Cystatin C for glomerular filtration rate estimation: coming of age.

Assessment of glomerular filtration rate (GFR) is essential for clinical practice. GFR is difficult to measure [measured GFR (mGFR)]; instead it is estimated [estimated GFR (eGFR)] by use of the serum concentration of endogenous filtration markers. However, all endogenous filtration markers are affected by factors other than GFR (non-GFR determinants), including generation, renal tubular reabsorption and secretion, and extrarenal elimination. Creatinine, a 113-Da breakdown product of muscle metabolism, was first identified in 1847 and proposed as a filtration marker in 1926 (1, 2 ). Creatinine-based eGFR (eGFRcr) is computed from serum creatinine concentration in combination with age, sex, and race as surrogates for creatinine generation by muscle. Cystatin C, a 13 300-Da serum protein produced by virtually all nucleated cells, was first identified in 1979 and proposed as a filtration marker in 1985 (3–5 ). Cystatin C is less influenced by muscle mass than creatinine and has often been proposed to be more accurate than creatinine for estimation of measured GFR in subgroups of the population, including vegetarians and those with muscle wasting, chronic disease, or limb amputation. The non-GFR determinants of cystatin C are not well known, and cystatin C– based eGFR (eGFRcys) is not more accurate for routine GFR estimation than eGFRcr. eGFR based on the combination of creatinine and cystatin C (eGFRcr-cys) is more accurate than either alone, reflecting the lesser influence of non-GFR determinants of either marker when both are used (6 ). eGFRcys is also more accurate in predicting prognosis than either eGFRcr-cys or eGFRcr, possibly reflecting opposite influences of non-GFR determinants of the markers (for example, muscle wasting leading to lower serum creatinine and inflammation leading to higher serum cystatin C) (7). Standardized reference materials are available for both creatinine and cystatin C, and estimating equations have been developed for use with standardized creatinine and cystatin C by the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) (6, 8 – 11 ). Current clinical practice guidelines recommend measuring serum creatinine and reporting eGFRcr by use of the CKD-EPI 2009 equation as the initial test and measuring serum cystatin C and reporting eGFRcys and eGFRcr-cys by use of the CKD-EPI 2012 equations as a confirmatory test (12 ). Use of a single equation for each filtration marker (or combination) facilitates widespread implementation of eGFR reporting, but other GFR-estimating equations are recommended if they are more accurate than the CKD-EPI equations. Creatinine is routinely measured in acute and chronic illness with eGFRcr reported by 90% of clinical laboratories in the US (13 ). Limitations to more widespread use of cystatin C include lack of agreement among commercial assays despite introduction of the standard, few evaluations of the CKD-EPI 2012 equations in large clinical populations, and incomplete understanding of the non-GFR determinants of cystatin C. In this issue of Clinical Chemistry, Grubb et al. (14 ) make significant progress in addressing the first 2 of these issues. Anders Grubb and colleagues are to be credited with landmark achievements in this field (3–5 ). More recently, Grubb led the Working Group for the Standardization of Cystatin C (WG-SCC), which was established and supported by the International Federation of Clinical Chemistry and the Institute for Reference Materials and Measurements and resulted in the development of the reference materials (9, 10 ). The current article has 2 aims, to improve agreement among commercially available assays and to develop a new GFR estimation equation that is independent of the clinical cystatin C measurement procedure used. For the first aim, the WG-SCC cooperated with 7 diagnostic companies to evaluate 6 assays (2 are identical). Linear regression was used to relate assigned values for the 6 assays in 800 –1000 plasma samples or 10 plasma pools. Correlations were excellent (R from 0.993 to 1.000), but there was heterogeneity in the comparisons, with slope coefficients ranging from 1 Division of Nephrology, Tufts Medical Center, Boston, MA; 2 Department of Nephrology, First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China; 3 Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, MN. * Address correspondence to this author at: Tufts Medical Center, 800 Washington St., Box 391, Boston, MA 02111. Fax 617-636-8329; e-mail alevey@ tuftsmedicalcenter.org. Received April 24, 2014; accepted April 30, 2014. Previously published online at DOI: 10.1373/clinchem.2014.225383 © 2014 American Association for Clinical Chemistry 4 Nonstandard abbreviations: GFR, glomerular filtration rate; mGFR, measured GFR; eGFR, estimated GFR; CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; WG-SCC, Working Group for the Standardization of Cystatin C; CAPA, Caucasian and Asian Pediatric and Adult subjects. Clinical Chemistry 60:7 916–919 (2014) Editorials