- Email: [email protected]
Excellent reliability of nurse-based bedside diagnosis of acute myocardial infarction by rapid dry-strip creatine kinase MB, myoglobin, and troponin T
Acute Ischemic Heart Disease
Excellent reliability of nurse-based bedside diagnosis of acute myocardial infarction by rapid dry-strip creatine kinase MB, myoglobin, and troponin T Christer Sylvén, MD, PhD, FACC, FESC, Susanne Lindahl, RN, Karin Hellkvist, RN, Olof Nyquist, MD, PhD, and Gundars Rasmanis, MD, PhD Huddinge, Sweden
With the aim to compare the diagnostic efficacy as regards acute myocardial infarction of two rapid dry-strip tests, one with both creatine kinase MB (CK-MB) and myoglobin (C + M) and the other with troponin T, and to test the reliability of bedside diagnosis by the coronary care unit (CCU) nurse, 151 patients with acute chest pain admitted to the CCU were investigated. There was no difference in diagnostic performance between rapid tests and quantitative determinations. With <6hour duration of symptoms, the sensitivity was better for C + M than for troponin T (72% vs 33%, p < 0.05). With symptoms lasting >12 hours on arrival, troponin T performed better, with 100% sensitivity and a negative predictive value of 100% in the 6-hour retest. For exclusion of damage, the two tests have similar and reliable diagnostic capacities 12 hours after the onset of symptoms. The bedside diagnosis or exclusion of acute myocardial infarction was carried out rapidly (within 20 minutes) and reliably by the CCU nurses. (Am Heart J 1998;135:677-83.)
Clinical evaluation of patients with acute chest pain presents a significant diagnostic problem.1 These patients constitute approximately 20% of the total number of patients attending the emergency department.2,3 Some 5% to 10% of the patients with chest pain discharged from emergency departments were reported to have acute myocardial infarction (AMI), with a yearly mortality rate of 5% to 10%.4-9 A missed diagnosis can result in serious consequences and malpractice claims.10,11 As a result, an increasing number of patients are being admitted to the coronary care unit (CCU) for observation to rule out AMI, although a large number of these patients do not have an acute coronary syndrome. Against this background, rapid and precise early confirmation or exclusion of ongoing myocardial damage will have considerable impact on the medical safety of the patient and health care logistics and economics.12 From Karolinska Institute at the Department of Cardiology, Huddinge University Hospital. Supported by the Swedish Heart and Lung Foundation. Submitted April 30, 1997; accepted Sept. 21, 1997. Reprint requests: Christer Sylvén, MD, Department of Cardiology, Huddinge University Hospital, S-141 86 Huddinge, Sweden. E-mail: [email protected] Copyright © 1998 by Mosby, Inc. 0002-8703/98/$5.00 + 0 4/1/86285
The potential use of rapid bedside immunochromatographic dry-strip assays to determine creatine kinase MB (CK-MB), myoglobin, and troponin T might therefore become a significant advance13-22 for medical decisionmaking for patients with acute chest pain. After applying anticoagulated blood, these tests have been suggested within 20 minutes to be capable of demonstrating the presence or absence of abnormal concentrations of these markers.15,19 To obtain maximum utility of these tests, they must be performed at bedside in the CCU to avoid unnecessary time delay associated with transport to and determination at the clinical laboratory. To test the efficacy of the tests, the present study evaluates (1) diagnostic bedside performance compared with quantitative laboratory tests of two rapid qualitative assays, one giving the result for CK-MB and myoglobin in the same test and the other for troponin T, and (2) the reliability of the tests when performed at bedside by the CCU nurse.
Methods One hundred fifty-one consecutive patients admitted to the CCU for chest pain suggestive of myocardial ischemia were studied. Myocardial necrosis was diagnosed according to World Health Organization standards with serial electrocardiography (ECG) and CK and CK-MB enzyme determinations
American Heart Journal April 1998
678 Sylvén et al.
Table I. Clinical characteristics of the patients
Age (yr) Male/female Duration of symptoms (hr) Duration samples 1-2 Hemoglobin (g × L-1) S-creatinine (µmol × L-1)
Mean
Median
SD
Range
66 105/46 14 7.0 140 104
68
13
19-89
9.5 6.0 143 94
16 2.7 15 35
0.5-75 3-22 96-174 57-311
Table III. Detection limits for qualitative determination of elevated plasma CK-MB mass and myoglobin concentrations
Table II. Diagnosis at CCU Q-wave AMI Non-Q-wave AMI Late admission AMI Total AMI Unstable AP Stable AP Non-IHD Atypical chest pain Musculoskeletal pain Pulmonary embolism Inflammatory disease Total non-AMI
26 22 4 52 29 41 7 8 6 1 7 99
AP, Angina pectoris; IHD, ischemic heart disease.
(Boerhinger Mannheim GMBH, Mannheim, Germany). Lateadmission AMI was defined as longer duration of symptoms than 48 hours when the first blood sample was taken. The study had been approved by the local ethics committee. Venous blood was sampled in EDTA tubes (Vacutainer, Becton Dickinson Vacutainer Systems, Rutherford, N.J.) at admission and after 6 hours. Rapid tests were done immediately for CK-MB and myoglobin (CARDIAC STATus CKMB/myoglobin test kit lot No. 645 E22, Spectral Diagnostics, Toronto, Canada) and for troponin T (TROPT, Boerhinger Mannheim, Mannheim, Germany). Plasma was also frozen for later determinations of quantitative CK-MB, myoglobin, and troponin T. Both CARDIAC STATus and TROPT are immunochromatographic dry-strip assays. In these methods, erythrocytes are separated from plasma, which then diffuses along the test strip. If abnormal concentrations of the markers are present, immobilized antibodies bind the marker and a purplish band appears in the test window. A test band indicates that the assay conditions are proper and functioning. The rapid qualitative assays give a yes or no answer. The tests were constructed to give a cutoff for a positive test, CKMB 10 µg × L-1 for CARDIAC STATus, 100 µg × L-1 for myoglobin,19 and 0.20 µg × L-1 for troponin.15 The tests are read 15 minutes after the application of 150 µL of whole blood anticoagulated with ethylenediamene tetraacetic acid. Two CCU nurses participated in the study. Before start of the study they were trained to follow a detailed standardized test procedure for the qualitative tests. Blood was pipetted to
Cutoff for a positive rate of CK-MB (µg × L-1) Myoglobin (µg × L-1) Troponin T (µg × L-1)
50%
90%
95%
11 10 68 14 13
23 16 91 25 21
26 20 94 31 23
Second line of values for CK-MB and troponin T shows results when patients diagnosed with inflammation and with false-positive tests were deleted.
the strips, and all tests were read after exactly 15 minutes after application of the blood sample. Quantitative determinations were made for the three investigated markers. CK-MB mass and myoglobin were determined by the Stratus II immunoassay (Dade International, Miami, Fla.) and troponin T by an ELISA technique (ELISA Troponin T, Boerhinger Mannheim GMBH, Mannheim, Germany). All determinations were made with daily quality assessments as recommended by the manufacturers.
Results Tables I and II give clinical characteristics of the patients. Of the 151 patients, 52 were diagnosed as having AMI. Details of the performance of the qualitative tests for CK-MB, myoglobin, and troponin T in relation to plasma concentration of the respective marker are given in Figs. 1 through 3. For the indicated quantitative intervals, the number of qualitative positive and negative outcomes are given above and below the x-axis. It can be seen from these figures how the three qualitative tests behave in the cutoff region. Table III summarizes the results of these frequency distributions. After the exclusion of two patients who had a final diagnosis of inflammation, the 50% cutoff values were for CK-MB 10 µg × L-1, 68 µg × L-1 for myoglobin, and 0.13 µg × L-1 for troponin T. For the three tests there was a gray zone to the 95% cutoff value of 50% to 100% of the 50% cutoff value. The overall performances of quantitative and qualita-
American Heart Journal Volume 135, Number 4
Figure 1
Frequency distribution of positive (+) and negative (-) results for CK-MB with CK-MB plus myoglobin dry-strip test as related to quantitative CK-MB mass concentrations determined by ELISA.
Figure 2
Sylvén et al. 679
Figure 3
Frequency distribution of positive (+) and negative (-) results for troponin T with troponin T dry-strip test as related to quantitative troponin T concentrations determined by ELISA.
Figure 4
Receiver operating curves for CK-MB left, myoglobin middle, and troponin T right with false-positive rate x-axis and true-positive rate (y-axis). Quantitative test are indicated by open circles and qualitative tests by solid circles. Measured values with quantitative and qualitative tests are indicated by squares and triangles, respectively. For quantitative determination of troponin T, values are given for both 0.10 (higher value) and 0.20 (lower value) µg × L-1 cutoff values.
Frequency distribution of positive (+) and negative (-) results for myoglobin with CK-MB plus myoglobin dry-strip test as related to quantitative myoglobin concentrations determined by ELISA.
tive tests in test 1 and the 6-hour retest were analyzed for sensitivity, specificity, and the predictive value of a positive [PV(+)] or negative [PV(-)] test. From a clinical point of view, sensitivity (positive test in disease) and PV(-) (health in negative test) are the practically important descriptors; these are given in Table IV. For qualitative tests, analyses are presented for CK-MB, myoglobin,
and troponin T. Because CK-MB and myoglobin analyses were combined in the same rapid test, the results are also given for the outcome of only one of the two tests, positive or both positive. The performance of the qualitative tests was not different from that of the quantitative tests, suggesting similar reliability of the two rapid dry-strip tests when performed at bedside by the CCU nurse. Receiver operating characteristics (ROC),23 illustrated by ROC curves, further demonstrated the similarity in diagnostic efficacy between quantitative and qualitative tests for all the three substances (Fig. 4). For
American Heart Journal April 1998
680 Sylvén et al.
Figure 5
Sensitivity in test 1 (top) and in 6-hour retest (bottom) in patients with duration of symptoms in test 1 of <6 hours, between 6 and 12 hours, and >12 hours.
quantitative troponin T, empirical characteristics are indicated for both a cutoff of 0.20 and 0.10 µg × L-1. As expected for ROC, both cutoffs have the same diagnostic efficacy and thus are located at the same ROC curve. Fig. 5 shows the diagnostic performance as sensitivity in relation to the duration of symptoms in test 1 and in the following 6-hour retest. Interestingly, in patients with a duration of symptoms of <6 hours in test 1, CK-MB had a sensitivity of 0.72, which was higher (p < 0.05) than that of troponin T, 0.33. The addition of myoglobin enhanced the sensitivity to 0.89. In the 6-hour retest, however, there was no significant difference between the tests, with CK-MB having a sensitivity of 1.00 and troponin T of 0.94. When test 1 was done in patients with >12 hours of symptoms, CK-MB and troponin T showed similar high sensitivities, 0.86 and 0.90, whereas in the 6-hour retest, troponin T tended to perform better, with a sensitivity of 1.00 compared with 0.89 for CK-MB. Thus myoglobin tended to add useful diagnostic information only when the duration of symptoms was <6 hours. In the 6-hour retest, myoglobin sensitivity was
Figure 6
PV(-) in test 1 (top) and in 6-hour retest (bottom) in patients with duration of symptoms in test 1 of <6 hours, between 6 and 12 hours, and >12 hours.
only 0.36 for the 6- to 12-hour interval. The confidence interval was, however, 0.16, suggesting that the sensitivity was of a similar order as the >12-hour samples. Fig. 6 shows that PV(-) was approximately 0.90 for CK-MB and troponin T in test 1 in patients with a duration of symptoms of 6 to 12 hours. With longer duration of symptoms, troponin T maintained this diagnostic level in test 1, whereas CK-MB decreased its PV(-) (Fig. 6). In the 6-hour retest, PV(-) for CK-MB was 1.00 only in patients with <6 hours of symptoms, whereupon it decreased. Interestingly, troponin T showed a PV(-) of 0.95 in the 6-hour retest in patients with a short duration of symptoms in test 1 but reached a PV(-) of 1.00 only in the retest performed in patients with >12 hours of symptoms in test 1. This implies that exclusion of AMI may only be achieved with troponin T 12 to 18 hours after onset of symptoms.
Discussion This study shows for the first time that bedside diagnoses by the CCU nurse with rapid dry-strip chromato-
American Heart Journal Volume 135, Number 4
Sylvén et al. 681
Table IV. Sensitivity and PV(-) with confidence limits and overall performance
Quantitative CK-MB 10 µg × L-1 CK-MB 10 Myoglobin 100 µg × L-1 Myoglobin 100 Troponin T 0.10 µg × L-1 Troponin T 0.10 Troponin T 0.20 µg × L-1 Troponin T 0.20 Qualitative CK-MB CK-MB Myoglobin Myoglobin CK-MB or myoglobin CK-MB or myoglobin CK-MB and myoglobin CK-MB and myoglobin Troponin T Troponin T
Test (1) or retest (2)
Sensitivity
PV(-)
1 2 1 2 1 2 1 2
0.79 ± 0.07 0.87 ± 0.03 0.77 ± 0.08 0.69 ± 0.08 0.77 ± 0.07 0.96 ± 0.03 0.65 ± 0.08 0.91 ± 0.05
0.92 ± 0.04 0.95 ± 0.04 0.86 ± 0.06 0.86 ± 0.06 0.93 ± 0.04 0.98 ± 0.02 0.87 ± 0.05 0.91 ± 0.05
1 2 1 2 1 2 1 2 1 2
0.85 ± 0.06 0.91 ± 0.05 0.87 ± 0.05 0.67 ± 0.08 0.88 ± 0.05 0.91 ± 0.05 0.73 ± 0.07 0.67 ± 0.08 0.60 ± 0.08 0.96 ± 0.03
0.92 ± 0.04 0.95 ± 0.04 0.77 ± 0.07 0.81 ± 0.07 0.94 ± 0.04 0.94 ± 0.04 0.87 ± 0.05 0.83 ± 0.07 0.83 ± 0.06 0.97 ± 0.03
graphic determinations of CK-MB plus myoglobin and troponin T are reliable diagnostic methods with same efficacy as quantitative laboratory tests. Thus when performed by the CCU nurse, the diagnostic characteristics do not differ from those of quantitative laboratory determinations of the same markers. The methods rely, however, on visual estimation and a subjective consideration of whether a chromatography band has appeared or not. In the typical case, this is not difficult. In cases with a weak band, the interpretative factor becomes prominent. It is therefore important that the test and the reading are done under detailed, standardized conditions regarding, for example, handling of test strips and blood samples, pipetting, and readings done under the same light source after exactly 15 minutes after application of the blood sample. From a security point of view, it might also be feasible for two nurses to decide independently whether a band has appeared or not. The cutoff for the three rapid tests defined as the plasma concentration interval with 50% positive tests were within the range suggested to be the cutoff for AMI with quantitative tests. The cutoff zone, the gray zone, was narrow for all the markers and in keeping with previous investigations.15,18 The tests investigated in this study were first-generation dry strip tests with 50% cutoffs of CK-MB 10, myoglobin 100, and troponin T 0.20 µg × L-1. Although these tests are qualitative and thus subject to variability of assay, intraobserver, and interobserver variabilities, the narrow gray cutoff zone
the excellent CCU nurse performance as evidenced by ROC curve analysis suggest that these analytic instruments are reliable utilities that have the potential to improve diagnosis and enhance logistics in the CCU. The second generation of tests will have 50% cutoffs of CKMB 5, myoglobin 50, and troponin T 0.10 µg × L-1. The diagnostic efficacy as tested by ROC curves will probably be the same as for the presently investigated tests but with increased sensitivity and decreased specificity. The clinical efficacy as regards identifying or excluding an acute coronary syndrome will probably increase because the 95% cutoff will be in the range of values where the 50% cutoff is for the first-generation tests. In patients with a shorter duration of symptoms than 6 hours, the diagnostic performance of CK-MB, especially when combined with myoglobin, was better than that of troponin T. Thus for CK-MB together with myoglobin, sensitivity was 0.90, whereas for troponin T it was approximately 0.30. These results imply that in the majority of patients with a short duration of symptoms, ongoing myocardial damage can be diagnosed rapidly with CK-MB and myoglobin. Such a rapid diagnosis will provide an important basis for therapeutic decisions such as thrombolysis, acute percutaneous transluminal coronary angioplasty, or bypass surgery. In the 6-hour retest, however, no significant difference in performance was observed between CK-MB without or with myoglobin and troponin T. Both tests had excellent diagnostic capacities,
American Heart Journal April 1998
682 Sylvén et al.
with a sensitivity approximately 0.95. With a longer duration of symptoms, troponin T tended to perform better than CK-MB, and, with this duration of symptoms, myoglobin was noncontributory. Although it is important to make a confirmatory diagnosis as early as possible to institute therapeutic measures, it is also important to safely exclude an ongoing myocardial infarction to be able to send patients home from the emergency department or to allocate CCU patients to a lower level of care. This diagnostic property is best described by PV(-). After a duration of 12 hours after the onset of symptoms, both troponin T and CKMB reached a PV(-) of 0.90. Not until the 6-hour retest in patients in which test 1 was taken >12 hours after the onset of symptoms did troponin T reach a PV(-) of 1.00. However, consideration of the presence of myocardial damage cannot rest only on chemical markers. Unstable angina pectoris is a dynamic process that can mature into myocardial damage or necrosis during the observation in the CCU. Taking these dynamics and their clinical manifestations into consideration, a negative troponin T 12 hours after the onset of symptoms in a patient in whom clinical symptoms and ECG recording do not suggest ongoing myocardial ischemia can be interpreted as excluding AMI. Exclusion with a shorter duration than 6 hours after the onset of symptoms is not safe with any of the markers. The 6- to 12-hour interval shows a PV(-) of approximately 0.90 for both CK-MB and troponin T. Thus these qualitative tests can give a reliable bedside confirmation or exclusion of myocardial ischemia within 6 to 12 hours. The test result is available at bedside after 20 minutes. With conventional laboratory determinations, there is a considerable delay because of the transport of the blood sample to the laboratory, centrifugation, determination, and reporting. As a rule, the analytic result is not available before 1 or more hours have elapsed. With the rapid drystrip chromatography tests, the diagnostic logistics can be greatly improved regarding both the medical safety and the economics of this important health care process. One scenario is confirmation or exclusion of AMI upon arrival at the emergency department or CCU and 12 hours after the onset of symptoms. In patients with a positive test result, quantitative troponin T is taken approximately 24 hours after the onset of symptoms to give an indication of the size of the myocardial damage.22
3. 4.
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15.
16. 17.
18.
19.
References 1. Murata GH. Evaluating chest pain in the emergency department. West J Med 1993;159:61-8. 2. Karlsson BW, Herlitz J, Pettersson P, Ekvall HE, Hjalmarsson °A. Patients
20.
admitted to the emergency room with symptoms indicative of acute myocardial infarction. J Intern Med 1991;230:251-8. Cardiology Preeminence Roundtable. Perfecting MI ruleout. Washington: The Advisory Board Company, 1994. Pozen MW, D’Agostino R, Selker HP, Sytkowski PA, Hood WB. A predictive instrument to improve coronary care unit admission practices in acute ischemic heart disease. N Engl J Med 1984;310:1273-8. Tierce WM, Fitzgerald J, McHenry R, Roth BJ, Psaty B, Stump DL, et al. Physicians’ estimates of the probability of myocardial infarction in emergency room patients with chest pain. Med Decis Making 1986; 6:12-7. Lee TH, Rouan GW, Weisberg MC, Brand DA, Acampora D, Stasiulewicz C, et al. Patients with acute myocardial infarction sent home from the emergency room: clinical characteristics and natural history. Am J Cardiol 1987;60:219-24. Rouan GW, Hedges JR, Toltzis R, Goldstein-Wayne B, Brand D, Goldman L. A chest pain clinic to improve the follow-up of patients released from an urban university teaching hospital emergency department. Ann Emerg Med 1987;16:1145-50. McCarthy BD, Beshansky JR, D’Agostino RB, Selker HP. Missed diagnosis of acute myocardial infarction in the emergency department: results from a multicenter study. Ann Emerg Med 1993;22: 579-82. Lee TH, Goldman L. The coronary care unit turns 25: historical trends and future directions. Ann Intern Med 1988;108:887-94. Rusnak RA, Stair TO, Hansen K, Fastow IS. Litigation against the emergency physician: common features in cases of missed myocardial infarction. Ann Emerg Med 1989;18:1029-34. Karcz A, Holbrook J, Burke MC, Doyle MJ, Erdos MS, Friedman E, et al. Massachusetts emergency medicine closed malpractice claims: 1988-1990. Ann Emerg Med 1993;22:553-9. Apple FS, Preese LM, Riley L, Gerken KL, Van Lente F. Financial impact of a rapid CK-MB-specific immunoassay on the diagnosis of myocardial infarction. Arch Pathol Lab Med 1990;114:1017-20. Puleo PR, Meyer D, Wathen C, Tawa CB, Wheeler S, Hamburg RJ, 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 1994; 331:561-608. Vaidya HC. Myoglobin: an early biochemical marker for the diagnosis of acute myocardial infarction [review]. J Clin Immunoassay 1994; 17:35-9. Muller-Bardoff M, Freitag H, Scheffold T, Remppis A, Kubler W, Katus HA. Development and characterization of a rapid assay for bedside determinations of cardiac troponin T. Circulation 1995;92:2869-75. Antman EM, Grudzien C, Sacks DB. Evaluation of a rapid bedside assay for detection of serum cardiac troponin T. JAMA 1995;273:1279-82. Collinson PO, Thomas S, Siu L, Vasudeva P, Stubbs PJ, Canepa-Anson R. Rapid troponin T measurements in whole blood for detection of myocardial damage. Ann Clin Biochem 1995;32:454-8. Gerhart W, Ljungdahl L, Collinson PO, Lovis C, Mach F, Sylven C, et al. An improved rapid troponin T test with a decreased detection limit. Scan J Clin Lab Invest 1997; 57:549-58. Pagani P, Pagani F. Characterization of a rapid immunochromatographic assay for simultaneous detection of high concentrations of myoglobin and CK-MB in whole blood. Clin Chem 1996; 42:1292. Lindahl B, Andén B, Ohlsson J, Venge P, Wallentin L. Risk stratification in unstable coronary artery disease: additive value of troponin T deter-
American Heart Journal Volume 135, Number 4
minations and pre-discharge exercise tests. FRISK Study Group. Eur Heart J 1997;18:762-70. 21. Stubbs P, Collinson P, Moseley D, Greenwood T, Noble M. Prognostic significance of admission troponin T concentrations in patients with myocardial infarction. Circulation 1996;94:1291-7. 22. Ravkilde J, Nissen H, Hörder M, Thygesen K. Independent prognostic
Sylvén et al. 683
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 1995;25:574-81. 23. Diamond GA. ROC steady: a receiver operating characteristic curve that is invariant to selection bias. Med Decis Making 1987;7:238-43.
Excellent reliability of nurse-based bedside diagnosis of acute myocardial infarction by rapid dry-strip creatine kinase MB, myoglobin, and troponin T Christer Sylvén, MD, PhD, FACC, FESC, Susanne Lindahl, RN, Karin Hellkvist, RN, Olof Nyquist, MD, PhD, and Gundars Rasmanis, MD, PhD Huddinge, Sweden
With the aim to compare the diagnostic efficacy as regards acute myocardial infarction of two rapid dry-strip tests, one with both creatine kinase MB (CK-MB) and myoglobin (C + M) and the other with troponin T, and to test the reliability of bedside diagnosis by the coronary care unit (CCU) nurse, 151 patients with acute chest pain admitted to the CCU were investigated. There was no difference in diagnostic performance between rapid tests and quantitative determinations. With <6hour duration of symptoms, the sensitivity was better for C + M than for troponin T (72% vs 33%, p < 0.05). With symptoms lasting >12 hours on arrival, troponin T performed better, with 100% sensitivity and a negative predictive value of 100% in the 6-hour retest. For exclusion of damage, the two tests have similar and reliable diagnostic capacities 12 hours after the onset of symptoms. The bedside diagnosis or exclusion of acute myocardial infarction was carried out rapidly (within 20 minutes) and reliably by the CCU nurses. (Am Heart J 1998;135:677-83.)
Clinical evaluation of patients with acute chest pain presents a significant diagnostic problem.1 These patients constitute approximately 20% of the total number of patients attending the emergency department.2,3 Some 5% to 10% of the patients with chest pain discharged from emergency departments were reported to have acute myocardial infarction (AMI), with a yearly mortality rate of 5% to 10%.4-9 A missed diagnosis can result in serious consequences and malpractice claims.10,11 As a result, an increasing number of patients are being admitted to the coronary care unit (CCU) for observation to rule out AMI, although a large number of these patients do not have an acute coronary syndrome. Against this background, rapid and precise early confirmation or exclusion of ongoing myocardial damage will have considerable impact on the medical safety of the patient and health care logistics and economics.12 From Karolinska Institute at the Department of Cardiology, Huddinge University Hospital. Supported by the Swedish Heart and Lung Foundation. Submitted April 30, 1997; accepted Sept. 21, 1997. Reprint requests: Christer Sylvén, MD, Department of Cardiology, Huddinge University Hospital, S-141 86 Huddinge, Sweden. E-mail: [email protected] Copyright © 1998 by Mosby, Inc. 0002-8703/98/$5.00 + 0 4/1/86285
The potential use of rapid bedside immunochromatographic dry-strip assays to determine creatine kinase MB (CK-MB), myoglobin, and troponin T might therefore become a significant advance13-22 for medical decisionmaking for patients with acute chest pain. After applying anticoagulated blood, these tests have been suggested within 20 minutes to be capable of demonstrating the presence or absence of abnormal concentrations of these markers.15,19 To obtain maximum utility of these tests, they must be performed at bedside in the CCU to avoid unnecessary time delay associated with transport to and determination at the clinical laboratory. To test the efficacy of the tests, the present study evaluates (1) diagnostic bedside performance compared with quantitative laboratory tests of two rapid qualitative assays, one giving the result for CK-MB and myoglobin in the same test and the other for troponin T, and (2) the reliability of the tests when performed at bedside by the CCU nurse.
Methods One hundred fifty-one consecutive patients admitted to the CCU for chest pain suggestive of myocardial ischemia were studied. Myocardial necrosis was diagnosed according to World Health Organization standards with serial electrocardiography (ECG) and CK and CK-MB enzyme determinations
American Heart Journal April 1998
678 Sylvén et al.
Table I. Clinical characteristics of the patients
Age (yr) Male/female Duration of symptoms (hr) Duration samples 1-2 Hemoglobin (g × L-1) S-creatinine (µmol × L-1)
Mean
Median
SD
Range
66 105/46 14 7.0 140 104
68
13
19-89
9.5 6.0 143 94
16 2.7 15 35
0.5-75 3-22 96-174 57-311
Table III. Detection limits for qualitative determination of elevated plasma CK-MB mass and myoglobin concentrations
Table II. Diagnosis at CCU Q-wave AMI Non-Q-wave AMI Late admission AMI Total AMI Unstable AP Stable AP Non-IHD Atypical chest pain Musculoskeletal pain Pulmonary embolism Inflammatory disease Total non-AMI
26 22 4 52 29 41 7 8 6 1 7 99
AP, Angina pectoris; IHD, ischemic heart disease.
(Boerhinger Mannheim GMBH, Mannheim, Germany). Lateadmission AMI was defined as longer duration of symptoms than 48 hours when the first blood sample was taken. The study had been approved by the local ethics committee. Venous blood was sampled in EDTA tubes (Vacutainer, Becton Dickinson Vacutainer Systems, Rutherford, N.J.) at admission and after 6 hours. Rapid tests were done immediately for CK-MB and myoglobin (CARDIAC STATus CKMB/myoglobin test kit lot No. 645 E22, Spectral Diagnostics, Toronto, Canada) and for troponin T (TROPT, Boerhinger Mannheim, Mannheim, Germany). Plasma was also frozen for later determinations of quantitative CK-MB, myoglobin, and troponin T. Both CARDIAC STATus and TROPT are immunochromatographic dry-strip assays. In these methods, erythrocytes are separated from plasma, which then diffuses along the test strip. If abnormal concentrations of the markers are present, immobilized antibodies bind the marker and a purplish band appears in the test window. A test band indicates that the assay conditions are proper and functioning. The rapid qualitative assays give a yes or no answer. The tests were constructed to give a cutoff for a positive test, CKMB 10 µg × L-1 for CARDIAC STATus, 100 µg × L-1 for myoglobin,19 and 0.20 µg × L-1 for troponin.15 The tests are read 15 minutes after the application of 150 µL of whole blood anticoagulated with ethylenediamene tetraacetic acid. Two CCU nurses participated in the study. Before start of the study they were trained to follow a detailed standardized test procedure for the qualitative tests. Blood was pipetted to
Cutoff for a positive rate of CK-MB (µg × L-1) Myoglobin (µg × L-1) Troponin T (µg × L-1)
50%
90%
95%
11 10 68 14 13
23 16 91 25 21
26 20 94 31 23
Second line of values for CK-MB and troponin T shows results when patients diagnosed with inflammation and with false-positive tests were deleted.
the strips, and all tests were read after exactly 15 minutes after application of the blood sample. Quantitative determinations were made for the three investigated markers. CK-MB mass and myoglobin were determined by the Stratus II immunoassay (Dade International, Miami, Fla.) and troponin T by an ELISA technique (ELISA Troponin T, Boerhinger Mannheim GMBH, Mannheim, Germany). All determinations were made with daily quality assessments as recommended by the manufacturers.
Results Tables I and II give clinical characteristics of the patients. Of the 151 patients, 52 were diagnosed as having AMI. Details of the performance of the qualitative tests for CK-MB, myoglobin, and troponin T in relation to plasma concentration of the respective marker are given in Figs. 1 through 3. For the indicated quantitative intervals, the number of qualitative positive and negative outcomes are given above and below the x-axis. It can be seen from these figures how the three qualitative tests behave in the cutoff region. Table III summarizes the results of these frequency distributions. After the exclusion of two patients who had a final diagnosis of inflammation, the 50% cutoff values were for CK-MB 10 µg × L-1, 68 µg × L-1 for myoglobin, and 0.13 µg × L-1 for troponin T. For the three tests there was a gray zone to the 95% cutoff value of 50% to 100% of the 50% cutoff value. The overall performances of quantitative and qualita-
American Heart Journal Volume 135, Number 4
Figure 1
Frequency distribution of positive (+) and negative (-) results for CK-MB with CK-MB plus myoglobin dry-strip test as related to quantitative CK-MB mass concentrations determined by ELISA.
Figure 2
Sylvén et al. 679
Figure 3
Frequency distribution of positive (+) and negative (-) results for troponin T with troponin T dry-strip test as related to quantitative troponin T concentrations determined by ELISA.
Figure 4
Receiver operating curves for CK-MB left, myoglobin middle, and troponin T right with false-positive rate x-axis and true-positive rate (y-axis). Quantitative test are indicated by open circles and qualitative tests by solid circles. Measured values with quantitative and qualitative tests are indicated by squares and triangles, respectively. For quantitative determination of troponin T, values are given for both 0.10 (higher value) and 0.20 (lower value) µg × L-1 cutoff values.
Frequency distribution of positive (+) and negative (-) results for myoglobin with CK-MB plus myoglobin dry-strip test as related to quantitative myoglobin concentrations determined by ELISA.
tive tests in test 1 and the 6-hour retest were analyzed for sensitivity, specificity, and the predictive value of a positive [PV(+)] or negative [PV(-)] test. From a clinical point of view, sensitivity (positive test in disease) and PV(-) (health in negative test) are the practically important descriptors; these are given in Table IV. For qualitative tests, analyses are presented for CK-MB, myoglobin,
and troponin T. Because CK-MB and myoglobin analyses were combined in the same rapid test, the results are also given for the outcome of only one of the two tests, positive or both positive. The performance of the qualitative tests was not different from that of the quantitative tests, suggesting similar reliability of the two rapid dry-strip tests when performed at bedside by the CCU nurse. Receiver operating characteristics (ROC),23 illustrated by ROC curves, further demonstrated the similarity in diagnostic efficacy between quantitative and qualitative tests for all the three substances (Fig. 4). For
American Heart Journal April 1998
680 Sylvén et al.
Figure 5
Sensitivity in test 1 (top) and in 6-hour retest (bottom) in patients with duration of symptoms in test 1 of <6 hours, between 6 and 12 hours, and >12 hours.
quantitative troponin T, empirical characteristics are indicated for both a cutoff of 0.20 and 0.10 µg × L-1. As expected for ROC, both cutoffs have the same diagnostic efficacy and thus are located at the same ROC curve. Fig. 5 shows the diagnostic performance as sensitivity in relation to the duration of symptoms in test 1 and in the following 6-hour retest. Interestingly, in patients with a duration of symptoms of <6 hours in test 1, CK-MB had a sensitivity of 0.72, which was higher (p < 0.05) than that of troponin T, 0.33. The addition of myoglobin enhanced the sensitivity to 0.89. In the 6-hour retest, however, there was no significant difference between the tests, with CK-MB having a sensitivity of 1.00 and troponin T of 0.94. When test 1 was done in patients with >12 hours of symptoms, CK-MB and troponin T showed similar high sensitivities, 0.86 and 0.90, whereas in the 6-hour retest, troponin T tended to perform better, with a sensitivity of 1.00 compared with 0.89 for CK-MB. Thus myoglobin tended to add useful diagnostic information only when the duration of symptoms was <6 hours. In the 6-hour retest, myoglobin sensitivity was
Figure 6
PV(-) in test 1 (top) and in 6-hour retest (bottom) in patients with duration of symptoms in test 1 of <6 hours, between 6 and 12 hours, and >12 hours.
only 0.36 for the 6- to 12-hour interval. The confidence interval was, however, 0.16, suggesting that the sensitivity was of a similar order as the >12-hour samples. Fig. 6 shows that PV(-) was approximately 0.90 for CK-MB and troponin T in test 1 in patients with a duration of symptoms of 6 to 12 hours. With longer duration of symptoms, troponin T maintained this diagnostic level in test 1, whereas CK-MB decreased its PV(-) (Fig. 6). In the 6-hour retest, PV(-) for CK-MB was 1.00 only in patients with <6 hours of symptoms, whereupon it decreased. Interestingly, troponin T showed a PV(-) of 0.95 in the 6-hour retest in patients with a short duration of symptoms in test 1 but reached a PV(-) of 1.00 only in the retest performed in patients with >12 hours of symptoms in test 1. This implies that exclusion of AMI may only be achieved with troponin T 12 to 18 hours after onset of symptoms.
Discussion This study shows for the first time that bedside diagnoses by the CCU nurse with rapid dry-strip chromato-
American Heart Journal Volume 135, Number 4
Sylvén et al. 681
Table IV. Sensitivity and PV(-) with confidence limits and overall performance
Quantitative CK-MB 10 µg × L-1 CK-MB 10 Myoglobin 100 µg × L-1 Myoglobin 100 Troponin T 0.10 µg × L-1 Troponin T 0.10 Troponin T 0.20 µg × L-1 Troponin T 0.20 Qualitative CK-MB CK-MB Myoglobin Myoglobin CK-MB or myoglobin CK-MB or myoglobin CK-MB and myoglobin CK-MB and myoglobin Troponin T Troponin T
Test (1) or retest (2)
Sensitivity
PV(-)
1 2 1 2 1 2 1 2
0.79 ± 0.07 0.87 ± 0.03 0.77 ± 0.08 0.69 ± 0.08 0.77 ± 0.07 0.96 ± 0.03 0.65 ± 0.08 0.91 ± 0.05
0.92 ± 0.04 0.95 ± 0.04 0.86 ± 0.06 0.86 ± 0.06 0.93 ± 0.04 0.98 ± 0.02 0.87 ± 0.05 0.91 ± 0.05
1 2 1 2 1 2 1 2 1 2
0.85 ± 0.06 0.91 ± 0.05 0.87 ± 0.05 0.67 ± 0.08 0.88 ± 0.05 0.91 ± 0.05 0.73 ± 0.07 0.67 ± 0.08 0.60 ± 0.08 0.96 ± 0.03
0.92 ± 0.04 0.95 ± 0.04 0.77 ± 0.07 0.81 ± 0.07 0.94 ± 0.04 0.94 ± 0.04 0.87 ± 0.05 0.83 ± 0.07 0.83 ± 0.06 0.97 ± 0.03
graphic determinations of CK-MB plus myoglobin and troponin T are reliable diagnostic methods with same efficacy as quantitative laboratory tests. Thus when performed by the CCU nurse, the diagnostic characteristics do not differ from those of quantitative laboratory determinations of the same markers. The methods rely, however, on visual estimation and a subjective consideration of whether a chromatography band has appeared or not. In the typical case, this is not difficult. In cases with a weak band, the interpretative factor becomes prominent. It is therefore important that the test and the reading are done under detailed, standardized conditions regarding, for example, handling of test strips and blood samples, pipetting, and readings done under the same light source after exactly 15 minutes after application of the blood sample. From a security point of view, it might also be feasible for two nurses to decide independently whether a band has appeared or not. The cutoff for the three rapid tests defined as the plasma concentration interval with 50% positive tests were within the range suggested to be the cutoff for AMI with quantitative tests. The cutoff zone, the gray zone, was narrow for all the markers and in keeping with previous investigations.15,18 The tests investigated in this study were first-generation dry strip tests with 50% cutoffs of CK-MB 10, myoglobin 100, and troponin T 0.20 µg × L-1. Although these tests are qualitative and thus subject to variability of assay, intraobserver, and interobserver variabilities, the narrow gray cutoff zone
the excellent CCU nurse performance as evidenced by ROC curve analysis suggest that these analytic instruments are reliable utilities that have the potential to improve diagnosis and enhance logistics in the CCU. The second generation of tests will have 50% cutoffs of CKMB 5, myoglobin 50, and troponin T 0.10 µg × L-1. The diagnostic efficacy as tested by ROC curves will probably be the same as for the presently investigated tests but with increased sensitivity and decreased specificity. The clinical efficacy as regards identifying or excluding an acute coronary syndrome will probably increase because the 95% cutoff will be in the range of values where the 50% cutoff is for the first-generation tests. In patients with a shorter duration of symptoms than 6 hours, the diagnostic performance of CK-MB, especially when combined with myoglobin, was better than that of troponin T. Thus for CK-MB together with myoglobin, sensitivity was 0.90, whereas for troponin T it was approximately 0.30. These results imply that in the majority of patients with a short duration of symptoms, ongoing myocardial damage can be diagnosed rapidly with CK-MB and myoglobin. Such a rapid diagnosis will provide an important basis for therapeutic decisions such as thrombolysis, acute percutaneous transluminal coronary angioplasty, or bypass surgery. In the 6-hour retest, however, no significant difference in performance was observed between CK-MB without or with myoglobin and troponin T. Both tests had excellent diagnostic capacities,
American Heart Journal April 1998
682 Sylvén et al.
with a sensitivity approximately 0.95. With a longer duration of symptoms, troponin T tended to perform better than CK-MB, and, with this duration of symptoms, myoglobin was noncontributory. Although it is important to make a confirmatory diagnosis as early as possible to institute therapeutic measures, it is also important to safely exclude an ongoing myocardial infarction to be able to send patients home from the emergency department or to allocate CCU patients to a lower level of care. This diagnostic property is best described by PV(-). After a duration of 12 hours after the onset of symptoms, both troponin T and CKMB reached a PV(-) of 0.90. Not until the 6-hour retest in patients in which test 1 was taken >12 hours after the onset of symptoms did troponin T reach a PV(-) of 1.00. However, consideration of the presence of myocardial damage cannot rest only on chemical markers. Unstable angina pectoris is a dynamic process that can mature into myocardial damage or necrosis during the observation in the CCU. Taking these dynamics and their clinical manifestations into consideration, a negative troponin T 12 hours after the onset of symptoms in a patient in whom clinical symptoms and ECG recording do not suggest ongoing myocardial ischemia can be interpreted as excluding AMI. Exclusion with a shorter duration than 6 hours after the onset of symptoms is not safe with any of the markers. The 6- to 12-hour interval shows a PV(-) of approximately 0.90 for both CK-MB and troponin T. Thus these qualitative tests can give a reliable bedside confirmation or exclusion of myocardial ischemia within 6 to 12 hours. The test result is available at bedside after 20 minutes. With conventional laboratory determinations, there is a considerable delay because of the transport of the blood sample to the laboratory, centrifugation, determination, and reporting. As a rule, the analytic result is not available before 1 or more hours have elapsed. With the rapid drystrip chromatography tests, the diagnostic logistics can be greatly improved regarding both the medical safety and the economics of this important health care process. One scenario is confirmation or exclusion of AMI upon arrival at the emergency department or CCU and 12 hours after the onset of symptoms. In patients with a positive test result, quantitative troponin T is taken approximately 24 hours after the onset of symptoms to give an indication of the size of the myocardial damage.22
3. 4.
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15.
16. 17.
18.
19.
References 1. Murata GH. Evaluating chest pain in the emergency department. West J Med 1993;159:61-8. 2. Karlsson BW, Herlitz J, Pettersson P, Ekvall HE, Hjalmarsson °A. Patients
20.
admitted to the emergency room with symptoms indicative of acute myocardial infarction. J Intern Med 1991;230:251-8. Cardiology Preeminence Roundtable. Perfecting MI ruleout. Washington: The Advisory Board Company, 1994. Pozen MW, D’Agostino R, Selker HP, Sytkowski PA, Hood WB. A predictive instrument to improve coronary care unit admission practices in acute ischemic heart disease. N Engl J Med 1984;310:1273-8. Tierce WM, Fitzgerald J, McHenry R, Roth BJ, Psaty B, Stump DL, et al. Physicians’ estimates of the probability of myocardial infarction in emergency room patients with chest pain. Med Decis Making 1986; 6:12-7. Lee TH, Rouan GW, Weisberg MC, Brand DA, Acampora D, Stasiulewicz C, et al. Patients with acute myocardial infarction sent home from the emergency room: clinical characteristics and natural history. Am J Cardiol 1987;60:219-24. Rouan GW, Hedges JR, Toltzis R, Goldstein-Wayne B, Brand D, Goldman L. A chest pain clinic to improve the follow-up of patients released from an urban university teaching hospital emergency department. Ann Emerg Med 1987;16:1145-50. McCarthy BD, Beshansky JR, D’Agostino RB, Selker HP. Missed diagnosis of acute myocardial infarction in the emergency department: results from a multicenter study. Ann Emerg Med 1993;22: 579-82. Lee TH, Goldman L. The coronary care unit turns 25: historical trends and future directions. Ann Intern Med 1988;108:887-94. Rusnak RA, Stair TO, Hansen K, Fastow IS. Litigation against the emergency physician: common features in cases of missed myocardial infarction. Ann Emerg Med 1989;18:1029-34. Karcz A, Holbrook J, Burke MC, Doyle MJ, Erdos MS, Friedman E, et al. Massachusetts emergency medicine closed malpractice claims: 1988-1990. Ann Emerg Med 1993;22:553-9. Apple FS, Preese LM, Riley L, Gerken KL, Van Lente F. Financial impact of a rapid CK-MB-specific immunoassay on the diagnosis of myocardial infarction. Arch Pathol Lab Med 1990;114:1017-20. Puleo PR, Meyer D, Wathen C, Tawa CB, Wheeler S, Hamburg RJ, 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 1994; 331:561-608. Vaidya HC. Myoglobin: an early biochemical marker for the diagnosis of acute myocardial infarction [review]. J Clin Immunoassay 1994; 17:35-9. Muller-Bardoff M, Freitag H, Scheffold T, Remppis A, Kubler W, Katus HA. Development and characterization of a rapid assay for bedside determinations of cardiac troponin T. Circulation 1995;92:2869-75. Antman EM, Grudzien C, Sacks DB. Evaluation of a rapid bedside assay for detection of serum cardiac troponin T. JAMA 1995;273:1279-82. Collinson PO, Thomas S, Siu L, Vasudeva P, Stubbs PJ, Canepa-Anson R. Rapid troponin T measurements in whole blood for detection of myocardial damage. Ann Clin Biochem 1995;32:454-8. Gerhart W, Ljungdahl L, Collinson PO, Lovis C, Mach F, Sylven C, et al. An improved rapid troponin T test with a decreased detection limit. Scan J Clin Lab Invest 1997; 57:549-58. Pagani P, Pagani F. Characterization of a rapid immunochromatographic assay for simultaneous detection of high concentrations of myoglobin and CK-MB in whole blood. Clin Chem 1996; 42:1292. Lindahl B, Andén B, Ohlsson J, Venge P, Wallentin L. Risk stratification in unstable coronary artery disease: additive value of troponin T deter-
American Heart Journal Volume 135, Number 4
minations and pre-discharge exercise tests. FRISK Study Group. Eur Heart J 1997;18:762-70. 21. Stubbs P, Collinson P, Moseley D, Greenwood T, Noble M. Prognostic significance of admission troponin T concentrations in patients with myocardial infarction. Circulation 1996;94:1291-7. 22. Ravkilde J, Nissen H, Hörder M, Thygesen K. Independent prognostic
Sylvén et al. 683
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 1995;25:574-81. 23. Diamond GA. ROC steady: a receiver operating characteristic curve that is invariant to selection bias. Med Decis Making 1987;7:238-43.