In 2000, nearly 47,000 corneal transplantations were performed in the United States, making penetrating keratoplasty one of the most common human tissue transplantation procedures. 1 Corneal grafting is one of the most successful allograft procedures performed, with success rates as high as 90% reported for initial grafts placed into nonvascularized corneal beds. 2–4 In part, this success is attributable to improved surgical technique, better donor tissue management, recognition of varied clinical manifestations of corneal graft rejection, and updated medical treatment. However, despite our enhanced understanding of corneal and ocular surface physiology, corneal graft rejection is still the leading cause of corneal transplantation failure. 5
The challenge for the corneal surgeon is to minimize allograft rejection by identifying high-risk patients preoperatively. It is important to educate patients to recognize symptoms of rejection and to seek treatment immediately. Once the patient is in the examining chair, it is imperative clinically to detect early and subtle signs of rejection and to implement prompt and aggressive therapy.
Risk Factors for Corneal Graft Rejection
The overall cumulative probability of having an episode of allograft rejection has been shown to be 21% at 10 years in a large-scale prospective study. 5 The majority of these rejection episodes occurred within the first 5 years after penetrating keratoplasty and, indeed, more than half took place within the first postoperative year. 5 Until we have more effective immunomodulatory medications at our disposal, prevention of allograft rejection remains the best way of preventing graft failure. Preventive measures should begin in the preoperative setting by careful selection of candidates with minimum risk factors. Several large-scale clinical studies have characterized the important risk factors for transplantation rejection and failure (Table 1). 3,5–9
It has been demonstrated that graft survival rates varied greatly depending on preoperative and postoperative vascularization of the host cornea. 3,6,7,9 Survival rates vary from 90% in grafts placed into avascular corneas to 50% to 75% in grafts transplanted into severely vascularized corneas. 2,10 The presence of stromal vascularization is considered more clinically relevant than the presence of superficial vessels because of the increasing risk of rejection associated with the severity of stromal vascularization. As a result of this observation, the presence of two or more quadrants of stromal vessels is considered to be a strong risk factor for allograft rejection. 8,11 Similarly, direct contact of the allograft with the host vascular system through the presence of peripheral anterior synechiae preoperatively or postoperatively is believed to increase the risk for graft failure, 6,9 increasing the risk by twofold in the presence of three or four quadrants of iris synechiae. 11
Another important risk factor for corneal graft rejection is the history of a previous graft failure, especially if the failure was a result of an allograft rejection. 7–9,11 One study demonstrated that the rate of graft failure secondary to allograft rejection increased from 8% in patients with no history of previous transplantation to 40% in patients with two or more previous grafts. 11 In addition, the history of previous anterior segment surgery 11 or the performance of concomitant vitrectomy at the time of the penetrating keratoplasty conferred a twofold increased risk of graft failure. 6,9
The age of the graft recipient as a risk factor for graft rejection has been controversial. Several clinical series have shown that a younger recipient conferred a higher risk of rejection, 12,13 but this was not corroborated by data from two transplant registries. 3,10 However, evidence from a large, prospective, statistically meticulous study showed that recipients younger than 40 years were at a higher risk for graft rejection. 11
Concurrent inflammation of the eye is associated with increased risk for corneal graft rejection. 6,14 It is imperative that active inflammation in the operative eye be controlled prior to penetrating keratoplasty, as transplantation into an inflamed eye is much more likely to result in allograft rejection and failure. 15
A history of glaucoma was shown by Aldave and coworkers 16 to result in earlier episodes of graft rejection and an overall higher risk of rejection in patients undergoing repeat penetrating keratoplasty. Others have shown that there is up to a threefold risk for graft failure when there was a prior history of glaucoma. 5–7,9,11 Certain preoperative diagnoses confer a higher risk for allograft rejection. Ing and colleagues 5 showed that there was more than a twofold difference in the probability of having a rejection episode in patients with a preoperative diagnosis of herpes simplex infection as compared to patients with preoperative diagnoses of keratoconus, Fuchs' dystrophy, or pseudophakic and aphakic bullous keratopathy. Other high-risk disease states include chemical burns. 11
Patient and Staff Education
Because early recognition of allograft rejection is the first step toward effective treatment, the importance of educating patients to be aware of the symptoms and signs of early transplant rejection cannot be overemphasized. Unless patients realize the consequence of delayed medical treatment in the presence of an allograft rejection, they may not pay attention to the early symptoms of rejection and may not seek medical attention until the graft failure has progressed. A helpful mnemonic by which patients can remember the signs and symptoms of graft rejection is RSVP:R for redness, S for sensitivity to light, V for visual symptoms, and P for pain. Kamp and associates 17 found that in a group of high-risk patients, nearly 70% of graft rejection episodes were preceded by patient symptoms and only 30% of allograft rejections were identified on routine clinical examination. However, the sensitivity and specificity of patient symptoms as an indicator of graft rejection was poor. 17 Despite this, patient symptoms cannot be ignored, and the symptomatic patient must be evaluated to rule out a rejection episode.
It is also important to educate the office staff regarding the importance of acknowledging corneal transplant patients' symptoms and relaying this information to the surgeon.
Early Clinical Features of Graft Rejection
As mentioned previously, early detection of graft rejection is the initial step toward effective therapy. This need for early recognition of an allograft rejection is equally important for both the physician and the patient. Once patients suspect they are experiencing these symptoms and seek medical attention, it is important that the physician be familiar with the signs of early graft rejection (Table 2).
Because an allograft rejection is mediated by an immune response, it demonstrates the typical characteristics of an inflammatory reaction, including cellular edema, vascular dilation and leakage, and cellular infiltration. In the eye, this typically results in conjunctival injection, perilimbal ciliary flush, anterior chamber flare and cells and, occasionally, elevated intraocular pressure. The presence of anterior chamber flare when there is no prior history of uveitis is highly suggestive of a rejection episode until proved otherwise. Acute epithelial defects can be a nonspecific sign of early rejection, especially in young children.
Subepithelial infiltrates are a result of cellular infiltration in the superficial donor cornea that spares the peripheral host cornea. They occur in 10% to 15% of rejections 18 and can occur independent of, or concurrent with epithelial or endothelial rejections. They consist of 0.2-to 0.5-mm hazy infiltrates that are suggestive of lesions seen in epidemic keratoconjunctivitis (Fig 1). Subepithelial infiltrates may be an early harbinger of more aggressive rejection; however, they respond well to treatment and typically heal without scarring.
Epithelial rejection is seen in 10% of graft rejection cases and usually occurs earlier in the postoperative period, within 1 to 13 months. 18 This type of rejection tends to take place earlier in the postoperative period as the epithelial rejection line appears when the recipient epithelium replaces donor epithelium (Fig 2). Once the recipient cells have replaced the donor epithelium, epithelial rejection does not tend to occur.
Isolated stromal rejection is rare but can be represented by stromal infiltrates and vessels. In severe cases of stromal rejection, stromal necrosis can occur. If the endothelium is not involved, corneal thickness may be normal.
The most common type of allograft rejection is manifested as endothelial rejection, occurring in 8% to 37% of cases. 18,19 Cellular infiltrates aggregate on the endothelium as keratic precipitates. They can appear as scattered deposits (see Fig 1) or can form a distinct line known as the Khodadoust line (Fig 3). This line tends to migrate from the peripheral cornea to the central cornea. Some would consider the Khodadoust line to be the hallmark of graft rejection; however, allograft rejection can definitely occur without the presence of this line. Endothelial rejection is often associated with stromal edema overlying the areas that have been traversed by the endothelial rejection line (Fig 4). An anterior chamber cellular reaction may accompany the keratic precipitates but is usually less severe than expected given the proportion of keratic precipitates present.
Treatment of Corneal Graft Rejection
Most episodes of allograft rejection can be reversed if prompt and aggressive treatment is initiated. Typically, the majority of graft rejection episodes are reversed and do not lead to graft failure. 9
Corticosteroid treatment is the therapy of choice in situations of acute corneal graft rejection. The routes of administration include topical, periocular, oral, and intravenous. Topical prednisolone acetate 1% was found to be preferred over subconjunctival or systemic corticosteroids by members of the Castroviejo Society. 20
Hill and colleagues 21 demonstrated in a prospective study that 500 mg intravenous methylprednisolone in a single dose was more effective and better tolerated than daily oral prednisolone (60–80 mg) when combined with topical steroids in corneal graft rejection. When patients were treated within 8 days of the onset of symptoms, the survival rate of grafts was 92% versus 55%. Hill's grave 22 also found that a second dose of methylprednisolone at 24 or 48 hours had no advantage over a single dose.
The Collaborative Corneal Transplantation Studies showed that a higher-frequency postoperative topical steroids regimen, closer patient follow-up, and more aggressive treatment of suspected or proven episodes of rejection precluded the need for histocompatibility matching. 8 This prospective double-blind study showed that matching for HLA-A, HLA-B, and HLA-DR antigens had no influence on the incidence of rejection episodes or on the overall rate of graft failure. It was found that in cases of mild rejection, topical prednisolone acetate 1% hourly and dexamethasone ointment at night was sufficient to reverse the rejection. In severe cases of rejection, topical prednisolone acetate 1% hourly, one dose of pulsed intravenous methylprednisolone (500 mg), and oral prednisone at 1 mg/kg/day for 5 days were recommended. 8
Hill 23 found that in high-risk keratoplasty, graft survival was greatly improved if systemic cyclosporine was used in addition to systemic and topical steroids (89%) as compared to use of topical steroids alone (10%). It was also shown that the maximum effect was obtained if the cyclosporine was used for 12 months instead of 6 months, demonstrating a 93% versus 69% graft survival rate, respectively. 24 Topical cyclosporine has been shown, in several case series, to be helpful for preventing rejection of high-risk grafts. 25 However, there is a need for larger-scale, controlled clinical studies before the efficacy of topical cyclosporine can be confirmed.
Despite improved donor tissue management, better understanding and earlier recognition of varied clinical manifestations of corneal graft rejection, and updated medical treatment, corneal graft rejection remains the major cause of allograft failure. Even with current aggressive medical management, prevention and early recognition of graft rejection are still the best therapeutic options to date. This process of prevention and early recognition should commence with the identification of high-risk clinical features preoperatively. It also includes frequent patient education on the importance of recognizing the symptoms of graft rejection and the need to seek prompt medical attention. The suspicion for allograft rejection should be maintained at a high level, and early and aggressive therapy should be initiated as soon as possible.
1. Statistical report, 2000. Washington, DC: Eye Bank Association of America, 2000
2. Volker-Dieben HJM, Kok-Alphen CC, Lansbergen Q, et al. Different influences on corneal graft survival in 539 transplants. Acta Ophthalmol 1982; 60: 190–202
3. Williams KA, Muehlberg SM, Wing SJ, et al. The Australian corneal graft registry, 1990–1992 report. Aust NZ J Ophthalmol 1993; 21 (suppl): 1–48
4. Report of the Organ Transplant Panel. Corneal Transplantation Council of Scientific Affairs. JAMA 1988;259:719–722
5. Ing JJ, Ing HH, Nelson LR, et al. Ten-year postoperative results of penetrating keratoplasty. Ophthalmology 1998; 105: 1855–1865
6. Boisjoly HM, Tourigny R, Bazin R, et al. Risk factors of corneal graft failure. Ophthalmology 1993; 100: 1728–1735
7. Yamagami S, Suzuki Y, Tsuru T. Risk factors for graft failure in penetrating keratoplasty. Acta Ophthalmol Scand 1996; 74: 584–588
8. The Collaborative Corneal Transplantation Studies Research Group. The collaborative corneal transplantation studies. Arch Ophthalmol 1992;110:1392–1403
9. Sit M, Weisbrod DJ, Naor J, et al. Corneal graft outcome study. Cornea 2001; 20: 129–133
10. Volker-Dieben HJM, D'Amaro J, Kok-Alphen CC, et al. Hierarchy of prognostic factors for corneal allograft survival. Aust NZJ Ophthalmol 1993; 111: 799–805
11. Maguire MG, Stark WJ, Gottsch JD, et al. Risk factors for corneal graft failure and rejection in the collaborative corneal transplantation studies. Collaborative corneal transplantation studies. Ophthalmology 1994; 10: 1536–1547
12. Boisjoly HM. Effect of factors unrelated to tissue matching on corneal transplant endothelial rejection. Am J Ophthalmol 1989; 107: 647–654
13. Musch DC, Meyer RF. Risk of endothelial rejection after bilateral penetrating keratoplasty. Ophthalmology 1989; 96: 1139–1143
14. Polack FM. The effect of ocular inflammation on corneal grafts. Am J Ophthalmol 1965; 60: 259–269
15. Williams KA. Factors predictive of corneal graft survival. Ophthalmology 1992; 99: 403–414
16. Aldave AJ, Rudd JC, Cohen EJ, et al. The role of glaucoma therapy in the need for repeat penetrating keratoplasty. Cornea 2000; 19: 772–776
17. Kamp MT, et al. Patient-reported symptoms associated with graft reactions in high-risk patients in the collaborative corneal transplantation studies. Cornea 1995; 14: 43–48
18. Alldredge OC, Krachmer JH. Clinical types of corneal transplant rejection. Their manifestations, frequency, preoperative correlates, and treatment. Arch Ophthalmol 1981; 99: 599–604
19. Arentsen JJ. Corneal transplant allograft rejection: possible predisposing factors. Trans Am Ophthalmol Soc 1983; 81: 361–402
20. Rinne JR, Stulting RD. Current practices in the prevention and treatment of corneal graft rejection. Cornea 1992; 11: 326–328
21. Hill JC, Maske R, Watson P. Corticosteroids in corneal graft rejection. Oral versus single pulse therapy. Ophthalmology 1991; 98: 329–333
22. Hill JC, Ivey A. Corticosteroids in corneal graft rejection. Double versus single pulse therapy. Cornea 1994; 13: 383–388
23. Hill JC. The use of cyclosporine in high-risk keratoplasty. Am J Ophthalmol 1989; 107: 506–510
24. Hill JC. Systemic cyclosporine in high-risk keratoplasty. Short versus long-term therapy. Ophthalmology 1994; 101: 128–133
25. Belin MW, Bouchard CS, Phillips TM. Update on topical cyclosporine A: background, immunology, and pharmacology. Cornea 1990; 9: 184–195