Efficacy Analysis of Corneal Crosslinking (CXL): Comparing Epi-on and Epi-off Procedures

Purpose: To analyze the factors influencing the corneal crosslinking (CXL) efficacy and comparison of epi-on and epi-off procedures.

Study Design: modeling the efficacy of epi-off and epi-on CXL.

Place and Duration of Study: New Taipei City, Taiwan, between November, 2021 and December, 2021.

Methodology: Solving the rate equations for the CXL efficacy which includes the roles of concentration of the photosensitizer, riboflavin (RF), RF depletion effects, dynamic of light intensity, and the non-uniform distribution of RF in the stroma, or the diffusion depth of RF. Both steady-state and transient state features are explored for the efficacy, crosslink depth (CD) and the effects of epithelium layer for the epi-on situation.

Results: The steady-state efficacy is proportional to the square-root of [RF-concentration] /[light-intensity], The competing factors of reduced RF, F(z), and reduced light intensity in the stroma determine the relative efficacy of epi-on and epi-off. For example, for F(z)<0.5, epi-on is more efficient than epi-off. In contrast, in the transient state (with efficacy <0.6), the efficacy is proportional to the light dose, and therefore epi-on is always less efficient than epi-off. The crosslink depth (CD) has an inverse trend, such that higher light intensity and lower RF concentration lead to deeper CD. The analytic formulas are developed under simplified conditions, in which numerical simulation is required for non-uniform distribution, and when RF depletion are included. Various strategies for improved steady-state efficacy efficacy and crosslink depth for epi-on CXL are explored including the use of higher RF concentration and lower light intensity; enhancing the RF diffusion by an electrode device, or diffusion enhancing medicine. The analytic formulas are compared with measured data.

Conclusion: For the steady-state epi-on is more efficient than epi-off, when the RF reduction factor is less than the light intensity gain factor. In contrast, in the transient state (with efficacy <0.6), the efficacy is proportional to the light dose, and therefore epi-on is always less efficient than epi-off.