This article is more than 5 years old. Information may no longer be current.
Recent discovery leads to PDEK as another cornea transplant option
Click Here to Manage Email Alerts
According to an Eye Bank Association of America statistical report, which includes data from 76 U.S. member eye banks, just under 48,000 keratoplasty procedures were performed domestically in 2014. This number has been relatively flat for several years, but the type of transplant continues to evolve significantly.
The top four indications for keratoplasty in the U.S. are Fuchs’ dystrophy (21.5%), post-cataract surgery corneal edema (12.2%), keratoconus (10.1%) and graft failure (9.9%).
In 2014, one or another form of endothelial keratoplasty accounted for 54.6% of transplants in America. The U.S.-based corneal surgeon selected EK for more than 90% of his patients with endothelial dysfunction. In keratoconus, penetrating keratoplasty was still preferred in 89% of patients and anterior lamellar keratoplasty in only 11%. Many of the failed grafts were also treated with some form of EK. With the increasing utilization of corneal collagen cross-linking, I anticipate that the need for keratoplasty to treat keratoconus will progressively decline and perhaps even disappear over the younger corneal surgeon’s career. Thus, the primary keratoplasty of the future is one or another form of EK.
In 2014, Descemet’s stripping endothelial keratoplasty was the preferred method of EK in 89% of patients, with Descemet’s membrane endothelial keratoplasty being utilized in the other 11%. Of note, DMEK is the fastest growing method of EK, nearly doubling in each of the last 2 years, but it still represented only 2,865 procedures vs. 23,100 for DSEK in 2014. It is apparent to me that EK will continue to grow in popularity because it provides better visual acuity, more rapid visual recovery and a lower incidence of graft rejection. That it requires greater skill will not slow its growth. Pre-Descemet’s endothelial keratoplasty was not even mentioned in the 2014 report, so it is definitely very new and somewhat controversial.
We are all familiar with the classic description of the five layers of the cornea: epithelium, Bowman’s layer, stroma, Descemet’s membrane and endothelium. The French ophthalmologist Jean Descemet is credited with Descemet’s membrane, which was described in the 1700s, and Sir William Bowman, the English ophthalmologist, described Bowman’s layer in the 1800s. For more than 200 years, no new layers of the cornea were proposed. In 2013, Harminder Singh Dua of the University of Nottingham in England proposed that a sixth layer of the cornea just anterior to Descemet’s membrane exists and named it Dua’s layer. This layer was discovered using deep lamellar air injection in human cadaver corneas and is described as being approximately 15 µm thick, denser than the overlying stroma and adherent to Descemet’s membrane.
While controversy over Dua’s discovery remains, it is now possible for a sophisticated eye bank to provide EK tissue in four different configurations. For DSEK, one can request a normal thickness donor, which will generally be between 100 µm and 150 µm in thickness, or a thin DSEK donor, which will generally be between 50 µm and 100 µm in thickness. The trend is toward thinner DSEK donors, which seem to generate better visual acuity outcomes. For DMEK, the eye bank separates the stroma from Descemet’s membrane, generating an EK that is only about 20 µm thick. This requires an adult donor, usually older than 50 years, because Descemet’s membrane is only about 3 µm thick at birth and very elastic, making it unmanageable for the surgeon. As one ages, the endothelial cells lay down additional basement membrane, and Descemet’s membrane reaches 8 µm to 10 µm in thickness by age 50 years with reduced elasticity. This generates a transplant that, with skill and experience, can be harvested, injected into the anterior chamber, unscrolled and attached to the stroma with a gas bubble. Outcomes in skilled hands with DMEK seem superior to DSEK, and I have suggested in a previous commentary that I expect DMEK to eventually dominate EK.
Now comes another challenger, PDEK, which includes Dua’s layer in the donor, adding about 15 µm more to the donor thickness. Advocates suggest that PDEK is easier to handle for the surgeon than DMEK and provides similar advantages over DSEK while allowing younger donors to be utilized, expanding the available donor tissue pool. Others find the often somewhat smaller donor diameter, the requirement of scissors rather than a trephine for harvesting in the OR, and the reduced visual acuity associated with a stroma-to-stroma interface to be meaningful negatives vs. DMEK. The fact that PDEK grafts can reproducibly be generated in many eye banks and operating rooms around the world is in my opinion supportive of Dua’s report and suggests that at a minimum there is a denser layer of stroma adjacent to Descemet’s membrane that can be dissected from the less dense posterior stroma with a big bubble technique.
It will be interesting to see if PDEK generates a following among U.S. corneal surgeons and starts to show up in the 2015 or 2016 EBAA statistical report. I project continuing strong growth in DMEK and believe it will be preferred over PDEK and DSEK for most patients requiring EK by 2020.
That a clinician/scientist can make a meaningful and potentially clinically useful observation about corneal anatomy 200 to 300 years after the description of Descemet’s membrane and Bowman’s layer reminds me that there remains much to be discovered, and many of these discoveries will be the result of careful observation by the actively practicing clinician/scientist.