This article is more than 5 years old. Information may no longer be current.
DMEK graft failure not primarily from tissue manipulation, study finds
Donor endothelium appears to be more stable than previously thought, authors say.
Click Here to Manage Email Alerts
Contrary to expectations, tissue damage during harvesting and surgical manipulation was not the primary cause of graft failure in Descemet’s membrane endothelial keratoplasty, according to a retrospective study.
Early repositioning with bubbling is recommended for partially or fully detached grafts, whereas attached grafts that clear slowly simply need time and close observation, Gerrit R.J. Melles, MD, PhD, and colleagues said in the American Journal of Ophthalmology.
The authors set out to identify the causes and proper management of graft detachment in the 3 to 4 weeks after Descemet’s membrane endothelial keratoplasty (DMEK).
Only three of 11 grafts failed because of endothelial cell loss stemming from tissue manipulation, the authors said.
Analysis of the unsutured grafts showed the endothelial layer to be more resilient than expected, Dr. Melles said in a subsequent telephone interview with Ocular Surgery News.
“The key message of the article was that we expected a lot of damage from the surgery due to the implantation, and apparently you can do much more with the endothelium than you actually would think you can,” Dr. Melles said.
Patients and methods
The study group comprised 11 eyes of 11 patients, three men and eight women with an average age of 69.4 years. All patients underwent DMEK for Fuchs’ endothelial dystrophy or bullous keratopathy, the authors said.
|
Image: Melles G |
Donor age averaged 67.5 years. Donor endothelial cell count averaged 2,720 cells/mm2. All grafts were harvested within 36 hours after death and stored at 31° C.
Corneal-scleral rims were evaluated after 1 week and mounted. Descemet’s membrane was stripped from the posterior stroma, creating an 8.5-mm to 10-mm diameter flap of posterior Descemet’s membrane with its endothelial monolayer. Descemet’s rolls with the endothelium on the outside formed spontaneously and were stored for 7 to 14 days before transplantation. Rolls were stained with 0.06% trypan blue solution, the authors reported.
DMEK grafts were inserted through a 3.5-mm tunnel incision just inside the limbus. A 9-mm diameter central portion of the recipient Descemet’s membrane was removed. Grafts were oriented with their peripheral edge flaps anterior to the central part of the membrane. After grafts were positioned onto the recipient posterior stroma, the anterior chamber was filled with air for 30 minutes to promote graft adhesion. Air-liquid exchange was used to pressurize the eye.
Graft failure was suspected in all 11 eyes, all of which underwent secondary Descemet’s stripping with endothelial keratoplasty (DSEK) 2 to 5 weeks after the initial Descemet’s procedure. Each failed Descemet’s graft was explanted and examined by inverted light microscopy. Videos of DMEK procedures were studied to isolate causes of graft failure, the authors said.
High endothelial cell counts
All but two failed grafts had an endothelial cell density of 1,700 cells/mm2 or higher, the authors reported.
“It’s not the endothelial cell density that you would expect to be low, but there may be some other factor here that made the grafts fail,” Dr. Melles said. “It may also just be the learning curve, that at some point you’re doing something different and you’re not exactly sure what it is.”
Among the 11 failed grafts, six detached completely; among the remaining five eyes, two grafts detached partially and three grafts attached but failed to clear.
Of the six complete detachments, two were deemed the result of grafts being positioned upside-down. Both grafts had somewhat high endothelial cell densities (2,400 cells/mm2 and 2,700 cells/mm2, respectively), suggesting failure because of poor orientation, not tissue damage from surgical manipulation, the authors said.
In the two eyes with partially detached grafts, one graft was also found to be positioned upside-down. No cause of detachment could be found in the other eye.
Among the three eyes with attached grafts that did not clear, one had an endothelial cell density of 800 cells/mm 2, suggesting excessive manipulation of tissue that was difficult to unfold in the recipient anterior chamber, the authors said.
Dr. Melles attributed some graft failures to unknown biomechanical factors.
“You have patients who have a bilateral DMEK, and in one eye it clears quickly and in the other eye it takes much longer,” he said. “Apparently, there is something going on, but it’s difficult to pinpoint. … It’s probably some kind of biomechanical mismatch or something that is going on, but it is hard to pinpoint what it is.”
Early re-bubbling
DSEK may not be the ideal intervention for all DMEK graft failures, the authors said. Rather, partially or completely detached grafts may only require early re-bubbling, or injection of air to reposition the graft onto the recipient posterior stroma.
“If you see that a DMEK graft is partly detached after surgery, you might as well put an air bubble in very early after surgery so that it is still effective,” Dr. Melles said. “In DSEK, you would just wait. And in DMEK, you would probably choose an early re-bubbling.”
In DMEK, the donor tissue tends to scar and grow stiff, making reattachment by re-bubbling more challenging with time, he said.
“Most people will then wait because it may attach again by itself,” Dr. Melles said. “But it’s probably better just to put an air bubble in very early because it’s also known that sometimes it becomes difficult to get it attached again, and it may just be that Descemet’s begins to scar a little bit or becomes more stiff. During surgery, you can see that it’s not a flexible membrane anymore.”
Slow clearing does not necessarily indicate graft failure, he said, adding that some grafts clear slowly for reasons that are not apparent.
“Apparently, in some cases, there might be something in the endothelium that makes the endothelium slower to start up, so to speak,” Dr. Melles said. “This is a thing that we do not really see in DSEK patients so much. Most often in DSEK, if the transplant is attached, then most often it is clear from day 1 and becomes better and better in time.”
In cases in which grafts are attached but do not clear, the grafts should be observed for several weeks, he said. In those cases, secondary surgery should be delayed to give the grafts time to clear.
For more information:
- Gerrit R.J. Melles, MD, PhD, can be reached at Netherlands Institute for Innovative Ocular Surgery, Laan Op Zuid 390, 3071 A Rotterdam, The Netherlands; +31-10-297-4444; fax: +31-10-297-4440; e-mail: melles@niios.nl.
Reference:
- Ham L, van der Wees J, Melles GRJ. Causes of primary donor failure in descemet membrane endothelial keratoplasty. Am J Ophthalmol. 2008;145:639-644.
- Matt Hasson is an OSN Staff Writer who covers all aspects of ophthalmology. He focuses on regulatory, legislative and practice management topics.