Abstract
Purpose :
The presence of molecular oxygen (O2) during corneal cross-linking (CXL) plays a pivotal role in the biomechanical properties of the cornea. We previously measured the role of O2 and its distribution across the stroma before and during CXL in porcine corneas using phosphorescent nanoparticle probes. The purpose of this study is to compare collagen type-I monomer gel solution with porcine measurements.
Methods :
Porcine globes were obtained from the local slaughterhouse 4 hours post mortem and kept at 4oC until used. The epithelium was removed and the corneas were stained with 0.1% riboflavin and different phosphorescent O2 nanoparticle probes for 1 hour to achieve maximum diffusion. Each globe was analysed at 37° C at 21% ambient O2 with phosphorescent lifetime microscopy (PLIM) with a 5x/0.25 Fluar objective, excitation at 488 nm and emission collected at 750-810 nm. The cornea was imaged over 15 sec at a fixed stromal depth of 50µm. The same method was applied to 6 samples of collagen type-I with an average pachymetry of 50 µm. CXL was achieved through periodic 20-30 illumination cycles with UV-A LED light (7 mW/cm2) whilst imaging. Photon distributions and phosphorescence decay curves were analysed and lifetime values and O2 concentrations calculated.
Results :
We optimised staining with various O2 probes and measurement conditions for porcine corneas and collagen type-I, and performed proof-of-principle PLIM experiments before and after CXL. We observed efficient uniform in-depth staining allowing high-resolution O2 maps to be generated and monitor dynamics during CXL. PLIM revealed a slight increase in O2 concentrations post UV illumination suggesting a role of reactive oxygen species (ROS) during the photochemical CXL process.
Conclusions :
The use of phosphorescent O2 probes allows for efficient and minimally-invasive determination of O2 concentrations prior to and during CXL. Results indicate that collagen type-I is a more efficient model for measurements of O2 due to restrictions in hydration control when using ex-vivo tissue. 2D and 3D maps of O2 concentrations during CXL will allow better understanding of the role of O2. Future work will focus on the suitability of the O2 PLIM method for in-vivo use along with increased imaging depth profiles beyond 100µm for porcine and collagen type-I samples.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.