ujms109_3.pdf Upsala J Med Sci 109: 247–254, 2004 Cystatin C vs creatinine as markers of renal function in patients on digoxin treatment Pär Hallberg,1 Håkan Melhus,1 Lars-Olof Hansson,1 Anders Larsson1 1Department of Medical Sciences, Clinical Chemistry and Pharmacology, University Hospital, Uppsala, Sweden ABSTRACT A free full-text copy of this article can be found at the web page of Upsala J Med Sci: http://www.ujms.se Background: The kidney function is a major determinant of the serum concentra- tion of digoxin as this drug is mainly eliminated unchanged through the kidneys. Since digoxin is widely prescribed among the elderly, and the glomerular filtration rate (GFR) declines with age, it is important that the clinician takes the patient’s GFR into account when prescribing digoxin. Serum cystatin C has been suggested to be superior to creatinine for estimation of GFR, which may have relevance for the optimization of treatment with digoxin. Methods: To evaluate which of the two GFR markers serum creatinine and serum cystatin C that best correlates with serum digoxin, we compared the serum levels of digoxin with the serum levels of creati- nine and cystatin C in 149 patients on therapeutic drug monitoring of digoxin at our hospital. Results: Overall, there was a stronger correlation between serum digoxin concentrations and cystatin C (p=0.00001) as compared to creatinine (p= 0.00003). Interestingly, of the patients with a serum digoxin concentration 1.5 nmol/L, 29% had a serum creatinine level within normal limits, as compared to 20% with normal cystatin C levels. Conclusions: In this study, serum cystatin C correlated better to serum digoxin than did serum creatinine. With improved GFR monitoring, digoxin concentrations should be better controlled. INTRODUCTION Digoxin is commonly used in the treatment of congestive heart failure and atrial fib- rillation. The drug has become a subject of discussion after recent publications that showed a gender-related difference in mortality [1], a gender-related difference in serum (s) digoxin [2], and subsequently an increased mortality for men with s- 247 Received 24 May 2004 Accepted 7 June 2004 Key words: Creatinine, Cystatin C, Digoxin, Glomerular filtraten rate, Kidney. digoxin 1.5 nmol/L[3]. This has led to a revised recommendation at our hospital to target s-digoxin to <1.5 nmol/L. In this context, an heightened attention to the patient’s s-digoxin level is warranted. Adverse drug events affect millions of patients each year and was responsible for up to 14% of acute hospital admissions at a Swedish internal medicine clinic in 2001 [4]. In the elderly, adverse drug events occur in 14.6% to 35%, depending on the population setting and measures employed for their identification [5, 7]. Toxic concentrations of digoxin may require hospital treatment and digoxin intoxication is one of the most frequent causes of hospital care due to toxic drug effects[8]. Among individuals aged 75 and older, almost 20% were on medication with digoxin in 1996 [9]. Despite declining use in the last few years, digoxin is still one of the most frequently prescribed drugs; it was listed twice among the top 200 pre- scriptions in 2000 [10]. In 1995, it was the drug most often monitored therapeuti- cally [11] because of its narrow therapeutic window and potentially serious side effects. Digoxin has a half-life of approximately 1.5 days and the drug is mainly elimi- nated unchanged in the urine [12]. Glomerular filtration rate (GFR) has a major impact on s-digoxin [12]. As geriatric patients often have reduced GFR, monitoring kidney function during digoxin treatment is important. In the last decades, creatinine has become the most commonly used marker of GFR [13, 14]. Despite its common use, creatinine has limitations as a renal function marker. Creatinine is influenced by factors such as age, gender, muscle mass, physi- cal activity and diet [15]. It is also insensitive for detection of small decreases in GFR, in the so-called creatinine-blind GFR area, due to the non-linear relationship between serum concentration and GFR [16]. Thus, there is a need for better GFR markers. Several markers such as �-trace protein, cystatin C and � 2 -microglobulin have been suggested as alternatives to creatinine [17–19]. The normal serum level of cystatin C is <1.20 mg/L for patients less than 50 years of age and <1.55 mg/L for patients over 50 years of age, while increasing levels is detected in serum from patients with reduced GFR. Cystatin C is a polypeptide with a molecular mass of 13 kDa and an ellipsoid molecular shape with axes of about 30 and 45 Å [20]. Studies on the handling of human cystatin C in the rat have shown that the serum clearance of cystatin C is 94% of that of the generally used GFR-marker 51Cr-EDTA [21]. A recent meta-analysis has indicated that s-cystatin C is superior to s-creatinine as a renal function marker [22]. We have studied the correlation between s-digoxin and the GFR markers s-creati- nine and s-cystatin C. The study was performed to evaluate which of the two GFR markers s-creatinine and s-cystatin C that best correlated with s-digoxin. One earlier study of 18 healthy elderly individuals found no correlation between either s-cys- tatin C or s-creatinine and digoxin clearance, and thus concluded that s-cystatin C did not offer any advantages over s-creatinine in this respect [23]. However, the small number of study subjects may have been insufficient to detect a correlation, and also, actual digoxin treated patients have not been studied. 248 MATERIALS AND METHODS Patient samples and assays Consecutive routine requests (n=163) for therapeutic drug monitoring of s-digoxin were also analyzed for s-cystatin C and s-creatinine. The material consisted of sam- ples from 98 females and 65 males, both in- and out-patients treated with digoxin for any medical condition. The mean age was 80 years (range 55–106 years), and the mean dose of digoxin was 0.18 mg/day (range 0.07–0.3 mg/day). S-digoxin was determined on the Advia 1650 (Bayer Corp., Tarrytown, NY, USA). S-digoxin values below 0.6 nmol/L were reported as <0.6 nmol/L. S-cystatin C measurements were performed by a latex-enhanced reagent (N Latex Cystatin C, Dade Behring, Deerfield, IL, USA) using a Behring BN ProSpec analyzer (Dade Behring). The total analytical imprecision (coeficient of variation, CV) of the method was 4.8% at 0.56 mg/L and 3.7% at 2.85 mg/L. S-creatinine measurements were performed by means of the modified kinetic Jaffe reaction on the Advia 1650 analyzer (Bayer Corp) and reported as S.I. units (µ mol/L). The total analytical imprecision (CV) of the method was 2.6% at 170 µ mol/L and 2.4% at 740 µ mol/L. All assays were performed independently and without prior knowledge of other test results at the Department of Clinical Chemistry and Pharmacology at our hospital. The study was approved by the local ethical board at Uppsala University (1–167). Statistical calculations Data on s-cystatin C, s-creatinine and s-digoxin were normally distributed, and sta- tistical correlation analysis was performed with the Pearson product-moment corre- lation test using Statistica 5.1 (StatSoft Inc., Tulsa, OK, USA). Digoxin values <0.6 (n=14) were excluded from the statistical analyses, giving a total number of patients of 149. P-values <0.05 were taken as statistically significant throughout the study. RESULTS Digoxin concentrations When analyzing only those patients who had reached steady-state concentrations of digoxin and in whom serum levels had been measured at through were analyzed (n=94), mean s-digoxin was 1.5 nmol/L (range 0.6-4.0 nmol/L), with higher con- centrations among females (1.6 nmol/L) than males (1.4 nmol/L). This difference did not reach statistical significance (p=0.24, unpaired t-test). Among these patients, 33% of the digoxin concentrations were 1.5 nmol/L and 9% >2.5 nmol/L. There was no correlation between age and s-digoxin (p=0.86, r=-0.02, n=94). Correlation between s-digoxin and s-cystatin C vs s-creatinine Overall, s-digoxin correlated stronger to s-cystatin C (p=0.00001, R=0.35, n=149) as compared to s-creatinine (p=0.00003, r=0.34, n=149). The difference was more marked (fig 1 and 2) when only patients who had reached steady-state of digoxin 249 250 Fig. 1. Correlation between S-digoxin and S-cystatin C for individual patients. Fig. 2. Correlation between S-digoxin and S-creatinine for individual patients. and in whom serum levels had been measured at through (n=94) were analyzed (s- digoxin vs s-cystatin C, p=0.000001, r=0.45, n=94 s-digoxin vs s-creatinine, p= 0.0003, p=0.37, n=94). The difference was similar in both genders. Also, of the patients with a s-digoxin concentration 1,5 nmol/L, 29% had a s-creatinine level within normal limits (69–113 µ mol/L), as compared to 20% with normal s-cystatin C levels. There were also inverse correlations between digoxin dose and s-cystatin C (p=0.007, r=–0.28, N=94), and digoxin dose and s-creatinine (p=0.04, r=–0.22, n=94). The correlation between s-cystatin C and s-creatinine is illustrated in fig 3. DISCUSSION GFR is generally accepted as the best overall index of renal function. GFR decreas- es with age and reduced GFR is the most important complication of renal disease. Reduced GFR affects the clearance of many drugs used today, including digoxin, so that in many cases the recommended dose has to be adjusted depending on the patient’s GFR. There is thus a need for robust GFR markers. Inulin, iohexol and 51Cr-EDTA clearances are considered the golden standards for GFR measurements [22, 24]. The disadvantage with these assays is that they are cumbersome, costly and slow which may delay the start of treatment. Assays such as s-creatinine and s- cystatin C can provide rapid test results. Creatinine in combination with the Cock- 251 Fig 3. Correlation between s-creatinine and s-cystatin C for individual patients. croft-Gault equation is often used to estimate GFR [25]. Using actual body weight in the Cockcroft-Gault equation overestimates the GFR for obese patients [26]. An alternative could be to use lean body mass, but this is not usually available. Creati- nine often overestimates GFR in patients with slight reductions in GFR [27]. This may cause the prescribing physician to treat the patient with unnecessary high drug doses, which will increase the cost and possibly cause side effects. The inverse correlation between digoxin dose and s-cystatin C and s-creatinine in this study indicates that the kidney function should be considered when digoxin is prescribed. However, the high prevalence of patients with uncontrolled s-digoxin 1.5 nmol/L, and the strong correlations between s-digoxin and s-cystatin C and s- creatinine indicate that even greater consideration should be taken to GFR function. S-digoxin correlated stronger to s-cystatin C as compared to s-creatinine, which is in agreement with previous studies showing s-cystatin C to be superior to s-creati- nine as a marker of renal function [22]. The use of cystatin C has been hampered by the limited availability of the test and the problem of relating cystatin C to an esti- mated GFR. The introduction of new cystatin C tests that can be applied on widely available chemical analyzers will increase the availability of the test. Our hospital has for the last year offered measurements of s-cystatin C as a STAT request and also reported cystatin C in mg/L as well as converted to GFR (mL/min). This has improved the clinical usefulness of the assay and we have noticed a rapid increase in the number of requests for cystatin C. We conclude that greater consideration should be taken to GFR when prescribing digoxin, and that s-cystatin C correlates better to s-digoxin than does s-creatinine among patients on digoxin treatment. ACKNOWLEDGEMENTS We greatly appreciate the technical assistance of Ms Charina Brännström. REFERENCES 1. Rathore, S.S., Wang, Y. and Krumholz, H.M. 2002. Sex-based differences in the effect of digoxin for the treatment of heart failure. N Engl J Med 18: 1403–1411. 2. Hallberg, P., Michaelsson, K. and Melhus, H. 2003. Digoxin for the treatment of heart failure. N Engl J Med 7: 661–663; author reply 661–663. 3. Rathore, S.S., Curtis, J.P., Wang, Y., Bristow, M.R. and Krumholz, H.M. 2003. Association of serum digoxin concentration and outcomes in patients with heart failure. JAMA 7: 871–878. 4. Mjorndal, T., Boman, M.D., Hagg, S., Backstrom, M., Wiholm, B.E., Wahlin, A. and Dahlqvist, R. 2002. Adverse drug reactions as a cause for admissions to a department of internal medicine. Pharmacoepidemiol Drug Saf 1: 65–72. 5. Lindley, C.M., Tully, M.P., Paramsothy, V. and Tallis, R.C. 1992. Inappropriate medication is a major cause of adverse drug reactions in elderly patients. Age Ageing 4: 294–300. 6. Gray, S.L., Sager, M., Lestico, M.R. and Jalaluddin, M. 1998. Adverse drug events in hospitalized elderly. J Gerontol A Biol Sci Med Sci 1: M59–63. 7. Hanlon, J.T., Schmader, K.E., Koronkowski, M.J., Weinberger, M., Landsman, P.B., Samsa, G.P. and Lewis, I.K. 1997. Adverse drug events in high risk older outpatients. J Am Geriatr Soc 8: 945–948. 252 8. Juurlink, D.N., Mamdani, M., Kopp, A., Laupacis, A. and Redelmeier, D.A. 2003. Drug-drug interactions among elderly patients hospitalized for drug toxicity. Jama 13: 1652–1658. 9. Wills, P., Fastbom, J., Claesson, C.B., Cornelius, C., Thorslund, M. and Winblad, B. 1996. Use of cardiovascular drugs in an older Swedish population. J Am Geriatr Soc 1: 54–60. 10. Steimer, W., Muller, C. and Eber, B. 2002. Digoxin assays: frequent, substantial, and potentially dangerous interference by spironolactone, canrenone, and other steroids. Clin Chem 3: 507–516. 11. Jortani, S.A. and Valdes, R., Jr. 1997. Digoxin and its related endogenous factors. Crit Rev Clin Lab Sci 3: 225–274. 12. Mooradian, A.D. 1988. Digitalis. An update of clinical pharmacokinetics, therapeutic monitoring techniques and treatment recommendations. Clin Pharmacokinet 3: 165–179. 13. Kemperman, F.A., Krediet, R.T. and Arisz, L. 2002. Formula-derived prediction of the glomerular filtration rate from plasma creatinine concentration. Nephron 4: 547–558. 14. Schwartz, G.J., Haycock, G.B., Edelmann, C.M., Jr. and Spitzer, A. 1976. A simple estimate of glomerular filtration rate in children derived from body length and plasma creatinine. Pediatrics 2: 259–263. 15. Hsu, C.Y., Chertow, G.M. and Curhan, G.C. 2002. Methodological issues in studying the epi- demiology of mild to moderate chronic renal insufficiency. Kidney Int 5: 1567–1576. 16. Sherman, D.S., Fish, D.N. and Teitelbaum, I. 2003. Assessing renal function in cirrhotic patients: problems and pitfalls. Am J Kidney Dis 2: 269–278. 17. Grubb, A.O. 2000. Cystatin C-properties and use as diagnostic marker. Adv Clin Chem 35: 63–99. 18. Grubb, A., Simonsen, O., Sturfelt, G., Truedsson, L. and Thysell, H. 1985. Serum concentration of cystatin C, factor D and beta 2-microglobulin as a measure of glomerular filtration rate. Acta Med Scand 5: 499–503. 19. Filler, G., Priem, F., Lepage, N., Sinha, P., Vollmer, I., Clark, H., Keely, E., Matzinger, M., Akbari, A., Althaus, H. and Jung. K. 2002. Beta-trace protein, cystatin C, beta(2)-microglobulin, and creatinine compared for detecting impaired glomerular filtration rates in children. Clin Chem 5: 729–736. 20. Bode, W., Engh, R., Musil, D., Thiele, U., Huber, R., Karshikov, A., Brzin, J., Kos, J. and Turk, V. 1988. The 2.0 A X-ray crystal structure of chicken egg white cystatin and its possible mode of interaction with cysteine proteinases. Embo J 8: 2593–2599. 21. Tenstad, O., Roald, A.B., Grubb, A. and Aukland, K. 1996. Renal handling of radiolabelled human cystatin C in the rat. Scand J Clin Lab Invest 5: 409–414. 22. Dharnidharka, V.R., Kwon, C. and Stevens, G. 2002. Serum cystatin C is superior to serum creati- nine as a marker of kidney function: a meta-analysis. Am J Kidney Dis 2: 221–226. 23. O’Riordan, S., Ouldred, E., Brice, S., Jackson, S.H. and Swift, C.G. 2002. Serum cystatin C is not a better marker of creatinine or digoxin clearance than serum creatinine. Br J Clin Pharmacol 4: 398–402. 24. Price, C.P. and Finney, H. 2000. Developments in the assessment of glomerular filtration rate. Clin Chim Acta 1–2: 55–66. 25. Cockcroft, D.W. and Gault, M.H. 1976. Prediction of creatinine clearance from serum creatinine. Nephron 1: 31–41. 26. Leader, W.G., Tsubaki, T. and Chandler, M.H. 1994. Creatinine-clearance estimates for predicting gentamicin pharmacokinetic values in obese patients. Am J Hosp Pharm 17: 2125–2130. 27. Verhave, J.C., Balje-Volkers, C.P., Hillege, H.L., de Zeeuw D,. and de Jong, P.E. 2003. The relia- bility of different formulae to predict creatinine clearance. J Intern Med 5: 563–573. Corresponding author: Anders Larsson Department of Medical Sciences, Uppsala University Hospital, S-751 85 Uppsala, Sweden Telephone: 46-18-6110000 FAX: 46-18-552562 E-mail: anders.larsson@akademiska.se 253