Transepithelial riboflavin/ultraviolet. a corneal cross-linking in keratoconus: morphologic studies on human corneas.

Purpose: To evaluate histologic and molecular changes in human keratoconic corneas after the procedure of transepithelial collagen cross-linking (CXL), without the removal of corneal epithelium.

Design: Experimental laboratory investigation.

Methods: Thirty corneal buttons were examined, 18 of which were from patients affected by severe keratoconus and submitted to penetrating keratoplasty (PK). Among these, 8 were analyzed without any treatment, 4 were treated with transepithelial CXL 2 hours before PK, and 6 were treated with transepithelial CXL 3 months before PK. Twelve normal corneal buttons from healthy donors were used as controls. The corneal buttons were then evaluated by hematoxylin-eosin staining and by immunostaining with markers of epithelial junction proteins (ß-catenin and connexin 43), of stromal keratocytes (CD34), of apoptosis (terminal deoxynucleotidyl transferase dUTP nick end labeling [TUNEL] assay), and of collagen type I fibers.

Results: The analysis of epithelial markers showed a clear defective expression in keratoconic corneas before and soon after the transepithelial CXL treatment, returning to normal in corneas analyzed 3 months after transepithelial CXL. The analysis of stroma components indicated a loss of keratocytes in the upper stroma of keratoconic corneas and a trend toward a normal situation 3 months after transepithelial CXL; similarly, collagen fibers appeared disorganized in keratoconus, while their pattern appears to be close to normal 3 months after treatment.

Conclusions: Histologic and immunohistochemical findings on human keratoconic corneas showed the presence of biochemical and morphologic alterations in the epithelium and the upper stroma that are significantly improved 3 months after transepithelial CXL. However, further studies are necessary to assess to what extent these results correlate with measurable biomechanical effects.