April 13, 2015
4 min read

DMEK will eventually show superiority in hands of expert surgeons

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

The Eye Bank Association of America was founded in 1961 with 25 member eye banks. Today, 78 member eye banks harvest approximately 120,000 corneas per year in a high-quality, highly regulated and expensive undertaking. With this network of eye banks and modern transportation capability, along with advanced corneal preservation media that sustain endothelial viability for up to 14 days, corneal transplantation, at least in the U.S., is now an elective procedure. The days of waiting lists and weekend emergency transplants are long gone.

Rounding off, approximately 50,000 transplants are performed in the U.S. each year. Eye Bank Association of America member eye banks also export approximately 20,000 corneas for use outside the U.S. every year. While corneal blindness in the U.S. amenable to keratoplasty essentially no longer exists, the story is extremely different in areas of the world where an inadequate eye banking network, unavailability of trained corneal surgeons and economic barriers leave several million people who might respond to a corneal transplant blind and disabled.

In the U.S., a study by the Lewin Group found the cost of a corneal transplant, including tissue, facility, anesthesia, surgeon and postoperative care, is approximately $15,000. We corneal surgeons know that we are only a very small part of this cost, in most cases less than 10%. In the same study, the calculated “value” to society of treating a patient disabled by corneal disease with a keratoplasty was $250,000 for those younger than 65 years and somewhat less as patients age and life expectancy declines. Corneal transplantation is therefore not only life-changing for patients, but also a great value for society.

The biggest change in the field of keratoplasty over the last decade has been the transition from penetrating keratoplasty to lamellar keratoplasty, especially for endothelial dysfunction, including aphakic and pseudophakic bullous keratopathy and Fuchs’ dystrophy. Approximately 20,000 of the 50,000 corneal transplants performed in the U.S. each year are now some form of lamellar endothelial keratoplasty.

The first meaningful transition, now pretty much complete in the U.S., was from PK to what most today call Descemet’s stripping endothelial keratoplasty. Our group of four, soon to be five, corneal surgeons made this transition nearly 5 years ago, and overall today we find DSEK preferable for most patients with endothelial dysfunction. Still, my personal impression affirmed by a substantial set of data collected and published by Doug Coster, MD, from a well-managed Australian Registry is not clearly in favor of DSEK over PK. While DSEK is much easier on the surgeon and the patient than PK, there is a significant learning curve, and postoperative interventions such as re-bubbling are a hassle for all. In addition, at 12 months and later postoperative, visual acuities achieved, especially if the PK patient is allowed to wear a gas permeable hard contact lens, are often superior with PK. In favor of DSEK, there is much more rapid visual recovery and less astigmatism, the hassle of suture removal and suture-related complications is reduced or eliminated, and there may be a slightly lower rejection rate. However, many patients with long-standing corneal edema also develop stromal or anterior corneal surface haze and/or significant epithelial basement membrane dystrophy. Depending on the corneal surgeon, some choose to treat these patients with DSEK combined with superficial keratectomy or phototherapeutic keratectomy when the surface disease is significant, while others choose PK. Both, in my opinion, are reasonable approaches.

Now comes Descemet’s membrane endothelial keratoplasty. This procedure is logical from an anatomical perspective and appears in early series by pioneering experts to generate better visual acuities and a lower incidence of late immune graft rejection. On the negative side, the technical challenges are higher than with PK or DSEK, with an even higher incidence of postoperative re-bubbling procedures and primary donor failures requiring a replacement graft. The DMEK procedure has not as yet been adopted by the majority of corneal surgeons in the U.S. An alternative that is also gaining in popularity is so-called “thin donor” DSEK, with donor buttons of less than 100 µm rather than more than 100 µm.


In our practice, the Minnesota Lions Eye Bank will provide on request a high-quality full-thickness donor, a normal-thickness DSEK donor, a thin DSEK donor or even a technician-prepared DMEK donor. Having these options available on order from a quality eye bank significantly simplifies the donor preparation for the surgeon, and the eye bank technicians who prepare these donors on a daily basis become technically adept. In our practice, we have all transitioned to thinner DSEK donors, but only one of us has transitioned to DMEK.

A question for every surgeon to consider is the surgical volume required to become truly expert at a technically demanding new procedure. As I have mentioned in prior commentaries, the annual number of procedures for a surgeon to be truly “expert,” based on studies in the orthopedic and cardiovascular literature, is 50 cases per year for the surgeon and 200 for the facility. The Minnesota Lions Eye Bank has an adequate volume to be expert in preparing full-thickness, anterior lamellar, standard DSEK, thin DSEK and DMEK tissue in a quality fashion. Our group performs approximately 400 keratoplasties a year, so our ASCs and support staff are excellent and meet the standard for safe and effective surgery. However, we now have four corneal surgeons and in July will add a fifth. My personal keratoplasty volume has declined from four per week when I was between the ages of 30 and 55 years old to less than one per week at age 67 because I have reduced my days in clinical practice. I feel very comfortable with PK, having done more than 2,000 in my career, and am comfortable enough with DSEK after 5 years to continue offering it to my patients. However, at a probable volume of 10 to 20 per year, I do not currently plan to incorporate DMEK into my surgical armamentarium. If and when my much younger four corneal fellowship-trained surgeon associates all adopt DMEK as their preferred procedure for endothelial keratoplasty, I will stop performing DSEK and refer all my patients who require endothelial keratoplasty to one of them. In my opinion, DMEK will not be a procedure for the occasional keratoplasty surgeon. I expect DMEK to prove itself superior to DSEK, and once it does, I will have performed the last endothelial keratoplasty of my career.

Disclosure: No products or companies that would require financial disclosure are mentioned in this article.