Which is the best glomerular filtration marker: Creatinine, cystatin C or both?

Thomas Stehlé,Pierre Delanaye

DOI: https://doi.org/10.1111/eci.14278

2024-07-03

European Journal of Clinical Investigation

Abstract:The non‐GFR determinants of creatinine (muscle mass, protein intake and tubular secretion) and of cystatin C (Dysthyroidism, systemic corticosteroids, and maybe others like diabetes, obesity, proteinuria, inflammatory syndrome) are the causes of inaccuracy in GFR estimation equations. Equations combining the two biomarkers provide a better estimate of GFR, particularly in the subgroup of patients whose estimates based on each of the biomarkers are highly discordant. These patients have also an increased risk of morbidity, particularly cardiovascular, and mortality. Background The glomerular filtration rate (GFR) is estimated by the serum or plasma concentration of creatinine and/or cystatin C using equations that include demographic data. The equations worldwide most widely used are those of the Chronic Kidney Disease Epidemiology Collaboration (CKD‐EPI) consortium and updated in 2021 to remove the Afro‐American racial correction factor. In 2021 and then in 2023, the European Kidney Function Consortium also developed equations based on creatinine and cystatin C, usable across the full age spectrum, and constructed by including the Q value (i.e. the median creatinine or cystatin C in healthy men and women, which is customizable for specific populations). Methods The aim of this narrative review is to examine the strengths and weaknesses of each biomarker. Results Both biomarkers have non‐GFR determinants, namely muscle mass, protein intake and tubular secretion for creatinine; dysthyroidism and systemic corticosteroids for cystatin C, as well as other more debated determinants (diabetes, obesity, proteinuria, inflammatory syndrome). These non‐GFR determinants are the reason why no equation based on a single endogenous biomarker has an accuracy within 30% greater than 90% over the entire age spectrum (in at least one patient in 10, estimated GFR is at least 30% higher or at least 30% lower than the measured GFR). Conclusion Equations combining the two biomarkers provide a better estimate of GFR, particularly in the subgroup of patients whose estimates based on each of the biomarkers are highly discordant. These patients must also be identified as being at increased risk of morbidity, particularly cardiovascular, and mortality.

medicine, general & internal, research & experimental