Diagnostic accuracy of Cystatin C as a marker of kidney disease in patients with multiple myeloma: calculated glomerular filtration rate formulas are equally useful.
The limitations of serum creatinine as a marker of the glomerular filtration rate (GFR) are widely appreciated (4,5). Creatinine clearance may be more sensitive, but it requires a timed urine collection, which is imprecise (6) and inconvenient (7). For clinical purposes, [sup.51]Cr-labeled EDTA clearance provides a surrogate gold standard measure of GFR (6, 8), but it is time-consuming, expensive, and not readily available in many hospitals.
Attempts to improve clinical measurement of GFR include the use of creatinine-based formulas, including those proposed by Cockcroft and Gault (9) and Levey and coworkers [Modification of Diet in Renal Disease (MDRD) formula (10) and its simplified version (11)]. These improve GFR estimation compared with serum creatinine alone (12), although considerable limitations persist. Cystatin C is a 13-kDa protein whose plasma concentration reflects GFR. Its superiority over serum creatinine in terms of diagnostic sensitivity for reduced GFR is generally accepted (13-21), but concerns have been expressed that it may be affected by malignant progression (22-26). To our knowledge, neither cystatin C nor the MDRD formulas have been evaluated against a gold standard measure of GFR in patients with multiple myeloma.
For this study, we recruited 39 Caucasian volunteers with a confirmed diagnosis of multiple myeloma through hematology outpatient clinics. The study was approved by the Local Research Ethics Committee. Patients were prospectively enrolled from April 2001 to February 2003 during times the investigators were available. Exclusion criteria were active rheumatoid disease, renal dialysis, or renal transplantation. Patients attended the hospital for a [sup.51]Cr-EDTA clearance within 1 month of recruitment, bringing with them a 24 h urine collection. Blood was taken for serum cystatin C, ([[beta].sub.2] microglobulin, urea, albumin, and creatinine measurement. Patients were not asked to fast on the day of attendance. Descriptive data for the patients are given in file 1 of the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol50/issue10/.
GFR was estimated from a single [sup.51]Cr-EDTA injection and three blood samples by mono-exponential analysis with the Brochner-Mortenson correction (27). GFR measurements were undertaken by technologists/ scientists and reported by accredited specialist physicians working in a nuclear medicine department accredited for training purposes by the Institute of Physical Sciences in Medicine. Serum and urinary creatinine were measured by a kinetic Jaffe method (Integra 800 analyzer, Roche Diagnostics) calibrated with the manufacturer's recommended humanserum-based material: between-day imprecision was <3% at concentrations of 113 and 333 [micro]mol/L. Serum cystatin C was measured by a particle-enhanced nephelometric immunoassay (Dako Ltd.) on an automated rate nephelometer (Immage[TM] analyzer; Beckman-Coulter) (28): between-day imprecision was 6.6% at 0.97 mg/L and 2.6% at 5.53 mg/L. Absence of interference from paraproteins had been confirmed (29). Serum [[beta].sub.2] microglobulin was measured by a latex particle-enhanced turbidimetric immunoassay (Binding Site Ltd.): between-batch imprecision was <8%. Serum albumin and urea were measured by bromcresol green and urease methods, respectively (Integra 800 analyzer). Creatinine clearance, Cockcroft and Gault (9), and MDRD (10,11) calculated clearances, body mass index [weight (kg)/height [(m).sup.2]], and body surface area (30) were calculated by use of standard formulas. [sup.51]Cr-EDTA, measured and calculated clearance GFR estimates were all adjusted to a mean surface area of 1.73 [m.sup.2]. Paraprotein typing was confirmed by studying serum and urine samples by use of agarose gel electrophoresis and immunofixation (Hydrasys; Sebia), and paraprotein concentrations were determined by densitometric scanning. Analyses were undertaken in an accredited (CPA UK Ltd.) laboratory by state-registered biomedical scientists blinded to the [sup.51]Cr-EDTA results.
Data were analyzed by Analyze-it[TM] software (Analyze-it Software Ltd.). [sup.51]Cr-EDTA was the reference method: log-transformed values were used because this gave a closer approximation to a gaussian distribution for this variable. The analysis was complicated by the known nonlinear and inverse relationship between serum markers of GFR (creatinine and cystatin C) and GFR itself, compared with the other clearance estimates, which are expressed in the same units as GFR (creatinine clearance and formulaic estimates of clearance). To enable comparison, reciprocal and log-transformed analyses were also calculated. For creatinine clearance and the formulaic estimates of clearance, comparison with [sup.51]Cr-EDTA clearance was also undertaken by use of difference plots, and the bias and imprecision of the estimates were compared by the paired t-test and F-test, respectively.
Using the relevant reference interval for the kidney function tests (see file 1 in the online Data Supplement), we assessed sensitivity for the detection of moderate kidney disease against a recently proposed threshold of 60 mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] (12). Differences in the sensitivities of the test markers for renal dysfunction were tested by the Fisher exact probability test (2-tailed P values are given). The existence of a relationship between cystatin C and tumor burden was tested by comparing cystatin C concentrations against (a) serum [[beta].sub.2]-microglobulin and (b) paraprotein concentration and (c) by comparing cystatin C concentrations in patients with stage I, II, and III disease.
The [R.sup.2] of linear regression analyses comparing the test marker (y) with [sup.51]Cr-EDTA (x) were 0.91 for Cockcroft and Gault and simplified MDRD clearances, 0.90 for MDRD clearance and creatinine clearance, 0.86 for serum creatinine, and 0.73 for cystatin C (Table 1).
On average, creatinine clearance and the three formulaic estimates of clearance all showed a slight positive bias compared with [sup.51]Cr-EDTA (P <0.05 in all cases; Table 1). In terms of the precision of the estimate of GFR, the 95% limits of agreement [mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] in all cases] did not differ (P >0.05) between the simplified MDRD (43.5), MDRD (47.3), Cockcroft and Gault (50.4), and measured creatinine clearance (50.9) estimates (Table 1).
There was no relationship (P >0.05) between serum paraprotein concentration and either log-transformed GFR ([R.sup.2] = 0.02) or cystatin C ([R.sup.2] = 0.03). We observed a significant positive relationship between serum cystatin C and ([[beta].sub.2]-microglobulin concentrations [R.sup.2] = 0.66 (P <0.0001); [[beta].sub.2]-microglobulin = 3.704(cystatin C) - 0.44]: however, when ([[beta].sub.2]-microglobulin concentration was adjusted for glomerular dysfunction (by expressing as a ratio to serum creatinine), this relationship was abolished ([R.SUP.2] = 0.00). There were no significant differences (P >0.05) between the cystatin C/creatinine ratios in patients with stage I, II, or III disease. There was a significant relationship between serum cystatin C and serum creatinine [[R.sup.2] = 0.77 (P <0.0001); cystatin C = 0.0077(creatinine) + 0.56].
GFR was <60 mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] in 20 of 39 patients, but serum creatinine was increased above the reference interval (i.e., true positive for moderate kidney disease) in only 8 of 20, compared with 20 of 20 for cystatin C (P <0.0001; Fig. 1). Cockcroft and Gault- and MDRD-calculated clearances were <60 mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] in 15 of 20 patients, both of which appeared to be inferior to cystatin C for detecting moderate kidney disease (P = 0.0471). The simplified MDRD and measured creatinine clearances detected 16 of 20 of these patients (P >0.05 compared with cystatin C).
Among 19 patients with GFR [greater than or equal to] 60 mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1], 10 had serum cystatin C concentrations exceeding the reference interval (i.e., false positive for moderate kidney disease) compared with none for serum creatinine (P = 0.0004). All 19 had MDRD and simplified MDRD clearance estimates >60 mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] (P = 0.0030): one measured creatinine clearance and one Cockcroft and Gault clearance estimate was <60 mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] (P = 0.0004). (P values are compared with cystatin C in all cases.) There were no particular distinguishing clinical features (drug history, presence of free light chains, skeletal lesions, or comorbid conditions) in the patients with increased serum cystatin C concentration and GFR [greater than or equal to] 60 mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1].
Finney et al. (31) also demonstrated a strong correlation between serum cystatin C and creatinine in myeloma patients (r = 0.89 compared with r = 0.88 in the present study), but an estimate of GFR was not reported by these authors. They also observed no relationship between serum cystatin C and disease burden. Tumor burden probably does not exert a strong influence on serum cystatin C concentration independently of the effect of myeloma itself on kidney function. Given this, does cystatin C measurement confer any advantages in the assessment of kidney function in these patients compared with conventional markers? Cystatin C was more sensitive than serum creatinine at detecting moderate reductions in GFR, and we observed a good correlation with GFR, although the strength of this relationship was lower than that observed with all other markers tested, including creatinine clearance. The good performance of creatinine clearance in this setting is unusual: in nearly all other studies, the performance of creatinine clearance was inferior to that of serum creatinine and calculated estimates of GFR (12,32) and serum cystatin C (17, 21, 33 ). Many of our patients had been receiving medical care of their disease for some time: they may have been more familiar, and consequently competent, with the 24-h urine collection procedure. All calculated clearance estimates gave reasonable [R.sup.2] coefficients against [sup.51]Cr-EDTA and showed reasonable sensitivity and specificity for the detection of moderate kidney disease.
[FIGURE 1 OMITTED]
In summary, cystatin C appears to reflect GFR in patients with multiple myeloma, as observed in other clinical settings, but calculated estimates of GFR appear to provide equally good information at much lower cost. In keeping with current best practice guidelines (12), we recommend that calculated estimates of GFR should supplement serum creatinine measurement alone and replace measured creatinine clearance.
We thank Wendy Van Der Steen of the Nuclear Medicine Department for the [sup.51]Cr-EDTA measurements. We are grateful to Drs. Y. Williams, A. Borg, R. Gale, and C. Pocock, from the Department of Hematology, for assistance with patient recruitment. We would also like to thank the staff of the Clinical Biochemistry Department, Kent and Canterbury Hospitals, for their cooperation and help. We are grateful to Dr. J. Sheldon from the Protein Reference Unit, St. George's Hospital, London, for the ([[beta].sub.2]-microglobulin measurements. This work was funded by the South East Regional NHS Project Grant Scheme (Grant Reference SEO 150).
(1.) Winearls CG. Acute myeloma kidney. Kidney Int 1995;48:1347-61.
(2.) Johnson WJ, Kyle RA, Pineda AA, O'Brien PC, Holley KE. Treatment of renal failure associated with multiple myeloma. Plasmapheresis, hemodialysis, and chemotherapy. Arch Intern Med 1990;150:863-9.
(3.) Irish AB, Winearls CG, Littlewood T. Presentation and survival of patients with severe renal failure and myeloma. Q J Med 1997;90:773-80.
(4.) Perrone RD, Macias NE, Levey AS. Serum creatinine as an index of renal function: new insights from old concepts. Clin Chem 1992;38:1933-53.
(5.) Duncan L, Heathcote J, Djurdjev 0, Levin A. Screening for renal disease using serum creatinine: who are we missing? Nephrol Dial Trans 2001;16: 1042-6.
(6.) Brochner-Mortensen J, Rodbro P. Selection of routine method for determination of glomerular filtration rate in adults. Scand J Clin Lab Invest 1976;36:35-43.
(7.) Payne RB. Creatinine clearance: a redundant clinical investigation. Ann Clin Biochem 1986;23:243-50.
(8.) Brochner-Mortensen J, Giese J, Rossing N. Renal inulin clearance versus total plasma clearance of [sup.51]Cr-EDTA. Scand J Clin Lab Invest 1969;23: 301-5.
(9.) Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31-41.
(10.) Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine. Ann Intern Med 1999;130:461-70.
(11.) Levey AS, Greene T, Kusek JW. A simplified equation to predict glomerular filtration rate from serum creatinine. J Am Soc Nephrol 2000;11:AO828.
(12.) National Kidney Foundation-K/DOQI. Clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002;39(Suppl 1):S1-266.
(13.) Price CP, Finney H. Developments in the assessment of glomerular filtration rate. Clin Chim Acta 2000;297:55-66.
(14.) Kyhse-Andersen J, Schmidt C, Nordin G, Andersson B, Nilsson-Ehle P, Lindstrom V, et al. Serum cystatin C determined by a rapid, automated particle-enhanced turbidimetric method, is a better marker than serum creatinine for glomerular filtration rate. Clin Chem 1994;40:1921-6.
(15.) Newman DJ, Thakkar H, Edwards RG, Wilkie M, White T, Grubb AO, et al. Serum cystatin C measured by automated immunoassay: a more sensitive marker of changes in GFR than serum creatinine. Kidney Int 1995;47: 312-8.
(16.) Bostom AG, Dworkin LID. Cystatin C measurement: improved detection of mild decrements in glomerular filtration rate versus creatinine-based estimates. Am J Kidney Dis 2000;36:205-7.
(17.) Coll E, Botey A, Alvarez L, Poch E, Quinto L, Saurina A, et al. Serum cystatin C as a new marker for noninvasive estimation of glomerular filtration rate and as a marker for early renal impairment. Am J Kidney Dis 2000;36:2934.
(18.) Dharnicharka VR, Kwon C, Stevens G. Serum cystatin C is superior to serum creatinine as a marker of kidney function: a meta-analysis. Am J Kidney Dis 2002;40:221-6.
(19.) Laterza OF, Price CP, Scott MG. Cystatin C: an improved estimator of glomerular filtration rate? Clin Chem 2002;48:699-707.
(20.) Fliser D, Ritz E. Serum cystatin C concentration as a marker of renal dysfunction in the elderly. Am J Kidney Dis 2001;37:79-83.
(21.) O'Riordan SE, Webb MC, Stowe HJ, Simpson DE, Kandarpa M, Coakley AJ, et al. Cystatin C improves the detection of mild renal dysfunction in older patients. Ann Clin Biochem 2003;40:648-55.
(22.) Kos J, Stabuc B, Cimerman N, Brunner N. Serum cystatin C, a new marker of glomerularfltration rate, is increased during malignant progression. Clin Chem 1998;44:2256-7.
(23.) Heidtmann HH, Salge U, Abrahamson M, Bencina M, Kastelic L, Kopitar-Jerala N, et al. Cathepsin B and cysteine proteinase inhibitors in human lung cancer cell lines. Clin Exp Metastasis 1997;15:368-81.
(24.) Corticchiato O, Cajot JF, Abrahamson M, Chan SJ, Keppler D, Sordat B. Cystatin C and cathepsin B in human colon carcinoma: expression by cell lines and matrix degradation. Int J Cancer 1992;52:645-52.
(25.) Kos J, Stabuc B, Schweiger A, Krasovec M, Cimerman N, Kopitar-Jerala N, et al. Cathepsins B, H, and L and their inhibitors stefin A and cystatin C in sera of melanoma patients. Clin Cancer Res 1997;3:1815-22.
(26.) De Vos J, Thykjaer T, Tarte K, Krasovec M, Cimerman N, Kopitar-Jerala N, et al. Comparison of gene expression profiling between malignant and normal plasma cells with oligonuclectice arrays. Oncogene 2002;21:6848-57.
(27.) Blaufox MD, Aurell M, Bubeck B, Fommei E, Piepsz A, Russell C, et al. Report of radionuclides in nephrourology committee on renal clearance. J Nucl Med 1996;37:1883-90.
(28.) Stowe H, Lawrence D, Newman DJ, Lamb EJ. Analytical performance of a particle-enhanced nephelometric immunoassay (PENIA) for serum cystatin C using rate analysis. Clin Chem 2001;47:1482-5.
(29.) Lamb E, Stowe H. Rheumatoid factor can interfere with cystatin C measurement. Ann Clin Biochem 2003;40:195-6.
(30.) DuBois D, DuBois EF. A formula to estimate the approximate surface area if height and weight are known. Ann Inter Med 1916;17:863-71.
(31.) Finney H, Williams AH, Price CP. Serum cystatin C in patients with myeloma. Clin Chim Acta 2001;309:1-6.
(32.) Lamb EJ, Webb MC, Simpson DE, Coakley AJ, Newman DJ, O'Riordan SE. Estimation of glomerular filtration rate in older patients with chronic renal insufficiency: is the MDRD formula an improvement? J Am Geriatrics Soc 2003;51:1012-7.
(33.) Van Den Noortgate NJ, Janssens WH, Delanghe JR, Afschrift MB, Lameire NH. Serum cystatin C concentration compared with other markers of glomerular filtration rate in the old old. J Am Geriatr Soc 2002;50:1278-82.
Previously published online at DOI: 10.1373/clinchem.2004.036947
Edmund J. Lamb,  * Helen J. Stowe,  David E. Simpson,  Anthony J. Coakley,  David J. Newman,  and Maeve Leahy  (Departments of  Clinical Biochemistry,  Nuclear Medicine, and  Haematology, East Kent Hospitals National Health Service Trust, Kent and Canterbury Hospital, Canterbury, Kent, UK;  South West Thames Institute for Renal Research, St. Helier Hospital, Carshalton, Surrey, UK; * address correspondence to this author at: Department of Clinical Biochemistry, East Kent Hospitals NHS Trust, Kent and Canterbury Hospital, Canterbury, Kent CT1 3NG, UK; fax 44-01227-783077, e-mail edmund. email@example.com)
Table 1. Regression and Bland-Altman difference analyses comparing various GFR test markers (y) with [sup.51]Cr-EDTA clearance (x). (a) GFR test marker (y) Regression equation [R.sup.2] Serum cystatin C, mg/L y = -3.1512x + 6.965 0.77 Log serum cystatin C, mg/L y = -0.7121x + 1.365 0.73 1/Serum cystatin C, mg/L y = 1.0308x - 0.978 0.55 Serum creatinine, [micro]mol/L y = 356.07x + 740 0.75 Log serum creatinine, [micro]mol/L y = 0.8211x + 3.452 0.86 1/Serum creatinine, [micro]mol/L y = 0.0132x - 0.013 0.75 Clearance, mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] Creatinine clearance y = 107.11x - 118.5 0.79 Log creatinine clearance y = 1.1240x - 0.180 0.90 C&G-calculated clearance y = 99.851x - 108.2 0.75 Log C&G-calculated clearance y = 0.9889x + 0.0474 0.91 MDRD-calculated clearance y = 99.76x - 106.2 0.81 Log MDRD-calculated clearance y = 1.0264x - 0.004 0.90 Simplified MDRD clearance y = 99.57x - 106.6 0.83 Log simplified MDRD clearance y = 1.0471x - 0.045 0.91 Mean (95% CI) (b) difference, mL x [min.sup.-1] GFR test marker (y) [(1.73 [m.sup.2]).sup.-1] Serum cystatin C, mg/L Log serum cystatin C, mg/L 1/Serum cystatin C, mg/L Serum creatinine, [micro]mol/L Log serum creatinine, [micro]mol/L 1/Serum creatinine, [micro]mol/L Clearance, mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] Creatinine clearance 6.7 (2.4-10.9) (c) Log creatinine clearance C&G-calculated clearance 4.5 (0.3-8.7) (c) Log C&G-calculated clearance MDRD-calculated clearance 6.3 (2.4-10.3) (c) Log MDRD-calculated clearance Simplified MDRD clearance 5.7 (2.1-9.3) (c) Log simplified MDRD clearance 95% limits of agreement, mL x [min.sup.-1] GFR test marker (y) [(1.73 [m.sup.2]).sup.-1] Serum cystatin C, mg/L Log serum cystatin C, mg/L 1/Serum cystatin C, mg/L Serum creatinine, [micro]mol/L Log serum creatinine, [micro]mol/L 1/Serum creatinine, [micro]mol/L Clearance, mL x [min.sup.-1] x [(1.73 [m.sup.2]).sup.-1] Creatinine clearance -18.8 to 32.1 Log creatinine clearance C&G-calculated clearance -20.7 to 29.7 Log C&G-calculated clearance MDRD-calculated clearance -17.3 to 30.0 Log MDRD-calculated clearance Simplified MDRD clearance -16.1 to 27.4 Log simplified MDRD clearance (a) For all regression analyses, log-transformed values for [sup.51]Cr-EDTA were used because they gave a closer approximation to a gaussian distribution for that variable. All clearance estimates have been adjusted to a body surface area of 1.73 [m.sup.2]. (b) CI, confidence interval; C&G, Cockcroft and Gault. (c) P < 0.05 compared with [sup.51]Cr-EDTA.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Technical Briefs|
|Author:||Lamb, Edmund J.; Stowe, Helen J.; Simpson, David E.; Coakley, Anthony J.; Newman, David J.; Leahy, M|
|Date:||Oct 1, 2004|
|Previous Article:||Simple extraction protocol for analysis of immunosuppressant drugs in whole blood.|
|Next Article:||Circulating corticotropin-releasing hormone mRNA in maternal plasma: relationship with gestational age and severity of preeclampsia.|