Refinements in techniques and tools lead to better-targeted corneal transplant surgery.

By Marianne Doran, Contributing Writer for EyeNet

Lamellar keratoplasty came into its own during the last decade, with a flurry of new procedures, technical variations and acronyms. Now, with increasing experience, corneal transplant surgeons are refining promising techniques and altering or abandoning procedures or devices that have not lived up to initial expectations.

As corneal grafts become thinner, development continues on donor insertion devices to protect the delicate tissues. At the same time, researchers are pursuing improvements in stem cell transplants and keratoprostheses to restore vision in patients for whom keratoplasty is not an option.



With recent waves of innovation in lamellar keratoplasty, it’s easy to overlook the continuing importance of its progenitor, penetrating keratoplasty. According to the Eye Bank Association of America, PK still accounted for more than half of the corneal transplants performed in the United States in 2010 (see “Corneal Transplant Procedures, 2010”).

Looking ahead over the next 10 years, Donald T. H. Tan, MD, medical director of the Singapore National Eye Centre, predicted that “PK will always be with us, as there are still many conditions that present late in the disease, in which the entire cornea is diseased or scarred and anterior or posterior replacement will be inadequate.”

Still, PK has well-known drawbacks, including irregular astigmatism, possible wound dehiscence and graft rejection. Thus, the goal of lamellar procedures is to replace just the diseased corneal layers while minimizing the wound size and the impact on healthy tissues.



Posterior corneal transplant procedures became a practical reality about 10 years ago with the advent of deep lamellar endothelial keratoplasty (DLEK). Although DLEK was a major step forward, it requires time-consuming, difficult corneal dissection and specialized instruments and is now rarely used. Today, the standard of care in posterior corneal transplantation is Descemet’s stripping endothelial keratoplasty (DSEK; also known as Descemet’s stripping automated endothelial keratoplasty, or DSAEK). But a newer procedure, called Descemet’s membrane endothelial keratoplasty, and its variants are showing excellent visual results and may prove to be less technically daunting than originally expected.

DSEK Grows in Success and Acceptance. “DSEK is technically easier than DLEK, and it produces better results,” said Christopher J. Rapuano, MD, professor of ophthalmology, chief of the cornea service and codirector of the refractive surgery department at Wills Eye Institute in Philadelphia. “While DSEK is not a perfect surgery, it has significant advantages over a full-thickness corneal transplant when the patient’s anterior cornea is reasonably normal.” Dr. Rapuano noted that DSEK has gained ground very quickly in the United States and around the world and that many modifications have been made over the years, resulting in a better, safer and more reproducible procedure.

Ivan R. Schwab, MD, cornea surgeon and professor of ophthalmology at the University of California, Davis, agrees. “The best technique is DSEK. It’s a marvelous procedure and is really much safer than penetrating keratoplasty. It is so good that it is replacing PK all over the United States and probably in much of the world.”

DMEK Goes Even Thinner. The latest entry into the field of posterior corneal transplantation is Descemet’s membrane endothelial keratoplasty (DMEK; Figs. 1–4), which preserves the recipient’s stroma and replaces only Descemet’s membrane and the endothelium, yielding a thinner graft than with DSEK. Dr. Rapuano noted that the theoretical advantage of DMEK is faster visual recovery and better vision postoperatively.

“Vision is definitely better with DMEK because it’s a thinner graft,” said Francis W. Price Jr., MD, medical director of the Price Vision Group in Indianapolis. “The thicker the tissue, the more significant the wrinkling and distortion, even if you don’t see it with the slit lamp. I believe this is the biggest reason we don’t achieve as many 20/20 visual results with DSEK.”

Overcoming technical challenges. “The problem is that the procedure is technically difficult, not only in obtaining the tissue but also in managing graft dislocations and the need for multiple rebubblings,” said Dr. Rapuano. “For me, and for many cornea surgeons, DMEK is not yet ready for prime time. But that is how DSEK was viewed seven or eight years ago. So it may well be that DMEK is the wave of the future, once the kinks are worked out.”

Dr. Price acknowledged the significant learning curve associated with DMEK but compared it to the difficulty of mastering any innovative and challenging technique.

Donor preparation. “The first problem we had with DMEK was donor preparation. Removing Descemet’s membrane from the back of the donor cornea without tearing it was difficult, and it took us a while to figure out how to do that,” Dr. Price said. “But a number of us who are doing a sizable volume of DMEKs are now finding that our donor loss rate is less than 1 percent. It’s not perfect, but we have pretty much solved that issue with some newer techniques.”

Insertion devices. Another hindrance to DMEK’s wider acceptance is the belief among many cornea surgeons that donor tissue must be inserted with glass tubes—not plastic IOL cartridges—to avoid damaging the endothelial cells. But Dr. Price noted that his group is getting excellent cell counts even when using plastic cartridges. “That was a big surprise because the donor Descemet’s membrane curls up with the endothelium on the outside, leaving the endothelium exposed to whatever is next to it. But we are doing well even with these plastic cartridges.”

Air reinjection. Dr. Price added that surgeons have overcome most of the hurdles with DMEK except for the need to reinject air a number of times to get the very thin graft to stick in place. “We haven’t solved that yet, but I feel confident that someone will. As more people start doing DMEK, one of them will find that missing link.”

In the meantime, Dr. Price maintains that endothelial cell counts with DMEK are as good as those with DSEK and that patients’ vision is significantly better. “Some surgeons contend that vision is not that much better with DMEK or that if you wait four years, DSEK eyes will achieve the good vision that DMEK patients have at six months—but most of my patients don’t want to wait four years to have that good vision,” Dr. Price said.


CORNEAL TRANSPLANT PROCEDURES, 2010.Statistics from the Eye Bank Association of America show the distribution of transplant procedures performed in the United States in 2010.


Penetrating keratoplasty (21,970)

Endothelial (lamellar) keratoplasty (19,159)

Deep anterior lamellar keratoplasty (1,041)

Keratoprosthesis (KPro) (342)

Keratolimbal allograft transplantation (130)

SOURCE: Eye Bank Association of America Statistical Report



Acceptance Grows, but Technical Issues Remain. In deep anterior lamellar keratoplasty (DALK), the surgeon removes all corneal tissue down to Descemet’s membrane and the endothelium. Thus, DALK is a potential alternative to PK for patients with keratoconus, ectasia, stromal dystrophies and scarring, offering the potential benefits of better graft survival and improved structural integrity of the cornea. Yet despite these advantages, many cornea surgeons remain hesitant to adopt the procedure.

Endothelial benefits. David D. Verdier, MD, clinical professor of surgery at Michigan State University College of Human Medicine in Grand Rapids, maintains that there is a compelling reason to do DALK in the appropriate patient. “Doing a PK using donor endothelium will increase the rate of cell loss and can lead to transplant rejection, whereas performing a successful DALK procedure enables patients to keep their own endothelium—and these corneas are very likely to last a lifetime.”

Dr. Rapuano added that the biggest advantages of DALK are the lack of endothelial rejection—and, ideally, lack of damage to endothelial cells during the surgery—and less need for topical steroids after surgery. Theoretically, the steroids can be decreased more quickly, leading to fewer steroid-related side effects.

Tricky technique. “The problem with this surgery is that it continues to be technically difficult to safely bare Descemet’s membrane,” Dr. Rapuano said, “and many cornea surgeons say that DALK is the most difficult of the corneal transplant operations they attempt. Even though we often attempt DALK, we may not achieve a good ‘big bubble’ and then cannot proceed with the ideal DALK procedure. Even if you get the big bubble, you may occasionally perforate Descemet’s membrane at some point in the procedure and need to switch to penetrating keratoplasty.”

However, encouraging results were shown in a recent retrospective interventional case series of 241 eyes with keratoconus. DALK was successfully completed in 234 of these eyes, with only seven requiring conversion to PK.1

Proceeding with caution. Although most cornea surgeons in the country are not yet performing DALK, Dr. Verdier said that “we may be at the point with our keratoconus and ectasia patients who require a graft where we were with Fuchs endothelial dystrophy or other endothelial problems in 2004 or 2005. In a few years, cornea specialists and comprehensive ophthalmologists will be seeing more DALK patients.”

Dr. Verdier said that an ophthalmologist will see a striking difference between an eye that has undergone lamellar keratoplasty and one that has had PK (Fig. 5). But the difference is not as dramatic with DALK (Fig. 6), because these patients still have problems with irregular astigmatism, and their rehabilitation and visual outcomes are comparable to those of patients who have undergone PK.

“This is why I believe it is going to take a little longer for DALK to catch on,” Dr. Verdier said. “It’s still a long process that requires multiple sutures, and the advantage of the procedure—the fact that these eyes are going to last longer—is realized over the long term.”