Development and Characterization of a Rapid Assay for Bedside Determinations of Cardiac Troponin T
Background The appearance of cardiac proteins in blood is the most specific and sensitive indicator of acute myocardial cell necrosis. The measurement of cardiac markers, however, is time consuming and requires sophisticated equipment. To facilitate the biochemical detection for acute myocardial cell necrosis, a whole-blood rapid assay device for cardiac troponin T detection was developed that provides a test result within 20 minutes.
Methods and Results Monoclonal antibody M7 is labeled with gold particles, and antibody 1B10 is labeled with biotin. Both antibodies, as well as buffer substances and detergents, are adsorbed onto paper fleeces mounted below an application well. Heparinized blood (160 μL) applied to this well solubilizes the dry chemistry reagents. Blood cells are separated from plasma via a glass-fiber fleece. The immunocomplexes formed are concentrated within the reading zone by binding of the biotin-labeled antibody with streptavidine immobilized to the test device. Troponin T bound to the test device serves as a control. The detection limit of this assay is 0.18 μg/L with a cross-reactivity with skeletal troponin T of 0.5%. In clinical analyses involving 25 healthy volunteers, 62 patients with chest pain but without myocardial ischemia, 35 patients with acute myocardial infarction, 24 patients with minor myocardial cell damage due to radiofrequency ablation, and 35 patients with unstable angina, the rapid assay was comparable to the troponin T enzyme immunoassay in regard to sensitivity and specificity.
Conclusions This newly developed assay allows accurate, rapid, and convenient diagnosis of acute myocardial cell necrosis.
The diagnosis of AMI is easy and straightforward when anginal pain is accompanied by typical changes on the ECG.1 2 3 In these patients, thrombolytic therapy should be administered before AMI is confirmed through time-consuming measurements of cardiac markers. However, in many patients with chest pain, a correct diagnosis of myocardial cell injury depends entirely on the result of biochemical assays. Thus, in approximately 20% to 40% of all patients finally classified as having AMI by elevations of cardiac enzymes in blood, the ECG on admission is nondiagnostic.4 5 6 Therefore, a rapid diagnostic tool capable of indicating myocardial cell necrosis with high specificity and sensitivity would be extremely beneficial in the diagnosis and management of these patients.
In comparison to the use of CK-MB, the marker molecule cTnT has several advantages. First, cTnT persists longer in circulation than CK-MB due to prolonged release of this marker from disintegrating myofibers, resulting in a longer diagnostic window of time. Second, the proportional rise of cTnT in blood above the discriminator value is higher than that observed for CK-MB. Third, cTnT is a cardiospecific molecule that can be differentiated from skeletal muscle isoforms by monoclonal antibodies, allowing better differentiation of cardiac and skeletal muscle damage than with the use of CK-MB. These characteristics of the marker cTnT have resulted in improved diagnostic efficiency, enabling the identification of a high-risk subgroup of patients with unstable angina pectoris and minor myocardial cell damage.7 8 These patients are characterized by minor but persistent elevation of cTnT in circulation. This ongoing cTnT release must result from degradation of contractile proteins and thus is most likely an indication of irreversible cell necrosis (microinfarction).
The clinical value of the enzyme immunoassay for cardiac TnT, however, is limited by its rather long turnaround time of 90 minutes at 20°C and 45 minutes at 37°C9 ; therefore, the purpose of the present study was to develop a whole-blood rapid assay system for measuring cTnT, yielding reliable and accurate test results within 20 minutes, as well as to test its diagnostic performance in patients with suspected myocardial cell damage.
Human myocardium or skeletal muscle was obtained from the Department of Pathology less than 10 hours after death of the patient. cTnT and skeletal muscle TnT were prepared from these tissue specimens by standard procedures.10 11 12 The molecular homogeneity of purified cTnT was assessed with sodium dodecyl sulfate–polyacrylamide gel electrophoresis.13 The identity of the purified proteins with cTnT was proved by partial sequence analysis of the amino acid composition, as outlined previously.9 Concentrations of skeletal muscle TnT and cTnT were determined by specific absorption of 0.706 L/gcm at 280 nm.9 Protein standards were achieved by diluting purified TnT in normal human plasma mixed with equal volumes of phosphate-buffered saline. Normal mouse IgG (0.1%) was added to reduce unspecific binding.
The specificity of the antibodies and the rapid cTnT assay was tested with cardiac and skeletal muscle homogenized in 3 vol of 50 mmol/L KCl, 50 mmol/L Tris · HCl, pH 7.0, 5 mmol/L EDTA, 1 mmol/L DTT, and 2% ε-aminocapronic acid. Tissue was obtained from an organ donor and from a patient who had died <10 hours before analysis. Muscle homogenates were centrifuged at 14 000g for 10 minutes. The resulting supernatants were saved for analyses of the cytosolic TnT fraction. The pellets were solubilized in 10 vol of 1.5 mol/L KCl, 50 mmol/L Tris · HCl, 2 mmol/L EDTA, 5 mmol/L ATP, and 2% ε-aminocapronic acid and centrifuged for 20 minutes at 14 000g to obtain the myofibrillar TnT protein fraction.
Characterization of Selected Antibodies
The same two monoclonal anti-human cTnT antibodies used in the enzyme immunoassay were used in the rapid cTnT assay. The selection and characterization of these two antibodies are described in detail elsewhere.9 In displacement studies, the antibody 1B10 reveals a cross-reactivity with TnT from skeletal muscle of 12%, whereas antibody M7 is cardiospecific.9
Monoclonal antibodies were purified from ascites by ammonium sulfate precipitation (40% v/v) and affinity chromatography using a protein G–Sepharose column (Pharmacia).
Biotin Labeling of Antibody 1B10
The high-affinity monoclonal antibody 1B10 was conjugated with biotin by incubating 20 mg/mL of the antibody in a 10-molar excess of d-biotinoyl-ε-aminocaproic acid N-hydroxysuccinimide ester (Biotin X-NHS, Boehringer Mannheim) with 0.1 mol/L KHPO, pH 8.5, for 25 minutes at room temperature. The reaction was stopped by adding lysine to a final concentration of 10 mmol/L.
Gold Labeling of Antibody M7
The cardiospecific anti-human cTnT antibody M7 was labeled with gold particles according to the general methods of Roth et al.14 In this procedure, the dispersion of gold particles is achieved through the heating of a solution containing 0.01% HAuCl and 0.01% Na-citrate according to Frens et al.15 After cooling and adjustment of pH to 6.5 with 0.2 mmol/L K2CO3, purified monoclonal antibody M7 (0.5 mg/mL) was added to the monodispersed gold solution with an optical density of 1.0 at a wavelength of 533 nm to give a final concentration of 10 μg/mL and incubated at room temperature for 30 minutes,14 followed by the addition of bovine serum albumin (1 mg/mL final concentration).
In the developed assay, cTnT is bound to various epitopes using the biotin-labeled antibody 1B10 and the gold-labeled antibody M7 (Fig 1⇓). The reaction is initiated by pipetting 160 μL of whole heparinized peripheral blood to the well of the test device. cTnT in blood combines with the high-affinity biotinylated antibody 1B10 and the cardiospecific gold-labeled antibody M7 (0.23 and 0.17 μg per test device, respectively) in buffer containing 150 mmol/L morpholinoethansulfonic acid, and 110 mmol/L sodium succinate, pH 5.6. Blood cells are separated from plasma with a glass-fiber fleece. The sandwich complex consisting of gold-labeled antibody M7, cTnT, and biotin-labeled antibody 1B10 is immobilized in a capture zone of streptavidine bound to the solid phase. The concentrated gold particles appear as a purple band. The intensity and speed at which the color develops are proportional to the concentration of cTnT in the patient’s blood. Unreacted gold-labeled antibody M7 may thereafter combine with bovine cTnT (25 ng per test device) immobilized distally to the streptavidine band on the cellulose nitrate membrane. The formation of this second band indicates the correct function of the test and unimpeded flow of plasma through the device.
Test Performance Analysis
Heparinized blood was collected from five patients with AMI and from four healthy volunteers. Aliquots (1 mL) were centrifuged, and circulating cTnT concentrations were determined by enzyme immunoassay. The remaining blood of these samples was stored at 4°C. Each blood sample from the five AMI patients was then diluted stepwise with heparinized blood from healthy blood donors covering a cTnT concentration range of from 5 to <0.005 μg/L. Then, 160 μL of each dilution was applied to the well of a rapid test device and analyzed by inspection at 5, 10, 15, and 20 minutes after application. When two bands (control and cTnT band) appeared in the reading zone during this time period, the results were rated as positive. When only the control band occurred, the results were rated as negative. When no control band was observed, the test was discarded. The investigator was blinded regarding the dilutions applied and the labeling code used. Before analyses, the test devices were arranged by a second person in a random order.
The specificity was tested using homogenates and purified TnT from human cardiac and skeletal muscle (psoas muscle). The cytosolic and myofibrillar muscle extracts were diluted in blood of healthy volunteers.
Precision at the Detection Limit
Seven blood samples from post–myocardial infarction patients with different cTnT concentrations (0.06, 0.12, 0.15, 0.18, 0.21 0.24, and 2.4 μg/L) were analyzed. The run precision was determined by five repetitive measurements of each of the seven cTnT concentrations in blood. The results of the rapid assays were analyzed as outlined.
Interobserver and Intraobserver Variabilities of the Assay
The intraobserver variability of the assay (Fig 4⇓) was tested by five repetitive measurements of seven blood samples with different cTnT concentrations. The interobserver variability was defined at five different cTnT concentrations by visual assessment of each test device by 12 different persons.
The performance of the rapid cTnT assay was tested in three groups of patients (Table 1⇓): (1) 25 healthy volunteers without any clinical evidence of previous or present cardiac disease and 62 patients in the Department of Internal Medicine with chest pain but without evidence of an acute cardiac event as assessed by history, clinical examination, ECG, echocardiography, and cardiac enzymes.
(2) Thirty-five patients with definite AMI and 24 patients with Wolff-Parkinson-White syndrome and successful radiofrequency ablation were included in this group. The diagnostic criteria for definite AMI were monophasic ST-segment elevations in at least two adjacent leads of the 12-lead ECG with appearance of new Q waves and reduction of R waves and/or a time-dependent rise of serum activities of CK-MB more than twice the upper limit of normal concentrations with a CK-MB fraction exceeding 5% of total CK activity. The diagnostic criteria for myocardial cell damage in patients undergoing radiofrequency ablation were a persistent disappearance of a preexisting abnormal atrioventricular conduction on intracardiac and surface 12-lead ECG after application of several bursts of radiofrequency energy to the myocardial zone characterized by abnormal conduction.
(3) A group of 35 patients with possible minor myocardial cell damage consisted of patients with unstable angina admitted to the emergency department of the University Hospital in Heidelberg without evidence for AMI by ECG or cardiac enzymes. These patients were classified according to the Braunwald criteria.16
All patients gave written informed consent to participation in the study after a thorough explanation of the study protocol. This investigation was approved by the Ethics Committee of the University of Heidelberg.
Total CK activity was determined in the Clinical Chemistry Laboratory using a Clin Chem Analyzer and the reagents provided by the manufacturer. The upper limit of normal concentrations of total serum CK is 80 IU/L for men and 75 IU/L for women. CK-MB activity was determined by the immunoinhibition method (CK-MB-NAC, Boehringer Mannheim), the upper limit of normal at 25°C being 10 IU/L or <5% CK-MB of total CK activity.
cTnT concentration was determined by the enzyme immunoassay method (Troponin T ELISA, Boehringer Mannheim) using the ENZYMUN test system (ES 300 Analyser, Boehringer Mannheim). cTnT was measured at room temperature with an assay turnaround time of 90 minutes. The discriminator value was 0.2 μg/L.
cTnT Rapid Assay Device
The cTnT rapid assay device is shown in Fig 2⇓ with the application well on the right and the reading zone located in the center of the device. The reading zone of these test devices is shown below after the addition of blood from a healthy volunteer (−), a patient with unstable angina pectoris and cTnT levels of 0.34 μg/L (+*), and a patient with AMI and cTnT levels of 1.5 μg/L (+). The intensity of the cTnT band apparently depends on the cTnT concentration in blood. When using cTnT rapid assay devices, it is important to rate even a faint (+*) band as a positive test result.
With purified cTnT and skeletal muscle TnT, positive results are found at 0.2 and 50 μg/L, respectively, yielding a cross-reactivity of 0.4% (Table 2⇓). Dilutions of low–ionic strength extracts of cardiac and skeletal muscle (cytosolic fraction of TnT) were compared to allow an estimation of cross-reactivity of the cTnT assay using native cardiac and skeletal muscle antigens. A positive test result was found at 1:40 000 and 1:200 dilutions of cardiac and skeletal muscle, respectively, giving a cross-reactivity of 0.5%.
The rapid assay results of heparinized blood sample dilutions from five AMI patients with known cTnT concentrations are shown in Fig 3⇓. The rapid assay result readings at 20 minutes were positive in all samples with cTnT concentrations of ≥0.18 μg/L and were negative in all samples with cTnT concentrations of ≤0.1 μg/L. In the range of 0.1 to 0.18 μg/L, the rapid assay results vary due to the analytical imprecision of visual assessment as to the presence or absence of a cTnT-positive band.
Precision at the Detection Limit and Speed of Color Development
The precision of the rapid assay at the detection limit was tested by five repetitive measurements of blood samples with cTnT concentrations of 0.06, 0.12, 0.15, 0.18, 0.21, 0.24, and 2.4 μg/L analyzed at 5, 10, and 20 minutes (Fig 4⇓). The speed of development of a cTnT-positive band depends on cTnT concentration in the blood. When cTnT levels are ≥2.4 μg/L, a band develops in all cases within 5 minutes. At 20 minutes, the results were negative in all patients with cTnT concentrations of <0.06 μg/L and positive in all patients with cTnT concentrations of >0.18 μg/L. In the range of 0.06 to 0.18 μg/L, the subjective assessment of a positive band varied, with more frequent positive results with increasing cTnT concentrations. On assessment of the same test strip by 12 different persons (interobserver variability), similar results were found. All persons agreed on a positive result if cTnT concentrations were >0.18 μg/L and on a negative result if cTnT concentrations were <0.06 μg/L. However, for concentrations between these values, there was a significant variability in test assessment of a cTnT concentration of 0.12; 65% of the persons rated the test device as positive and 35% rated it as negative.
No positive results were obtained in the 25 persons tested.
Patients Without Clinical Evidence of Acute Myocardial Cell Damage
In this group of 62 persons, there were no false cTnT rapid assay results except in two samples of 2 patients with CK elevations of 6500 and 10 000 IU/L (86.6- and 133.3-fold the upper limit of normal, respectively) (Fig 5⇓). In both patients, the cTnT determined by ELISA was elevated to 4.3 and 5.0 μg/L. Both patients had rhabdomyolysis due to viral infection. The ECG recordings revealed reversible ST-segment changes, but the echocardiogram remained normal. Thus, at the present time it is uncertain whether the positive cTnT results are true- or false-positive.
Patients With Definite Myocardial Cell Damage
The cTnT rapid assay was positive in all except 2 patients with cTnT of >0.18 μg/L and all except 1 patient with CK activity of >75 IU/L (0.85-fold the upper limit of normal) (Fig 6⇓). In 19 patients, the cTnT rapid assay was positive when CK activity showed normal values. In 4 of these 19 patients, blood samples were obtained more than 3 days later; in 7 patients, blood samples were obtained within 4 hours of the onset of AMI symptoms. In comparison to the cTnT enzyme immunoassay, the rapid assay was positive in all except 2 of TnT ELISA–positive patients. The cTnT concentrations in these two patients were 0.20 and 0.25 μg/L. One sample with cTnT concentrations of <0.1 μg/L was positive in the rapid assay. In the patients with AMI, the time delay from the onset of symptoms until all test results became positive was comparable for the rapid cTnT test and the cTnT ELISA. Less than 4 hours after the onset of symptoms, the cTnT ELISA, cTnT rapid assay, and CK were positive in 0 of 8, 1 of 8, and 0 of 8 patients, respectively. At 8 hours after the onset of pain, the respective results were 5 of 5, 5 of 5, and 4 of 5 patients.
Patients With Equivocal Myocardial Cell Damage
The cTnT rapid assay and cTnT ELISA was positive in 7 of 35 patients with unstable angina (Fig 7⇓). All positive patients were in Braunwald class III. There were 3 patients with elevated CK activity but normal cTnT; 2 of them had normal coronary angiograms.
Clinical Assessment of Sensitivity and Specificity of the Rapid Assay
The comparative analysis of all patients tested in this study revealed a sensitivity of 96.6% and a specificity of 96% if the cTnT value of <0.2 μg/L is considered to be indicative of myocardial cell necrosis. In the 99 patients with a cTnT value of <0.2 μg/L, 95 patients were TnT negative and 4 patients were TnT positive. In the 59 patients with cTnT ELISA of ≥0.2 μg/L, 57 patients had a positive and 2 patients had a negative rapid assay result.
A new test device was developed to provide a reliable, convenient, and rapid method of measuring cTnT. Three steps were critically important to achieve this goal. First, a glass-fiber fleece was incorporated in the device to allow separation of blood cells from plasma. Therefore, whole heparinized blood can be applied to the device, and the step of centrifugation can be omitted. Second, the antigen interaction with the antibodies adsorbed to the paper fleece below the application well starts immediately after application of the patient’s blood and continues throughout the entire process of blood cell separation from plasma and its diffusion to the reading zone. Third, the detecting antibody is not conjugated with an enzyme as in ELISA technology but instead is labeled with gold particles. These gold particles are visible after local concentration of the labeled antibodies, which is accomplished by the interaction of streptavidine linked to the cellulose nitrate membrane with the biotin-labeled capture antibodies.
The developed assay uses the same high-affinity antibodies that are used in the enzyme immunoassay test kit for cTnT. However, in the rapid assay device, the antibodies are switched in such a way that the cardiospecific antibody M7 is used as the labeled antibody, whereas the cross-reactive biotin-labeled antibody 1B10 serves as the capture molecule. Due to this specific arrangement of selected antibodies, skeletal TnT bound by the cross-reactive capture antibody will not be detected by the cardiospecific gold-labeled antibody. This improves the specificity of the assay, as could be documented through analysis of the cross-reactivity with both skeletal muscle homogenates and purified skeletal muscle TnT, which is as low as 0.5%. Even patients with significant skeletal muscle damage showing CK activities in blood of ≥30-fold the upper limit of normal skeletal muscle TnT did not have false-positive test results.
The developed rapid assay is a “yes-or-no” test that is analyzed by visual inspection. Therefore, the sensitivity limit of the rapid assay device is determined not only by the characteristics of test reagents but also by the variability of visual assessment as to the presence or absence of a cTnT-positive band at low cTnT concentrations in blood. This in part explains the variability of rating a cTnT result as positive or negative for cTnT concentrations between 0.1 and 0.18 μg/L. However, in the analytical assessment of rapid assay performance, at <0.1 μg/L, 5 of 5 test results were rated negative, and at >0.18 μg/L, 5 of 5 test results were rated positive, indicating a high reproducibility of the rapid assay device. When these discriminator values were applied prospectively in the clinical study groups, sensitivity and specificity were lower, most likely due to the variability of the results of cTnT ELISA at these low concentrations and the variability of visual assessment of the rapid assay results. The developed assay is a qualitative test; the speed of color development and the intensity of the cTnT band are dependent on the cTnT blood concentrations. With cTnT levels of >2 μg/L, a positive band is visible within 5 minutes, whereas at low concentrations, a faint band is not found until 20 minutes after the addition of blood.
The rapid assay device shows a higher sensitivity than the first evaluation lots assessed by us and others.17 With these first devices, we found the detection limit to be as high as 0.28 μg/L, which did not appear to be sufficient for the detection of micronecroses. The addition of a detergent (brij 35) to this rapid assay formulation resulted in increased sensitivity of the rapid assay device to 0.18 μg/L.
The sensitivity achieved with the improved rapid cTnT assay device is sufficiently high to be useful in clinical practice. Of the patients tested with suspected AMI (n=35) and minor cell damage after radiofrequency ablation (n=24), only 4 (6.7%) were observed for whom the cTnT ELISA was >0.1 μg/L and the rapid assay was rated negative. Furthermore, all patients with unstable angina and elevated cTnT by ELISA were also identified by the rapid cTnT assay. The relatively high number of patients with AMI and positive rapid cTnT assay results but negative results for CK activity are easily explained by an earlier appearance in circulation and the longer diagnostic time window of cTnT in comparison to CK. Therefore, our clinical data indicate that this rapid cTnT assay is comparable in its diagnostic efficiency to the ELISA for cTnT.
The diagnosis of massive AMI can be based on clinical symptoms and ECG recordings.1 18 19 20 21 22 In these patients, biochemical assays serve only to confirm this diagnosis. The major advantage of the cTnT rapid assay in this group of patients with a very high probability of AMI is the convenience and ease of testing. The rapid cTnT test device does not require 24 hours of laboratory service with the attending personnel and equipment and allows testing on the wards or in the ambulance by paramedics. It allows an almost instantaneous analysis of clinical findings, biochemical test results, and ECG recordings at the patient’s bedside. However, any biochemical marker needs a certain time to become detectable in circulation. As shown, the time delay from the onset of symptoms to a positive TnT result was similar for the rapid assay and ELISA method. cTnT was elevated in all patients tested at 8 hours after the onset of symptoms.
Only a minority of patients with chest pain, however, have clear evidence for AMI on ECG recordings. In the vast majority, the ECG is normal or reveals unspecific changes of the ST segment or T wave only. Some of these patients even have concomitant skeletal muscle injury, leading to unspecific elevations of infarct markers in blood. The correct detection or exclusion of acute myocardial cell damage in this large group of patients depends entirely on biochemical tests that must be both sensitive and specific. In previous trials, it was shown that cTnT is a sensitive marker for the detection of myocardial cell necrosis and is more specific than the measurement of CK-MB.7 8 9 The higher sensitivity of cTnT compared with the golden standard, CK-MB, will yield positive test results in patients not classified as having AMI according to World Health Organization criteria. As a consequence, the specificity of the more sensitive cTnT measurement will appear to be lower than that of the cardiac enzyme measurements that are part of the classification criteria for AMI. However, the detection of minor myocardial injury is clinically important as we and others showed that patients with unstable angina and elevated cTnT represent a high-risk subgroup, whereas patients with a negative cTnT result have a more favorable prognosis.7 22 23 24 25 26 27 With the developed assay, this group could be identified even when the ECG and cardiac enzymes yield equivocal results. Most likely, the patients with chest pain and elevated cTnT should be monitored on coronary care units as they have a high risk for cardiac events. On the contrary, patients with a negative cTnT finding during the first 12 hours after onset of symptoms could be discharged from a coronary care to a regular ward bed or discharged from the chest pain unit for ambulatory follow-up.
The optimal treatment strategy for patients with minor myocardial cell damage is unknown. However, based on cTnT measurements, this group of patients could be identified, and new treatment strategies in these patients could be evaluated in randomized clinical trials.
Selected Abbreviations and Acronyms
|AMI||=||acute myocardial infarction|
|cTnT||=||cardiac troponin T|
Dr Katus invented the troponin T assay that is the subject of this article. This assay is patented in Europe and Japan; a patent is pending in the United States.
- Received February 22, 1995.
- Revision received June 26, 1995.
- Accepted June 28, 1995.
- Copyright © 1995 by American Heart Association
Yusuf S, Pearson M, Sterry H, Parish S, Ramsdale D, Rossi P, Sleight P. The entry ECG in the early diagnosis and prognostic stratification of patients with suspected acute myocardial infarction. Eur Heart J. 1984;5:690-696.
Rude RE, Poole WK, Muller JE, Turi Z, Rutherford J, Roberts R, Raabe DS, Gold HK, Stone PH, Willerson JT, Braunwald E, and the Milis Study Group. Electrocardiographic and clinical criteria for recognition of acute myocardial infarction based on analysis of 3,697 patients. Am J Cardiol. 1983;52:936-942.
Turi ZG, Rutherford JD, Roberts R, Muller JE, Jaffe AS, Rude RE, Parker C, Raabe DS, Stone PH, Hartwell TD, Lewis SE, Parkey RW, Gold HK, Robertson TL, Sobel BE, Willerson JT, Braunwald E, and Cooperating Investigators from the Milis Study Group. Electrocardiographic, enzymatic and scintigraphic criteria of acute myocardial infarction as determined from study of 726 patients (a Milis Study). Am J Cardiol. 1985;55:1463-1468.
Gibler WB, Young GP, Hedges JR, Lewis LM, Smith MS, Carleton SC, Aghababian RV, Jorden RO, Allison EJ Jr, Otten EJ. Acute myocardial infarction in chest pain patients with nondiagnostic ECGs: serial CK-MB sampling in the emergency department: the Emergency Medicine Cardiac Research Group. Ann Emerg Med. 1992;21:504-512.
Katus HA, Remppis A, Neumann FJ, Scheffold T, Diederich KW, Vinar G, Noe A, Matern G, Kübler W. Diagnostic efficiency of troponin T measurements in acute myocardial infarction. Circulation. 1991;83:902-912.
Gerhardt W, Katus HA, Ravkilde J, Jörgensen PJ, Hamm C, Peheim E. S-troponin T and mass concentration S-CKMB as markers for ischemic myocardial injury. Clin Chem. 1991;37:1405-1411.
Katus HA, Looser S, Hallermayer K, Remppis A, Scheffold T, Borgya A, Essig U, Geuß U. Development and in vitro characterisation of a new immunoassay of cardiac troponin T. Clin Chem. 1992;38:386-393.
Eisenberg E, Kielley WW. Troponin-tropomyosin complex. J Biol Chem. 1974;249:4742-4748.
Staprans J, Takahashi H, Russell MP, Watanabe S. Skeletal and cardiac troponins and their components. J Biochem. 1992;72:723-735.
Roth J. The colloidal gold marker system for light and electron microscopic cytochemistry. In Bullock GR, Petrusz P, eds. Techniques in Immunocytochemistry, Vol 2. New York, NY: Academic Press; 1983:216-284.
Braunwald E. Unstable angina. Circulation. 1989;80:410-414.
Antman E, Grudzien C, Sacks D. A rapid bedside assay for the detection of cardiac troponin T. Circulation. 1994;90(pt 2):1497. Abstract.
McComb JM, McMaster EA, MacKenzie G, Adgey AAJ. Myoglobin and creatine kinase in acute myocardial infarction. Br Heart J. 1984;51:189-194.
Jablonsky G, Leung FY, Henderson AR. Changes in LD1/LD2 ratio during the first day after myocardial infarction. Clin Chem. 1985;31:1621-1624.
Mair J, Artner-Dworzak E, Lechleitner P, Smidt J, Eagner R, Dienstl F, Puschendorf B. Cardiac troponin T in diagnosis of acute myocardial infarction. Clin Chem. 1991;37:845-852.
Timmis AD. Early diagnosis of acute myocardial infarction. BMJ. 1990;301:941-942.
Ravkilde J, Horder M, Gerhardt W, Ljungdahl L, Petterson T, Tryding N, Moller BH, Hamfelt A, Graven T, Asberg A, Helin M, Pentillä I, Thygesen K. Diagnostic performance and prognostic value of serum troponin T in suspected acute myocardial infarction. Scand J Clin Lab Invest. 1993;53:677-685.
Johnson PA, Albano MP, Sacks D, Orav JE, Goldman L, Lee TH. Troponin T in patients with active chest pain. J Am Coll Cardiol. 1994;(suppl A):412A. Abstract.