|Click here for information on how to order reprints of this article.|
Comparison of CK-MB Subforms and Troponin I in the
Evaluation of Patients with Acute Coronary Syndromes
Vikramaditya Dulam, MD
Simon Chakko, MD
Andrew Quartin, MD
Jose A. Gueton, MT
Raphael Valenzuela, MD, PhD
Robert J. Myerburg, MD
Medicine and Laboratory Services
Miami VA Medical Center and
the University of Miami School of Medicine
KEY WORDS: creatine kinase, troponin, acute coronary syndromes
Background: Among the various markers of myocardial injury, CK-MB subforms MB2, MB1 and their ratio (CKMBS) have been reported to be the most sensitive early marker of acute myocardial infarction.
Hypothesis: The aim of this study was to determine if CKMBS is superior to troponin I in the prediction of major cardiac events in patients with acute coronary syndromes.
Methods: The prognostic values of CKMBS and troponin I were compared in 100 consecutive patients with acute coronary syndromes. CKMBS and troponin I were measured on arrival to the emergency room and if the duration of chest pain was < 8 hours, a second measurement was done 8 hours later. In addition, CK-MB was measured every 8 hours for 24 hours. Physicians were blinded to the Troponin I data. Patients were followed until their discharge from the hospital for the occurrence of major cardiac events (myocardial infarction, coronary angioplasty, coronary bypass surgery, or death). Results: Major cardiac events occurred in 24 patients. The sensitivities of CK-MB, CKMBS, and troponin I in the prediction of major cardiac events were 58%, 62%, and 87%, and specificities were 88%, 73%, and 88%, respectively. By multiple logistic regression analysis, troponin I (chi-squared 26.2, odds ratio 54.9) and CKMBS (chi-squared 6.4, odds ratio 6.59) were independent predictors of major cardiac events. In the whole group, troponin I was better than CKMBS in the prediction of the occurrence or absence of cardiac events (88% versus 73%, P = .009, odds ratio 2.67 + 1.48). In the 77 patients with normal CK-MB, there were 9 with elevated troponin I and 21 with elevated CKMBS. Among these 77 patients, troponin I was better than CKMBS in the prediction of the occurrence or absence of cardiac events (72% versus 57%, P = .0027, odds ratio 4.0 + 1.65). However, among the 23 patients with elevated CK-MB, troponin was not better than CKMBS. There was a significant increase in the incidence of cardiac events with increasing levels of troponin I (chi-squared 49.79, P < .0001).
Conclusions: For the prediction of major cardiac events in acute coronary syndromes, troponin I is superior to CKMBS, especially in patients with normal CK-MB.
Cardiac troponin I and creatine kinase MB subforms (MB1, MB2, and their ratio) are laboratory tests that have improved the diagnostic accuracy of myocardial infarction.1,2 A comparison of the various diagnostic markers of myocardial injury (e.g., CK-MB subforms, myoglobin, total CK-MB [activity and mass], troponin T, troponin I) demonstrated that CK-MB subforms assay is the most reliable test at 6 hours after the onset of myocardial infarction.3 Troponin I was found to have very high sensitivity and specificity 10 hours after the onset of chest pain.
In addition to the diagnosis of myocardial infarction, troponin I and T have also been reported to be useful in the risk stratification of patients with acute coronary syndromes.4-6 In the Global Use of Strategies to Open Occluded Coronary Arteries in Acute Coronary Syndromes (GUSTO-II a) trial and the Thrombolysis in Myocardial Ischemia Phase III B (TIMI III B) trials, troponin T and I were found to be independent predictors of early mortality.4,5 Although CK-MB subforms are superior to troponins in the early diagnosis of myocardial infarction,3 it is not clear whether they are superior in the risk stratification of patients with acute coronary syndromes. The purpose of this study was to prospectively compare the prognostic value of CK-MB subforms and troponin I in patients with acute ischemic syndromes admitted to a coronary care unit.
An observational study was performed in 119 patients with chest pain admitted to the Coronary Care or Telemetry Units of the Miami V.A. Medical Center. These patients were evaluated in the emergency room by a physician and referred to cardiology for admission with cardiac monitoring. Nineteen patients were excluded due to incomplete data collection. To be eligible, duration of chest pain was >15 minutes and suspected to be of myocardial ischemia in origin. Blood samples were drawn for total CK and CK-MB according to a "rule out myocardial infarction" protocol on arrival and every 8 hours for 24 hours. In addition, blood specimens were obtained for CK-MB subforms and troponin I on arrival to the emergency room and if the duration of chest pain was less than 8 hours, a second specimen was drawn 8 hours later.
Patients were treated by an attending cardiologist and house staff. Although troponin I was measured with other cardiac enzymes, the results of the troponin I were not released. Thus the treating physicians were blinded to the troponin I results. This study was approved by the Human Studies Subcommittee (IRB) of the VA Medical Center, which granted permission to the investigators to obtain the results of troponin I and correlate them with clinical outcomes. Other enzyme results were available to the physicians in the usual manner. A diagnosis of myocardial infarction was made if the total CK-MB was >10 units, with the presence of acute ST segment elevation or development of new pathologic Q waves on the ECG. Patients were treated in the usual manner and the decision to perform noninvasive or invasive cardiac procedures was made on clinical grounds by the attending cardiologist. The patients were followed until their discharge from the hospital. Major cardiac events such as death, myocardial infarction, coronary angioplasty, and coronary artery bypass grafting were recorded.
Total CK-MB was chemically quantitated in a Vitros 950 instrument (Johnson & Johnson, Rochester, New York) and confirmed by low-voltage gel electrophoresis using standard protocol, reagents, and equipment (Beckman, Brea, California). Blood for CK-MB subforms was collected on EDTA-containing tubes and was transported to the laboratory on ice. Analysis was completed within 1 hour to prevent convection of subforms. Both CK-MB 1 and 2 were simultaneously quantitated by automated, high-voltage, agarose gel electrophoresis using reagent kits and a fully computerized electrophoresis-densitometry unit (Cardio Rep) following manufacturers' instructions (Helena Laboratories, Beaumont, Texas). Presence of CK-MB2 >2.0 IU/L and MB2/MB1 ratio >1.7 was considered abnormal. Troponin I measurements were made using automated fluorometric microparticle enzyme immunoassay (Abbott Diagnostics, Abbott Park, Illinois). Troponin I level >2.0 ng was considered abnormal.
Statistical analysis was performed using commercially available software. Sensitivity, specificity, predictive values, and confidence intervals were calculated by previously described methods. Differences in proportions were tested with c2 and Fisher's exact test. Strength of association of enzyme markers to outcome was assessed using logistic regression, and multiple logistic regression with stepwise selection was used to generate parsimonious models. Relative utility of enzyme markers in the prediction of cardiac events was compared using McNemar's test. A value of P < .05 was considered significant.
The study cohort consisted of 100 patients; 19 patients were excluded due to incomplete data collection or lack of follow-up information. There were 95 men and 5 women with a mean age of 63.4 + 12.1 years. Seventy were white and 30 black. History of hypertension was present in 71%, diabetes mellitus in 37%, hyperlipidemia in 68%, past or present cigarette smoking in 86%, and alcohol use in 31%. A prior history of myocardial infarction was present in 38%, angina pectoris in 60%, abnormal stress test in 16%, coronary angiography in 48%, coronary angioplasty in 15%, and coronary bypass surgery in 20%.
CK-MB was abnormal in 23 patients, troponin I in 30 and CK-MB subforms in 35. Among the 23 patients with abnormal CK-MB, 21 had abnormal troponin I levels (sensitivity 96%) and 14 had abnormal CK-MB subforms (sensitivity 61%). During the hospitalization, echocardiogram was performed in 93 patients, stress thallium test in 54 patients, and cardiac catheterization in 48 patients.
Major cardiac events (death, coronary angioplasty or coronary artery bypass surgery) occurred in 24 patients. There were 9 acute myocardial infarctions, 14 coronary angioplasties, 4 coronary artery bypass graft surgeries, and 4 deaths.
The role of various cardiac markers to predict these events was evaluated. The sensitivity, specificity, positive predictive value, and negative predictive value of CK-MB, MB subforms, and troponin I are shown Table 1. Troponin I had better sensitivity than CK-MB and CK-MB subforms. CK-MB and troponin I had better specificity than CK-MB subforms.
Among the 77 patients with normal CK-MB, there were 9 patients with elevated troponin I and 21 patients with elevated CK-MB subforms. Major cardiac events occurred in 8 (88.8%) patients with troponin elevation. Such events occurred in 6 (28.6%) patients with elevated CK-MB subforms. Among the patients with normal CK-MB, the risk of a major cardiac event was higher with elevated troponin I compared with elevated CK-MB subforms (P = .0043; odds ratio: 20.0; confidence interval: 2.036 to 196.49).
Results of troponin I and CK-MB subforms analysis were discordant in 33 (33%). In 19 patients, CK-MB subforms were abnormal but the troponin I was normal; in 14 patients, troponin I was elevated but CK-MB subforms were normal. Cardiac events occurred in 3 of 19 patients (15.8%) with abnormal CK-MB subforms and normal troponin I and in 8 of 14 patients (57.1%) with elevated troponin I and normal CK-MB subforms. The risk of having a major cardiac event was lower with elevated CKMBS compared with elevated troponin I (P = .0240; relative risk: 0.276; confidence interval: 0.088 to 0.085).
In many clinical settings, multiple enzymatic markers are often used to evaluate myocardial necrosis. By multiple logistic regression analysis, troponin I (chi-squared 26.2, odds ratio 54.9) and CKMBS (Chi square 6.4, odds ratio 6.59) were independent predictors of major cardiac events (Table 2). In addition, the ability of troponin I and CKMBS to correctly classify patients for the presence or absence of cardiac events was evaluated using the McNemar test. Classification was considered to be correct if the test predicted no event and none occurred or if it predicted an event and one did occur. In the whole group, troponin correctly predicted the occurrence or absence of a cardiac event in 88 (88%) and CKMBS in 73 (73%). Troponin was a better predictor of cardiac events (P = .0090, odds ratio 2.67+1.48). In the 77 patients with normal CK-MB, troponin I correctly classified 72 patients and CKMBS, 57 patients. In patients with normal CK-MB, troponin I was better than CKMBS in predicting the occurrence or absence of cardiac events (72% versus 57%, P = .0027, odds ratio 4.0 + 1.65). Among the 23 patients with elevated CK-MB, troponin correctly classified 16 patients and CKMBS in 16 patients (P = 1.0, odds ratio 1.0). Thus among all patients with acute coronary syndromes and in the subgroup with normal CK-MB, troponin I was a better prognostic test than CKMBS. In patients with abnormal CK-MB, troponin I or CKMBS did not provide additional prognostic information.
The correlation between the level of troponin I and the incidence of cardiac events was evaluated. Cardiac events occurred in 2 (3.5%) of the 56 patients with a troponin I level of < 0.5 ng/mL, 1 (7.1%) of 14 patients with a level of 0.6 to 2 ng/mL, 11 (64.7%) of 17 with a level of 2 to 20 ng/mL, and 10 (77%) of 13 patients with a level > 20 ng/mL (Figure 1). There was a significant difference in the incidence of cardiac events in the groups of patients with troponin I levels < 0.5, 0.6 to 2, and >2 ng/ml (chi-squared 49.79, P < .0001). Using logistic regression, significant correlation between the level of troponin I and cardiac events was also found (P = .027)
Acute coronary syndromes are caused by the rupture of an atherosclerotic plaque in the
coronary artery that impairs or interrupts the myocardial perfusion. The clinical
presentation varies from unstable angina with no electrocardiographic changes to Q-
wave myocardial infarction. Each year, approximately 2.5 million people presenting to
emergency departments in the United States are hospitalized with the diagnosis of acute coronary syndromes: 1.5 million have a final diagnosis of unstable angina, whereas non-ST segment elevation and ST segment elevation myocardial infarction account for the remaining 1 million.7 Presence of certain clinical markers (prolonged chest pain,
third heart sound, heart failure) and electrocardiographic markers (ST segment and T-
wave changes) are predictors of poor prognosis.8 Recent studies have shown that enzymatic markers of myocardial necrosis have prognostic value in patients with unstable angina.
CK-MB has been the traditional marker of myocardial necrosis. In the initial hours after myocardial necrosis, the amount of CK-MB released from the myocardium is minimal. The myocardial form of CK-MB is CK-MB2, which on its release into blood is cleaved by lysine carboxypeptidase into CK-MB1.9 Normally, CK-MB2 and CK-MB1 are in equilibrium. The elevation of the serum level of CK-MB2 and the ratio of CK-MB2 to CK-MB1 have better sensitivity and specificity than the traditional CK-MB measurement during the first 6 hours of myocardial infarction.2,3 Troponin C, I and T are proteins that regulate the calcium-dependent interactions between actin and myosin during the cardiac contraction and relaxation. The amino acid sequence of troponin I and T in the cardiac muscle differ from that of skeletal muscle. This has allowed the development of immunoassays that are specific for cardiac troponins.10 They are very sensitive and specific markers of myocardial necrosis.11 The initial rise of troponin occurs approximately 3 hours after myocardial injury but in some patients there is a delay of several hours.12 Thus at least two measurements, one of which is at least 6 hours after the onset of symptoms, are recommended.
Recent studies of relatively small numbers of patients with acute coronary syndromes reported that the measurement of cardiac troponins have prognostic significance.13,14 Retrospective analysis of patients who were enrolled in large multi-center studies evaluating various therapies for acute coronary syndromes confirmed this finding.4-6 In the GUSTO IIa trial, elevated cardiac troponin T was an independent predictor of 30-day mortality and was superior to electrocardiographic abnormalities and CK-MB measurement in risk stratification.4 Troponin measurement was useful in patients with ST segment elevation, depression, or normal electrocardiograms. In this study, only a single troponin T measurement was made 2 hours after hospital admission. Among the patients enrolled in the TIMI IIIB trial, those with elevated troponin I had a higher 42-day mortality.5 Troponin I was an independent predictor of mortality after adjustment for other baseline variables such as ST-segment depression and age. These studies convincingly demonstrate that cardiac troponin can be used in the early risk stratification of patients with acute coronary syndromes.10 Troponin I level, measured on admission, can be used to identify patients who could benefit from revascularization. In one study, patients with acute coronary syndromes and elevated troponin I levels who undergo early catheter-based revascularization have more favorable long-term clinical outcomes.15
The CK-MB subforms and troponin I are more sensitive for the detection of myocardial necrosis. Among patients presenting to the emergency department with chest pain, a direct prospective comparison of CK-MB subforms, myoglobin, troponin T, troponin I, and CK-MB revealed that CK-MB subforms were the most sensitive and specific in the early diagnosis of myocardial infarction.3 The present study confirmed that troponin I is a very useful predictor of prognosis in acute coronary syndromes. The purpose of the study was to compare troponin I and CK-MB subforms in that role. To the best of our knowledge there is no reported study evaluating the role of CK-MB subforms in the risk stratification of patients with acute coronary syndromes. In the present study, both troponin I and CK-MB subforms were superior to CK-MB in the risk stratification of patients with acute coronary syndromes. However, troponin I was a better prognostic indicator than CK-MB subforms. It should be noted that the treating physicians were blinded to the troponin I results and their decision making was not influenced by the result of troponin I.
A significant number of patients with normal CK-MB had elevated CK-MB subforms or troponin I. Similar findings were reported in another study in which 22 of the 91 patients hospitalized for unstable angina were found to have elevated troponin I and normal CK-MB.16 These patients with normal CK and CK-MB and abnormal troponin I were found to have a higher incidence of death and myocardial infarction. In addition, only 68% of these patients with elevated troponin I were free of cardiac events at one year.16 Patients with elevated troponin T and normal CK-MB in the GUSTO IIa study4 were also found to have significantly higher mortality and increased rates of congestive heart failure and cardiogenic shock.
Results of CK-MBS and troponin I were discordant in 33% patients. Such discordance has been reported in other studies. In the TIMI IIIB trial,5 1401 patients with acute coronary syndromes were studied. Among the 453 patients with elevated CK-MB, troponin I was normal in 118; among the 573 with elevated troponin I (> 0.4 ng), 238 had normal CK-MB. The reason for discordant results of troponin and CK-MB and CK-MB subforms may be differences in sensitivities and the duration of enzyme elevation. Elevations in troponin I are detectable in the plasma for up to 7 to 10 days after myocardial necrosis, but CK-MB and CK-MB subforms return to normal in 2 to 3 days.12 Thus it is possible that some of the patients with elevated troponin I had suffered myocardial injury more than 3 days prior to their hospitalization. The exact mechanism of discordant results between highly sensitive cardiac enzymes is not understood.17
The positive predictive value of cardiac enzymes for major cardiac events is high in this study since our hospital is a referral center and only patients who were admitted to the coronary care unit were included. When troponin I is used a screening test in the emergency department, the positive predictive value is lower. In one study,18 troponin I in the first sample of blood obtained from patients presenting with chest pain to the emergency department had a sensitivity of only 17% to predict a major cardiac event. However, the specificity was 91%, similar to the specificity reported in the present study.
Limitations of the study include a relatively small number of patients and the use of some "soft" endpoints, such as coronary angioplasty and bypass surgery. Patients were treated by clinical staff who were unaware of the troponin I results. However, they were not blinded to the CK-MB subforms results, which may have confounded the outcomes. The upper limit for normal troponin was 2 ng/mL in the present study. The two cut-off values that could be used are at the lower limit of detectability (0.5 ng/mL) and the manufacturer's suggested upper limits of normal (2.0 ng/mL). The latter was used in this study since it has better specificity.19 In the TIMI IIIB trial5 the risk for mortality rose progressively with increasing troponin levels in patients with acute coronary syndromes. Each increase of 1 ng/mL in the cardiac troponin I level was associated with a significant increase in the risk ratio for death after adjustment for the baseline characteristics. In the present study also, the risk of cardiac events increased with increasing levels of troponin I. When compared to using different levels of troponin I as an independent variable, the predictive value was better when patients were grouped by normal and abnormal values of troponin I. This is probably due to the fact that very high values of troponin I may indicate reperfusion, a wash out phenomenon, rather than large myocardial damage.
Both CK-MB subforms and troponin I are independent predictors of major cardiac events in patients with acute coronary syndromes. Troponin I is superior to CK-MB subforms for the early risk stratification of acute coronary syndromes.
1. Adams JE III, Schechtman KB, Landt Y, et al: Comparable detection of acute myocardial infarction by creatine kinase MB isoenzyme and cardiac troponin I. Clin Chem 40:1291-1295, 1994.
2. Puleo PR, Meyer D, Wathen C, et al: Use of a rapid assay of subforms of creatine kinase MB to diagnose or rule out acute myocardial infarction. N Engl J Med 331:561-566, 1994.
3. Zimmerman J, Fromm R, Meyer D, et al: Diagnostic marker cooperative study for the diagnosis of myocardial infarction. Circulation 99:1671-1677, 1999.
4. Ohman EM, Armstrong PW, Christenson RH, et al for the GUSTO-IIa Investigators: Cardiac troponin T levels for the risk stratification in acute myocardial ischemia. N Engl J Med 335:1333-1341, 1996.
5. Antman EM, Tanasijevic MJ, Thompson B, et al: Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 335:1342-1349, 1996.
6. Lindahl B, Venge P, Wallentin L: Relation between troponin T and the risk of subsequent cardiac events in unstable coronary artery disease. Circulation 93;1651-1657, 1996.
7. Cohen M, Ferguson JJ, Harrington RA: Trials of glycoprotein II b-IIIa inhibitors in non-ST segment elevation acute coronary syndromes: Applicability to the practice of medicine in the United States. Clin Cardiol 22(V1)2-12, 1999.
8. Braunwald E, Jones RH, mark DB, et al: Diagnosing and managing unstable angina. Circulation 90:613-622, 1994.
9. Wevers RA, Desling M, Klein Gebbink JAG, Soons JBJ: Post-synthetic changes in creatine kinase isoenzymes. Clin Chem Acta 86:323-327, 1978.
10. Van de Werf F: Cardiac troponins in acute coronary syndromes. New Engl J Med 335:1388-1389, 1996.
11. Katus HA, Looser S, Hallermayer K, et al: Development and in vitro characterization of a new immunoassay of cardiac troponin T. Clin Chem 38:386-393, 1992.
12. Dhond MR, Amsterdam EA: Cardiac troponins: Cardiovascular reviews and reports 21:20-25, 2000.
13. Wu AHB, Abbas SA, Green S, et al: Prognostic value of cardiac troponin T in unstable angina pectoris. Am J Cardiol 76:970-972, 1995.
14. Ravkilde J, Nissen H, Horder M, Thygesen K: Independent prognostic value of serum creatine kinase isoenzyme, MB mass, cardiac troponin T and myosin light chain levels in suspected acute myocardial infarction. Analysis of 28 months of follow-up in 196 patients. J Am Coll Cardiol 25:574-581, 1995.
15. Fuchs S, Kornowski R, Mehran R, et al: Cardiac troponin I levels and clinical outcomes in patients with acute coronary syndromes. J Am Coll Cardiol 34:1704-1710, 1999.
16. Galvani M, Ottani F, Ferrini D, et al: Prognostic influence of elevated values of cardiac troponin I in patients with unstable angina. Circulation 95:2053-2059, 1997.
17. Sobel BE, LeWinter MM: Ingenuous interpretation of elevated blood levels of macromolecular markers of myocardial injury: A recipe for confusion. J Am Coll Cardiol 35:1355-1358, 2000.
18. Polanczyk CA, Lee TH, Cook EF, et al: Cardiac troponin I as a predictor of major cardiac events in emergency department patients with acute chest pain. J Am Coll Cardiol 32:8-14, 1998.
19. Kontos MC, Anderson FP, Fritz L, et al: What is the optimal cut-off value for troponin? A comparison using two different troponin I assays. J Am Coll Cardiol 35:361A, 2000
Figure 1. Patients were grouped by their troponin I level and the percentage of patients who developed major cardiac events is shown.
Table 1. Univariate analysis of the role of cardiac enzymes in the identification of patients who suffered major cardiac events.
Enzyme Sensitivity Specificity Positive Predictive Negative Predictive
% (CI) % (CI) Value % (CI) Value % (CI)
CK-MB 58(36-77) 88(78-94) 60(38-80) 87(77-93)
MB Subforms 62(40-81) 73(62-83) 42(26-60) 86(75-93)
Troponin I 87(67-97) 88(78-94) 70(50-85) 95(88-99)
Table 2. Relative value of CK-MB, CK-MB subforms, and troponin I in the prediction of
major cardiac events.
Enzyme Initial Estimation Final Estimation
Chi-Squared P value Chi-Squared Odds ratio
CK-MB 2.65 .103 -- --
CK-MB Sub 8.06 .004 6.40 6.59
Troponin I 20.24 <.001 26.26 54.98
©2000-2013. All Rights Reserved. Veterinary Solutions LLC