Analytical study of three cystatin C assays and their impact on cystatin C-based GFR-prediction equations.

Background: Cystatin C-based equations are used to estimate GFR. However, three cystatin C immunoassays are on the market. Difference in cystatin C assays could have strong consequences on the accuracy and precision of cystatin C-based equations. We have performed an analytical study of these three assays and studied potential differences between assays on the precision of cystatin C-based equations.
Methods: We have studied imprecision, recovery, linearity and interferences of the three immunoassays (nephelometric assay from Siemens and turbidimetric assays from Dako and Gentian). The impact of differences in cystatin C assays has been studied for the equations published by Levey (Siemens assay) and Grubb (Dako assay).
Results: Analytical performance of the Dako assay is slightly less high. For cystatin C values below 2.5 mg/L, no statistical difference is found between results given by the Dako and the Gentian assays. So, both assays can be used in the Grubb equation. Cystatin C results are different with the Siemens assay. The Levey equation, built with the Siemens assay, can only be used with cystatin C values measured with this assay. Using the Dako or Gentian assay results in the Levey equation can lead to differences in estimating GFR up to 6 mL/min/1.73 m2. Differences can reach 9.5 mL/min/1.73 m2 if the Siemens assay is used in the Grubb equation.
Conclusion: The Siemens and Gentian assays seem analytically more valid than the Dako assay for cystatin C determination. Differences in cystatin C assays can lead to significant differences in cystatin C-based equations. However, these differences seem less important than the differences observed with creatinine and creatinine-based equations.