OBJECTIVE: Chronic endothelial cell loss of the graft is very common after penetrating keratoplasty. The aetiology of this is unknown. Clinically, non-identifiable immune reactions have been suspected. Recently, we were able to demonstrate that proper human leucocyte antigen (HLA) matching is a suitable means to reduce classical immune reactions in normal risk keratoplasty patients. In this study, we therefore investigated whether HLA-matched grafts also experience less chronic endothelial cell loss. METHODS: A homogenous group of 223 normal risk keratoplasty patients was divided into six groups with different degrees of HLA matching (group 1 with unknown HLA data, group 2 with up to two mismatches, group 3 with three mismatches, group 4 with four mismatches, group 5 with five mismatches and group 6 with six mismatches on the HLA A, B, DR loci). All serological HLA A, B, C and all moleculargenetic HLA DRB, DRQB typings of donors and recipients were performed in a single laboratory accredited by the American Society for Histocompatibility and Immunogenetics. Only patients with at least three postoperative endothelial cell density values were included in the study. The slopes of the regression lines for each individual scatterplot of endothelial cell density values plotted against postoperative time (linear regression, lost cells/mm2/day), and after logarithmic transformation (exponential regression, annual relative loss of cells) were evaluated, respectively. RESULTS: There were no statistically significant differences between the six groups. CONCLUSION: Whereas proper HLA matching at present standards is already a suitable means to reduce identifiable immune reactions and to prolong graft survival even in normal risk keratoplasty patients, the same HLA matching procedures are not effective in reducing the extent of chronic endothelial cell loss. For several reasons this does not yet exclude, however, the possibility that the underlying cause of chronic endothelial cell loss is immunological.
OBJECTIVE: Chronic endothelial cell loss of the graft is very common after penetrating keratoplasty. The aetiology of this is unknown. Clinically, non-identifiable immune reactions have been suspected. Recently, we were able to demonstrate that proper human leucocyte antigen (HLA) matching is a suitable means to reduce classical immune reactions in normal risk keratoplasty patients. In this study, we therefore investigated whether HLA-matched grafts also experience less chronic endothelial cell loss. METHODS: A homogenous group of 223 normal risk keratoplasty patients was divided into six groups with different degrees of HLA matching (group 1 with unknown HLA data, group 2 with up to two mismatches, group 3 with three mismatches, group 4 with four mismatches, group 5 with five mismatches and group 6 with six mismatches on the HLA A, B, DR loci). All serological HLA A, B, C and all moleculargenetic HLA DRB, DRQB typings of donors and recipients were performed in a single laboratory accredited by the American Society for Histocompatibility and Immunogenetics. Only patients with at least three postoperative endothelial cell density values were included in the study. The slopes of the regression lines for each individual scatterplot of endothelial cell density values plotted against postoperative time (linear regression, lost cells/mm2/day), and after logarithmic transformation (exponential regression, annual relative loss of cells) were evaluated, respectively. RESULTS: There were no statistically significant differences between the six groups. CONCLUSION: Whereas proper HLA matching at present standards is already a suitable means to reduce identifiable immune reactions and to prolong graft survival even in normal risk keratoplasty patients, the same HLA matching procedures are not effective in reducing the extent of chronic endothelial cell loss. For several reasons this does not yet exclude, however, the possibility that the underlying cause of chronic endothelial cell loss is immunological.