Rapid and accurate assessment of glomerular filtration rate in patients with renal transplants using serum cystatin C

Assessment of renal function in patients with renal transplants is of great importance. Various studies have reported cystatin C as an easily and rapidly assessable marker that can be used for accurate information on renal function impairment. To date, no study is available to define the role of cystatin C in patients with renal transplants.Thirty steady-state patients (50% male/50% female) with status post-kidney transplantation were studied. To assess renal function, cystatin C, creatinine clearance, serum creatinine, beta2-microglobulin (beta2M), and [125I]iothalamate clearance were determined. Correlations and non-parametric ROC curves for accuracy, using a cut-off glomerular filtration rate (GFR) of 60 ml/min, were obtained for the different markers allowing for calculations of positive predictive values (PPV), positive likelihood ratios (PLR), specificity and sensitivity, respectively. Further, to evaluate the usefulness of these markers for monitoring, intraindividual coefficients of variation (CVs) for cystatin C and creatinine measurements were compared in 85 renal transplant patients. Measurements consisted of at least six pairs of results, which were obtained at different time points during routine follow-up.Cystatin C correlated best with GFR (r=0.83), whereas serum creatinine (r=0.67), creatinine clearance (r=0.57) and beta2M (r=0.58) all had lower correlation coefficients. The diagnostic accuracy of cystatin C was significantly better than serum creatinine (P=0.025), but did not differ significantly from creatinine clearance (P=0.76) and beta2M (P=0.43). At a cut-off of 1.64 mg/l, cystatin C has a PPV of 93%, PLR of 6.4, specificity 89% and sensitivity 70%, respectively. For beta2M, PPV 83%, PLR 1.7, specificity 67% and sensitivity 75% was seen at a cut-off of 3.57 mg/l. Accordingly, at a cut-off of 125 micromol/l for serum creatinine, a PPV 76%, PLR 1.4, specificity 44% and sensitivity 80% was revealed. Finally, at a cut-off of 66 ml/min/1.73 m2 for creatinine clearance, the following characteristics were found: PPV 94%, PLR 7.7, specificity 89% and sensitivity 85%. The intraindividual variation of creatinine was significantly lower than that of cystatin C (P<0.001). With increasing concentrations, their ratios of CV tended towards a value of 1, demonstrating identical variability at low GFR.Together, our data show that in patients with renal transplants, cystatin C, in terms of PPV and PLR, has a similar diagnostic value as creatinine clearance. However, it is superior to serum determinations of creatinine and beta2M. The intraindividual variation of cystatin C is greater than that of creatinine. This might be due to the better ability of cystatin C to reflect temporary changes especially in mildly impaired GFR, most critical for early detection of rejection and other function impairment. Thus, cystatin C allows for rapid and accurate assessment of renal function (GFR) in renal transplants and is clearly superior to the commonly used serum creatinine.