In many transplant centers there is a reluctance to perform percutaneous core needle biopsies in renal allografts for fear of complications that may jeopardize the graft. We have evaluated the safety of percutaneous renal allograft biopsy by retrospectively studying 1129 biopsy specimens in 513 patients between 1974 and 1988. All biopsies were performed with a conventional 2.0 mm TruCut disposable needle (Travenol Labs.; Deerfield, IL) without radiographic aid for localization of the kidney. Kidney tissue was obtained in 1095 cases (97.0%). In 1037 biopsies (91.9%) enough renal tissue for histological evaluation was obtained. In 34 biopsies (3.0%) no renal tissue and in 58 (5.1%) only renal medulla was found. All the complications were demonstrated by with macroscopic bleeding into the urinary tract system. Thirty-two patients (2.8%) developed hematuria requiring hospitalization and some type of active treatment (catheter-à-demeure, n = 14; cystoscopy, n = 11; percutaneous nephrostomy, n = 3; surgery, n = 4). On 8 biopsy occasions blood transfusion was required. Three graft removals (0.3%) were attributed to the procedure of biopsy for emergency diagnostic purposes. All three grafts were severely damaged by rejection and had little or no function. No grafts were lost among the biopsies taken for long-term follow-up. No deaths occurred. Biopsies yielding only renal medulla were found to carry a greater risk of bleeding than adequate biopsy specimens (P less than 0.001), as did biopsies from transplants with acute vascular rejection. Conversely, biopsies taken for routine check-ups of long-term renal allografts were associated with a lower risk than biopsies taken because of poor or deteriorating renal function (P less than 0.05). An analysis of 340 biopsies, taken in accordance with a protocol during periods of stable renal function, revealed no deterioration in graft function at 1 and 12 months after the biopsy. In this study, we have found that conventional percutaneous needle biopsy of the renal allograft involves a certain risk of complications, even including graft loss. We have also defined a number of risk factors for such complications. However, we think that the benefits outweigh the risks, and needle biopsy should therefore remain an important diagnostic tool among all the others in the posttransplantation management of the renal transplant recipient.