In order to evaluate long-term renal graft function, 149 cyclosporin A and prednisolone (CyA/P)-treated renal transplant recipients were compared with 119 azathioprine and prednisolone (Aza/P)-treated patients. Only patients who had a functioning graft for at least 1 year and who were maintained on their initial immunosuppressive protocol were included. The minimum follow-up period was 4 years. Renal graft function was estimated by yearly determinations of serum creatinine and creatinine clearance. The CyA/P-treated patients had a significantly higher serum creatinine and a significantly lower creatinine clearance at every point in time post-transplantation than Aza/P-treated patients (P less than 0.001). The evolution of renal graft function, as reflected in the line of regression for serum creatinine and creatinine clearance versus time, was estimated for each individual patient. There was an almost stable renal function, as assessed by the median of the slopes of the regression line for serum creatinine versus time in both groups. The median increase in serum creatinine was only 1.4 mumol/l per year for Aza/P-treated patients and 2.4 mumol/l per year for CyA/P-treated patients (difference NS). The median decline in creatinine clearance was 2.18 ml/min per 1.73 m2/year in the Aza/P group and 1.07 ml/min per 1.73 m2/year in the CyA/P group (P = 0.05). In patients with a functioning graft for at least 5 years, creatinine clearance remained unchanged in both groups during the study period. In conclusion, renal graft function, as assessed by measurements of serum creatinine and creatinine clearance, remained essentially unchanged for at least 5 years after transplantation, regardless of the immunosuppressive protocol used.(ABSTRACT TRUNCATED AT 250 WORDS)