January 2018


All you need to know about cornea transplants
Alternative procedures seek to reduce stress on limited supply of donor corneas

by Liz Hillman EyeWorld Staff Writer

PDEK pre- and postop with a young donor shows clearance of the severe corneal haze.
Source: Amar Agarwal, MD

A patient with previously failed PK underwent PDEK and a single-pass four-throw pupilloplasty.
Source: Amar Agarwal, MD

Corneas of rabbits receiving tissue engineered graft in tissue engineered endothelial keratoplasty group swelled to almost 1,000 μm by day 4. Corneal thickness recovered to just under 400 μm by week 3. Corneas of rabbits receiving dose of corneal endothelial cell injection increased in corneal thickness slightly at day 4, but gradually thinned down to just under 500 μm thick by week 3. Control rabbit corneas had thickness of about 400 μm–425 μm and did not cross-react when stained with anti-human nuclei antibody.
Source: Jodhbir Mehta, MD

Primary Descemetorhexis without a graft, PDEK, and cultivated endothelial cells seek to serve more patients with corneal disease

Since Gerrit Melles, MD, PhD, first described posterior lamellar keratoplasty in 1998, there has been a revolution in techniques to avoid a penetrating keratoplasty (PK) in cases where a full-thickness transplant might not be needed, such as deep lamellar endothelial keratoplasty (DLEK), Descemet’s stripping endothelial keratoplasty (DSEK), and Descemet’s membrane endothelial keratoplasty (DMEK).1
Newer procedures are in the works with the goal of more efficiently utilizing or reducing dependence on donor tissue, resulting in positive patient outcomes. One global survey found that there is only one donor cornea available per 70 patients who might need it.2 The authors of this survey wrote that “it is also essential to develop alternative and/or complementary solutions, such as corneal bioengineering.”
Amar Agarwal, MD, Dr. Agarwal’s Eye Hospitals, Chennai, India, Kathryn Colby, MD, PhD, chair, Department of Ophthalmology and Visual Science, University of Chicago Pritzker School of Medicine, Chicago, and Jodhbir Mehta, MD, associate professor, Singapore National Eye Centre, discuss pre-Descemet’s endothelial keratoplasty (PDEK), primary Descemetorhexis without a graft, and cultivated corneal endothelial cells, respectively, sharing surgical basics, patient selection, and current findings/status of the technique.

Pre-Descemet’s endothelial keratoplasty

According to Dr. Agarwal, PDEK is “a variant of endothelial keratoplasty,” combining the advantages of DSEK and DMEK. The advantage of DSEK is a thicker graft from the donor, making it easier for the surgeon to prepare, insert, and manipulate in the recipient’s eye. But this graft, which ranges from 100–150 µm thick, adds to the overall corneal thickness after transplant and, as Dr. Agarwal put it, this means the endothelium has to “pump out more fluid. That is the problem in DSEK,” he said.
Conversely, in DMEK, the graft is only 15 µm, but Descemet’s is stuck to the stroma until an older age, necessitating the use of older donors. The thinness of the graft also makes it very fragile. “We just touch it, and it tears,” Dr. Agarwal said.
In PDEK, the graft is not only thinner than a DSEK graft (25 µm), but physicians use younger donors. In fact, Agarwal et al. showed how it was possible to use infant corneas for this purpose.3
The PDEK graft, which includes the pre-Descemet’s layer (Dua’s layer), Descemet’s membrane, and endothelium, is created using a 30-gauge needle connected to a syringe bevel up to enter the stroma and inject air to form a type 1 big bubble. The air separates these layers from the residual stromal bed. Trypan blue is injected into the big bubble to stain the graft and scissors finish the dissection. The graft is then loaded into an injector.
Various eye banks are already producing PDEK grafts, making it possible for the surgeon to only have to stain the graft with trypan blue and load it into the injector, Dr. Agarwal said.
For transplant, Dr. Agarwal stressed the importance of using a trocar anterior chamber maintainer. With air on via the AC maintainer, perform the descemetorhexis, scoring, and stripping with a reverse Sinskey hook. Once the patient’s Descemet’s is removed, turn the air off for insertion of the graft. Once it begins to unroll, inject air so the graft sticks against the overlying stroma, restarting air infusion afterward, Dr. Agarwal explained. Use a reverse Sinskey hook to spread out the graft, smoothing wrinkles.
“If you are an endothelial keratoplasty surgeon—or even if you are not—you can certainly learn PDEK,” Dr. Agarwal said, adding that there are some minor surgical differences when compared to DSEK and DMEK and reiterating the importance of a trocar AC maintainer in PDEK cases.
Dr. Agarwal said PDEK can be performed in all cases requiring some form of EK, including previously failed grafts. In November 2017, Dr. Agarwal published the first book on PDEK, which includes the various ways of how to surgically perform the procedure and the surgical techniques, such as a glued IOL or iris repair with his relatively new four-throw pupilloplasty technique.4

Primary descemetorhexis

As in PDEK, DMEK, and DSEK, before a graft can be inserted into the recipient eye, the patient’s Descemet’s membrane must be stripped off. But what if it were left that way, stripped and with no graft?
About 5 years ago, small case reports started to be published about failed DMEK grafts that spontaneously cleared in Fuchs’ patients, use of a rho kinase inhibitor to clear the cornea without a graft, and deliberate descemetorhexis without a graft with mixed results.5–8
This research had Dr. Colby testing the technique of a primary descemetorhexis in 2014 on one of her long-term Fuchs’ patients who was scheduled for cataract surgery. At 1 month, this patient’s cornea had cleared and his vision was 20/25. At 4 years out now, Dr. Colby said the patient is still 20/20 with a clear cornea. According to Dr. Colby, this and a follow-up retrospective case series was a “proof of concept that it could work.”9
At the same time Gregory Moloney, MD, Sydney, Australia, was doing similar work, publishing his first case in 2015 and later describing the use of a topical ROCK inhibitor (ripasudil, a ROCK inhibitor not FDA approved in the U.S.) to “save” failing cases.10–11
Dr. Colby said she has performed 10 such descemetorhexis cases, and all of her patients have chosen to obtain ripasudil on their own. Though she hasn’t formally analyzed her data yet, Dr. Colby said anecdotally, it seems the ripasudil speeds endothelial clearance by a factor of two.
“My average clearance in my original series was 3 months. The average now is about 6 weeks; I’ve had some clear as quickly as 3 weeks,” Dr. Colby said.
A primary descemetorhexis without a graft is not for everyone though, Dr. Colby stressed. The patients this seems to work well on are Fuchs’ patients with confluent guttae right in the center but the peripheral endothelium is preserved. Patients who would not do well with this technique include Fuchs’ patients who have wall-to-wall guttae out to the periphery and non-Fuchs’ patients requiring a graft, such as those with pseudophakic bullous keratopathy.
Preoperatively, patients need to be advised about the potential that the procedure will fail and they will need to have a traditional EK procedure. Corneal edema within the first few months is also common, but Dr. Colby said she has not had a patient upset about that, explaining to them preoperatively this is likely to happen.
Postoperatively, Dr. Colby starts patients on prednisolone six times a day and begins to taper them gradually after a week or two. If patients choose to obtain and use ripasudil, Dr. Colby said they start using it immediately postop.
The procedure itself is “well within the skillset” of a cornea surgeon, Dr. Colby said, explaining that the descemetorhexis is the first step of any EK procedure, just smaller. She noted recent research that showed performing a capsulorhexis-like tear on Descemet’s, rather than scoring and removing it, provides a smoother edge that may facilitate migration better than a jagged edge.
“The field of cornea has seen an incredible revolution in the last 15 years. I think forward-thinking people will look at this carefully because if you don’t have to put cells from someone else and give them steroids for the rest of their life, that’s a good thing,” Dr. Colby said. “If a patient comes in and they’ve got a 720-µm cornea and there’s no clear area, don’t do it on that one, do a regular EK. But if a 45-year-old person comes in with symptoms with central guttae, what are you going to do? DMEK and give them a cataract, they’ll be pseudophakic. The [primary descemetorhexis] procedure is quick, it’s not resource intense.”
Dr. Colby said her team is gearing up to do a side-by-side comparison of DMEK in one eye and the descemetorhexis without a graft in the other, looking at outcomes, complications, patient satisfaction, and cost.
“I know there’s people even in this country who don’t have access to a corneal surgeon for a transplant. … You can do this procedure and if it works that patient doesn’t need regular follow-up with a cornea specialist, whereas even if you do the best EK in the world, most people still give steroids and there’s risks of glaucoma and rejection and even though they’re low with a procedure like DMEK, they’re not zero,” she said.

Cultivated corneal endothelial cells

Dr. Mehta is part of a team that’s hoping to eliminate the issue of donor tissue availability by cultivating corneal endothelial cells from a donor that otherwise would have too low of an endothelial cell count to be transplanted. Dr. Mehta said their research has shown the ability to expand cells almost 240-fold.
“With that number, it allows you to take one donor cornea and make enough cells to treat 90 patients. Now we have something that could be a cell therapy approach,” he said, and “you’ve alleviated this issue on donor stress.”
Preliminary research for this technique involved establishing the best media to create a proliferative, morphologically sound endothelial monolayer.12 Dr. Mehta said they now use two different types of media to culture cells, one for expansion and the other to stabilize the morphology of the cells.
Animal testing has shown successful re-endothelialization with cultivated human donor cells, delivered either on a carrier or via injection.13–14 For the carrier method, endothelial cells are cultivated on a thin slice of human tissue—a biodegradable carrier is also in the works, Dr. Mehta said—and implantation and postoperative protocol follows exactly as in a DSEK. For injection, the patient’s Descemet’s is left intact, but Dr. Mehta uses a specially designed instrument from ASICO (Westmont, Illinois) to scrape off the endothelial cells. From there, the cultivated endothelial cells are injected and the patient lays face down for 3 hours.
Dr. Mehta said there are certain indications where one technique might be favored over the other. For example, patients with no bullous keratopathy, no Descemet’s scarring could fare well with cell injection, while a patient with a lot of central guttae would be better with cultivated cells transplanted on a carrier.
While culturing endothelial cells is resource intensive, Dr. Mehta said they performed analysis to determine the actual cost of cultured endothelial cell transplants vs. traditional donor transplants.15 He said even if the number of possible cultured cells from one donor could only treat 60–70 patients, the cost to treat one cornea in this way is about $400, compared to $3,000 for a full donor cornea.
“A lot of countries don’t have access to tissue, but even if you do have access, that’s still a significant reduction in cost,” he said.
At this point, Dr. Mehta said clinical trials have been approved for cultivated endothelial cells on a carrier and will soon begin enrolling.
Separately and only for Fuchs’ patients, Dr. Mehta described research with regenerative therapy, where the patient’s own Descemet’s membrane is removed and replaced with an acellular donor Descemet’s that had endothelial cells removed. The patient’s own endothelial cells at the periphery then migrate over the donor Descemet’s, which is acting as a scaffold but also prevents stromal cell migration to the posterior. Dr. Mehta said replacing with an acellular Descemet’s rather than stripping and leaving without a graft resulted in better recovery in a rabbit host.16


1. Terry MA. Endothelial keratoplasty: history, current state, and future directions. Cornea. 2006;25:873–8.
2. Gain P, et al. Global survey of corneal transplantation and eye banking. JAMA Ophthalmol. 2016;134:167–73.
3. Agarwal A, et al. Pre-Descemet endothelial keratoplasty with infant donor corneas: a prospective analysis. Cornea. 2015;34:859–65.
4. Agarwal A. Pre-Descemet’s endothelial keratoplasty (PDEK). New Delhi, India. Jaypee Brothers Medical Publishers. November 2017.
5. Balachandran C, et al. Spontaneous corneal clearance despite graft detachment in Descemet membrane endothelial keratoplasty. Am J Ophthalmol. 2009;148:227–234.
6. Arbelaez JG, et al. Long-term follow-up and complications of stripping Descemet membrane without placement of graft in eyes with Fuchs’ endothelial dystrophy. Cornea. 2014;33:1295–9.
7. Koenig SB. Planned Descemetorhexis without endothelial keratoplasty in eyes with Fuchs’ corneal endothelial dystrophy. Cornea. 2015;34:1149–51.
8. Okumara N, et al. Effect of the rho-associated kinase inhibitor eye drop (ripasudil) on corneal endothelial wound healing. Invest Ophthalmol Vis Sci. 2016;57:1284–92.
9. Borkar DS, et al. Treatment of Fuchs’ endothelial dystrophy by Descemet stripping without endothelial keratoplasty. Cornea. 2016;35:1267–73.
10. Moloney G, et al. Descemetorhexis for Fuchs’ dystrophy. Can J Ophthalmol. 2015;50:68–72.
11. Moloney G, et al. Descemetorhexis without grafting for Fuchs’ endothelial dystrophy– supplementation with topical ripasudil. Cornea. 2017;36:642–648.
12. Peh GS, et al. Cultivation of human corneal endothelial cells isolated from paired donor corneas. PLoS One. 2011;6:e28310.
13. Numata R, et al. Cultivation of corneal endothelial cells on a pericellular matrix prepared from human decidua-derived mesenchymal cells. PLoS One. 2014;9:e88169.
14. Mimura T, et al. Treatment of rabbit bullous keratopathy with precursors derived from cultured human corneal endothelium. Invest Ophthalmol Vis Sci. 2005;46:3637–44.
15. Tan TE, et al. A cost-minimization analysis of tissue-engineered constructs for corneal endothelial transplantation. PLoS One. 2014;9:e100563
16. Bhogal M, et al. Allogeneic Descemet’s membrane transplantation enhances corneal endothelial monolayer formation and restores functional integrity following Descemet’s stripping. Invest Ophthalmol Vis Sci. 2017;58:4249–4260.

Editors’ note: Dr. Mehta has a patent on the culture media used for cultivation of human corneal endothelial cells. Dr. Agarwal and Dr. Colby have no financial interests related to their comments.

Contact information

: dragarwal@vsnl.com
Colby: kcolby@bsd.uchicago.edu
Mehta: jodmehta@gmail.com

Alternative procedures seek to reduce stress on limited supply of donor corneas Alternative procedures seek to reduce stress on limited supply of donor corneas
Ophthalmology News - EyeWorld Magazine
283 110
220 183
True, 1