Femtosecond-UVA-riboflavin (FUR) cross-linking approach to penetrating keratoplasty and anterior lamellar keratoplasty

Purpose: To introduce femtosecond laser wound design combined with riboflavin/ultraviolet light-A (UVA) collagen cross-linking at the wound for penetrating (PKP) and anterior lamellar keratoplasty (ALK). Primary outcomes were intraocular pressure (IOP in mmHg) at burst point for the PKP group, and tensile strength (kPa) until dehiscence for the ALK group. Methods: Human corneoscleral rims (N =20) were mounted on artificial anterior chambers. PKP specimens underwent FUR, femtosecond laser-cut without cross-linking, or conventional corneal transplantation. PKP maximum burst IOP with progressive suture removal was assessed by a digital manometer, in triplicate and by three observers. ALK involved whole human globes (N = 10) divided into three groups using a 200-micron, 8 mm diameter donor lenticule, with or without cross-linking. Cross-linked specimens were exposed to UVA light (3 mW/cm² irradiance, 3.4 J, 370 nm wavelength) for 30 min with 0.1% riboflavin (20% Dextran) applied every 2-min. ALK tensile strength was determined using a digital tensiometer. Results: In PKP, burst IOP was 31.32 mmHg greater for corneas that underwent the UVA-riboflavin treatment than for those that did not (p 0.05). There was no significant relationship (p=0.719) established between cut design (femtosecond versus conventional). On multivariate analysis, there was a mean of 15.82 mmHg higher sustainable pressure for each stabilization suture present (p 0.0001). In ALK, specimens comprised of human donor and human recipient tissue combined with UVA-riboflavin therapy experienced the greatest level of adhesion strength (954.7± 290.4 kPa) as shown by the force required to separate the tissues, and compared to non-cross-linked specimens. Electron microscopy of ALK specimens showed non-fused and fused longitudinal cross-linked collagen fibers as well as bridges, densities, attachment plaques and primitive plasmalemmal densities. Conclusions: Cross-linking effects of the FUR technique enable a stronger graft-recipient adhesion compared to conventional penetrating and anterior lamellar keratoplasty. Electron microscopy enabled visualization of cross-linked interface and potential bonding. The FUR approach may further lead to sutureless transplantation techniques in the future. Setting/venue: ImagePlus Laser Eye Centre, Winnipeg, and University of Ottawa Eye Institute, Ottawa, Canada.