The current prospective investigation was conducted to determine whether development of IgG donor-specific lymphocytotoxins detected at the onset of acute rejections was predictive of a poor-prognosis acute rejection. Between January 1990 and August 1993, 206 kidney transplants were performed. Cadaver kidney recipients were managed with antilymphocyte globulin as induction therapy and all recipients (i.e., cadaver and living related donor kidneys) received triple immunosuppressive therapy, i.e., CsA, AZA, and prednisone. Rejections were treated with intravenous Solu-Medrol and OKT3. Presence of donor-specific IgG lymphocytotoxin was detected by using dithiothreitol-pretreated sera (obtained at onset of rejection) and frozen donor cells. In addition, percentage of panel reactive antibody was determined on this dithiothreitol-pretreated sera. Of the 82 patients with biopsy-proven acute rejections, 19 were found to have developed donor-specific IgG lymphocytotoxin and a marked increase in panel reactive antibody. One-year graft survival in this group was dismal (16%), despite OKT3 therapy. Over 90% of these patients lost their graft within 2 months of rejection diagnosis. In 63 recipients who had acute rejections without development of IgG anti-HLA antibody, 1-year graft survival was 72%. The majority of these patients lost their grafts from chronic rejection. No anti-HLA activity was found in patients who did not have rejection episodes. Based on this study, evidence indicates that assaying for IgG donor-specific antibody at time of rejection is a valuable tool for selecting a subset of patients with poor-prognosis acute rejections. Identifying this subset will become important as we enter an era of new immunosuppressive agents.