- Email: [email protected]
Serum cystatin C concentration as a marker of renal dysfunction in the elderly
Serum Cystatin C Concentration as a Marker of Renal Dysfunction in the Elderly Danilo Fliser, MD, and Eberhard Ritz, MD • Serum cystatin C is a novel marker of renal function claimed to be superior to plasma creatinine. We assessed both parameters in young normotensive subjects (n = 12; 6 men; mean age, 25 -+ 2 years) and elderly normotensive and hypertensive subjects (n = 41; 19 men; mean age, 67 -+ 6 years). Glomerular filtration rate (GFR) was measured in all individuals using the inulin clearance (C=n) technique. Compared with young subjects, mean GFR was modestly but significantly (P < 0.001) less in elderly subjects (young, 119 -+ 11 mL/min/1.73 m 2 versus elderly, 104 _+ 12 mL/min/1.73 m2). Mean plasma creatinine concentration was identical in both groups (young, 0.93 -+ 0.11 mg/dL versus elderly, 0.93 _+ 0.10 mg/dL; P < 0.90). Mean serum cystatin C concentration was significantly (P < 0.001) greater in elderly subjects (0.84 _ 0.10 mg/L) compared with young subjects (0.69 - 0.08 mg/L). In all but one elderly subject, plasma creatinine concentration was within the 95% confidence interval of plasma creatinine concentration in young subjects. Eleven of 41 elderly subjects (27%) had GFRs less than the lower 95% confidence interval, respectively, and 12 of 41 elderly subjects (29%) had a serum cystatin C concentration greater than the upper 95% confidence interval in young subjects. The correlation between serum cystatin C concentration and C=n (r = -0.65; P < 0.001) was considerably better than between plasma creatinine concentration and C=n (r = -0.30; P < 0.02). Serum cystatin C concentration is a better marker of renal dysfunction (ie, reduced GFR) than plasma creatinine concentration, at least in elderly subjects with plasma creatinine concentrations within the normal range.
© 2001 by the National Kidney Foundation, Inc. INDEX WORDS: Cystatin C; creatinine; elderly; glomerular filtration rate (GFR); renal function.
G
LOMERULAR filtration rate (GFR) decreases with age, mostly within the normal range in the absence of renal disease. 1,2 This age-dependent loss of renal function is accelerated by such comorbid conditions as atherosclerosis, hypertension, and heart failure. 3-6 Unfortunately, in the elderly, plasma creatinine level is notoriously unreliable as an indicator of GFR because the daily production of creatinine is diminished as a result of reduced muscle mass. ~,6 Consequently, plasma creatinine concentration underestimates the decline in GFR with age. When plasma creatinine concentration is still in the upper normal range in an elderly individual, a major decrease in GFR may already be present. 7 More reliable (plasma or serum) markers of renal function are needed in the elderly, in whom the assessment of a timed creatinine clearance is particularly difficult.8 Serum cystatin C concentration, a new endogenous marker of renal function, is believed to be superior to plasma creatinine concentration as an indicator of renal function. TM Cystatin C is a proteinase inhibitor with a low molecular weight that is produced in all nucleated cells. Therefore, its appearance rate in blood is constant. It is freely filtered in the renal glomeruli and reabsorbed and metabolized in the proximal tubule. Thus, the serum concentration of cystatin C is mainly determined by GFR; it does not depend
on muscle mass and sex. 12,13 TO explore whether the measurement of cystatin C serum is superior to the measurement of plasma creatinine to assess age-dependent changes in GFR, we compared both indicators of renal function in young normotensive subjects and elderly normotensive and hypertensive subjects against the gold standard of GFR, ie, inulin clearance (Cin).
SUBJECTS AND METHODS
Participants and Protocol The study protocol was approved by the Ethics Committee of the University of Heidelberg, Germany. All participants gave their written informed consent. Twelve young, healthy, normotensive subjects were recruited among students of the University of Heidelberg, and 22 elderly healthy normotensive subjects were recruited among members of the Academy for Elderly in Heidelberg. In addition, 19 elderly patients with mild to moderate hypertension were recruited from the outpatient department of the Division of Nephrology. To exclude primary renal disease, sonography, urinaly-
From the Division of Nephrology, Department of Internal Medicine, Ruperto-Carola University Heidelberg, Germany. Received April 28, 2000; accepted in revised form July 28, 2000. Address reprint requests to Danilo Fliser, MD, Department of Internal Medicine, Medical School Hannover, CarlNeuberg-Strt~e 1, 30625 Hannover, Germany. E-mail: [email protected] © 2001 by the National Kidney Foundation, Inc. 0272-6386/01/3701-001153.00/0 doi: 10.1053/ajkd. 2001.20591
American Journal of Kidney Diseases, Vo137, No 1 (January), 2001: pp 79-83
79
80
FLISER AND RITZ
sis (including phase-contrast microscopy), and serum chemistry tests were performed in all participants. Individuals with suspected renal disease were excluded from the study. Only subjects with normal plasma creatinine concentrations, ie, less than 1.3 mg/dL, were included on the study. All participants were white, and the groups of young and elderly subjects were matched with respect to sex and body weight. All participants were studied under outpatient conditions. They were advised to ingest a diet containing a standardized amount of sodium and protein 1 week before and during the examination. Ambulatory 24-hour blood pressure was measured using an automatic blood pressure monitoring device (model 90207; SpaceLabs Inc, Tampa, FL). On the morning of a separate day, GFR was measured after 12 hours of fasting, with the subject in a quiet room in the supine position, by means of the infusion Cin technique, described previously. 14 In brief, a priming dose of 1,500 mg of inulin/m2 (Inutest; Laevosan Co, Linz, Austria) was followed by a continuous 3-hour inulin infusion (10 mg/m2/ min) using an ultraprecise pump (Perfusor FT; Braun Melsungen, Melsungen, Germany). After an equilibration period of 90 minutes, blood samples to determine plasma inulin concentrations were collected every 15 minutes for another 90 minutes. In addition, blood samples to measure plasma creatinine and serum cystatin C concentrations were collected. Cardiovascular drugs with the potential to contound the measurements of blood pressure and GFR (mostly diuretics and g-blockers, but also angiotensin-converting enzyme inhibitors and calcium channel blockers) were washed out in the group of elderly hypertensive patients for at least 7 days before the study. This protocol was also used in a larger study of renal function in the elderly, published elsewhere. ('
Measurements and Calculations Inulin was measured enzymatically using inulinase.15Cin was calculated from the delivered dose as follows: C = (I, × lc)/S~ where C is the clearance, I~ is the infusion rate (in milliliters per minute), lc is the concentration of the analyte in the infusion fluid (in milligrams per milliliter), and Sc is the plasma concentration of the analyte (in milligrams per milliliter). Plasma creatinine concentration was analyzed with an autoanalyzer (Hitachi 705; Boebringer Mannbeim, Mannheim, Germany). Creatinine clearance was calculated after Cockroft and Gault:
concentration in young healthy volunteers (n = 309) was 0.50 to 0.96 mg/L. The reproducibility of 17 repetitive measurements of cystatin C in three serum pools was 2.3% (intra-assay coefficient of variation), and the daily variability of duplicate measurements over 10 working days was 7.9% (interassay coefficient of variation). ~
Statistical Analysis The SPSS package (SSPS lnc, Chicago, IL) was used for statistical analysis. The normality of data distribution was tested with the Shapiro-Wilk test. Comparison between both groups was performed using a two-tailed t-test for comparison of random data. Pearson's correlation analysis was performed between variables of renal function in all individuals studied. The differences were considered statistically significant at P of 0.05. All data are presented as mean _+ SD. RESULTS
Table 1 lists m e a n age, sex, m e a n b o d y weight, a n d data o n renal f u n c t i o n in b o t h groups. M e a n G F R (Cin) was s i g n i f i c a n t l y less in e l d e r l y indiv i d u a l s c o m p a r e d with the y o u n g h e a l t h y s u b jects d e s p i t e i d e n t i c a l m e a n p l a s m a c r e a t i n i n e c o n c e n t r a t i o n s ( P = 0.90). C o n v e r s e l y , m e a n s e r u m c y s t a t i n C c o n c e n t r a t i o n was s i g n i f i c a n t l y greater in e l d e r l y t h a n y o u n g subjects. T h e c a l c u lated c r e a t i n i n e c l e a r a n c e ( C o c k r o f t - G a u l t ) m a r k edly underestimated GFR. The 24-hour ambulatory m e a n arterial b l o o d p r e s s u r e w a s 88 _+ 6 m m H g in the g r o u p o f h e a l t h y y o u n g subjects, 92 _+ 9 m m H g in the g r o u p o f h e a l t h y e l d e r l y subjects, a n d 101 _+ 9 m m H g in the g r o u p o f e l d e r l y p a t i e n t s with m i l d to m o d e r a t e h y p e r t e n sion. F i g u r e 1 s h o w s i n d i v i d u a l G F R s (Fig 1A) b y Cin, as well as p l a s m a c r e a t i n i n e (Fig 1B) a n d s e r u m c y s t a t i n C c o n c e n t r a t i o n s (Fig 1C). A l l b u t o n e e l d e r l y s u b j e c t had a p l a s m a c r e a t i n i n e c o n Table 1. Mean Age, Sex, Mean Body Weight, and Indicators of Renal Function in Both Groups'Studied
body weight (in kilograms) × (140 - age)/0.814 x plasma creatinine (micromolar) ~6 The value was multiplied by 0.85 for women. Serum cystatin C concentration was measured with an immunonephelometric assay (N Latex Cystatin C; Dade Behring Marburg GmbH, Marburg, Germany) using the Bchring Ncphelometer II.9A7AsThe basic measuring range of the assay covers serum cystatin C concentrations from 0.2 to 6.8 mg/L, and the standard curve is linear from 0.7 to 6.8 mg/L. The 95% confidence interval for serum cystatin C
Age (y) Sex (M/W) Body weight (kg) GFR (mL/min/1.73 m 2) CCr (mL/min) S creatinine (mg/dL) S cystatin C (mg/L)
Young Subjects (n = 12)
Elderly Subjects (n - 41)
25 _+ 2 6/6 71 _+ 9 119 -+ 11 114 _+ 22 0.93 -+ 0.11 0.69 +- 0.08
67 _+ 6* 19/22 73 _+ 12 104 _+ 12" 74 ÷ 18" 0.93 -+ 0.10 0.84 + 0.10"
Abbreviations: CCr, Cockroft-Gault clearance; S, serum. *P < 0.01, young versus elderly subjects.
CYSTATIN C IN THE ELDERLY
81 S - c y s t a t i n C (rag/l)
A
GF R ( ml/min/1.73m 2)
1.3 o
150
o
O
o
o%
130
000
0.9
00
~
o
O-110
0.7
90
Q5
70
/z
-
//"
B
S- creatinine (mg/dt)
I
I
I
I
I
80
100
120
140
160
GFR (mr/min/1.73 m ~) ].3
Fig 2. Correlation between serum cystatin C concentration (S-cystatin C) and GFR in (0) young healthy subjects (n = 12) and (©) elderly healthy and hypertensive subjects (n = 41). The correlation between both parameters was highly significant (r = -0.65; P <
--
O OO
1.1
0.9
l,
0.001).
OO O
0.7 0.5
C
S-cystatin C (mg/l) ].2
--
1.0 0.8 0.6 O
0.L Young
Elderly
centration within the 95% confidence interval of plasma creatinine concentration in young subjects (ie, 0.71 to 1.15 mg/dL). Eleven of 41 elderly subjects (27%) had GFRs less than the lower limit of the 95% confidence interval in young subjects (ie, 96 mL/min/1.73 m2), and 12 of 41 elderly subjects (29%) had a serum cystatin C concentration greater than the upper limit of the 95% confidence interval in young subjects (ie, 0.85 rag/L). When all individuals were evaluated, the correlation between serum cystatin C concentration and GFR (r = -0.65; P < 0.001; Fig 2) was considerably better than the correlation between plasma creatinine concentration and GFR (r = -0.30; P < 0.02). Analyzed separately, the correlation coefficient (r) between serum cystatin C
Fig 1. (A) Individual GFRs and (B) serum creatinine (S-creatinine) and (C) serum cystatin C (S-cystatin C) concentrations in young healthy subjects (n = 12) and elderly healthy subjects and elderly patients with mild to moderate hypertension (n = 41). Line indicates mean value. (©), Healthy normotensive subjects; (S), hypertensive patients.
82
FLISER AND RITZ
concentration and GFR was -0.29 (P = 0.36) in young subjects, -0.58 (P < 0.005) in healthy elderly, and -0.64 (P < 0.003) in elderly patients with hypertension. The correlation coefficients for plasma creatinine concentration and GFR w e r e - 0 . 3 1 (P = 0.33; young subjects),-0.33 (P = 0.14; healthy elderly), and-0.37 (P = 0.12; elderly with hypertension). DISCUSSION
GFR measured by Ci, is modestly but significantly less in healthy elderly subjects and elderly patients with hypertension compared with young healthy subjects. 6 Conversely, mean plasma creatinine concentrations were identical in young and elderly individuals, and the 95% confidence interval for plasma creatinine concentration practically overlaps in these two populations. Thus, plasma creatinine concentration does not show small differences in GFR with age. In addition, the calculated creatinine clearance was not very helpful in this respect because it markedly underestimated true GFR. Our data confirm previous reports that the Cockcroft-Gault formula is unreliable for the estimation of GFR in elderly subjects. 6,19,2° The present data suggest that serum cystatin C concentration is more sensitive in this respect and provides a better reflection of GFR in the elderly. Our results in the elderly agree with the findings of other investigators who examined renal function in children ~3 and adult patients with cancer, 2~ kidney transplants, 22-24 and various renal diseases. 25-27 In these studies, serum cystatin C concentration was superior to plasma creatinine concentration for the detection of impaired renal function. For example, in patients with incipient renal failure, Tian et a125 documented that a mild reduction in GFR was detected more easily using serum cystatin C than plasma creatinine, a finding consistent with our results. The same conclusion was reached in a recently published study and the accompanying editorial in this journal. 27'28 The correlation between serum cystatin C concentration and GFR was considerably better than the correlation between plasma creatinine concentration and GFR, particularly in elderly individuals. Age-related changes in renal hemodynamics arc markedly amplified by coexisting cardiovascular pathological states. 4-6 In the elderly with comorbid conditions, increased serum cystatin C
concentration may therefore provide a clinically useful pointer to a decrease in GFR, even if plasma creatinine concentration has remained unchanged. The greatest serum cystatin C concentration in one of our elderly patients with hypertension, 1.16 mg/L (corresponding to a GFR of 86 mL/min/1.73 m2), was more than 65% greater than the mean serum cystatin C concentration (0.70 mg/L) in the young subjects. In conclusion, serum cystatin C concentration is a better marker than plasma creatinine concentration for the detection of subtle changes in GFR in the elderly with a plasma creatinine level still within the normal range. Additional studies should be performed to examine whether the measurement of serum cystatin C is particularly helpful for the detection of impaired renal function in sick and/or malnourished elderly individuals, in whom it is usually difficult to obtain a reliable assessment of creatinine clearance and measurement of GFR with exogenous markers is cumbersome. In such patients, a significant increase in serum cystatin C concentration could unmask a major decrease in GFR, which usually remains undetected by plasma creatinine determinations. ACKNOWLEDGMENT
The authors thank Dade-Behring Company, Marburg, Germany, for measuring cystatin C serum concentrations. REFERENCES
I. Rowe JW, Andres R, Tobin J, Norris AN, Shock NW: The effect of age on creatinine clearance in men: A crosssectional and longitudinal study. J Gerontol 31:155-163, 1976 2. Fliser D, Ritz E: Renal function and fluid-electrolyte homeostasis changes with age. J Geriatr NephrolUrol 3:107116, 1993 3. Lindeman RD, Tobin J, Shock NW: Association between blood pressure and the rate of decline in renal function with age. Kidney Int 26:861-868, 1984 4. LindemanRD, TobinJ, Shock NW: Longitudinalstudies on the rate of decline in renal function with age. J Am Geriatr Soc 33:278-285, 1985 5. Kasiske BL: Relationship between vascular disease and age-associatedchanges in the human kidney. Kidney Int 31:1153-1159, 1987 6. Fliser D, Franek E, Joest M, Block S, Mutschler E, Ritz E: Changes of renal function in the elderly--Influence of hypertension and cardiac function. Kidney Int 51 : 11961204, 1997 7. Fliser D, Ritz E: The relationship between hypertension and renal function and its therapeutic implications in the elderly.Gerontology44:123-131, 1998
CYSTATIN C IN THE ELDERLY
8. Goldberg TH, Finkelstein MS: Difficulties in estimating glomerular filtration rate in the elderly. Arch Intern Med 147:1430-1433, 1987 9. Randers E, Erlandsen EJ: Serum cystatin C as an endogenous marker of the renal function--A review. Clin Chem Lab Med 37:389-395, 1999 10. Keevil BG, Kilpatrick ES, Nichols SP, Maylor PW: Biological variation of cystatin C: Implication for the assessment of glomerular filtration rate. Clin Chem 44:1535-1539, 1998 11. Newman DJ, Thakkar H, Edwards RG, Wilkie M, White T, Grubb AO: Serum cystatin C measured by automated immunoassay: A more sensitive marker of changes in GFR than serum creatinine. Kidney Int 47:312-318, 1995 12. Randers E, Kristensen JH, Erlandsen EJ, Danielsen H: Serum cystatin C as a marker of the renal function. Scand J Clin Lab Invest 58:585-592, 1998 13. Bokenkamp A, Domanetzki M, Zinck R, Schumann G, Byrd D, Brodehl J: Cystatin C: New marker of glomerular filtration rate in children independent of age and height. Pediatrics 101:875-881, 1998 14. Fliser D, Zeier M, Nowack R, Ritz E: Renal function reserve in healthy elderly subjects. J Am Soc Nephrol 3:1371-1377, 1993 15. Kfihnle HE von Dahl K, Schmidt F: Fully enzymatic inulin determination in small volume samples without deproteination. Nephron 62:104-107, 1992 16. Cockroft DW, Gault MH: Prediction of creatinine clearance from serum creatinine. Nephron 16:31-41, 1976 17. Mussap M, Ruzzante N, Varagnolo M, Plebani M: Quantitative automated particle~Enhanced immunonephelometric assay for the routine measurement of human cystatin C. Clin Chem Lab Med 36:859-865, 1998 18. Finney H, Newman DJ, Gruber W, Merle P, Price CP: Initial evaluation of cystatin C measurement by particle enhanced immunonephelometry on the Behring Nephelometer systems. Clin Chem 43:1016-1022, 1997 19. Beck CL, Pucino F, Carlson JD, Silbergleit IL, Strommen GL, Fenelon JC, Lipp R, Gill DS: Evaluation of
83
creatinine clearance estimation in an elderly male population. Pharmacotherapy 8:183-188, 1988 20. O'Connell MB, Dwinell AM, Bannick-Mohrland SD: Predictive performance of equations to estimate creatinine clearance in hospitalized elderly patients. Ann Pharmacother 26:627-635, 1992 21. Stabuc B, Vrhovec L, Stabuc-Silih M, Cizej TE: Improved prediction of decreased creatinine clearance by serum cystatin C: Use in cancer patients before and during chemotherapy. Clin Chem 46:193-197, 2000 22. Bostom AG, Gohh RY, Bausserman L, Hakas D, Jaques PF, Selhub J, Dworkin L, Rosenberg IH: Serum cystatin C as a determinant of fasting total homocysteine levels in renal transplant recipients with a normal serum creatinine. J Am Soc Nephrol 10:164-166, 1999 23. Risch L, Blumberg A, Huber A: Rapid and accurate assessment of GFR in patients with renal transplants using serum cystatin C. Nephrol Dial Transplant 14:1991-1996, 1999 24. Le Bricon T, Thervet E, Benlakehal M, Bousquet B, Legendre C, Erlich D: Changes in plasma cystatin C after renal transplantation and acute rejection in adults. Clin Chem 45:2243-2249, 1999 25. Tian S, Kusano E, Ohara T, Tabei K, Itoh Y, Kawai T, Asano Y: Cystatin C measurement and its practical use in patients with various renal disease. Clin Nephrol 48:104108, 1997 26. Harmoinen AP, Kouri TT, Wirta OR, Lehtimaki TJ, Rantalaiho V, Tuljanmaa VM, Pasternack AI: Evaluation of plasma cystatin C as a marker for GFR in patients with type 2 diabetes. Clin Nephrol 52:363-370, 1999 27. Coil E, Botey A, Alvarez L, Poch E, Quinto L, Saurina A, Vera M, Piera C, Darnell A: 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 36:29-34, 2000 28. Bostom AG, Dworkin LD: Cystatin C measurement: Improved detection of mild decrements in glomerular filtration rate versus creatinine-based estimates. Am J Kidney Dis 36:205-207, 2000
© 2001 by the National Kidney Foundation, Inc. INDEX WORDS: Cystatin C; creatinine; elderly; glomerular filtration rate (GFR); renal function.
G
LOMERULAR filtration rate (GFR) decreases with age, mostly within the normal range in the absence of renal disease. 1,2 This age-dependent loss of renal function is accelerated by such comorbid conditions as atherosclerosis, hypertension, and heart failure. 3-6 Unfortunately, in the elderly, plasma creatinine level is notoriously unreliable as an indicator of GFR because the daily production of creatinine is diminished as a result of reduced muscle mass. ~,6 Consequently, plasma creatinine concentration underestimates the decline in GFR with age. When plasma creatinine concentration is still in the upper normal range in an elderly individual, a major decrease in GFR may already be present. 7 More reliable (plasma or serum) markers of renal function are needed in the elderly, in whom the assessment of a timed creatinine clearance is particularly difficult.8 Serum cystatin C concentration, a new endogenous marker of renal function, is believed to be superior to plasma creatinine concentration as an indicator of renal function. TM Cystatin C is a proteinase inhibitor with a low molecular weight that is produced in all nucleated cells. Therefore, its appearance rate in blood is constant. It is freely filtered in the renal glomeruli and reabsorbed and metabolized in the proximal tubule. Thus, the serum concentration of cystatin C is mainly determined by GFR; it does not depend
on muscle mass and sex. 12,13 TO explore whether the measurement of cystatin C serum is superior to the measurement of plasma creatinine to assess age-dependent changes in GFR, we compared both indicators of renal function in young normotensive subjects and elderly normotensive and hypertensive subjects against the gold standard of GFR, ie, inulin clearance (Cin).
SUBJECTS AND METHODS
Participants and Protocol The study protocol was approved by the Ethics Committee of the University of Heidelberg, Germany. All participants gave their written informed consent. Twelve young, healthy, normotensive subjects were recruited among students of the University of Heidelberg, and 22 elderly healthy normotensive subjects were recruited among members of the Academy for Elderly in Heidelberg. In addition, 19 elderly patients with mild to moderate hypertension were recruited from the outpatient department of the Division of Nephrology. To exclude primary renal disease, sonography, urinaly-
From the Division of Nephrology, Department of Internal Medicine, Ruperto-Carola University Heidelberg, Germany. Received April 28, 2000; accepted in revised form July 28, 2000. Address reprint requests to Danilo Fliser, MD, Department of Internal Medicine, Medical School Hannover, CarlNeuberg-Strt~e 1, 30625 Hannover, Germany. E-mail: [email protected] © 2001 by the National Kidney Foundation, Inc. 0272-6386/01/3701-001153.00/0 doi: 10.1053/ajkd. 2001.20591
American Journal of Kidney Diseases, Vo137, No 1 (January), 2001: pp 79-83
79
80
FLISER AND RITZ
sis (including phase-contrast microscopy), and serum chemistry tests were performed in all participants. Individuals with suspected renal disease were excluded from the study. Only subjects with normal plasma creatinine concentrations, ie, less than 1.3 mg/dL, were included on the study. All participants were white, and the groups of young and elderly subjects were matched with respect to sex and body weight. All participants were studied under outpatient conditions. They were advised to ingest a diet containing a standardized amount of sodium and protein 1 week before and during the examination. Ambulatory 24-hour blood pressure was measured using an automatic blood pressure monitoring device (model 90207; SpaceLabs Inc, Tampa, FL). On the morning of a separate day, GFR was measured after 12 hours of fasting, with the subject in a quiet room in the supine position, by means of the infusion Cin technique, described previously. 14 In brief, a priming dose of 1,500 mg of inulin/m2 (Inutest; Laevosan Co, Linz, Austria) was followed by a continuous 3-hour inulin infusion (10 mg/m2/ min) using an ultraprecise pump (Perfusor FT; Braun Melsungen, Melsungen, Germany). After an equilibration period of 90 minutes, blood samples to determine plasma inulin concentrations were collected every 15 minutes for another 90 minutes. In addition, blood samples to measure plasma creatinine and serum cystatin C concentrations were collected. Cardiovascular drugs with the potential to contound the measurements of blood pressure and GFR (mostly diuretics and g-blockers, but also angiotensin-converting enzyme inhibitors and calcium channel blockers) were washed out in the group of elderly hypertensive patients for at least 7 days before the study. This protocol was also used in a larger study of renal function in the elderly, published elsewhere. ('
Measurements and Calculations Inulin was measured enzymatically using inulinase.15Cin was calculated from the delivered dose as follows: C = (I, × lc)/S~ where C is the clearance, I~ is the infusion rate (in milliliters per minute), lc is the concentration of the analyte in the infusion fluid (in milligrams per milliliter), and Sc is the plasma concentration of the analyte (in milligrams per milliliter). Plasma creatinine concentration was analyzed with an autoanalyzer (Hitachi 705; Boebringer Mannbeim, Mannheim, Germany). Creatinine clearance was calculated after Cockroft and Gault:
concentration in young healthy volunteers (n = 309) was 0.50 to 0.96 mg/L. The reproducibility of 17 repetitive measurements of cystatin C in three serum pools was 2.3% (intra-assay coefficient of variation), and the daily variability of duplicate measurements over 10 working days was 7.9% (interassay coefficient of variation). ~
Statistical Analysis The SPSS package (SSPS lnc, Chicago, IL) was used for statistical analysis. The normality of data distribution was tested with the Shapiro-Wilk test. Comparison between both groups was performed using a two-tailed t-test for comparison of random data. Pearson's correlation analysis was performed between variables of renal function in all individuals studied. The differences were considered statistically significant at P of 0.05. All data are presented as mean _+ SD. RESULTS
Table 1 lists m e a n age, sex, m e a n b o d y weight, a n d data o n renal f u n c t i o n in b o t h groups. M e a n G F R (Cin) was s i g n i f i c a n t l y less in e l d e r l y indiv i d u a l s c o m p a r e d with the y o u n g h e a l t h y s u b jects d e s p i t e i d e n t i c a l m e a n p l a s m a c r e a t i n i n e c o n c e n t r a t i o n s ( P = 0.90). C o n v e r s e l y , m e a n s e r u m c y s t a t i n C c o n c e n t r a t i o n was s i g n i f i c a n t l y greater in e l d e r l y t h a n y o u n g subjects. T h e c a l c u lated c r e a t i n i n e c l e a r a n c e ( C o c k r o f t - G a u l t ) m a r k edly underestimated GFR. The 24-hour ambulatory m e a n arterial b l o o d p r e s s u r e w a s 88 _+ 6 m m H g in the g r o u p o f h e a l t h y y o u n g subjects, 92 _+ 9 m m H g in the g r o u p o f h e a l t h y e l d e r l y subjects, a n d 101 _+ 9 m m H g in the g r o u p o f e l d e r l y p a t i e n t s with m i l d to m o d e r a t e h y p e r t e n sion. F i g u r e 1 s h o w s i n d i v i d u a l G F R s (Fig 1A) b y Cin, as well as p l a s m a c r e a t i n i n e (Fig 1B) a n d s e r u m c y s t a t i n C c o n c e n t r a t i o n s (Fig 1C). A l l b u t o n e e l d e r l y s u b j e c t had a p l a s m a c r e a t i n i n e c o n Table 1. Mean Age, Sex, Mean Body Weight, and Indicators of Renal Function in Both Groups'Studied
body weight (in kilograms) × (140 - age)/0.814 x plasma creatinine (micromolar) ~6 The value was multiplied by 0.85 for women. Serum cystatin C concentration was measured with an immunonephelometric assay (N Latex Cystatin C; Dade Behring Marburg GmbH, Marburg, Germany) using the Bchring Ncphelometer II.9A7AsThe basic measuring range of the assay covers serum cystatin C concentrations from 0.2 to 6.8 mg/L, and the standard curve is linear from 0.7 to 6.8 mg/L. The 95% confidence interval for serum cystatin C
Age (y) Sex (M/W) Body weight (kg) GFR (mL/min/1.73 m 2) CCr (mL/min) S creatinine (mg/dL) S cystatin C (mg/L)
Young Subjects (n = 12)
Elderly Subjects (n - 41)
25 _+ 2 6/6 71 _+ 9 119 -+ 11 114 _+ 22 0.93 -+ 0.11 0.69 +- 0.08
67 _+ 6* 19/22 73 _+ 12 104 _+ 12" 74 ÷ 18" 0.93 -+ 0.10 0.84 + 0.10"
Abbreviations: CCr, Cockroft-Gault clearance; S, serum. *P < 0.01, young versus elderly subjects.
CYSTATIN C IN THE ELDERLY
81 S - c y s t a t i n C (rag/l)
A
GF R ( ml/min/1.73m 2)
1.3 o
150
o
O
o
o%
130
000
0.9
00
~
o
O-110
0.7
90
Q5
70
/z
-
//"
B
S- creatinine (mg/dt)
I
I
I
I
I
80
100
120
140
160
GFR (mr/min/1.73 m ~) ].3
Fig 2. Correlation between serum cystatin C concentration (S-cystatin C) and GFR in (0) young healthy subjects (n = 12) and (©) elderly healthy and hypertensive subjects (n = 41). The correlation between both parameters was highly significant (r = -0.65; P <
--
O OO
1.1
0.9
l,
0.001).
OO O
0.7 0.5
C
S-cystatin C (mg/l) ].2
--
1.0 0.8 0.6 O
0.L Young
Elderly
centration within the 95% confidence interval of plasma creatinine concentration in young subjects (ie, 0.71 to 1.15 mg/dL). Eleven of 41 elderly subjects (27%) had GFRs less than the lower limit of the 95% confidence interval in young subjects (ie, 96 mL/min/1.73 m2), and 12 of 41 elderly subjects (29%) had a serum cystatin C concentration greater than the upper limit of the 95% confidence interval in young subjects (ie, 0.85 rag/L). When all individuals were evaluated, the correlation between serum cystatin C concentration and GFR (r = -0.65; P < 0.001; Fig 2) was considerably better than the correlation between plasma creatinine concentration and GFR (r = -0.30; P < 0.02). Analyzed separately, the correlation coefficient (r) between serum cystatin C
Fig 1. (A) Individual GFRs and (B) serum creatinine (S-creatinine) and (C) serum cystatin C (S-cystatin C) concentrations in young healthy subjects (n = 12) and elderly healthy subjects and elderly patients with mild to moderate hypertension (n = 41). Line indicates mean value. (©), Healthy normotensive subjects; (S), hypertensive patients.
82
FLISER AND RITZ
concentration and GFR was -0.29 (P = 0.36) in young subjects, -0.58 (P < 0.005) in healthy elderly, and -0.64 (P < 0.003) in elderly patients with hypertension. The correlation coefficients for plasma creatinine concentration and GFR w e r e - 0 . 3 1 (P = 0.33; young subjects),-0.33 (P = 0.14; healthy elderly), and-0.37 (P = 0.12; elderly with hypertension). DISCUSSION
GFR measured by Ci, is modestly but significantly less in healthy elderly subjects and elderly patients with hypertension compared with young healthy subjects. 6 Conversely, mean plasma creatinine concentrations were identical in young and elderly individuals, and the 95% confidence interval for plasma creatinine concentration practically overlaps in these two populations. Thus, plasma creatinine concentration does not show small differences in GFR with age. In addition, the calculated creatinine clearance was not very helpful in this respect because it markedly underestimated true GFR. Our data confirm previous reports that the Cockcroft-Gault formula is unreliable for the estimation of GFR in elderly subjects. 6,19,2° The present data suggest that serum cystatin C concentration is more sensitive in this respect and provides a better reflection of GFR in the elderly. Our results in the elderly agree with the findings of other investigators who examined renal function in children ~3 and adult patients with cancer, 2~ kidney transplants, 22-24 and various renal diseases. 25-27 In these studies, serum cystatin C concentration was superior to plasma creatinine concentration for the detection of impaired renal function. For example, in patients with incipient renal failure, Tian et a125 documented that a mild reduction in GFR was detected more easily using serum cystatin C than plasma creatinine, a finding consistent with our results. The same conclusion was reached in a recently published study and the accompanying editorial in this journal. 27'28 The correlation between serum cystatin C concentration and GFR was considerably better than the correlation between plasma creatinine concentration and GFR, particularly in elderly individuals. Age-related changes in renal hemodynamics arc markedly amplified by coexisting cardiovascular pathological states. 4-6 In the elderly with comorbid conditions, increased serum cystatin C
concentration may therefore provide a clinically useful pointer to a decrease in GFR, even if plasma creatinine concentration has remained unchanged. The greatest serum cystatin C concentration in one of our elderly patients with hypertension, 1.16 mg/L (corresponding to a GFR of 86 mL/min/1.73 m2), was more than 65% greater than the mean serum cystatin C concentration (0.70 mg/L) in the young subjects. In conclusion, serum cystatin C concentration is a better marker than plasma creatinine concentration for the detection of subtle changes in GFR in the elderly with a plasma creatinine level still within the normal range. Additional studies should be performed to examine whether the measurement of serum cystatin C is particularly helpful for the detection of impaired renal function in sick and/or malnourished elderly individuals, in whom it is usually difficult to obtain a reliable assessment of creatinine clearance and measurement of GFR with exogenous markers is cumbersome. In such patients, a significant increase in serum cystatin C concentration could unmask a major decrease in GFR, which usually remains undetected by plasma creatinine determinations. ACKNOWLEDGMENT
The authors thank Dade-Behring Company, Marburg, Germany, for measuring cystatin C serum concentrations. REFERENCES
I. Rowe JW, Andres R, Tobin J, Norris AN, Shock NW: The effect of age on creatinine clearance in men: A crosssectional and longitudinal study. J Gerontol 31:155-163, 1976 2. Fliser D, Ritz E: Renal function and fluid-electrolyte homeostasis changes with age. J Geriatr NephrolUrol 3:107116, 1993 3. Lindeman RD, Tobin J, Shock NW: Association between blood pressure and the rate of decline in renal function with age. Kidney Int 26:861-868, 1984 4. LindemanRD, TobinJ, Shock NW: Longitudinalstudies on the rate of decline in renal function with age. J Am Geriatr Soc 33:278-285, 1985 5. Kasiske BL: Relationship between vascular disease and age-associatedchanges in the human kidney. Kidney Int 31:1153-1159, 1987 6. Fliser D, Franek E, Joest M, Block S, Mutschler E, Ritz E: Changes of renal function in the elderly--Influence of hypertension and cardiac function. Kidney Int 51 : 11961204, 1997 7. Fliser D, Ritz E: The relationship between hypertension and renal function and its therapeutic implications in the elderly.Gerontology44:123-131, 1998
CYSTATIN C IN THE ELDERLY
8. Goldberg TH, Finkelstein MS: Difficulties in estimating glomerular filtration rate in the elderly. Arch Intern Med 147:1430-1433, 1987 9. Randers E, Erlandsen EJ: Serum cystatin C as an endogenous marker of the renal function--A review. Clin Chem Lab Med 37:389-395, 1999 10. Keevil BG, Kilpatrick ES, Nichols SP, Maylor PW: Biological variation of cystatin C: Implication for the assessment of glomerular filtration rate. Clin Chem 44:1535-1539, 1998 11. Newman DJ, Thakkar H, Edwards RG, Wilkie M, White T, Grubb AO: Serum cystatin C measured by automated immunoassay: A more sensitive marker of changes in GFR than serum creatinine. Kidney Int 47:312-318, 1995 12. Randers E, Kristensen JH, Erlandsen EJ, Danielsen H: Serum cystatin C as a marker of the renal function. Scand J Clin Lab Invest 58:585-592, 1998 13. Bokenkamp A, Domanetzki M, Zinck R, Schumann G, Byrd D, Brodehl J: Cystatin C: New marker of glomerular filtration rate in children independent of age and height. Pediatrics 101:875-881, 1998 14. Fliser D, Zeier M, Nowack R, Ritz E: Renal function reserve in healthy elderly subjects. J Am Soc Nephrol 3:1371-1377, 1993 15. Kfihnle HE von Dahl K, Schmidt F: Fully enzymatic inulin determination in small volume samples without deproteination. Nephron 62:104-107, 1992 16. Cockroft DW, Gault MH: Prediction of creatinine clearance from serum creatinine. Nephron 16:31-41, 1976 17. Mussap M, Ruzzante N, Varagnolo M, Plebani M: Quantitative automated particle~Enhanced immunonephelometric assay for the routine measurement of human cystatin C. Clin Chem Lab Med 36:859-865, 1998 18. Finney H, Newman DJ, Gruber W, Merle P, Price CP: Initial evaluation of cystatin C measurement by particle enhanced immunonephelometry on the Behring Nephelometer systems. Clin Chem 43:1016-1022, 1997 19. Beck CL, Pucino F, Carlson JD, Silbergleit IL, Strommen GL, Fenelon JC, Lipp R, Gill DS: Evaluation of
83
creatinine clearance estimation in an elderly male population. Pharmacotherapy 8:183-188, 1988 20. O'Connell MB, Dwinell AM, Bannick-Mohrland SD: Predictive performance of equations to estimate creatinine clearance in hospitalized elderly patients. Ann Pharmacother 26:627-635, 1992 21. Stabuc B, Vrhovec L, Stabuc-Silih M, Cizej TE: Improved prediction of decreased creatinine clearance by serum cystatin C: Use in cancer patients before and during chemotherapy. Clin Chem 46:193-197, 2000 22. Bostom AG, Gohh RY, Bausserman L, Hakas D, Jaques PF, Selhub J, Dworkin L, Rosenberg IH: Serum cystatin C as a determinant of fasting total homocysteine levels in renal transplant recipients with a normal serum creatinine. J Am Soc Nephrol 10:164-166, 1999 23. Risch L, Blumberg A, Huber A: Rapid and accurate assessment of GFR in patients with renal transplants using serum cystatin C. Nephrol Dial Transplant 14:1991-1996, 1999 24. Le Bricon T, Thervet E, Benlakehal M, Bousquet B, Legendre C, Erlich D: Changes in plasma cystatin C after renal transplantation and acute rejection in adults. Clin Chem 45:2243-2249, 1999 25. Tian S, Kusano E, Ohara T, Tabei K, Itoh Y, Kawai T, Asano Y: Cystatin C measurement and its practical use in patients with various renal disease. Clin Nephrol 48:104108, 1997 26. Harmoinen AP, Kouri TT, Wirta OR, Lehtimaki TJ, Rantalaiho V, Tuljanmaa VM, Pasternack AI: Evaluation of plasma cystatin C as a marker for GFR in patients with type 2 diabetes. Clin Nephrol 52:363-370, 1999 27. Coil E, Botey A, Alvarez L, Poch E, Quinto L, Saurina A, Vera M, Piera C, Darnell A: 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 36:29-34, 2000 28. Bostom AG, Dworkin LD: Cystatin C measurement: Improved detection of mild decrements in glomerular filtration rate versus creatinine-based estimates. Am J Kidney Dis 36:205-207, 2000