Although conventional histology (CH) of needle core biopsies has been accepted as the gold standard for diagnosis of renal allograft rejection, this assumption has never been tested. Fine-needle aspiration cytology (FNAC) and monoclonal antibody (panleukocyte) staining of needle core biopsies (MABS) have been suggested to be superior to CH. A total of 50 patients received a cadaveric renal transplant followed by immunosuppression with triple therapy. Biopsies were taken routinely at days 7,14,21,28, and 90, with additional biopsies taken between these times if rejection was suspected (total biopsy sessions = 219). Specimens were taken for CH, FNAC, and MABS at each biopsy session, but only the result of one technique (previously randomly allocated) was communicated back to the clinical team, using a standardized grading system. Subsequently the presence or absence of rejection was determined by retrospective analysis of the clinical and biochemical course by 4 clinicians, without reference to the biopsy result. Graft survival was not significantly different in the three groups. The sensitivities for CH, FNAC, and MABS were 75%, 59%, and 77%, respectively, while the specificities were 87%, 96%, and 80%, respectively. Inadequate samples for analysis occurred frequently with the MABS technique--and, to a lesser extent, with CH--and both techniques tended to overdiagnose rejection. FNAC most often gave an answer but did miss clinically important rejection episodes. Needle-core biopsy processed for CH remains the most reliable biopsy technique for the diagnosis of rejection of renal allografts. FNAC is a useful technique for monitoring grafts with stable function or nonfunction. MABS does provide information equivalent to CH, but, in this study, had a high incidence of inadequate samples.