Corneal graft rejection

Clinical background

Corneal transplantation is the most common and successful form of human solid-tissue transplantation, which is widely practiced as a sight-restorative therapy for patients with congenital or acquired corneal opacification, infection, or damage ( Box 8.1 ). The major indications for this procedure include keratoconus (corneal thinning and warping which cause visual distortion), bullous keratopathy (corneal edema, which is both painful and reduces visual acuity), failed previous grafts, corneal scarring, corneal dystrophy, and infection. Currently, the most common form of corneal transplantation is “penetrating,” which is the engraftment of a full-thickness corneal button. However, partial-thickness grafts or “lamellar” transplantation is also performed in a significant number of patients. Outcomes of corneal transplantation are typically excellent; 2-year graft rejection rates are approximately 10% in uncomplicated first grafts. However, this preponderance has led to the common misconception that immune rejection is not a significant clinical problem. Indeed, immune rejection is the leading cause of corneal graft failure, and, as reported in the 2006 Eye Banking Statistical Report, the number of regrafts due to previous failure is increasing.

Box 8.1

Clinical significance

  • Corneal transplantation is the most common and successful form of solid-tissue transplantation

  • Immune rejection is the leading cause of corneal graft failure

Historical development

Several early studies have been formative in our current understanding of corneal immune rejection and instrumental in stimulating the rapid progress made during the past two decades. Just over 100 years ago, Zirm reported the first successful human corneal transplantation, and subsequently corneal immune rejection was described by Paufique et al in 1948 and further confirmed by Maumenee in 1951. Arguably one of the most important studies in corneal immune rejection was carried out by Khodadoust and Silverstein in 1969. They demonstrated, by transplanting individual layers of the cornea in rabbits, that the epithelium, stroma, and endothelium could separately undergo immunologic rejection – important observations which are seminal in the diagnosis, prevention, and treatment of corneal immune rejection even today.

Key symptoms and signs

While symptoms are by no means universal, patients undergoing or in the very early stages of a graft rejection episode may experience irritation or pain, redness of the eye, decreased vision, and photophobia. Such symptoms may occur as early as 1 month or as late as 20 years after transplantation. Signs of rejection include circumcorneal injection, which is the dilation and engorgement of blood vessels around the circumference of the cornea and conjunctiva. A mild-to-moderate form of anterior-chamber reaction (cellular) and flare (acellular) may also be associated. In addition, edema and presence of keratic precipitates on the donor endothelium, either diffusely scattered precipitates or in an irregular line (commonly referred to as a “Khodadoust line”), are key signs of graft rejection ( Box 8.2 ).

Box 8.2

Signs and symptoms of corneal allograft rejection

  • Signs and symptoms of corneal graft rejection can occur as early as 1 month or as late as 20 years after transplantation

  • Edema and presence of keratic precipitates on the donor endothelium (Khodadoust line) are key signs of corneal graft rejection

Clinical features

There are potentially three distinct forms of corneal graft rejection, which may occur singly or in combination. They include: (1) epithelial rejection; (2) stromal rejection; and (3) endothelial rejection. Endothelial rejection, however, is the most common and profound form. It can be identified by endothelial surface precipitates in scattered clumps or in a classic linear form (Khodadoust line) ( Figure 8.1 ) that usually begins at a vascularized portion of the peripheral graft–host junction and progresses, if untreated, across the endothelial surface over several days. A mild-to-moderate anterior-chamber cellular and flare reaction may be associated with the process. Damage to the endothelium results in compromised regulation of corneal hydration and thereby is associated with edema, in addition to inflammation.

Figure 8.1

Khodadoust line: A clinical feature of endothelial rejection. This is a rejection line (arrows) formed by adherent macrophages and T cells, referred to as keratic precipitates, which traverse across the donor endothelium leaving behind a zone of graft destruction. The presence of a Khodadoust line is associated with graft edema since damage to the endothelium results in compromised regulation of corneal hydration.

Other forms of corneal graft rejection, including epithelial, subepithelial, and stromal rejection, occur less frequently (10–15% of all rejected cases; Box 8.3 ). These forms of rejection are not problematic per se; however, they often serve as harbingers of a more serious endothelial rejection.

Box 8.3

Graft endothelial rejection

  • Endothelial rejection is the most common and profound form of corneal graft rejection


The number of corneal transplants carried out worldwide is thought to exceed 70 000 per year. Eye Bank Association of America (EBAA) alone provided 45 000 donor corneas in 2006. Of corneal transplants placed in uncomplicated or “normal-risk” graft beds (i.e., absence of inflammation and neovessels), approximately 20–40% experience at least one bout of immune rejection. In spite of this, only 10% of grafts fail due to immune rejection by 1 year postsurgery in the normal-risk setting, since rejection is often reversible with intensive steroid treatment. By 15 years postsurgery, graft failure due to immune rejection nearly doubles to 17%, according to the Australian Corneal Graft Registry.

In high-risk graft beds (i.e., presence of inflammation and neovessels), which make up approximately one-third of all transplants, 50–90% of grafts fail even with maximal topical and systemic immune suppression. Indeed, these rates in high-risk transplantation are far worse than those experienced in vascularized solid-organ transplantation (e.g., heart, liver, or kidney). Moreover, data reported in the 2006 Statistical Report from the EBAA show a significant increase in the proportion of patients needing second and third grafts, by definition also considered high-risk.

Risk factors

There are several important risk factors which are used universally to determine if a patient is at high risk for corneal graft rejection and these factors have been established by large, multicenter, prospective studies such as the Corneal Transplant Follow-Up Study in the UK, the Collaborative Corneal Transplantation Studies (CCTS) in the USA, and the Australian Corneal Graft Registry. While numerous risk factors for immune rejection have been considered and studied extensively (e.g., gender-matching, age of donor, circumference of graft tissue), the two most important prognostic factors are stromal neovascularization ( Figure 8.2A ) and host bed inflammation ( Figure 8.2B ; Box 8.4 ).

Figure 8.2

Major risk factors for corneal graft rejection include neovascularization and inflammation of host graft bed. Corneal neovascularization is almost invariably associated with high graft rejection and the level of blood vessel ingrowth at the time of transplantation is directly correlated with graft survival (A). Corneal inflammation is another important risk factor and, even in previously inflamed graft beds which are quiet at the time of transplantation (e.g., herpetic eye disease), substantially diminishes the chance of graft survival (B).

Rights were not granted to include this figure in electronic media. Please refer to the printed book.

(Redrawn with permission from Williams KA, Lowe MT, Barlett CM, et al (eds) The Australian Corneal Graft Registry: 2007 Report. Flinders: Flinders University Press, 2007.)

Box 8.4

Risk factors for corneal allograft rejection

  • The two most important risk factors in corneal graft rejection are stromal neovascularization and host bed inflammation

Corneal neovascularization ( Figure 8.3C ) is almost invariably associated with high graft rejection and the level of blood vessels at the time of transplantation is significantly correlated with graft survival. Khodadoust reported that endothelial rejection occurred in 3.5% of avascular cases, 13.3% of mildly vascular cases, 28% of moderately vascular cases, and 65% of heavily vascularized cases. As per the CCTS definition, graft bed vascularization in two or more quadrants is classified as “high-risk.”

Figure 8.3

Animal models of corneal transplantation are powerful tools in the study of corneal graft rejection. In the mouse, accepted allografted cornea (A) is indicated by a clear and readily visible pupillary margin (arrow) and iris vasculature via slit-lamp observation. In contrast, for an allograft undergoing immunologic rejection (B), the pupillary margin is not readily visible (arrow). Grafts undergoing immunologic rejection are associated with graft edema, as indicated by corneal thickness observed via obtuse angle slit-lamp illumination in rejecting grafts (B* versus A*). As in humans, graft rejection in the murine model is associated with pathologic neovascularization (C, arrow) of graft bed and donor tissue. Graft failure due to immunologic rejection results in a severely opaque donor tissue (D).

Corneal inflammation is another important risk factor. Even previous inflammation in graft beds which are quiet at the time of transplantation (e.g., herpetic eye disease) has substantially diminished chances of graft survival ( Figure 8.2B ). This could be due to a subclinical presence of inflammatory cells and persistence of vascular channels. Transplantation in inflamed graft beds at the time of surgery (e.g., active microbial keratitis) and similarly in inflamed graft beds triggered postoperatively (i.e., in the case of suture abscess or recurrent herpes simplex virus infection) also has a substantially increased risk for graft rejection. In addition, previous ipsilateral rejection is an important risk factor, which is thought to be due to corneal inflammation or presensitization to donor tissue via previous engraftment.

Other noteworthy risk factors involve poor ocular surface function, usually found in conjunction with inflammation and corneal neovascularization. Hence, ocular surface diseases (e.g., severe dry eye, ocular pemphigoid, Stevens–Johnson syndrome, and neuroparalytic disease) or injury (e.g., severe chemical or radiation burns) is associated with poor prognosis for corneal transplantation.

Differential diagnosis

Several clinical situations exist which make the diagnosis of corneal allograft rejection difficult; the largest confounder is recurrent ocular inflammation in herpetic keratouveitis. The occurrence of corneal allograft rejection is very common in herpetic patients, and repeated bouts of inflammation carry a more dismal prognosis for the graft. Indicators for herpetic inflammation include the presence of typical dendriform epithelial lesions, unusually intensive anterior-chamber reaction, or endothelial keratic precipitates not confined to the graft.

Another significant confounder in diagnosing immune rejection includes graft endothelial decompensation, since this condition also leads to corneal graft edema. However, any questionable presence of edema in a corneal graft is nonetheless treated with corticosteroids as rejection.

Prevention and treatment of corneal allograft rejection

Postoperative prophylactic immunosuppressive regimens can be devised according to the degree of risk of rejection. In low-risk cases, low-frequency use of topical steroids is adequate initially, and tapered off to 6–12 months. High-risk corneal grafts require more intensive treatment. Topical and systemic corticosteroids in conjunction with topical and/or systemic ciclosporin are used for prevention of corneal rejection.

Corticosteroid therapy is also the treatment of choice for acute corneal immune rejection, and can be administrated topically, periocularly, and/or systemically. Most episodes of corneal graft rejection can be reversed if therapy is initiated early and aggressively; thus, it is imperative for the patient to identify and report any onset of symptoms associated with immune rejection (i.e., decreased vision, pain, and redness). In mild episodes of graft rejection topical steroids are preferred and can be applied as often as every 15 minutes to 2 hours. For severe episodes of rejection, such as those experienced by high-risk recipients, intensive steroid therapy administered via frequent topical eye drops, periocular injection, and/or systemically (oral or intravenous) may be given.


There is little information in this regard since human cornea grafts are not typically biopsied and clinical examination relies heavily on biomicroscopic or “slit-lamp” evaluation. Moreover, because rejection is treatable, donor tissue is only replaced once the graft has irrevocably failed, not before or at the time of a rejection episode. Hence, clinical use of pathology in corneal transplantation is not common practice.


Graft rejection is triggered by genetically nonidentical (allogeneic) donor peptides known as histocompatibility antigens, or “alloantigens.” Alloantigens which pose the greatest barrier to graft survival in transplantation en bloc are encoded by the major histocompatibility complex (MHC), also referred to as human leukocyte antigen (HLA), system in humans. Class I MHC antigens (or HLA-A, -B, and -C) are constitutively expressed by all nucleated cells and platelets, while class II MHC antigens (or HLA-DR, -DQ, and -DW) are constitutively expressed on leukocytes.

Unlike in other forms of solid-tissue/organ transplantation, histocompatibility-matching donor tissue to the intended recipient for promotion of graft survival is variably performed in the cornea. This is in part because the normal cornea expresses very low levels of HLA antigens. However, during inflammation and in graft rejection the expression levels of these antigens are strongly upregulated and can trigger immune rejection. The role of histocompatibility-matching has been studied extensively in the clinic and while CCTS reported no overall beneficial effect of this practice, a myriad of other independent studies have indicated the contrary by showing that histocompatibility-matching (particularly at HLA-A and HLA-B loci) does significantly reduce the risk of rejection. Moreover, in the murine model of corneal transplantation, it has been clearly demonstrated that MHC alloantigens per se trigger immune rejection, particularly in the high-risk setting.

Interestingly, unlike in other forms of solid-tissue/organ transplantation, minor histocompatibility antigens (minor H) have been shown to play a significant role in triggering immune rejection ( Box 8.5 ). These alloantigens are encoded throughout the genome and include ABO blood antigens and Lewis antigens in humans, and H3 antigens in mice. Studies conducted in mice have indicated that minor H alloantigens are a critical barrier to graft survival (particularly in the normal-risk setting). Moreover, it has also been reported that ABO-matching is a relatively feasible and inexpensive clinical practice which can be effective in reducing the risk of graft failure.

Aug 26, 2019 | Posted by in OPHTHALMOLOGY | Comments Off on Corneal graft rejection

Full access? Get Clinical Tree

Get Clinical Tree app for offline access