Plasma cystatin C is superior to 24-h creatinine clearance and plasma creatinine for estimation of glomerular filtration rate 3 months after kidney transplantation.

To the Editor: In a recent report (1 ), we provided data on the early follow-up of renal function by plasma cystatin C in transplanted adults. Cystatin C was found to be an alternative and more sensitive marker of acute changes in glomerular filtration rate (GFR) than plasma creatinine, especially during acute rejection episodes. We present here some additional data on the same group of patients 3 months after surgery (n 5 25; 5 patients moved to another institution). We measured GFR by the reference Crlabeled EDTA clearance performed over five 30-min periods starting 1 h after injection of 50–100 mCi of Crlabeled EDTA, with all individual results normalized for body surface area. Plasma cystatin C was determined by a latex particle-enhanced immunonephelometric assay (BN100; Dade-Behring) as described in detail previously (1). Plasma and urinary creatinine were measured by the Jaffé reaction on an Hitachi 747 analyzer (Boehringer); 24-h creatinine clearance was calculated from plasma and 24-h urinary creatinine. The accuracy of plasma markers to estimate GFR was calculated from the relative increases in their plasma concentrations vs their upper reference limits (creatinine, females, 100 mmol/L; males, 109 mmol/L; cystatin C, 0.94 mg/L) (1) and the relative decrease in Cr-labeled EDTA-measured GFR vs the lower limit of 80 mL z min z 1.73 m. This widely used threshold value of GFR (2) was preferred to ageand sex-matched reference values because of the relatively limited number of patients. Results are presented as the median and range after they were checked for non-gaussian distribution. Statistical analysis was performed with SigmaStat (Jandel Scientific), using linear regression analysis and the Kruskal–Wallis test; P ,0.05 was considered significant. Three months after surgery, GFR measured by the Cr-labeled EDTA clearance was 31–97 mL z min z 1.73 m (median, 49 z min z 1.73 m). Renal function was classified as severely impaired (,40 mL z min z 1.73 m; n 5 5), impaired (40 mL z min z 1.73 m , GFR , 50 mL z min z 1.73 m; n 5 8), moderately impaired (50 mL z min z 1.73 m , GFR , 80 mL z min z 1.73 m; n 5 11), or normal (.80 mL z min z 1.73 m; n 5 1). Results of 24-h creatinine clearance significantly correlated with those of Cr-labeled EDTA clearance (r 5 0.874; P ,0.0001) but overestimated GFR by ;40% (P ,0.0001), with six (24%) false-negative (misleadingly normal) results in patients in the 60–80 mL z min z 1.73 m range by EDTA clearance. The reciprocal of plasma creatinine significantly correlated with Cr-labeled EDTA clearance (r 5 0.784; P ,0.0001), but plasma creatinine overestimated GFR by 30% (P ,0.0001) with seven (28%) false negatives (48, 54, and .60 mL z min z 1.73 m; Fig. 1). The reciprocal of plasma cystatin significantly correlated with Cr-labeled EDTA clearance (r 5 0.879; P ,0.0001). GFR could be estimated from plasma cystatin C according to the following formula: GFR (mL z min z 1.73 m) 5 78 3 (1/cystatin C, in mg/L) 1 4. Overall, plasma cystatin C underestimated GFR by 14% (221% to 147%; P ,0.001) with no false negatives (Fig. 1). In our institution, GFR is measured 3 months after transplantation by the Cr-labeled EDTA clearance. This method is time-consuming and expensive, but it provides an accurate measurement of GFR at the beginning of the long-term follow-up period of the transplant recipient (.3 months post surgery). Despite their strong correlation with Cr-labeled EDTA clearance, we as others (2 ), confirm the poor abilities of 24-h creatinine clearance and plasma creatinine to evaluate GFR, a phenomenon that has been attributed to creatinine tubular secretion (2 ). In the present study, the use of creatinine to monitor GFR led to gross overestimation of GFR (30–40%). The diag-