Percutaneous biopsy is a valuable investigation in the management of allograft rejection for all solid organs. Pancreas transplants have not been biopsed percutaneously, though open and percystoscopic biopsies have proved useful. We have compared percutaneous needle core biopsy with fine-needle aspiration cytology for the diagnosis of rejection in 18 patients receiving combined kidney and pancreas transplants and in one who was transplanted with the pancreas alone. Percutaneous needle core biopsy was successful in 37 of 40 attempts (93%), while fine-needle aspiration yielded diagnostic material on 33 of 47 attempts (70%). Transient hyperamylasemia occurred in 29%, returning to baseline in three days. One patient twice developed transient macroscopic hematuria. There was agreement between needle core biopsy and fine-needle aspiration on the diagnosis of rejection on six occasions and for the absence of rejection on 16. There was an 8% false-positive rate for fine-needle aspiration. In 13 instances of histologically proved renal rejection, concurrent pancreas biopsy revealed rejection in 69%. Pancreas rejection was not, however, seen in the absence of renal rejection. In this pilot study, percutaneous biopsy of the bladder-drained pancreas allograft was shown to be a practicable and valuable investigation without major complications.