Importance of allograft biopsy in renal transplant recipients: correlation between clinical and histological diagnosis.

Renal allograft dysfunction after transplantation may be caused by acute rejection (AR), chronic rejection (CR), cyclosporine (CyA) or tacrolimus (FK) toxicity, and other causes such as recurrence of renal disease. Allograft biopsy is the "gold standard" to establish the correct diagnosis. However, many transplant centers routinely do not consider graft biopsy at the onset of renal dysfunction; instead, empirical steroid therapy or CyA dose reduction is the initial response to graft dysfunction. In this study, we prospectively predicted the histological findings prior to renal biopsy and correlated the clinical and histological diagnoses after the final report was issued by the pathologist. Patients with renal dysfunction after transplantation (increased serum creatinine >20% from baseline) were submitted to allograft biopsy. Three clinicians (C1, C2, and C3) involved in the care of these patients independently predicted the histological findings prior to the biopsy. A total of 100 cases (62 men, 38 women; 71 whites, 29 blacks) with a mean age of 41 years (21 to 70 years) were included in this study. Biopsy samples were taken after a mean period of 1.6 +/- 0.32 years (median, 0.25 years; range, 4 days to 17 years) after transplantation. Two patients with en bloc pediatric kidneys required postbiopsy blood transfusions for self-limiting bleeding; all other patients had no complications. All patients received azathioprine and prednisone; additionally, 74 received CyA and 19 FK. Final histopathologic diagnoses were AR (30), CyA/FK toxicity (36), AR plus CyA/FK toxicity (17), CR (11), recurrent disease (11), and other (6). In 28 cases (28%), the results of the biopsies showed more than one diagnosis. A completely correct diagnosis was predicted by C1, C2, and C3 in 47%, 42%, and 41% (mean, 43%) of the cases, incorrect diagnosis in 25%, 27%, and 25% (mean, 26%) of the cases, and partially correct diagnosis in 28%, 31%, and 34% (mean, 31%) of the cases, respectively. AR was confirmed histologically in 26 of 47 cases (55%) in the presence of therapeutic or high CyA/FK blood levels, whereas in 41 of 53 cases (77%), the histology showed CyA/FK toxicity in the presence of therapeutic or low CyA/FK blood levels. The mean serum creatinine at the time of the biopsy was 2.92 +/- 0.30 mg/dL, compared with the baseline of 1.76 +/- 0.10 mg/dL (P < 0.0001). After appropriate treatment, mean serum creatinine was 2.38 +/- 0.33 mg/dL (P < 0.0001). These data show that clinical prediction was poor, with totally correct diagnosis in only 43% of the cases. In 26%, the diagnosis was incorrect. We conclude that the renal biopsy is essential for establishing the correct diagnosis of renal allograft dysfunction and the appropriate management thereof.