 |
 |
|
Home |
Current Issue |
Past Issues |
In the Clinic |
ACP Journal Club |
CME |
Collections |
Audio/Video |
Mobile |
Subscribe |
Tools |
Help |
ACP Online
|
Rapid Responses to:
|
|
Electronic letters published:
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif){kind=icon}
In Response
Application of the "Mayo Clinic" equation and MDRD on an elder population
Estimation of the Glomerular Filtration Rate could be Easier with the Inclusion of Pulse.Mass Index
ASSESSMENT OF THE PREDICTIVE PERFORMANCES OF THE “MAYO CLINIC” EQUATIONS
Comments about the new "Mayo Clinic" equation to estimate glomerular filtration rate
CREATININE AND GFR: AN IMPERFECT MARRIAGE OF CONVENIENCE
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Andrew D Rule, MD Mayo Clinic, Fernando G. Cosio
Send rapid response to journal:
rule.andrew{at}mayo.edu Andrew D Rule, et al.
|
To the Editor: We are gratified that this article has generated a healthy discussion. It was not our intent to initiate a debate about “whose GFR equation is best”. All of the available equations, including ours, have limitations. Instead, we believe we should concentrate our efforts on whether or not a generalizable equation can be developed based on serum creatinine (this may not be possible). We also believe that it is important to be cautious when interpreting the prevalence of a reduced estimated GFR in various populations. The discrepancy between our results and Dr. Froissart’s may relate to differences in methodology. In our study, we compared CKD patients with an estimated (not measured) GFR greater than or equal to 60 mL/min/1.73 m-superscript2 to healthy persons. Since the objective was to compare the accuracy of equations in predicting measured GFR, we did not identify or stratify the CKD and healthy samples by measured GFR. The Quadratic GFR equation (derived using two-thirds healthy persons and one-third CKD patients) should be tested in populations where the diagnosis of CKD is unknown, such as the general population. Even with careful attention to creatinine assay calibration, other investigators showed that at 60 mL/min/1.73 m2, the MDRD equation underestimated measured GFR by 30 mL/min/1.73 m2 among healthy persons.(1) We agree with Delanaye et al. that the relationship between serum creatinine and GFR differs among many populations and/or clinical presentations (i.e. good health versus CKD). In our equation, the 0.8 mg/dL cutoff for serum creatinine was determined as the peak of the parabola, otherwise a further decrease in creatinine would lead to a decrease in estimated GFR. The version of the Cockcroft-Gault equation that was referenced predicted GFR (mL/min/1.73 m2) among CKD patients.(2) Despite its limitations, if GFR was not indexed to body size in some way, smaller persons would be more likely to have a reduced GFR and thus CKD. A similar argument exists between cardiac output and cardiac index. The letter by Maaravi et al. nicely illustrated a point. An equation derived with CKD patients (MDRD equation), increased the prevalence of reduced GFR and weakened epidemiologic associations between GFR and risk factors. However, an equation derived with healthy persons and CKD patients (Quadratic equation), decreased the prevalence of a reduced GFR and strengthened epidemiologic associations. To better understand the epidemiology of early CKD, we need studies that measure GFR instead of estimate GFR in populations where the diagnosis of CKD is unknown. Andrew D Rule, MD rule.andrew@mayo.edu Fernando G Cosio, MD Mayo Clinic Rochester, MN, 55905 USA References 1. 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;16:459-66. 2. 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(6):461-70. Conflict of Interest:None declared |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Yoram Maaravi, MD Hadassah University Hospital, Mount Scopus, Jerusalem, Israel, Michael Bursztyn MD, Jochanan Stessman MD
Send rapid response to journal:
maaravi{at}md2.huji.ac.il Yoram Maaravi, et al.
|
Rule and colleagues (1) reported the testing of the MDRD study equation (2) across a wide range of GFR and found that it underestimated GFR by 29% in potential kidney donors. They concluded that the MDRD equation is inaccurate in healthy people and developed a new equation for estimating GFR when the diagnosis of chronic kidney disease is unknown. As suggested by Stevens and Levey in their editorial (3) we applied both these equations to estimate GFR on a community-dwelling elder population as part of the Jerusalem Longitudinal Study. 454 subjects, all aged 70 at study entry, underwent extensive social, clinical, physical, functional and laboratory examinations. Baseline serum creatinine was measured in all subjects as part of the laboratory profile. Survival was determined for 12 years following the initial cross-section. Using MDRD 323 subjects were classified as “healthy” with a mean GFR of 75.7 ml/min and 131 were classified as “renal failure” with a mean GFR of 49.8 ml/min. The mean GFR for the whole population was 68.2 ml/min. Using the new quadratic equation 400 subjects were classified as “healthy” with a mean GFR of 85.3 ml/min and only 54 were classified as “renal failure” with a mean GFR of 47.7 ml/min. The mean GFR for the whole population was 80.8 ml/min. We analyzed the relation of GFR derived by each formula to 12 year survival using Cox proportional hazard model and included the following independent variables at age 70: gender, independence in ADL, physical activity, self reported health, diabetes mellitus, hypertension, ischemic heart, cerebrovascular and malignant diseases, anemia, smoking, body mass index and serum cholesterol. Hazard ratio using MDRD was 1.51 (95% CI 1.04 -2.2, p=0.03) and using the quadratic equation was 2.01 (95% CI 1.25-3.23, p=0.004) In our population of community dwelling elders a significantly smaller number of subjects were classified with renal failure by the new quadratic equation yet it appears as a stronger predictor of mortality than the MDRD. The differences between these two formulas are substantial and may change the frequency of subjects evaluated and treated for renal failure. It is therefore important to further validate these formulas in populations of elders living in the community. References 1. Rule AD, Larson TS, Bergstralh EJ, Slezak JM, Jackobsen SJ, Cosio FG. Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med 2004;141:929-37. 2. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method t o estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461-70. 3. Stevens LA, Levey AS. Clinical implications of estimating equations for glomerular filtration rate. Ann Intern Med 2004;141:959-61. Conflict of Interest:None declared |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Enrique J. Sánchez-Delgado, M.D., Prof. Dr. med Hospital Metropolitano Vivian Pellas, Managua, Nicaragua
Send rapid response to journal:
esanchez{at}metropolitano.com.ni Enrique J. Sánchez-Delgado
|
Rule and colleagues improved the formulas for the estimated glomerular filtration rate (eGFR) using serum creatinine. The idea is to facilitate the daily clinical work, keeping the accuracy, both in normal persons (potential kidney donors) and in chronic renal disease. The reciprocal of serum creatinine (1/SCr) is frequently used as a simple but inaccurate eGFR. I propose to investigate the possible usefulness and accuracy of 1/SCr divided by the PULSExMASS INDEX or PMI (1). For example, for a normal SCr of 1.1 mg/dl and a normal PMI of 1.0, the eGFR would be 0.9 or 90% of normal. For a SCr of 1.5, it would be 0.66. If the PMI was 1.3, which is common in patients with a high global cardiovascular risk according to the Framingham Risk Equation, the eGFR would be 0.7 (70 % of normal) in the case of 1.1 mg SCr or 0.51 (less than 60 % of normal) for a SCr of 1.5 mg. The higher the PMI, the lower the expected eGFR for a giver value of SCr. Using one of the original examples from Rule, if a 50-year-old woman presented to donate a kidney and had a Mayo Clinic serum creatinine of 1.1 mg/dL , she would have an eGFR of 90 mL/min per 1.73 m2 (equation 3). If this woman had a BMI of 27 (69 kg, 1.6m), and a RHR of 80, her PMI would be 1.25 and her eGFR 0.73 (73% of normal). This eGFR is normal, but less than ideal, reflecting her higher cardiovascular risk. The PULSExMASS INDEX (PMI) is calculated as follows: Body Mass Index (BMI) multiplied by Resting Heart Rate (RHR) and divided by 1730 (24x72). The PMI considers the weigh in kilograms, the high in meters (BMI =Kg/m2) and the RHR. The normal values of BMI (average 24) are similar in males and females. The RHR (average 72) reflects the basal metabolic rate and related factors, both in healthy, fit, potential donors, and in sick people. The PMI reflects all these elements and correlates highly both with the body surface area, and the global cardiovascular risk (known to be elevated in renal patients), being much easier to calculate. If 1/SCr/PMI resulted acceptably accurate to estimate the GFR, it would facilitate the daily work with renal patients, until we know more from cystatin C. References 1.Gilbert Ross, Jeff Stier, Donald M Lloyd-Jones, Daniel Levy, Enrique Sánchez-Delgado, et al. Lifetime risk of developing coronary heart disease. Lancet 1999 (13 March); 353:924-925 Conflict of Interest:None declared |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Marc C Froissart, MD, PhD Department of Physiology and Biophysics, Georges Pompidou Hospital (AP-HP), INSERM U356 and IFR 58, Jerome Rossert, and Pascal Houillier
Send rapid response to journal:
marc.froissart{at}egp.aphp.fr Marc C Froissart, et al.
|
In the paper of Rule et
al, analysis of a series of 580 potential kidney donors referred to the Mayo
Clinic shows that the abbreviated MDRD equation significantly underestimated
glomerular filtration rate (GFR) in healthy persons (1). This led the authors to
develop new equations (refit MDRD equation for healthy persons, refit MDRD
equation with healthy indicator and quadratic equation) that performed better
than the abbreviated MDRD formula in healthy subjects. However, as pointed out
by Stevens and Levey in their Editorial Comment, proper evaluation of these
equations requires testing in an independent population (2). We used a series of 162
potential kidney donors and 882 patients with chronic kidney disease (CKD)
stage 1 or 2, whose GFR had been measured by renal clearance of 51Cr-EDTA,
to test the equations developed by Rule et al. The characteristics of our
population and the method used to measure GFR have been described in detail
elsewhere (3). Briefly, all subjects were Caucasian, 45% were female, mean
weight was 69.9 kg (SD 14.6) and mean body mass index was 24.8 kg/m2
(SD 4.8). Importantly, the serum creatinine assay that we used has been
calibrated to the laboratory where serum creatinine samples were measured in
the MDRD study (Dr Van Lente, Cleveland Clinic Foundation laboratory). As shown in the Table,
none of the three equations performed better than the abbreviated MDRD equation
in our series, and in healthy subjects the refit MDRD equation with healthy
indicator largely overestimated GFR. It is extremely unlikely that the
differences observed between our series and the one of Rule et al are due to
the use of different exogenous tracers to measure GFR. In a series of 111
subjects who had simultaneous measurements of inulin and 51Cr-EDTA
renal clearances, the
mean bias of EDTA renal clearance was 2.7 ± 3.5 mL/min/1.73m2 when compared to inulin (Froissart et al, unpublished data). In
contrast, part of the discrepancy may be explained by a different calibration
of the creatinine assay and by the fact that the subjects included in our
series tended to have lower body weight and body mass index. Thus, our data
suggest that the quadratic equation and the refit MDRD equations developed by
Rule et al should be more extensively tested before being used in clinical
practice outside the Mayo Clinic. References
Table: Accuracy and precision of the refit
MDRD equations and quadratic equation, in a series of potential kidney donors
and patients with CKD stage 1 or 2
* Measured and Estimated (eGFR) GFR ± SD [range]
given in mL/min/1.73m2. † Biases ± SE given in mL/min/1.73m2. ‡ P30% was percentage of estimated GFR within 30% of
measured GFR. Conflict of Interest:None declared |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Pierre Delanaye, M.D. Service de Dialyse. CHU Sart Tilman. Liège. Belgium., Jean Marie Krzesinski
Send rapid response to journal:
pierre_delanaye{at}yahoo.fr Pierre Delanaye, et al.
|
We read with great interest the article of Rule et al on the use of different creatinine-based equations to estimate glomerular filtration rate (GFR). We fully agree with the limitations of MDRD equation for good health population (1). Nevertheless, we would like to add some comments. Firstly, as told by Rule et al, GFR equation lacks representation of elderly patients and non-white racial groups. We think that neither MDRD nor new proposed “Mayo Clinic” equation will be adequate in specific types of populations including obese, anorexic, cirrhotic or paediatric patients (2). In these populations, the relationship between serum creatinine and GFR is not the same than in normal weight, adult, white, and in good health people. Specific equations must thus be developed in such situations as it has been done in black population (3). Secondly, using 0,8 mg/dL as creatinine value in the “Mayo Clinic” formula seems questionable. Such doing may occult phenomena as hyperfiltration frequently seen in obese and diabetic patients. Thirdly, we are not surprised about the inability of Cockcroft-Gault formula to estimate GFR. In fact, this formula does not estimate GFR but creatinine clearance and we know that the later is different from GFR because tubular creatinine secretion. Moreover, the mean creatinine clearance of the 236 patients included in the Cockcroft’s study is 72,7 ml/min. The assertion that Cockcroft-Gault equation has been “developed in chronic kidney disease samples” is thus not correct (4). Lastly, we think that using indexed GFR for BSA is also theoretically questionable and may cause bias, especially in obese patients (5). References 1. Rule AD, Larson TS, Bergstralh EJ, Slezak JM, Jacobsen SJ, Cosio FG. Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med. 2004;141:929-37. [PMID: 15611490] 2. Pierrat A, Gravier E, Saunders C, Caira MV, Ait-Djafer Z, Legras B, et al. Predicting GFR in children and adults: a comparison of the Cockcroft- Gault, Schwartz, and modification of diet in renal disease formulas. Kidney Int. 2003;64:1425-36. [PMID: 12969162] 3. Lewis J, Agodoa L, Cheek D, Greene T, Middleton J, O'Connor D, et al; African-American Study of Hypertension and Kidney Disease. Comparison of cross-sectional renal function measurements in African Americans with hypertensive nephrosclerosis and of primary formulas to estimate glomerular filtration rate. Am J Kidney Dis. 2001;38:744-53. [PMID: 11576877] 4. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16:31-41. [PMID: 1244564] 5. Turner ST, Reilly SL. Fallacy of indexing renal and systemic hemodynamic measurements for body surface area. Am J Physiol. 1995;268:R978-88. [PMID: 7733408] Conflict of Interest:None declared |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Charles J. Diskin, M.D.
Send rapid response to journal:
HNDT512{at}bellsouth.net Charles J. Diskin
|
It is not without some consternation and little amusement that I watch the evolution of the use of creatinine as a marker for GFR. Since it has always been a marriage of convenience at the expense of accuracy, I think that no one should be shocked that such an estimate of the clearance is not an accurate measure of GFR.1 When Homer Smith was searching for an accurate marker for GRF, he noted that such a marker should be:1) completely filtered at the glomerulus,2)free from synthesis or secretion in the tubules,3)free from reabsorption in the tubules, and 4)physiologically inert.2 His work demonstrated that inulin most closely met those goals, but it was inconvenient because it required a timed urine collection and the administration of an exogenous substance.2 Oddly enough, when Rehberg first proposed the use of creatinine, it was also as an exogenous administration,3 because there were many substances in the serum that also gave Jaffe’s reaction that were not really creatinine, one could not rely on endogenous creatinine. That problem became balanced by another one of Homer Smith’s exclusion criteria. While there is no tubular secretion of creatinine in goosefish, toadfish, dogfish and chickens, there is tubular secretion of creatinine in man. Since that 20% secretion often counterbalanced the overestimation in the serum by noncreatinine chromogens, a marriage of convenience was born, because the measurement of an endogenous creatinine clearance did not require the inconvenience of the administration of an exogenous substance. A half of a century later, micropuncture studies confirmed that the glomerulus was freely permeable to inulin whereas the tubule wall was impermeable unlike endogenous creatinine. We knew of course that creatinine was indeed an inadequate measure of GFR,4 yet we continued to use it because it was convenient and by the 1980’s we had had so much experience with it that we really knew more about human diseases and symptoms at any given creatinine clearance than at the true GFR. In reality, we do not use the creatinine clearance to estimate GFR, since we have so little experience with real inulin clearances, but we use it because we have become comfortable with creatinine and have a such a vast experience with creatinine clearances that we can better predict symptoms at a known creatinine clearance than we could if we had an inulin clearance. Similarly, when Cockcroft and Gault introduced their equation in 1976 we again chose convenience over the trouble of explaining to the patient how to really collect all urine samples in a 24 hour period; however, all equations are one more step away from the real clearance and are also subject to errors. As Drs. Stevens and Levy point out in their editorial,5 equations like Dr. Rule’s or their own MDRD equation are designed to fit well with mean measured GFRs, but practicing physicians, unlike administrators and epidemiologists do not deal with means but individuals. Therefore any change in production or secretion that results from a patient medication, time of day, physical activity or medical condition will be converted by Dr. Rule’s equation as well as Dr. Levey’s equation to indicate a change in GFR that is not really present. As a result both equations will interpret a patient with ketoacidosis6 or myeloma7 as suffering a loss in GFR and another with hyperbilirubinemia8 as demonstrating an improvement in function where there is in reality no such loss or improvement. Similarly, the introduction of trimethoprim,9 cimetidine9 or fenofibrate10 will similarly be interpreted by those equations as a loss in GFR rather than the mere changes in tubular secretion or creatinine production that they really are. What is more unfortunate are research studies that use either equation to determine an effect on renal function and show a gain in GFR by the use of dopamine or dobutamine in heart failure when it is really interference with the Jaffe reaction.11 Since many of the studies that have evaluated the use of acetylcysteine to prevent contrast nephrotoxicity have only collected serum creatinines to estimate clearance, that the “protective” effect may really represent an artificial tubular event. Similarly, since creatinine production has diurnal variations (creatinine clearance increases in hours of deepest sleep12 and serum creatinine is 11% higher in nonfasting subjects in afternoon hours13) only a time collection over twenty-four hours could account for such variations. An isolated serum creatinine might be drawn when production is at its peak or nadir and the error would be compounded when placed into any equation that extrapolated its production to be constant on a daily basis. Therefore the problem is not limited to one particular equation. The problem is and has always been with creatinine itself and our own search for convenience. Creatinine clearance is not and has never been synonymous with GFR and all of the regression analysis will not make it so because the serum creatinine depends upon many other factors than filtration. We should not be surprised that the more approximations that we make, the less accurate our data becomes. The problems come when we actually delude ourselves (and others) into thinking that these equations actually represent an actual GFR. 1. Rule AD, Larson TS, Bergstralh EJ, Slezak JM, Jacobsen SJ, and Cosio FG.Using Serum Creatinine to Estimate Glomerular Filtration Rate: Accuracy in Good Health and in Chronic Kidney Disease. Ann Intern Med 2004;141:929-937. 2. Smith H. The inulin clearance in The Kidney, H Smith ed.Oxford University Press, NY 1951,pp 47-55. 3. Rehberg PB. Studies on kidney function. The rate of filtration and reabsorption in the human kidney. Biochem J 1926;20:447-57. 4. Carrie BJ, Golbetz HV, Michaels AS and Myers BD. Creatinine: an inadequate filtration marker in glomerular disease. Am J Med 1980;69:177- 182. 5. Stevens LA, Levey AS. Clinical implications of estimating equations for glomerular filtration rate. Ann Inern Med 2004;69:959-961. 6. Molitch ME, Rodman E, Hirsch CA, Dubinsky E. Spurious serum creatinine elevations in ketoacidosis. Ann Intern Med 1980;93:280-1. 7. Hummel KM, von Ahsen N, Kuhn RB, Kaboth U, Grunewald RW, Oellerich M, Muller GA. Pseudohypercreatininemia due to positive interference in enzymatic creatinine measurements caused by monoclonal IgM in patients with Waldenstrom's macroglobulinemia.Nephron. 2000 Oct;86(2):188-9 8. Halstead AC, Nanji AA. Artifactual lowering of serum creatinine levels in the presence of hyperbilirubinemia. JAMA. 1984 Jan 6;251(1):38-9. 9. Urakami Y, Kimura N, Okuda M, Inui K. Creatinine transport by basolateral organic cation transporter hOCT2 in the human kidney.Pharm Res. 2004 Jun;21(6):976-81. 10. Hottelart C, El Esper N, Rose F, Achard JM, Fournier A.Fenofibrate increases creatininemia by increasing metabolic production of creatinine.Nephron. 2002;92(3):536-41. 11. Daly TM, Kempe KC, Scott MG. Bouncing Creatinine levels. N Engl J Med 1996;334:1749-1750. 12. Sirota JH, Baldwin DS, Villareal H. Diurnal variations in renal function of men. J Clin Invest 1950;29:187-90. 13. Pasternak A, Kuhlback B. Diurnal variations of serum and urine creatinine Scand J Clin Lab Invest 1975;27:1-9. Conflict of Interest:None declared |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
