Continuous evolution of endothelial keratoplasty changing field of corneal transplantation

Lamellar techniques have introduced a paradigm shift in corneal transplantation. Endothelial keratoplasty has undergone rapid advancements in the last decade, becoming the most commonly performed corneal transplantation procedure in the West and relegating penetrating keratoplasty to a minor role.

In 1997, Gerrit Melles, MD, started with deep lamellar endothelial keratoplasty as the first technique to replace the endothelium.

“This technique was relatively demanding, so we continued with Descemet’s stripping endothelial keratoplasty in 2001, in which only the host Descemet’s membrane had to be excised instead of a posterior corneal button. However, the corneal graft in both these techniques contains posterior stroma, which seems to somehow degrade the optical quality of a transplanted cornea,” Melles said. “To reach an anatomically ‘normal’ cornea, we worked towards Descemet’s membrane endothelial keratoplasty in 2005, which we thought should eliminate the image degradation caused by the donor stroma, since only Descemet’s membrane is transplanted.”

Today, DSEK, with the variation Descemet’s stripping automated endothelial keratoplasty, and DMEK represent one-third to two-thirds of the total number of keratoplasty procedures in Western countries, depending on the area. According to a statistical report from the Eye Bank Association of America (EBAA), in 2011 they were performed as often as penetrating procedures in the U.S. — about 21,000 each — but rapid growth has occurred over the last year.

“EBAA will probably release their numbers for 2012 in April, which will be quite different, since penetrating keratoplasty has now been relegated to indications that do not qualify for endothelial transplantation,” Marianne O. Price, PhD, executive director of the Cornea Research Foundation of America, said. “In our practice, 90% of procedures for endothelial dysfunction are DMEK and 10% are DSEK. This represents about 80% of the 500 transplants the practice performed last year, with anterior lamellars and penetrating grafts for non-endothelial dysfunction each making up 10%.”

According to Massimo Busin, MD, the relative percentage of endothelial procedures in Europe is currently around 30%, but it is much higher for some surgeons.

“Of the 398 transplantations I performed last year, 60% were endothelial, 35% were anterior lamellar and only 5% were penetrating keratoplasty,” he said.

In Asia, numbers are still low, due to the late introduction of the surgery and the high costs of cutting tissues. Donald T.H. Tan, MBBS, FRCSG, FRCSE, FRCOphth, an OSN APAO Edition Board Member, estimated that the relative percentage is around 5% to 10%. However, some centers of excellence, such as in Singapore, are aligning with Western standards.

“In our practice we do about 300 transplantation procedures a year, of which 50% are endothelial, 30% are anterior lamellar procedures and only about 20% are PK, reserved for end-stage stromal and endothelial disorders, which are still prevalent in Asia. My personal rates for EK are even higher, around 62%. This is also thanks to the availability of tissues from our local eye bank, which offers both precut EK and ALK grade tissue,” he said.

In the Middle East, like in Asia as a whole, endothelial dysfunctions are relatively less common than conditions involving the anterior cornea.

“In Iran, we have a rate of about 16% for endothelial keratoplasty. Opposed to the West, Fuchs’ dystrophy isn’t common. Most of the patients who are referred to us for endothelial keratoplasty have posterior bullous keratopathy, which can be attributed to the learning curve period in which surgeons have begun to use phacoemulsification to treat cataract in the last 2 decades,” Hassan Hashemi, MD, said.

DSEK and DMEK advocates

Both DSEK/DSAEK and DMEK have advocates with differing opinions. Some surgeons have made a definite transition to DMEK and use DSEK only in specific cases. The main arguments in favor of DMEK are the superior visual outcomes and lower rejection rate. Other surgeons believe that the latest developments of DSAEK with ultra-thin donor discs have results comparable to DMEK, without the additional surgical challenges posed by DMEK.

“In our clinic, DMEK is now our preferred procedure. It feels much better controlled than DSEK/DSAEK, and the visual outcomes surpass earlier techniques. The rule of thumb is that about 80% of eyes reach 20/25 or better at 6 months,” Melles said. “Once the technique is mastered, complications are fewer and there is no need for large financial investments. Standard donor corneal rims can be used for harvesting the tissue.”

The thin DMEK grafts, which do not have much impact on the original corneal thickness, induce a minimal amount of corneal aberration and lead to better visual acuity and faster healing.

“The stroma that’s left on the donor in DSAEK has to conform to the posterior surface of the recipient, causing wrinkles and folds that degrade the image. In addition, the microkeratome cut produces irregularities from the side to the center. With DMEK, we avoid all those distortions,” Francis W. Price Jr., MD, an OSN U.S. Edition Cornea/External Disease Board Member, said.

Also the smaller incision size, less than 3 mm, contributes to better vision.

 “Contrary to DSAEK, in DMEK we don’t have significant hyperopic shift or induced astigmatism. Thanks to the smaller incision, spherical equivalent changes are less than 0.5 D,” Hashemi said.

Francis Price said that DMEK allows operating on the two eyes just 1 week apart, similar to what is done with cataract surgery. Thanks to the smaller DMEK incision, healing and visual recovery are fast, which allows for the shorter interval between surgeries, he said.

According to some surgeons, ultra-thin DSAEK has comparable visual results to DMEK while also offering the ease of preparation and manipulation of DSAEK.

“We now do thin donors, below 120 µm thick. In a series of 250 cases, we have found that the percentage of patients achieving 20/20 is equal, if not superior, to DMEK,” Busin said.

DMEK still results in a higher endothelial cell loss rate, approximately 30% to 35% at 1 year, whereas cell loss rates with newer DSAEK techniques and donor inserts continue to improve, Tan said.

“In our first series of 100 eyes, ultra-thin DSAEK with the use of our ultra-thin DSAEK inserter had a cell loss of 13.5% at 6 months and 15% at 1 year, and DSAEK still has a lower re-bubbling rate. However, DSAEK techniques continue to improve, and we hope to see cell loss and re-bubbling rates reduce further,” he said.

Rejection rate

The significantly lower rejection rate is a key issue in favor of DMEK. In 2011, in his first series of 120 eyes of 105 Fuchs’ dystrophy patients, Melles and his group reported a rejection rate of less than 1% at 2 years. The same rate over the same period of time was reported in 141 eyes by Francis Price and colleagues in 2012.

“We compared retrospectively these patients with cohorts of DSEK and PK patients treated at the same center, with similar demographics, follow-up duration, postoperative regimen and indications for surgery. The relative risk ratio for immunological graft rejection at 2 years with DMEK turned out to be 15 times lower than DSEK and 20 times lower than PK,” Marianne Price said.

“I used to do more DSEK than DMEK, but once we found that rejection rate was so much lower, about 1.5 years ago, I definitely shifted to DMEK and encouraged others to do so,” Francis Price said.

Tan suggested that the 8% to 12% DSEK rejection rate reported by Price seemed relatively high, saying that the rate is steady at 2% to 2.5% in his own patients, with both standard and ultra-thin procedures.

“Rejection rates are likely to vary depending on the extent and duration of steroid use after DSAEK surgery, and it may be that the tendency to greatly reduce steroid use after DSAEK as compared to PK may affect rejection rates. In our initial DSAEK series, we didn’t really reduce steroids. We therefore see a low rejection rate, but conversely, our glaucoma rates remain similar to PK,” he said.

Francis Price noted that the low rejection rate of DMEK may allow a significant reduction of postoperative steroids and steroid-related complications such as IOP increase.

“It appears that either the corneal stroma or the total amount of tissue transplanted affects the immune response and propensity to stimulate immune-rejection episodes. While many doctors are attempting to make thin-cut DSEK an alternative to DMEK, DMEK is the thinnest possible graft at this point and therefore the most desirable for the best possible vision and least immunologic reaction. We are evaluating in a multicenter trial the safety and feasibility of reduced steroid administration after DMEK,” he said.

For those who prefer DMEK, there are still indications for DSEK/DSAEK in specific cases.

“In complicated eyes with aphakia, previous posterior segment surgery and glaucoma tubes, DSEK/DSAEK may still be a valuable fallback option. In such eyes, obtaining graft attachment is a challenge and may outweigh the final visual outcome, especially because the visual potential of these eyes most often is relatively low,” Melles said.

“DSEK is preferable in complicated eyes with a lot of synechiae requiring anterior chamber reconstruction. Likewise, in eyes that are aphakic, a DSEK graft can be pulled into the eye with forceps or a suture and held in place until air can be placed beneath it, whereas a DMEK graft could go into the posterior chamber more easily,” Francis Price said.

The challenges of DMEK

DMEK is known to be a technically challenging surgery with a long learning curve. This is the main downside of the procedure and the main limitation to gaining fast and widespread popularity.

The transition from DSEK to DMEK requires “a shift of mind,” according to Hashemi.

“We must believe that we can perform DMEK for appropriately selected patients and accept going through a learning curve period as we did from intracapsular to extracapsular and then to phaco in cataract surgery or from PK to DSAEK. The transition is not that long, and we’ll finally be able to perform this surgery with confidence and good results,” he said.

Patient selection is also crucial, Hashemi said. The best patients to start with are those with prominent eyes, low edema, a deep anterior chamber, a large diameter cornea, pseudophakia and no scars in the corneal stroma.

“Typically, myopic patients with Fuchs’ who have cataract or have already been operated for cataract,” he said.

Patience, practice and donor corneas older than 40 years of age are the keys to successful surgery, Francis Price said.

“Some DMEK surgeons prefer donors over 50 years. Older donor tissue is easier to prepare and to uncurl inside the recipient eye, probably because the Descemet’s membrane continues to thicken with age,” he said.

Concerns about damaging donor grafts in the harvesting and implantation process can be overcome by using the “no-touch” technique suggested by Melles and colleagues.

“With our technique, there is no direct handling of the tissue. For harvesting, we leave a peripheral ring of trabecular meshwork tissue in situ, and the graft is trephined on an underlying soft contact lens. For implantation, we use a smooth glass injector and then manipulate the tissue by air or [balanced salt solution]. Endothelial cell loss is minimized,” Melles said.

The challenge of unfolding and positioning the graft has been evaluated. The classic “double roll” is the easiest approach, but if this cannot be done, several other unfolding techniques have been developed by Melles and colleagues.

Tan recently presented a new device for graft insertion, the disposable D-Mat carrier. Acting as a stromal carrier replacement, it supports the endothelial graft for enhanced ease of manipulation and prevents scrolling, allowing for surgical techniques akin to ultra-thin DSAEK to be adopted.

“The donor Descemet’s membrane adheres gently to the D-Mat surface and therefore does not wrinkle up, and a no-touch technique is employed as the D-Mat is held rather than the donor tissue, which can also be coiled into a DSAEK inserter and delivered into the eye. However, further work needs to be done before this becomes a viable technique,” Tan said.

Inserters

Inserters have been a key feature in DSEK. The Busin glide (Moria) is a funnel-shaped device, designed to fit into a 3-mm incision.

“The graft is pulled rather than pushed into place, grasping a preplaced nylon suture from the opposite side. Under continuous irrigation, the graft unfolds easily, with minimal manipulation and cell loss,” Busin said.

The Tan EndoGlide (Network Medical Products) was inspired by the difficulty of inserting the graft in small Asian eyes with high vitreous pressure and a tendency for anterior chamber collapse and iris prolapse, which induces a cell loss of about 60% with traditional taco-folding techniques.

“The EndoGlide provided essentially a much higher element of control. We had a much lower cell loss, between 13.5% and 15% at 6 months and 1 year, respectively, the lowest to date. An advantage is that this inserter is extremely useful in more challenging cases where other insertion techniques may be difficult, simply because of better anterior chamber and donor control,” Tan said.

Results in high-risk eyes and those with failed grafts, previous vitrectomy or implanted glaucoma drainage tubes showed an almost comparable rate of cell loss, around 18%.

“We now have produced a second version of the Tan EndoGlide for ultra-thin DSAEK, which makes coiling of very thin donor tissue easier and more reproducible,” he said.

A further advance has been made by preloading the donor endothelial tissue at the eye bank. Preloaded donors at several eye banks may soon be readily available with the Tan EndoGlide.

In addition, Busin is working on the development of preloaded grafts with the Veneto Eye Bank in Italy.

The future

Francis Price said that it is a great privilege for cornea specialists to be working at a time when surgical procedures are undergoing such dramatic changes and to be involved in this progress.

“The benefit to our patients is huge. With cataract surgery, we would not perform an intracapsular cataract unless people were 20/200 or worse. Now we do phaco at 20/25 or 20/30. The same is happening with transplant surgery. Not very long ago, patients had to be non-functional before we performed keratoplasty, and now we do it if people have problems driving or working,” he said.

Further improvements are expected in coming years. According to Tan, there is still a wide variation in techniques, results, dislocation rates and primary graft failures.

“We need to develop consistency in the procedure, and we will gradually achieve uniformly better results. But we need to routinely measure endothelial cell loss. Strangely enough, many surgeons today ask for donor tissue with a cell count but don’t do the cell count after surgery,” he said.

There is also the question of whether to use ultra-thin DSAEK or DMEK.

“We are trying to make DMEK an easier, more reproducible procedure, but at the same time, ultra-thin DSAEK is giving us results that appear to be nearly as good. What is certain is that the future is thinner with either technique, and less is the new more,” Tan said.

The role of eye banks will also become prominent, because the demand for precut and possibly preloaded tissue is going to increase.

“This will create new medical legal scenarios, as it may be difficult to establish who has the responsibility in case of failure,” Busin said.

Melles has recently been investigating a new Descemet’s membrane endothelial transfer (DMET) technique that might completely change the approach to corneal transplantation. The idea came from the observation that in some Fuchs’ dystrophy eyes, corneal clearance occurred despite endothelial graft detachment or subtotal attachment after DMEK.

“This may prove that host cells can start migrating again under the regenerating stimulus of the donor and recreate a healthy endothelium,” he said.

The possibility of performing DMET, ie, the implantation of a graft into the anterior chamber without positioning it to the host posterior stroma, may be useful in eyes that are technically challenging. Also, it may be of interest from a scientific point of view.

“If host cells play a main role in corneal clearance in eyes operated on for Fuchs’ endothelial dystrophy, graft surgery might be ancient history in a not too distant future, and stimulating agents may be used instead to reconstruct the diseased endothelium,” Melles said.

The main drawback of DMET at present is that corneal clearance is delayed to 2 to 3 months, and endothelial cell density is lower than after DMEK.

“Nevertheless, anatomical restoration is near perfect, and the risk of intra- and postoperative complications may be lower,” Melles said. – by Michela Cimberle

• References:
• Ang M, et al. Ophthalmology. 2012;doi:10.1016/j.ophtha.2012.06.012.
• Anshu A, et al. Ophthalmology. 2012;doi:10.1016/j.ophtha.2011.09.019.
• Anshu A, et al. Surv Ophthalmol. 2012;doi:10.1016/j.survophthal.2011.10.005.
• Baydoun L, et al. Am J Ophthalmol. 2012;doi:10.1016/j.ajo.2012.06.025.
• Busin M, et al. Arch Ophthalmol. 2008;doi:10.1001/archopht.126.8.1133.
• Busin M, et al. Invest Ophthalmol Vis Sci. 2012;doi:10.1167/iovs.11-7753.
• Dapena I, et al. Arch Ophthalmol. 2011;doi:10.1001/archophthalmol.2010.334.
• Dapena I, et al. Cornea. 2011;doi:10.1097/ICO.0b013e31820d8540.
• Dirisamer M, et al. Am J Ophthalmol. 2012;doi:10.1016/j.ajo.2012.02.032.
• Khor WB, et al. Am J Ophthalmol. 2011;doi:10.1016/j.ajo.2010.08.027.
• Khor WB, et al. Cornea. 2013;doi:10.1097/ICO.0b013e31828321f8.
• Liarakos VS, et al. JAMA Ophthalmol. 2013;doi:10.1001/2013.jamaophthalmol.4.
• Tan DT, et al. Lancet. 2012;doi:10.1016/S0140-6736(12)60437-1.
• Tong CM, et al. Can J Ophthalmol. 2012;doi:10.1016/j.jcjo.2012.04.009.
• van Dijk K, et al. Cont Lens Anterior Eye. 2013;doi:10.1016/j.clae.2012.10.066.
• For more information:
• Massimo Busin, MD, can be reached at Villa Serena Hospital, Via Del Camaldolino 8, 47100 Forli, Italy; email: mbusin@yahoo.com.
• Hassan Hashemi, MD , can be reached at Noor Ophthalmology Research Center, Noor Eye Hospital, 96 Esfandiar Blvd., Valiasr Ave., Tehran, Iran, 1968653111; +98-21-88-65-15-15; fax: +98-21-88-65-15-14; email: hhashemi@norc.ac.ir.
• Gerrit Melles, MD, can be reached at the Netherlands Institute for Innovative Ocular Surgery, Laan Op Zuid 88, 3071 AA Rotterdam, Netherlands; +31-10-297-4444; fax: +31-10-297-4440; email: melles@niios.nl.
• Francis W. Price Jr., MD, and Marianne O. Price, PhD, can be reached at Price Vision Group, 9002 N. Meridian St., Suite 100, Indianapolis, IN 46260, U.S.A.; +1-317-844-5530; fax: +1-317-844-5590; email: francisprice@pricevisiongroup.net; marianneprice@cornea.org.
• Donald T.H. Tan, MBBS, FRCSG, FRCSE, FRCOphth, can be reached at Singapore National Eye Centre, 11 Third Hospital Ave., Singapore 168751; +65-6227-7255; fax +65-6222-9393; email: donald.tan.t.h@snec.com.sg.

Disclosures: Busin receives royalties from Moria for the glide and forceps carrying his name developed to deliver the DSAEK graft. Hashemi, Melles, and Francis and Marianne Price have no relevant financial disclosures. Tan receives royalties from Network Medical Products for the glide developed to deliver the DSAEK graft.

Is precut tissue preferable to tissue dissected at the time of surgery for endothelial keratoplasty?

Precut tissue increases efficiency, reliability, quality

The number of endothelial keratoplasty procedures increases continuously every year. There are several techniques for surgery and preparation of the donor tissue, from manually dissected DSEK grafts to microkeratome-assisted DSAEK grafts, from femtosecond laser-prepared DSEK grafts to manually prepared DMEK grafts. Precut tissue has many advantages in endothelial keratoplasty because, especially for less experienced surgeons, the surgery itself might be quite a challenge. By using tissue that has been pre-dissected in an eye bank, the surgeon can focus solely on other aspects of the surgery.

Precut tissue reduces stress on the day of surgery and saves time in the operating room. The risk of losing tissue due to preparation, especially in DMEK, is more or less eliminated. Precut tissue also helps to work out a proper OR schedule, without having to cancel cases in the OR due to failed preparation. Precut tissue is prepared with minimal trauma by experienced eye bank technicians and every cut (DSAEK, DSEK) or roll (DMEK) is re-evaluated before shipping, providing a high level of quality. Tissue preparation in eye banks typically takes place in clean room conditions. The number of air changes per hour is up to 10 times higher than under operating room conditions, which may reduce the risk of contamination.

Precut tissue may increase the efficiency, reliability and quality of transplanted donor grafts.

Martin Dirisamer, MD, is from the Department of Ophthalmology, AKH Linz, Austria. Disclosure: Dirisamer has no relevant financial disclosures.

Prefer to cut own DMEK grafts

When I started doing DSEK 6 or 7 years ago, I definitely preferred to prepare my own lamellae because eye banks at the time were not equipped to provide precut tissue for this surgery. I was already exploring a double cut to do thinner lamellae, and I had no doubt that my experience with the microkeratome was better than what I could expect from any beginner technician in an eye bank.

With time, as expected, eye banks improved their skills, and today if I do DSEK, I use eye bank precut lamellae. However, DSEK is today only 5% of my endothelial procedures, and I have shifted to DMEK in 95% of my cases. Again, I do not trust any of the eye banks that provide tissue to me on the ability to prepare DMEK rolls, and most of them do not provide this service anyway. I expect they might be ready in 1 or 2 years, but this will also depend on eye bank policies and their willingness to invest in this procedure.

We can reasonably predict that DMEK will become progressively widespread, and if the demand grows, eye banks are likely to respond. However, these are issues I cannot control. At present, I prepare my own DMEK rolls, which gives me the advantage of using the rest of the cornea for anterior lamellar procedures in some cases. My main concern is to have the best possible tissue in my hands. If eye banks will be able to provide good endothelial rolls at the same price as the whole cornea, I will not hesitate to switch to precut tissue. Preparing a DMEK roll is a demanding procedure. I like using younger donors. Although older donors make the procedure easier, my technique allows for relatively simple and safe harvesting. I expect eye banks to give me the same quality, but whether they will take this risk or not, only time can tell.

José L. Güell, MD, PhD, is an OSN Europe Edition Board Member and director of the Cornea and Refractive Surgery Unit, Instituto de Microcirugia Ocular, Barcelona. Disclosure: Güell has no relevant financial disclosures.