Background. The beneficial effects of donor specific transfusion (DST) have become controversial in the cyclosporine era. This study was performed to evaluate the potential benefits of a new protocol for administering DSTs in the perioperative period. Methods. Non-HLA identical living donor kidney transplant recipients were randomized prospectively to control or to receive a DST 24 hr before transplant and 7-10 days posttransplant. All patients received similar immunosuppression according to protocol. Results. The protocol had 212 evaluable patients (115 transfused and 97 control). There were no differences in 1- and 2-year graft and patient survival, causes of graft failure, incidence and types of infection, repeat hospitalization, or the ability to withdraw steroids. Immunological hyporesponsiveness (by mixed lymphocyte culture) occurred more frequently in transfused patients (18%) than controls (3%) (P=0.04). Blood stored for ≥3 days was associated with fewer early rejections than blood stored ≤2 days. Overall, class II antigen mismatches were associated with more rejection episodes than class I antigen mismatches. However, transfused patients, but not control patients, with more class I antigen mismatches were more likely to have rejections. Conclusions. Administration of DSTs by the method described had no practical influence on patient or graft survival for up to 2 years. However, donor-specific hyporesponsiveness was more common in transfused patients (18 vs. 3%). Longer follow-up will be needed to determine whether DST will be associated with long-term benefit.