Efficacy of Customized Corneal Cross-linking vs. Standard Corneal Cross-linking

In 2003 Wollensak et al used corneal cross-linking (CXL) in humans to halt the progression of keratoconus. During the procedure the top layer of the cornea, the epithelium, is debrided. Then the cornea is soaked with riboflavin, a photosensitizer. Hereafter a 9.0 mm diameter Ultraviolet-A (UVA) beam radiates the cornea for 30 minutes with a fluence of 3 mW/cm2 resulting in a total energy of 5,4 J/cm2. This protocol is called the Dresden protocol. Currently, accelerated versions of the Dresden protocol are used in common practice. There are different accelerated protocols with fluences of 9mW/cm2, 10mW/cm2 and 15 mW/cm2. The higher the fluence, the shorter the treatment time, however according to the Bunsen-Roscoe reciprocity law the total amount of energy stays the same.During the procedure oxygen radicals are formed that interact with the surrounding molecules, leading to the formation of new chemical bounds between the collagen fibrils (i.e. corneal crosslinks). The final goal of the procedure is to cause the cornea to stiffen and achieve flattening of the treated region.

For any treatment, it is imperative that the unaffected region of the tissue is not unnecessarily treated by an intervention or drug application. To minimalize the risk of damage to surrounding tissues it would be beneficial that the UVA beam is restricted to the affected, keratoconic zone in the patient's cornea. This can be achieved by customizing the beam shape and size in a way that only the degenerated zone is treated, i.e. by customized cross-linking (cCXL). Recently published studies provide clinical evidence that similar clinical outcomes (amount of corneal flattening) can be achieved when only the cone is treated instead of the entire cornea.They also show the potential benefits of cCXL, e.g. the treatment is patient-specific, a smaller surface of the cornea is irradiated, lower incidence of corneal haze, a faster reepithelialisation and a shorter procedure time. However, none of these studies are randomized and study results are limited by using small sample sizes. Therefore, we feel that there is a great need for a randomized controlled trial with an appropriate design and sample size to confirm these findings.

The aim of this study is to investigate if cCXL is non-inferior to sCXL (10 mW/cm2) in terms of flattening of the corneal surface and halting the disease progression.