Following renal transplantation, immunosuppression is usually increased to treat presumed rejection episodes. However, a) many conditions mimic rejection in the post-transplant period, and b) many rejection episodes are irreversible. As increased immunosuppressive therapy is associated with an increased risk of infection, it would be ideal to limit antirejection therapy to only the rejection episodes that are reversible. The role of percutaneous allograft biopsy was studied as an aid to decide which patients to treat for rejection, to limit unnecessary immunosuppression and to predict allograft survival. One hundred thirty-five patients with suspected rejection underwent 206 allograft biopsies without complication. Two hundred four biopsies were available for study. Biopsies were coded on a 1-4 scale (minimal, mild, moderate, severe) for acute and chronic tubulointerstitial infiltrate and vascular rejection, as well as no rejection (e.g., recurrence of original disease). Treatment decisions were made on the basis of the biopsy combined with clinical data. All patients have been followed two years and outcome correlated with biopsy findings (death, nephrectomy, and return to dialysis defined as kidney loss). The results were the following: 1) biopsies represented changes within the kidney. Of 16 kidneys removed within one month of biopsy, no nephrectomy specimen showed less rejection than that seen on biopsy. 2) Eighty-one biopsies (39.7%) led to tapering or not increasing immunosuppression (either no rejection, minimal rejection, or irreversible changes). 3) Kidneys having either severe acute or chronic vascular rejection (<30% function at three months) had significantly (p < 0.05) decreased survival three to 24 months postbiopsy than those with minimal or mild vascular rejection or tubulointerstitial infiltrate (83% function at three months). 4) Kidneys with moderate chronic vascular rejection and those with severe acute tubulointerstitial infiltrate had significantly (p < 0.05) decreased survival at six to 24 months. 5) Kidneys with moderate chronic vascular rejection (MCV) without an acute infiltrate (ATI) had significantly better survival than those having both MCV and ATI. 6) Similarly, kidneys having severe ATI alone had better survival than those with ATI plus vascular rejection. It was concluded that a) percutaneous allograft biopsy can be done without graft loss or infection; b) biopsy represents changes throughout the kidney; c) biopsy aids in deciding when to treat for rejection and in deciding when to withhold increased immunosuppression, and d) allograft biopsy predicts the outcome of treatment of a rejection episode.