Analysis of allograft biopsy specimens from long-term surviving patients with stable renal function: predictive value of long-term graft prognosis.

Chronic allograft dysfunction is multi-factorial, and histology of long-term renal allograft shows variable findings. It is important to characterize the pathological features of graft kidneys with normal function to understand the natural course of transplants, which in turn would contribute to elucidate the causes of chronic allograft nephropathy (CAN). To address this issue, we performed 'non-episode' biopsies on well-functioning renal allografts, and evaluated the correlation between clinical outcome and histopathological findings. Patients who underwent a non-episode biopsy had a serum creatinine concentration less than 2.0 mg/dL, urinary protein of less than 500 mg/day and a stable clinical course. In total, 90 such biopsies were performed. Mean follow-up period after biopsy was 29 ± 16 months. We evaluated the histopathological findings and clinical outcome on each finding. Moreover, we compared the findings in the patients on tacrolimus with those of patients taking cyclosporin. Twenty-three biopsy specimens were essentially normal. Graft dysfunction during the follow-up period was recognized more frequently in patients showing more than one pathological process than in those with isolated findings. Graft outcome was not associated with drug-induced nephropathy, but with acute rejection (P = 0.0193) and CAN (P = 0.0032). Patients found to have CAN-b had a worse outcome than those with CAN-a. CAN-b was less common in the tacrolimus group than in the cyclosporin group. Non-episode biopsy has a predictive value of the long-term outcome of a renal allograft. CAN is associated with graft dysfunction; neither is drug-induced nephropathy. Patients treated with tacrolimus had lower rates of CAN-b than did cyclosporin-treated subjects.