Thirty-seven patients were submitted to kidney transplantation after transfusion at 2-week intervals with 4-week stored blood from their potential donors. All patients and donors were typed for HLA-A-B and DR antigens. The patients were also tested for cytotoxic antibodies against donor antigens before each transfusion. The percentage of panel reactive antibodies (PRA) was determined against a selected panel of 30 cell donors before and after the transfusions. The patients were immunosuppressed with azathioprine and prednisone. Rejection crises were treated with methylprednisolone. The control group consisted of 23 patients who received grafts from an unrelated donor but who did not receive donor-specific pretransplant blood transfusion. The incidence and reversibility of rejection episodes, allograft loss caused by rejection, and patient and graft survival rates were determined for both groups. Non-parametric methods (chi-square and Fisher tests) were used for statistical analysis, with the level of significance set at P<0.05. The incidence and reversibility of rejection crises during the first 60 post-transplant days did not differ significantly between groups. The actuarial graft and patient survival rates at five years were 56% and 77%, respectively, for the treated group and 39.8% and 57.5% for the control group. Graft loss due to rejection was significantly higher in the untreated group (P = 0.0026) which also required more intense immunosuppression (P = 0.0001). We conclude that tranfusions using stored blood have the immunosuppressive effect of fresh blood transfusions without the risk of provoking a widespread formation of antibodies. In addition, this method permits a reduction of the immunosuppressive drugs during the process without impairing the adequate functioning of the renal graft.