Repeated HLA mismatches and second renal graft survival in centers of the South-Eastern Organ Procurement Foundation.

To determine if repeated HLA mismatches and other putative risk factors were predictive of second graft failure in second grafts performed at Southeastern Organ Procurement Foundation (SEOPF) members centers, we identified a cohort of 753 retransplants in which one or more HLA antigens were mismatched in primary grafts. Of this group, 158 (21.1%) received second grafts with repeated mismatches of one or more HLA-A, B, or DR antigens that were previously mismatched in the primary graft (RMMs). All regrafts were cadaveric kidneys transplanted between 1982 and 1995. Multivariate analysis of 19 covariates in 438 regrafts identified four independent factors that were predictive of graft survival frequency in second transplants. Three of the four factors were associated with a reduced risk for graft loss in retransplants: cyclosporin A (CsA) use in graft (p = 0.0001, RR = 0.26), peak PRA < 50% (p = 0.008, RR = 0.52) and white donor race (p = 0.035, RR = 0.63). One factor was associated with an increased risk of second graft failure, namely, blood transfusion prior to the first graft (p = 0.026, RR = 5.14). None of the other 15 factors exerted significant additional risk to regraft survival frequency in these SEOPF data. In multivariate analysis, RMMs were not associated with altered graft survival frequency in regrafts (p = 0.944, RR = 0.99). We than used univariate analyses to determine whether RMMs had adverse effects on GS in particular subsets of recipients that were thought to be at increased risk for the second transplant failure. Univariate analyses were performed with methods that are sensitive to early events (Wilcoxon) and late events (log-rank). The variables tested were CsA use for the regraft, duration of primary graft function, panel reactive antibody levels (PRA), immunopathologic cause of first graft failure, and HLA mismatch of the second graft. These analyses indicated that repeated HLA mismatches were not an associated risk factor in any of these subgroups. These SEOPF data indicate that RMMs are not predictive of increased frequency of graft loss in cadaveric donor second transplants. We conclude that our results do not support a policy of routine avoidance of RMMs, which may result in increased waiting time for a second donor without providing an improved graft survival rate. The available literature suggests that HLA antibody identification, the use of sensitive flow cytometric and antiglobulin-augmented cross-match tests, together with appropriate donor selection, optimal immunosuppression and patient management may be sufficient to avoid the early loss of second grafts.