Elevation of Creatine Kinase BB CK in Hospitalized Patients

Transcription

1 ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 12, No. 5 Copyright 182, Institute for Clinical Science, Inc. Elevation of Creatine Kinase BB CK in Hospitalized Patients Importance of Distinguishing BB CK from MB CK FRED S. APPLE, Ph.D., NEIL S. GREENSPAN, M.D., Ph.D., and DAVID N. DIETZLER, Ph.D.* Division of Laboratory Medicine, Departments of Pathology and Medicine, Washington University School of Medicine St. Louis, MO ABSTRACT Nine hundred and forty patients (1023 samples) were analyzed by cellulose acetate electrophoresis for creatine kinase MB (MB CK) in the routine clinical chemistry laboratory during screening for acute myocardial infarction. Thirty patients showed the presence of BB CK and the associated disease states were identified. Rigid criteria were applied to demonstrate the activity observed was genuinely BB CK. It is noted and demonstrated that careful attention must be paid to the methodology used to distinguish CK isoenzymes, because some assays do not differentiate betw een BB CK and MB CK. It was shown by us that the false positive rate for MB CK could be as great as 1 percent in laboratories using non-discriminant assays for MB CK. Thus, the laboratory must know the limitations of the assays employed to avoid potential errors in MB CK results that could be misused as biochemical evidence of a myocardial infarction. Introduction There have been numerous reports concerning the presence of creatine kinase BB (BB CK) in the plasma of patients with a wide variety of disease states.3,5,,17,1-23 These observations raise an increasingly important concern for the specificity of techniques used in detecting MB CK and BB CK isoenzymes. Using electrophore sis to screen for the presence of MB CKas an aid in the diagnosis of acute myocardial infarction, separate studies have reported that a significant percentage of the samples screened showed the presence of BB CK (1 percent,11 2 percent,4 8 percent13). However, none of these three reports noted the disease states associated with the presence of BB CK in plasma. The results of these various studies raise several important unanswered questions. First, is the presence of BB CK in patients being screened for MB CK associated with a disease state known to show /82/ $00.0 Institute for Clinical Science, Inc. * To whom reprint requests should be sent at Division of Laboratory Medicine, Department of Pathology, Washington University School of Medicine, Box 8118, St. Louis, MO

2 B B C K ELEVATION IN HOSPITALIZED PATIENTS 3 BB CK in the serum? Second, do the percentages given for the occurrence of BB CK in the screened populations accurately reflect the percentage that will be seen in routine clinical chemistry laboratories in general? Third, what fraction of the screened samples that are positive for MB CK will also be positive for BB CK? Finally, are the BB CK isoenzymes observed on screening truly BB CK or artifacts? All of these questions are important to answer in light of the availability of quantitative methods that do not differentiate MB CK and BB CK in the course of biochemical screening for acute myocardial infarction. In the study reported here, 1023 samples from 40 patients were screened by electrophoresis for MB CK. O f these, 3.2 percent of the samples (3.2 percent of patients) showed the presence of BB CK. From the samples screened, 1.8 percent showed the presence of BB CK in the absence of MB CK, while 1.5 percent showed BB CK in the presence of MB CK. Rigid criteria were applied to demonstrate the activity observed was truly BB CK. The results of our findings are discussed in terms of the potential for false biochemical evidence for acute myocardial infarction (i.e., false positive results) in procedures that do not distinguish MB CK and BB CK. It is also demonstrated that one such procedure (DuPont ACA chromatographic method) shows false positive results. Materials and Methods During routine fractionation of CK isoenzymes,* serum CK isoenzymes were separated by cellulose acetate electrophoresis, f Samples for electrophoretic analysis were preserved by adding /6-mercaptoethanol (10 mm final concentration) to the sera. Samples were assayed at 37 C * Tests conducted in the clinical chemistry laboratory of Barnes Hospital, St. Louis, MO. f Equipm ent and reagents purchased from Gelman Sciences, Ann Arbor, MI. with and without substrate (creatine phosphate) to rule out non-ck artifacts 4 The isoenzymes were located by reduced nicotinamide-adenine dinucleotide phosphate (NADPH) fluorescence with a lower limit of sensitivity of 15 U per L.15 In order to show that MB CK was truly absent in samples showing BB CK without MB CK, another electrophoresis strip* was developed using reagents that allowed detection of MB CK as low as 5 U per L (established in our laboratory with purified human MB CK). The electrophoretic strips were interpreted for MB CK and BB CK as positive or negative. Both MB CK and BB CK were quantitated by a chromatographic? method.18 In addition, BB CK was quantitated by a radioimmunoassay that differentiates betw een MB and BB CK.16 Total creatine kinase activities were determ ined by the method of Rosalki.14 Results Over a period of two months, 1023 samples (40 patients) were screened for CK MB to aid the medical staff in the diagnosis of myocardial infarction. From these screenings, MB CK was observed in 124 samples (12.1 percent) while BB CK was observed in 33 samples (3.2 percent). Electrophoretograms were run with and without substrate to verify the activities of MB CK and BB CK. In no case did gels show any activity in the areas appropriate for CK isoenzymes MM, MB, or BB when developed in the absence of substrate. In all cases, the BB CK was clearly separated from albumin that appeared as a faint green band on the cathodal side of BB CK. In table I are shown BB CK fractions which appear alone in several non-cardiac, pathological conditions (18 patients) in addition to coincident appearance with positive MB CK fractions (12 patients). In all but one case, detection of BB CK correlates with clinical evidence of damage int Calbiochem, Lajolla, CA. DuPont ACA, Wilmington, DE.

3 4 0 0 APPLE, GREENSPAN, AND DIETZLER curred in organs enriched in BB CK. To our knowledge, a ruptured abdominal aortic aneurysm has never been previously reported to be associated with elevated BB CK in serum. In 13 patients with electrophoretic evidence of BB CK in the absence of MB CK, TABLE I Qualitative Analysis of BB CK and MB CK Isoenzymes* Clinical Diagnosis BB CK MB CK Reference for Prior BB Report Central Nervous System Intracerebral Positive Positive hemorrhage, chronic heart failure Stroke, myocardial Positive Positive infarct Neuroblastoma Positive Positive Gastrointestinal Rectal cancer Positive Negative 20 Abdominal aortic Positive Positive None aneurysm Ruptured abdominal Positive Negative None aneurysm Duodenal cancer, Positive Positive 10 ischemic heart disease Stomach cancer Positive Negative 12 Gastrectomy Positive Negative 8 Bowel resection Positive Negative 6 Renal Acute renal failure Positive Negative 1 Chronic renal Positive Negative 1 failure, congestive heart failure Renal failure, Positive Positive 1 ketoacidosis Renal failure, Positive Negative 1 diabetic ketoacidosis Respiratory Oat cell carcinoma, Positive Positive 2 congestive heart failure Oat cell carcinoma Positive Negative 2 Chronic obstructive Positive Positive pulmonary disease, cardiac irregularities Chronic obstructive Positive Negative pulmonary disease Lung cancer, cardiac Positive Positive 3, irregularities Lung cancer Positive Negative 3, Respiratory arrest, Positive Positive CPR Respiratory arrest, Positive Positive myocardial infarction Heart Cardiac arrest Positive Positive Cardiac arrest Positive Negative Cardiac arrest Positive Negative Angina Positive Negative Miscellaneous Hypotensive episode Positive Negative Ovarian cancer Positive Negative 20 Prostate cancer, Positive Negative 7,20 congestive heart failure Hard nodular prostate Positive Negative 7 *Number of patients studied was 30. enough serum remained after electrophoresis to assay these samples by either an MB CK ion exchange chromatographic procedure with the DuPont ACA, radioimmunoassay (RIA) procedure specific for BB CK, or both. The results are shown in table II. The RIA specific for BB CK detected BB in all nine of the samples tested. The DuPont ACA method showed measurable levels in all six samples tested. Two of the six levels were increased above the normal range established by DuPont (0 to 8 IU per L) and thus would be falsely interpreted as positive for MB CK. Discussion There have been reports of BB CK in the serum of patients hospitalized with a wide variety of pathological entities. D epending on the patient population screened and the methodology chosen to evaluate CK isoenzymes, a wide range of incidence of BB CK has been observed. Tsung1 reported a 17 percent incidence of BB CK in serum samples from a preselected patient population (mainly patients with trauma to organs known to contain BB CK). In a different selected patient population qf coronary care unit patients (a patient population that would be much less likely to have undergone trauma to organs enriched in BB CK), Delahunty and Foreback4 found only a 1.6 percent incidence of BB CK. In undefined patient populations, Obzansky and Lott13 reported an eight percent incidence, but Lamar11 reported only a one percent incidence of BB CK. Differences in the patient populations may have accounted for the wide divergence in incidence observed in these last two reports; in addition, a difference in the methodology used may have also played a role. Agarose gel electrophoresis used by Obzansky and Lott is known to have a high sensitivity (down to 5 U per L detectable15). However, Devine and M cdaniel5 used agarose and found only a 2.2 percent

4 incidence in a sleeted population (critical care unit patients). Thus, it is most likely that the patient population studied was the major factor in the high incidence reported by Obzansky and Lott. In the present study, a 3.2 percent incidence was found when using cellulose acetate in an unselected patient population that should accurately reflect the patient population screened in a routine clinical chemistry laboratory for evidence of myocardial infarction. For each BB CK observed, the patient s diagnosis was obtained from the chart. The majority of the pathological conditions noted were consistent with those reported in the literature (table I). It is of interest to note in the study of Devine and M cdaniel5 of patients from a critical care unit that the occurrence of BB CK was associated with a much narrower range of disease states. Precautions were taken to ensure the genuine nature of the BB CK isoenzyme. Positive BB CK samples observed by our electrophoretic separation distinctly migrated apart from albumin and associated artifacts. All strips contained a control with BB CK that migrated coincident with patient samples. In addition, a purified MB CK marker of 15 units was run to reassure us of the proper staining intensity of our fluorescent detection. Samples run without substrate showed no staining pattern coincident with BB CK. Furthermore, the BB CK specific radioimmunoassay showed substantial quantities of BB CK in all samples positive for BB CK by electrophoresis that were tested. All of these findings indicate that bona-fide BB CK was observed. In previous studies, only Devine and M cdaniel took any steps to ensure that the moiety observed was BB CK. These investigators processed the electrophoretograms in the presence and absence of the substrate creatine phosphate, but they did not confirm the BB CK by an immunological technique. Because some methodologies cannot discriminate BB CK from MB CK, the ap BB CK ELEVATION IN HOSPITALIZED PATIENTS pearance of BB CK activity in a clinical setting is important. Laboratories utilizing such methodologies may obtain false positive MB CK results owing to the presence of elevated BB isoenzyme. This was illustrated in the present study with the DuPont ACA method that only partially discriminates betw een MB CK and BB CK (table II). Two of the six samples with positive BB CK but no MB CK on electrophoresis showed MB CK levels on the DuPont ACA that would be interpreted as biochemical evidence consistent with a myocardial infarction. Moreover, of the 40 patients screened for MB CK in this study, 78 had a positive MB while 18 other patients had only a positive BB. Thus, in 1 percent (18 out of 6) of these patients, false positive results for MB CK could have been observed in methods that do not discriminate to any degree betw een MB and BB CK. In addition, 12 of the 78 positive for MB CK were also positive for BB CK. Thus, the false positive rate could be higher because the concommitant BB CK could elevate the measured MB CK from a negative to a positive level. Laboratories using non-discriminant assays for MB CK should be aware of the magnitude of these potential sources of error. TABLE I I Radioim m unoassay and C hrom atographic Evidence fo r BB CK* Total CK Activity (IU/L) MB CK by ACA (IU/L) BB CK by RIA (ng/ml) NA NA NA 24 2 NA 304 NA NA 8 72 NA 7 48 NA 2 37 NA NA 3 7 NA 5 *Total number of patients studied was 13. NA = Not assayed

5 402 A PPL E, GREENSPAN, AND D IE T ZLER In conclusion, a greater number of BB CK positive specimens from various pathological conditions are being observed in the clinical laboratory. The fact that a larger variety of hospitalized patients is being screened for the cardiac MB CK isoenzyme and that MB CK and BB CK are being observed both alone and in combination calls for careful attention to the methodology used to distinguish CK isoenzymes. The laboratory staff must know the limitations of the assays employed and take the responsibility to educate the medical staff of possible false positive MB CK results which could be misused for the diagnosis of a myocardial infarction. Acknowledgment Thanks are extended to Bill Schillizzi from the laboratory of Dr. Robert Roberts for performing the immunologic assays for BB CK and for the gift of purified human MB CK. Fred S. Apple was supported in part by NIH training grant ES References 1. CHUGA, D. J. and BACHNER, P.: Creatine kinase isoenzyme BB in the serum of renal-disease patients, distinct from an albumin-like artifact. Clin. Chem. 24:1286, COOLEN, R. B. and PRAGAY, D.: The production of brain-type creatine kinase in the serum of patients with oat-cell carcinoma. Clin. Chem. 22:1174, C o o l e n, R. B., P r a g a y, D. A., N o s a n c h u k, J. S., a n d BELDING, R.: E le v a tio n o f b ra in -ty p e c re a tin e k in a se in s e ru m fro m p a tie n ts w ith c arc in o m a. C a n c e r 44: , D e l a h u n t y, T. J. and F o r e b a c k, C. C.: Automated creatine kinase-mb estimation by immuno-inhibition: a clinical evaluation. Clin. Chem. 26: , D e v in e, J. E. and M c D a n i e l, R. C.: The detection of creatine kinase isoenzyme CKt in serum. Ann. Clin. Lab. Sci. :80-83, DORAN, G. R.: Appearance of creatine kinase BB isoenzyme in the serum of a patient suffering from infarction of the colon. Clin. Chim. Acta 2:415-41, FORMAN, D. T.: The significance of creatine kinase (CKBB) in metastatic cancer of the prostate. Ann. Clin. Lab. Sci. : , H o a g, G. N., F r a n k s, C. R., and D e C o t e a u, W. E.: Creatine kinase isoenzymes in serum of patients with cancer of various organs. Clin. Chem. 24:1654, I t a n o, M.: The detection of CPK (BB) in serum. Amer. J. Clin. Path. 65: , K a n e m i ts u, F., K a w a n is h i, I., and Mizu- SHIMA, J.: Three atypical serum creatine kinases in a patient with carcinoma of the colon. Clin. Chim. Acta 112: , L a m a r, W., W o o d a r d, L., and S t a t l a n d, B. E.: Clinical implications of creatine kinase-bb isoenzyme. New Eng. J. Med. 2: , L e d e r e r, ; #. ; H. and G e r s t b r e i n, H. L.: Creatine kinase isoenzyme BB activity in serum of a patient with gastric cancer. Clin. Chem. 22: , O b zan S K Y, D. and L o t t, J. A.: Clinical evaluation of an immunoinhibition procedure for creatine kinase-mb. Clin. Chem. 26: , R o s a l k i, S. B.: An improved procedure for serum creatine phosphokinase determination. J. Lab. Clin. Med. 6:66-701, S h e l l, W. E., K l i g e r m a n, M., R o r k e, M. P., and B u r h a n, M.: Sensitivity and specificity of M B creatine kinase activity determined with column chromatography. Amer. J. Cardiol. 44: 67-75, S c h i l l i z z i, W., R i t t e r, C. S., and R o b e r t s, R.: Radioimmunoassay to differentially assay plasma MB and BB CK. (Submitted for publication.) 17. S i l v e r m a n, L. M., D e r m e r, G. B., Z w e ig, M. H., V a n S t e i r t e g h e m, A. C., an d T o k e s, Z. A.: Creatine kinase BB: a new tumor-associated marker. Clin. Chem. 25: , S t o r e y, J. D., S a s s, M. L., G r a d y, H. J., B a i l l i e, E. E., J a x, S. M., and M o o r e, J. J.: An evaluation of a creatine kinase isoenzymes method for the DuPont ACA. DuPont Company, Wilmington, DE, August T s u n g, S. H.: Several conditions causing elevation of serum CK-MB and CK-BB. Amer. J. Clin. Path. 75: , Z w e ig, M. H. and V a n S t e i r t e g h e m, A. C.: Assessment by radioimmunoassay of serum creatine kinase BB (CK-BB) as a tumor marker: studies in patients with various cancers and a comparison of CK-BB concentrations to prostate acid phosphatase concentrations. J. Nat. Cancer Inst. 66: , Z w e ig, M. H., A d o r n a t o, B., V a n S t e i r t e g h e m, A. C., and E n g e l, W. K.: Serum creatine kinase BB and MM concentrations determined by radioimmunoassay in neuromuscular disorders. Ann. Neurol. 7: , Z w e ig, M. H., S i l v e r m a n, L. M., D e r m e r, G. B., and V a n S t e i r t e g h e m, A. C.: Serum creatine kinase isoenzyme BB as an indicator of active metastatic disease. Clin. Chem. 25: , Z w e i g, M. H. and V a n S t e i r t e g h e m, A. C.: Increased CK-BB concentrations in serum following cardiac surgery. Clin. Chim. Acta 3: 2-33, 17.