Serum creatine kinase activity is not a reliablemarker for muscle damage in conditions

associated with low extracellularglutathione concentration

Johan J. Gunst, Michel R. Langlois, Joris R. Delanghe,* Marc L. De Buyzere,and Geert G. Leroux-Roels

Creatine kinase (CK, EC 2.7.3.2) assays usually containthiol-reducing compounds to restore the enzyme activ-ity. In this study, we investigated the effect of endoge-nous extracellular glutathione on serum CK activity. Weexamined CK activity and glutathione concentrations inserum from 200 healthy subjects (107 males, 93 females)and 38 patients with multiple organ failure, musclewasting, and low serum CK activity (<50 U/L) (24 males,14 females). Muscle damage was further evaluated usingserum myoglobin concentrations and aldolase activity.In the overall group, serum glutathione concentrationscorrelated with serum CK activity (r 5 0.791) but notwith myoglobin concentrations and aldolase activity. Inpatients with multiple organ failure, low serum CKactivities were accompanied by extremely low serumglutathione concentrations (<0.5 mmol/L, P <0.001).Endogenous glutathione can be regarded as a CK-pre-serving agent during the lifetime of the enzyme in thecirculation (22 h on average). Serum CK activity shouldbe interpreted with caution in patients with liver dis-ease and multiple organ failure. In these conditions, theloss of CK activity due to extracellular glutathionedepletion cannot be restored by the presence of thiol-reducing compounds in the CK assays.

Serum creatine kinase (CK, ATP:creatine N-phospho-transferase, EC 2.7.3.2) activity is an indicator commonlyused in the diagnosis of heart and skeletal muscle disor-ders. Because the thiol groups of the enzyme are easilyoxidized, CK is relatively unstable, and its activity is

apparently lost as a result of internal disulfide formation.Generally, reagents for the determination of CK activitycontain thiol-reducing agents such as N-acetylcysteine(NAC) and glutathione to maximally restore the enzymeactivity in vitro (1). Because the life-span of the enzyme is;1.4 times the biological half-life, CK stays ;22 h (firstorder kinetics) in the extracellular compartment beforedeactivation and elimination (2–4). During the time theenzyme stays in the circulation, CK is prone to oxidativestress that may produce an irreversible reduction ofcatalytic activity in vivo before blood sampling.

Until now, the in vivo effects of endogenous thiol-reducing compounds like glutathione on CK activity havenot been studied. In the present study, we investigatedthe effect of endogenous extracellular glutathione concen-trations on serum CK activity. Particularly, we evaluatedthe serum glutathione status in conditions associated withlow serum CK activity despite the presence of importantmuscle wasting (5, 6). The presence of muscle damagewas further evaluated using serum myoglobin concentra-tions and aldolase (EC 4.1.2.13) activity measurements.Furthermore, we investigated the in vitro stability ofserum CK in the presence of various glutathione concen-trations.

Materials and Methodssubjects and patientsBlood was sampled by venipuncture, allowed to clot, andcentrifuged (1000g, 10 min, room temperature). The su-pernatant serum was collected for analysis. Serum wasobtained from 200 healthy sedentary blood donors (107males, 93 females; ages 30 6 7 years) without recentskeletal muscle trauma (within 2 months before the sam-pling). Concomitantly, 38 intensive care patients (24males, 14 females; ages 57 6 15 years) with low serum CKactivity (,50 U/L) suffering from multiple organ failure

Laboratory of Clinical Chemistry, University Hospital Gent, De Pintelaan185, B-9000 Gent, Belgium.

*Author for correspondence. Fax 32 9 240 4985; e-mail joris.delanghe@rug.ac.be.

Received October 2, 1997; revision accepted December 19, 1997.

Clinical Chemistry 44:5939–943 (1998) Enzymes and Protein

Markers

939

(impaired renal and liver function and muscle wasting inthe presence of inflammation) were studied. This studywas approved by the ethical committee of the UniversityHospital of Ghent.

biochemical assaysThe catalytic activity of CK in serum was determined at37 °C according to the IFCC method (7) on a HITACHI747 analyzer using CK-NAC reagents (Boehringer Mann-heim). The total glutathione concentration in serum wasmeasured according to the method described by Griffith(8), which was performed on a HITACHI 911 analyzer(Boehringer Mannheim) according to the manufacturer’sinstructions. All necessary reagents were purchased fromSigma Chemicals, including glutathione (reduced form),b-NADPH-tetrasodium salt, Ellman’s reagent [5,59-dithio-bis(2-nitrobenzoic acid)], and glutathione reductase (EC1.6.4.2; type III from baker’s yeast). The serum myoglobinconcentration was measured by fixed-time immunoneph-elometry on a BN II nephelometer using the N LatexMyoglobin kit (Behringwerke AG) (9). The aldolase activ-ity in serum was determined at 37 °C on a HITACHI 911analyzer using commercial reagents (Boehringer Mann-heim) (10).

stability of ck in vitroSerum was obtained from healthy young men and pooled(initial CK activity, 121 U/L; glutathione concentration,2.95 mmol/L). Glutathione (reduced form) was added tothe serum pool to obtain final concentrations of 2.95, 4.95,and 6.95 mmol/L. A serum pool with a low glutathioneconcentration (,0.5 mmol/L; initial CK activity, 133 U/L)was studied as well. In another low glutathione pool(initial CK activity, 171 U/L), supplementation with re-duced glutathione (final concentration, 3.0 mmol/L) andoxidized glutathione (Sigma Chemicals; final concentra-tions, 1.5 and 3.0 mmol/L) was examined. The catalyticactivity of CK was monitored during incubation of thevarious serum pools at 37 °C for 48 h.

reversibility of thiol oxidationBecause addition of thiol-reducing agents to the serumbefore analysis might restore CK activity (11, 12), thepotential reversibility of thiol oxidation was investigatedby adding fresh aqueous solutions of NAC (final concen-

tration, 10 mmol/L), b-mercaptoethanol (final concentra-tion, 280 mmol/L), and reduced glutathione (final con-centration, 10 mmol/L) to the serum 30 min before CKassay. NAC and b-mercaptoethanol were purchased fromFluka Chemie AG. The addition of these agents producedminimal sample dilution ( ,1%). These sample pretreat-ments were performed in two sera from healthy controls,two sera from intensive care patients, and in pooled serawith low (,0.5 mmol/L) and physiological (3.15 mmol/L)glutathione concentrations after incubation for 48 h at37 °C in vitro. Results were expressed as the percentagechange of CK activity of the untreated sera.

statisticsResults were expressed as median and interquartileranges. Comparison of data between patients and healthysubjects was performed using the Mann–Whitney U-test.Correlations of serum glutathione with other parameterswere examined using regression analysis. Statistical sig-nificance was considered as P ,0.05.

Resultsserum glutathione and muscle parametersThe data on glutathione, CK, aldolase, and myoglobin inhealthy subjects and in intensive care patients sufferingfrom multiple organ failure are summarized in Table 1. Inthe patients, serum CK activity was low (P ,0.001)despite increased values found for the other musclemarkers, myoglobin and aldolase. Similarly, serum gluta-thione concentrations were lower in intensive care pa-tients than in healthy subjects (P ,0.001). Serum myoglobinconcentrations and aldolase activities were significantlyhigher in intensive care patients, as expected for multiple-organ failure and muscle wasting. We observed nogender-related difference in the total serum glutathioneconcentration. The low serum glutathione and CK activityin multiple-organ-failure patients were present in bothsexes.

relationship between serum ck and glutathioneIn the overall study population (controls and patients), asignificant correlation was observed between the serumCK activity (y, U/L) and the serum glutathione concen-tration (x, mmol/L): y 5 32.28x 1 5.85, r 5 0.791, Syux 530.04 (Fig. 1). The regression equations differed between

Table 1. Serum glutathione and muscle markers in healthy controls and in patients with multiple organ failure.Controls Patients

Males Females Males Females

n 107 93 24 14Glutathione, mmol/L 3.02 (2.66–3.50)a 2.88 (2.49–3.40) 0.20 (0.15–0.43)b 0.36 (0.10–0.98)b

CK, U/L 118 (96–156) 78 (61–103) 12 (10–34)b 34 (24–47)b

Myoglobin, mg/L 14.8 (11.8–22.9) 10.0 (7.1–13.8) 18.6 (16.2–52.9)c 18.8 (12.1–60.8)b

Aldolase, U/L 4.1 (3.0–5.3) 2.9 (2.2–3.6) 5.4 (4.8–5.9)d 5.3 (4.1–8.1)b

a Median (interquartile range)b P ,0.001, c P ,0.005, and d P ,0.05 versus controls (Mann–Whitney U-test).

940 Gunst et al.: Serum CK and glutathione

controls (y 5 41.97x 2 24.35, r 5 0.655) and intensive carepatients (y 5 31.03x 2 0.43, r 5 0.663). In contrast, nosignificant correlation was observed between the serumglutathione concentration and the serum myoglobin con-centration or serum aldolase activity.

serum ck stability in vitroThe in vitro stability of serum CK in the presence ofvarious extracellular glutathione concentrations is shownin Fig. 2. The rate of loss of CK activity depended on theserum glutathione concentration. In the sample with aphysiological serum glutathione concentration (2.95mmol/L), CK activity was reduced by 30% after 30 hincubation. The stability of the enzyme was higher in thepresence of supraphysiological amounts of glutathione:these samples showed a residual CK activity of .90%during the same incubation period. In the pool with the

lowest glutathione concentration (0.38 mmol/L), CK hasthe highest rate of activity loss. After 30 h (correspondingwith two plasma half-lives of the enzyme activity), theserum CK activity in the latter sample was alreadyreduced by 70%. Addition of oxidized glutathione to alow glutathione pool (from intensive care patients) didnot change the stability of CK (residual activity ,30%after 30 h), whereas addition of reduced glutathione to thesame pool produced a residual CK activity of 55% after30 h incubation.

reversibility of thiol oxidationAddition of thiol-reducing compounds (NAC, b-mercap-toethanol, or glutathione) to serum 30 min before analysisdid not change the measured CK activity in healthycontrols (increase of original activity of the untreated sera,2%). Similarly, this sample pretreatment did not restore

Fig. 1. Correlation between serum CK activity(y-axis, U/L) and extracellular glutathione concen-tration (x-axis, mmol/L) observed in healthy con-trols (F) and in patients with multiple-organ failure(M) (n 5 238): y 5 32.3x 1 5.8, r 5 0.791, Syux 530.04.

Fig. 2. In vitro stability of serum CK during incubationfor 48 h in the presence of various concentrations ofreduced glutathione, expressed as the percentage ofresidual CK activity at 37 °C.Serum glutathione concentrations were 0.38 mmol/L (F),2.95 mmol/L (M), 4.95 mmol/L (E), and 6.95 mmol/L (f).CK activity was measured at 0, 6, 24, 30, and 48 h ofincubation.

Clinical Chemistry 44, No. 5, 1998 941

the low CK activity in sera from intensive care patients(,5% increase). The in vitro loss of CK activity in thepresence of low and physiological glutathione concentra-tions for 48 h could not be reversed by this additionalsample treatment before analysis (,4% increase of CKactivity of untreated sera).

DiscussionIn the present study, we have demonstrated that theserum CK activity is dependent on the extracellularglutathione concentration. This is not the case with theother biochemical muscle markers, aldolase and myoglo-bin. Furthermore, low serum CK activities are observed inintensive care patients with low extracellular glutathioneconcentrations. In these patients, the depletion of theextracellular glutathione pools is one of the CK-modifyingfactors in serum that is responsible for the low serum CKactivities.

In contrast to CK, the serum myoglobin concentrationand aldolase activity were increased in critically ill pa-tients. Although muscle wasting was evident in thesepatients, increases of myoglobin concentrations partiallyreflect the decreased renal function in patients with mul-tiple organ failure. Aldolase is a less specific marker formuscle injury than CK because it is also present in othertissues (13).

Addition of thiol-containing components such as NACto the reaction mixture of CK assays is commonly used torestore the serum CK activity (1). Addition of thiolsduring sample storage has also been found to have arestoring effect on CK activity. However, we found thatthe effect of major in vivo glutathione depletion on CKactivity cannot be restored by the incorporation of reacti-vating compounds in the CK assays. Even addition ofthiol-reducing agents directly to serum before analysisdoes not restore CK activity.

In vivo extracellular glutathione is partly reduced,partly oxidized (14). The in vitro long-term incubationexperiments performed in this study confirm the obser-vation that endogenous reduced glutathione protectsagainst the aging of CK in biological fluids. In contrast,oxidized glutathione did not protect against the loss of CKactivity in vitro.

Because CK is mainly metabolized by the liver macro-phages (15), the plasma half-life of CK increases in severeliver insufficiency. Consequently, an accumulation of CKactivity in plasma is to be expected in this case. Therefore,the finding of low serum CK activity in multiple organfailure and liver insufficiency is paradoxical but can beexplained by oxidative damage of the enzyme producedby extracellular glutathione depletion. Low extracellularglutathione concentrations have been associated withliver disease and increased oxidative stress. Patients suf-fering from liver cirrhosis show lower extracellular gluta-thione concentrations (16). Certain drugs (17) (e.g.,paracetamol and isoniazid) can deplete the extracellularglutathione concentration, as do aging and fasting (18).

Previous occasional findings of unexpected low serumCK concentrations after proven myocardial infarction(19–20) and in patients suffering from infective endocar-ditis (21), sepsis (5, 6), connective tissue disease (22),prolonged illness (23), severe liver disease (24–26), andmetastatic disease (12) are also in agreement with thepresent findings. These conditions may lead to an under-estimation of the myocardial infarct size when serum CKactivity measurements are used (27). Consequently, meth-ods for infarct-sizing based on the determination of serumCK cannot be recommended in conditions associated withlow serum glutathione concentrations. In these situations,methods based on other biochemical markers such asmyoglobin (28) (in the absence of renal failure) should bepreferred to assess the myocardial infarct size. Becausethe muscle cell can also produce glutathione (29), intra-cellular glutathione depletion (increased oxidative stressin the myocyte) may decrease the CK activity before theenzyme is released into the circulation.

In addition to glutathione depletion, other causes forlow CK activity have been described. In multiple organfailure, tissue factors, such as lysosomal enzymes, that areable to deactivate CK are released into the circulation (30).Treatment with certain drugs also has been associatedwith a low serum CK activity (31–34). Furthermore, theabsence of physical exercise in the (immobilized) criticallyill patients contributes to low CK activity in serum.

In conclusion, attention should be paid to the effect ofendogenous glutathione when interpreting serum CKactivity, especially in clinical conditions associated withlow extracellular glutathione concentrations. This in vivoeffect cannot be restored by the in vitro addition ofreactivating sulfhydryl compounds to the reagents for CKdetermination or directly to the serum before analysis.

References1. Szasz G, Gerhardt W, Gruler W. Creatine kinase in serum. 5.

Effects of thiols on isoenzyme activity during storage at varioustemperatures. Clin Chem 1978;24:1557–63.

2. Cember H, ed. Introduction to health physics, 2nd ed. New York:Pergamon Press, 1983:71.

3. Lang H, Wurzburg U. Creatine kinase, an enzyme of many forms.Clin Chem 1982;28:1439–47.

4. Wevers R, Hagelauer U, Stein W, Bohner J, Faust U, Van Land-eghem A, et al. Indices for the age of the creatine kinase M-chainin the blood. Clin Chim Acta 1985;148:197–210.

5. Delanghe J, De Buyzere M, De Scheerder I, Van Den Abeele AM,Vandenbogaerde J, Wieme R. Significance of low creatine kinasein intensive-care patients [Letter]. Clin Chem 1986;34:713–4.

6. Delanghe J, De Buyzere M, De Scheerder I, Van Rostenberghe H,Faust U, Rodenbach J, et al. Post-transcriptional modification ofserum creatine kinase in infected intensive care patients. ClinChim Acta 1990;187:115–24.

7. Horder M, Elser R, Gerhardt W, Mathieum W, Simpson E. IFCCmethod for the measurement of catalytic concentration of en-zymes. Part 7. IFCC method for creatine kinase (ATP: creatineN-phosphotransferase, EC 2.7.3.2) J IFCC 1989;1:130–9.

8. Griffith OW. Glutathione and glutathione disulfide. In: Bergmeier

942 Gunst et al.: Serum CK and glutathione

HU, ed. Methods of enzymatic analysis. Vol VIII, 3rd ed. Weinheim:Verlag Chemie, 1984:521–9.

9. Delanghe J, Chapelle J-P, Van Der Schueren S. Quantitativenephelometric assay for determining myoglobin evaluated. ClinChem 1990;36:1675–8.

10. Pinto P, Kaplan A, Van Dreal P. Aldolase II. Spectrophotometricdetermination using an ultraviolet procedure. Clin Chem 1969;15:349–60.

11. Rollo J, Davis J, Ladenson J, McDonald J, Bruns D. Effects ofb-mercaptoethanol and chelating agents on the stability andactivation of creatine kinase in serum. Clin Chim Acta 1978;87:189–98.

12. Bruns D, Morgan W, Davis J, Ladenson J. Low apparent creatinekinase activity and prolonged lag phases in serum of patients withmetastatic disease: elimination by treatment of sera with sulfhy-dryl agents. Clin Chem 1976;22:1889–95.

13. Moss DW, Henderson AR. Enzymes. In: Burtis CA, Ashwood ER,eds. Tietz textbook of clinical chemistry, 2nd ed. Philadelphia: WBSaunders, 1994:735–896.

14. Paroni R, De Vecchi E, Cighetti G, Arcelloni C, Fermo I, Grossi A,Bonini P. HPLC with o-phthalaldehyde precolumn derivatization tomeasure total, oxidized, and protein-bound glutathione in blood,plasma, and tissue. Clin Chem 1995;41:448–54.

15. Smit M, Buursma A, Bouma J, Gruber M. Receptor-mediatedendocytosis of lactate dehydrogenase M4 by liver macrophages: amechanism for elimination of enzymes from plasma. Evidence forcompetition by creatine kinase MM, adenylate kinase, malate andalcohol dehydrogenase. J Biol Chem 1987;262:13020–6.

16. Loguercio C, Taranto D, Vitale LM, Beneduce F, Del VecchioBlanco C. Effect of liver cirrhosis and age on the glutathioneconcentration in the plasma, erythrocytes, and gastric mucosa ofman. Free Radic Biol Med 1996;20:483–8.

17. Crippin J. Acetaminophen hepatotoxicity: potentiation by isonia-zid. Am J Gastroenterol 1993;88:590–2.

18. Vogt BL, Richie JP. Fasting-induced depletion of glutathione in theaging mouse [Abstract]. Biochem Pharmacol 1993;46:257–63.

19. Weber H, Wegmann T, eds. Atlas der klinischen Enzymologie.Stuttgart: Georg Thieme Verlag, 1968:34–5.

20. Cavanagh K, Veenendaal M, Kapenga J. Subnormal serum creat-

ine kinase activity in a patient with acute myocardial infarction.Clin Chem 1987;33:1306–7.

21. De Scheerder I, Delanghe J, De Buyzere M, Hollanders G, ClementD, Leroux-Roels G. Low serum creatine kinase in patients withinfective endocarditis. Clin Chim Acta 1991;197:117–22.

22. Wei N, Pavlidis B, Tsokos G, Elin RJ, Plotz PH. Clinical significanceof low creatine phosphokinase values in patients with connectivetissue diseases. JAMA 1981;246:1921–3.

23. Frohlich J. Prolonged illness and creatine kinase activity [Letter].Clin Chem 1975;21:1343–4.

24. Barnert J, Behr W. Acute viral hepatitis and low values for serumcreatine kinase [Letter]. Clin Chem 1985;31:1909–10.

25. Rotenberg Z, Weinberger I, Fuchs J, Davidson E, Erdberg A,Sperling O, et al. Low serum creatine phosphokinase values inpatients with acute viral hepatitis. Am J Gastroenterol 1984;79:276–9.

26. Nanji AA, Blank D. Low serum creatine kinase activity in patientswith alcoholic liver disease [Letter]. Clin Chem 1981;27:1954.

27. Roberts R, Ishikawa Y. Enzymatic estimations of infarct sizeduring reperfusion. Circulation 1983;68(Suppl 1):83–9.

28. Delanghe J, De Buyzere M, Cluyse L, Thierens H, Clement D. Acutemyocardial infarct size and cumulative myoglobin release inserum. Eur J Clin Chem Clin Biochem 1992;30:823–30.

29. Burk R, Hill K. Reduced glutathione release into rat plasma byextrahepatic tissues. Am J Physiol 1995;269:396–9.

30. Michelutti L, Falter H, Certossi S, Marcotte B, Mazzuchin A.Isolation and purification of creatine kinase converting factor fromhuman serum and its identification as carboxypeptidase N. ClinBiochem 1987;20:21–9.

31. Hinderks G, Frohlich J. Low serum creatine kinase values associ-ated with the administration of steroids [Letter]. Clin Chem1979;25:2050–1.

32. Mueller K, Swanson R. Possible captopril inhibition of creatinekinase: a case report [Abstract]. Clin Chem 1985;31:941.

33. Fraser A. Chemotherapy as a cause of low serum creatine kinaseactivity [Letter]. Clin Chem 1980;26:1629–30.

34. Young DS, ed. Effects of drugs on clinical laboratory tests, 4th ed.Washington, DC: AACC Press, 1995:185–9.

Clinical Chemistry 44, No. 5, 1998 943