Instead of replacing all layers of the cornea (as in PK), lamellar keratoplasty selectively replaces only the diseased component of the cornea, namely Anterior Lamellar Keratoplasty (ALK) in corneal stromal diseases without endothelial dysfunction, and Endothelial Keratoplasty (EK) in cases where only the endothelium is compromised.1
In this article, we will focus on the evolution of endothelial keratoplasty techniques, and emerging therapies on the horizon, for the treatment of corneal endothelial diseases.
The cornea comprises of five main layers: the epithelium, Bowman’s membrane, stroma, Descemet Membrane (DM) and the endothelium.
The endothelium is the monolayer of cells that lines the posterior corneal surface, which is derived from the neural crest during embryologic development.
The metabolically active endothelium serves the important function of maintaining the health and clarity (and therefore transparency) of the corneal stroma through ensuring stromal deturgescence, by both acting as a barrier to fluid movement into the cornea and osmotically drawing water into the aqueous humour as an active pump.
Corneal endothelial cells, however, have a poor capacity for regeneration when lost due to any traumatic insult or disease.
With significant attrition of endothelial reserves below a critical level needed to maintain corneal deturgescence, the cornea is said to have “decompensated”, heralding the onset of edema which disrupts the orderly lamellar organisation and critical spacing of collagen fibrils within the stroma and in turn, degrades the optical transparency of the cornea (Refer to Figure 1).
Diseases of the corneal endothelium may be broadly classified into primary and secondary corneal endotheliopathies.2
The most common primary endotheliopathy is Fuchs’ Endothelial Corneal Dystrophy (FECD), an inherited bilateral disease which becomes evident from the fifth decade onwards (Refer to Figure 2). It is characterised by slowly progressive corneal endothelial dysfunction and guttate excrescences on the posterior corneal surface (Refer to Figure 3). Various genetic loci and mutations have been found in both adult and early- onset disease. A gender preponderance towards females is also well recognised in this disease.
Other primary endotheliopathies include posterior polymorphous dystrophy, congenital hereditary endothelial dystrophy in the paediatric age group and iridocorneal endothelial syndrome, of which the clinical details and characterisation are beyond the scope of this article.
Secondary corneal endotheliopathies may be the result of any extrinsic factor or ocular condition which contributes to the loss of endothelial cell structure and function, including intraocular surgery (in particular cataract extraction), laser peripheral iridotomy performed in eyes with angle closure, chronic ocular inflammation (such as uveitis and viral endothelitis), glaucoma and contact lens use.
Of these, corneal decompensation after cataract surgery or Pseudophakic Bullous Keratopathy (PBK) is a major indication for corneal transplantation worldwide and locally (Refer to Figure 4). Depending on the amount of surgical trauma, a variable extent of endothelial cell loss is usually expected 1 to 5 days after cataract surgery.
Even after these initial cell losses, it has been shown that endothelial cell attrition continues at a rate of 2.5% for at least 10 years after surgery, around 4 times higher than the average rate of 0.6% in unoperated eyes.2
Clinical Presentation Clinical manifestations of corneal endothelial dysfunction may range from mild to severe.
Before the onset of frank corneal decompensation and edema, many patients may be asymptomatic and features of dystrophy seen in FECD or posterior polymorphous dystrophy may be incidentally detected during a routine ophthalmic examination.
Fuchs’ endothelial corneal dystrophy and bullous keratopathy (predominantly pseudophakic) are the major indications for corneal transplantation globally.
Interestingly, there are clear differences in the disease indications for corneal transplantation between developed (mainly Western) countries (such as the United States, the United Kingdom, Italy, France, Australia), in which FECD was the major indication, and less developed countries (mostly in Asia, Africa and the Middle East), where PBK was the main indication for surgery, accounting for 31%-38% of cases.3
In Singapore, PBK alone accounted for 32.4% and 46.6% of penetrating and endothelial keratoplasty done at the Singapore National Eye Centre (SNEC) between 2000 and 2011 respectively. Comparatively, only 10.1% of PK and 32.0% of EK cases were performed for FECD locally.3
Over the past decade, endothelial keratoplasty has overtaken penetrating keratoplasty as the corneal transplantation technique of choice for endothelial diseases, as a result of its superior safety profile, as well as more rapid and predictable visual outcomes as compared to PK.
Briefly, penetrating keratoplasty is a full thickness transplantation procedure, in which the host diseased cornea is trephined and completely excised before a full thickness corneal graft is sutured to the recipient bed (Refer to Figure 5). It is a relatively undemanding technique which can be employed for all forms of corneal disease (stromal and/or endothelial) and potentially offers the best optical result due to the absence of a lamellar corneal interface.
However, endothelial failure with cell loss of 30%-40% at the time of surgery, followed by progressive attrition of endothelial cells for 10 years after the procedure, is a major cause of graft failure.
Penetrating keratoplasty is also associated with a high risk of graft rejection estimated at 20% by 5 years, which may adversely affect long-term graft survival. The full-thickness, “open-sky” surgical approach in PK may also be associated with serious complications, such as suprachoroidal haemorrhage and endophthalmitis, often with blinding consequences.1
Endothelial Keratoplasty (EK), on the other hand, involves stripping the host Descemet’s membrane and endothelium, and attachment of the donor endothelium and Descemet’s membrane, with or without residual donor stromal tissue.
Over the years, EK techniques have evolved to be progressively selective in the layers of corneal replacement. From:
i. Descemet Stripping Automated Endothelial Keratoplasty (DSAEK) With progressive refinement in surgical technique and donor preparation by eye banks, DSAEK has rapidly emerged as the main EK technique among surgeons and replacing PK as the procedure of choice for endothelial disease (Refer to Figure 6).
Chiefly, DSAEK offers faster visual rehabilitation compared to PK, due to the reduction of surgically induced astigmatism as fewer sutures are used. Many of the disadvantages of PK, such as suture-related complications, graft rejection and wound dehiscence, are also greatly reduced in DSAEK.1
ii. Descemet Membrane Endothelial Keratoplasty (DMEK) DMEK represents the next step in the evolution of EK technique. Despite all the benefits of DSAEK, the presence of residual stroma tissue in the donor may contribute to a postoperative hyperopic shift in the patient’s refraction and in some cases, suboptimal visual recovery DMEK overcomes this issue through a more anatomically accurate replacement of only donor Descemet Membrane (DM) and endothelium without any stromal tissue, potentially leading to a more rapid visual recovery with minimal refractive change (Refer to Figure 7).
Other purported advantages of DMEK include better visual outcomes and a lower risk of graft rejection, requiring a shorter, less intensive postoperative steroid regimen with a lower likelihood of elevated intraocular pressure and glaucoma.
However, difficulty in donor preparation and the technically demanding nature of DMEK surgery has thus far limited its widespread adoption.4
Singapore National Eye Centre (SNEC) performs approximately 85% to 90% of all corneal transplantations in Singapore.
All transplantations done in SNEC are tracked by the Singapore Corneal Transplant Study (SCTS), an ongoing prospective registry maintained by the Singapore Eye Bank, which is the national eye bank in Singapore supplying corneal tissues.3
The SNEC Endothelial Keratoplasty (EK) programme was started in 2006 and since then, we have had a high adoption rate, with over 50% of all corneal transplantations being EK from 2012.
Our surgical technique has evolved with time in order to adapt to the unique configuration of the Asian eye (generally smaller, shallower anterior chamber and higher posterior vitreous pressure), which made EK surgery more challenging.
We started with the standard taco-folding technique at that time (donor folded into a taco shape before insertion into anterior chamber) and progressed to adopting pull-through techniques using either the Busin Glide (donor coiled into an open-ended metal spatula inserted through a temporal wound and pulled into eye by a coaxial forceps from a nasal paracentesis) or Sheets Glide (donor placed on a plastic sheet inserted through the wound into the anterior chamber).
One of the key innovations which our centre has contributed to the field of EK surgery was the development of a donor inserter device, the EndoGlide, which improved the surgical control of both the donor tissue and anterior chamber dynamics during insertion (Refer to Figure 8).
This resulted in a lower endothelial cell loss of 13.1% and 15.6% at 6 and 12 months after surgery respectively, much lower than those published with the taco-folding and Busin Glide techniques.5 Given the technical advantages, short learning curve and better donor endothelial cell preservation, EndoGlide DSAEK very quickly became the donor insertion method of choice in our centre and has also been widely adopted worldwide.
Our data has shown that the 5-year graft survival of DSAEK (79.4%) was superior to that of PK (66.5%) with a lower rate of endothelial cell loss seen in DSAEK (48.7%) over PK (60.9%).6
In our hands, DSAEK also provided significantly better long-term best spectacle corrected visual acuity and lower astigmatism, compared to PK over 5 years of follow up.7
In recent years, our centre has increasingly performed DMEK, as an alternative to DSAEK for patients with endothelial dysfunction and corneal decompensation, with more than 150 cases done to date.
Two main surgical techniques of donor insertion are used in our centre, namely with a glass injector and the EndoGlide, in which a thin layer of stromal layer is retained as a carrier for the thin donor graft to improve handling characteristics, before the donor is pulled through into the anterior chamber.
Unpublished data from our series have suggested better graft survival in DMEK over DSAEK and PK, with lower rates of graft rejection and postoperative elevation of intraocular pressure or glaucoma.
There is now mounting interest in medical, cell and gene therapies, which have emerged as potential alternatives to keratoplasty in the management of endothelial dysfunction over the past few years.8
Rho-associated Kinase (ROCK) inhibitors have been extensively studied as a novel pharmacological adjunct in the management of endothelial disease, in particular FECD.
The agent, Y-27632, first demonstrated efficacy in stimulating corneal endothelial regeneration via enhancement of endothelial cell proliferation, migration and adhesion in in vitro experiments. These effects were also replicated in subsequent in vivo animal models and ex vivo studies. However, the safety and efficacy of ROCK inhibitors have not yet been evaluated in adequately power human clinical trials.
With the vast improvements made in techniques used to enhance both the quality and quantity of cultured human corneal endothelial cells, cell therapy with the introduction of engineered endothelial cells onto the posterior corneal surface to replace or reform the endothelial cell layer, represents a promising approach to treating corneal endothelial disease.
Different methods of delivery of these cells, either by direct injection into the anterior chamber or via a cell carrier, have been proposed.
With the cell carrier approach, synthetic or biological endothelial grafts may be constructed by seeding the cultured endothelial cells at physiological density (which may be controlled, hence overcoming the issue of biological variability of endothelial cell count in cadaveric donors) onto a thin, transparent biomimetic material. These tissue-engineered grafts may then be inserted into the recipient using techniques similar to DSEK or DMEK. A clinical trial to evaluate the safety and efficacy of this technique will soon be conducted at our centre.
Lastly, with advances made in our understanding of the genetic basis of FECD and drawing from the experience and clinically promising results of novel gene therapy modalities in the treatment of other trinucleotide repeat diseases, such an approach may also be applicable to FECD in the future.
Endothelial Keratoplasty (EK), which provides excellent clinical results in the form of improved visual outcomes, faster visual recovery, higher graft survival rate and a lower rate of postoperative complications compared to PK, will likely remain the mainstay of treatment for corneal endothelial diseases in the near future. Various novel cell and gene therapies, as well as pharmacological adjuncts, are on the horizon in this rapidly evolving field.
GPs can call for appointments through the GP Appointment Hotline at 6322 9399 for more information.
Dr Woo Jyh Haur is currently a consultant at the Singapore National Eye Centre. He was accredited as a specialist in Ophthalmology in 2015 and is a Fellow of the Royal College of Ophthalmologists (FRCOphth), Royal College of Surgeons of Edinburgh (FRCSEd) as well as the Academy of Medicine of Singapore (FAMS). His clinical and research interests include corneal and external eye diseases, corneal transplantation and refractive surgery.
1. Tan DT, Dart JK, Holland EJ, Kinoshita S. Corneal transplantation. Lancet. 2012 May 5;379(9827):1749-61.
2. Bourne WM. Biology of the corneal endothelium in health and disease. Eye (Lond). 2003 Nov;17(8):912-8.
3. Tan D, Ang M, Arundhati A, Khor WB. Development of Selective Lamellar Keratoplasty within an Asian Corneal Transplant Programme: The Singapore Corneal Transplant Study (An American Ophthalmological Society Thesis). Trans Am Ophthalmol Soc. 2015;113:T10.
4. Ang M, Wilkins MR, Mehta JS, Tan D. Descemet membrane endothelial keratoplasty. Br J Ophthalmol. 2016 Jan;100(1):15-21.
5. Khor WB, Mehta JS, Tan DT. Descemet stripping automated endothelial keratoplasty with a graft insertion device: surgical technique and early clinical results. Am J Ophthalmol. 2011 Feb;151(2):223-32.
6. Ang M, Soh Y, Htoon HM, Mehta JS, Tan D. Five-Year Graft Survival Comparing Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty. Ophthalmology. 2016 Aug;123(8):1646-1652.
7. Fuest M, Ang M, Htoon HM, Tan D, Mehta JS. Long-term Visual Outcomes Comparing Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty. Am J Ophthalmol. 2017 Oct;182:62-71.
8. Soh YQ, Peh GS, Mehta JS. Evolving therapies for Fuchs’ endothelial dystrophy. Regen Med. 2018 Jan;13(1):97-115.
Subscribe to our mailing list to get the updates to your inbox