Study Design

The study was a retrospective, comparative case series and was designed from a consecutive series of keratoplasties performed in a single center. Inclusion criteria were the following: keratoplasty procedures performed by 1 surgeon (V.M.B.) between January 1995 and August 2009 for optical indication in eyes with corneal diseases not involving the corneal endothelium (ie, keratoconus, scars after infectious keratitis, stromal dystrophies, and trauma). For patients who received 2 grafts in the same or contralateral eye during the study period, only the first graft was included in the study. The study group (ALK group) included all ALK procedures that met the inclusion criteria (n = 149). These surgical procedures were performed between September 2002 and August 2009. The control group (PK group) was selected from PK procedures performed between January 1995 and August 2002. Two hundred seventy-four PKs that met the inclusion criteria were performed during this period. Among these 274 PKs, 149 cases matched for preoperative diagnosis and recipient rejection status were selected for the control group (PK group). Patients who underwent ALK between January 1995 and August 2009 were excluded from the control group so that a patient could contribute only for 1 graft to the study. PKs were selected according to the date of transplantation, with the most recent grafts being selected and the latest grafts being excluded. They were included until the same numbers of eyes with the various preoperative diagnoses and the same number of high-risk recipients were obtained in both groups. Donor corneas were stored at 31 C in an organ culture medium before surgery. All donor corneas were processed with the same techniques in a single eye bank during the study period. The composition of the culture medium used for corneal storage was not modified during that time. Grafts were not specifically selected to be ABO, Rhesus D, HLA, or HY compatible with the recipients except for high-risk recipients, who received ABO-compatible grafts. A graft was HY compatible if it corresponded to one of the following combinations: male donor with male recipient, female donor with female recipient, and female donor with male recipient. Conversely, male donor with female recipient resulted in HY mismatch. Surgical procedures have been described elsewhere. Data were recorded prospectively and then analyzed retrospectively.

Recipients and Transplantation Outcomes

High-risk recipients were defined as having a vascularized cornea (> 2 quadrants of corneal vascularization). Among 149 recipients included in each group, 25 (17%) were high-risk recipients. No eyes had limbal deficiency. Patients were hospitalized up to the time of graft reepithelialization. They were then examined prospectively at 1 and 2 weeks; 1, 3, 6, 9, 12, 18, 24, 30, and 36 months; and 3, 4, 5, 6, 7, 8, 10, and 15 years after surgery. The criteria for graft failure were irreversible graft stromal edema and corneal opacification. Pathologic events involving the graft, but not the recipient cornea, and progressing toward the center of the graft were considered as features of rejection. Epithelial rejection was defined as an epithelial rejection line appearing as a raised ridge of epithelium. Subepithelial rejection was defined as subepithelial infiltrates resembling those seen in epidemic keratoconjunctivitis. Stromal rejection was defined as stromal infiltrates, vascularization crossing the wound, or both involving the graft (or its interface with the recipient remaining posterior stroma) and progressing toward the center of the graft. Vascularization of the recipient peripheral cornea with no graft involvement was not considered as rejection, nor was an infectious suture abscess or superficial vascularization progressing along sutures. Endothelial rejection was diagnosed when at least 2 of the following features were present: Khodadoust line, keratic precipitates, stromal edema, ciliary injection, and aqueous cells. The diagnosis of rejection was made only if the transplant remained clear for an interval of at least 2 weeks after surgery. Irreversible rejection was defined as a rejection episode leading to graft failure.

After transplantation, all patients were treated with topical dexamethasone (1 mg/mL) and neomycin (3400 IU/mL). This treatment was tapered for several months without standardization of postoperative corticosteroid management. The initial corticosteroid regimen was 1 drop hourly in high-risk recipients and 1 drop every 6 hours in low-risk recipients. The average steroid regimen for low-risk recipients was 1 drop every 6 hours during the postoperative first month, 1 drop every 8 hours for the next 3 months, 1 drop every 12 hours for the next 3 months, and then 1 drop every day. For high-risk recipients, the average steroid regimen was as follows: 16 drops per 24 hours (ie, 1 drop every hour from 7:00 am to 10:00 pm) for 1 week, 8 drops per 24 hours for 1 week, 6 drops per 24 hours for 2 weeks, 5 drops per 24 hours for 2 weeks, 4 drops per 24 hours for 1 month, 3 drops per 24 hours for 3 months, 2 drops per 24 hours for 3 months, and then 1 drop per 24 hours. When the intraocular pressure increased to more than 21 mm Hg, dexamethasone was replaced by fluorometholone, rimexolone, or hydrocortisone. Corticosteroid use was never stopped in high-risk recipients, and it was discontinued when all the sutures were removed in low-risk recipients. In high-risk eyes, the objective was to find the lowest steroid regimen permitting absence of rejection episodes. Treatment of rejection episodes has been described elsewhere. Management of rejection episodes was similar in both surgical groups. It included topical dexamethasone every hour in all cases, dexamethasone subconjunctival injections every day in moderately severe cases, and intravenous methylprednisolone 500 mg per 24 hours for 3 days in severe cases.

Statistical Analyses

For quantitative variables, the unpaired t test was used to compare the ALK group with the PK group. Qualitative variables were analyzed using the chi-square test. Graft survival was analyzed using the Cox proportional hazard regression model. The association of variables with survival was tested by the log-rank test. Three-year graft survival was computed using the Kaplan-Meier method. The power of the study was calculated taking into account the actual follow-up of patients. It was defined as the probability of finding an 85% significant decrease in the risk of irreversible rejection after ALK as compared with PK. A multivariable Cox proportional hazard regression was carried out, including variables that were significant at a univariate level ( P < .05). The following variables were studied: donor age, recipient age, ABO compatibility, Rhesus D compatibility, HY compatibility, preoperative diagnosis, recipient rejection status, trephination size, and preoperative intraocular pressure. A P value < .05 was considered to be significant. Statistical analysis was performed using Statistica software version 6.1 (StatSoft France, Maisons-Alfort, France).