Prediction of subclinical renal allograft rejection by vascular endothelial growth factor in serum and urine

BACKGROUND: Until now subclinical renal allograft rejection has only been recognized through a protocol biopsy. The aim of this study was to assess whether measurement of vascular endothelial growth factor (VEGF) in serum and urine could be adopted as a new noninvasive tool to predict subclinical rejection. METHODS: Concentration of VEGF in serum and urine was determined by ELISA in 132 recipients of a renal allograft with stable renal transplant function who were to undergo protocol biopsy and 80 healthy controls. A conventional receiver operating characteristic (ROC) curve was used to determine the sensitivities and specificities for patients with subclinical rejection. RESULTS: Levels of VEGF in serum (126.96 +/- 20.13 pg/mL; 95% confidence interval [95% CI], 83.10-170.83) and urine (16.14 +/- 4.09 ng/mmol creatinine; 95% CI, 7.21-25.06) of 13 patients with subclinical rejection significantly differed from those of 119 patients with no allograft rejection (No-AR) and health controls. The areas under the ROC curve were 0.771 (95% CI, 0.0.64-0.901) and 0.819 (95% CI, 0.662-0.976), respectively. Levels of VEGF in serum and urine after antirejection therapy (50.45 +/- 6.58 pg/mL and 2.60 +/- 0.83 ng/mmol creatinine, respectively) were lower than those at the time of protocol biopsy. No difference in urinary and serum VEGF expression was observed between cyclosporine and tacrolimus treatment. CONCLUSION: It is first reported that the monitoring of VEGF in serum and urine might be a new noninvasive approach to supplement a protocol biopsy for detection of subclinical rejection.