Endothelial keratoplasty offers a targeted surgical solution

OSN’s 25th anniversary coverage continues with a look at the innovations that have propelled EK through numerous refinements.

Endothelial keratoplasty offers a targeted surgical solution for patients with endothelial dysfunction who previously needed to undergo full- thickness penetrating keratoplasty. The procedure is still evolving and continues to make gains in efficacy, safety and rapidity of visual healing.

The procedure’s advantages over traditional PK can be divided into two categories, according to OSN Cornea/External Disease Section Member Thomas John, MD.

“First, the fact that we’re taking the full-thickness corneal wound and sutures out of the equation has advantages. With PK, because of the wound, significant astigmatism occurs. Second, with PK, the wound never heals entirely. So if the patient has an accident or trauma, that’s the spot where the eye will rupture,” Dr. John said.

The procedure is rapidly garnering favor among corneal surgeons, according to OSN Cornea/External Disease Section Member Francis W. Price Jr., MD.

“Traditionally in the U.S., Fuchs’ dystrophy probably accounted for only 10% to 15% of the grafts. If you look at what’s being done with [endothelial keratoplasty] EK right now, about 90% being done are primarily for Fuchs’,” he said.

Beginnings

The selective replacement of just the posterior cornea was first successfully performed in a human in 1956 and described in the literature by Charles Tillett, MD, in an issue of the American Journal of Ophthalmology. Dr. Tillett performed “posterior lamellar keratoplasty,” as he called it, on a patient and reported a clear cornea at 1-year postoperative.

“He made a 12-mm limbal incision, dissected the stroma halfway across the cornea, and then lifted that edge up and completed his pocket dissection across the cornea, and then used scissors to remove the posterior tissue,” Mark A. Terry, MD, said. Dr. Tillett’s technique was largely similar to modern EK surgery, except for the large incision, he added.

“He accomplished the same thing [as we do today] in terms of getting the central surface of the cornea to be untouched by incision or sutures,” he said.

Research in animals continued over the next several decades. In 1993, Ko and colleagues presented a revised EK technique in a rabbit study at the Association for Research in Vision and Ophthalmology annual meeting. They described the “first concept of a truly limbal-based method of endothelial transplantation,” Dr. Terry said.

“They used a smaller 7- or 8-mm incision, made a pocket, cut out the back part with scissors and sutured the tissue in place. It was similar to what Dr. Tillett did, but they didn’t create as much of a corneal flap,” Dr. Terry said. Their grafts were successful in roughly half of the rabbits.

In 1999, Gerrit Melles, MD, PhD, a corneal surgeon based in the Netherlands, revived the term posterior lamellar keratoplasty in a case study that fundamentally altered the way surgeons performed EK.

“He was the first to describe taking a posterior disc of the patient’s cornea through a scleral pocket incision. The pocket is created within the patient’s cornea, limbus-to-limbus. Then he trephined the posterior part of that split cornea 360° and then removed the disc,” Dr. John said.

“Then he surgically created a donor disc consisting of the posterior corneal stroma, Descemet’s membrane and the healthy endothelium all in one disc. This disc was brought into the anterior chamber of the patient’s eye and placed to close the space that was surgically created in the patient’s cornea, much like closing a manhole lid,” he said.

Dr. Melles’ main innovation, however, was using an air bubble to aid in stabilizing the donor tissue on to the posterior bed of the recipient cornea, obviating the need for sutures entirely, Dr. Terry said.

“With this new technique, Dr. Melles was able to avoid the two inherent liabilities of PK surgery — surface corneal sutures and vertical stromal wounds,” Dr. Terry said.

Dr. John described Dr. Melles’ technique as a “huge step forward,” but noted that it did not become popular among corneal surgeons.

“It was technically more demanding. You had to create this pocket without splitting the patient’s cornea onto the surface or into the anterior chamber. Because if that happens, you have to finish the procedure by doing a full-thickness cornea graft,” he said.

Dr. Terry cited further reasons for the procedure’s impracticality.

“[Dr. Melles] used fresh whole eyes for the donor tissue. Here in the U.S., we don’t routinely have access to whole globes. We use corneal-scleral caps that have already been prepared by the eye bank. The second thing is that [Dr. Melles] used an air bubble to do his dissection. ... Cutting out the back part of the cornea with the chamber filled with air and keeping the chamber from collapsing while you’re cutting the tissue out was enormously difficult and a testament of his surgical skill,” he said.

DLEK

In response to these limitations, Dr. Terry began to refine Dr. Melles’ technique for wider acceptance.

“Since I knew the technique that he was using with fresh whole globes could not be propagated around the world in that fashion, we were the first people to ever work with an artificial anterior chamber in EK surgery. We developed the technique by mounting the corneal cap tissue on the artificial anterior chamber and preparing the tissue in the lamellar fashion for this type of transplantation. That opened up the procedure to the U.S.,” he said.

Dr. Terry also used the viscoelastic Healon (sodium hyaluronate, Advanced Medical Optics), instead of air, to assist when cutting out the posterior cornea.

“It made the procedure so much easier to cut out the back part of the cornea without having to worry about the chamber collapsing, getting iris damage or lens damage or having air escape,” he said.

Dr. Terry called the revised procedure deep lamellar endothelial keratoplasty (DLEK) and performed the first surgery in the United States in March 2000.

“But the procedure was difficult, even after we simplified it,” Dr. Terry said. “The DLEK and the [posterior lamellar keratoplasty] procedure were being done by just a handful of surgeons for about 2 years until Dr. Melles came up with stripping of recipient Descemet’s membrane, without the need to dissect and remove the posterior stromal tissue.”

With this latest shift, the “lid/manhole” relationship between the donor disc and recipient cornea was no more.

“The disc was placed on the inner corneal surface and was only held with the air bubble, much like taking a pizza and slapping it onto the ceiling,” Dr. John said.

In avoiding the need to form a pocket, Dr. Melles had provided another “key innovation” that made the procedure much easier, Dr. Terry said. “The popularization of this procedure took place at that point because it was easier,” he said.

Counterclockwise descemetorhexis 360° without exiting the anterior chamber. Images: John T

DSEK

However, Dr. Melles’ descemetorhexus procedure also posed new challenges. To facilitate clean removal of Descemet’s membrane only within the area that would be covered by the donor graft, Dr. Price said, he developed a modified Sinskey hook to lightly score the membrane along the perimeter of the area to be removed.

“In my early cases beginning in late 2003, donor graft adhesion was much more challenging compared with DLEK, so I kept refining the procedure to help reduce the dislocation rate,” Dr. Price said.

“First we increased the length of time the eye was completely filled with air to 8 minutes, plus had patients remain face up in the recovery area for an additional 30 to 60 minutes with a partial air bubble. Residual fluid in the graft interface seemed to interfere with adhesion, so I began using a LASIK roller to gently massage fluid from the interface out into the anterior chamber. Finally, I found that placing four small fenestrations in the peripheral cornea could help drain fluid from the interface,” Dr. Price said.

Dr. Price called the revised procedure Descemet’s stripping endothelial keratoplasty (DSEK). He also discovered that EK could successfully treat failed penetrating grafts and provide patients with faster recovery compared with a standard PK regraft.

“The question that remained [with DSEK] is will the tissue stick as well or will it fall off since you’re not making a pocket for the tissue to fit up into,” Dr. Terry said.

In a recent paper published in Cornea, Dr. Terry examined with scanning electron microscopy the histology of dislocation and found that stripping Descemet’s membrane produces a glassy smooth surface that is less conducive to adhesion than the donor pocket used in DLEK.

“So what we have been advocating to [Descemet’s stripping automated endothelial keratoplasty] DSAEK surgeons is that you a take an instrument and you scrape the peripheral edge of your recipient bed,” Dr. Terry said. “Leave the center pristinely smooth for vision, but if you scrape the peripheral bed to create white fibrils of stroma, tissue will grip on the edges. This technique to promote donor adhesion can reduce your dislocation rate.”

He added that in his series of his first 200 DSAEK patients, he used peripheral scraping and achieved a dislocation rate of only 1.5%.

DSAEK

Mark Gorovoy, MD, further established EK when he began using the Moria automated microkeratome to prepare donor tissue. Previously, surgeons had been using a manual dissection technique. In 2006, Dr. Gorovoy published a retrospective study of 16 patients who received microkeratome-dissected donor corneas and reported a rapid visual recovery. With the addition of the microkeratome, he renamed the procedure DSAEK. In a large consecutive case series, Dr. Price demonstrated that visual recovery was accelerated with DSAEK.

“By utilizing the microkeratome to prepare the donor tissue and eliminating all manual lamellar dissections, the surgical skill level required for DSAEK … was significantly reduced from what was required for [posterior lamellar keratoplasty] or DLEK surgery, and this procedure gained further general acceptance,” Dr. Terry said.

Contrary to popular belief, he noted, the automated microkeratome was first used for EK in DLEK surgery.

The DSAEK procedure is undergoing further investigation by looking at the possibility of using the femtosecond laser for the donor dissections rather than the microkeratome. According to Dr. Terry, however, the results have not been initially as good as were expected.

“The problem is the femtosecond laser cuts smoothly in the anterior cornea, but it cuts rough in the posterior cornea. Also, the femtosecond laser docks with the cornea, flattens it and applanates the cone to the cornea,” he said. “When you flatten the cornea, it pushes the posterior cornea into folds and ends up having the laser cut across these folds.”

“The femtosecond laser is being further developed by different companies, not just IntraLase, and has undergone modifications to see if it can produce a smoother interface in the posterior cornea. That’s going to be key,” he said.

DMEK

Dr. Melles has further evolved the EK procedure with a technique he calls Descemet’s membrane endothelial keratoplasty (DMEK).

“In DMEK, only Descemet’s membrane is transplanted [without donor stroma]. As a result, the visual rehabilitation is much quicker, and the final [best corrected visual acuity] is better,” Dr. Melles said.

According to Dr. Melles’ Web site (www.niios.com), the majority of endo- thelial disorders are limited to Desce-met’s membrane and the endothelium, so limiting the transplantation to these two layers would be most efficient.

Dr. Melles and colleagues evaluated DMEK in 10 patients with Fuchs’ endothelial dystrophy or pseudophakic bullous keratopathy. In all cases, surgeons created a 3.5-mm clear corneal tunnel incision, filled the anterior chamber with air and stripped Descemet’s membrane from the posterior stroma. A donor roll, 9 mm in diameter, was harvested from a donor corneo-scleral rim, placed into a recipient anterior chamber, positioned onto the posterior stroma and secured with an air bubble.

At 1-month postop, six eyes had a BCVA of 20/40 or better, and three eyes were 20/20 or better. The donor tissue completely detached in three eyes, which surgeons corrected with a secondary DSEK procedure.

“DMEK may have potential to become the most preferable technique to manage corneal endothelial disorders because it provides quick and nearly complete visual rehabilitation,” Dr. Melles and colleagues wrote.

What the future holds

“Thinking further out into the future, EK is probably just a temporary measure,” Dr. Terry said. “Eventually with bioengineering and with gene therapy, we’re looking at injecting endothelial cells into the anterior chamber directly and making them stick and populate the endothelial surface.” The technique has already been successfully performed in rabbits, he said.

“And then the final step of EK evolution would be gene therapy, where we don’t do any surgery at all. We just inject a drug into the anterior chamber that causes the endothelial cells to divide and repopulate on their own,” he said. “Now that scientists have mapped the human genome, I think that’s where everything is eventually headed.”

For more information:

• Thomas John, MD, can be reached at 16532 S. Oak Park Ave., Tinley Park, IL 60477; 708-429-2223; fax: 708-429-2226.
• Gerrit Melles, MD, PhD, can be reached at Laan Op Zuid 390, 3071 AA Rotterdam, The Netherlands; 011-31-10-297-4444. Dr. Melles is a consultant for Dutch Ophthalmic USA.
• Francis W. Price Jr., MD, can be reached at Price Vision Group, 9002 N. Meridian St., Suite 100, Indianapolis, IN 46260; 317-844-5530; fax: 317-844-5590. Dr. Price is a consultant for IntraLase.
• Mark A. Terry, MD, can be reached at Devers Eye Institute, 1040 NW 22nd Ave., N-200, Portland, OR 97210; 503-413-6223; fax: 503-413-6937.

References:

• Gorovoy MS. Descemet-stripping automated endothelial keratoplasty. Cornea. 2006;25:886-889.
• Melles GR, Lander F, et al. Posterior lamellar keratoplasty for a case of pseudophakic bullous keratopathy. Am J Ophthalmol. 1999;127:340-341.
• Melles GRJ, Ong S, Ververs B, et al. Descemet membrane endothelial keratoplasty (DMEK). Cornea. 2006;25:987-990.
• Price FW, Price MO. Descemet’s stripping with endothelial keratoplasty in 50 eyes: A refractive neutral corneal transplant. J Refract Surg. 2005;21:339-345.
• Price FW, Price MO. Descemet’s stripping with endothelial keratoplasty (DSEK) in 200 eyes: Early challenges and techniques to enhance donor adherence. J Refract Surg. 2006;32:411-418.
• Price FW, Price MO. Descemet’s stripping with endothelial keratoplasty: Comparative outcomes with microkeratome- and manually-dissected donor tissue. Ophthalmology. 2006;113:1936-1942.
• Terry MA, Hoar KL, Wall J, Ousley P. Histology of dislocations in endothelial keratoplasty (DSEK and DLEK): A laboratory-based, surgical solution to dislocation in 100 consecutive DSEK cases. Cornea. 2006;25:926-932.
• Terry MA, Ousley PJ. Deep lamellar endothelial keratoplasty in the first United States patients: Early clinical results. Cornea. 2001;20(3):239-243.
• Tillet CW. Posterior lamellar keratoplasty. Am J Ophthalmol. 1956;41:530-533.

• Andy Moskowitz is an OSN Staff Writer who covers all aspects of ophthalmology.