Analytical Performance of 4 Automated Assays for Measurement of Cystatin C
L. J Sokoll
We evaluated the analytical performance of 4 cystatin C assays (Siemens N Latex on BNII, Roche Tina-quant on Cobas c501, Genzyme on Cobas c501, and Tosoh ST AIA-PACK on Tosoh AIA-600II) according to guidelines published by the Clinical and Laboratory Standards Institute.
We evaluated total imprecision, limit of detection, and limit of quantification for each assay using patient serum pools and linearity/recovery using serial dilutions of a patient serum pool with cystatin C–free serum. We compared patients (n = 102) using the Siemens assay as a comparison method.
All assays had limits of detection and quantification <0.08 and <0.39 mg/L, respectively. Total CVs were generally higher than the manufacturers' claims for all assays. The Roche assay overrecovered cystatin C, particularly at low concentrations (mean recovery 119%, 142% at 0.587 mg/L). Deming regression equations were y = 1.184x + 0.089, Sy|x = 0.246 for Genzyme; y = 0.937x + 0.231, Sy|x = 0.231 for Roche; and y = 1.010x + 0.216, Sy|x = 0.115 for Tosoh. The Genzyme assay appeared to report higher results than the Siemens assay, which is consistent with a higher reference interval specified by the manufacturer.
Although all assays were acceptable for clinical use, their diagnostic performances were not optimal. Limitations include imprecision greater than claimed, overrecovery for the Roche assay on low concentration samples, and differences in results for patient samples. The latter situation requires assay-specific cystatin C–based glomerular filtration rate prediction equations at least until calibration is standardized using the international cystatin C calibrator now being developed.