Evaluating Corneal Densitometry in Keratoconic Patients Versus Normal Control Subjects

Purpose: To evaluate differences in best-corrected HCVA, corneal densitometry (CD), pachymetry and keratometry measurements between ten normals (27.3±6.7years), fourteen keratoconic patients (24.8±5.8years) who previously (3/12 earlier) underwent corneal collagen-crosslinking (CCXL), and ten keratoconic patients without CCXL (26.5±7.3years). Methods: Pachymetry, CD and keratometry measurements were made with the Pentacam. CD data were obtained across three annular zones (6-10mm, 2-6mm and 0-2mm) within the anterior (120μm depth), central and posterior (60μm depth) layers, respectively. Pachymetry data were evaluated at the thinnest corneal location. Keratometry measurements were recorded in the steepest/flattest meridians. HCVA measurements were made using an ETDRS logMAR chart positioned at 4m using either trial lenses (normals), or the keratoconic patient’s habitual correction (either spectacles or RGP lenses). Study parameters were compared between groups using either ANOVA (parametric data), or the Kruskal-Wallis H-test (non-parametric data). Correlation analyses were performed using either Pearson’s correlation (parametric data) or Spearman’s correlation (non-parametric data). Results: Neither age (p = 0.630) nor gender (p = 0.820) were significantly different between the three study groups. In contrast, ETDRS logMAR HCVA (p<0.001), the thinnest corneal location (p<0.001), the steepest keratometry readings (p<0.001), and the flattest keratometry readings (p<0.001) were all found to be significantly different between the three study groups.With respect to CD, the data in the anterior (p<0.001), central (p<0.006) and posterior layers (p<0.006) were found to be significantly different, across each of the three annular zones, between all three study groups.However, for all three study groups, no significant correlations were found between the CD values and the HCVA data (p≥0.111); or between the CD values and the flat keratometry data (p≥0.077), or between the CD values and the steep keratometry data (p≥0.140).For the CCXL group only, there were weak positive correlations between the thinnest corneal location and the CD data in the anterior corneal layer within the 2-6mm zone (R=+0.537, p=0.048), and within the 6-10mm zone (rho=+0.620, p=0.018). Conclusion: Although CD values were found to be significantly different between these three study groups, the CD data could not be used to predict best-corrected HCVA or keratometry measurements in any of these groups.