Estimating glomerular filtration rate from serum creatinine measurements: analytical issues and standardization programs

Reliable serum creatinine measurements in glomerular filtration rate estimation (eGFR) are critical to ongoing global public health efforts to increase the diagnosis and treatment of chronic kidney disease. It is accepted that use of serum creatinine concentration alone as a GFR marker is inadequate. International recommendations favour the reporting of creatinine-based eGFR using equation that was developed from the Modification of Diet in Renal Disease (MDRD) study, i.e. the "four-variable" MDRD equation that uses age, sex, race, and serum creatinine parameters. However, a limitation of this equation for general implementation in healthcare is related to the use of differently calibrated creatinine measurement procedures among laboratories. In particular, creatinine results which were used to generate the clinical basis for the eGFR MDRD equation were not traceable to high-order reference measurement procedures and reference materials. Consequently, the eGFR is very dependent on the accuracy of the creatinine method in use. The only way to achieve universal implementation of the eGFR prediction equation, with the associated clinical benefits for the patients, is, therefore, to promote worldwide standardization of methods to determine creatinine together with the introduction of a revised eGFR equation appropriate for use with standardized creatinine methods. Standardization of calibration does not, however, correct for analytical interferences of methods (non-specificity bias). Establishing calibration traceability to the creatinine reference system will align the average performance of methods to each other, but will not substitute for improvement of suboptimal routine methods. To account for the sensitivity of alkaline picrate-based methods to non-creatinine chromogens, some manufacturers have adjusted the calibration to minimize the pseudo-creatinine contribution of plasma proteins, producing results more closely aligned to the reference method (isotope dilution-mass spectrometry), but this strategy makes an assumption that the non-creatinine chromogen interference is a constant among samples, which is an oversimplification. Analytical non-specificity for substances found in individual patient samples can affect the accuracy of eGFR computed from serum creatinine values for any alkaline picrate method including the so-called "compensated" Jaffe methods. The use of assays that are more specific for serum creatinine determination, such as those based on enzymatic reactions, may provide more reliable eGFR values. Supporting the choice of more specific assays by clinical laboratories represents one of the main tasks of our profession in order to achieve the ultimate clinical goal, which is to routinely report an accurate eGFR in all the pertinent clinical situations.