During my residency and cornea fellowship in the late 1990s, penetrating keratoplasty (PKP) was the only surgical remedy for endothelial disease (mostly for pseudophakic bullous keratopathy and advanced Fuchs corneal dystrophy). At that time, advances in lamellar surgery were occurring in the field of refractive surgery with the advent of handheld microkeratomes for LASIK.
I remember transitioning to endothelial keratoplasty (EK) as the technique emerged. In the OR, my colleagues and I mounted eye-banked corneas on artificial anterior chambers and cut them manually with a microkeratome to prepare the graft. If the cut was imperfect, either a second donor would be used, or the case would be rescheduled for another day. Those early experiences underscore just how radical—and promising—the shift to EK was.
Since then, EK has transformed the treatment of patients with endothelial disease to allow earlier intervention and faster, safer patient recovery. In this installment of the “Origins” column, Mark A. Terry, MD, shares the origin story of EK and describes his important role in advancing early surgical concepts to the current global standard of care.
Tal Raviv, MD
Over the past 25 years, a revolution has occurred in the treatment of vision loss due to corneal edema. The selective replacement of the diseased endothelial layer has supplanted PKP as the standard of care. The development of EK began with a unique idea that was cultivated by a few passionate, persistent individuals. Many people, however, contributed to the incremental refinement of EK. Each has their own story. This is mine.
THE BEGINNING
In 1998, a Dutch corneal specialist, Gerrit Melles, MD, PhD, published a laboratory study using cadaver eyes that evaluated an experimental surgical procedure he called posterior lamellar keratoplasty (PLK). During PLK, a scleral-corneal tunnel incision was created, and then the posterior stroma and endothelium were excised and replaced with donor tissue cut in a similar fashion.1
My twin sons were born that same year. While catching up on published research during a 2 am feeding, I read Dr. Melles’ publication. I immediately recognized the fundamental importance of a transplantation procedure that moved the incision and sutures to the sclera, leaving the anterior stroma and corneal surface pristine.
I could not sleep that night. The next morning, I sketched out new instruments for this exciting surgery and faxed the diagrams to Bausch + Lomb. Within 6 weeks, I began conducting laboratory experiments on weekends using prototype instruments at the Lions Eye Bank of Oregon (an entity currently known as VisionGift). After operating on hundreds of eye bank eyes and revising the design of the instruments multiple times, I felt that the surgery could be safely performed in humans. First, however, I shared videotapes of my experiments with prominent transplant surgeons and requested their opinions. Peter Laibson, MD, of the Wills Eye Institute was the most encouraging.
I wrote a surgical protocol and a 10-page consent form and submitted an application to my Institutional Review Board (IRB), which agreed 8 months later to allow me to operate on five patients. I performed the first EK surgery in the United States on March 14, 2000. I dubbed the procedure deep lamellar endothelial keratoplasty (DLEK) to distinguish it from multiple forms of PLK.2
In my first case, the edema cleared by 2 weeks postoperatively, and the eye had 1.00 D of regular astigmatism. The patient’s visual acuity was better at 6 weeks following DLEK than most PKP patients achieved at 6 months. I presented videos of the laboratory procedures and patient surgery at the AAO Annual Meeting that year. The editor of Cornea, Mark Mannis, MD, subsequently published my report on my first two cases.3 Soon thereafter, I began receiving requests from surgeons who wanted to visit me to learn to perform DLEK.
THE ENDOTHELIAL KERATOPLASTY GROUP
In the 1980s, I had witnessed the chaotic rollout of radial keratotomy, during which some surgeons had experimented with the placement of 32 or more incisions as well as hexagonal carving incisions without IRB approval—excesses that necessitated a PKP in some cases and caused the loss of some eyes. I wanted to avoid similar problems with EK, so I founded the international Endothelial Keratoplasty Group (EKG) for the ethical development of EK.
Each member of the EKG received copies of my protocol, IRB application, and consent form to use as templates when starting their own EK program. Before they could obtain DLEK instruments from Bausch + Lomb, they had to apply to their IRB. I trained more than 70 surgeons in DLEK free of charge, and thus began an ethical rollout of EK surgery.
For more than 20 years, the EKG met for breakfast discussions at annual meetings of the ASCRS and AAO. Participants’ frank discussions of data, techniques, and complications propelled the advancement of EK throughout the world (Figure). I retired the EKG meetings in 2023 when Descemet membrane endothelial keratoplasty (DMEK) became the standard form of EK.
Figure. Pictured from left to right are EK pioneers Dr. Terry; Dr. Melles; Friedrich Kruse, MD; and Dr. Price.
BETTER, FASTER, EASIER
Compared to PKP, DLEK offered several advantages: faster postoperative recovery, no irregular astigmatism, a stronger wound, and a lower risk of rejection.4 Few surgeons, however, completed large series of DLEK. Although they concurred that the procedure was superior to PKP, they also agreed that DLEK was too difficult to perform and took too long to complete. In addition, because DLEK had no specific Current Procedural Terminology (CPT) code, insurance payment for the surgery was tenuous at best.
In 2002, Dr. Melles reported on a technique that eliminated the difficult resection of the posterior recipient stroma. Instead, the recipient Descemet membrane was stripped away, and the donor stromal-endothelial tissue was placed on the smooth bed.5 In 2005, Francis Price Jr, MD, published early visual and refractive outcomes with Descemet stripping endothelial keratoplasty.6 The use of a mechanical microkeratome instead of manual dissection—a technique popularized and dubbed Descemet stripping automated endothelial keratoplasty (DSAEK) by Mark Gorovoy, MD—made preparing donor tissue easier.7
Large studies by our center and Dr. Price’s demonstrated that, compared to PKP and DLEK, DSAEK was better for the patient and faster and easier for the surgeon.8,9 Even so, DSAEK was not universally accepted.
INCREASING SAFETY AND ENSURING PAYMENT
Eye banks and the politics of CPT codes were critical to the universal acceptance of EK. The preparation of donor DSAEK tissue in the OR presented a significant risk. Two eye banks—one in North Carolina and the other Oregon—evolved from harvesting to processing facilities that could precut the donor tissue for the surgeon. This advance made EK surgery far easier, faster, and safer.
Insurance payment for EK surgery was the final hurdle. David Glasser, MD, led the charge for The Cornea Society to obtain a specific CPT payment code from CMS. Once this was achieved, EK was adopted worldwide. By 2011, EK was performed more often than PKP for the treatment of corneal edema.
PURELY ANATOMIC REPLACEMENT
I tell my residents and fellows, “The ideal surgery leaves the eye with the same exact anatomy as God and evolution intended.”
Dr. Melles’ early laboratory work removed and replaced Descemet membrane alone, without a stromal carrier, a step he referred to as the descemetorhexis. In a 2006 publication of his first case in a human patient, he called the procedure Descemet membrane endothelial keratoplasty.10 Surgeons initially considered the procedure to be too difficult and the preparation of the fragile donor Descemet membrane too risky. Similar to their role in DSAEK, eye banks began to provide surgeons with precut, prestained, premarked, and preloaded DMEK tissue to make the surgery faster, easier, and safer.11
DMEK is a purely anatomic replacement that, of all forms of EK, provides patients with the most natural vision in the shortest amount of time and is associated with the lowest rejection rate. As of 2023, DMEK was the most common form of EK in the United States.12
POSSIBLE FUTURE ADVANCES
For new surgical procedures to gain acceptance, they must be better for patients and easier and faster than previous iterations.13
Descemet stripping only is faster and easier than DMEK and appears to deliver comparable visual results with no risk of rejection. Descemet stripping only with adjunctive Rho kinase inhibitors is a reasonable alternative to DMEK in select patients who have centrally located guttae and excellent peripheral endothelial cell density. In patients who do not have good peripheral endothelial cell density, however, such as those with diffuse Fuchs guttae or pseudophakic bullous keratopathy, DMEK is a better option.
Also exciting is the potential of cultured endothelial cell injection to address the worldwide shortage of donor corneas. The procedure’s simplicity and speed are unmatched. It remains to be seen, however, whether the speed of patients’ recovery, their vision in the presence of glare, and the cells’ long-term survival rate are comparable to those with DMEK.
1. Melles GR, Eggink FA, Lander F, et al. A surgical technique for posterior lamellar keratoplasty. Cornea. 1998;17(6):618-626.
2. Azar DT, Jain S, Sambursky R, Strauss L. Microkeratome-assisted posterior keratoplasty. J Cataract Refract Surg. 2001;27(3):353-356.
3. Terry MA, Ousley PJ. Deep lamellar endothelial keratoplasty in the first United States patients: early clinical results. Cornea. 2001;20(3):239-243.
4. Ousley PJ, Terry MA. Stability of vision, topography, and endothelial cell density from 1 year to 2 years after deep lamellar endothelial keratoplasty surgery. Ophthalmology. 2005;112(1):50-57.
5. Melles GR, Lander F, Nieuwendaal C. Sutureless, posterior lamellar keratoplasty: a case report of a modified technique. Cornea. 2002;21(3):325-327.
6. Price FW Jr, Price MO. Descemet’s stripping with endothelial keratoplasty in 50 eyes: a refractive neutral corneal transplant. J Refract Surg. 2005;21(4):339-345.
7. Gorovoy MS. Descemet-stripping automated endothelial keratoplasty. Cornea. 2006;25(8):886-889.
8. Terry MA, Shamie N, Chen ES, Hoar KL, Friend DJ. Endothelial keratoplasty a simplified technique to minimize graft dislocation, iatrogenic graft failure, and pupillary block. Ophthalmology. 2008;115(7):1179-1186.
9. Price FW, Price MO. Descemet’s stripping with endothelial keratoplasty in 200 eyes: early challenges and technique to enhance donor adherence. J Cataract Refract Surg. 2006;32:411–418.
10. Melles GR, Ong TS, Ververs B, van der Wees J. Descemet membrane endothelial keratoplasty (DMEK). Cornea. 2006;25(8):987-990.
11. Potts LB, Bauer AJ, Xu DN, et al. The last 200 surgeon-loaded Descemet membrane endothelial keratoplasty tissue versus the first 200 preloaded Descemet membrane endothelial keratoplasty tissue. Cornea. 2020;39(10):1261-1266.
12. Van Meter W, Buckman N, Philippy B, DeMatteo J, America obotEBAo. 2024 Eye Banking Statistical Report—Executive Summary. Eye Banking and Corneal Transplantation. 2025;4(2):e0042.
13. Terry MA. Endothelial keratoplasty: Why aren’t we all doing DMEK? Cornea. 2012;31(5):469-471.
