Corneal epithelial aberrations: a novel diagnostic tool for keratoconus and forme fruste keratoconus

Purpose To assess the ability of corneal epithelial aberrations to discriminate forme fruste keratoconus (FFKC) and keratoconus (KC) from normal eyes. Methods This prospective, case-control study enrolled 91 right eyes from 91 normal participants, 87 eyes with FFKC and 148 eyes with KC. Epithelial aberrations for the 6-mm pupil were measured using an anterior segment optical coherence tomography (MS-39, CSO). The epithelial root mean square of higher and lower-order aberrations (total RMS), root mean square of higher-order aberrations (HOAs RMS, from the 3rd to the 7th Zernike polynomials), coma, trefoil, spherical aberration, and secondary astigmatism were recorded. Stepwise logistic regression was utilized to develop the epithelial aberrations index (EAI) for obtaining the optimal discriminant function to diagnose FFKC (EAI-FFKC) and KC (EAI-KC). Area under the receiver operating characteristic curve (AUC) analysis was used to determine the diagnostic accuracy of the indices. Results FFKC and KC eyes had significantly higher epithelial aberrations than normal eyes. Comparing FFKC with the normal group, epithelial HOAs RMS and coma attained AUC values of 0.714 and 0.788, respectively. The EAI-FFKC showed the highest discrimination ability to differentiate FFKC from normal eyes indicated by an AUC value of 0.822 with 77.0% sensitivity and 75.8% specificity. Comparing KC with the normal group, epithelial HOAs RMS attained AUC values of 0.976–0.998 with 95.2%–100% sensitivity and 92.3%–96.7% specificity, epithelial coma attained AUC values of 0.974–0.997 with 92.9%–100% sensitivity and 96.7%–98.9% specificity. The EAI-KC showed the highest discriminative ability to differentiate KC from normal eyes indicated by AUC of 0.996 with 98.6% sensitivity and 98.9% specificity. Conclusion Epithelial wavefront analysis can identify abnormal epithelial changes across all stages of KC, from very early to severe. Epithelial aberrations can be used as a diagnostic tool for KC and FFKC.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.