Purpose  

The purpose of this study is to determine the effectiveness of HLA typing in preventing corneal allograft rejection.

Methods  

This retrospective single-center study analyzed 459 consecutive HLA-typed patients who underwent perforating keratoplasty (PKP) between 1983 and 2001. Grafts were postoperatively transparent after donor-recipient selection by HLA-A, -B and -DR typing. Patients were divided into a low- and a high-risk group based on their preoperative diagnosis.

Results  

The estimated 1-, 5- and 10-year graft survival (Kaplan-Meier) was 93, 88 and 67% in low-risk patients and 73, 43 and 38% in high-risk patients. We found a significant correlation between the number of HLA mismatches and the rate of allograft rejections: a donor-recipient match of two or more alleles in HLA-A, -B or -DR reduces the rejection rate by at least 10% in low-risk (10 years after PKP; P<0.04) and 40% in high-risk patients (3 years after PKP; P<0.0001). Especially HLA-B mismatches are important prognostic factors for both low- (P<0.008) and high-risk patients (P<0.003). Considering both HLA-B and -DR mismatches significantly reduces the rate of allograft rejection, particularly in high-risk patients (P<0.0001). Matching on a split typing level offers no significant advantage over broad level matching.

Conclusion  

Clinical results confirm theories developed to explain the function of the HLA (MHC) receptor. The closest possible donor-recipient match of HLA antigens based on broad level typing significantly reduces the rate of allograft rejection and thus improves the prognosis for long-term transparency of corneal grafts in both high- and low-risk patients.