eurotimes.org
EUROTIMES STORIES

DMEK finds its mark

Descemet’s membrane endothelial keratoplasty (DMEK) has come a long way in a few short years despite initial concerns about its technical difficulty

Dermot McGrath

Posted: Saturday, April 1, 2017

 

Descemet’s membrane endothelial keratoplasty (DMEK) has come a long way in a few short years despite initial concerns about its technical difficulty. While Descemet’s stripping (automated) endothelial keratoplasty (DS(A)EK) remains the most widely performed endothelial keratoplasty technique, the perceived advantages of DMEK in terms of visual outcomes, faster rehabilitation and precise anatomic reconstruction of the diseased cornea have won over many new adherents.

Nevertheless, while more surgeons are switching to DMEK, the transition from DSAEK to DMEK has not been as fast as the transition from penetrating keratoplasty (PK) to DSAEK, notes Harminder Dua MD, PhD, Chair and Professor of Ophthalmology, University of Nottingham, Queen’s Medical Centre, Nottingham, UK.

“The technicality of harvesting donor Descemet’s membrane (DM) is still a challenge and individual surgeons are more concerned about this step than the actual surgery on the patient. Loss of tissue incurs a fair expense. The clear advantage of DMEK is the restoration of normal anatomy, which in turns obviates any induced refractive change as seen with DSAEK,” he said.
Furthermore, procedures such as ultra-thin DSAEK and pre-Descemet’s endothelial keratoplasty (PDEK) also provide near-anatomical restorations without any induced refractive change, added Dr Dua. “These techniques offer other advantages in tissue handling and unrolling in the eye. They will remain or emerge as viable options to DMEK,” he said.

While the field of corneal transplantation is evolving rapidly, Dr Dua believes that it is still too early to write the obituary for PK. “My view is that PK will always be there, though the indications will shrink. DSAEK too is here to stay for a while longer. Globally there are still more surgeons doing PK and DSAEK than those doing DMEK,” he said.
Sadeer B Hannush MD, Attending Surgeon on the Cornea Service at Wills Eye Hospital, Thomas Jefferson University, Philadelphia, USA, agrees that the trend towards DMEK is moderate rather than frenzied.

“I’m in my fourth year of offering the procedure, yet in the USA less than half of all corneal surgeons currently offer DMEK. The Eye Bank Association of America (EBAA) 2015 statistical report indicates that 4,694 DMEK or DMAEK procedures were performed that year out of a total of 27,208 endothelial procedures. The conversion rate to DMEK in the USA has picked up over the past 18 months, while in Europe only a few centres and surgeons offer the procedure,” he said.
Dr Hannush said that surgeons are drawn to DMEK for a number of reasons: quicker visual rehabilitation, better visual results at the end point, lower incidence of rejection, and lower incidence of steroid induced ocular hypertension, because of the option of weaning the patient off steroids earlier.

Like Dr Dua, Dr Hannush sees DSAEK happily co-existing with DMEK for the foreseeable future. “One of my most sought-after talks at the moment is ‘DSAEK in the age of DMEK: Indications and Results’. The indications are usually severe comorbidities: unicameral eyes (aphakes, anterior chamber IOLs, and sutured posterior chamber IOLs), filters and tubes, vitrectomised eyes, and hypotonous eyes for any reason. As far as PK is concerned, it is still indicated when all layers of the cornea are involved in the disease process. I don’t believe that it is appropriate, however, for the indication of endothelial dysfunction, unless there is significant stromal scarring or fibrosis,” he said.

EVOLUTION OF DMEK
While there is a widespread perception that DMEK evolved from DSEK, Gerrit Melles MD, PhD, Director of the Netherlands Institute for Innovative Ocular Surgery, Rotterdam, points out that the evolution of the various EK techniques was not linear and that DMEK actually preceded DSEK.
“The funny thing is that we did our experimental DMEK surgeries around 1997, years before those on DSEK, but back then we did not have our own eye bank facilities in The Netherlands and to perform DMEK clinically seemed a bridge too far. In 2001, the first DSEK was performed and at the time it felt a step up from deep lamellar endothelial keratoplasty (DLEK) and PK. Then, after we received our eye bank licence in 2004, the first clinical DMEK was performed in 2006 and with the first patient reaching a visual acuity of 20/18 within the first month, that seemed the way forward,” he said.

As Dr Melles sees it, the different techniques should be seen as complementary rather than competitive. “It is not a competition between techniques but we apparently needed DSEK/DSAEK as an in-between step. Both techniques have their advantages and disadvantages and are probably both preferable over PK, because that was the starting point. For DMEK, it takes a bit of a different surgical mindset since all manoeuvres should preferably be ‘no-touch’, but I think that if surgeons give it a serious try they will be amazed at how rewarding it is to master DMEK,” he said.
The fact that DMEK is more technically difficult to master compared to DSEK definitely makes the procedure best suited to less complex cases such as early cases of Fuchs’ dystrophy, believes Donald TH Tan MD, Director of the Singapore National Eye Centre.

“There will definitely still be a role for DSAEK in the more challenging cases such as more severe pseudophakic bullous keratopathy, post-glaucoma surgery, and aphakic or vitrectomised eyes where there will be greater difficult in manipulation of the DMEK donor, or retaining air in the anterior chamber (AC) which is important for DMEK graft attachment,” he said.
DSAEK will also still remain a procedure of choice for those surgeons who fail to feel comfortable in performing DMEK surgery consistently, especially as ultra-thin DSAEK can deliver almost as good final vision as DMEK and may also result in a lower rejection rate as compared to conventional DSAEK with thicker donor tissue, he added.

STANDARDISED PROCEDURE
Part of the growing popularity of DMEK in recent years stems from improved efforts to standardise the procedure, noted Dr Melles.
“Looking back, I always feel like we fell short in standardising DLEK and DSEK/DSAEK before introducing it. For that reason, we have been working on a step-by-step approach to teach DMEK to colleagues and other surgeons have been working on similar standardisations as well,” he said.
Dr Dua agreed that surgical experience has allowed identification of clear procedures and steps to make the procedure more predictable for new learners. He also highlighted the key role of patients undergoing procedures that were continually being refined in enhancing the learning process.
“I do not think that we have reached the final stage as more refinement is still required. Detachment rates are still higher than they should be and the ideal procedure that gives zero detachment rates is still elusive,” he said.

Part of the ongoing refinement to improve results is coming from better instrumentation, points out Dr Tan. As he sees it, the challenge in DMEK is not so much in donor insertion as per DSAEK, but in donor manipulation in the AC to unroll the graft and position it correctly.
A few years ago, Dr Tan developed a Descemet’s Mat (D-Mat) device, which was based on a soft contact lens design, to lay the DM on the D-Mat, and coil it into the EndoGlide DSAEK inserter, and pull just the donor DM into the AC. “This prevents the inside-out scrolling up of the donor completely, so that no unscrolling was needed,” he said.
Similarly, Massimo Busin MD in Italy has recently reported using a standard soft contact lens placed on a DMEK donor inserter to help stabilise the donor in the inserter before pulling the unscrolled donor into the AC using a DSAEK forceps.

“These new techniques are completely different from the conventional unscrolling approach, and provide better AC control of the donor tissue,” said Dr Tan.

HYBRID DMEK
Taking the innovation drive a step further, Dr Tan now uses an approach he calls ‘hybrid DMEK’ in which donor DM is pre-stripped on a DSAEK pre-cut donor, and both posterior stromal lenticule and DM are loaded into an EndoGlide DSAEK cartridge in a non-scrolled and correct orientation. However, only the DM is pulled into the AC with DSAEK micro-forceps, leaving the posterior stromal lenticule behind in the EndoGlide chamber. The DM donor is then carefully positioned centrally in the AC, and an air bubble injected to attach the DM donor, before the forceps finally release the donor.

The advantages of the technique are that the donor tissue is always delivered the right way up, it is not allowed to scroll, and it can be gently manipulated centrally in position in the AC with the forceps until the air bubble accurately positions the donor. Furthermore, the technique is fairly similar to conventional DSAEK so the learning curve is improved, and there is more consistency and control in this surgical approach to conventional DMEK methods, said Dr Tan.
“I’ve performed over 30 cases of H-DMEK with no primary failures, no re-bubblings, and an improved endothelial cell loss of under 30%. We have now begun performing H-DMEK in more complex cases such as post-PK eyes, vitrectomised eyes and post-glaucoma surgery cases,” he said.

Some surgeons also highlighted the increasing role of intraoperative optical coherence tomography (OCT) in facilitating DMEK surgery.
“Intraoperative OCT is useful to visualise graft rolling, unfolding, orientation and monitoring of the graft attachment,” said Pierre Fournié MD, University Hospital Toulouse, France. “It is then easier to check the interface and graft adherence. It may also increase he adoption of DMEK surgery by flattening the learning curve,” he added.

GETTING TO GRIPS WITH COMPLICATIONS
In terms of complications associated with DMEK, Dr Fournié said that the detachment rate remains the main problem with the technique, but is related to surgeon experience.
“If initial procedures that needed a re-bubbling or re-grafting were high, as surgeons gain experience fewer patients require repeat procedures. Furthermore, many of these detachments reattach spontaneously.” he said.

Glaucoma can also occur after DMEK because of migration of a portion of the inserted air bubble behind the iris, inducing acute angle closure, noted Dr Fournié.
“This early complication can be easily solved with pupil dilation or aspiration of the air bubble. In addition, in phakic eyes, DMEK can induce a cataract, but can easily be removed without endangering the survival of the DMEK graft,” he said.

Dr Melles agreed that there is a direct correlation between surgical experience and complications in DMEK. “We are just in the process of finalising a paper on 2,500 DMEK eyes performed by different surgeons all over the world and the good news is that the incidence of most complications in DMEK tend to go down with technique experience. Partial detachment can most often be dealt with, but it takes some experience to recognise when to re-bubble or not. Secondary glaucoma can be largely avoided, although eyes with pre-existing glaucoma may need monitoring. The main problem is the group of eyes with angle-supported phakic IOLs, since virtually all of these eyes may develop uncontrolled glaucoma, whether the IOL is removed or not,” he said.
In dealing with detachments, Dr Dua warns that there is a clear risk associated with multiple re-bubbling, which is very likely to compromise endothelial function.

“The air bubble is the only way today to keep the graft in place. It is also the major cause of high or very high pressure and cataract. At present, we are all preoccupied with getting the graft to attach, often by putting more air. Understanding the biology of attachment and other means of limiting detachment will help address the issues of raised pressure and cataracts. This is an area to which attention and research will shift in the near future,” he concluded.

Harminder Dua:harminder.dua@nottingham.ac.uk
Sadeer B Hannush: sbhannush@gmail.com
Gerrit Melles: melles@niios.com
Donald TH Tan: donald.tan.t.h@singhealth.com.sg
Pierre Fournié: fournie.p@chu-toulouse.fr

 app advert
Latest Articles


escrs members advert