Recently, we have developed collagen crosslinking induced by combined riboflavin/UVA treatment, thus increasing the biomechanical rigidity of the cornea to treat progressive keratoconus. The present safety study was performed to evaluate possible cytotoxic effects of combined riboflavin/UVA treatment on the corneal endothelium in vitro. Endothelial cell cultures from porcine corneas were treated with 500 µM riboflavin solution, exposed to various endothelial UVA irradiances (370 nm) ranging from 0.1 to 1.6 mW/cm² for 30 min and evaluated 24 h later using trypan blue staining and Yopro fluorescence staining. The effect of either treatment alone (UVA irradiation ranging from 0.2 to 6 mW/cm²) was also tested. An abrupt cytotoxic threshold irradiance level was found at 0.35 mW/cm² after combined treatment with riboflavin plus UVA irradiation and at 4 mW/cm² with UVA irradiation alone. Riboflavin alone was not toxic. A cytotoxic effect of the combined riboflavin/UVA treatment on corneal endothelial cells is to be expected with a corneal thickness of less than 400 µm. Therefore, pachymetry should be routinely performed before riboflavin/UVA treatment to exclude patients at risk.

Our results show that collagen cross linking might be a useful conservative treatment modality to stop the progression of keratoconus. By this means the need for keratoplasty might be significantly reduced. Given the simplicity of the technique and minimal costs of the treatment it might also be well suited for developing countries.Further studies are envisaged to exclude long-term side effects and to evaluate the long term durability of the mechanical stiffness effect.

The cross-linking technique using riboflavin plus UV irradiation is suitable for at least temporarily stiffening the cornea in vivo and seems to be a promising method for conservative treatment of keratectasia.

The biomechanical behaviour of the cornea can be altered by low-concentration glutaraldehyde, Karnovsky's solution, and by riboflavin and UV irradiation, which offers potential conservative treatment of keratoconus. To optimize this effect further investigation is necessary regarding the dose-effect relation and the in-vivo conditions.

Surprisingly, corneal crosslinking does not increase the interlamellar cohesive force. In the α-amylase group the cohesive force was mainly decreased because of the digestion of proteoglycans. Crosslinking seems to stabilise only inter- and intrafibrillar, but not interlamellar cohesion.

CRF and CH changes may reflect structural changes of the cornea. Thus, the ORA provides valuable information for a better understanding and characterization of the biomechanical condition of the cornea, especially with regard to diseases such as keratoconus and glaucoma.

Purpose: Contact lens-assisted corneal cross-linking (CACXL) has been proposed for the cross-linking treatment in thin corneas. The aim of this study was to assess the biomechanical efficacy of this treatment. Methods: Post-mortem porcine eyes were treated with standard cross-linking and with cross-linking placing a contact lens soaked with isoosmolar riboflavin solution on the debrided cornea with or without an adherent precorneal riboflavin film of up to 100 lm thickness. Three soft contact lenses (Air Optix Aqua, SofLens and Galifa) with different degrees of hydrophilic properties were tested. After cross-linking with a surface UVA irradiance of 3 mW/cm² for 30 min (fluence 5.4 J/cm²), a 400 lm deep anterior corneal flap was created using a lamellar rotating microkeratome. Biomechanical stress-strain measurements and thermal shrinkage tests were performed. Results: In the Air Optix Aqua group (30% hydration) without riboflavin film, Young 0 s modulus and stress at 8% strain were increased significantly versus untreated controls and the effect was 92.4% respectively 86.35% of the standard CXL value. In the SofLens group (59% hydration) without riboflavin film, Young 0 s modulus and stress at 8% strain were increased significantly versus untreated controls and the effect was 67.04% respectively 65.28% of the standard CXL value. In the Galifa group (72% hydration) without riboflavin film, Young 0 s modulus and stress at 8% strain were increased significantly versus untreated controls and the effect was about 68.48% respectively 75.52% of the standard CXL value. In all samples with a precorneal riboflavin film under the contact lens, there was no significant biomechanical effect compared to the untreated controls. Similarly, in the hydrothermal experiments at 70°C, there was a typical mushroom pattern with increased resistance to thermal shrinkage in the anterior stroma after standard CXL, a markedly reduced mushroom effect using a riboflavin-soaked contact lens only and no effect with the use of a riboflavin-soaked contact lens plus a precorneal riboflavin film. Conclusion: The biomechanical effect of CACXL in porcine corneas is about one-third less than after standard CXL. The efficacy of CACXL might be improved by reducing or omitting the riboflavin film on the contact lens. Further risk assessment studies are necessary.

Because of the absence of any side-effects in serious cases the method is recommended prior to surgery.