A new thermokeratoplasty procedure holds promise for treating keratoconus.
 

While thermokeratoplasty has been used to reduce refractive error in the past, prior procedures have been plagued by problems with regression, lack of predictability, and competition from other options. With the development of a new thermokeratoplasty technology that delivers a precise, annular pulse of microwave energy, however, this concept is now being revived as a possible treatment for keratoconus and/or mild-to-moderate myopia. Still in the early testing phases, the Keraflex® procedure (Avedro, Inc.) is being evaluated both as a stand-alone procedure and for use in combination with collagen crosslinking.

Like other thermokeratoplasty procedures, Keraflex works by shrinking corneal collagen. In this case, the treatment involves applying microwave energy to an annulus of tissue in the corneal mid-periphery, thereby flattening the cornea. Unlike past procedures for hyperopia—for example, holmium laser thermokeratoplasty and conductive keratoplasty—which worked by applying focal spots in the periphery of the cornea, the size and shape of the Keraflex procedure is geometrically different. The choice of optical zone, diameter, and configuration determines the corneal shape change, thus allowing the Keraflex procedure to have a flattening effect. Depending on the intensity and configuration of the microwave energy, this procedure can yield corrections between roughly 1.00 D and 5.00 D.

The Surgical Procedure

Because success with Keraflex depends on centration, the first step in the procedure is to define the geometric center of the cornea. (I also like to note the first Purkinje image in order to locate the corneal apex.) The surgeon then marks the apex with a Sinskey hook and a gentian violet marking pen.

Next, the surgeon centers and locks the handheld targeting stage, which is very similar in its application to a standard microkeratome suction ring. X-Y dials are then used to center the device precisely over the centration mark, and the microwave handpiece is inserted into the targeting stage and engaged.

The interface on the handpiece then cools the cornea to a preset temperature—cooling sets standard baseline parameters and helps to make the procedure reproducible between patients—after which the circular electrode applies the microwave treatment energy. The electrode is computer controlled and automatically delivers the treatment to the right depth. Energy is applied in a ring that is approximately 4.5 mm in diameter and between 200 and 500 microns wide; the depth of the treatment is about 150 microns. The actual application of microwave energy takes less than a second, after which suction is disengaged and the instrument is removed.

Because this treatment affects the epithelium as well as the underlying stromal tissue, patients will have a small annular defect in the epithelium following the procedure. The epithelium typically heals within 24 hours, but non-preserved artificial tears and topical nonsteroidal antiinflammatory drugs (NSAIDs) may be used to ensure patient comfort during this period; a bandage contact lens may also be applied. Antibiotic drops are used to prevent infection.

Combining Keraflex and Collagen Crosslinking

Used alone, the Keraflex procedure should benefit keratoconus patients by flattening the cornea and possibly reducing corneal asymmetry. To achieve long-term improvement in vision, however, the eye must be stabilized so that further progression of the patient’s keratoconus does not negate the effect of treatment. With this goal in mind, the Keraflex procedure is being tested in conjunction with collagen crosslinking, in the hope that together these procedures will be able to flatten the cone, regularize the corneal topography, and then stabilize the cornea in its new geometry.

Unlike a standard collagen crosslinking procedure (in which treatment is applied to the entire cornea), the collagen crosslinking procedure being developed for this application targets just the area of the corneal mid-periphery where the microwave treatment was performed (Figure 1). After

completing the Keraflex procedure as described above, the surgeon uses a dry sponge to debride the resi

dual epithelial cells overlying the treatment zone. An ultraviolet-blocking mask is then applied to protect the untreated central and peripheral cornea, leaving just the microwave-treated area of the corneal mid-periphery exposed. Finally, riboflavin drops are administered to the annulus of exposed stroma every 2 minutes for 30 minutes, after which ultraviolet light is applied to the area for 30 minutes.

In addition, adjunctive use of the standard pan-corneal crosslinking technique is being investigated.

Early Results

More research is needed to evaluate the efficacy of this combined procedure, but my hope is that the Keraflex procedure and collagen crosslinking will work synergistically. Laboratory studies have shown that collagen crosslinking increases the rigidity of corneal tissue, and work by John Marshall has shown that the Keraflex procedure also increases corneal rigidity.1 As a result, use of both procedures together may be able to further slow or halt disease progression, as well as reduce the myopia and astigmatism already produced by the disease.

While the Keraflex procedure has only been tested in a handful of patients to date, early results appear promising. In a series of seven patients who were treated in Turkey and followed for 2 months, investigators found that the Keraflex/collagen crosslinking combined procedure achieved an average of 6.0 D of corneal flattening and a mean change in manifest refraction spherical equivalent (MRSE) of 4.39 D. No data is available regarding possible astigmatic corrections, but regularity of the corneal topography appears to be improved in some cases (Figure 2). Also, while additional follow-up will be needed to assess long-term changes in refraction, topographies and visual acuity appear stable so far.

Keys to Success

Like any procedure, Keraflex requires proper patient selection to achieve optimal outcomes. Currently, the Keraflex procedure is being evaluated in keratoconus patients who have a potential for good visual acuity but are experiencing some optical problems that cannot be easily corrected with glasses or contact lenses. Patients with corneal scars are not good candidates for Keraflex, as they are likely to need a penetrating or lamellar keratoplasty to achieve good vision, and patients with thin corneas in the area of microwave application (midperipheral thickness < 400 microns) are also unsuitable for the procedure, since treating these eyes could result in complications or unpredictable refractive changes.

The Keraflex procedure should have a good safety profile, but there are still a few risks that must be considered. The procedure affects only a small area of corneal epithelium, so the riskof complications related to the epithelial defect is modest; nonetheless, care should be taken with patients who are at high risk of infection or delayed epithelial healing. Also, any change in corneal conformation presents a risk of optical side effects—such as glare, halo, or diplopia—so patients need to be educated about this possibility. While the goal of the Keraflex procedure is to reduce the incidence of such symptoms, there is always the possibility that patients may develop new dysphotopsias.

Finally, while the Keraflex procedure is currently being evaluated only for the treatment of keratoconus, future applications may include treatment of myopia and/or astigmatism. As with keratoconus, treatment of spherical myopia would involve uniform flattening of the cornea. To treat astigmatism, however, different treatment patterns would be needed.

Keraflex® Microwave Thermokeratoplasty: A New Era in Corneal Refractive Surgery David Muller, PhD, MBA

EDITORS NOTE:Over two decades ago, David Muller led the team that brought the first FDA-approved excimer laser to mar ket. Now, he has founded Avedro to develop a new procedure that offers similarly exciting potential.

When we began talking publicly about the Keraflex® procedure, the focus was on the collagen reshaping aspect of the procedure, but it has been our intent from the outset to combine corneal reshaping with collagen stabilization. It’s no secret that the refractive effects of earlier thermokeratoplasty procedures were short-lived, but with collagen crosslinking, we now have a procedure that can “lock in” the effects of the thermokeratoplasty to provide the needed stability.

A Robust Procedure

We believe the Keraflex procedure is extremely robust. Not only does collagen crosslinking provide a mechanism to lock in the refractive change, but Keraflex is capable of far more significant refractive corrections than prior thermokeratoplasty procedures. For example, where conductive keratoplasty (CK) provides a few diopters of myopic shift at best, Keraflex has been able to produce more than 6 D of refractive change in normal corneas.

In addition, because the Keraflex procedure doesn’t cause the scarring found in prior thermokeratoplasty procedures, it can be used to shrink collagen inside the visual axis. This enables us to create collagen shrinking procedures that either steepen or flatten the cornea, depending on where the microwave heating is applied. A band of collagen shrinkage outside the central 6.0 or 6.5 mm causes the cornea to steepen (as in the CK procedure); a ring of collagen shrinkage with a diameter less than 4.5 or 5.0 mm causes corneal flattening.

This flattening is relatively uniform across the cornea. Unlike LASIK, which requires a blend zone around the optical zone, Keraflex produces a relatively consistent flattening effect from limbus to limbus. Because of this benefit, and because there is no scarring, we haven’t seen dysphotopsias with Keraflex, even though the collagen shrinking takes place within or adjacent to the visual axis. Equally important, the cornea remains prolate, and there is typically no loss of contrast sensitivity.

A Promising Procedure

We have now received the CE mark, but we recognize that the Keraflex procedure is still in its very early phases. What we have shown so far is that the procedure can yield a relatively accurate correction, and in the few normal patients who have also been crosslinked, that correction appears to be very stable. (Without crosslinking, the effect regresses.) Will crosslinking stabilize the Keraflex procedure? While there are still no long-term results with large cohorts of Keraflex patients, we now have considerable data on what crosslinking can do in ectatic corneas, and, although it remains to be proved conclusively, it is reasonable to expect the same stabilization in the case of thermokeratoplasty.

In early clinical trials, we have been able to get most of our patients to within 0.5 D of target. We have found, too, that the relationship between energy delivered and refractive correction is almost linear, making it relatively easy to design a correction. As someone who took part in the development of the excimer laser, I think it fair to say that it took us several years of trials with that technology to get to the level of predictability that microwave thermokeratoplasty has achieved in its first few tests.

Targeting Myopes

Although we are starting with keratoconus correction, we believe that in the long run the largest market for the Keraflex procedure will be vision correction for low and low-to-moderate myopes.

Although laser vision correction produces exceptional outcomes in this range, the fraction of low myopes who have had laser vision correction is relatively small. However, our market studies tell us that many of the low myopes who have not come in for LASIK may consider Keraflex because we will be able to offer a non-cutting procedure. Even the fact that the procedure isn’t necessarily permanent can be a benefit in this market—unless we lock the result in with crosslinking, any unsatisfactory outcome will likely reverse itself in a relatively short time.

Because we believe that our market consists primarily of low myopes with no interest in LASIK, we don’t see this procedure as competition for LASIK; rather, we see it as a complement to laser vision correction. The low-myope population is huge, and it’s largely untapped. To succeed, all we need to do is tap a small fraction of this group.

Current Status

Myopia correction trials in normal eyes are going on now in Turkey. We are launching Keraflex as a treatment for keratoconus in Europe, and at the same time we plan to start a multicenter myopia trial in Europe as a prelude to the similar multicenter trial we expect to eventually perform in the US.

With a platform that can both reshape the cornea and lock that new shape into place, I believe that we are at the beginning of a new field. The upside potential is enormous. In 5 or 10 years, there will be different types of crosslinking agents, different initiators, and different inhibitors, and by then we will have learned how to control the corneal reshaping process with great precision. We are at the dawn of a new technology that can provide precision reshaping of the cornea, without cutting it or inserting something into it. And unlike current refractive procedures and other experimental procedures currently being investigated, the Keraflex procedure strengthens rather than weakens the cornea.

Thermokeratoplasty procedures have been a source of repeated disappointments, so skepticism about a new one is understandable. But I can recall some of the reactions in the early days of PRK when we talked about using a laser to correct ametropias on otherwise healthy eyes. Some people were appalled by the idea, and many more were skeptical. But today roughly a million laser vision corrections are performed annually in the US—it has become a common, routine, and relatively easy surgery. I firmly believe that something very similar will happen with Keraflex.

David Muller, PhD, is president and CEO of Avedro, Inc., Waltham, MA.

The Bottom Line

Keraflex is a new thermokeratoplasty procedure being tested for the treatment of keratoconus. The Keraflex procedure uses microwave energy delivered in a small annulus to the corneal mid-periphery. This procedure can flatten the cone and may also help to regularize the cornea. By combining the procedure with a modified collagen crosslinking procedure, investigators hope to improve the stability of the eye as well as regularize its geometry. While still in clinical trials outside of the United States, early results with this technology appear promising.

Peter S. Hersh, MD, FACS, is director of the Cornea and Laser Eye Institute-Hersh Vision Group, professor of clinical ophthalmology and chief of cornea and refractive surgery at UMDNJ-New Jersey Medical School, and a visiting research collaborator at Princeton University. He is also a medical monitor and paid consultant for Avedro, Inc.  Refractive Eyecare managing editor Kay Downer assisted in the preparation of this manuscript.

References

1. John Marshall, personal communication.