American Academy of Ophthalmology Web Site: www.aao.org
Lamellar Transplants: Forge a New Corneal Landscape
Partial-thickness corneal transplants have been performed on and off for decades. Now a convergence of factors is finally propelling the procedures forward.
The idea is simple and intuitive: Why transplant an entire cornea—risking significant astigmatism, unnecessary endothelial rejection or damage to the structural integrity of the globe—when only the diseased portion of the cornea needs to be replaced? This straightforward concept has proven difficult to put into practice until recently, largely because of disappointing visual results. Now, after a history of starts and stops, it looks as if recent advances in lamellar keratoplasty, both anterior and posterior, are the real deal.
“The major changes that are now occurring in corneal transplants are due to better equipment and newer ideas about how to perform lamellar keratoplasty,” said Ivan R. Schwab, MD, professor of ophthalmology and director of the cornea and external disease service at the University of California, Davis. “These will change the way we do corneal transplantations.”
PK Begins to Pale
Corneal transplantation celebrated its centennial year in 2005. Dr. Schwab noted that whereas the specialties of cataract and retina/vitreous have each undergone two or three paradigm shifts in the last three decades, corneal transplantation has changed very little. Penetrating keratoplasty (PK) has retained dominance over the years, despite its acknowledged drawbacks, primarily because of the technical challenges involved in anterior lamellar procedures and their problems with interface haze. The status quo was further bolstered by the liberal use of topical steroids that reduce the risk of rejection with full-thickness grafts and by the development of photorefractive procedures that made PK-associated astigmatism somewhat manageable.
Attractive lamellar edge. But lamellar keratoplasty offers several advantages over PK: Astigmatism is nearly eliminated, there are few or no sutures to promote infection or vascularization, the globe remains structurally sound, graft rejection is less likely and full functional vision returns within weeks to months instead of months to years. The trade-off has been lower Snellen visual acuity. With historical lamellar procedures, the average visual acuity is 20/40 to 20/50. But with the latest techniques, surgeons are reporting 20/20 vision in some patients. These techniques are technically challenging, and donor disc dislocation is a potential complication.
The Lamellar Story
Today’s posterior lamellar transplants (now increasingly referred to as endothelial keratoplasty, or EK) have their roots in pioneering work performed by Jose I. Barraquer, MD, in the 1960s. In these early partial-thickness grafts, the surgeon created an anterior flap, which was then retracted to permit the recipient posterior stroma to be trephined out. A posterior lamellar button from the donor was then positioned in the recipient bed, and the flap was laid back over the new tissue and sutured into place. This procedure ultimately fell out of favor because of irregular astigmatism, unpredictable corneal topography and problems related to the sutures or flaps. Interface haze was also a significant problem.
The next big advance arrived with little fanfare in 1993. In a poster presented at an Association for Research in Vision and Ophthalmology meeting, Wilson Ko, MD, and others reported success with a new scleral-limbal approach in a rabbit model. The surgeons made an incision in the sclera and tunneled deep into the cornea, where they were able to remove the endothelium without touching the corneal surface. The endothelium was then replaced with donor tissue.1
In 1998, Dutch surgeon Gerrit R. J. Melles, MD, PhD, became the first surgeon to perform this scleral-limbal procedure in humans. The Melles procedure, which he named posterior lamellar keratoplasty (PLK), involved a 9-millimeter sclerocorneal pocket incision and relied on an intracameral air bubble for the dissection and resection of recipient tissue and for placing and attaching the donor endothelium.
From PLK to DLEK. Galvanized by the progress made by Dr. Melles, Mark A. Terry, MD, set out to improve upon the procedure. “The technique Melles used was all done under air, without using any viscoelastic material,” said Dr. Terry, director of corneal services at the Devers Eye Institute in Portland, Ore. “It took tremendous skill and was a little risky. I also thought the instrumentation could be improved upon.”
Dr. Terry’s laboratory established that the viscoelastic Healon (Advanced Medical Optics) could be used to stabilize the anterior segment during surgery and still be completely removed without interfering with the graft adherence. His group also worked with Bausch & Lomb in redesigning the instrumentation to make the procedure easier and faster.
In 2000, after a year of experimentation, Dr. Terry became the first surgeon in the United States to perform the surgery, which he renamed deep lamellar endothelial keratoplasty (DLEK). “What we did in our laboratory was to continue to develop the procedure to make it more accessible to other surgeons,” Dr. Terry said. “I also wanted to establish it as scientifically valid by using a prospective, large-study protocol.”
To that end, he initially trained surgeons free of charge if they had an institutional review board–approved protocol at their institution. These surgeons formed the Endothelial Keratoplasty Group, which now has more than 100 members around the world who have done the surgeries under IRB-approved protocols.
Diversification, front to back. In the six years since DLEK was first performed, several new or improved EK procedures have evolved: posterior procedures that remove the endothelium and Descemet’s membrane; anterior procedures that take the epithelium and stroma. “There is a whole set of iterations of how deep, what level and how you do it,” noted Dr. Schwab. He suspects that when everything settles out, one or two procedures will remain.
The procedures developed within the last several years are highlighted below. The posterior endothelial keratoplasties are used primarily to treat Fuchs’ dystrophy and pseudophakic bullous keratopathy. Anterior keratoplasties are useful in treating keratoconus, scarring from refractive surgery or trauma, and stromal scarring from bacterial or viral infections.
DLEK. Dr. Melles introduced a new method for DLEK that used a smaller (5-mm) incision and donor tissue that was folded in half for insertion. His procedure proved the viability of folding the donor tissue but was associated with a loss of endothelial cell density higher than what he had shown in his prior studies of EK without folding of the tissue.
DSEK/DSAEK. Dr. Terry noted that as the field has evolved, surgeons are taking less and less tissue from the recipient bed. Now most practitioners are simply harvesting the donor posterior surface in a procedure called Descemet’s stripping. Francis W. Price Jr., MD, and Mark S. Gorovoy, MD, have been leading proponents of Descemet’s stripping with EK (DSEK).
The procedure is also known as Descemet’s stripping automated EK (DSAEK) when a microkeratome is used to prepare the donor. These three surgeons agree that DSEK/ DSAEK should be the standard of care for treating Fuchs’ dystrophy and postcataract corneal decompensation. It also may be valuable in some patients with a failed PK.
DSAEK relies on a Moria Automated Lamellar Therapeutic Keratoplasty (ALTK) microkeratome system to prepare the donor tissue. Descemet’s membrane and the endothelium are gently stripped from the recipient cornea. The donor disc of posterior stroma and endothelium are then inserted through a 5-mm incision in the sclera. An injected air bubble is used to unfold the donor tissue and press it into place. Use of the microkeratome is associated with faster visual recovery at one month and provides better depth control. This, in turn, helps prevent perforation due to donor tissue that is sliced too thin and reduces the risk of primary graft failure.
Other advantages of DSEK/DSAEK over DLEK include obviation of complex recipient trephination techniques, less potential for trauma to the anterior chamber and lens, less concern about inducing ectasia in patients with prior refractive surgery and the ability to perform corneal refractive surgery later on to correct refractive errors.
Dr. Price, medical director of the Price Vision Group and president of the board of the Cornea Research Foundation of America in Indianapolis, moved to the DSEK/DSAEK procedure after doing about 100 DLEK procedures. He has now performed more than 400 transplants with the DSEK/ DSAEK technique.
“The PLK/DLEK procedures involve not only dissecting the donor but also dissecting the patient’s eye,” he said. “That’s really a lot more complicated, and more things can go wrong. Most important, it is a hand dissection. We would have these two hand-dissected surfaces that we would put together, and the visual recovery wasn’t quite what we wanted.”
He added that the results he is obtaining with DSEK/ DSAEK at six months are better than anything that has been reported with PK, even going out to two years or more after surgery. “DSEK is all I offer my patients with Fuchs’ dystrophy, pseudophakic bullous keratopathy or other causes of endothelial dysfunction.
“We don’t get many people who are 20/20, which is something we are going to be working on to figure out why,” Dr. Price noted. “But if you look at the number of people who are 20/40 best corrected, it’s as good as or better than anything ever reported for penetrating keratoplasty in Fuchs’ dystrophy and pseudophakic bullous keratopathy. That’s really pretty exciting—and that’s without contact lenses.”
Dr. Gorovoy, a corneal surgeon in solo practice in Fort Myers, Fla., commented that he has used all of the different techniques and that DSEK/DSAEK has revolutionized his practice.
“A regular transplant patient takes one to two years to get decent vision, if at all, and a lot of patients never get good vision without wearing a rigid contact lens,” he said. “The DSEK/DSAEK patients, in general, are seeing 20/40 by six weeks, and 90 percent are seeing that by three months. More than 20 percent of my patients are seeing 20/20 by three months. They’re getting excellent vision.”
Recovery boom. Both surgeons pointed out that DSEK/ DSAEK not only offers a quicker recovery from corneal surgery, but it also allows patients to be treated earlier in the course of their endothelial disease. In the past, corneal transplants were put off as long as possible because of the disability that followed, and bilateral disease could take years to correct. Surgeons also were reluctant to operate on younger patients or on those who could refract better than 20/70.
With DSEK/DSAEK, Dr. Gorovoy said, “the level of care is getting so much better that we can be more aggressive in helping patients.” Dr. Price added that he now has patients who request that their second eye be done a month after the first procedure because their vision has improved so much.
Dislocation bust. “The biggest Achilles’ heel of this procedure is the dislocation rate,” Dr. Gorovoy said. “The good news is that you can put repeat air bubbles in, and these patients ultimately tend to do well.” He noted that in his first 100 patients, the dislocation rate was 25 percent, but it now averages about 10 percent. A frustrating aspect of dislocation, he said, is that it has been impossible to predict and can occur in surgeries that seemed to go perfectly.
Dr. Price believes that he has solved the dislocation problem with the use of new techniques to remove fluid from the donor/recipient interface. As reported at the Academy’s Annual Meeting last fall, he has had only one dislocation in his last 187 cases.
DALK. In deep anterior lamellar keratoplasty (DALK), surgeons remove everything on the anterior surface down to the endothelial layer and Descemet’s membrane.
Preempt rejection. “The concept is that when you excise everything except the endothelium and Descemet’s membrane, that leaves a layer of recipient tissue that will protect your transplant from being recognized by the host,” Dr. Terry explained. “You still get problems with astigmatism because you have to suture the anterior donor tissue in place. But if you can avoid rejection, that’s a big issue.”
John D. Sheppard, MD, associate professor of ophthalmology, microbiology and molecular biology at Eastern Virginia Medical School, in Norfolk, Va., said, “The decreased risk of rejection significantly reduces the need for steroid drops, theoretically reducing the risk of postoperative cataract and glaucoma. Those risks are also further increased the minute you surgically enter the eye but can be avoided altogether in a smoothly completed anterior lamellar procedure.”
Rajesh Fogla, MD, senior consultant and cornea and laser refractive surgeon at Apollo Hospitals in Jubilee Hills, Hyderabad, India, has been performing DALK for corneal scars, ectasia, dystrophies and degenerations since 1998. He began with the maximum depth technique (near Descemet’s membrane), with manual dissection of the corneal layers. For the past three years, Dr. Fogla said, he has used the “big bubble” technique of Mohammed Anwar, FRCS, and Klaus D. Teichmann, MD, with excellent results.
A bubble works well. “In the big bubble technique, partial-thickness trephination is followed by air injection into the corneal stroma to detach Descemet’s membrane from the stromal layers,” Dr. Fogla explained. “This is followed by removal of the anterior stromal tissue, exposing the smooth Descemet’s membrane.”
The donor tissue is punched out and its Descemet’s membrane is then peeled off with a pair of fine-tipped forceps. The remaining corneal tissue is placed on the stromal bed and sutured.
Dr. Fogla noted that the procedure results in a very smooth interface. In his patients with keratoconus, the average postoperative best-corrected visual acuity is 20/25, with a range of 20/20 to 20/50. These results are comparable to those achieved with traditional PK. Other dissection methods under investigation include using a balanced salt solution or viscoelastic material to separate Descemet’s membrane from the overlying tissue.
Maximum depth lamellar keratoplasties do carry a risk of microperforation, said Dr. Fogla. But in most instances, he added, the dissection can be completed after filling the anterior chamber with air.
Out, haze. With all lamellar procedures—anterior or posterior—minimizing interface haze is critical. “The jury is still out on the best way to avoid interface haze,” Dr. Sheppard said. “Obviously the smoothest, slickest cut, minimizing manual dissection, will be the one that does the best job. Intraoperative and postoperative topical wound healing response modifiers, including steroids and biotech drugs, will become another area of innovation for the pharmaceutical industry to optimize our surgical outcomes.”
A Turning Point
The recent advances in partial-thickness corneal transplants are making the procedures easier to perform and more accessible. In turn, surgeon interest in learning the latest techniques has soared, with some courses booked up nearly a year in advance. Nonetheless, more work needs to be done.
“These procedures are challenging for the surgeon, they have a steeper learning curve perhaps than a standard transplant, and there are still problems with interface haze and wound healing,” said Dr. Sheppard.
Dr. Schwab agreed. “I think the femtosecond Intralase may be instrumental in solving some of the problems of lamellar surgery,” he said. He compared innovations such as these to the project of turning an oil tanker. It doesn’t happen quickly. But he does believe that the conceptual shift in corneal transplantation is definitely under way.
“It’s in progress, and we’ll look back on it historically and say that this was the five- to 10-year period when the oil tanker changed direction,” Dr. Schwab added. “We’ll ask ourselves, ‘Why did it take us so long to do this? Why did it take a hundred years to make this change?’”