Previous experimental and human data suggests a detrimental effect on the course of acute renal failure related to exposure of blood to artificial dialysis membranes of poor biocompatibility. We performed a 2.5-year prospective randomized trial to compare the clinical course of acute renal failure (post-operative ischemic acute tubular necrosis, ATN) in patients receiving a cadaveric renal transplant requiring supportive hemodialysis in the immediate post-transplant setting. Patients were randomized to either a cuprophane or polymethylmethacrylate (PMMA) conventional hollow fiber dialyzer. All patients received a standard immunosuppressive regimen which included induction therapy with either horse anti-thymocyte gamma globulin (ATGAM) or the murine anti-CD3 monoclonal antibody (OKT3). Of 53 patients randomized, 17 were excluded (2 for intervening biopsy-proven rejection prior to recovery from ATN, 10 for primary graft nonfunction and 5 for other reasons), leaving 36 evaluable cases of uncomplicated ATN, 18 in each group. There was no difference by age, race, gender, cause of ESRD, immunosuppressive regimen, cold or warm ischemia time, use of pre-transplant dialysis, percent oliguria or the incidence of intra-dialytic hypotension between the 2 groups. There was no difference in the mean time to recovery from ATN posttransplant (8.9 days in the cuprophane group vs 9.5 days in the PMMA group, p = NS) or in the average number of hemodialysis treatments required (3.6 in both groups, p = NS). There was also no difference in long term allograft outcome in terms of the nadir serum creatinine, the number of episodes of subsequent acute rejection or in the development of chronic rejection. An intent-to-treat analysis of all 53 originally randomized patients similarly yielded no significant differences. A subsequent, non-randomized study using a membrane of intermediate biocompatibility (Hemophan) also showed no difference in recovery time from ATN. Bioincompatible membranes do not seem to have a significant clinical impact on the course of recovery of this form of acute renal failure. The striking benefits of biocompatibility in the course of ARF seen in other human trials may relate more to the non-renal systemic toxic effects of bioincompatibility.