Evolution of the renal function is a better predictor of long-term survival than serum creatinine

BACKGROUND: In recent years acute rejection has decreased to 10% to 20%. Therefore it is necessary to look for new endpoints in renal transplantation. Serum creatinine and changes in creatinine have been reported to be powerful predictors of long-term kidney transplant survival. Chronic renal allograft nephropathy is the primary cause of long-term graft failure but may appear at any stage in the evolution. METHODS: Data from 315 patients receiving cadaver donor renal transplants between February 1987 and March 2001 that functioned for 1 year were examined for the influence of demographic characteristics and transplant variables. Creatinine clearance was estimated using the Cockroft-Gault formula. Survival was assessed with the actuarial method. The multivariate analyses were performed using Cox proportional hazard models. RESULTS: The 10-year graft survival showed a relative risk of 2.5 in the univariate analysis when there was more than 10% decrease in renal function at 3 months compared with nadir values. When the decrease was more than 25% of creatinine clearance at the third month, during the evolution and serum creatinine at 3 months introduced in the multivariate model, the latter was not significant, while the other variables had a RR of 4.4 and 10, respectively. CONCLUSION: The evolution of renal function at 3 months and throughout the evolution were better predictors of graft failure than an isolated serum creatinine value.