Immunochemical CK-MB Assay for Myocardial ...

6 downloads 0 Views 1MB Size Report
efficiency of an immunochemical assay for CK-MB (I-MB) was compared with ... rates for the E-MB test were 93.3% and 94.7% respectively. The test efficiency ...
Immunochemical CK-MB Assay for Myocardial Infarction MAJID ALI, M.D., SALVATORE LARAIA, M.D., RUDOLPH ANGELI, M.D., A. OLUSEGUN FAYEMI, M.D., EVALYNNE V. BRAUN, M.D., EDWARD DAVIS, M.D., AND PEGEEN H. PALLADINO, CLA(ASCP)

THE DIAGNOSTIC VALUE of enzyme profiles comprised of time-based studies of lactic dehydrogenase (LD) and creatine kinase (CK), and their isoenzyme for acute myocardial infarction is well established.1,3,419'20 The available technics for fractionation of creatine kinase include column chromatography,9,10,19,27 electrophoresis,3,4,19 immunoinhibition, 61617 and radioimmunoassay.7,14 A radiometric approach employing anti CK-MM and I125-labeled anti CK-BB has also been used.24 The advantages and shortcomings of these methods have been extensively reviewed." 13,15,28 The immunochemical technics are diagnostic tools of a high order of sensitivity and specificity. However, the two immunochemical methods for assay of CK-MB in current usage, immunoinhibition and radioimmuReceived June 12, 1981; received revised manuscript and accepted for publication September 10, 1981. Address reprint requests to Dr. Ali: Department of Pathology, Holy Name Hospital, 718 Teaneck Road, Teaneck, New Jersey 07666.

Divisions of Pathology, Immunology and Laboratories, Medicine, Cardiology and Radiology, Holy Name Hospital, Teaneck, New Jersey; Department of Pathology, College of Physicians and Surgeons of Columbia University and Department of Pathology, The Mount Sinai School of Medicine, New York, New York and Department of Pathology, Fairleigh Dickinson University School of Dentistry, Teaneck, New Jersey

noassay, have some disadvantages. These include incomplete inhibition of the M-subunit, interference by the BB isoenzyme, spurious elevation caused by adenylate kinase, and relatively low sensitivity of these methods for MB isoenzyme.6,7,14,16,17,30 Wicks and coworkers have recently introduced a different immunochemical approach which circumvents the interference caused by adenylate kinase and BB isoenzyme.29 The present study was undertaken to compare the diagnostic efficiency of this immunochemical (I-MB) with the electrophoretic method for CK-MB (E-MB) in the diagnosis of acute myocardial infarction. Materials and Methods Two hundred and eighteen consecutive patients in whom an ischemic myocardial injury was suspected and a cardiac isoenzyme profile was ordered were included in this study. During the study all isoenzyme profiles were initiated by the clinicians and the laboratory staff had no input in this decision. One hundred and thirty one patients were admitted to the coronary and intensive care units with the suspicion of acute myocardial infarction. Eighty seven patients developed chest pain during their hospitalization for both cardiac and noncardiac disorders. Three patients presented with a history of chest pain of several hours to three days duration and were subsequently diagnosed to have acute MI on the basis of positive EKG and LD, results. Negative results with both I-MB and E-MB procedures in these three patients were ascribed to the delay incurred between the onset of ischemic symptoms and drawing of blood sample (missed CK changes). For this reason, these patients were excluded from the study. The remaining 215 patients comprised the study group. The standard diagnostic profile for these patients included daily electrocardiograms and serial determina-

0002-9173/82/0500/0573 $00.85 © American Society of Clinical Pathologists

573

Downloaded from http://ajcp.oxfordjournals.org/ by guest on November 1, 2016

Ali, Majid, Laraia, Salvatore, Angeli, Rudolph, Fayemi, A. Olusegun, Braun, Evalynne V., Davis, Edward, and Palladino, Pegeen H.: Immunochemical CK-MB assay for myocardial infarction. Am J Clin Pathol 77: 1982; 573-579. The diagnostic efficiency of an immunochemical assay for CK-MB (I-MB) was compared with that of an electrophoretic procedure for this isoenzyme (E-MB) in 215 consecutive patients in whom acute myocardial infarction (MI) was clinically suspected. All patients were investigated with a standard protocol consisting of serial assays of total creatine kinase (CK) and lactic dehydrogenase (LD) activities, CK-MB and LD, isoenzymes, and electrocardiograms. Technetium pyrophosphate cardiac scans were obtained for eleven patients; autopsy was performed in six cases. The diagnosis of acute MI was established in 45 patients. The coefficients of variation for the intra-assay and interassay precision of I-MB assay ranged from 2.05%-7.2%. Concordance between the I-MB and E-MB results at the decision values for acute MI was observed in 203 of 215 patients (94.4%). At the cut-off point of 10 U/L, the I-MB assay was 95.5% sensitive and 95.3% specific for acute MI; the corresponding rates for the E-MB test were 93.3% and 94.7% respectively. The test efficiency rate for I-MB assay was 95.3% and that of E-MB procedure 94.4%. A complete time-based isoenzyme protocol for the diagnosis of acute myocardial infarction has several variables; the need for interpretation of individual profiles and for the preparation of a formal report is discussed. (Key words: Creatine kinase (CK); Lactic dehydrogenase (LD); Immunochemicals CK-MB (I-MB); Electrophoretic CK-MB (EMB); Immunochemical LD! (I-LD,); Myocardial infarction (MI); Cardiac myocytolysis)

574

ALI ET AL. Tube II

IB

Adenylate Kinase

BB Adenylate Kinase

Goat anti-CK-MM (Immunoinhibition by antibody excess)

Goat anti CK-MM Donkey anti-goat IgG bound to polyvinylidene fluoride flocules

FIG. 1. The principles of immunochemical CK-MB assay are schematically expressed. Interference by MM, BB, and adenylate kinase in I-MB assay is circumvented in the two steps shown in the diagram. The difference between the CK activity between tube 1 and 2 is a true measure of CK-MB.

tions of total CK and LD activities and CK and LD isoenzymes at the time of admission and on two successive mornings thereafter. For patients admitted between 1 A.M. and 8 A.M., the second sample was drawn at noon. For hospitalized patients, the first sample was drawn at the onset of chest symptoms. The EKGs were interpreted by a group of physicians, and were reviewed by one of two cardiologists, who were unaware of the results of the enzyme studies at the time.

Enzyme Assays Total CK and LD activities were determined with kinetic procedures using Beckman reagents and enzyme analyzer. Total LD activity was assayed using the lactate-pyruvate reaction; CK assays were performed by the method of Rosalki.21 The immunochemical LD, assays were performed as described previously using the antibodies and reagents purchased from the Roche Diagnostics, Inc.1,25 Briefly, in this procedure, goat antiLD5 serum is incubated with the patients serum for five minutes at ambient temperature to separate LD, from all LD isoenzymes containing the M subunit. In the second step, donkey anti-goat IgG conjugated to polyvinylidene fluoride floccules is added and incubated for five minutes at ambient temperature; the mixture is then centrifuged to remove LD isoenzymes containing the M subunit (LD2 to LD5). The analysis of the supernatant fluid for LD activity is, then, a measure of

LD, in the test sample. In a previous study of 100 consecutive patients suspected of acute myocardial ischemia, the LD, assay at the decision value of 90 U/L was found to be 100% sensitive and 89% specific.1 In this study the agreement between the results of LD, assay and electrophoretic LD procedure (reversed LD,/ LD2 ratio) was observed in 94% of patients. The LD, results in the present study were regarded as positive when the values were greater than over 95 U/L, equivocal between 85-95 U/L, and negative when the values fell below 85 U/L. The electrophoretic fractionation of CK was performed with cellulose acetate electrophoresis (Helena Laboratories, Inc., Beaumont, Texas) according to the manufacturer's directions. The CK electropherograms were examined under U.V. light and the results were recorded as CK-MB band present or absent; these results were validated by scanning the electrophoresis strips with Helena autoscanner-flur-vis densitometer. Immunochemical CK-MB Assay The principles and procedures of the immunochemical CK-MB assay (I-MB) are schematically expressed in Figure 1. For this assay goat anti-CK-MM, donkey anti-goat IgG bound to polyvinylidene fluoride flocules, and lyophilized CK control sera were kindly provided by Roche Diagnostics, Inc., Nutley, N. J. The details of production and characterization of these antibodies have been described elsewhere.24,29 The I-MB assay was performed as follows: 200 n\ of serum sample was incubated with 250 nl of goat antiCK-MM serum at ambient temperature for 20 minutes in tube #1. In tube #2, 200 n\ of serum sample was incubated with 50 ii\ of goat anti-CK-MM serum at ambient temperature for five minutes. In the second step, 200 nl of donkey anti-goat IgG serum is added, gently mixed, and incubated with the contents of tube #2 at ambient temperature for 5 min. This tube is then centrifuged for 5 minutes at 1000 X g at ambient temperature. The residual CK activity is then measured using a 1:11 sample to reagent ratio. The results are multiplied by 2.25 to correct for this dilution of serum with antibodies. The value for CK-MB is obtained by subtracting the result of tube #2 from that of tube #1 and multiplying it by a factor of 2. The CK electrophoresis were performed with a serum sample from a patient with acute MI before and after incubation with the antibody reagents of I-MB assay. Complete immunoinhibition of CK-MM was ascertained by observing an absence of the MM peak following treatment of the sample with goat anti-CK-MM antibody (Fig. 2, #2). Complete precipitation of MB was assured by the disappearance of MB peak following incubation of the

Downloaded from http://ajcp.oxfordjournals.org/ by guest on November 1, 2016

CK Tube I - CK Tube II = CK-MB

A.J.C.P. • May 1982

IMMUNOCHEMICAL CK-MB

Vol. 77 • No. 5

575

test serum with both antibodies and centrifugation (Fig. 2, #3). I-MB Cut-off The values for mean and SD for 100 consecutive I-MB assays for patients without acute MI were 3.14 U/L and 2.12 U/L respectively (Mean + 3 SD = 9.50). A cut-off point of 10 U/L for acute myocardial infarction for I-MB assay was selected in light of this statistical analysis. Diagnosis of Acute Myocardial Infarction

Results The data for within-batch and between-batch variability for the I-MB assay are given in Table 1. The intra-assay and inter-assay precision of this assay for patients with acute MI is good and compares favorably with other versions of enzyme immunoassays.1'2 The frequency distribution of I-MB assay results (peak values) for patients with and those without acute myocardial infarction is shown in Figure 3. In the MI group, two of 45 samples gave CK-MB values below the cutoff point of 10 U/L. In the non MI group, values of over 10 U/L were observed in eight of 170 patients. In two of these eight patients, the total CK activities were 5250

FIG. 2. CK electrophoresis performed on a serum sample from a patient with acute MI (1) is compared with electrophoresis performed on the same sample following incubation with anti CK-MM antibody (2) and following incubation with both anti CK-MM and the precipitating antibodies (3). Complete immunoinhibition of MM isoenzyme in the first step (2) and precipitation of MB in the second step (3) are assured. Electrophoretic mobility of CK-MB changes after it is linked with anti-MM.

Table I. The Data for Intra-assay and Inter-assay Variations for the Immunochemical CK-MB Assay for Sera from Two Patients with Acute MI and Three Reference Sera Sample Intra-assay*

Inter-assayt

* Mean of seven determinations.

Mean (Range) U / L

MI #1

99.3 (96.3-105.5)

MI #2

145.6 (138.1-155.9)

2SD

Coefficient of Variation 2.05%

12.7

4.37%

Reference Serum #1

9.9 (8.7-10.6)

1.02

5.6%

Reference Serum §2

95 (89-116)

13.68

7.2%

Reference Serum #3

103 (94-112)

10.68

5.2%

t Mean of 15 consecutive determinations.

Downloaded from http://ajcp.oxfordjournals.org/ by guest on November 1, 2016

The diagnosis of acute myocardial infarction was established on the basis of clinical course, I-LDi and EKG data. In addition, the results of technetium pyrophosphate (11 patients) and autopsy (six patients) were considered in formulation of the diagnosis. The rates of test efficiency, sensitivity and specificity for both I-MB and E-MB tests were computed by the method detailed by Galen and Gambino5 (See Appendix).

ALI ET AL.

576 210 190 170 150 130

11

=! ° 3

OO

90

I " O

50

3

30; ',

|

27

-

',

1 24 c 3

E E

21 18

15

9 6

'+:0W;;.L!

3 0 SERA from Patients With Ml (45)

SERA from Patients Without Ml (170)

* CK-MB Not Detectable FIG. 3. Peak I-MB assay values for 45 patients with acute MI and 170 patients without acute MI are shown. The interrupted line is drawn to indicate the cut-off point of 10 U / L .

and 850 U/L respectively while the total CK activities were over 300 U/L in another three patients. The diagnosis of acute myocardial infarction was established in 45 patients in this study on the basis of positive I-LD|, EKG, cardiac scan (three patients), and autopsy findings (two patients). The data for LD, and EKG for the 215 patients in this study are correlated with the I-MB and E-MB test results in Table 2. Agreement between the results of I-MB assay and E-MB procedure was observed in 203 of 215 patients (94.4%). At the cut-off points of 10 U/L, the I-MB assay was 95.5% sensitive and 95.3% specific for acute MI; the corresponding values for the E-MB procedure were

93.3% and 94.7% respectively. The test efficiency rate for I-MB assay was 95.3% and that of E-MB method 94.4%. In terms of decision values, the results of the I-MB assay did not agree with those of the I-LD, assay in 15 patients (LD] positive, MB-negative, 4; LD,-negative, MB-positive, 11). The analysis of this important subset of patients and the pathologic states diagnosed in them is given in Table 3. In one patient with LDrpositive and MB-negative results, the absence of myocardial infarction and the presence of generalized carcinomatosis was established at autopsy. We observed several interesting isoenzyme profiles in this study. Table 4 shows two such profiles. The profile of patient #1 illustrates the typical temporal relationships between the total CK, CK-MB, total LD and LD, values. It is noteworthy that the CK-MB peak was seen in the third sample (38 hours after onset of chest pain in this instance). A repeat isoenzyme profile initiated in the seventh hospital day to investigate a second episode of chest pain revealed no additional ischemic injury but the tail ends of the CK and LD curves following acute MI. Patient #2 was admitted to the CCU with chest pains of four hour duration and a strong clinical suspicion of acute MI. His first isoenzyme profile was interpreted as negative (I-LD, results were in equivocal range twice but were accompanied by falling total LD values). A second isoenzyme profile, initiated on the fourth hospital day, still showed low total CK values and negative CK-MB results but the LD| results were positive. The patient died on the seventh hospital day and at autopsy a massive MI was found. The myocardial infarct was histologically estimated to be 5-7 days old. Another atypical isoenzyme pattern seen in one patient in this study comprised of I-LD, positivity preceding that of CK-MB isoenzyme positivity. In seven other patients, CK-MB and LD, positivity appeared simultaneously after negative results for these isoenzymes had been seen in the first samples. The data concerning the time of appearance of peak I-MB and I-LD, assay results are given in Table 5.

Table 2. Correlation of Immunochemical MB (I-MB) and Electrophoretic MB (E-MB) Data with Immunochemical LD, (I-LD,) and EKG Results in 215 Consecutive Patients with Clinical Suspicion of Acute Myocardial Infarction No Myocardial Infarction (N = 170)

Acute Myocardial Infarction (N = 45) LD, Positive EKG Positive

LD, Positive EKG Negative

LD, Positive EKG Equivocal

LD, Negative EKG Positive

LD, Negative EKG Negative

LD, Negative EKG Equivocal

I-MB

26 2

5 145

3 15

E-MB

26 2

7 143

2 16

LD, Positive Autopsy Negative

Downloaded from http://ajcp.oxfordjournals.org/ by guest on November 1, 2016

12

A.J.C.P. • May 1982

IMMUNOCHEMICAL CK-MB

Vol. 77 • No. 5

The results of the technetium pyrophosphate cardiac scan were in agreement with the EKG and LD, data in ten of eleven patients (positive in three and negative in seven patients). The cardiac scan in the eleventh patient had been performed on the fourth day and was regarded as false negative. Discussion

a t

j

0 n

Table 3. Pathologic States Associated with Discordant I-LDi and I-MB Assay Results in 15 of 215 Patients with Clinical Suspicions of Acute Myocardial Infarction LD,-Positive, MB-negative —Acute Myocardial infarction (EKG+) —Carcinomatosis* —Septicemia* —Dissecting Aortic Aneurysm Total LD]-Negative, MB-Positive —Acute Myocardial Infarction (EKG+) —Cardiac Myocytolysis; diffuse* —Cardiac Arrythmias —Cardiac Arrest and Defibrillation —Severe Congestive Failure —Angina of Anemia —Septicemia* Total

11

* Acute MI ruled out at autopsy.

9..0..9.27

Antibodies raised against the two homodimers, MM and BB, do not cross-react.61416,31 Predictably, MB isoenzyme reacts with antibodies against both MM and BB fractions. This immunologic specificity of anti-MM and anti-BB antibodies provides the basis for the immunoinhibition, radioimmunoassay, and immunochemical technics for separation of CK isoenzymes. The problem of interference by adenylate kinase and BB isoenzyme, and difficulties of complete immunoinhibition of M-subunit have not been satisfactorily resolved in the immunoinhibition and radioimmunoassay procedures for CK-MB. 67141617 The sensitivity of the immunoinhibition technic in some studies has been as low as 70%.27 The radioimmunoassay using anti-BB antibody is beset by the serious problem of interference by BB isoenzyme because the antibody seems to recognize BB with greater facility than MB isoenzyme.24 The column chromatography procedures have their own

disadvantages: they are time-consuming; in some columns a sizable portion of BB isoenzyme is eluted along with MB, thus producing a false elevation of CK-MB results; in others 10-15% of CK-MB may be retained in the column creating the possibility of false negative results. In the immunochemical method developed by Wicks and co-workers,29 and used in the present study, interference from both adenylate kinase and BB isoenzyme is circumvented. This is made possible by immunologic inhibitions of MM and M subunit of MB with goat anti-MM antibody in one step of the assay, and precipitation of both MM and MB isoenzymes with goat ahti-MM and donkey anti-goat IgG bound to polyvinylidene fluoride floccules in the second step. The difference between the CK values in the two stages of the test, then, is a true estimate of CK-MB isoenzyme alone.

Table 4. Extended Cardiac Isoenzyme Profiles of Two Patients with Acute Myocardial Infarction. All Enzyme Levels Are Given in U/L Patient #1

Patient #2

* Not detectable, t Not performed.

Days Total CK 1-MB E-MB Total LD LD,

1

2

3

50 3.3 neg. 129 11

159 11.8 pos. 649 41

701 40.3 pos. 40 i 144

Days Total CK 1-MB E-MB Total LD LD,

1

2

3

33 * neg. 229 88

152 * neg. 206 88

163 1.7 neg. 174 70

4-6

7 127

t neg.

8 81

t

9 153

t

neg.

neg.

647 188

343 75

251 51

5 200 6.4

6 185 3:1

7tt

neg.

neg.

347 100

331 92

142 2.7 neg. 7614 2655

t t Death on 7th day. Massive myocardial infarct found at autopsy (estimated age 5-7 days).

Downloaded from http://ajcp.oxfordjournals.org/ by guest on November 1, 2016

The creatine kinase molecule is composed of two polypeptide chains: an M chain and a B chain. Of the three possible dimers, BB is the fastest anodally moving isoenzyme while MM is the slowest cathode fraction; the electrophoretic mobility of the hybrid dimer MB isoenzyme falls in between the two.30 The most widely used method for CK isoenzyme fractionation at present is based on this physical characteristic of CK isoenzymes.319 The structural differences between the three CK isoenzymes have also been utilized in developing column chromatographic technics for their fraction-

577

ALI ET AL.

578 Table 5. Occurrence of Peak I-MB and I-LD, Values in Cardiac Isoenzyme Protocol in 45 Patients with Acute MI I-MB Samples (days) Number of patients

14

#2 21

I-LD

#3 10

#4* —

#2 12

#3 25

* Samples were drawn in the morning of 4th hospital day as an extension of the formal 3-sample isoenzyme profile.

HOLY NAME HOSPITAL TEANECK, N.J.

CARDIAC ISOENZYME PROFILE M.D.I

ML LD,

LD

SAMPLE/ DATE

TOTAI U/L U/L

• 1/ */lt

ACUTE MI PATTERN

• O -

•o COMHENTi

231

CK

CK-MB

Ztl

FINAL DIAGNOSISi

ROOM li

D'AGATI

ACUTE MI PATTERN

.oo --

U/L

TOTAL U/L

t3 07 *¥.*///

77

Enzymatic evidence of acute myocardial infarction PR£S«fT ABSHft

The cardiac isoenzyme profile U inconclusive for acute M.I. CK-MB positivity in second sample and total CK and LD activities are indicative of an ischemic myocardial injury. The laboratory will extend the isoenzyme profile for an additional day to rule out a delayed positive LD. response.

J*£-

Mail Pathologist

FIG. 4. Format of a cardiac isoenzyme profile report.

small myocardial infarct in this setting. We were unable to find any study in English literature in which the absence of myocardial infarction in patients with positive CK-MB results (and negative LD, and EKG findings) was established with morphologic study. The diagnostic possibilities when LD, positivity is accompanied by negative CK-MB results are well recognized.3,4"13 Our patients falling in this category are listed in Table 3. False positive LD, results were seen in one patient with carcinomatosis and in another with dissecting aortic aneurysm. The abnormalities of LD and CK isoenzyme studies in two of our patients with septicemia are noteworthy; one patient had positive IMB and negative LDi results while the converse obtained in the second patient. The EKGs in both patients were negative for acute MI. In two patients with acute MI, negative I-MB assay results were regarded as false negative. The total CK activities have been used to estimate the infarct size.22 The extended cardiac enzyme profile of patient #2 given in Table 4 is noteworthy in this context. The total CK activities in this patient were only moderately increased and CK-MB results were negative with both I-MB and E-MB technics on all six occasions. Yet this patient at autopsy was found to have massive myocardial infarction. This case study belies the common assumption that total CK and CK-MB levels in serum reflect the extent of myocardial necrosis. It seems likely that some other factors, such as inactivation of enzymes in the cardiac interstitium, may profoundly influence the levels of these enzymes in blood in some instances. The results of this study emphasize the need for a formal interpretation of individual cardiac isoenzyme profiles in clinical practice. Clearly, the highest level of diagnostic accuracy can only be achieved by interpretation of both the CK-MB and LD, results in the context of total CK and LD activities as well as the clinical setting. In our laboratory, an extra margin of safety is assured by creating an equivocal zone in reporting both the I-LD] and I-MB data. Thus, the equivocal zone for I-LD, extends from 85 to 95 U/L and that for I-MB assay from 9 to 12 U/L. At the completion of the cardiac isoenzyme profile, each profile is interpreted by a pathologist and a final report prepared (Fig. 4). Not infrequently, we have observed rising total CK levels and CK-MB positivity unaccompanied by LD, positivity in the three timed samples in the formal isoenzyme protocol. We extend the protocol in these instances for a day, and in a rare instance for two more days, and are sometimes able to observe positive enzymatic evidence for myocardial infarction in some patients. In other instances persistently elevated and nonfluctuating total LD activities provided us with the first clue that the positive I-LDi results may not be truly

Downloaded from http://ajcp.oxfordjournals.org/ by guest on November 1, 2016

The CK-MB is generally believed to be the most sensitive serum marker of ischemic myocardial injury, with or without progression to acute myocardial infarction.4'81218,2326 The results of the present study support this viewpoint. The significance of CK-MB positivity in patients with negative LD, and EKG findings has not been fully elucidated. There were 11 such patients in our study (Table 3). In each instance there were one or more pathologic states which would be expected to inflict ischemic myocardial injury at a cellular level (cardiac arrest, arrythmias, congestive heart failure, and septicemia). In one such patient diffuse cardiac myocytolysis (cellular injury with loss of individual myocardial fibers) was observed at autopsy but there was no evidence of acute myocardial infarction. A more difficult problem is how to exclude the presence of a

A.J.C.P. • May 1982

Vol. 77 • No. 5

IMMUNOCHEMICAL CK-MB

Acknowledgment. We are grateful to Joyce Yondolino for preparation of the manuscript and to Joanne Clark, Director of Medical Records, for assistance in data collection.

References 1. Ali M, Braun EV, Laraia S, Fayemi AO, Nalebuff DJ, Palladino PH: Immunochemical LD, assay for myocardial infarction. Am J Clin Pathol, 1981; 76:426-429 2. Ali M, Nalebuff DJ, Ramanarayanan R, Fayemi AO, Mesa-Tejada R: Assay of IgE antibodies against June and Timothy grasses by an immunoperoxidase technique. Clin Allergy 1980; 10:203-209 3. Galen RS, Reiffel JA, Gambino R: Diagnosis of acute myocardial infarction. JAMA 1975; 232:145-147 4. Galen RS: The enzyme diagnosis of myocardial infarction. Hum Pathol 1975; 6:141-155 5. Galen RS, Gambino SR: Beyond Normality: The Predictive Value and Efficiency of Medical Diagnosis. New York, John Wiley & Sons, 1975, pp 13, 33 6. Gerhardt W, Ljungdahl L, Borjesson J, et al: Creatine kinase Bsubunit activity in human serum I. Development of an immunoinhibition method for routine determinations of S-creatine kinase B-subunit activity. Clin Chim Acta 1977; 78:2941 7. Homburger HA, Jacob GL: Creatine kinase radioimmunoassay and isoenzyme electrophoresis compared in the diagnosis of acute myocardial infarction. Clin Chem 1980; 26:861-866 8. Gutovitz AL, Sobel BE, Roberts R: Progressive nature of myocardial injury in selected patients with cardiogenic shock. Am J Cardiol 1978;41:469-475 9. Klein B, Foreman JA, Jeunelot CL, Sheehan JE: Separation of serum creatine kinase isoenzymes by ion-exchange column chromatography. Clin Chem 1977; 23:504-510 10. Kraft J, Aastrup H, Schroder P: Diagnostic value for acute myocardial infarction of creatine kinase and lactate dehydrogenase isoenzymes compared with total enzymes. Acta Med Scand 1978; 203:167-174 11. Lott JA, Stang JM: Serum enzymes and isoenzymes in the diagnosis and differential diagnosis of myocardial ischemia and necrosis. Clin Chem 1980; 26:1241-1250 12. Mercer DW, Varat MA: Detection of cardiac-specific creatine kinase isoenzyme in sera with normal or slightly increased total creatine kinase activity. Clin Chem 1975; 21:1088-1092 13. Morin LG: Evaluation of current methods for creatine kinase isoenzyme fractionation. Clin Chem 1977; 23:205-210

APPENDIX

T E S T S E N S I T I V I T Y (positivity in disease)

TP

= TPT^X100 T E S T S P E C I F I C I T Y (negativity in h e a l t h )

TN =

„,

TP + TN

TT7TFP-X100

TEST EFFICIENCY = — X GRAND TOTAL

100

T P , T r u e positive; F N , False negative; T N , T r u e negative; F P , False positive.

Downloaded from http://ajcp.oxfordjournals.org/ by guest on November 1, 2016

indicative of acute MI. The absence of CK changes reinforced our suspicion of an extra-cardiac source of LDi in these cases. The final diagnosis in one such patient in this study, and two patients seen by us previously, proved to be disseminated carcinomatosis. The I-MB assay results may show spurious elevation when the total CK activity is very high. One of our patients gave a false positive I-MB result for this reason (I-MB 15.6 U/L, total CK 5250 U/L). This problem is not confined to the immunochemical methods and has been encountered with both the column and electrophoretic technics. To circumvent this source of error, it has been proposed that CK-MB value be expressed as a percentage of the total CK activity. We believe inclusion of such a percentage figure in routine reports will further add to the complexity of this issue. This difficulty can be best resolved, in the few instances where it arises, by a succinct remark to this effect by the pathologist interpreting the profile.

579

14. Morin LG: Creatine kinase isoenzyme-antibody reactions in immuno-inhibition and immunonephelometry. Clin Chem 1979; 25:1415-1419 15. Navin TR, Hager DW: Creatine kinase MB isoenzyme in the evaluation of myocardial infarction. Curr Probl Cardiol 1979; 3:7-32 16. Neumeier D, Prellwitz W, Wurzburg U: Determination of creatine kinase isoenzyme MB activity in serum using immunological inhibition of creatine kinase M subunit activity. Clin Chim Acta 1976; 73:445-451 17. Obzansky D, Lott JA: Clinical evaluation of an immunoinhibition procedure for creatine kinase MB. Clin Chem 1980; 26:150152 18. Reiffel JA, McCarthy DM, Leahey EB, Jr: Does DC cardioversion affect isoenzyme recognition of myocardial infarction? Am Heart J 1979; 97:810-811 19. Roark SF, Wagner GS, Izlar HL, Roe CR: Diagnosis of acute myocardial infarction in a community hospital: Significance of CPK-MB determination. Circulation 1976; 53:965-969 20. Roberts R, Gowda KS, Ludbrook PA, Sobel BE: Specificity of elevated serum MB creatine phosphokinase activity in the diagnosis of acute myocardial infarction. Am J Cardiol 1975; 36:433-437 21. Rosalki SB: An improved procedure for serum creatine phosphokinase determination. J Lab Clin Med 1967; 69:696-705 22. Sobel BE, Bresnahan GF, Shell WE, et al: Estimation of infarct size in man and its relation to prognosis. Circulation 1972; 46:640-648 23. Strauss HD, Roberts R: Plasma creatine kinase activity and other conventional enzymes. Arch Intern Med 1980; 140:336-339 24. Usategui-Gomez M, Wicks RW, Farrenkopf B, Hager H, Warshaw, M: Immunochemical determination of CK-MB isoenzyme in human serum: A radiometric approach. Clin Chem. 1981; 27:823-827 25. Usategui-Gomez M, Wicks RW, Warshaw M: Immunochemical determination of the heart isoenzyme of lactate dehydrogenase (LDH,) in human serum. Clin Chem 1979; 25:729-734 26. Varat MA, Mercer DW: Cardiac specific creatine phosphokinase isoenzyme in the diagnosis of acute myocardial infarction. Circulation 1975; 51:855-859 27. Von Arnim T, Neumeier D, Sandel P: Diagnostische Wertigkeit der CK-MB-Bestimmung bei Patienten mit Verdacht auf Myokardinfarkt. Verh Dtsch Ges Inn Med 1978; 84:1408-1411 28. Wagner GS: Optimal use of serum enzyme levels in the diagnosis of acute myocardial infarction. Arch Intern Med 1980; 140:317319 29. Wicks R, Usategui M, Miller M, Warshaw M: Immunochemical determination of CK-MB isoenzyme in human serum II: An enzymatic approach. Clin Chem 1982; 28:54-58 30. Wilkinson JH: Isoenzymes. Philadelphia, JB Lippincott, 1970, pp 72-75 31. Zweig MH, Van Steirteghem AC, Schechter AN: Radioimmunoassay of creatine kinase isoenzymes in human serum: Isoenzyme BB. Clin Chem 1978; 24:422-428