Cornea
Inserters reduce EK-related endothelial cell loss

At a glance: Protecting
endothelium

• Switching from forceps to inserters & minimal manipulation of tissue
prevents damage to endothelium
• Shortened incision size and insertion technique also minimize endothelial damage

Performing endothelial keratoplasty with forceps has similar long-term cell loss as PK, but taking several measures including using an inserter should improve EK results

W. Barry Lee, M.D., Atlanta, performs DSEK
Source: W. Barry Lee, M.D.

Improving technology has allowed for surgeons to perform a layer-specific keratoplasty, or selective keratoplasty, said Sadeer B. Hannush, M.D., assistant professor of ophthalmology, Jefferson Medical College, Philadelphia, and attending surgeon, Cornea Service, Wills Eye Institute, Philadelphia..
For example, if the patient has an endothelial problem such as Fuchs’ dystrophy or the endothelium is compromised by a condition such as pseudophakic corneal edema or pseudophakic bullous keratopathy, then the preferred procedure is endothelial keratoplasty (EK), said Dr. Hannush, who will be presenting on this topic at the World Cornea Congress in Boston in April. Dr. Hannush is the chair of The Cornea Society, which sponsors the meeting.
In other examples, if the patient has ectasia, then the physician might prefer deep anterior lamellar keratoplasty (DALK). If the patient has a condition that affects the corneal surface, then the surgeon might consider limbal stem cell transplantation.
As far as techniques for harvesting the donor tissue, Descemet’s stripping endothelial keratoplasty (DSEK) and Descemet’s stripping automated endothelial keratoplasty (DSAEK) “are here to stay for a while,” Dr. Hannush said. “[Descemet’s membrane endothelial keratoplasty (DMEK)] is not yet at a point where it is ready for prime time because of the difficulty handling the tissue. And ex-vivo preparation of endothelial cells for infusion into the anterior chamber is still a while away.”

Cell loss and EK

“The less you touch the endothelium, the better. That is the bottom line,” said Francis Price, Jr., M.D., Price Vision Group, Indianapolis, and president of the board, Corneal Research Foundation of America, Indianapolis. “So the gold standard is the standard graft. And when you do that, you usually place the graft on a bed of viscoelastic and sew it in place—a good surgeon can do that with minimal damage. This is why in the first six months postoperatively, there is minimal cell loss as described in the corneal donor study published in Ophthalmology in May 2008. But then there is a lot of cell loss after that period.”
In contrast, in DSEK/DSAEK, there is rapid cell loss in the first six months—usually about 30% cell loss—but after that, there is very little cell loss, Dr. Price said.
W. Barry Lee, M.D., Atlanta, confirmed Dr. Price’s data citing an American Academy of Ophthalmology-sponsored paper that was released in September and that he helped author.
In that review of more than 30 studies of EK and PK, Dr. Lee said they found that average cell loss at six months post-operatively was 37% with EK and 42% at one year. PK varies, but it ranges from 20% to 35% at six months and 30% to 60% at one year. While those percentages are not always significantly different, Dr. Lee notes that almost all of the EK procedures were done with forceps, so those numbers might change overtime as they look at more inserter device techniques.
Dr. Price said the first surgeons to perform DSEK were doing so in December 2003, so they are just now gathering five-year follow-up data. The May 2008 Ophthalmology article examined two-year data and what was pretty amazing was that despite the more than 30% cell loss after six months, at one year, it only increased a couple of percentage points, and then even more slightly after two years. “We hadn’t expected that; it almost levels out,” Dr. Price said. “And we are seeing that it is pretty close to that at three and four years so far, too.”
One qualification to keep in mind is that most of the DSEK patients have Fuchs’ dystrophy, and those patients do better than patients with pseudophakic bullous or glaucoma. The corneal donor study had a larger number of patients that did not have Fuchs’ dystrophy. Another is that this data is reviewing some of the earliest cases of DSEK, and since then, the procedure has undergone some changes to improve the cell loss rates, as Dr. Lee also stated.
Most prominently, surgeons are switching away from forceps and shortening the incision length. “So as we’ve improved the techniques, what we’re also seeing is that the initial drop at six months is markedly decreasing,” Dr. Price said.

Improving cell loss rates

In addition to moving away from forceps and shortening the incision size to about 4.5 to 5.5mm, Dr. Hannush had some other tips, but they all fall under the broad heading of minimizing manipulation of the tissue.
For example, a common cause of cell loss for physicians preparing their own donor is collapse of the artificial chamber. Others might use a gentian violet surgical mark on the stromal side, it is alcohol based, and alcohol will seep through the stroma and damage the endothelium to a degree. Dull trephine blades also can damage the endothelium.
When inserting the tissue into the anterior chamber, if the chamber collapses or flattens, the endothelium could get damaged, Dr. Hannush said. If the pupil is dialated and there is an IOL in the posterior chamber, the endothelium could make contact with the IOL resulting in damage to the endothelium. And using an instrument on the endothelium side to position the donor will damage endothelium.
“There are other ways to avoid damaging the endothelium, but primarily it is the incision size and insertion technique,” Dr. Hannush said. But above all, Dr. Hannush said addressing insertion is the key to improving cell loss rates, and he thinks the question will be resolved in the next 12 to 18 months when eventually all American surgeons that perform DSEK will be using an inserter.
“We seem to cause the least amount of damage to the endothelium if we use some kind of insertion device where the endothelium doesn’t touch anything else.” His experience with inserters has been limited to the Neusidl Corneal Inserter (NCI, Fischer Surgical, St. Louis), which has received Food and Drug Administration (FDA) approval. He added there are about six inserters that have either received FDA approval or are in their trials.
Adding DMEK to the mix
While neither Dr. Lee nor Dr. Hannush have tried DMEK, Dr. Price has been doing the procedure for about one year and a half, and there is a paper being published in Ophthalmology at the time of this reporting in December.
Dr. Price said he is finding that the cell loss is in the range of DSEK cases, but he added that like the early DSEK cases, this procedure still is in the earliest developmental stages, so future initial cell loss rates could improve as well.
“Considering that doing DMEK is just Descemet’s membrane—and that thing is so fragile, so hard to handle that trying to manipulate it in the eye is pretty challenging—we were pretty surprised our cell counts were that good,” Dr. Price said.
The biggest advantage with DMEK is that they think they will get better visual acuities. Physicians are not screening donor tissue for its shape, so when they are placed in the eye with a DSEK graft, in most cases, it is not going to have the same curvature as the back of the patient’s eye. Therefore, the tissue has to mold to the back of the patient’s cornea resulting in small folds and wrinkles. The thicker the stroma, the more apparent the folds and wrinkles will be.
With DSEK, we are not getting our patients to 20/20 very often; most tend to reach 20/30 or 20/40. With DMEK, at three months, they are finding that about 50% of the eyes that don’t have retinal problems to 20/25 or better. And with Descemet’s membrane automated endothelial keratoplasty (DMAEK), that increases to more than 70%. He said that there is a chance DSEK could have similar outcomes if the cuts get thinner, but that needs to be researched more.
In Dr. Lee’s opinion, so far the early studies on DMEK have shown a higher primary graft failure rate and higher endothelial cell loss, so he is waiting for the technique to become more standardized before he is ready to try it.
Dr. Hannush expressed his own concerns as well, he will consider adding the procedure this year. “I am not extremely motivated to go to DMEK right now because our results with DSEK are really excellent,” he said.

Editors’ note: Drs. Hannush, Price, and Lee have no financial interests related to their comments.

Contact Information

Hannush: 215-752-8564, SBHannush@comcast.net
Lee: 404-351-220, 0003lee@aol.com
Price: 317-814-2823, wendymickler@pricevisiongroup.net