As optometrists, we are fortunate to be practicing at a time of such diverse innovation. There is little doubt that our understanding of pathology and our ability to effectively treat many conditions have grown by leaps and bounds on a yearly basis. The upside for this growth in technology is obvious. But on the downside, it is often difficult to separate fact from fiction when new data is presented on a monthly basis.

One of these developing areas in eye care is endothelial keratoplasty, in its most current incarnation known as DSAEK. While penetrating keratoplasty has proven to be a very successful treatment strategy, historically, EK produces similar end results with less risk and a shorter recovery.

Perhaps because the procedure is so new, however, there seems to be some confusion in the eye care community regarding its limitations. For example, regarding allograft rejection, I’ve heard it stated in no uncertain terms that an EK procedure cannot elicit a rejection response.

This is absolutely untrue, and I’d like to address this misconception.

Signs of Graft Rejection vs. Sources of Graft Failure

Graft Rejection
• Keratic precipitates (KP). These are the most significant representation of endothelial rejection. They cause breakdown in the endothelial pump (PK/EK).
• Stromal edema. As a sign of rejection, this is usually seen in the presence of KP; however, infrequently, acute stromal edema may occur as a result of rejection in the absence of KP. May also lead to failure (PK).
• Sub-epithelial infiltrates. These represent anterior stromal rejection. They don’t lead to failure, but they can lead to an aggressive endothelial rejection (PK).
• Gray epithelial line. This represents an advancing front of immunity after sensitization of the foreign epithelium. It’s a rare rejection type, as the epithelium is usually sloughed and replaced with host tissue naturally (PK).

Graft Failure
• Edematous cornea. This is the end result of rejections that cause failure; however, it may also simply be the endothelial pump failing on its own (PK/EK).
• Vascularization. Severe graft vascularization can be its own source of failure. It is not a sign of rejection, but it does create an increased risk of rejection. Pre-surgically vascularized corneas are at higher risk of rejection (PK).
• Astigmatism. Often forgotten of as a case of graft failure, pathologic levels of astigmatism, which are uncorrectable, comprise a failure of one of the key indications of transplant— refraction—and therefore constitutes failure (PK).
• Dehisced endothelial graft. This usually occurs in the first day after surgery. If attempted repositioning fails, is considered a primary or iatrogenic failure (EK).
• Scarring. Scarring of the graft as a result of infection or severe ocular surface disease constitutes failure (PK).

A Bit of History
Pioneered by Gerrit Melles, M.D., Ph.D., and widely disseminated by Mark Terry, M.D., and Francis Price, M.D., endothelial keratoplasty has altered the treatment paradigm for cases of endothelial decompensation. EK procedures—namely, deep lamellar endothelial keratoplasty (DLEK), Descemet’s stripping endothelial keratoplasty (DSEK), Descemet’s stripping automated endothelial keratoplasty (DSAEK) and recently, Descemet’s membrane endothelial keratoplasty (DMEK)—all attempt to replace the failed endothelial pump mechanism with a graft that includes a functional pump.

Historically, causes of endothelial failure, such as Fuch’s corneal dystrophy and pseudophakic bullous keratopathy, have accounted for 30% to 50% of cases of penetrating keratoplasty (PK) performed in the U.S.1,2 Although these conditions have typically been treated effectively with PK, the nature of this surgery requires a slow recovery of best correctable vision while leaving the eye more susceptible to penetrating trauma, which requires prolonged steroid use to temper immunologic rejection, increases the risk of substantial ocular surface disease, and may lead to infection or desiccation of the cornea.

EK procedures were developed with these issues in mind. Basically, they all involve stripping the host Descemet’s membrane (DM) and placing a donor endothelial sheet on a thin stromal button directly against the posterior face of the host cornea. These procedures require no sutures to hold the graft, because the endothelial pump creates an anterior-driven force substantial enough to hold the donor button in place. The total benefit: less effect on refraction—changes to the posterior corneal curvature have less effect on refraction, and lack of sutures result in less potential for induced astigmatism—with a much more rapid progression to visual stability (three to six months, vs. as long as 18 months with PK), less risk of ocular surface disease or penetrating injury, and potentially less risk for rejection.3

Know Your Terms
To understand the strengths and weaknesses of EK, we need to be familiar with the terminology of transplants. Here’s a review.

• Graft rejection. Rejection is an immunologic attack by host cells on the graft tissue. But, in discussions of transplantation, this term is often used incorrectly when referring to graft failure, a failure of graft tissue from either an immunologic or non-immunologic source.

Defining a failed graft as having been rejected requires evidence of immunologic attack on the corneal transplant. In corneal grafting, signs include subepithelial infiltrates and an epithelial rejection line in cases of full thickness transplantation, and most commonly as keratic precipitates (in EK and PK). White cells in the anterior chamber of a grafted eye, in the absence of another cause, while not always representative of a rejection episode should be treated aggressively—inflammation from any cause is enough to cause sensitization of the immune system to the foreign graft.

• Graft failure. Graft failure is very intuitive term. It simply means the graft is no longer functioning in its intended fashion. In a PK graft, this can be due to loss of clarity, refractive potential or tectonic integrity. There are many reasons for graft failure. Vascularization of the graft, pathologic levels of irregular astigmatism, rejection, desiccation of the ocular surface, infection leading to scarring and iatrogenic causes are some of the most common sources of graft failure. (Note that rejection is a cause of failure, but does not carry the same meaning as failure.)

• Primary graft failure. Practitioners may, when discussing their experience with EK patients, relate how the patient “rejected” on the first day and had to have the graft repositioned. What the practitioner is actually referring to is not graft rejection, but primary or iatrogenic graft failure.

Primary graft failure, by strict definition, refers to a donor cornea with an endothelial pump that was insufficient prior to transplantation. In these cases, the transplant does not function adequately to allow corneal deturgescence or (in cases of EK) graft adherence. This can manifest as an edematous/non-clearing cornea or a graft that is non-adherent despite repeated attempts at placement. This is not related to an immune response; therefore, is not graft rejection.

Iatrogenic graft failure is essentially the same as primary graft failure, but rather than the donor tissue causing endothelial dysfunction, it is caused by intraoperative trauma to the graft.

In either case, the recipient ends up with graft tissue with insufficient endothelial function.

The ‘Myth’ of Rejection
Simply stated, endothelial transplantation can lead to rejection. Even a cursory review of the literature will reveal this.4-6 Within the eye care community, however, it has been stated that, due to the absence of adjacent vasculature with EK, rejection cannot occur. Actually, increasing proximity to vasculature and the circulating immune cells contained within increases the risk of rejection, and the eye is exposed to immune cells along both vascularized and non-vascularized routes (i.e., through the anterior chamber). Rejection can be stimulated through either route.

Immunologic rejection of grafted tissue occurs when the immune system recognizes non-self antigens present in foreign tissue and stimulates an immune response to eliminate this perceived invasion. Within the genome of each cell, there is a region that encodes for antigen-presenting proteins on the cell membrane: the major histocompatibility complex (MHC) region. When transcribed, these proteins present antigens that enable the immune system to recognize the cell as part of “self” and avoid destruction, and direct an immune response to cells marked with a “non-self” antigen, marking that cell for destruction.

In non-auto grafting transplant techniques, the grafted tissue or organ stimulates an immune response in its expression of “non-self” surface peptides. This process can either take place directly—T-cells interacting with the MHC of graft tissue—or indirectly—stimulation of T-cells by an antigen-presenting cell (e.g., a B-cell that has processed a surface antigen of the grafted tissue). In either event, T-cell activation results in an immune response against the source of the antigen, which in this case is the graft.

In addition to the rich supply of antigen-presenting cells found in paracorneal vasculature and lymph tissue, the iris and trabecular meshwork contains a species of macrophages which have been shown to internalize foreign antigens shed by the corneal endothelium. In this case, an immune response can be mounted across the anterior chamber, rather than though limbal or corneal vasculature.7

How Does It Happen?
To understand how graft rejection can occur in EK procedures, the better-established PK serves as an example. The cornea has the longest history of successful transplantation of any organ or tissue in the body, stretching back more than 100 years—despite not having immunosuppressive agents, such as topical steroids, available for at least half that time.8 PK remains the most frequently performed transplant procedure and is among the most successful.

One of the reasons for PK’s history of success is the relative immune privilege enjoyed by corneal tissue. This level of shielding from the immune system is due to a handful of factors: limited vascular tissue in the non-inflamed cornea, absence of lymph tissue in the non-inflamed cornea, and a relative paucity of antigen-presenting cells in the cornea. All of these factors ensure that only a serious infectious or inflammatory threat can produce an immune reaction, which frequently results in disruption of corneal clarity.
Corneal grafts take advantage of these factors and their minimizing effects on immune response. Corneal grafts do not need to be immunologically matched with host tissue (which would result in 100% rejection rates of other tissue grafts), yet first-year corneal grafting success rates are higher than 90%.7

Immune shielding works beautifully in a non-inflamed eye; however, in an inflamed eye, the immune response becomes much easier to trigger. In eyes that have undergone corneal transplantation, the inflammatory response is often elevated simply because of the very factors that necessitated the transplant, and it can be an expected postoperative outcome.8 For example, keratoconic patients with corneal scarring and patients with endothelial decomposition who exhibit bullous keratopathy (both considered essentially non-inflammatory) do have elevated levels of inflammation compared to a normal eye. Further, such factors as dryness and multiple sutures (all part of the normal postoperative landscape for PK patients) increase inflammatory activity.9

Any source of inflammation can be sufficient to sensitize the immune system to the graft tissue and stimulate an acute graft rejection episode. This, if left unchecked, can lead to graft failure; in fact, it’s the most common cause of graft failure.10-12

The immune system requires no blood vessels to access the cornea; therefore, it shouldn’t come as a surprise that rejection can and does occur following EK procedures. But, some factors that stimulate rejection in PK patients are absent in EK procedures. Lack of sutures, lack of proximity to epithelial antigen-presenting Langerhans cells and greater distance from active blood supply would appear to provide EK grafts with greater protection. Nevertheless, any foreign antigens on the graft are enough to incite an immune response and cause a rejection. In fact, Dr. Terry’s group has published the first reported case of graft rejection as a cause of graft failure in an EK patient, and two recent studies by researchers in Indianapolis have described the features (primarily keratic precipitates) and frequency (12% at two years) of graft rejection in EK patients.4-6

The true question is not whether EK grafts reject, but whether they will reject less frequently than PK grafts. Prior studies place a rejection risk in endothelial causes of PK up to 65%.10-12 Compared to the Indianapolis study, these findings seem to indicate a reduced rejection potential. Therefore, both clinically and theoretically, the concept of reduced rejection risk in EK does seem to be validated to some degree—but no long-term studies on the procedure are available, so this should not be understood as an absolute truth. While it may be reasonable to expect less risk of immunologic rejection with EK procedures, it is still a risk, and post-EK patients should expect chronic use of topical corticosteroids as part of their treatment.

EK procedures offer many potential advantages over PK. One of these: a potentially reduced risk of immunologic rejection. But, these grafts can and do reject. Determining a true relative risk to PK will require long-term studies, which are currently unavailable. EK’s strengths are many, but we need to be upfront and accurate in our patient education prior to referring for surgery.

Aaron Bronner is staff optometrist at Hollingshead Eye Center, a secondary care facility in Boise, Idaho. His residency was at Davis Duehr Dean in Madison, Wis., under Christopher Croasdale, M.D.

1. Maeno A, Naor J, Lee HM, et al. Three decades of corneal transplantation indications and patient characteristics. Cornea. 2000 Jan;19(1):7-11.
2. Mamalis N, Anderson CW, Kreisler KR, et al. Changing trends in the indications for penetrating keratoplasty. Arch Ophthalmol. 1992 Oct;110(10):1409-11.
3. Bronner A. Descemet’s stripping automated endothelial keratoplasty. Optom Vis Sci. 2008 Sep; 85(9):808-13.
4. Allan B, Terry M, Price FW, et al. Corneal transplant rejection and severity after endothelial keratoplasty. Cornea 2007 Oct; 26(9):1039-1042.
5. Price MO, Jordan CS, Moore G, Price FW. Graft rejection episodes after Descemet’s stripping with endothelial keratoplasty. Part one: clinical signs and symptoms. Br J Ophthalmol. 2009 Mar;93(3):387-90.
6. Price M, Jordan C, Moore G, Price F. Graft rejection episodes after Descemet’s stripping with endothelial keratoplasty. Part two: the statistical analysis of probability and risk factors. Br J Ophthalmol. 2009 Mar;93(3):391-5.
7. Williams K, Coster, D. The immunobiology of corneal transplantation. Transplantation. 2007 Oct 15;84(7):806-13.
8. Niederkorn J. Immune mechanisms of corneal allograft rejection. Curr Eye Res. 2007 Dec;32(12):1005-16.
9. Panda A, Vanathi M, Kumar A, et al. Corneal graft rejection. Surv Ophthalmol. 2007 Jul-Aug;52(4):375-96.
10. Maguire MG, Stark WJ, Gottsch JD, et al. Risk factors for corneal graft failure and rejection in the collaborative corneal transplantation studies. Ophthalmology. 1994 Sep;101(9):1536-47.
11. Naacke H, Borderie VM, Bourcier T, et al. Outcome of corneal transplantation rejection. Cornea. 2001 May;20(4):350-3.
12. Claesson M, Armitage WJ, Fagerholm P, Stenevi U. Visual outcome in corneal grafts: a preliminary analysis of the Swedish Corneal Transplant Register. Br J Ophthalmol. 2002 Feb;86(2):174-80.