 |
|
The Troponins
Laura Coudrey, MD
Arch Intern Med. 1998;158:1173-1180.
INTRODUCTION
New cardiac-specific markers have recently been introduced into the arena of cardiac diagnostic tests. The cardiac troponins, because of their extraordinary high specificity for myocardial cell injury, have gained particular interest. They can be used in a variety of clinical situations, including differentiation of skeletal from cardiac muscle injury; detection of minor myocardial cell damage in coronary insufficiency syndromes, which allows stratification of patients into high- and low-risk categories; detection of perioperative myocardial infarction; estimation of infarct size; and assessment of therapeutic success of reperfusion therapy. The application of the troponin proteins in relation to percutaneous transluminal coronary angioplasty and in the diagnosis of acute myocarditis has also been recently described.
BIOCHEMICAL MAKEUP
The troponins are 3 distinct proteins (I, C, and T) that are expressed in cardiac and skeletal muscle and are encoded for by different genes. These molecules form a complex that regulates the calcium-dependent interaction of myosin with actin. The troponin C that is expressed in cardiac muscle is identical to the troponin C that is expressed in skeletal muscle tissue, which limits its clinical usefulness.1 The genes encoding for cardiac and skeletal troponin I and T are different; accordingly, the pronounced divergence of amino acid composition between cardiac and skeletal troponins I and T allows the differentiation of these molecules by immunological techniques.2 Thus, recently developed monoclonal antibodies to cardiac troponins I and T have no or trivial cross-reactivity with their skeletal muscle forms.2-3 A sensitive enzyme-linked immunosorbent assay has been developed and is currently available for clinical use.2, 4-6
IMMUNOASSAYS
Cardiac troponin T is measured by a newly developed immunometric 1-step sandwich assay.7 In this assay, an affinity-purified, cardiospecific anti–troponin T fraction of polyclonal antibody is immobilized on polyvinyl chloride test tubes. Patient serum and peroxidase-labeled monoclonal anti–troponin T antibody are added to these antibody-coated test tubes. During the incubation period, the troponin T molecule is bound on different epitopes by both the solid-phase polyclonal antibody fraction and the liquid-phase monoclonal antibody–enzyme complex. After the unbound peroxidase-labeled monoclonal antibodies are removed by washing, the antibody-enzyme complex adhering to the assay tubes corresponds to the amount of troponin T recognized by the polyclonal and monoclonal anti–troponin T antibodies. The amount of enzyme immobilized, as a direct measure of bound troponin T, is quantified in the spectrophometer as peroxidase substrate conversion at a wavelength of 405 nm.8 The measuring range of this assay extends from 0.5 to 25 µg/L. There is a 1% to 2% nonimmunological cross-reactivity of the assay with troponin T extracted from human or bovine mixed skeletal muscle resulting from nonspecific absorption of purified skeletal troponin T to the test tubes. The discriminating value between undetectable and elevated troponin T values is defined at 0.5 µg/L. Results are obtainable in 85 minutes at an estimated cost of $17.48 per reportable result.9 A second-generation troponin T enzyme-linked immunosorbent assay has recently been developed.10 Muller-Bardorff et al10 were able to improve specificity using a high-affinity cardiac-specific antibody, M11.7, and lower the discriminating value to 0.1 µg/L with a turnaround time of 45 minutes. Alternatively, a rapid, qualitative, bedside immunoassay for cardiac troponin T that uses a handheld device has been developed for clinical use.4 This test is also based on a dual monoclonal antibody "sandwich" principle that uses a streptavidin-biotin capture system. It is initiated by the addition of 150 µL of whole blood to a sample well in the device that separates red blood cells from plasma. Cardiac troponin T in the patient's plasma combines with both biotinylated anti–troponin T antibodies and gold-labeled anti–cardiac troponin T antibodies to form a sandwich. The biotin adheres to streptavidin, which is immobilized in a line across the "read" zone of the device. If cardiac troponin T is present in the patient's blood, the gold particles in the sandwich produce a red or purple line within 20 minutes that is read visually. A positive result is obtained in patients with 0.2 ng/mL or more of cardiac troponin T in their blood, at an estimated cost of $21.65 per reportable result.4, 9
Similarly, a sensitive fluorometric sandwich enzyme immunoassay has been developed to measure cardiac troponin I.6 This assay uses 2 monoclonal antibodies that recognize 2 different epitopes on the cardiac troponin I molecule. Calculated cross-reactivity with human skeletal troponin I is less than 0.1%. In serum specimens from healthy persons without evidence of cardiac disease, the cardiac troponin I concentration is below the minimal concentration detectable by the assay, or 0.35 ng/mL; ie, levels greater than this discriminator value indicate myocardial necrosis.11 A 20-minute time interval is necessary to run the assay, at an estimated cost of $6.38 per reportable result.
DIAGNOSIS WHEN SKELETAL MUSCLE INJURY IS PRESENT
Although tests for the muscle-brain (MB) isoenzyme of creatine kinase (CK) have served physicians' purposes for nearly 3 decades, increases in CK-MB fractions are not as specific for myocardial injury as first believed. It is important to be aware that some noncardiac conditions can cause elevation of CK-MB levels in the peripheral blood, which can be confused with myocardial cell injury.12-13 The MB isoenzyme of creatine kinase is normally present in skeletal muscle in low amounts (1%-3%). Substantial injury to skeletal muscle can increase CK-MB activity by enhancing the production of B subunits, and elevate values to abnormal levels.13 Particularly with injuries involving the chest and thorax, it can be difficult to distinguish true myocardial damage from concomitant skeletal muscle injury with the currently available techniques. Furthermore, assays that detect B subunits can measure elevated levels that are the result of brain injury; pregnancy; carcinoma of the prostate, lung, or gastrointestinal tract; tuberculosis; or uterine abnormalities.12-13 Pierse and Jaffe14 described a 67-year-old man with a history of triple-vessel coronary bypass surgery with persistently elevated CK-MB fractions despite the absence of electrocardiographic changes, chest pain, or hemodynamic instability. His only complaint was muscle weakness, which subsequently was diagnosed as a procainamide hydrochloride–induced lupus myositis. Discontinuation of the procainamide therapy gradually returned the CK-MB concentrations to within the normal reference interval. Hypothyroidism, chronic renal failure, and alcoholism can also cause increased values of serum CK-MB as a result of myopathies associated with these diseases.14-15 Likewise, other inflammatory myopathies, such as polymyositis and dermatomyositis, can raise CK-MB values to abnormal levels, leading to diagnostic confusion when patients complain of chest tightness or associated symptoms.14-16 Finally, the interference by fluorescent compounds, which create artifactual bands with electrophoretic assays, can elevate CK-MB levels without underlying myocardial cell necrosis in patients with renal failure.14
It appears that molecular markers expressed in skeletal muscle during fetal development are often reexpressed after muscle injury (eg, CK-MB).5 Cardiac troponin I is not expressed in skeletal muscle throughout ontogeny. After the ninth postnatal month, it is expressed only in myocardium.13 As far as can be determined, skeletal muscle in humans does not express cardiac troponin I at any developmental stage or in response to any pathological stimuli.5 The specificity of cardiac troponin I for myocardium is high, yet its sensitivity for cardiac injury appears to be comparable to that of CK-MB. In a study by Cummins et al,3 immunochemical cross-reactivity with skeletal troponin I was only 2%, and the results were nonspecific. In 34 subjects with skeletal muscle damage, cardiac troponin I levels were not elevated above normal, although CK-MB isoenzyme levels were elevated 4 times higher than normal in some patients. Furthermore, skeletal muscle cross-reactivity has decreased to less than 0.1% with the introduction of newer radioimmunoassays.6 Adams et al5 studied 215 patients with acute skeletal muscle injury or chronic muscle disease, including patients with Duchenne muscular dystrophy, polymyositis, and chronic myopathy, as well as well-trained marathon runners and patients receiving chronic dialysis. Elevations of total creatine kinase levels were common, and elevations of CK-MB levels occurred in 59% of patients with acute muscle injury, 78% of patients with chronic muscle disease and marathon runners, and 3.8% of patients with chronic renal failure. Only the 6 patients with echocardiographically proved myocardial infarction, demonstrated by significant wall-motion abnormalities, had elevated cardiac troponin I levels.
Thus, the presence of cardiac troponin I in the circulation above the reference limit is highly specific for myocardial injury and allows CK-MB elevations resulting from skeletal muscle damage to be distinguished from those resulting from cardiac injury.13 Elevations of cardiac troponin I levels do not occur in patients with acute or chronic muscle disease or in patients with renal failure unless concomitant cardiac damage has occurred. Cardiac troponin I is therefore unique among molecular markers and ideally suited for the detection of myocardial necrosis in these complex situations.
Cardiac troponin T has also been found to be a sensitive marker of myocardial cell necrosis. However, its specificity has not been fully defined.5 In contrast to cardiac troponin I, cardiac troponin T is expressed in fetal skeletal muscle and is reexpressed in adult rat skeletal muscle after injury.13 Since the troponin system is highly conserved across species, cardiac troponin T could be reexpressed in injured human skeletal muscle, with a decrease in its specificity for myocardial necrosis. This suggestion is supported by Kobayashi et al,17 who have found increased plasma levels of cardiac troponin T in the absence of evidence of myocardial involvement in patients with polymyositis. In 2 studies in which assays were based on polyclonal antibodies with 1% to 3% cross-reactivity with skeletal troponin T, 15% of patients with skeletal muscle disease had borderline elevation of cardiac troponin T concentration without any evidence of myocardial injury.8, 18 However, the specificity of troponin T when compared with CK-MB was 84% vs 49%, respectively. In both studies, the discriminator value of 0.5 µg/L was used. Specificity of the troponin T test was improved to 95%, without loss in sensitivity, when the discriminator value was 1 µg/L. Katus et al,2 using the same monoclonal antibody–based assay for cardiac troponin T, found only minor cross-reactivity (<0.5%). Three of 14 patients with skeletal muscle damage with normal electrocardiographic findings and low CK-MB fractions (<4%) had increased concentrations of cardiac troponin T. Katus and colleagues attributed this to concomitant cardiac muscle damage not detected by cardiac enzyme measurements or to a false-positive test result secondary to minor cross-reactivity. Nevertheless, spurious rises in troponin T concentrations have been reported in patients with diverse underlying clinical problems, such as rhabdomyolysis, renal failure, septicemia, and widespread malignancy, as well as in the aforementioned conditions, without additional supportive evidence of myocardial injury.17, 19-22 Although the newer troponin T enzyme-linked immunosorbent assay has proved to be more specific than previous assays, patients with renal failure and muscular dystrophy were found to have elevated troponin T levels despite the absence of cardiac injury.10 Consequently, the use of the more cardiospecific troponin I proteins may be warranted to delineate concomitant cardiac muscle injury in patients with intricate comorbidity.
DIAGNOSIS OF ACUTE MYOCARDIAL INFARCTION
Early identification and confirmation of acute myocardial infarction is essential for correct patient care and disposition decisions. Often, patients present with atypical signs and symptoms and nondiagnostic electrocardiograms.23 Additionally, a rise in CK activity above the upper limit of the normal range is rarely found before 4 to 6 hours have elapsed after the onset of chest pain. Accordingly, the unselected treatment of such patients or of patients with unstable angina has not been shown to improve outcome.1 The overall sensitivity of the standard CK-MB assay still appears to be inadequate for clinical decision making in the emergency department. The troponins may be useful to confirm a suspected myocardial infarction, aid in the diagnosis in difficult cases, and assist in the triaging of patients with chest pain, a procedure for which demand is increasing.23 Because of the peculiar characteristics of these proteins, they can be used in the evaluation of subtle myocardial damage24 and serve as prognostic indicators.11, 25 Furthermore, the success of thrombolytic therapy may be monitored and myocardial infarct size can be estimated using currently available assays. The rapid cardiac troponin T immunoassay requires no specialized laboratory equipment, is easily performed by medical or paramedical personnel, can be performed at the patient's bedside, uses whole blood, and is completed within 20 minutes.4
Mair et al23 studied 37 patients with acute myocardial infarctions and compared early sensitivities of several biochemical markers. Cardiac troponin I and T levels increased significantly (P .07) earlier and were markedly more sensitive (P.05) than CK and CK-MB activity before or after thrombolytic treatment. Troponin T appeared in the serum as early as 3 hours after the onset of chest pain in 50% and 60% of patients in a study by Katus et al8 and Wu et al,26 respectively. This relatively early release could be the result not only of rapid myofibrillar breakdown, but also of rapid loss of a cytoplasmic pool of troponins that have not yet been incorporated into the myofibril.3 Interestingly, levels of troponin T remained elevated for more than 130 hours, which is more than 4 times longer than those of total serum CK.8 High troponin concentrations persist for at least 5 days, despite its biological half-life of 120 minutes, reflecting a continuing release of this protein from disintegrating myofilaments.24 Consequently, there is a wide diagnostic window, permitting both very early and late detection of myocardial infarction after the onset of chest pain.
DETECTION OF REPERFUSION
The presence of the combination of relief of chest pain, normalization of ST-segment changes, and reperfusion arrhythmias has been found to be indicative of coronary recanalization, but these criteria are not sensitive, because all 3 occur in only 15% of patients.13 A retrospective assessment of the efficacy of reperfusion can be performed, based on CK and CK-MB washout curves, which can also provide a rough estimate of the extent of cell injury. More specific parameters of cell death are provided by the measurement of the myofibrillar troponin proteins, which exist in a large structurally bound pool, but in low cytosolic concentrations.1 This compartmentalization is responsible for the biphasic release pattern, most pronounced for troponin T, that is seen after successful reperfusion (Figure 1, top).24 The cytosolic pool is rapidly washed out and is responsible for the first distinct peak, occurring after approximately 14 to 16 hours. A long plateau follows (until day 5 after acute myocardial infarction) or a second peak value occurs during day 4 (around 96 hours) (Figure 1, inset), representing disintegrating myofibrils.24 In patients who have not undergone reperfusion, troponin T levels increase unsteadily (Figure 1, bottom). Furthermore, biochemical assessment of reperfusion therapy would be most valuable at times when mechanical intervention, such as rescue angioplasty, could still be of benefit to the patient.1 In an angiographically controlled study, Abe et al27 demonstrated that serial measurements of troponin T are excellent indexes of reperfusion at 1 hour (Figure 2). The kinetics of troponin T release differ markedly in relation to reperfusion and can therefore be used to indirectly assess coronary artery patency. Similarly, cardiac troponin I has been shown to be advantageous for the early, noninvasive determination of coronary reperfusion following thrombolytic therapy. With documented angiography at 90 minutes, Apple et al28 demonstrated a significantly larger percentage of change from baseline value in cardiac troponin I concentration with reperfusion as compared with concentrations of CK-MB and myoglobins (Figure 3). They found the sensitivity for detecting reperfusion at 90 minutes for the cardiac troponin T, myoglobin, and CK-MB levels to be 82.4%, 76.5%, and 64.7%, respectively.
|
|
|
|
Figure 1. Serum concentration changes of cardiac troponin T in patients with reperfused myocardial infarction (top) and patients with nonreperfused acute myocardial infarction (bottom). The inset shows a biphasic release of cardiac troponin T in a case of reperfused acute myocardial infarction (modified from Burlina et al24).
|
|
|
|
|
|
|
Figure 2. Time course of serum troponin T release after reperfusion (R) by means of percutaneous transluminal coronary angioplasty (PTCA) or no reperfusion. Rapid increase in troponin T gives early evidence of successful reperfusion (modified from Abe et al27).
|
|
|
ESTIMATION OF INFARCT SIZE
The evaluation of infarct size after acute myocardial infarction is important for predicting the subsequent clinical course, since it reflects the reduction of left ventricular function and the risk of ventricular arrhythmias.29 Infarct size has been traditionally based on an estimation of the increase in serum CK enzyme levels during the acute phase of the infarction.30 Because the serum CK level changes rapidly during the acute phase of infarction, frequent blood sampling is required. Furthermore, if the patient is admitted more than 24 hours after the onset of myocardial infarction, the peak CK level may be missed, making it difficult to calculate infarct size by this method. The late troponin T peak, however, may be used to estimate infarct size under a greater variety of conditions in comparison to CK-MB. Omura and colleagues31 measured late peak troponin T concentration occurring on the third to fifth day after acute myocardial infarction in a small study involving 34 patients. Left ventriculography, 2-dimensional echocardiography, and resting thallium-201 myocardial single photon emission computed tomography were performed approximately 4 weeks after the onset of myocardial infarction and were used for correlation with the late serum troponin T concentration. Both the left ventricular ejection fraction obtained from left ventriculography and the wall-motion index obtained from 2-dimensional echocardiography were inversely related to the late troponin T peak value. Extent scores and severity scores, which were estimated from the initial resting thallium-201 single photon emission computed tomographic image, showed excellent linear correlation with the latter. Because the late elevation of serum troponin T levels appears to be due to degradation of the contractile apparatus, it is correlated with infarct size and is independent of whether or not coronary reperfusion occurred. In fact, this correlation was present both in patients with an early troponin T peak on day 1 and in patients without an early peak. Therefore, a single troponin T level may be useful for estimating infarct size in a wide variety of circumstances.31
RISK STRATIFICATION
The clinical spectrum of patients arriving at the hospital with acute myocardial ischemic syndromes represents a continuum of disease from unstable angina to acute infarction. The extent of myocardial necrosis is an important determinant of the risk for increased mortality.11 The troponins are sensitive and specific markers for even small amounts of myocardial necrosis,11, 25, 32 probably more so than their CK-MB counterparts,33-35 and the measurement of levels drawn within 24 hours after the patients are admitted to the hospital may be used to stratify those who are at risk for future cardiac events. Several small studies36-39 have demonstrated an increased number of cardiac events in patients with elevated troponin T levels, even in those without elevated CK-MB fractions. More recently, 855 patients who presented within 12 hours of the onset of acute myocardial ischemia were studied in a large prospective analysis.25 Cardiac troponin T and CK-MB levels were quantified on admission. The investigators found that elevated troponin T levels (>0.1 µg/L) were associated with significantly higher mortality rates within 30 days, both in the total study population (11.8% vs 3.9%) and in all electrocardiographic subgroups examined, including patients with ST-segment depression or elevation, T-wave inversion, and electrocardiographic features that confounded the detection of ischemia (such as bundle-branch block and paced rhythms). Antman et al11 published similar findings. They measured serum levels of cardiac troponin I in 1404 patients with unstable angina and non–Q-wave myocardial infarction. Cardiac troponin I levels of 0.4 mg/mL or above were associated with a significantly higher mortality rate within 42 days than were lower levels (3.7% vs 1.0%). Moreover, significant increases in mortality rates were found with increasing levels of cardiac troponin I.
Elevated troponin levels indicated severe coronary artery narrowing (>75% obstructions of a major coronary artery), with high specificity ( 90%) demonstrated angiographically on day 6 of hospital admission8; however, the negative predictive value was low (0.23%); ie, some patients without circulating troponin T were also found to have severe coronary artery disease. In this regard, Antman et al11 did not find significant differences in the number of diseased coronary vessels or in the incidence of coronary thrombi on angiography between patients who had increased levels of troponin I and those who did not. However, in their study, only a minority of patients underwent coronary angiography in the first few hours after their symptoms began. These findings may reflect the liability of thrombus formation at the unstable plaque, leading to occlusion of the coronary artery and thereby explaining the troponins' relation to future cardiac events rather than to lumen diameter or number of vessels diseased.32
Patients who are stratified to a high-risk group may benefit from aggressive treatment that includes early revascularization as opposed to those at low risk for subsequent events, who can be treated more conservatively.30 The Fragmin During Instability in Coronary Artery Disease (FRISC) Study has already suggested that measuring cardiac troponin T levels may have therapeutic value in patients with acute coronary syndromes.32 In that study, a subgroup of 644 patients with increased levels of cardiac troponin T on admission were identified in whom long-term treatment with low-molecular-weight heparin was associated with an impressive 48% reduction in 42-day mortality rates, compared with a group of 327 patients without troponin T elevations in whom this treatment had no benefit. Again, it is possible that elevated cardiac troponin levels may be associated with unstable coronary plaques, a hypothesis that needs to be evaluated prospectively in studies correlating the morphological features of plaque with cardiac troponin levels in patients with acute coronary syndromes.40
DETECTING PERIOPERATIVE MYOCARDIAL INFARCTIONS
Myocardial infarctions that occur during or after vascular and nonvascular surgery are associated with increased mortality and poor functional outcome. The use of CK or CK-MB determinations to aid in diagnosing acute myocardial infarction and assessing the extent of myocardial cell injury is limited because of the frequent increases in the levels of these enzymes owing to some degree of skeletal muscle damage. Measuring the more cardiospecific troponins could eliminate the great number of false-positive results and thereby improve diagnostic accuracy.1 In an echocardiographically controlled study, the concentration of troponin I was demonstrated to be superior to that of CK-MB for the detection of perioperative infarcts.41 Similar findings have been reported for troponin T.42-45 A single measurement of the serum level of troponin T obtained on the fourth postoperative day was highly sensitive for perioperative infarctions and probably represents the continuous release of cardiac troponin T from necrotic cardiac muscle cells.42-43 The electrocardiogram is only useful for detecting major acute myocardial infarctions; however, small postoperative myocardial infarction incidents may be identifiable with serum troponin T levels. Since the amount of myocardium that is damaged appears to correlate with patients' functional outcome, it is important to quantify such loss. Moreover, prolonged aortic cross-clamp time may result in perioperative ischemic death of cardiac myocytes, with only minor electrocardiographic findings but with elevated troponin T levels.42-43 Intraoperative variables, such as the extent of surgical procedure and the duration of bypass and ischemia (cross-clamping), have been shown to correlate with serum troponin T concentrations.42-43 Thus, the serum troponin levels may be used to identify and measure perioperative myocardial cell damage and may be useful in assessing the efficiency of cardioprotective measures.
APPLICATION IN CORONARY ANGIOPLASTY
Percutaneous transluminal coronary angioplasty is a well-established technique for myocardial revascularization.46 Myocardial infarction, as diagnosed by elevation of cardiac enzymes after percutaneous coronary interventions, is relatively common and reported in approximately 8% to 15% of patients undergoing percutaneous transluminal coronary angioplasty (PTCA).47 These infarctlets were previously thought to be relatively benign, resulting in no permanent sequela48-49; however, more recent trials confirm an increased risk for future cardiac events in patients with elevated postintervention myocardial isoenzyme levels.47, 50-53 There is controversy as to the source of mild isoenzyme elevations of CK and CK-MB activity in patients undergoing PTCA, but such elevations have been linked to distal microembolization, side branch occlusion, multivessel procedures, abrupt vessel closure, directional coronary atherectomy, and interventions on vein grafts.52-55 Enzyme elevations, in the setting of percutaneous interventions, may or may not reflect irreversible ischemia.56-60
The troponin proteins have proved to be more sensitive and specific markers of myocardial necrosis. Ravkilde et al61 uncovered myocardial ischemia and/or minor myocardial damage that was inapparent by ambulatory electrocardiography in 26% and 13% of 23 patients with elevated CK-MB mass and cardiac troponin T measurements, respectively, after successful PTCA. Five of 6 patients, however, with increased levels of cardiac troponin T had pronounced chest pain during the procedure. Perioperative side branch occlusions were not seen. Hunt et al62 failed to demonstrate any significant elevations of CK-MB or cardiac troponin T levels on immunoassay after successful angioplasty in 22 patients, regardless of the instability of the angina, the severity of the stenosis, number of target lesions, number and duration of inflations, and occurrence of chest pain during the procedure. Talasz et al63 also could not demonstrate a significant increase in CK-MB or cardiac troponin T concentrations in 16 patients undergoing uncomplicated PTCA but correlated an increase in CK-MB mass and cardiac troponin levels in 5 and 3 patients, respectively, to the occurrence of side branch occlusions. Genser et al64 investigated 52 patients undergoing PTCA and found elevated cardiac troponin T levels in 13 patients (25%), all of whom experienced either a m |
cTnI), muscle-brain isoenzyme of creatine kinase (