BACKGROUNDSerum creatinine (Scr) measurement plays a key role in glomerular filtration rate estimation (eGFR), chronic kidney disease (CKD) diagnosis as well as CKD treatment. However, the test results of Scr from different laboratories vary significantly. In order to get comparable results, the European in vitro diagnostic (IVD) directive requires traceability to reference methods and materials. The purpose of this study was to verify the effect of traceability implementation by investigating the trueness of creatinine measurement on nine homogenous systems in Beijing.

METHODSCommutable frozen human serum reference material, National Institute of Standards & Technology (NIST) Standard Reference Material (SRM) 967, was used to verify the trueness of Scr measurement results from nine homogeneous analytical systems of seven companies which are the most widely used systems in Beijing's third-grade hospitals. The methods referred to the Jaffe's and Enzymatic methods.

RESULTSfrom nine routine measurement systems were assessed using two criteria: biological variability and Clinical Laboratory Improvement Amendments' 88 (CLIA' 88). We simulated a series of broken lines representing the limits of SD and bias that would produce a relative increase (or decrease) of 10% and 20% in the measurement error when estimating GFR (MEeGFR) using the isotope dilution mass spectrometry (IDMS)-traceable Modification of Diet in Renal Disease (MDRD) Study equation.

RESULTSof the College of American Pathologists (CAP) 2008-B LN24 Survey were compared with our investigation results.

RESULTSCompared with the total error criteria of biological variability, Ortho (traceable to IDMS) met the minimum acceptable criteria; Roche (Jaffe), Roche (Enzymatic), Shino and Daiichi met the desirable criteria at level I. At level II, Ortho (traceable to gas chromatography/isotope dilution mass spectrometry, GC/IDMS), Dade Behring and Beckman (traceable to rate Jaffe) met the minimum acceptable criteria; Roche (Enzymatic) met the optimum criteria. The other five systems met the desirable criteria. Compared with the second criterion, all the results met the requirement of CLIA' 88. Trueness evaluation showed: the MEeGFR of Dade Behring exceeded 10% while the MEeGFRs of Beckman (traceable to rate Jaffe), Beckman (traceable to IDMS) and Ortho (traceable to Jaffe/High Performance Liquid Chromatography) exceeded 20% at level I. At level II the MEeGFRs of Dade Behring, Ortho (traceable to GC/IDMS) and Beckman (traceable to rate Jaffe) exceeded 10%. None of the nine systems got a MEeGFR higher than 20%. The conclusions of NIST SRM 967 agreed with those of LN 24 except for the Beckman measurement system.

CONCLUSIONSTrueness investigation of routine creatinine assays on nine homogeneous systems demonstrates an encouraging outcome that meets clinical requirements. Among the nine homogeneous routine systems, Roche and Daiichi produce the most accurate results. The implementation of traceability is effective.