Minor complications can occur with collagen cross-linking

An expert reviews some of the conditions that may arise after the procedure.

Introduction

Amar Agarwal

Collagen cross-linking has taken the treatment of ectasia and keratoconus, as seen in Figure 1, to another level. This great boon to treatment, pioneered by Theo Seiler, MD, PhD, has helped so many patients. At the same time, one should also learn the complications of collagen cross-linking.

My special guest in this column is A. John Kanellopoulos, MD, who has been involved with the clinical and investigational application of collagen cross-linking for a decade. He currently serves as medical director of the Laservision.gr Institute, Athens, Greece, and as clinical professor in the department of ophthalmology at New York University Medical School.

– Amar Agarwal, MS, FRCS, FRCOphth
Complications Consult Editor

by A. John Kanellopoulos, MD

A. John Kanellopoulos

We have employed collagen cross-linking over the last 10 years in the treatment of ectasia after refractive surgery, as well as in the treatment of primary keratoconus, with relative success. In our center, we have now treated more than 800 cases of primary keratoconus and/or ectasia following refractive surgery, and in more than 500 cases, we combined cross-linking treatment with the use of partial topography-guided PRK in order to facilitate visual rehabilitation (the Athens protocol).

Minor complications encountered due to technique

We have experimented in the laboratory with the use of different riboflavin solution concentrations and different levels of energy, and we have reproduced previous work that suggests doubling the concentration of riboflavin may enhance collagen cross-linking by tenfold and create opaque patches in the cornea that will present significantly over-cross-linked tissue. We have not encountered this complication clinically using the standard 0.1% solution, even in our recent studies while introducing and using high fluence UV light at the level of 7 mW to 15 mW, which is now clinically available in Europe by Avedro. One has to be careful to calibrate the distance of the UV light source from the level of the treated cornea as well as the fluence of the light source. This helps to avoid overexposure and potential toxic levels of UV light that could create significant cornea opacities.

Infection

Infections (Figure 2) are another group of potential complications that can occur. We have encountered only one infection, and it was cured within a few days with the use of topical vancomycin solution. It was attributed to either contamination of the surgical field during the removal of the epithelium and/or contamination of riboflavin drops. We have since changed our technique of epithelium removal and utilize the excimer laser to scrape the epithelium, thus going for a no-touch technique of the cornea epithelium and a reduction of the possibility of transferring pathogens to the corneal surface and stroma with the collagen cross-linking process. Obviously, single-use packaging of the riboflavin solution is essential in order to avoid contamination from patient to patient.

Endothelial toxicity

Endothelial toxicity from high levels of free oxygen radical formation at the endothelial level (and the recently recognized type 2 cross-linking factor that appears to be the riboflavin molecule radical) can potentially happen with high fluence, extended exposure, cornea thinning and/or higher effective riboflavin concentration. It has been suggested that in corneas thinner than 400 µm, one should use hypotonic riboflavin solution in order to induce some cornea edema; however, this may influence the effective riboflavin concentration and hamper the efficacy of collagen cross-linking. Hypotonic solution may help make the cornea reach a thickness of more than 400 µm and thus avoids this potential complication. We have encountered a couple of cases that had transient cornea edema, but interestingly enough, the endothelium cell count did not drop when measured at the 6-month interval.

Figure 1. Acute hydrops.

Figure 2. Corneal ulcer. Images: Dr. Agarwal’s Eye Hospital

Figure 3. Corneal opacity.

Figure 4. Deep anterior lamellar keratoplasty postop.

Delayed re-epithelialization

Treatments using partial topography-guided PRK, utilizing the Alcon WaveLight platform, combined with collagen cross-linking bring a new group of potential complications such as persistent epithelial defects, especially when mitomycin is used to avoid corneal scarring (Figure 3). We have occasionally encountered PRK-related corneal scarring more often in cases that we treated with PRK after cross-linking, thus the adaptation of the Athens protocol (combined same-day partial topographic-guided normalization of the cornea and the collagen cross-linking).

Recurrence of ectasia

The last complication that I want to mention is the potential for recurrence of ectasia after cross-linking. Some groups have noted up to 8% with the classic Dresden protocol. This has also been reported following pregnancy. There have been some anecdotal reports of ectasia resulting after a cornea has been stabilized with collagen cross-linking and then treated with a PRK procedure. This would make sense if the PRK procedure removed a significant amount of tissue and created a significant biomechanical change in the cornea, producing a more vulnerable situation for ectasia.

In our practice, the combination of a frugal PRK normalization of the cornea and collagen cross-linking appears to be synergistic in both cornea stability and visual rehabilitation, outweighing the potential risk of further ectasia from cornea thinning with a 1% regression of ectasia.

Conclusion

The benefits from this procedure are greater than the potential risks, and when used with caution, I think we reward patients by reducing significantly the number of keratoplasties (Figure 4) performed for visual rehabilitation of keratoconus and ectasia. Furthermore, our 10 years of clinical and research experience make me confident that collagen cross-linking will have wider applications in infectious keratitis and biomechanical modulation and stability following a multitude of cornea procedures.

• A. John Kanellopoulos, MD, can be reached at 115 East 61st St., New York, NY 10065; 917-770-0586; email: ajk@brilliantvision.com.
• Edited by Amar Agarwal, MS, FRCS, FRCOphth, director of Dr. Agarwal’s Eye Hospital and Eye Research Centre. Prof. Agarwal is the author of several books published by SLACK Incorporated, publisher of Ocular Surgery News, including Phaco Nightmares: Conquering Cataract Catastrophes, Bimanual Phaco: Mastering the Phakonit/MICS Technique, Dry Eye: A Practical Guide to Ocular Surface Disorders and Stem Cell Surgery and Presbyopia: A Surgical Textbook. He can be reached at 19 Cathedral Road, Chennai 600 086, India; fax: 91-44-28115871; email: dragarwal@vsnl.com; website: www.dragarwal.com.
• Disclosures: Dr. Agarwal has no relevant financial disclosures. Dr. Kanellopoulos is a consultant for Seros Medical, Avedro and Alcon/WaveLight.