Comparing the Time-Dependent Biomechanical Behavior of Ex-vivo Porcine Corneas under Cross-linking Treatment using Wave-Based Optical Coherence Elastography

Corneal collagen crosslinking (CXL) is a fundamental treatment to halt the progression of keratoconus. Several accelerated CXL protocols are available, ensuring the cornea receives a consistent UV-light dose of 5.4 J/cm2 by manipulating irradiation time and intensity. However, their impact on stiffening the cornea may not be uniform. In this study, we compare three protocols using a dextran-based epi-off photosensitizer applied during 30 minutes of riboflavin instillation, and 30 minutes (Dresden: DRE, 3mW/cm2), 10 minutes (accelerated A10, 9mW/cm2), and 3 minutes (Accelerated A03, 30mW/cm2) of UV-light irradiation time. This study utilizes a custom-made Optical Coherence Elastography (OCE) setup based on an air-coupled ultrasound transducer (ACUS), coaligned with a ~365 nm UV-light source in the sample arm. This configuration allowed for the time-dependent measurement of Lamb wave speed and thickness of n=12 ex-vivo porcine corneas, allowing the estimation of time-dependent shear modulus G(t). Our findings revealed that the Dresden protocol induced the greatest increase (p < 0.001) in G (95.643±11.022%), followed by A10 (33.932±5.467%) and A03 (24.299±7.022%) during the UV-irradiation phase. Interestingly, the average rate change of G with respect to time (mG) in the irradiation phase was significantly higher (p<0.05) in A03 (86.34x10-3min-1), followed by A10 (50.49x10-3min-1), and DRE (36.61x10-3min-1). Our results suggest that, even though the speed of the stiffening process increases with the irradiation intensity during accelerated CXL, the UV- exposure time is not enough to achieve similar results as in the Dresden protocol. This highlights the need of higher energy dose (>5.4 J/cm2) for accelerated protocols.