Corneal surgeons overcoming challenges, maximizing outcomes with DMEK

Endothelial keratoplasty has largely supplanted penetrating keratoplasty, or full-thickness corneal transplantation, as the gold standard treatment for corneal endothelial disease.

The leading endothelial keratoplasty technique is Descemet’s stripping endothelial keratoplasty or Descemet’s stripping automated endothelial keratoplasty. A newer iteration, Descemet’s membrane endothelial keratoplasty, is making steady inroads in corneal transplantation.

Experts attribute the growth of DMEK to various factors. First, unlike DSEK, in which the endothelium and Descemet’s membrane are removed from the recipient cornea and replaced with a donor disc comprising endothelium and posterior stroma, DMEK only involves replacement of tissue with donor endothelium and Descemet’s membrane.

“The advantage of DMEK over DSAEK or DSEK is that DMEK is pure anatomic replacement surgery so that you’re taking out Descemet’s membrane and replacing it with Descemet’s membrane,” Mark A. Terry, MD, said. “It’s always better to do exactly what nature would do rather than approximate it. For example, a posterior chamber lens is better overall than an anterior chamber lens. A DMEK graft is better overall than a DSAEK graft because it really is a perfect anatomic replacement.”

More importantly for patients and surgeons striving to maximize outcomes, DMEK offers better visual and refractive outcomes, faster visual recovery and lower graft rejection rates than DSEK, according to Thomas John, MD, OSN Cornea/External Disease Board Member.

“The driving force is all about the patient and the quality of vision and how quickly patients can get back to their normal lifestyle,” John said. “But, having said that, DMEK offers, among this subgroup of corneal transplantation, the best visual outcomes and the lowest graft rejection. As far as the patient is concerned, it’s comfortable and, overall, a very good experience.”

Francis W. Price Jr., MD, OSN Cornea/External Disease Board Member, said that DMEK has become more reliable and reproducible because of evolving techniques.

“Initially, when we first started doing DMEK, we were doing it because it improves the best corrected vision one or two lines from DSEK and uses a smaller incision. Just like with cataract surgery, the smaller the incision, the fewer problems you have with refractive shift,” Price said. “Then, after we had done this for a few years, we actually found that not only was the vision better, the rejection rate was tremendously less.”

DMEK vs. DSEK

Terry Kim, MD, OSN Cornea/External Disease Board Member, said that an appreciable number of cornea surgeons perform DMEK in addition to DSEK.

“I still think DSEK is leading the percentage of procedures among surgeons who are doing partial-thickness transplants,” Kim said. “I believe there are a lower percentage of surgeons who have converted to DMEK for 100% of their posterior lamellar grafts and a higher percentage of surgeons who are doing both, meaning performing DMEK selectively.”

David R. Hardten, MD, OSN Cornea/External Disease Section Editor, said DMEK comprises 80% to 85% of corneal transplantation cases in his practice.

“And then there are certain situations where I’ll do DSEK instead — for example, if they’ve had a tube shunt or a very active filtering bleb, a sutured IOL, an iris iridodialysis or an artificial iris, where the anterior chamber is not really trapped by the iris and an IOL or the iris and natural lens,” Hardten said. “Those are the cases that I’ll do DSEK on, as opposed to DMEK. But otherwise, I’ve gotten to the point where the vast majority of my cases are DMEK as opposed to DSEK.”

Hardten noted that he prefers to do DSEK in eyes that have undergone previous corneal transplantation.

John outlined challenging cases that may be more suitable for DSEK.

“Subsets of challenging patients would include those with glaucoma shunt tubes in the anterior chamber, synechial angle compromise or hyperopic eyes with less space in the anterior chamber, smaller corneal diameter, anterior chamber IOL or scleral-fixated posterior chamber IOL,” John said.

Kim said he primarily performs ultrathin-graft DSEK, which involves a graft with a thickness ranging from 60 µm to 90 µm.

“That’s a controversial area in terms of whether ultrathin-graft DSEK equals the outcomes of DMEK. But I find that these DSEK patients do very well visually. The surgical procedure is, in my opinion, much more predictable and technically easier with a much lower complication rate,” he said.

Kim chooses to perform DMEK in select scenarios, such as in emmetropic pseudophakic eyes to avoid the hyperopic shift commonly associated with DSEK.

“I have also seen several patients with endothelial or Fuchs’ dystrophy who have undergone cataract surgery as a primary procedure with a multifocal IOL. In this scenario, I would also prefer to perform a DMEK procedure, particularly to avoid the refractive hyperopic shift and potential contrast sensitivity effects of a DSEK graft,” Kim said.

Donor preparation

Terry said that an increasing number of eye banks prepare donor tissue for DMEK.

“One of the challenges was preparing your own donor tissue at the time of surgery, stripping your own donor tissue and risking having the tissue destroyed at the time of surgery,” he said. “But now we have the eye bank pre-stripping the tissue. We no longer have to ever worry about canceling a case, not being able to complete a case or not being able to do the DMEK because we destroyed the tissue if the tissue is already pre-stripped.”

John said that while he does not use prepared donor tissue for DMEK, prepared tissue makes it easier for many surgeons to adopt DMEK.

“Myself, I don’t use pre-prepared tissue. I prepare my own tissue,” John said. “But since eye banks made these tissues available, that takes off 50% of the challenge of DMEK for all the corneal surgeons who want to do DMEK.”

The greatest challenge in graft preparation is configuring the donor button as a single disc without tearing it. A torn graft can have a negative financial impact on that surgical case, John said.

“You should also have backup tissue whenever possible when you do DMEK, especially if you want to prepare the donor tissue yourself,” he said. “This provides the added insurance of being able to use the standby tissue to complete the case in the event of a torn donor DMEK graft.”

Preparing DMEK grafts does not require specialized or expensive equipment, Price said.

“You just need some basic tying forceps that don’t have any sharp edges and some type of blunt or semi-sharp instrument to score the peripheral Descemet’s. Then you can do your donor preps. But a lot of people still want the eye banks to do it for them,” Price said.

Surgeons who prepare their own donor tissue can reduce costs, Price said.

“From a cost-benefit ratio for the health care system, it’s much better to have the surgeons do it than the eye banks because of the difference in reimbursement if surgeons do the donor prep compared to the eye banks,” he said.

Price noted that surgeons typically prefer corneal tissue from donors in their 40s and older.

“We certainly take donors in the 40s and on occasion take a few in the 30s. But the ones in the 30s can be difficult sometimes to get them to work. A lot of surgeons won’t go below 50. As a person ages, their Descemet’s gets thicker. The thicker Descemet’s is, the easier it unfolds. That’s why we want older donors,” Price said.

Some surgeons perform pre-Descemet’s endothelial keratoplasty to prepare grafts from younger donors. In PDEK, an air bubble is used to remove the Descemet’s membrane, as in deep anterior lamellar keratoplasty, Price said. The thicker graft is easier to unroll and insert than a thinner graft.

“The reason that has some appeal is that when you detach Descemet’s by putting a bubble in the center of the cornea, it comes off with some stromal fibers,” he said.

Dua’s layer is also removed in the PDEK procedure, Price said.

“There’s some debate over whether Dua’s layer is real or not, but what is real is that when you detach it in the center instead of the periphery, you get a thicker piece of tissue,” he said. “[It] does give you the ability to use donors from young patients. [Recently] I did a donor that was 7 years old and used that for a PDEK case.”

Hardten said it is important to not disrupt the posterior stroma when stripping the host tissue.

“Any little discontinuities or irregularities in the posterior stroma can lead to a little unevenness on the back surface that can lead to decreased adherence and also lead to interface haze or scarring,” he said.

Graft manipulation, injection

DMEK involves a no-touch technique with a somewhat steep initial learning curve, especially for a new DMEK surgeon, John said.

“A surgeon may do pretty well with DSEK, but with DMEK you’re going back to the drawing board, so to speak, the main difference being one is a touch technique and the other is a no-touch technique, for the most part,” he said. “With DSEK, you have stroma, and it’s like conventional surgery where you can actually handle the tissue. But with the DMEK, for the first time you’re dealing with tissue which you really cannot physically handle if you can avoid it.”

Melles and colleagues recently published a study in JAMA Ophthalmology showing that a group of surgeons successfully learned a standardized no-touch DMEK technique in a relatively short time. The no-touch technique is safe, easy to learn and reproducible, they reported.

Price noted that methods of injecting donor tissue into recipient eyes have improved.

“I think that the way that we inject [tissue] into the eye has gotten better,” Price said.

New types of injectors make the insertion process easier and reduce the risk of damaging the donor tissue, Terry said.

“The best injectors are made of glass and they’re attached to a syringe,” Terry said. “You can aspirate the tissue into the injector and irrigate it out. So, you’re never touching the tissue itself. You’re not grabbing the tissue and putting it into an injector. You’re simply aspirating it in and irrigating it out.”

According to Terry, the main cause of primary graft failure is tissue being unrolled and injected upside-down.

“Because the tissue is so thin and so fragile, it’s often difficult to tell whether the tissue is right-side-up or upside-down. So, what we’ve also done is we’ve had the eye banks pre-mark the tissue with an ‘S’ stamp. So, now it’s virtually impossible to leave the tissue upside-down,” Terry said. “We haven’t had any graft failures since we’ve been using the ‘S’ stamp in the last 2 years.”

Price said that pipettes, the STAAR ICL inserter, IOL cartridges, syringes and angiocatheters can be used to inject donor tissue into the eye.

A more recent advance, the Muraine technique, involves the use of a Busin glide (Moria) to insert donor tissue, Price said.

“The advantage of that particular technique is that you actually pull it in like a DSEK graft. You pull it in, and you know how it’s oriented,” he said. “If you get an air bubble under it quickly enough, you don’t have any manipulation of the tissue. That’s a nice way to do it, but I’m not sure that way has really taken off yet. We’re waiting for data from Dr. Marc Muraine to see how his cell counts do.”

Using an injector to insert an ultrathin DSEK graft provides more reproducible results, especially in complex eyes, Kim said.

“We recently published a retrospective case series of 71 eyes using an injector system to implant thin DSEK grafts. We found that DSEK surgery with this injector was effective and compatible with a very low complication rate and a good graft survival rate. We found it especially beneficial in eyes with complex anterior chamber characteristics or pathology (ie, glaucoma drainage device, trabeculectomy, previous retinal surgery, penetrating keratoplasty, anterior chamber IOL, etc.),” he said.

Hardten said he prefers using a SofPort injector (Bausch + Lomb) to inject the donor tissue.

“I think it’s important to be careful not to rub the endothelium on the edges of the injector. I prefer to put it together first and then load the tissue into the tip as opposed to trying to put it in and then fold it closed. I tend to use a technique … where the tissue is opened up and I put a bubble underneath it,” Hardten said.

Re-bubbling

DSEK and DMEK involve the injection of an air bubble to force the graft tissue to adhere to the recipient tissue. DSEK requires only a small bubble that remains at the conclusion of the surgery to support the graft, while in DMEK, the bubble is markedly larger.

John said there is an increased rate of re-bubbling encountered with DMEK. Studies have shown re-bubbling rates in DMEK of 50% to 80%.

“The downside is that patients may have to go back to the operating room or minor sterile surgical procedure area, depending on the situation, for re-bubbling those cases that need re-bubbling,” he said.

The re-bubbling rate depends on surgeon comfort level, experience and the re-bubbling threshold, John said. For example, some surgeons will opt to re-bubble if 5% or less of the graft is detached at the margins, while others will wait for 10% or more detachment.

“In my opinion, being aggressive to re-bubble may be indicated in some of the cases, especially if you document it at two different time spots where there is progression of Descemet’s membrane detachment. So, then there’s a higher risk of an overall Descemet’s membrane detachment. On the other hand, if you have partial non-attachment and absolutely no progression, you might elect to observe,” John said.

John said that the John DMEK Smoother (Bausch + Lomb/Storz) can be used to unfold and center the DMEK graft onto the recipient cornea without using air.

“It’s an instrument with a spherical ball at the tip,” John said. “It’s used on the corneal surface to both unfold the Descemet’s membrane and keep it in the center, doing it every single time without the use of any air. That’s a huge advantage, especially because it is one of the more difficult steps of the procedure and helps simplify the DMEK surgery.”

Anatomic and visual outcomes

Kim cited a contralateral eye study by Goldich and colleagues, published in the American Journal of Ophthalmology, that included 17 patients who underwent DMEK in one eye and DSAEK in the other eye. Preoperative best corrected visual acuity was similar in both groups, but the DMEK group had slightly better postoperative BCVA at 6 months. Endothelial cell counts were similar in both groups at 6 months, and patient satisfaction ratings were slightly higher in the DMEK group.

“I’m curious to know what the longer-term endothelial cell counts are going to show,” Kim said. “Personally, in doing the procedure, there’s more stress and trauma on the corneal endothelium with the DMEK procedure, both during the harvesting of the graft as well as during the insertion and manipulation of that graft to have it unfold and center prior to the air injection. The traumatic tapping on the recipient cornea to help unfold, manipulate and center the DMEK graft is typically done in the setting of a very shallow anterior chamber, and so we’ll have to wait and see the impact on the corneal endothelium.”

Kim cited a study published in Eye by Maier and colleagues that showed while DMEK and DSAEK had similar patient satisfaction ratings, patients indicated that they preferred DMEK if given a choice “because it probably provided a little better best corrected visual acuity and maybe a little better contrast sensitivity.”

Another study by Goldich and colleagues showed that eyes that underwent DSAEK had slightly thicker corneas and slightly steeper posterior corneal curvature than eyes that underwent DMEK, which may explain the hyperopic shift seen after DSAEK.

In a study published by Guerra and colleagues in Cornea, postoperative BCVA was 20/24 in DMEK eyes and 20/32 in DSAEK eyes.

“A majority of patients in this study, 85%, perceived better visual quality in the DMEK eye vs. the DSEK eye, and they would prefer DMEK over DSEK. And the 1-year endothelial cell loss was comparable between the two. But I’m still waiting to see what the longer-term results are going to show,” Kim said.

Descemet’s membrane endothelial transfer

One relatively new addition to the EK subcategory is Descemet’s membrane endothelial transfer (DMET), a no-touch, no-graft technique that involves the implantation of a free-floating graft into the anterior chamber without attaching it to the host posterior stroma. DMET allows endothelial cells to migrate, proliferate and re-endothelialize the posterior stroma. DMET is indicated for Fuchs’ endothelial dystrophy.

“The concept behind that is that you don’t need to have the Descemet’s membrane of the donor. You simply have to inject the tissue and then that stimulates the healing process of endothelial migration and allows the endothelium to crawl across. That’s an interesting concept, but I don’t think that it’s true,” Terry said.

Terry attributed the cell repopulation to the natural healing process.

“That’s simply a wound-healing response. It’s similar to when you have a hydrops in keratoconus and the Descemet’s membrane splits open,” Terry said. “It just takes a very long time, and it also gives you very low cell counts postop. So, I’m not a proponent of DMET. I don’t think it’s a concept that’s going to prove fruitful in the future.”

DMET has only been successful in Fuchs’ not pseudophakic bullous patients, indicating that the donor has no role in repopulating the endothelial cells on the patients’ eyes, Price said. He further said that it may take months for the corneal edema to clear after either DMET or Descemet’s stripping alone, and the prolonged periods of edema may cause long-term degenerative changes in the anterior cornea, as reported by both Arbelaez and Price. – by Matt Hasson

For more information:

David R. Hardten, MD, can be reached at Minnesota Eye Consultants, 710 E. 24th St., Suite 100, Minneapolis, MN 55404; email: drhardten@mneye.com.
Thomas John, MD, can be reached at tjcornea@gmail.com. Terry Kim, MD, can be reached at Duke University Eye Center, Erwin Road P.O. Box 3802, Durham, NC 27710-3802; email: terry.kim@duke.edu.
Francis W. Price Jr., MD, can be reached at Price Vision Group, 9002 N. Meridian Street, Suite 100, Indianapolis, IN 46260; email: francisprice@pricevisiongroup.net.
Mark A. Terry, MD, can be reached at Devers Eye Institute, 1040 NW 22nd Ave., Suite 200, Portland, OR 97210; email: mterry@deverseye.org.
Disclosures: Hardten, Price and Terry report no relevant financial disclosures. John reports he is a consultant for Bausch + Lomb. Kim reports he is a consultant for Ocular Systems.

Do you prepare your own donor tissue for DMEK?

Eye banks offer better quality

The preparation of DMEK grafts can be performed by the surgeons themselves, either in the operating room immediately before surgery or 1 day in advance. Alternatively, the stripping of Descemet’s membrane may be performed in the eye bank up to 1 week before surgery. Grafts prepared in an eye bank may have a higher quality because the tissue may then have extra rounds of evaluation to ensure its suitability for transplantation, including a thorough assessment of endothelial cell viability, as well as microbiological testing to exclude possible contamination during tissue handling.

In addition, eye bank-prepared grafts may allow better planning of surgical procedures because in the event of preparation failure, if necessary, the surgery could be rescheduled in advance. Grafts with a different diameter than anticipated or with small tears or that have the tendency to roll more tightly can be redirected to a selected host or surgeon, and late patient cancellations can be managed by redistribution of the tissue to a different center.

Given the above, surgeons may therefore be recommended to collaborate with an eye bank that provides pre-stripped DMEK grafts. Furthermore, from experience we learned that these preparations may best be performed by standardized techniques to obtain consistent tissue quality and to minimize tissue loss due to preparation failure.

Gerrit R.J. Melles, MD, PhD, is an OSN Europe Edition Board Member. Disclosure: Melles reports he is a consultant for DORC International and SurgiCube International.

Surgeons, banks have low tear rates

Regardless of whether it may be preferred, many surgeons do not have the option of utilizing eye bank-prepared DMEK tissue because a minority of eye banks are preparing the tissue. I was fortunate because the San Diego Eye Bank was one of the pioneers, and I never had to prepare my own tissue.

While studies emerge to guide decision making, and as eye banks start preparing DMEK tissue, early adopting surgeons are forced to prepare their own tissue.

With any new surgical technique, peer-reviewed, published, head-to-head studies are the only way to truly make recommendations. Having said this, to date, there are no published studies comparing DMEK grafts dissected by an eye bank only, partially by an eye bank and then completed by the surgeon, and tissue preparation by the surgeon only in terms of endothelial cell loss and defects in the graft. Further, not all eye banks are preparing DMEK tissue.

Deng and colleagues observed that in 43 DMEK-prepared eye bank grafts, there was a 0.2% rate of tears significant enough to render the graft unusable. Meanwhile, Schlötzer-Schrehardt and colleagues published results of surgeon separation of Descemet’s membrane in 350 corneas and also reported a 0.2% rate of tissue loss.

Deng’s study also evaluated the change in endothelial cell density resulting from the surgeon completion of DMEK graft separation from pre-dissected eye bank tissue and found a mean loss of 3.8% and up to 14.5%. Future studies are needed to compare if tissue manipulation once yields better endothelial cell density than two sequential dissection steps regardless of who is dissecting the tissue.

Further studies are needed to evaluate the success of graft processing and standardization. As more and more eye banks become facile at DMEK tissue preparation, more and more surgeons may allow their eye banks to supply the tissue due to the time saving and lack of tissue loss. There is also no doubt that many surgeons may be waiting for their eye banks to begin reliable tissue preparation before learning DMEK surgery.

Elizabeth Viriya, MD, is a cornea/external disease and refractive surgery fellow at Gordon-Weiss-Schanzlin Vision Institute and Scripps Clinic, La Jolla, Calif. Francis S. Mah, MD, is an OSN Cornea/External Disease Board Member. Disclosures: Viriya and Mah report no relevant financial disclosures.