Purchase this article with an account.
Cheng W. Hong, Abhijit Sinha-Roy, William J. Dupps, Jr.; Tissue Models of Corneal Ectasia and Cross-linking Treatments. Invest. Ophthalmol. Vis. Sci. 2011;52(14):4386.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Clinical results suggest that ectasia progression can be halted by riboflavin/UVA cross-linking. We evaluate the suitability of collagenase in inducing corneal ectasia, and the relative effectiveness of two cross-linking protocols in offsetting the ectasia in a pilot study.
Human corneoscleral tissue not suitable for transplantation was obtained from the Cleveland Eye Bank. After the epithelium was removed, a solution of 10mg/cc collagenase type II was mixed with 15% dextran and applied to the corneas for three hours. A corneal button mounted and pressurized on an artificial anterior chamber underwent cross-linking as described by Wollensak et al. where 0.1% riboflavin is applied every five minutes and UVA radiation is applied for 25 minutes starting from the second addition of riboflavin at an average beam intensity of 3mW/cm2 and diameter of 9mm. The other underwent a similar procedure, but the UVA radiation was centered on the steepest area over a 6mm diameter. Using Scheimpflug and optical coherence tomography, tomographical maps were taken before collagenase exposure, after collagenase exposure, and after cross-linking to evaluate changes in maximum axial curvature, depth of riboflavin penetration and depth of cross-linking.
Collagenase exposure induced significant steepening, with a maximum increase of axial power of 10D. Over multiple measurements and a range of intraocular pressures, the standard cross-linking protocol induced regression of 5D in the central area, while the focused protocol produced regression of 7D.
A simple method for generating topographical features of corneal ectasia in donor tissue is demonstrated. Both crosslinking protocols were effective in causing regression of the disease, but a smaller diameter treatment focused on the locus of maximum curvature was more effective. These preliminary results corroborate our computational modeling studies predicting greater topographic regression with a curvature-guided cross-linking protocol.
This PDF is available to Subscribers Only