Acknowledging more than 5 years of preliminary work in animals, in cadaver eyes, and in unsighted or partially sighted human eyes, the US Food and Drug Administration (FDA) has approved the use of the ArF excimer laser for correction of refractive error in a limited series of patients with normal corrected visual acuities. For the first time, the laser is being used on patients with no ocular pathology other than refractive error.
As one prominent ophthalmologist in the field of refractive corneal laser surgery remarked, at long last researchers "now have the chance to get to the right animal."
While this is a significant advance in the field of excimer laser corneal research, at the same time it surely forebodes further contention in an arena already fragmented by tremendous commercial pressures. When the mode of delivery for a promising new medical technology costs hundreds of thousands of dollars per unit, the risk of being hit by falling hype and flying patent challenges is high indeed.
Excimer lasers ("excimer" is a contraction of the phrase "excited dimer," and should probably be pronounced "ek-SIGH-mer," although only the Europeans seem to do that) were used originally for industrial purposes, such as etching computer chips. In the early 1980s, the laser's ability to ablate organic solids was noted by R. Srinivasan, PhD, of IBM's T.J. Watson Research Center. This capability drew the interest of Stephen Trokel, MD, at Columbia University, who in 1983 first described the use of the excimer laser to etch corneal tissue.
Although excimer lasers can produce energy at several wavelengths in the UV range, it is the 193-nm wavelength in the near-UV that has drawn most of the interest of Trokel and other medical researchers. Studies suggest that 193 nm is the safest of the frequencies produced by excimer lasers, in terms of mutagenic and carcinogenic effect on tissues, and most researchers seem to feel it is quite safe. The higher frequencies of 248 nm and 308 nm that excimers are capable of producing have shown mutagenic and carcinogenic effects in studies of incoherent light, and so are thought likely to pose the same threat in coherent form. This is not a settled issue, however, and toxicity studies continue.
The excimer laser works on tissue not by thermal energy, as most ophthalmic lasers do, but by breaking down chemical bonds, disintegrating molecules that are struck by its beam. This is the basis for its accuracy in removing small amounts of tissue. It is hoped that this mechanism of action will make possible eye surgery that can "fool" the cornea; that is, removal of corneal tissue without disruption of the surrounding tissue, so that no scarring or unwanted healing response is mounted.
Corneal surgeons are well aware that wound healing response can be a significant variable in the outcome of radial keratotomy and other refractive procedures. The prospect of removing exact amounts of tissue from the cornea without invoking a stromal healing response is therefore very attractive to the community of refractive surgeons.
As a result of this interest, the pursuit of workable systems for delivery of a medically useful version of the excimer beam has been intense. As Trokel told Refractive and Corneal Surgery recently, the excimer is "not a friendly laser system." It delivers a broad beam of light which, in its raw form, is not homogeneous; considerable processing is required to distribute the basic laser beam evenly enough to ablate tissue uniformly. A medical laser system comprises the basic laser along with its processing and delivery apparatus, which can be quite complex.
Once the beam has been made as homogeneous as possible, the challenge is to deliver the broad beam to the cornea in a shape that allows control over its effect. A number of centers in academia and industry, here and abroad, are working to refine delivery systems that provide useful ablation patterns. Among the approaches being investigated are computer-controlled constricting diaphragms that deliver different amounts of energy to different areas of the cornea, rotating wheels with apertures in graduated sizes to achieve a similar effect, corneal masks that permit energy delivery only to designated areas of the cornea, masks of varied thickness that ablate at the same rate as the cornea, and moving slits that pass a beam of laser light across the surface of the cornea.
As Trokel said recently, one can envision dozens of ways to deliver excimer laser energy to the cornea, from a moving spot to a broad beam. The challenge lies in deciding which is likely to be the most efficient and devising a way to make it work.
The process, no matter what method is used, is one of tissue subtraction. The object is to obtain a new corneal surface - one that is either curved differently or smoothed. There are three basic types of corneal laser surgery:
* Linear keratectomy. The laser can produce refractive changes by making linear or accurate "cuts" in the cornea, which are actually excised troughs. These excisions can be placed using the same principles as radial or transverse keratotomy, but researchers hope that the laser will serve to eliminate several variables inherent in knife-blade procedures, making the result more predictable. For one thing, whereas a knife plows through tissue and tends to distort it, the laser treats the entire length of the excision at one time, digging the trough uniformly deeper with each pulse. For another, the variable of the surgeon himself is, to a degree, eliminated, as a steady hand is replaced by an unwavering beam of light. At least in theory, the elimination of physical contact and of tissue distortion should produce a more predictable refractive outcome.
While the use of traditional keratotomy patterns (and perhaps, eventually, new patterns designed specifically for excimer surgery) has the advantage of keeping the laser outside the optical zone, it has the concomitant disadvantage that these procedures will rely on a translation of effect from the periphery of the cornea where the excision takes place to the center, where flattening or steepening is desired. There may still, then, be problems with biologic variability.
* Surface keratomileusis. The alternative to treating the periphery is to work directly on the optical zone, changing the cornea's refractive power by removing graded amounts of tissue. While this approach may be more direct it is also much more demanding, because the treated cornea must be as clear as it was before treatment. The issue of postlaser corneal haze has been discussed extensively; some centers say they almost never see haze, some say they almost always do.
It seems to be true that the more homogeneous the excimer beam is made, the less haze is seen. It also seems that deeper excisions show more tendency to haze than shallow ones.
Other concerns exist as well. Recent research shows that different layers of the cornea ablate at different rates. Hydration also affects ablation rate. Any attempt to produce accurate refractive change will have to take these factors into account.
In addition, unless very small optical zones are used, large area ablation for a large refractive change will surely mean going through Bowman's layer. What effect this will have on clarity and stability remains to be seen.
At last year's Second International Workshop on Laser Corneal Surgery, in Boston, Roger F. Steinert, MD, said that in order for surface keratomileusis to work, "we have to have minimal or no repair processes" following excimer surgery. If repair processes cause regression of effect, he said, or unpredictability and instability, then no matter how clear the corneas are, "it's not going to make it, clinically." The addition of "fudge factors" to a supposedly predictable procedure will make laser surface keratomileusis no better than other available surgical methods of refractive correction.
Steinert called for "zero wound repair response," because "we are talking about something that's going to be applied to normal corneas." He said excimer surgeons must set themselves "a surgical performance goal that is a higher standard than has ever been asked of any procedure anywhere."
* Lamellar surface keratectomy. Not all corneas treated by the excimer need be "normal," however. More than one center is investigating use of the laser to remove ocular surface pathology, to smooth the corneal surface. Pterygia, band keratopathy, chemical burns, dystrophies - all have been treated by application of the broad beam of the excimer.
So, who is on the cutting (or ablating) edge of excimer laser research? The field has grown at a tremendous rate since the concept of tissue ablation with the excimer was introduced only a few years ago. Three companies in the United States now have clinical units for sale.
US Ophthalmic Excimer Laser Companies
These companies are currently sponsoring clinical trials of their excimer laser systems: Summit Technology Ine of Watertown, Mass; Taunton Technologies Ine of Monroe, Conn; and VisX Ine of Sunnyvale, Calif. All are publicly traded corporations and all have a significant stake in the success of their products. (Note: The FDA holds the terms of its investigational device exemptions [IDEs] in strict confidence. Any information in this article about the scope of an IDE comes from company spokespeople).
VisX, of which Trokel is part owner and partner, has an enthusiastic team of investigators at Louisiana State University Eye Center, including Herbert E. Kaufman, MD and Marguerite B. McDonald, MD. They have moved fastest in advancing the status of their laser keratomileusis clinical trials and are the first group in this country to receive FDA approval specifically to perform surface keratomileusis on a series of eyes with normal corrected visual acuities.
Several people associated with VisX declined to tell this reporter how many patients were to be included in this series of "normal" patients; they said only that it was "more than ten."
VisX also has deployed one of its lasers at Johns Hopkins Hospital's Wilmer Eye Institute, where Walter J. Stark, MD, will begin treating a series of patients with corneal scars, infections and surface irregularities.
Taunton has also treated "potential 20/20" patients in the second phase of its clinical trial, which has now completed enrollment. According to Taunton president John W. Warner, PhD, the company's second phase was designed to be as broad as possible, including patients with corneal opacities as well as medium to high refractive errors. Taunton is in the process of applying to the FDA for permission to begin a third phase of its trial, which Warner hopes will lead to a series of "hundreds" of patients. Investigators for Taunton include Richard L. Lindstrom, MD, at the University of Minnesota, and J. James Rowsey, MD, at the University of Oklahoma. Daniel Taylor, MD, performed the first clinical trials with the Taunton instrument on partially sighted eyes.
The third major commercial player in the United States is Summit, which has a number of trials either in progress or about to begin. Under medical monitor John D. Hunkeler, MD, of the University of Kansas, trials of an ab externo glaucoma filtering procedure and a transverse keratectomy procedure for astigmatism have been in progress for over a year.
By the time this article sees print, two more of Summit's US trials will probably have commenced, according to the regulatory affairs manager at Summit. Lamellar surface keratectomy and surface keratomileusis will be investigated in separate trials, the first under medical monitor Daniel S. Durrie, MD. A monitor for the second trial has yet to be announced. Theo Seiler, MD has already used the Summit laser for approximately 20 of these procedures in Berlin and Michael Gordon, MD has done a small series in La Jolla, Calif.
VisX is engaged in a patent dispute with Taunton, which has licensed six patents granted to Francis L'Espérance, MD, of Columbia-Presbyterian Medical Center. Although Trokel, now with VisX, was the first to describe excimer corneal ablation, L'Espérance beat Trokel to the patent office by a matter of weeks.
Visx and Trokel have provoked several "interference actions" from the US Patent and Trademark Office (PTO) against sections of L'Esperance's patents. An interference is an internal review process at the PTO in which the records of both sides of a dispute are reexamined to try to determine who has the stronger claim to precedence. The review is proceeding, and no further action has been taken by either party.
Perhaps even more significant in the patent struggle is yet another patent issued late last year to researchers at IBM, covering methods and apparatus for ablation of tissue with ultraviolet radiation. This patent may supercede those of L'Espérance and Trokel, making their dispute moot. Industry sources say that IBM has expressed willingness to license its patents nonexclusively to all interested parties.
Summit has been a bystander in these disputes, but it has recently acquired a patent of its own, for an ablatable corneal mask to be used in its laser keratomileusis procedure, pending FDA approval. The mask will ablate at the same rate as corneal tissue, and will be shaped according to the refractive result desired. Thus, claims the manufacturer, the submicron "steps" that are etched into the cornea by the constricting diaphragm used in other excimer systems (and currently in Summit's own system) will be eliminated, leaving the cornea smoother. This new delivery mask is under investigation by George O. Waring III, MD, at Emory University in Atlanta.
Whether or not this makes a difference clinically, it should make a difference economically, for this will be the first disposable element to be associated with excimer refractive procedures. The one-use corneal masks could be a significant source of revenue for Summit, since the company plans to charge about $200 apiece for them.
In our next installment we will examine the latest data from these three companies' ongoing clinical trials. We will also turn our attention to some of the hype that has accompanied the release of early results because of commercial pressures involved, and introduce the thoughts of some excimer researchers on how these releases should be handled.