Corneal transplant

Selective lamellar keratoplasty, both endothelial and anterior, is taking hold around the world and is almost certainly the future of corneal transplantation, 2015 Binkhorst Medallist Donald TH Tan FRCS, Arthur Lim Professor, Singapore National Eye Centre, told the 2015 American Society of Cataract and Refractive Surgery Symposium in San Diego, USA.

But while endothelial keratoplasty (EK) has become the gold standard for treating endothelial dysfunction, loss of endothelial cells remains a challenge, Dr Tan observed. Similarly, deep anterior lamellar keratoplasty (DALK) remains technically challenging. Advances in surgical technology and technique are improving cell survival rates and making all lamellar procedures more manageable, which should lead to better long-term outcomes.

For example, in 2006, the taco folding tissue insertion technique used in Descemet’s stripping automated endothelial keratoplasty (DSAEK) led to endothelial cell loss above 60 per cent at six months in Asian eyes, Dr Tan reported. Switching to the sheets glide technique cut that to about 30 per cent in 2007, and in 2009 the EndoGlide inserter, developed by Dr Tan and colleagues, then cut that to about 15 per cent – less than the 20 to 40 per cent seen with penetrating keratoplasty (PK).

“We have managed to halve our endothelial cell loss in DSAEK with each insertion transition,” said Dr Tan.

DSAEK also produces better graft survival for repeat procedures than PK, Dr Tan noted. At year five, 51 per cent of grafts survived in eyes treated with PK after a failed PK, compared with 86 per cent of eyes treated with EK after failed PK (Ang, Mehta, Tan et al. In Press, AJO).

These results, along with a manageable learning curve and favourable economics, are making DSAEK the preferred procedure for treating endothelial dysfunction around the world, accounting for approximately half of procedures in the US, UK, Sweden and Singapore in 2012, and one-quarter or more in Italy, Germany and Australia, Dr Tan noted.

DMEK challenge

Descemet’s membrane endothelial keratoplasty (DMEK) is anatomically the most perfect EK procedure, adding no excess stroma, which results in better visual acuity outcomes, Dr Tan noted. The procedure is also low-tech, requiring only an IOL shooter, BSS and air to accomplish.

Even so, DMEK is technically very challenging, Dr Tan said. Dealing with its extremely thin and fragile donor tissue presents new challenges. The tissue is difficult to manipulate and touching it destroys endothelium. Even the masters of the technique, such as Gerrit Melles MD, Francis Price MD and Friedrich Kruse MD, see endothelial losses of 34 to 40 per cent, Dr Tan noted.

DMEK has also led to more donor tissue wastage and higher re-bubbling rates that DSAEK, Dr Tan noted. As a result it is not widely used, accounting for just six per cent of grafts and 11 per cent of EK procedures in the US in 2014.

However, the newly developed Descemet’s Mat (D-Mat) for use with the EndoGlide may make the membrane easier to handle, making DMEK a more similar procedure to DSAEK, and help reduce cell loss and donor tissue waste, Dr Tan said. This is still in the developmental stage, though whether it will enter the mainstream remains an open question, he added.

Similarly, DALK has proven superior to PK in graft survival outcomes and reduction in complications. Yet the surgery itself is technically challenging, prompting many to seek new solutions. Indeed, some have moved away from the “big bubble” lamellar separation technique, which can result in accidental perforations, to manual blunt dissection, Dr Tan said.

But manual dissection, too, remains more an art than a science, Dr Tan said. As a result, DALK remains less popular than DSAEK, with only a handful of countries, including Singapore, Brazil and Italy, using it in more than one-quarter of cases requiring keratoplasty, although it should be recognised that the indications for DALK, i.e stromal diseases with healthy endothelium, are also less than EK. Possible technological solutions include microkeratomes and femtosecond lasers, but these also have their limits.

Donald Tan: donald.tan.t.h@snec.com.sg