Though corneal collagen cross-linking has the potential to help many patients with keratoconus, the patients likely to benefit from it the most—children—may not receive it, for a variety of reasons. Here, I’ll review what we know about cross-linking in patients under the age of 14, and look at ways the procedure might be able to help them.

The Disease

Keratoconus is a progressive corneal ectatic disease that thins the cornea and creates irregular astigmatism. In advanced stages, the disease can also cause a reduction in the patient’s best-corrected visual acuity. 

Keratoconus remains fairly rare. Its prevalence varies in different parts of the world and by the criteria used to diagnose the condition. In the United States the rate of keratoconus has been reported as 54.5 per 100,000 in Minnesota,1 but can range anywhere from between 4 to 600 per 100,000 depending on the diagnostic criteria used.2 

Keratoconus management underwent a paradigm shift in the United States with FDA approval in the spring of 2016 of the Avedro KXL system and Photrexa riboflavin solutions for the treatment of progressive keratoconus and corneal ectasia following refractive surgery.3 This approval means that it’s possible to administer the treatment right when ectasia begins, which has been hypothesized to occur around the time of puberty. If it can be stopped at that point, that would prevent 21.6 percent of patients from progressing to the need for corneal transplantation.4 The challenge is that the original FDA studies performed by Avedro didn’t include any patients under the age of 14. So, without safety data for the pediatric population, the indication ultimately approved by the FDA was for patients 14 or older.5,6 This is of course well after most children have entered puberty, since age 13 for girls and 14 for boys is considered delayed puberty.7



The diagnosis of keratoconus is becoming more specific and more sensitive as our ability to image the cornea improves. Initially, diagnostic criteria were based on corneal steepness on keratometry and the scissoring reflex seen on retinoscopy. In more advanced cases, the ectactic portion of the cornea—the cone itself—could be visualized at the slit lamp. 

Other findings of keratoconus that can be seen at the slit lamp include: 

1) Vogt’s striae: fine folds in Descemet’s membrane at the apex of the cone that flatten with gentle pressure on the globe;

2) Kayser-Fleischer ring: iron deposition into the corneal subepithelial layer at the base of the cone;

3) corneal hydrops: a rupture of Descemet’s membrane, leading to corneal edema and decreased vision; and

4) apical scarring: this occurs as the cone becomes steeper, or after an episode of corneal hydrops.

Everyone agrees on the diagnosis of keratoconus in these late stages when there’s clear clinical evidence of the disease. However, it’s the early diagnosis of cases of forme fruste or subclinical ectasia that remains controversial. With the development of topography and the progression to multiple forms of tomography, our attempts to diagnose keratoconus earlier have become more successful. 

Piggyback lenses (RGP on top of cosmetic colored soft contact lens) in a patient with keratoconus.

In an effort to reach a consensus, researchers surveyed corneal experts from around the world and then held a panel meeting to discuss the results of the survey. In the debates that followed, researchers and clinicians came to some conclusions regarding the diagnosis, monitoring and treatment of keratoconus and other ectatic diseases, even though the scientific evidence wasn’t clear-cut.8 Among those conclusions was the importance of posterior corneal curvature measurements; the panel defined the presence of abnormal posterior corneal elevation as a requirement for diagnosing early or subclinical keratoconus.8 This makes corneal tomography essential for the earliest diagnosis. Currently, imaging of the posterior corneal curvature is most often performed using Scheimpflug technology or anterior segment OCT. Recently, imaging of the corneal epithelium—primarily using OCT technology—in order to create an epithelial thickness map has become an area of intense interest in the quest for another marker for early keratoconus.9



The treatment of keratoconus has undergone many changes over the years, with the approval of corneal collagen cross-linking, both in the United States and abroad, being the most dramatic. 

Before cross-linking, treatments for keratoconus included glasses, contact lenses (soft, hybrid, rigid gas permeable and scleral), intrastromal ring segments, full thickness penetrating keratoplasty, deep anterior lamellar keratoplasty, and the transplantation of Bowman’s layer. None of these, however, adequately addressed the progressive nature of the disease. Cross-linking, with its ability to halt progression, was a paradigm shift. Since most experts agree that there’s a much higher likelihood of keratoconus progression in the younger age group, the goal is to treat keratoconus at its earliest stages in the youngest ages to prevent progression, reduced best-corrected visual acuity, and the need for corneal transplantation.

Since Avedro’s cross-linking system is the one that’s FDA-approved, I’ll be discussing it exclusively here. Currently the system is for epithelium-debrided (a.k.a., “epi-off”) corneal collagen cross-linking, following what’s known as the Dresden protocol. The Dresden protocol first entails removing the corneal epithelium. Then, you administer Photrexa Viscous at a rate of one drop every two minutes over a “soaking” period of 30 minutes. Next, examine the eye at the slit lamp for the presence of yellow flare in the anterior chamber. If flare is present, then you proceed with the ultraviolet treatment. If there’s no flare, then you continue to administer one drop of Photrexa Viscous every two minutes for two to three more doses before checking for flare again. You then perform corneal pachymetry. If the cornea is at least 400 µm thick, you proceed to ultraviolet treatment. If it’s thinner than 400 µm, you administer two drops of Photrexa (NOT Viscous) every five to 10 seconds until the pachymetric measurement is greater than 400 µm. If the cornea can’t be thickened to at least 400 µm, then it’s not advisable to proceed with the treatment. If you’re able to proceed to the ultraviolet treatment, then apply 3 mW/cm2 of energy (wavelength of 365 nm) to the cornea for 30 minutes, for a total energy of 5.4J/cm2. During the 30 minutes of UV treatment, apply one drop of Photrexa Viscous every two minutes.10

Tomography can be useful for catching the telltale signs of keratoconus.



The goal of the pilot studies was to show a reduction in the maximum keratometry measurement (Kmax) of at least 1 D. All three studies (UVX-001, UVX-002 and UVX-003) met this endpoint. For keratoconus specifically (included in UVX-001 and UVX-002) there was a reduction in Kmax of -1.9 D and -2.3 D, respectively, at 12 months postop.

Looking specifically at the pediatric population, a group of international researchers studied keratoconic patients ranging from 10 to 17 years old.11 They performed the standard Dresden protocol treatment on them, and then followed them for an average of 7.5 years. During that time, though the keratometry measurements didn’t improve, the researchers did note stability of the keratoconus.

The long-term effect of cross-linking can be seen in the trends observed internationally where cross-linking has been available for over a decade. There are a number of studies that show the rates of corneal transplants (PKP and DALK) for keratoconus decreasing after the introduction of cross-linking, intracorneal ring segments and advanced contact lens technologies. In a study from Italy’s Corneal Transplant Epidemiological Study, there was a 27-percent reduction in corneal grafting for keratoconus from 2002 to 2008.12 Much of the decline started after 2004 when cross-linking was becoming more common in Europe.

The complications that arise from crosslinking are usually related to the removal of the corneal epithelium; they include delayed healing, corneal haze, scarring and infection. In one study of cross-linking complications, researchers studied 117 eyes of 99 patients. The percentage of eyes losing two or more Snellen lines was 2.9 percent. The failure rate of CXL (which they defined as the percentage of eyes with continued progression) was 7.6 percent. They determined that a high preoperative maximum keratometry reading was a significant risk factor for failure. Sterile infiltrates were seen in 7.6 percent of eyes and central stromal scars in 2.8 percent. The researchers found that patient age older than 35 years and preoperative corrected distance vision better than 20/25 were significant risk factors for complications.13

The classic sign of inferior steepening on topography in a patient suffering from keratoconus.


Reimbursement Issues

The biggest challenge with corneal collagen cross-linking in the United States currently is the reimbursement model. Initially, cross-linking was done as a cash-pay procedure, but since FDA approval, more insurance carriers have begun to cover it. I participated in a telephone conference with the Avedro team to discuss my experiences with the UCLA model as well as the steps Avedro has taken to help facilitate this shift.

Avedro reported that in the second half of 2018, coverage increased to 62 health insurance plans. That coverage now includes 190 million patients, or more than 90 percent of U.S. covered lives. However, billing challenges remain for both the procedure and reimbursement for the riboflavin solutions. Previously, Photrexa and Photrexa Viscous had a miscellaneous J code. Recently, CMS issued a permanent J code, J2787, effective January 1, 2019, which will help to streamline the process and avoid the erroneous rejections that can occur with a manual review. Billing errors can also arise from the procedure code being a category 3 code rather than the normal category 1 code. The process of securing the CPT code is going to take longer to finalize and may be years away. The procedure is also not reimbursed when used on patients under the approved minimum age of 14.

Since it’s a new procedure, and many insurers are requiring manual reviews, a lot of rejections have resulted from an insurer’s staff not being properly informed about the status of cross-linking, which they mistakenly think of as experimental. There’s been some improvement lately, however. Avedro recently created a program called ARCH in order to provide information and resources for navigating coverage problems with commercial payers and improve patient access.  

In addition to the ARCH program, Avedro has hired employees to provide guidance in physicians’ offices, if need be. These new positions include payor relations directors who work with the insurance companies to ensure coverage and positive reimbursement, and field reimbursement managers who are up-to-date on the current regional payor landscape. The latter can work with a physician’s back office and billing staff to ensure proper coding and submission, so reimbursements are processed correctly.

There will certainly continue to be hiccups along the way, but moving to an insurance payment model will make cross-linking available to a larger population, possibly including the pediatric/adolescent population who need the procedure the most, but whose parents may not be able to afford the out-of-pocket cost. 


Future Directions

To avoid complications and increase patient comfort, especially in the pediatric population, there’s a lot of interest in developing an epithelium-on technique. The concern with this approach, however, is whether the riboflavin solution can achieve enough penetrance to fully saturate the cornea, and whether the UV light can penetrate adequately, as well.14 To overcome this barrier, a couple of different techniques have been attempted, including iontophoresis15 and disrupting the epithelium with benzalkonium chloride.16 Studies of epi-on crosslinking have thus far shown that corneal measurements stabilize for a period of time, but then tend to progress after about a year.17 A recent meta-analysis in the Journal of Cataract and Refractive Surgery concluded that for now, epi-off crosslinking should remain the standard of care.18

Yellow/green flare in the anterior chamber after the 30-minute “soaking” period during cross-linking.

Another modification to cross-linking that would benefit the pediatric population is an accelerated protocol, so the entire procedure could be performed in much less than the hour currently required. The idea of the accelerated protocols is to deliver the same amount of total energy (5.4 J/cm2) but use higher ultraviolet intensity for shorter periods of time. This means that you would need 30 mW/cm2 for a three-minute treatment, 18 mW/cm2 for a five-minute treatment and 9 mW/cm2 for a 10-minute treatment.19 

Some of the studies looking into accelerated protocols have also been accompanied by shorter soaking times, ranging from as short as 10 minutes to the normal 30 minutes. In a review by Marcony Santhiago, MD’s group, they note that many of the studies have a short follow-up period, so it’s still difficult to assess the longevity of the accelerated protocol. As an indirect marker, the demarcation line has been used to signal the depth of penetrance of the treatment and a marker of successful treatment. In the standard protocol, the demarcation line is usually found around 300 to 350 µm deep. In accelerated treatments the demarcation line can be as shallow as 100 to 150 µm, which Dr. Santhiago’s review found in a study of the three-minute protocol. It was only by expanding the total treatment to 14 minutes with a 9 mW/cm2 treatment delivering 7.5 J/cm2 that the accelerated protocols were able to deliver a demarcation line at depths comparable to the Dresden protocol.19

In conclusion, corneal collagen cross-linking is changing the paradigm for the treatment of keratoconus and corneal ectatic disease. It’s crucial to treat patients at the youngest age at which progressive keratoconus is diagnosed, to help prevent them from needing keratoplasty or other advanced interventions in the future. REVIEW


Dr. Bert is an assistant professor of ophthalmology at the Doheny Eye Center of UCLA, Doheny and Stein Eye Institutes, David Geffen School of Medicine.

He has no financial interest in Avedro’s cross-linking products.

1. Kennedy RH, Bourne WM, Dyer JA. A 48-year clinical and epidemiologic study of keratoconus. Am J Ophthalmol 1986;101:3:267-73.

2. McMahon TT, Shin JA, Newlin A, Edrington TB, Sugar J, Zadnik K. Discordance for keratoconus in two pairs of monozygotic twins. Cornea 1999;18:4:444-51.

3. CDER Rare Disease And Orphan Drug Designated Approvals. Accessed November 2018.

4. Tuft SJ, Moodaley LC, Gregory WM, Davison CR, Buckley RJ. Prognostic factors for the progression of keratoconus. Ophthalmology 1994;101:3:439-47. 

5. Center for Drug Evaluation and Research Application Number: 203324Orig2s000 Summary Review. Accessed November 2018.

6. NDA 203324. Accessed November 2018.

7. Palmert MR, Boepple PA. Variation in the timing of puberty: Clinical spectrum and genetic investigation. J Clin Endocrinol Metab 2001;86:2364-2368

8. Gomes JA, Tan D, Rapuano CJ, Belin MW, Ambrósio R Jr, Guell JL, Malecaze F, Nishida K, Sangwan VS; Group of Panelists for the Global Delphi Panel of Keratoconus and Ectatic Diseases. Global consensus on keratoconus and ectatic diseases. Cornea 2015;34:4:359-69.

9. Li Y, Tan O, Brass R, Weiss JL, Huang D. Corneal epithelial thickness mapping by Fourier-domain optical coherence tomography in normal and keratoconic eyes. Ophthalmology 2012;119:12:2425-33.

10. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-A-induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol 2003;135:5:620-7.

11. Zotta PG, Diakonis VF, Kymionis GD, Grentzelos M, Moschou KA. Long-term outcomes of corneal cross-linking for keratoconus in pediatric patients. J AAPOS 2017;21:5:397-401.

12. Frigo AC, Fasolo A, Capuzzo C, et al. Corneal transplantation activity over 7 years: Changing trends for indications, patient demographics and surgical techniques from the Corneal Transplant Epidemiological Study (CORTES). Transplant Proc 2015;47:528–35. 

13. Koller T, Mrochen M, Seiler T. Complication and failure rates after corneal crosslinking. J Cataract Refract Surg 2009;35:8:1358-62.

14. Li W, Wang B. Efficacy and safety of transepithelial corneal collagen crosslinking surgery versus standard corneal collagen crosslinking surgery for keratoconus: A meta-analysis of randomized controlled trials. BMC Ophthalmol 2017;17:1:262.

15. Bikbova G, Bikbov M. Standard corneal collagen crosslinking versus transepithelial iontophoresis-assisted corneal crosslinking, 24 months follow-up: Randomized control trial. Acta Ophthalmol 2016;94:7:e600-e606.

16. Rush SW, Rush RB. Epithelium-off versus transepithelial corneal collagen crosslinking for progressive corneal ectasia: A randomised and controlled trial. Br J Ophthalmol 2017;101:4:503-508.

17. Shalchi Z, Wang X, Nanavaty MA. Safety and efficacy of epithelium removal and transepithelial corneal collagen crosslinking for keratoconus. Eye (Lond) 2015;29:1:15-29.

18. Kobashi H, Rong SS, Ciolino JB. Transepithelial versus epithelium-off corneal crosslinking for corneal ectasia. J Cataract Refract Surg 2018;44:12:1507-1516.

19. Medeiros CS, Giacomin NT, Bueno RL, Ghanem RC, Moraes HV Jr, Santhiago MR. Accelerated corneal collagen crosslinking: Technique, efficacy, safety, and applications. J Cataract Refract Surg 2016;42:12:1826-1835.