Back to Freepaper Session
Detection of keratoconus using total and corneal aberrometry
Session Title: Imaging II
Session Date/Time: Monday 07/10/2013 | 16:30-18:00
Paper Time: 16:48
Venue: Main Lecture Hall (Ground Floor)
First Author: : E.Diniz BRAZIL
Co Author(s): : B. Lopes I. Ramos R. Ambrósio Jr
to determine the ability of total and corneal wavefront coefficients analysis for distinguishing normal and keratoconic eyes.
Instituto de Olhos Renato Ambrósio, Rio de Janeiro, Brazil
The study comprised of one eye randomly selected from 48 patients diagnosed with bilateral keratoconus (keratoconus group) and one eye randomly selected from 57 normal refractive candidate subjects (control group). Corneal and total Zernike coefficients were calculated for the 4mm zone from measurements using the iTrace (Tracey Technologies, Corp., Houston, TX). After checking the distribution of each parameter in both groups using the Kolmogorov-Smirnov test, the Students t-test or the Mann-Whitney test were used to test if the distribution of each parameter was different among normal and keratoconic eyes. The receiver operating characteristic (ROC) curve was calculated for all parameters that had statistically different distributions to establish cut-off points, sensitivity and specificity for discriminating between keratoconus and normal subjects. . Discriminant Fisher analysis was performed to optimize the combination of the following statistically significant parameters: total high order aberration (HOA), corneal central power, steep Sim K and corneal coma.
All but total defocus (P=0.29), corneal spherical aberrations (P=0.06) and spherical refractive error (P=0.83) had statistical distributions (P<0.05) among normal and keratoconic eyes. Corneal coma and corneal HOA were the coefficients with best area under the ROC curve [AUROC] with 0.987 and 0.983 and cutoffs of >0.111? and >0.231? respectively. Sensitivity and specificity were 100% and 89.47% for corneal coma and 95.83% and 94.74% for corneal HOA. The Fisher“s discrimination parameter had AUROC of 0.992, cutoff of >1.241 with 97,96% sensitivity and 96,49% specificity.
Aberrometry analysis offers an alternative methodology for discriminating keratoconus from refractive candidates with normal corneas. Coma-like and higher order aberrations had best individual performances. Combined parameters have potential to improve the diagnostic performance of aberrometry-derived parameters. Corneal ROC curves tend to show better accuracy compared to total coefficients ROC curves. Further studies involving cases with milder forms of ectasia should be performed to test if such parameters are sensitive to detect form fruste keratoconus or ectasia susceptible refractive candidates.