There is no optimal desensitization protocol for cadaveric renal transplant recipients who display high levels of donor-specific alloantibodies as defined by a positive T- or B-cell cytotoxic crossmatch. We used anti-CD20 monoclonal antibodies (Rituximab) to try to prevent antibody-mediated rejection in three crossmatch-positive renal transplants recipients (standard and sensitized techniques). The three patients received a first, second, or third cadaveric donor renal transplant. Patient one had an historical positive T- and B-cell cytotoxicity crossmatch: negative T- and B-cell cytotoxicity crossmatch at the day of transplantation. The panel reactive antibody (PRA) level was 100%. Patients 2 and 3 showed positive B-cell cytotoxicity crossmatches: historical and on the day of transplantation; PRA levels were 50% and 71%, respectively. All recipients were treated pretransplant with rituximab (375 mg/m(2)) and 4 days of intravenous immunoglobulin (0.5 g/kg body weight) per day posttransplant plus 5 days of thymoglobulin. Maintenance immunosuppression consisted of tacrolimus, mycophenolate mofetil, and prednisone. Antibody-mediated rejection occurred in all patients at day 6, day 10, or 8 months after renal transplantation. For patient 1, the rejection was not reversible and the graft was lost at day 15. Patient 2 had poor renal function with an MDRD estimate of glomerular filtration rate at 36 mL/min/1.73 m(2) at 18 months posttransplantation, and the graft of patient 3 was lost at 9 months posttransplantation due to resistant antibody-mediated rejection with thrombotic microangiopathy. In these three cases of crossmatch-positive patients, rituximab failed to prevent antibody-mediated rejection. Others studies will be needed to determine the place of rituximab in these patients.