Enzymatic creatinine assays allow estimation of glomerular filtration rate in stages 1 and 2 chronic kidney disease using CKD-EPI equation

• We assessed impact of creatinine error on MDRD and CKD-EPI glomerular filtration rate.
• Compensated Jaffe assays led to greater errors than enzymatic ones.
• CKD-EPI equation is robust against creatinine error at higher glomerular filtration rate.
• CKD-EPI estimates could be expressed numerically up to 120 mL/min/1.73 m2.

The National Kidney Disease Education Program group demonstrated that MDRD equation is sensitive to creatinine measurement error, particularly at higher glomerular filtration rates. Thus, MDRD-based eGFR above 60 mL/min/1.73 m2 should not be reported numerically. However, little is known about the impact of analytical error on CKD-EPI-based estimates. This study aimed at assessing the impact of analytical characteristics (bias and imprecision) of 12 enzymatic and 4 compensated Jaffe previously characterized creatinine assays on MDRD and CKD-EPI eGFR.In a simulation study, the impact of analytical error was assessed on a hospital population of 24 084 patients. Ability using each assay to correctly classify patients according to chronic kidney disease (CKD) stages was evaluated.For eGFR between 60 and 90 mL/min/1.73 m2, both equations were sensitive to analytical error. Compensated Jaffe assays displayed high bias in this range and led to poorer sensitivity/specificity for classification according to CKD stages than enzymatic assays. As compared to MDRD equation, CKD-EPI equation decreases impact of analytical error in creatinine measurement above 90 mL/min/1.73 m2.Compensated Jaffe creatinine assays lead to important errors in eGFR and should be avoided. Accurate enzymatic assays allow estimation of eGFR until 90 mL/min/1.73 m2 with MDRD and 120 mL/min/1.73 m2 with CKD-EPI equation.