Sealing and Healing of Clear Corneal Incisions with an Improved Dextran Aldehyde-PEG Amine Tissue Adhesive

Abstract

Purpose: To create a non-cytotoxic, spontaneously curing tissue adhesive that is strongly bonding and persistent enough that 1–2 μL is capable of sealing a clear corneal incision throughout the first five days of healing.

Methods: A novel prototype delivery device capable of delivering 1–2 μL of a two-component adhesive delivered aqueous solutions of dextran aldehyde and star PEG amine, which mixed by diffusion and crosslinked to form an adhesive hydrogel. Adhesive hydrogels were tested for rates of degradation in phosphate-buffered saline, leak pressures when used to seal clear corneal incisions in enucleated rabbit eyes, and in vitro cytotoxicity when placed in contact with NIH3T3 fibroblast cells. Two formulations were used in vivo to seal clear corneal incisions in New Zealand White rabbits. Wound integrity after 1, 3, 5 and 7 days of healing was assessed by measuring the leak pressures of enucleated eyes.

Results: Tissue adhesives formed by combining aqueous solutions of dextran aldehyde (MW 10,000, 50% oxidized) and an 8-arm star poly(ethylene glycol) (MW 10,000) having two primary amine groups at the end of each arm gave mean leak pressures as high as 141 ± 35 mm Hg and exhibited no in vitro cytotoxicity. When 1–2 μL was used in vivo to seal clear corneal incisions in New Zealand White rabbits, the adhesive maintained an eye leak pressure of at least 120 mm Hg and remained visibly present at the wound site for 5 days.

Conclusions: The combination of an 8-arm star poly(ethylene glycol), MW 10,000, having two primary amine groups at the end of each arm and dextran aldehyde (MW 10,000, 50% oxidized) forms a tissue adhesive that cures spontaneously, is non-cytotoxic, and is strongly bonding and persistent enough that 1–2 μL is capable of sealing a clear corneal incision through the first 5 days of healing.

Keywords::

• Corneal incision
• Dextran aldehyde
• PEG
• Tissue adhesive
• Hydrogel

ACKNOWLEDGEMENTS

We thank Usha Sharma, Ken Tribbett and Ann Yetter for laboratory support work. WC and AB acknowledge support from Research to Prevent Blindness.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.