Issue: April 2016
April 07, 2016
9 min read

Latest developments in endothelial keratoplasty aimed at graftless surgery, regeneration

Issue: April 2016
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

Endothelial keratoplasty has created a revolutionary paradigm shift in the treatment of corneal endothelial disease. Through several iterations the procedure has become increasingly selective, and further developments are now on the horizon.

According to Gerrit Melles, MD, PhD, the cornea specialist who laid the foundation for modern endothelial keratoplasty, there are two new directions in which the technique is developing, straying from the main path of transplanting donor tissue. On one hand, there is a surgical graftless approach, and on the other, a non-surgical biological approach.

“Both these techniques are still in their infancy. They do raise hope and I look at them with great interest, but if you ask me to express an opinion, I prefer to be cautious, wait and see,” he said.

The descemetorhexis-only approach was inspired by the unexpected, spontaneous functional recovery of the endothelium in several patients despite the failure or absence of a donor graft.

“One of our fellows at the time, Martin Dirisamer, was among the first to observe and speculate on such a phenomenon. The conclusion we reached was that the host cells must be involved in this repopulation, but a question mark remained on the role of the donor cells and the physical contact of the graft,” Melles said.

Results of descemetorhexis-only are so far inconsistent, he said. In his own cases and in some other series, the technique did not work, while other surgeons reported a high rate of success in selected Fuchs’ dystrophy patients.

“Somehow we must start to discriminate which cases need a donor and which cases can do without a donor,” Melles said.

According to Gerrit Melles, MD, PhD, the cornea specialist who laid the foundation for modern endothelial keratoplasty, there are two new directions in which the technique is developing, straying from the main path of transplanting donor tissue.

Image: Melles G


Descemetorhexis-only consists of removing a given area of Descemet’s membrane over the visual axis. Over time, cells migrate from the periphery and repopulate the area, forming a new layer of clear, healthy tissue.

“I tried with a patient of mine about 2 years ago,” Gregory Moloney, MBBS, FRANZCO, FRCSC, said. “This was prompted by several cases of corneal clearance after Descemet’s trauma or failed DMEK attachment. The final decision to proceed was encouraged by one case referred by a cataract surgeon who had inadvertently stripped off the central Descemet’s membrane in a curvilinear circle during femtosecond laser-assisted cataract surgery. We watched it very closely with the confocal microscope and saw cells migrating centrally as well as some cells undergoing what appeared to be cell division. We then tried with a Fuchs’ patient to see if the cornea could improve without a transplant just by removal of the central guttata. Best corrected vision improved from 20/40 to 20/20 and is stable after more than 2 years.”

Gregory Moloney

Kathryn Colby, MD, PhD, started doing descemetorhexis-only in January 2014 in a patient with Fuchs’ and cataract.

“He agreed to let me try this novel technique on him. I removed the cataract, implanted a lens and stripped the Descemet. He did incredibly well. By 1 month his cornea was clear, endothelial cells had repopulated the central cornea, and his vision returned 20/20. Nine months later, when I saw that his cornea was stable, I started offering this technique to additional patients,” she said.

Surgery is not different from the first stage of endothelial keratoplasty, except for the size of the descemetorhexis, which is much larger in Descemet’s stripping endothelial keratoplasty and Descemet’s membrane endothelial keratoplasty.

“Rather than the average 8 mm, we are just removing the central endothelium. We don’t really know what the optimal size is. It may vary between 4 mm and 6 mm, according to individual patients, but one thing we know is that the bigger the area you remove, the more the cells have to spread to cover,” Colby said.

Kathryn Colby

Clinical exams and endothelial imaging by confocal microscopy guide the surgeon in deciding the appropriate size.

Discriminating responders from non-responders

Both Moloney and Colby have had a fairly high percentage of success in their respective series.

In a total of 13 eyes of 11 patients, Colby and colleagues had a failure rate of 25%. In those patients, endothelial keratoplasty was performed. In the remaining 75%, the patient’s own endothelium migrated to cover the defect, the cornea cleared, and there was no need for transplantation. One-third of responders were clear within 1 month.

Moloney has treated eight eyes so far, six of which completely cleared with visual improvement.

“The other two patients are still in the early stages of healing, so we don’t know yet what the end results will be, but their progress is promising,” he said. “So far a salvage endothelial keratoplasty has not been required.”

Both surgeons observed that the patients who responded well to the therapy were those with predominantly central guttae and a relatively healthy periphery, from which cells could migrate and repopulate the center.

“I would not perform descemetorhexis-only in patients who have guttae over the entire cornea, suggesting that the disease is a global disease,” Colby said.

“In our study we are turning away quite a lot of patients, focusing on those who have predominantly a phenotype of Fuchs’ that is dense central guttata, enough to degrade vision, but also a population of healthy peripheral cells. Healing from migration seems to do most of the work, and we are learning how to do the surgery to make it have more chances of success,” Moloney said.

However, other factors, including genetic factors, may play a role.

“We are looking at the genetics of the patients in our series to see if we can detect any differences between the responders and the non-responders. We have also looked at factors that are known to increase oxidative stress, including diabetes and smoking history, but found no correlation with response or non-response,” Colby said.

Being able to discriminate between potential responders and non-responders is a crucial issue, according to Melles.

“If we cannot make this discrimination, it is difficult to offer patients this procedure. You might have edema for months, which may induce astigmatism and changes in the anterior cornea that are difficult to treat and may jeopardize the good outcomes of a secondary DMEK procedure,” he said.

“It is our impression that if you choose the patients correctly, there is a role for this procedure, and I think we are going to move away from thinking that we have one solution to Fuchs’ dystrophy to having, like for other ocular diseases, some choice,” Moloney said.

Injection of cultured cells

A second promising technique has been developed by Shigeru Kinoshita, MD, PhD, and consists of injecting cultured human corneal endothelial cells to heal endothelial damage.

Shigeru Kinoshita

“There are several laboratories that could culture endothelial cells, but the key point is that we combined the cells with a Rho kinase (ROCK) inhibitor to preserve cell morphology and enhance cell attachment and proliferation,” Kinoshita said.

Before injecting the cells, a 1.6-mm corneal incision is performed in the upper portion of the corneoscleral area. Some of the degenerated endothelial cells are scraped off from the Descemet’s membrane using a special silicone needle. One suture is applied to the incision, the aqueous humor is drawn from the anterior chamber, and 300 µL of the solution containing the cultured cells is injected. Immediately after the injection, the patient is placed face down for 3 hours.

Stripping the Descemet’s to remove the guttae may not be necessary, according to Kinoshita.

“We observed that with time, after cell injection, the density and severity of guttae tends to decrease because the healthy cells we inject do not express abnormal extracellular matrix. Maybe in the future we will try to combine cell injection with Descemet’s membrane removal, but all we like to do at this point is prove that cell injection is effective,” he said.


Early results encouraging

“We have done around 30 cases and have just finished summarizing the data from the first 11 cases, which showed beautiful results. The cornea regains transparency, leading to full recovery of visual acuity. We are almost ready to disclose these first clinical results,” Kinoshita said.

In his hands, the outcomes of this procedure are almost comparable to DMEK and superior to DSEK. In some patients DSEK was performed in one eye while the other eye received the cell injection. The patients said that the injected eye was better.

“It may be a bit too early to draw conclusions, but this procedure could become an alternative to DMEK, which requires a skillful surgeon. The injection procedure is easier and maybe more reproducible, not only because patients don’t need extensive time for surgery, but also because one donor cornea could provide enough cells to treat approximately 100 patients,” Kinoshita said.

In addition, rejection may be even less than with DMEK based upon basic science studies.

“We expect that there will be less or no endothelial rejection even though we use an allogenic cell source, but we need to prove this in further studies,” Kinoshita said.

The main issue is how to culture the human corneal endothelial cells, not so much from a technical point of view but from the point of view of securing safety, avoiding fibroblastic alteration and maintaining the original chromosomes.

“The cultured cells sometimes show chromosome alterations or abnormalities. If so, even though the culture is OK, we cannot use these endothelial cells because of safety issues. With our protocol, we have been able so far to prevent these alterations,” Kinoshita said. – by Michela Cimberle

Disclosures: Kinoshita reports he is a consultant for Senju Pharmaceutical. Colby, Melles and Moloney report no relevant financial disclosures.



Should cataract surgery be performed sequentially or concurrently with endothelial keratoplasty?

Francis W. Price Jr.


Combined surgery saves time and resources, is preferred by patients

DMEK and cataract surgery can be combined or staged for Fuchs’ dystrophy and cataracts. We have compared the results and found that the visual outcomes, complications and endothelial cell loss were similar with both strategies. So the decision is up to the surgeon and the patient. The surgeon is fully compensated for each procedure if the surgeries are done sequentially but not if they are combined. However, combining the surgeries is much more convenient for the patient and more efficient in terms of reducing the overall cost, facility resource utilization and time. We have found that patients and their caregivers overwhelmingly ask for combined surgery. As a result, when a patient has any cataract at all and visually significant Fuchs’ dystrophy, we typically perform combined phacoemulsification, IOL implantation and DMEK with topical anesthesia and IV sedation. For astigmatism exceeding 2 D, we typically recommend a toric IOL. We are currently evaluating ReSure sealant (Ocular Therapeutix) with these combined cases with promising results so far.

Francis W. Price Jr., MD, an OSN U.S. Edition Cornea/External Disease Board Member, is president of Price Vision Group, Indianapolis, USA. Disclosure: Price reports he has received research support from Ocular Therapeutix.


Sequential approach might spare transplantation in some patients

Alessandro Galan

Unless the disease has already progressed to the advanced stages, I prefer to do cataract surgery alone, making the patient aware that graft surgery might be needed at a later stage. In many cases I have seen Fuchs’ corneas, even with a low cell count, go through cataract surgery without any worsening of the primitive conditions and remain clear for a fairly long time, if not forever. A careful clinical evaluation is sufficient for the expert surgeon to judge whether the endothelium is or is not at risk of progressing to decompensation. Cataract surgery in these cases requires the adoption of endothelium-sparing strategies, such as the use of adhesive viscoelastic substances, reduced fluidics, a retrolimbal incision and a short tunnel, to avoid reaching the more sensitive center of the cornea. The overall operating time should be as short as possible, using for instance a quick chop technique. Early intervention is recommended in Fuchs’ patients to avoid the longer operating time and increased use of ultrasound required by more mature cataracts. Overall, the management of Fuchs’ patients, from preoperative assessment to the procedure itself, should be handled by expert surgeons, and disease progression should be monitored regularly thereafter.

Alessandro Galan, MD, an OSN Europe Edition Board Member, is head of the ophthalmology department, Sant’Antonio Hospital, Padua, Italy. Disclosure: Galan reports no relevant financial disclosures.