Surgical treatments range from cross-linking to keratoplasty when keratoconus progresses

Many clinicians are limited to two options when treating keratoconus surgically: either intrastromal corneal ring segments or more aggressive keratoplasty. A third alternative, corneal cross-linking, has been reserved for clinical trials in the U.S., with FDA approval of the technology pending as of the press date for this issue.

Meanwhile, there are new diagnostic devices for early detection.

“Keratoconus is an easy diagnosis when it is manifest,” George O. Waring IV, MD, FACS, OSN Technology Board Member, said, noting that the disease is typically both a clinical and topographic diagnosis. “The challenge arises, however, when keratoconus is pre-manifest or forme fruste keratoconus. These are partly genetic and partly environmental etiologies that lead to this biomechanical disorder.”

Because there is no commercially available genetic test for keratoconus, clinicians must rely on ancillary testing modalities and other diagnostics, which historically have been related to topography and tomography.

“But our interest is in measuring the strength of corneal tissue directly, which has been difficult to do with any degree of specificity due to the limitations of our diagnostic devices,” Waring said.

One diagnostic product that evaluates biomechanics directly is the Ocular Response Analyzer (Reichert Technologies), which uses an infrared signal and an air puff to deform the cornea.

“Thus, we can observe the deformation and the reformation of the cornea. This helps us better understand these behaviors in different patient populations,” Waring said. Waring assisted in validating the device and the accompanying software that evaluates the behavior of the cornea response due to deformation.

Waring and his group have also been validating the Corvis STL (Oculus), which uses a Scheimpflug video to capture the corneal deformation and reformation responses to an air puff.

“We are validating parameters for normal eyes, keratoconic eyes, eyes that appear to be at risk or have early forms of keratoconus, and observing the behavior in post-refractive surgery,” he said.

Image: Beckman KA

Surgical considerations

“While individual studies will define keratoconic progression differently, in general a main sign of progression is an increase in maximum keratometry,” OSN Cornea/External Disease Board Member Kenneth A. Beckman, MD, FACS, said. “Other signs may include increased thinning, striae, scarring and increasing astigmatism.”

In the absence of access to cross-linking, “when vision decreases to the point where the patient is having difficulty with daily activities, surgery may be performed,” Beckman said. “However, typically, there is first an attempt to improve the vision with contact lenses. If adequate vision is not attainable or if contact lens wear is not tolerable, then surgery is considered.”

Theoretically, cross-linking is most effective at stopping progression rather than reversing any changes, according to Beckman.

“Therefore, if a patient can be treated while they still have good vision, before significant progression occurs, they may have the best success,” he said. “This tends to be younger patients, as progression tends to stop in older patients.”

Nonetheless, there are some older patients who show signs of progression and may benefit from cross-linking, he said.

Beckman said that it is not uncommon for cross-linking treatment to lead to improvement, rather than simply stopping progression. “Hence, patients with advanced keratoconus may still benefit from treatment,” he said.

Other clinicians are impressed with cross-linking as well.

“My experience is that it is extremely effective, not only at stopping progression of keratoconus and post-LASIK ectasia, but also at helping improve vision and improving corneal shape,” OSN Technology Section Editor William B. Trattler, MD, said. Trattler is the director of cornea at the Center for Excellence in Eye Care in Miami, which has been involved with multiple clinical trials for cross-linking. In 2010, his center switched from epithelial-off to epithelial-on cross-linking because the latter technique is less invasive yet just as effective, he said.

Corneal cross-linking

Corneal cross-linking is a minimally invasive procedure that combines liquid riboflavin, UV light and oxygen in a single session of therapy to help strengthen the cornea. The procedure can be repeated in patients whose condition worsens typically 6 months to 2 years after the first session. Less than 1% of eyes progress after cross-linking, according to Trattler.

The advantage of the epithelium remaining on the cornea during cross-linking is that it makes the procedure noninvasive and no bandage contact lens is needed afterward.

“However, patients experience discomfort or pain the first evening. But by the next day, 97% of patients have pretty much returned to normal,” Trattler said. And because epi-on is a safer and easier procedure to perform than removing the epithelium, it can be performed in children and older adults, as well as in patients with very steep corneas.

Epi-off has a longer track record than epi-on but has a little more risk. These risks include infection, corneal haze and delays in epithelium healing. Despite the benefits of epi-on, though, the commercially available riboflavin formulations available internally have yet to be optimized for epi-on treatment, Trattler said.

“Some formulations are able to penetrate through the epithelial cells while others are not,” he said. His practice, which is part of the CXLUSA study, uses a riboflavin formulation optimized for epi-on and has therefore treated exclusively with epi-on since 2010.

According to Trattler, one of the major challenges is identifying patients with keratoconus early. The slit lamp exam is typically normal in mild to moderate keratoconus. He recommended having a low threshold to order a topography, because this will allow for the identification of more patients with early keratoconus who can have cross-linking before they have experienced significant loss of vision.

“Pulsating UV light also appears to be helpful,” he said. Although 3 mW of continuous UV light for 30 minutes is the traditional protocol, “many clinicians have been increasing the energy and shortening the exposure time,” Trattler said. “However, better outcomes with these high-energy/shorter UV treatment times have yet to be determined.” Trattler also advised that a critical step is confirming that the cornea is saturated in riboflavin before proceeding to the UV light exposure.

Of the available surgical options to treat keratoconus, David R. Hardten, MD, FACS, OSN Cornea/External Disease Section Editor, is most excited about corneal cross-linking, in part because of its potential to reduce disease burden.

“By identifying keratoconus early, we can prevent the disability that patients incur from the irregular astigmatism and, therefore, the problems encountered later on that require keratoplasty or corneal rings or hard contact lenses, for example. Some of these alternative treatments are not as low risk as we would like,” Hardten said.

Corneal ring segments

Hardten considers cross-linking to be the first-line surgical intervention for keratoconus, followed by corneal ring segments. Currently, Intacs corneal implants (Addition Technology) are the only corneal ring segment approved by the FDA for use in the U.S.

A review article of corneal ring segments that appeared in Current Opinion in Ophthalmology found that although “patient response is variable, shape improvement can be obtained in many patients to allow contact lens tolerance again,” Hardten said.

“Corneal ring segments may be effective in changing the corneal curvature to achieve a more regular surface,” Beckman said. “With this treatment, a patient may be able to either attain a better correction with spectacles or an easier fit for contacts to improve vision and delay the need for keratoplasty.” However, unlike cross-linking, ring segments “do not stop progression,” he said. “Some surgeons are using ring segments in combination with cross-linking.”

According to Hardten, a patient who is suitable for a ring is contact lens intolerant but would return to contacts with a 20% improvement in corneal shape. Another criterion is that the patient will have already undergone cross-linking or be scheduled for the procedure around the same time as corneal ring segment placement.

The plastic 150° segment, available in thicknesses up to 0.45 mm, “acts kind of like a splint in the cornea. The segment pulls back and stretches out the bulging area of the cornea,” Hardten said. The segment is typically placed at a corneal depth between 75% and 80%, normally in the lower section of the cornea with a thicker segment and often with a thinner segment on the superior portion of the cornea.

“Most surgeons nowadays create the channels for these segments at that 75% to 80% depth with a femtosecond laser,” Hardten said.

At Hardten’s practice and at several other physician offices known to him, the 10-minute ring segment insertion procedure is often augmented by placing conductive keratoplasty spots to shrink the bulging areas of the cornea for increased effectiveness. Cross-linking is usually scheduled either the same day of or a few days after conductive keratoplasty. “By just placing the ring segment, there is still some risk of progression of the disease,” Hardten said.

The success rate for a ring segment is quite high for Hardten, who began implanting rings in the 1990s.

“Most patients are able to return to a contact lens with comfort and avoid a keratoplasty,” he said, although keratoplasty remains an option if keratoconus continues to progress.

Keratoplasty

The third line of surgical therapy is keratoplasty, reserved for the patient who is unable to achieve visual correction with glasses/contacts, a corneal ring or cross-linking.

“My preferred initial keratoplasty, if technically feasible, is deep anterior lamellar keratoplasty,” Hardten said. For most younger patients, Hardten favors a femtosecond laser-assisted wound, such as an IntraLase-enabled keratoplasty (Abbott Medical Optics), in an attempt to improve the wound strength of the peripheral graft-host interface and reduce traumatic wound dehiscence later in life.

“Keratoplasty is a definitive treatment suitable for patients who are unable to attain a corrected vision that allows them to perform their activities of daily living,” Beckman said. “In many cases, though, even if the cornea has progressed to the point where keratoplasty may be done, cross-linking may be an appropriate initial treatment, if available, as it may allow enough improvement to avoid keratoplasty.” And if cross-linking fails, “it does not eliminate the opportunity to perform keratoplasty at a later date,” he said.

For grafts in general to treat keratoconus, the two mainstay choices are a full-thickness penetrating keratoplasty or a partial-thickness anterior lamellar keratoplasty, according to Jack S. Parker, MD, a Young Ophthalmologist member of the OSN Europe Edition Board.

“Although PK has a much longer history of more than 100 years, the simple substitution of the patient’s cornea with an identical cornea or a normal cornea from a donor invites a lot of potential problems,” he said. “Sutures are sources for potential infection, and they distort the cornea in a myriad of unpredictable and difficult-to-manage ways.”

Parker also said that at the interface between the transplant and the patient’s eye, “there is a permanent weakness that is significant.” In fact, roughly 5% of patients end up at some point in their life having their head or eye knocked and the graft becoming dehisced or expelled from the surface of the eye.

“Although PK is described as an effective option to restore vision in patients with keratoconus, the downside of the transplant is that the visual improvement frequently comes at the cost of the obligation to manage a litany of very serious potential complications,” Parker said.

ALK, conversely, is considered safer and may provide better outcomes, Parker said. DALK, in particular, is safer than PK because “it leaves part of the patient’s cornea intact, so the cornea is more stable,” he said. “Stitches also remain in the eye for a briefer period of time.”

Parker was one of the authors of a 2014 article in JAMA Ophthalmology that described a midstromal isolated Bowman’s layer graft to reduce advanced keratoconus.

“This surgical technique is very different from DALK,” Parker said. “Rather than cutting off the top half of the cornea and sewing in a new top half, the Bowman’s layer transplant just slides a small graft into the middle of the patient’s cornea, so it does not remove any tissue. The graft simply provides a scaffold around which the recipient eye can heal and pulls the graft into a better configuration.”

A previously distorted cornea is regularized with the graft, “so it becomes flatter and more even, and therefore the patient is more able to wear a hard contact lens,” Parker said. Another advantage of the procedure is that it appears to halt the process of progressively irregular corneas. “The cornea stops its degenerative changes,” Parker said.

Since 2011, Parker has treated about 60 eyes with the Bowman’s graft in patients primarily with advanced disease. “However, we are now starting to treat patients with milder disease, too,” he said.

Other considerations

As for the differences in pediatric vs. older adult cases for detection and treatment of keratoconus, younger patients are likely to have more rapid progression. Children with risk factors, such as being eye rubbers or having allergies or Down syndrome, may benefit from early screening with topography.

“I also feel that young patients with new identification of unusual astigmatism or rapidly progressive changes in their refractive error are indicated for topography,” Hardten said. But apart from more attention given to early diagnosis so that the child can begin cross-linking sooner, the progression of management for keratoconus is the same as for adults, he said.

Beckman concurred about the need for early detection in children.

“Pediatric cases tend to progress more rapidly,” he said. “As one ages, there is a natural cross-linking effect in the cornea, and progression eventually stops. Catching younger patients at an earlier stage and then treating them with cross-linking can potentially preserve excellent vision and may be preferable to waiting until progression is demonstrated.”

“We know that keratoconus can behave much more aggressively in children,” Waring said. He has performed corneal cross-linking as first-line therapy in patients as young as 12 years old. “By definition, keratoconus is a progressive bilateral disorder, so if we have the opportunity to catch it early and keep it from progressing, we should, much like glaucoma.”

Likewise, as patients reach early adulthood and seek refractive surgical correction, “we want to be able detect corneas at risk to prevent the exacerbation of an underlying weakening disorder such as ectasia,” Waring said.

One of the at-risk groups for keratoconus is patients with sleep apnea, due in part to patients sleeping on their face, according to a 2012 article in Cornea.

“Education and early identification of keratoconus can help get these patients treatment with cross-linking at the early stages of the disease to reduce the morbidity associated with the disease,” Hardten said.

In May, at the annual meeting of the American Society of Cataract and Refractive Surgery, Waring is scheduled to present on diagnosing keratoconus using spectral-domain OCT.

“We feel that SD-OCT with epithelial mapping and stromal mapping properties can be very useful in the early detection of keratoconus,” he said.

Waring’s group has also proposed a novel morphologic index based on the epithelium-to-stroma ratio, which it believes is more meaningful than historic curvature indices.

“You can have a steep, thick cornea that is relatively strong and perhaps with stable irregular astigmatism, or you can have a thin ectatic cornea that is not very steep because of the reactive epithelial hyperplasia that is blunting the curvature changes,” Waring said.

Waring and his group have found that such curvature changes are undetected with traditional topographic analysis. “But by doing a morphologic analysis with SD-OCT, we can actually locate the thin, weak areas of the cornea,” Waring said.

Looking ahead

Waring said that Avedro is developing a novel device that uses the analysis of light scatter from Brillouin optical microscopy to calculate the elastic modulus of ocular tissue. “This looks very promising for early detection of keratoconus and understanding direct measurement of biomechanics,” he said.

Furthermore, having resubmitted a new drug application to the FDA for the riboflavin ophthalmic solution/KXL system for advanced cross-linking, Avedro was anticipating a Prescription Drug User Fee Act date this month.

If approved, “the NDA would be the first FDA approved therapeutic treatment for progressive keratoconus and corneal ectasia following refractive surgery,” according to a company press release.

Trattler, who was an investigator for the first Avedro clinical trial in 2008, which at that time failed to meet the trial’s primary endpoints, is encouraged for the FDA approval.

Cross-linking has become the standard of care to treat keratoconus around the world, except in the U.S., according to Beckman.

“Unfortunately, there are many patients here who could have benefited from cross-linking but instead have gone on to progression and need keratoplasty,” he said. “Once cross-linking is approved and becomes more widespread, we may see a generation of patients scheduling treatment early in the course and never progressing to keratoplasty.”

“As long as UV cross-linking remains prohibitive in the United States, or at least relegated to clinical trials only, it will be very difficult for patients with progressing keratoconus to receive the disease-arresting treatment they need,” Parker said. “Until then, we need something else to stop these diseases from becoming worse. Maybe we will be able to use Bowman’s layer transplant for that purpose because it does seem to arrest the course of the disease.”

Hardten referenced a comprehensive review of cross-linking detailing multiple studies published earlier this year in Cornea, underscoring “that there is still much to learn” and that “the exact technique that is best is still evolving.”

Hardten believes that the ophthalmic community can always do better in treating keratoconus.

“I would like us to continue to improve our vertical options,” he said. “But our hope for the future, even though we might have better surgical options for treating advanced keratoconus, is that those surgical processes will be used less and less because of the fact that we are doing a better job of identifying and managing patients at early stages of the disease to help prevent progression.” – by Bob Kronemyer

• References:
• Alhayek A, et al. Int J Ophthalmol. 2015;doi:10.3980/j.issn.2222-3959.2015.02.35.
• Amalich-Montiel F, et al. Eye Vis (Lond). 2016;doi:10.1186/s40662-016-0033-y.
• Beniz LA, et al. Arq Bras Oftalmol. 2016;doi:10.5935/0004-2749.20160009.
• Farjadnia M, et al. Oman J Ophthalmol. 2015;doi:10.4103/0974-620X.159105.
• Gupta PK, et al. Cornea. 2012;doi:10.1097/ICO.0b013e31823f8acd.
• McGhee CNJ, et al. Cornea 2015;doi:10.1097/ICO.0000000000000504.
• Meiri Z, et al. Cornea. 2016;doi:10.1097/ICO.0000000000000723.
• Park S, et al. Curr Opin Ophthalmol. 2013;doi:10.1097/ICU.0b013e3283622a2c.
• Spadea L, et al. Saudi Medical J. 2016;doi:10.15537/smj.2016.2.12985.
• Stulting RD. Am J Ophthalmol. 2012;doi:10.1016/j.ajo.2012.05.005.
• van Dijk K, et al. JAMA Ophthalmol. 2014;doi:10.1001/jamaophthalmol.2013.5841.
• Vinciguerra P, et al. Am J Ophthalmol. 2012;doi:10.1016/j.ajo.2012.03.020.

• For more information:
• Kenneth A. Beckman, MD, FACS, can be reached at Comprehensive Eye Care of Central Ohio, 450 Alkyre Run Drive, No. 100, Westerville, OH 43209; email: kenbeckman22@aol.com.
• David R. Hardten, MD, FACS, can be reached at Minnesota Eye Consultants, 710 E. 24th St., Suite 100, Minneapolis, MN 55404; email: drhardten@mneye.com.
• Jack S. Parker, MD, can be reached at Parker Cornea, 700 18th St. S., No. 503, Birmingham, AL 35233; email: jack.parker@gmail.com.
• William B. Trattler, MD, can be reached at Baptist Medical Arts Building, 8940 N. Kendall Drive, Suite 400-E, Miami, FL 33176; email: wtrattler@gmail.com.
• George O. Waring IV, MD, FACS, can be reached at MUSC Storm Eye Institute, 167 Ashley Ave., Charleston, SC 29425; email: georgewaring@me.com.

Disclosures: Beckman reports he was an investigator for two Avedro FDA clinical trials. Hardten reports he has been a speaker for Addition Technology and is a consultant for Abbott Medical Optics; he also participates in research with CXL-USA. Parker reports no relevant financial disclosures. Trattler reports he has a financial interest in CXLO. Waring reports he is an investigator for and on the medical advisory board of Avedro, as well as a consultant to AcuFocus and Visiometrics.

What are the advantages of epi-on vs. epi-off corneal cross-linking?

Epi-on reduces risk

If you were having a cross-linking procedure, assuming you did not have to, would you want your corneal epithelium scraped off? I have asked this question internationally on many occasions, for which I notice few raised hands for epi-off, assuming epi-on treatment is effective. However, not all epi-on proprietary technology is effective.

Assuming effective transepithelial technology, the only exception to preferring epi-on corneal cross-linking is in cases in which cross-linking is to be combined with another procedure that would require the epithelium to be removed, such as PRK. Any treatment in which you do not remove the epithelium is clearly less invasive and problematic than any treatment in which you do.

Effective epi-on treatment reduces or eliminates the risks of delayed epithelial healing, slow visual recovery, infection, haze, scarring and corneal perforation. Epi-off is also very painful.

With effective epi-on cross-linking, patients generally return to their normal activities the next day, along with vision that is the same as or better than their preop vision. This is in contrast to epi-off treatment, for which the downtime is on the order of 1 week or more, and the return to preop vision may take months. In addition, with epi-on, patients typically experience only 4 to 6 hours of discomfort compared with epi-off, which consists of roughly 4 to 7 days of discomfort, light sensitivity, foreign body sensation, tearing and the inability to drive or work. My epi-on patients also require nothing stronger than a single dose of an oral nonsteroidal.

The advantages of a highly effective, noninvasive and safer transepithelial (epi-on) cross-linking procedure radically shift the risk-benefit ratio for treating patients with ectatic diseases. With increased safety and similar efficacy to epi-off, we see no reason to wait for progression and vision loss in a patient who has been diagnosed with ectasia.

Roy S. Rubinfeld, MD, is a clinical associate professor of ophthalmology at Georgetown University Medical Center. Disclosure: Rubinfeld reports a financial interest in CXL-USA LLC and CXL Ophthalmics LLC.

• References:
• Rubinfeld R, et al. Quantitative analysis of trans-epithelial corneal riboflavin loading. Presented at International CXL Congress; Dec. 5, 2015; Boston.
• Rubinfeld R, et al. Mastering transepithelial and epithelial-off corneal collagen crosslinking for keratoconus and post-LASIK ectasia. Presented at American Academy of Ophthalmology meeting; Nov. 15, 2015; Las Vegas.

Epi-off creates deep, strong demarcation

The use of corneal collagen cross-linking to halt progressive keratoconus was first described by Wollensak, Spoerl and Seiler in 2003. In the original technique, the epithelial debridement was performed (epi-off technique) before administering riboflavin. Cross-linking with the epi-off method has noticeably demonstrated its efficacy in arresting the progression of keratoconus. Advantages of the epi-off technique are the fast penetration and high concentrations of riboflavin in the anterior stroma. Therefore, a deep and strong demarcation line is observed: approximately 300 µm as opposed to 150 µm in epi-on cross-linking. The high absorption of UV light protects the endothelium, lens and retina.

In epi-off cross-linking, epithelial remodeling is often a reflection of stromal changes. Because the epithelium compensates for irregularities of the anterior stromal surface, postoperative analysis of epithelial thickness maps appears to be a sensitive method for identifying stromal changes after corneal cross-linking. However, clinicians must be aware that corneal topography maps may remain stable, or even appear worse, in the early postoperative period due to epithelial remodeling. In addition, the long-term changes in the epithelial thickness profile may have hidden the described changes of the stromal surface shape, thus leading to the observed mild changes in corneal topography. This dilemma highlights the importance of using epithelial thickness maps to detect and monitor stromal changes after cross-linking.

Karolinne Maia Rocha, MD, PhD, is the director of cornea service at Medical University of South Carolina, Storm Eye Institute. Disclosure: Rocha reports no relevant financial disclosures.

• References:
• Rocha KM, et al. J Refract Surg. 2013;doi:10.3928/1081597X-20130129-08.
• Rocha KM, et al. J Refract Surg. 2014;doi:10.3928/1081597X-20140120-08.
• Wollensak G, et al. Am J Ophthalmol. 2003;doi:10.1016/S0002-9394(02)02220-1.