c/deg contrast sensitivities (CS) (computer-generated randomly oriented sine-wave gratings) of 6 KEs and 6 NEs while dynamically correcting their monochromatic aberrations using an adaptive optics system (Imagine Eyes CRX1 device). We also measured sphero-cylinder corrected visual performances of 6 KEs and 3 NEs viewing through the higher-order aberrations of each of the 6 KEs. The usual defocus and astigmatism terms (i.e. spectacles correction) were corrected using the adaptive optics system. In order to simulate the vision of each KE, we reshaped the deformable mirror of this system to match the measured residual aberrations of each KE. We then assessed the visual performances of each NE while viewing through each of the 6 KEs aberrations pattern (i.e. KE wearing their usual correction). Subjects viewed the stimulus, generated on a calibrated LCD micro-display subtending a visual angle of 114x86 arcmin, through a 5.5-mm artificial pupil conjugated to the pupil plane.
Results: :
After dynamic correction of monochromatic aberrations, the average residual wavefront aberration in NEs and KEs was respectively 0.09±0.019 µm RMS and 0.12±0.015 µm RMS over a 5.5-mm pupil. The difference between KEs and NEs correction was not statistically significant (p=0.54). In monochromatic aberrations correction condition, NEs obtained statistically better visual performances than KEs: in terms of VA, -0.24±0.11 logMAR for NEs compared to -0.07±0.12 logMAR for KEs (p=0.006); in terms of CS, 1.48±0.20 log units for NEs compared to 1.14±0.20 log units for KEs (p=0.006). When viewing through KEs aberrations, NEs experienced statistically (p=0.03) poorer VA (0.35±0.20 logMAR) than KEs (0.26±0.21 logMAR) and statistically (p=0.05) better CS (0.60±0.19 log units) than KEs (0.48±0.20 log units).
Conclusions: :
Compared to NEs, KEs showed poorer visual performances with the same perfect optical quality and better VA when wearing their usual sphero-cylinder correction, suggesting that KEs could be adapted to their poor retinal image. However, when measuring visual performances with a test which is less comparable to real life (i.e. CS measurement), KEs performed worse than NEs with the same aberrations. Consequently, a hypothetical neural adaptation might only occur for usual "real life" visual tasks.
Keywords: adaptation: blur • aberrations • keratoconus