Corneal Crosslinking: Is Using an Epithelium-On Technique Effective?

Sumit (Sam) Garg, MD


February 14, 2019

Corneal collagen crosslinking (CXL) has been available in Europe for nearly 15 years, beginning with its discovery in 1997 by the German ophthalmologist Theo Seiler and the subsequent publishing of its first clinical results in 2003 by Wollensak and colleagues.[1] The goal of CXL is to halt the progression of keratoconus, and in general, it does and often flattens keratometry by 1-2 diopters. The procedure has also been expanded to post-laser in situ keratomileusis (LASIK) ectasia.

The classic treatment involves removing the corneal epithelium, soaking the cornea with a riboflavin solution (0.1% diluted in 20% dextran), and then applying ultraviolet (UV) light to the cornea for 30 minutes, a procedure that has been coined the Dresden Protocol and is also commonly referred to as epithelium-off (epi-off) CXL. Our domestic experience has grown since its approval by the US Food and Drug Administration (FDA) in 2006, but the current treatment leaves much to be desired. While epi-off CXL has been shown to be effective in many peer-reviewed publications, there have been various modifications to further improve its efficacy and to shorten the time required for the procedure.

One of the major issues with epi-off CXL, however, is the removal of the corneal epithelium, which causes discomfort and pain, increases infection risk, and can lead to corneal haze and/or edema. We have learned much of this through our growing experience with photorefractive keratectomy and the CXL procedure itself.

In Support of Epi-On

Within the crosslinking community, there is debate as to whether epithelium-on CXL (epi-on) is as effective as epi-off CXL. Some ask whether epi-on CXL is effective at all.

Stulting and colleagues[2] presented their experience with epi-on CXL in patients with both keratoconus (512 eyes) and post-LASIK ectasia (80 eyes). Their procedure deviates from the FDA-approved protocol in that the corneal epithelium was not removed, the riboflavin soak was aided by use of a proprietary sponge, and the UV light was delivered in a pulsed fashion. They did not specifically require progression for patients to be eligible for the study; however, the majority (92%) showed signs of progression by medical history and 56% by increasing myopia and/or steeper keratometry. Overall, the study results were favorable, with improvement in uncorrected and corrected distance visual acuities, flattening of keratometry, and no vision-threatening complications.

These data are certainly subject to criticism by those who believe that epi-on CXL cannot be effective because of the UV-blocking nature of the epithelium. However, I believe that these data are encouraging. Despite the opinion that epi-on CXL is not as effective as epi-off CXL, it may be effective enough in the right patient. If we can identify those patients who could potentially benefit from a "safer" alternative, then why not? Logic would support that if these patients showed evidence of progression after treatment, then perhaps epi-off CXL could be pursued. I think we can all agree that if epi-on CXL can match the efficacy of epi-off CXL, the latter will become obsolete. This study by Stulting and colleagues is a step forward in our understanding of epi-on CXL, and I look forward to the day that we have an epi-on CXL procedure that is safe, effective, and time-efficient.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.