The importance of glomerular filtration rate (GFR) estimating equations in the care of hospitalized patients

Abstract

In their letter, Lippi et al. [[1]Lippi G, Franchini M, Targher G. Detection of chronic kidney disease in hospitalized patients: is one estimating glomerular filtration rate equation better than another? Eur J Intern Medicine, doi: 10.1016/j.ejim.2010.08.002 (this issue).Google Scholar] correctly point at the importance of creatinine assay calibration and standardization of creatinine assay for decreasing the inter-laboratory variation and providing a more accurate and comparable estimation of GFR. They also address that there is increasing evidence that new estimating GFR equations (Mayo Clinic Quadratic (MCQ) and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations [2Rule A.D. Larson T.S. Bergstralh E.J. Slezak J.M. Jacobsen S.J. Cosio F.G. Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease.Ann Intern Med. 2004; 141: 929-937Crossref PubMed Scopus (895) Google Scholar, 3Levey A.S. Stevens L.A. Schmid C.H. Zhang Y.L. Castro A.F. Feldman H.I. et al.CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration). A new equation to estimate glomerular filtration rate.Ann Intern Med. 2009; 150: 604-612Crossref PubMed Scopus (0) Google Scholar].) are more accurate than the MDRD study formula, especially in a setting of elderly and hospitalized patients. In our study [[4]de Francisco A.L. Fernandez E. Cruz J.J. Casas M.T. Gómez-Gerique J. León A. et al.Under-recognized renal insufficiency in hospitalized patients: implications for care.Eur J Intern Med. 2010; 21: 327-332Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar] although the creatinine measurements were not standardized in the majority of patients and were obtained using different assays the methodology followed international recommendations. The Working Group of the National Laboratory Kidney Disease Education Program (NKDEP) has made some recommendations to clinical laboratories about the type of equation to estimate GFR [[5]Stevens L.A. Coresh J. Feldman H.I. Greene T. Lash J.P. Nelson R.G. et al.Evaluation of the modification of diet in renal disease study equation in a large diverse population.J Am Soc Nephrol. 2007; 18: 2749-2757Crossref PubMed Scopus (461) Google Scholar] Thus, those clinical laboratories that use methods with respect to IDMS traceability must use the equation developed from the revaluation of a new equation developed from the MDRD-4 equation: MDRD-IDMS; laboratories using methods without regard to traceability reference method must use the MDRD-4 equation. There are exogenous markers of glomerular filtration rate (GFR), such as inulin and iothalamate, but they are difficult to use properly and rarely employed outside of research. Estimations of GFR based on the serum creatinine are routinely used and the Modification of Diet in Renal Disease (MDRD) equation seems at present the best available one. This equation was developed using patients who had CKD identified by elevated serum creatinine levels. In a cross-sectional analysis of 5504 participants in 10 studies that included measurements of standardized serum creatinine and urinary clearance of iothalamate [[6]National Kidney Disease Education Program, Suggestions for Laboratories, December 2005: http://nkdep.nih.gov/resources/NKDEP_Suggestn4Labs_1205.pdf.Google Scholar], the MDRD Study equation provided unbiased and reasonably accurate estimates across a wide range of subgroups when eGFR is <60 ml/min per 1.73 m(2). We agree with Lippi et al. that in individual patients, interpretation of GFR estimates near 60 ml/min per 1.73 m(2) should be interpreted with caution to avoid misclassification of chronic kidney disease in the context of the clinical setting. In order to solve the problem the CKD Epidemiology Collaboration (CKD-EPI) equation was published in 2009 and intended to be more generalizable across various clinical settings than the MDRD equation. Weight, diabetes, and transplant were considered as potential variables, but the final equation uses the same variables as the MDRD equation [7Levey A.S. Stevens L.A. Schmid C.H. Zhang Y.L. Castro III, A.F. Feldman H.I. et al.A new equation to estimate glomerular filtration rate.Ann Intern Med. 2009; 150: 604-612Crossref PubMed Scopus (14838) Google Scholar, 8Stevens L.A. Schmid C.H. Zhang Y.L. Coresh J. Manzi J. Landis R. et al.Development and validation of GFR estimating equations using diabetes, transplant and weight.Nephrol Dial Transplant. 2009; 25: 449-457Crossref PubMed Scopus (99) Google Scholar]. The source studies that were used for the CKD-EPI equation can be broken down into two groups: High-risk populations such as patients with clinical CKD, are characterized by an average measured GFR (mGFR) <90 ml/min per 1.73 m(2), and low-risk populations such as potential kidney donors, are characterized by an average mGFR>90 ml/min per 1.73 m(2). The CKD-EPI equation was developed using a sample size of 8254, 71% (n 5858) of whom came from high-risk populations. The CKD-EPI equation was externally validated using a sample size of 3896, 72% (n 2810) of whom came from high-risk populations. We do agree with Lippi et al. that the CKD-EPI equation performed better in the external validation sample than did the MDRD equation. However the CKD- EPI equation authors recognized that “a single equation is unlikely to work equally well in all populations”. For instance, the CKD-EPI equation leads to a lower prevalence of eGFR <60 ml/min per 1.73 m2 in low-risk white women than the MDRD equation but when demographics in GFR-estimating equations start to model the CKD risk, this comes at the cost of less optimally modeling muscle mass [[9]Rule A.D. The CKD-EPI equation for estimating GFR from serum creatinine: real improvement or more of the same?.Clin J Am Soc Nephrol. 2010; 5: 951-953Crossref PubMed Scopus (30) Google Scholar] In fact to assess disease severity after a diagnosis of CKD has been made, one can choose between the CKD-EPI equation, which provides the most optimal results in the higher range of GRF but loses some accuracy from the inclusion of low-risk patients, or the MDRD equation, which loses some accuracy from the statistical methods that are used to model age. Michells et al. [[10]Michels W.M. Grootendorst D.C. Verduijn M. Elliott E.G. Dekker F.W. Krediet R.T. Performance of the Cockcroft–Gault, MDRD, and new CKD-EPI formulas in relation to GFR, age, and body size.Clin J Am Soc Nephrol. 2010; 5: 1003-1009Crossref PubMed Scopus (347) Google Scholar] compared the estimations of Cockcroft–Gault, Modification of Diet in Renal Disease (MDRD), and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations to a gold standard GFR measurement using (125)I-iothalamate, within the strata of GFR, gender, age, body weight, and body mass index (BMI). Overall (n=271, 44% male, mean measured GFR 72.6 ml/min per 1.73 m(2) [SD 30.4 ml/min per 1.73 m(2)]), mean bias was smallest for MDRD (P<0.01). The absolute bias of all formulas is influenced by age; CKD-EPI and MDRD are also influenced by GFR. Cockcroft–Gault is additionally influenced by body weight and BMI. We do agree with Lippi et al. that CKD-EPI gives the best estimation of GFR, although its accuracy is close to that of the MDRD. Cystatin C is a waste product of nucleated cells. It is freely filtered by the glomerulus, and subsequently is 100% reabsorbed by the renal tubules and degraded .It was long thought that the cystatin C concentration was independent of demographic variables, that is, unrelated to muscle mass, weight or disease states. Recent studies, however, have suggested that cystatin C may also not be an ideal kidney function marker. It has been shown that cystatin C concentration is dependent not only on kidney function, but also on age, race, sex, smoking and inflammation [[11]Gansevoort R.T. de Jong P.E. Challenges for the present CKD classification system.Curr Opin Nephrol Hypertens. 2010; 19: 308-314Crossref PubMed Scopus (24) Google Scholar]. The dependence of cystatin C on demographic variables seems, however, less than in the case with creatinine. A problem is that there are now several GFR estimation equations that are based on the serum cystatin C concentration. Which one of them performs the best has not been settled yet. It seemed not to perform substantially better than the MDRD equation in estimating true GFR. Introduction of cystatin C-based eGFR estimates is therefore promising for the future. Issues that need to be resolved first, however, are standardization of measurement, lowering of costs involved with measurement, and a consensus on which estimation equation should be used. Given the above we can conclude that measuring a reliable GFR in clinical practice is not that easy. Criticizing the MDRD estimation equation is easy, but offering better alternatives is not. As Lippi et al. agree, use of MDRD study equation to estimate renal function lower than 60 ml/min in routine clinical practice in hospitalized patients is easy to introduce, without associated costs and provides a more precise information than serum creatinine alone.