A novel equation to estimate glomerular filtration rate using beta-trace protein.
A number of equations, such as the Cockcroft Gault (3) and the Modification of Diet in Renal Disease (MDRD) (4) equations, have been developed in an attempt to improve GFR estimation from serum creatinine. These equations, however, have not proven to be accurate in RTRs (5-7) or in other patient populations that differ from the patient groups in which the equations were derived (8,9). The limitations of creatinine have led to the pursuit of alternate endogenous markers of GFR.
Beta-trace protein (BTP) has emerged as a promising novel marker of GFR (10) and has been shown to be a more sensitive marker of GFR than creatinine in a number of different patient groups (11-15). To date, serum concentrations of BTP have had limited clinical utility owing to the absence of a formula to convert the serum concentration of BTP into an estimate of GFR. The aim of this study was to develop a novel GFR estimation equation based on serum BTP concentration.
We recruited 163 stable RTRs from the Ottawa Hospital. Plasma clearance of radiolabeled [sup.99m]technetium-diethylenetriaminepentaacetic acid ([sup.99m]Tc-DTPA) was used to measure GFR (5), which was then corrected for standard body surface area (16). Serum samples were collected by venipuncture from each participant.
A modified Jaffe reaction was used to measure serum creatinine on a Beckman Coulter LX20 Pro Clinical System with the manufacturer's reagents (Beckman Coulter). CVs for serum creatinine were 4.9% at 55 [micro]mol/L (0.6 mg/dL), 1.7% at 150 [micro]mol/L (1.7 mg/dL), and 1.3% at 600 [micro]mol/L (6.8 mg/dL). Serum urea was measured using an enzymatic conductivity rate method on a Beckman Coulter LX20 Pro Clinical System with manufacturer's reagents. The CVs for serum urea were 3.5% at 5.2 mmol/L (14.6 mg/dL), 1.9% at 13.5 mmol/L (37.8 mg/dL), and 1.7% at 23.6 mmol/L (66.1 mg/dL). Albumin was measured by bromcresol purple dye-binding on a Beckman Coulter LX20 Pro Clinical System. The CVs for albumin were 2.0% at 24 g/L, 1.3% at 33 g/L, and 1.3% at 41 g/L. BTP was measured using a nephelometric assay on a BN Dade Behring ProSpec analyzer. The total analytical imprecision values (intraassay plus interassay; n = 41) of the assay calculated from 2 control materials with concentrations of 1.51 and 7.89 mg/L were 2.33% and 6.5%, respectively.
Stepwise multiple regression models were used to predict GFR. Variables considered for entry into the models were BTP, sex, race, and serum creatinine, urea, and albumin concentrations. BTP was forced into the models, and the remaining predictors were selected based upon a required P value of <0.001 for retention in the model. GFR and all predictor variables were log transformed to eliminate heteroscedasticity of residual error terms in the regression models. The GFR model results were expressed in original units [mL * [min.sup.-1] * [(1.73 [m.sup.2]).sup.-1]] to aid interpretation. An increase in [R.sup.2] of [greater than or equal to]0.02 was required to consider a more complicated model (9). Because measured GFR was adjusted for body surface area, height and weight were not included as predictor variables. To improve practicality arising from the unavailability of urea measurements in clinical practice, a similar model excluding urea was also derived.
The bias, precision, and accuracy of the equations and the abbreviated MDRD equation were calculated as recommended in the National Kidney Foundation guidelines on chronic kidney disease (CKD) (17). For the MDRD analysis, creatinine values were calibrated to the Cleveland Clinic guidelines as recommended (17) and as described elsewhere (5).
ROC analysis was performed to further characterize the diagnostic performance of the prediction equations. The relative increase of serum creatinine and BTP concentrations above the upper reference values [0.74 mg/L for BTP (15),88 [micro]mol/L in females and 106 [micro]mol/L for males for creatinine] in the 5 different CKD stages (17) was also performed. CKD stages 4 and 5 were combined, because there were so few patients in these stages.
The mean (SD) age of the cohort was 53 (12) years, 67% were male and 90% were white. The mean (SD) 99mTcDTPA GFR was 59 (23) mL * [min.sup.-1] * [(1.73 [m.sup.2]).sup.-1], the serum creatinine concentration was 148 (66) [micro]mol/L, and the serum BTP concentration was 1.26 (0.71) mg/L. All 5 stages of CKD were represented in the cohort. Further demographic and clinical characteristics can be found in Tables 1 and 2 of the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol53/issue11.
Of the various regression models shown in Table 1, a simple model incorporating BTP, urea, and sex (model 3) resulted in an excellent model fit ([R.sup.2] = 0.81). The addition of creatinine to the model (model 4) increased the [R.sup.2] by only 0.005. Similarly, the addition of albumin, age, and race increased the [R.sup.2] to a maximum of only 0.821. The model that included creatinine instead of urea had an [R.sup.2] value of 0.79 (data not shown).
Model 3 provided a balance between good model fit and simplicity. The exclusion of albumin and race as predictor variables did not substantially decrease the overall [R.sup.2]. Based on model 3 using BTP (in mg/L) and urea (in mmol/L), the estimated GFR [expressed in mL * [min.sup.-1] * [(1.73 [m.sup.2]).sup.-1]] would be calculated as follows:
Estimated GFR = 112.1 x [BTP.sup.-0.662] x [Urea.sup.-0.280] x (0.880 if the patient is female)
If urea was unavailable, the best fit model using creatinine (in [micro]mol/L) would be calculated as follows:
Estimated GFR = 167.8 x [BTP.sup.-0.758] x [Creatinine.sup.-0.204] x (0.871 if the patient is female)
Both BTP-based equations had a low bias [-0.9 and -1.2 mL * [min.sup.-1] * [(1.73 [m.sup.2]).sup.-1]], reasonable precision [SD of the residuals = 11.3 and 11.1 mL * [min.sup.-1] * [(1.73 [m.sup.2]).sup.-1]], and excellent accuracy, with 40% and 42% within 10% of the measured GFR, respectively (Supplemental Table 3). In comparison, the MDRD equation had a higher mean bias of -9.2 mL * [min.sup.-1] * [(1.73 [m.sup.2]).sup.-1], decreased precision of 12.2 mL * [min.sup.-1] * [(1.73 [m.sup.2]).sup.-1], and had only 25% of estimates within 10% of the measured GFR (Supplemental Table 3). For each equation, bias varied according to stage of CKD, with greater bias seen at higher GFRs (Supplemental Table 3). Overall, the BTP equations showed improved performance over the abbreviated MDRD equation at higher GFRs (Supplemental Table 3). ROC analysis showed that the areas under the curve for the 3 equations were not significantly different at GFR cutoffs of 30, 40, 50, and 60 mL * [min.sup.-1] * [(1.73 [m.sup.2]).sup.-1] (Supplemental Table 4). The serum concentration of BTP showed significantly higher relative increases above the upper reference value compared to serum creatinine at all levels of kidney function except CKD stage 1 (Supplemental Fig. 1).
BTP has been studied to a very limited extent in renal transplantation patients. Poge et al. (15) found similar diagnostic performances for creatinine and BTP by ROC analysis in 85 RTRs. Similar to our results, however, their BTP showed significantly higher proportional increases above the upper reference values than did serum creatinine for each of 6 GFR subgroups (15).
Another promising novel marker of GFR is serum cystatin C. A potential drawback of cystatin C, however, is that its serum concentration is increased by corticosteroids (15,18). In the studies to date, serum BTP concentration has not been influenced by corticosteroids (15, 19), and as such this marker may be preferred in settings in which steroid use is common such as renal transplantation. In the present study there was no significant association between steroid dose and BTP (r = 0.103, P = 0.15).
Strengths of the present study include the measurement of BTP, creatinine, and urea on the same day as the DTPA-GFR study and the inclusion of a wide spectrum of GFR values within the cohort. Several weaknesses should also be noted: (a) BTP was measured only once in each patient, and the within-person variability of BTP is unknown. (b) BTP concentrations may be increased in the cerebrospinal fluid of patients with various neurologic conditions (20). Whether the equations would be valid in patients with intracerebral pathology will need further study. (c) The effect of other pathologic conditions on the serum BTP has not been well established. (d) Calibration differences between laboratories may exist for BTP as they do for serum creatinine (2). (e) Measurement of BTP remains costly and not widely available, possibly limiting its practicality as a marker of GFR at this time. (fl The performance of the equations was assessed in the same group of patients in which the equations were derived and must be confirmed in an independent cohort of RTRs. Finally, the equations were derived in a cohort of RTRs and must to be validated in other patients groups. In conclusion, we have shown that BTP can be used in a simple equation to estimate GFR. This study was conducted in a cohort of RTRs. Therefore the derived equations require external validation in other populations to determine their accuracy, clinical utility, and generalizability.
Grant/funding support: This study was funded by The Physicians Services Incorporated Foundation (Grant no. R0359) and Astellas Pharma Canada. Instrumentation and reagents to measure BTP were provided by Dade Behring.
Financial disclosures: The Physicians Services Incorporated Foundation, Astellas Pharma Canada, and Dade Behring played no role in the study design, the data collection, analysis and interpretation, the manuscript preparation, or in any publication decisions.
Acknowledgments: We thank the staff and patients from the renal transplant program that participated in the study. We wish to acknowledge Alan Thibeau, Nur Jamal, Ian Graham, Lisa Banfield, Sheila Dowell, Philip St. Laurent, Julie Noel, Elyse Bienvenue, Martine Blouin, Claudine Messier, and Sunil Thakrar for assistance with the DTPA GFR measurements; Marcella Cheng-Fitzpatrick and Anna Micucci for data management; and Zeyad Alrayes, Judy Cheesman, Darlene Hackett, Margo McCoshen, Paul McLoughlin, Amy Pocock, Lisa South, and Jennifer Bowes for their invaluable assistance in conducting this study.
Previously published online at DOI: 10.1373/clinchem.2007.090126
(1.) Perrone RD, Macias NE, Levey AS. Serum creatinine as an index of renal function: new insights into old concepts. Clin Chem 1992;38:1933-53.
(2.) Coresh J, Astor BC, McQuillan G, Kusek J, Greene T, Van Lente F, et al. Calibration and random variation of the serum creatinine assay as critical elements of using equations to estimate glomerular filtration rate. Am J Kidney Dis 2002;39:920-9.
(3.) Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31-41.
(4.) 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 [Comment]. Ann Intern Med 1999;130:461-70.
(5.) White C, Akbari A, Hussain N, Dinh L, Filler G, Lepage N, et al. Estimating glomerular filtration rate in kidney transplantation: a comparison between serum creatinine and cystatin C-based methods. J Am Soc Nephrol 2005; 16:3763-70.
(6.) Mariat C, Alamartine E, Barthelemy JC, De Filippis JP, Thibaudin D, Berthoux P, et al. Assessing renal graft function in clinical trials: can tests predicting glomerular filtration rate substitute for a reference method? Kidney Int 2004;65:289-97.
(7.) Poge U, Gerhardt T, Palmedo H, Klehr HU, Sauerbruch T, Woitas RP. MDRD equations for estimation of GFR in renal transplant recipients. Am J of Transplant 2005;5:1306-11.
(8.) 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.
(9.) 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.
(10.) Huber AR, Risch L. Recent developments in the evaluation of glomerular filtration rate: is there a place for beta-trace [Comment]? Clin Chem 2005;51:1329-30.
(11.) Priem F, Althaus H, Birnbaum M, Sinha P, Conract HS, Jung K. Beta-trace protein in serum: a new marker of glomerular filtration rate in the creatinineblind range. Clin Chem 1999;45:567-8.
(12.) Filler G, Priem F, Lepage N, Sinha P, Vollmer I, Clark H, et al. Beta-trace protein, cystatin C, beta(2)-microglobulin, and creatinine compared for detecting impaired glomerular filtration rates in children. Clin Chem 2002; 48:729-36.
(13.) Woitas RP, Stoffel-Wagner B, Poege U, Schiedermaier P, Spengler U, Sauerbruch T. Low-molecular weight proteins as markers for glomerular filtration rate. Clin Chem 2001;47:2179-80.
(14.) Kobata M, Shimizu A, Rinno H, Hamada C, Maeda K, Fukui M, et al. Beta-trace protein, a new marker of GFR, may predict the early prognostic stages of patients with type 2 diabetic nephropathy. J Clin Lab Anal 2004;18:237-9.
(15.) Poge U, Gerhardt TM, Stoffel-Wagner B, Palmedo H, Klehr HU, Sauerbruch T, et al. beta-Trace protein is an alternative marker for glomerular filtration rate in renal transplantation patients. Clin Chem 2005;51:1531-3.
(16.) DuBois D, DuBois EF. A formula to estimate the approximate surface area if height and weight be known. Arch Intern Med 1916;17:863-71.
(17.) National Kidney Foundation. K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease: Evaluation, Classification and Stratification. Am J Kidney Dis 2003;39:S1-266.
(18.) Risch L, Herklotz R, Blumberg A, Huber AR. Effects of glucocorticoid immunosuppression on serum cystatin C concentrations in renal transplant patients. Clin Chem 2001;47:2055-9.
(19.) Risch L, Saely C, Reist U, Reist K, Hefti M, Huber AR. Course of glomerular filtration rate markers in patients receiving high-dose glucocorticoids following subarachnoical hemorrhage. Clin Chim Acta 2005;360:205-7.
(20.) Urade Y, Hayaishi 0. Biochemical, structural, genetic, physiological, and pathophysiological features of lipocalin-type prostaglandin D synthase. Biochim Biophys Acta 2000;1482:259-71.
Christine A. White,  Ayub Akbari, [2,3] Steve Doucette,  Dean Fergusson,  Naser Hussain,  Laurent Dinh,  Guido Filler,  Nathalie Lepage,  and Greg A. Knoll [2,3,4] *
 Division of Nephrology, Department of Medicine, Queens University, Kingston, Canada;
 Division of Nephrology, Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada;
 Kidney Research Centre, The Ottawa Health Research Institute, Ottawa, Canada;
 Clinical Epidemiology Program, The Ottawa Health Research Institute, Ottawa, Canada;
 Division of Nuclear Medicine, Department of Medicine, University of Ottawa, Ottawa, Canada;
 Division of Nephrology, Department of Pediatrics, University of Western Ontario, London, Canada;
 Department of Laboratory Medicine, Children's Hospital of Eastern Ontario and the University of Ottawa, Ottawa, Canada;
* address correspondence to this author at: Division of Nephrology, The Ottawa Hospital, Riverside Campus, 1967 Riverside Dr., Ottawa, Ontario, Canada K1H 7W9; fax 613-738-8337, e-mail email@example.com
Table 1. Stepwise regression models. Model Variables Model [R.sup.2] 1 BTP 0.756 2 BTP, urea 0.791 3 BTP, urea, sex 0.810 4 BTP, urea, sex, creatinine 0.815 5 BTP, urea, sex, creatinine, albumin 0.821 6 BTP, urea, sex, creatinine, albumin, 0.821 age 7 BTP, urea, sex, creatinine, albumin, 0.821 age, race
|Printer friendly Cite/link Email Feedback|
|Author:||White, Christine A.; Akbari, Ayub; Doucette, Steve; Fergusson, Dean; Hussain, Naser; Dinh, Laurent;|
|Date:||Nov 1, 2007|
|Previous Article:||Effect of sample aliquot size on the limit of detection and reproducibility of clinical assays.|
|Next Article:||Rapid cortisol assay during adrenal vein sampling in patients with primary aldosteronism.|