Estimation of Renal Function in Patients With Chronic Kidney Disease

Transcription

1 JOURNAL OF MAGNETIC RESONANCE IMAGING 30: (2009) NSF Special Issue Estimation of Renal Function in Patients With Chronic Kidney Disease Jeroen P. Kooman, MD, PhD* The risk of nephrogenic systemic fibrosis after gadolinium exposure is inversely related to renal function. Various methods are available to assess the glomerular filtration rate (GFR). Prediction formulas based on serum creatinine, such as the abbreviated Modification of Diet in Renal Disease (MDRD) formula, are most commonly used and appear acceptable for clinical purposes in the majority of patients with chronic renal failure. However, especially in patients at the extremes of body composition, the results from creatinine-based equations should be interpreted with caution. In those patients, additional methods, such as timed urine collections, predictions based on cystatin, single-shot radiotracer methods, or, optimally, inulin clearance could be considered. In this review, the strengths and limitations of different methods to assess GFR are discussed. Apart from inulin clearance, no method can be considered the gold standard in the assessment of GFR. In cases of doubt, the decision to use gadolinium-enhanced magnetic resonance imaging should always be based on clinical risk benefit judgment. Key Words: renal function; MDRD; GFR J. Magn. Reson. Imaging 2009;30: VC 2009 Wiley-Liss, Inc. THE RISK OF NEPHROGENIC SYSTEMIC FIBROSIS (NSF) is strongly related to renal function. Therefore, an adequate estimation of renal function is of major importance for the clinician in estimating the risk of NSF and thus potential contraindications for magnetic resonance imaging (MRI). However, the risk of contrast nephropathy is also greatly dependent on the level of renal function. With the recent advent of new prediction formulas, the estimation of renal function has entered a new area. In the following, various methods to assess renal function, their usefulness for Department of Internal Medicine, Division of Nephrology, University Hospital Maastricht, Maastricht, The Netherlands. *Address reprint requests to: J.P.K., Department of Internal Medicine, Division of Nephrology, University Hospital Maastricht, PO Box 5800, 6202 AZ Maastricht, The Netherlands. jeroen.kooman@mumc.nl Received April 1, 2009; Accepted August 27, DOI /jmri Published online in Wiley InterScience ( clinical purposes, and possible drawbacks will be discussed. GLOMERULAR FILTRATION RATE AND CLASSIFICATION OF CHRONIC KIDNEY DISEASE The interest in the estimation of renal function has greatly increased since it was recognized that chronic kidney disease (CKD) is a strong risk factor for cardiovascular mortality, despite the fact that patients with earlier stages of CKD are generally asymptomatic. Renal function, although covering much more than the excretion of waste products alone, is generally expressed as the glomerular filtration rate (GFR). GFR in healthy subjects is considered to be in the range of ml/min/1.73 m 2. In general, CKD is assumed to be present below an GFR of 60 ml/min/ 1.73 m 2 (1), although GFR was shown to be lower than 60 ml/min/1.73 m 2 in a substantial percentage of otherwise healthy elderly patients with no signs of comorbidity (2). However, the great majority of cases of NSF occurred in patients with a GFR below 30 ml/ min/1.73 m 2 (CKD Stages 4 and 5). Thus, when identifying patients at risk for NCF, it is necessary to have a reliable estimate of GFR at this level of renal function. The prevalence of the different stages varies widely. In the PREVEND (Prevention of Renal and Vascular Endstage Disease) study in the Netherlands, in a population-based cohort, the prevalence of Stage 3 CKD (egfr ml/min/1.73 m 2 ) was 5.3%, and 0.04%, and <0.04% for Stages 4 and 5 CKD (egfr ml/min/1.73 m 2 and <15 ml/min, respectively) (3). These numbers are comparable to NHANES (National Health and Nutrition Examination Survey) data from the United States (4,5). As renal function declines with age, the percentage of patients with CKD progressively increases with advancing age: up to 42% in men and 44% in women of age 85 and higher, which mainly concerns subjects with Stage 3 CKD (6). These reports have initiated a discussion of whether age should be taken into account regarding the definition of CKD Stage 3. However, this discussion does not focus on the definition of Stages 4 and 5 CKD (5,7). There is not yet consensus in the nephrological field whether community screening for CKD (either by assessment of egfr, albuminuria, or both) would be VC 2009 Wiley-Liss, Inc. 1341

2 1342 Kooman warranted. However, there is certainly a strong point for screening of certain subgroups, such as patients with hypertension and diabetes mellitus. As for the risk of NSF, it is important to stress that the risk for CKD Stages 4 and 5 progressively increases with age: in patients seeking nonnephrological care the prevalence of Stage 4 CKD was 0.1% in patients of years of age, but increased to >1% in patients age 65 years or older (8). Hospitalization appears to be a risk factor for a decline in renal function: a significant increase in serum creatinine after admission was observed in 3.4% of all hospitalized patients (9). These changes in renal function may not be noticeable by clinical means (such as changes in urine output) and can often only be detected by laboratory testing. METHODS FOR DETERMINATION OF RENAL FUNCTION An ideal marker of GFR is freely filtered and not reabsorbed in the kidney, and has a constant plasma level. The inulin (a polysaccharide of 5200 Da) infusion method with concomitant urine collection is considered the gold standard method to measure GFR. However, this method is unsuitable for clinical practice due to the need for continuous infusion and multiple blood sampling, as well as cost limitations (10). Radiolabeled and nonradioactive tracers such as 99m Tc-DTPA, 51 Cr-EDTA, and 125 I-iothalamate require only a single injection and blood sample, and their use would appear to be more practical as compared to the inulin infusion method. However, results obtained by the inulin and single-injection tracer methods may differ by ml/min, with generally an overestimation by the latter (11,12), which appears to be partly related to the duration of subsequent plasma sampling after the tracer injection (13). However, the disagreement between single-injection methods and the classical inulin techniques appears to be higher in patients with normal renal function as compared to patients with renal insufficiency (14). Due to costs and practical limitations, neither inulin nor radiotracers are widely used in clinical practice and mainly serve as reference methods to validate other methods to assess renal function. However, a call for an increased use of single-injection radiotracer methods was made in a recent review (10). Still, on a routine basis, less costly and less complicated methods based on serum creatinine are by far the most commonly used. UREA AND CREATININE Urea (60 Da), which is quantitatively by far the largest breakdown product of protein metabolism, is not a reliable marker of GFR due to significant tubular reabsorption. Creatinine (113 Da) is a breakdown product of creatine and phosphocreatine, which are nitrogenous organic acids involved in the energy metabolism of skeletal muscle cells. Creatinine is freely filtered by the glomerulus, but also to a smaller degree (20% 30%) secreted by the proximal tubule. The latter compromises to some degree its use in the assessment of GFR. In addition, there also is some extrarenal elimination of creatinine. However, the overestimation of GFR which might arise from the additional tubular secretion is to some degree counterbalanced by the overestimation of plasma creatinine (up to 20%) by interference of noncreatinine chromogens by the Jaffe s kinetic rate colorimetric reaction, which is the most commonly used method for the assessment of creatinine. Indeed, creatinine measurements by the unmodified Jaffe method are in general significantly higher as compared to the reference method, gas chromatography-isotope dilution mass spectrometry (GC-IDMS), whereas results obtained by the modified Jaffe reaction are generally in better agreement with the reference method (15). Also, the generation rate of creatinine in the body may not always be constant, and may be influenced by factors such as dietary intake (1). The relation between GFR and serum creatinine levels is curvilinear, which compromises its sensitivity in detecting early CKD. Although increased creatinine levels are indicative of a decreased GFR, creatinine levels are profoundly influenced by the muscle mass of the patient. Thus, in patients with reduced muscle mass, eg, due to chronic illness, GFR may be significantly reduced despite (near) normal serum levels of creatinine. Therefore, serum creatinine levels alone cannot be used to estimate renal function. As an extreme example, a serum creatinine level of 132 lmol may correspond to a GFR range of ml/ min/1.73 m 2 (16). CREATININE CLEARANCE: TIMED URINE COLLECTIONS Clearance can be defined as the volume of plasma which is completely cleared of a substance per time unit (usually ml/min). This volume of plasma is not an actual, but a virtual volume. In contrast to mass removal, which is largely dependent on the concentration of the solute, the clearance is, under given circumstances, concentration-independent (16). In the assessment of renal function, the clearance of creatinine by timed (24-hour) urine collections has long been a mainstay in diagnostics. However, as mentioned previously, creatinine clearance may overestimate GFR due to the additional tubular secretion of creatinine, a mechanism which becomes relatively more important when renal function declines. Administration of drugs like cimetidine and trimethoprim block the tubular secretion of creatinine, but this approach is rarely used in clinical practice. In patients with Stage 5 CKD and especially those on dialysis, the mean of urea and creatinine clearance is advocated to assess GFR. Timed urine collections to assess creatinine clearance may also have a role in the estimation of GFR in less advanced renal failure. However, a potential drawback is that timed collections are often incomplete and the collection process may be cumbersome

3 Estimation of Renal Function 1343 for the patient. Therefore, they have been largely replaced by the prediction formulas discussed below (1,10,11,17). However, timed urine collections may still be of additional help in the assessment of renal function, especially in those with malnutrition or, the opposite, with increased body muscle mass, in which estimated GFR may respectively provide an over- and underestimation of true renal function (18,19). PREDICTION FORMULA BASED ON CREATININE Cockcroft Gault Equation Creatinine production is strongly related to muscle mass, and therefore, as mentioned previously, also the relation between creatinine and GFR. The Cockcroft Gault formula corrects for differences in body composition by including, next to creatinine, also age, body weight, and sex in the estimation. The Cockcroft Gault formula was validated in 249 male subjects with a wide age range, using an unmodified Jaffe method. In the original article the correlation coefficient between predicted and measured creatinine clearance was 0.83 (20). The formula for the Cockcroft Gault equation is: Creatinine clearance ðml=minþ ð140 age ½yrsŠÞ body weight ðkgþconstant ¼ Serum creatinine ðlmol=lþ constant : 1:23 ðif maleþ; or 1:04 ðif femaleþ ð140 age ½yrsŠÞ body weightðkgþ0:85 ðif femaleþ or as Serum creatinine ðmg=dlþ72 The Cockcroft formula can also be expressed per 1.73 m 2 body surface area (21). Based on a theoretical estimate of a 15% lower muscle mass in females, a correction factor for females is included in the formula (22). Due to the fact that the Cockcroft Gault formula was developed to assess creatinine clearance, it may significantly overestimate GFR, especially in patients with CKD Stages 4 and 5, but was also found to underestimate creatinine clearance by 24-hour collections, especially in elderly patients (23). Also in patients with abnormalities of body composition, the Cockcroft Gault formula may not be a reliable estimate of GFR because the increased weight may be largely dependent on an increase in fat mass. Whereas the use of the Cockcroft Gault equation in the assessment of renal function has been largely replaced by the modified Modification of Diet in Renal Disease (MDRD) equation, it is still often used in drug prescription tables (17), with continuing discussion as to which one of the two should be used in the guidance of dose adjustment (24,25). MDRD Equation This formula, unlike the Cockcroft Gault formula, gives an estimation of GFR, normalized to 1.73 m 2 body surface area. The formula was validated in 1628 patients with a mean GFR of 40 ml/min/1.73 m 2 participating in the MDRD study using the 125 I-iothalamate method as the gold standard. The original equation contained six parameters, including urea nitrogen and serum albumin, making it somewhat impractical for daily use. However, the modified MDRD equation, for which only sex, age, serum creatinine, and race are needed, appears to perform equally well, and is advocated as an estimate of GFR by current guidelines such as K/DOQI and KDIGO (11). Even though a surrogate of body composition such as body weight is not included in the equation, a good correlation was observed between the MDRD formula and the gold standard method. In most but not all studies, the MDRD formula appears to outperform the Cockcroft Gault method in the estimation of GFR, although in individual patients significant differences were observed between the gold standard method and MDRD equations. Indeed, in 20% of patients with CKD, estimated GFR fell outside the accuracy range of 30% of the actual GFR (10). Limitations of both the Cockcroft Gault and MDRD formulas in the classification of renal insufficiency were also noted in a recent study (26). Another point of concern with the use of the MDRD formula is the calibration of creatinine assays. The original MDRD equations were validated with the modified kinetic Jaffe method used to assess creatinine. Because this commonly used assay overestimates creatinine (as discussed previously), a different prediction equation has to be used when the (more expensive) enzymatic assay traceable to the reference method GC-IDMS is used to assess creatinine (11). Estimation of GFR by MDRD equations can easily be performed by the use of specific websites ( As an example, the original and IDMS-traceable MDRD formula (serum creatinine in lmol/l) are presented below: GFR ðml=min=1:73 m 2 Þ¼186 ðs cr =88:4Þ 1:154 ðageþ 0:203 ð0:742 if femaleþ ð1:212 if African-AmericanÞ½original equationš GFR ðml=min=1:73 m 2 Þ¼175 ðs cr Þ 1:154 ðageþ 0:203 ð0:742 if femaleþð1:212 if African AmericanÞ½IDMS traceableš Alternative equations have been derived for Asian populations (27). After the original MDRD validation study (28), which included only 6% diabetic patients, additional validation studies were performed in diabetic patients (29). The MDRD equation was validated in patients with renal disease and appears not to be reliable as an estimate of GFR at values above 60 ml/min/1.73 m 2 (29), although this would not compromise its use in identifying patients at risk for NSF. However, in patients with GFR at the lowest range, the MDRD equation may overestimate true GFR. This seems especially relevant for hospitalized patients with severely impaired renal function (mean GFR 18 ml/ min), in which the MDRD equations were found to overestimate egfr by an average of 6.8 ml/min/

4 1344 Kooman 1.73 m 2 (30), probably due to a reduced muscle mass in these patients. Moreover, the MDRD formula may also not yield reliable results in patients with significant abnormalities of body composition (17). However, in general, the modified MDRD equation has largely replaced other methods in the routine assessment of renal function, and may also be used as an initial method to identify patients at risk for NSF or contrast nephropathy. However, especially in ill and catabolic patients or those with extremes of body composition, results of the MDRD equation should be interpreted with caution. Attempts have been made to improve the accuracy of creatinine-based equations by including body weight in the equation. However, although the agreement with measured GFR appeared to be somewhat better compared to the original MDRD equation, substantial deviation from the reference method still occurred in individual patients (31). Despite the limitations of the MDRD equation, it should be noted that the risk assessment for NSF in the literature has been largely based on egfr assessment by the MDRD equation, with the great majority of cases occurring in patients with an egfr <30 ml/ min/1.73 m 2 (32,33). It is of great importance to stress that the MDRD formula, which assumes a stable serum creatinine level, is not reliable in patients with acute renal failure, which is an independent risk factor for NSF (34,35). In the case of acute renal failure serum levels of creatinine may increase by lmol/l in 1 day (16). POINT OF CARE TESTING Assessment of creatinine has to be performed using specialized laboratory techniques, which are often not available on the spot. However, in certain cases, such as high-risk patients scheduled for computed tomography (CT) or MRI in whom no recent assessment of kidney function is available, rapid bedside testing might be an asset for the radiologist. At present, this is not yet a standard procedure, but recent developments in point-of-care testing might lead to a wider availability of bedside testing methods for renal function assessment. Performing the test in whole blood is quicker as compared to testing in serum, for which additional centrifugation steps are necessary. In a recent study, a whole blood sensor, based on enzymatic conversion of creatine to hydrogen peroxide (which is oxidized at the amperometric electrode) was successfully validated against a conventional IDMStraceable creatinine method. Possibly, based on further studies, point-of-care testing for creatinine will become more widespread in clinical practice (36). CYSTATIN Cystatin (13,000 Da) is a member of the family of cysteine protease inhibitors, and an inhibitor of lysosomal proteinases. It is completely filtered by the glomeruli, after which it is reabsorbed and catabolized by the tubuli. It is generated by all nucleated cells, at a rate which is assumed to be constant. It is less influenced by age and muscle mass as compared to creatinine, although other factors such as thyroid dysfunction, inflammation, and the use of glucocorticoids may influence cystatin levels (1). As is the case for creatinine, the extrarenal elimination of cystatin increases at lower GFR. Cystatin levels in healthy subjects range from mg/l (15). It is well known that chronic kidney disease is a cardiovascular risk factor, and cystatin levels were shown to be better predictors of cardiovascular disease as compared to serum creatinine (37). Also, cystatin levels appear to be a better predictor for GFR as compared to serum creatinine levels. However, there is no consensus whether plasma cystatin or cystatinbased equations would improve the prediction of GFR as compared to creatinine-based equations in patients with CKD. Moreover, there is no uniformity between cystatin-based equations, which have not been validated in large cohorts such as the MDRD population. This may also be due to the fact that (nephelometric) assays of cystatin have not yet been standardized (18). Recently, a pooled analysis was performed based on cystatin-based equations validated in various studies. The most simple cystatin-based equation for estimated GFR from the refitted dataset was: egfr ¼ 76.7*CystatineC 1.19, although slightly better results may be obtained by including age, sex, and race in the equation (38). Cystatin may be more sensitive in the detection of early kidney damage as compared to serum creatinine, and the use of cystatin-based equations may be especially relevant in the assessment of GFR at levels higher than 60 ml/min/1.73 m 2, or in patients with lower levels of GFR and abnormalities in body composition (10,11,22). RENAL FUNCTION IN DIALYSIS PATIENTS Dialysis is generally started when GFR falls below 6 15 ml/min/1.73 m 2, also depending on the presence of uremic syndromes (39). Hemodialysis and peritoneal dialysis are by far the most commonly used techniques in the treatment of patients with endstage renal disease. The efficacy of dialysis treatment, especially with regard to the removal of small molecular weight uremic toxins such as urea, is a determinant of survival. However, it has been shown that next to dialysis efficacy, also the residual renal function of the patient is a strong predictor of survival, independent of dialysis dose (40). Thus, renal clearance and dialysis clearance may not be equivalent, which might be due to the fact that persistent renal clearance still plays an important role in water and sodium removal of the patient (overhydration is commonly present in dialysis patients), and because hemodialysis and peritoneal dialysis are inefficient in removing larger and proteinbound uremic toxins. With regard to all these factors, even relatively small amounts of residual function may be of direct relevance, and explain the strong relation between residual renal function and survival

5 Estimation of Renal Function 1345 in both peritoneal dialysis and hemodialysis patients. Thus, the preservation of even small amounts of residual renal function in dialysis patients is of major clinical importance (41). When estimating renal function in dialysis patients, serum creatinine levels or prediction formulas such as MDRD are unreliable because creatinine, a small molecule, is efficiently removed by dialysis techniques. Cystatin might be more useful for this purpose, because it is less well removed by dialysis techniques due to its higher molecular weight (42). However, as holds true for patients with CKD Stage 5 not on dialysis, the mean of urea and creatinine clearance by timed urine collections is the preferred approach to estimate renal function in dialysis patients (39,43), and is widely performed in dialysis centers throughout Europe (44). However, assessment of renal function in dialysis patients will generally not be used. CONCLUSION Although in individual patients, especially those with abnormal body composition, GFR estimated from prediction equations may deviate from measurements by the gold standard method, in the majority of patients an estimation of GFR by the abbreviated MDRD formula appears to be acceptable for clinical purposes. Nevertheless, clinicians should be aware of the inherent limitations of population-derived formulas for treatment decisions in individual patients, especially in those at the extremes of body composition, and ill and malnourished patients. Importantly, egfr equations should not be used in the setting of acute renal failure, which is an established risk factor for NSF. It is important to note that in the literature the risk assessment for NSF has largely been based on egfr measurements by the MDRD formula, the great majority of cases occurring in patients with egfr below 30 ml/min/1.73 m 2. In some patients, alternative methods, such as timed urine collections, predictions based on cystatin, or single-injection radiotracer methods could be considered as additional tools. However, apart from inulin clearance, none of these alternative methods can be considered a gold standard in the assessment of GFR, and with single-injection radiotracer methods or timed urine collection, GFR may be overestimated to some degree. Therefore, some doubt may remain in patients with borderline levels of renal function. Under these circumstances, the decision to use gadolinium-enhanced MRI should always be based on clinical risk benefit judgment. REFERENCES 1. Stevens LA, Coresh J, Greene T, Levey AS. Assessing kidney function measured and estimated glomerular filtration rate. N Engl J Med 2006;354: Wetzels JF, Kiemeney LA, Swinkels DW, Willems HL, den Heijer M. Age- and gender-specific reference values of estimated GFR in Caucasians: the Nijmegen Biomedical Study. Kidney Int 2007;72: de Zeeuw D, Hillege HL, de Jong PE. The kidney, a cardiovascular risk marker, and a new target for therapy. Kidney Int Suppl 2005;S25 S Coresh J, Astor BC, Greene T, Eknoyan G, Levey AS. Prevalence of chronic kidney disease and decreased kidney function in the adult US population: Third National Health and Nutrition Examination Survey. Am J Kidney Dis 2003;41: Glassock RJ, Winearls C. An epidemic of chronic kidney disease: fact or fiction? Nephrol Dial Transplant 2008;23: Wetzels JF, Kiemeney LA, Swinkels DW, Willems HL, den Heijer M. Age- and gender-specific reference values of estimated GFR in Caucasians: the Nijmegen Biomedical Study. Kidney Int 2007;72: Coresh J, Stevens LA, Levey AS. Chronic kidney disease is common: what do we do next? Nephrol Dial Transplant 2008;23: Risch L, Saely CH, Neyer U, et al. Prevalence of decreased glomerular filtration rate in patients seeking non-nephrological medical care an evaluation using IDMS-traceable creatinine based MDRD as well as Mayo Clinic quadratic equation estimates. Clin Chim Acta 2007;378: Smeets RE, Hoorntje SJ. incidence of impaired kidney function in 2050 patients admitted to a general hospital; a prospective study. Ned Tijdschr Geneeskd 1988;132: Prigent A. Monitoring renal function and limitations of renal function tests. Semin Nucl Med 2008;38: Herget-Rosenthal S, Bökenkamp A, Hofmann W. How to estimate GFR-serum creatinine, serum cystatin C or equations? Clin Biochem 2007;40: Rahn KH, Heidenreich S, Bruckner D. How to assess glomerular function and damage in humans. J Hypertens 1999;17: Agarwal R, Bills JE, Yigazu PM, et al. Assessment of iothalamate plasma clearance: duration of study affects quality of GFR. Clin J Am Soc Nephrol 2009;4: Perrone R, Steinman TI, Beck GJ, et al. Utility of radioisotopic filtration markers in chronic renal insufficiency: simultaneous comparison of 125 I-iothalamate, 169 Yb-DTPA, 99m Tc-DTPA, and inulin. The Modification of Diet in Renal Disease Study Group. Am J Kidney Dis 1990;16: Delanghe JR, Cobbaert C, Galteau MM, et al. Trueness verification of actual creatinine assays in the European market demonstrates a disappointing variability that needs substantial improvement. An international study in the framework of the EC4 creatinine standardization working group. Clin Chem Lab Med 2008;46: Feehally J, Floege J, Johnson RJ (eds.). Comprehensive clinical nephrology, 3rd ed. Amsterdam: Mosby Elsevier p Traynor J, Mactier R, Geddes CC, Fox JG. How to measure renal function in clinical practice. BMJ 2006;333: Manjunath G, Sarnak MJ, Levey AS. Estimating the glomerular filtration rate. Dos and dont s for assessing kidney function. Postgrad Med 2001;110: Toussaint ND, Agar JW, D Intini V. Calculating glomerular filtration rate in a young man with a large muscle mass. Med J Aust 2006;185: Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16: Rostoker G, Andrivet P, Pham I, Griuncelli M, Adnot S. A modified Cockcroft-Gault formula taking into account the body surface area gives a more accurate estimation of the glomerular filtration rate. J Nephrol 2007; Coresh J, Stevens LA. Kidney function estimating equations: where do we stand? Curr Opin Nephrol Hypertens 2006;15: Raj GV, Iasonos A, Herr H, Donat SM. Formulas calculating creatinine clearance are inadequate for determining eligibility for Cisplatin-based chemotherapy in bladder cancer. J Clin Oncol 2006;24: Spruill WJ, Wade WE, Cobb HH, 3rd. Comparison of estimated glomerular filtration rate with estimated creatinine clearance in the dosing of drugs requiring adjustments in elderly patients with declining renal function. Am J Geriatr Pharmacother 2008;6: Bookstaver PB, Johnson JW, McCoy TP, Stewart D, Williamson JC. Modification of Diet in Renal Disease and modified Cockcroft- Gault formulas in predicting aminoglycoside elimination. Ann Pharmacother 2008;42:

6 1346 Kooman 26. Botev R, Mallié JP, Couchoud C, et al. Estimating glomerular filtration rate: Cockcroft-Gault and Modification of Diet in Renal Disease formulas compared to renal inulin clearance. Clin J Am Soc Nephrol 2009;4: Imai E, Horio M, Nitta K, et al. Modification of the Modification of Diet in Renal Disease (MDRD) study equation for Japan. Am J Kidney Dis 2007;50: Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation: Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130: Poggio ED, Wang X, Greene T, Van Lente F, Hall PM. Performance of the modification of diet in renal disease and Cockcroft-Gault equations in the estimation of GFR in health and in chronic kidney disease. J Am Soc Nephrol 2005;16: Poggio ED, Nef PC, Wang X, et al. Performance of the Cockcroft- Gault and modification of diet in renal disease equations in estimating GFR in ill hospitalized patients. Am J Kidney Dis 2005; 46: Virga G, Gaspari F, Thomaseth K, Cara M, Mastrosimone S, Rossi V. A new equation for estimating renal function using age, body weight and serum creatinine. Nephron Clin Pract 2007;105: c Wertman R, Altun E, Martin DR, et al. Risk of nephrogenic systemic fibrosis: evaluation of gadolinium chelate contrast agents at four American universities. Radiology 2008;248: Perez-Rodriguez J, Lai S, Ehst BD, et al. Nephrogenic systemic fibrosis: incidence, associations, and effect of risk factor assessment report of 33 cases. Radiology 2009;250: Sadowski EA, Bennett LK, Chan MR, et al. Nephrogenic systemic fibrosis: risk factors and incidence estimation. Radiology 2007; 243: Saab G, Abu-Alfa A. Are patients with moderate renal failure at risk for developing nephrogenic systemic fibrosis? Radiology 2007;244: Skurup A, Kristensen T, Wennecke G; National Kidney Disease Education Program Laboratory Working Group. New creatinine sensor for point-of-care testing of creatinine meets the National Kidney Disease Education Program guidelines. Clin Chem Lab Med 2008;46: Shlipak MG, Sarnak MJ, Katz R, et al. Cystatin C and the risk of death and cardiovascular events among elderly persons. N Engl J Med 2005;352: Stevens LA, Coresh J, Schmid CH, et al. Estimating GFR using serum cystatin C alone and in combination with serum creatinine: a pooled analysis of 3,418 individuals with CKD. Am J Kidney Dis 2008;51: European Best Practice Guidelines on Hemodialysis. Measurement of renal function. Nephrol Dial Transplant 2001;17: Termorshuizen F, Dekker FW, van Manen JG, Korevaar JC, Boeschoten EW, Krediet RT; NECOSAD Study Group. Relative contribution of residual renal function and different measures of adequacy to survival in hemodialysis patients: an analysis of the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD)-2. J Am Soc Nephrol 2004;15: Marrón B, Remón C, Pérez-Fontán M, Quirós P, Ortíz A. Benefits of preserving residual renal function in peritoneal dialysis. Kidney Int Suppl 2008:S Hoek FJ, Korevaar JC, Dekker FW, Boeschoten EW, Krediet RT. Estimation of residual glomerular filtration rate in dialysis patients from the plasma cystatin C level. Nephrol Dial Transplant 2007;22: Korevaar JC, Jansen MA, Dekker FW, Boeschoten EW, Krediet RT. Estimation of residual glomerular filtration rate and renal Kt/Vurea from creatinine clearance in end-stage renal disease patients. The Netherlands Cooperative Study on the Adequacy of Dialysis. Adv Perit Dial 1999;15: Couchoud C, Jager KJ, Tomson C, et al., on behalf of the QUEST working group on dialysis adequacy. Assessment of urea removal in haemodialysis and the impact of the European best practice guidelines. Nephrol Dial Transplant 2009;24: