Abstract
Purpose :
We have previously shown that corneal CXL can be achieved in ex-vivo rabbit corneas using a 76 MHz FS laser and that crosslinked regions can be imaged with collagen autofluorescence (CAF), which is indicative of local stiffening. While this method achieved rapid, localized, and spatially controllable CXL, the required laser power of 800 mW is 20 fold higher than the ANSI limits (46.1 mW) for use in humans. The purpose of this study was to test whether regeneratively amplified 760nm FS laser pulses (5 kHz, ~2 µJ pulse energy) can be used to photoactivate riboflavin within the cornea to induce precise collagen CXL and CAF with a single pulse while remaining under ANSI limits.
Methods :
The same variable numerical aperture (NA), custom laser scanning delivery system with adjustable focal depth was used as in our previous studies. 800 nm FS pulses from a regenerative amplifier (5 kHz) were tuned to 1520 nm in an optical parametric amplifier (Coherent Inc, Santa Clara, Ca). The 1520 nm laser pulses were then frequency doubled in a custom bismuth triborate (BiB3O6) nonlinear crystal (Newlight Photonics, Ontario Canada) to 760 nm and then aligned into our delivery system. As a proof of concept, rabbit corneas soaked in 0.5% Riboflavin/PBS with dextran (20%) were raster scanned with 0.1 and 0.2 NA, 5 mm/s scan velocity, and 12 mW of average power. CAF was used to detect corneal collagen CXL.
Results :
CAF was detected for both numerical apertures. Representative CAF images are presented in Figure 1. A single pulse through the 0.1 NA and 0.2 NA delivery produced CXL regions 173 ± 14 µm and 126 ± 14 µm in axial length, respectively. Both NA’s resulted in a lateral width of roughly 3µm.
Conclusions :
Corneal collagen CXL is possible with a single amplified 760nm FS laser pulse while remaining well under the 46.1 mW ANSI limit for our system. Given the 5 mm/s scan velocity, the current system deposits 1 high energy pulse per micron per 200 μs, compared to roughly 15,000 low energy pulses using a pulse repetition rate of 76 MHz. This innovation allows for adequate energy density per unit time to induce CXL without exceeding ANSI limits. Future optimization of the optical system will enable nearly 4x the current pulse energy while still respecting ANSI limits.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.