Author + information
- Received October 25, 1999
- Revision received April 11, 2000
- Accepted June 15, 2000
- Published online November 1, 2000.
- Thomas Meyer, MD, PhD∗,*,
- Lutz Binder, MD†,
- Nadine Hruska∗,
- Hilmar Luthe, PhD† and
- Arnd B Buchwald, MD∗
- ↵*Reprint requests and correspondence: Dr. Thomas Meyer, Center of Internal Medicine, University of Göttingen, Robert-Koch-Str. 40, D-37075 Göttingen, Germany
The purpose of this study was to evaluate the prevalence and diagnostic utility of cardiac troponin I to identify patients with right ventricular (RV) dysfunction in pulmonary embolism.
Right ventricular overload resulting from elevated pulmonary resistance is a common finding in major pulmonary embolism. However, biochemical markers to assess the degree of RV dysfunction have not been evaluated so far.
In this prospective, double-blind study we included 36 study patients diagnosed as having acute pulmonary embolism.
Among the whole study population, 14 patients (39%) had positive troponin I tests. Ten of 16 patients (62.5%) with RV dilatation had increased serum troponin I levels, while only 4 of 14 patients (28.6%) with elevated troponin I values had a normal RV diameter as assessed by echocardiography, indicating that positive troponin I tests were significantly associated with RV dilatation (p = 0.009). Patients with positive troponin I tests had significantly more segmental defects in ventilation/perfusion lung scans than patients with normal serum troponin I (p = 0.0002).
Our data demonstrate that more than one-third of patients clinically diagnosed as having pulmonary embolism presented with elevated serum troponin I concentrations. Troponin I tests helped to identify patients with RV dilatation who had significantly more segmental defects in lung scans. Thus, troponin I assays are useful to detect minor myocardial damage in pulmonary embolism.
Pulmonary embolism still remains a major cause of cardiovascular death, despite new therapeutic strategies to prevent hemodynamic deterioration, such as thrombolysis or pulmonary embolectomy. Thromboembolic events are difficult to distinguish from other cardiovascular and respiratory syndromes due to nonspecific symptoms and diagnostic signs, suggesting that pulmonary embolism is often undetected and treated with inappropriate therapeutic procedures (1–6). Although the exact incidence of pulmonary embolism is unknown due to its variable clinical presentation and degree of hemodynamic instability, this disorder is usually considered to have a remarkable morbidity (2–4,6,7). In order to commence aggressive treatment without delay, it is a great challenge for treating physicians to diagnose this common disorder promptly. The high mortality of major acute thromboembolic events is generally accepted to be associated with sudden right ventricular (RV) dysfunction (2–4,6,7). As a result of pulmonary arterial obstruction and elevated pulmonary vascular resistance, the mechanical afterload in the RV increases, and the tension in the ventricular wall rises to pathological levels, leading to the acute dilation of the RV. Right ventricular dilation and hypokinesis in the presence of normal or hypotensive arterial pressure may lead to myocardial ischemia.
Cardiac-specific isoforms of troponin have been reported to be sensitive biochemical markers for minor myocardial damage (8–17). Troponin I and troponin T function together as essential components of the contractile apparatus in striated muscle and are released in the blood flow as a result of myocardial necrosis (8,18). Numerous studies have demonstrated that measurements of cardiac troponins are useful for the objective risk assessment in patients presenting with unstable angina pectoris and myocardial infarction (8–15,17). However, the diagnostic value of troponin measurements in pulmonary embolism is unknown because, so far, no systematic investigation has evaluated the prevalence of elevated serum concentrations that result from acute RV overload. Thus, the aim of this study was to test the hypothesis that a recently developed high-sensitive assay for the detection of cardiac troponin I in serum samples could identify patients with severe embolic episodes.
This study was conducted over an eight-month period from January 1999 at the University Hospital Göttingen. The subjects of this prospective, monocenter trial consisted of patients finally diagnosed as suffering from acute pulmonary embolism. Patients with symptoms of suspected pulmonary embolism, such as arterial hypotension, tachypnea/dyspnea, syncope, arterial hypoxemia or cardiogenic shock, were included in the study if at least one of the following criteria was met: a positive diagnostic pulmonary angiogram, a lung scan indicating intermediate or high probability for the diagnosis of pulmonary embolism, precapillary pulmonary hypertension as assessed by right heart catheterization, or echocardiographic signs of RV dilatation with normal left ventricular systolic function. Patients were not included in the study if the clinical symptoms were judged to be caused by either sepsis, hypovolemia, or acute respiratory failure. Particular efforts were undertaken not to include patients suffering from coronary heart disease (CHD). Thus, patients admitted with angina pectoris were strictly excluded, as were all patients with electrocardiographic findings suggesting ischemic heart attack.
Study patients were recruited from either the emergency department, both medical and surgical wards, or the intensive care units. Data from the history, physical examination and electrocardiograms were obtained from the medical records. No attempt was made by the investigators to influence the process of diagnosing pulmonary embolism in patients presenting with symptoms clinically suggestive of thromboembolic episodes. Treating physicians were completely blinded to the serum levels of troponin I. Patients remained under the medical care of the physicians, who decided the further therapeutic management. Written informed consent to participate in the study was obtained from all patients after the nature of the investigation had been fully explained. The study protocol was approved by the Human Ethics Committee at the university where this investigation was carried out.
Blood samples and laboratory methods
Blood samples were obtained at enrollment, immediately after pulmonary embolism was diagnosed. Samples were centrifuged, and aliquots were frozen at −70°C for subsequent measurements of cardiac troponin I and D-dimers. The measurements of serum markers were performed by laboratory personnel blinded to the clinical data. Cardiac troponin I concentrations were measured using the ACS:180 cTnI assay from Bayer Diagnostics (Fernwald, Germany). This is a two-site sandwich immunoassay based on an automated chemiluminescence detection system. The assay uses constant amounts of polyclonal goat anti–troponin I antibody labeled with acridinium ester and a combination of different monoclonal mouse anti–troponin I antibodies covalently coupled with paramagnetic particles. According to the manufacturer, the assay measures troponin I concentrations up to 50 μg/L, with a minimum detectable concentration of 0.15 μg/L. All incubations, pipetting and measurement steps were performed automatically on the ACS:180 analyzer. The complete procedure requires approximately 10 min to obtain the test result. Blood samples that were analyzed for troponin I were also examined for D-dimers using an enzyme-linked immunosorbent assay from Roche Diagnostics, Mannheim (Germany). For the measurements of creatine kinase (CK), N-acetylcysteine-activated assays from Roche Diagnostics were performed on a Hitachi 917 analyzer. An immuno-inhibition assay based on the presence of antibodies to the CK-M subunit was used to determine the concentrations of the CK-MB isoenzyme.
All consecutive patients in whom the diagnosis of pulmonary embolism was made underwent echocardiographic examination. The transthoracic echocardiography was standardized using the apical, parasternal, subcostal and, occasionally, the suprasternal approaches. The echocardiographic criteria of major embolic events included pulmonary hypertension, RV afterload stress and tricuspidal regurgitation. “Right ventricular dilation” was defined as a diastolic diameter >30 mm. Pathological changes in serial 12-lead electrocardiograms were recorded such as ST-depressions, right precordial T-wave inversions, complete right bundle branch blocks (RBBBs), atrial fibrillation, S-waves in lead I and aVL and right axis deviation. The blood pressure in the pulmonary artery was determined in a minority of study patients using a Swan-Ganz catheter. Ventilation/perfusion lung scans were performed in all patients presenting with symptoms suggestive of pulmonary embolism. For ventilation studies, xenon 133 was used; and for perfusion studies, technetium 99m macroaggregated to albumin was used. A “positive lung scan” was defined as a ventilation-perfusion mismatch showing segmental and lobar filling defects.
All results are expressed as means ± standard deviations. The differences between the two troponin groups with the diagnostic parameters were analyzed using Pearson chi-square tests for categorial variables. Student t tests for continuous variables were calculated to determine the statistical significance of the partial pressure of O2 and the number of segmental defects in lung scans in predicting troponin elevations. A p value <0.05 was considered to be statistically significant.
Clinical findings at diagnosis
A total of 36 patients diagnosed as suffering from pulmonary embolism were included in the study. Baseline characteristics of the patient population are summarized in Table 1. The study population consisted of 23 women (64%); the mean age was 62.7 ± 15.8 years. All patients except three complained of dyspnea at enrollment (92%). Thirty-three individuals (92%) presented with syncope, of whom 12 also had clinical signs of breathlessness. Pleuritic chest pain was recorded in 13 patients (36%); all of these subjects also had dyspnea. In 12 patients (33%) pulmonary embolism developed in the immediate postoperative period. Six patients (17%) were suffering from malignant tumors. In 17 subjects (47%) deep-vein thromboses of the lower limbs were diagnosed using B-mode and/or Doppler ultrasonographic or phlebographic studies. Circulatory collapse with assisted ventilation was observed in two patients (6%). Three patients (8%) demonstrated signs of cardiogenic shock as judged by the need for catecholamine infusions.
The electrocardiographic changes in the study population were as follows: 3 patients (8%) with S-waves in lead I and aVL, 3 patients (8%) with right axis deviation, 4 patients (11%) with atrial fibrillation, 6 subjects (17%) with RBBB and 10 subjects (28%) with right precordial T-wave inversion. Right ventricular dilatation as determined by echocardiographic examination was observed in 16 individuals (44%). Lung scanning was performed in 36 patients, of whom 29 (81%) had positive results. Twenty-three patients (64%) underwent catheterization of the right heart. Among these, 16 subjects (70%) had pulmonary hypertension as defined by a systolic blood pressure >25 mm Hg.
Data from troponin I measurements
In the total study population troponin I tests were positive in 14 patients (39%) and were below the cut-off level of 0.15 μg/L in 22 patients (61%). Ten of 16 patients with RV dilation (62.5%) had increased troponin I levels, while only 4 of 14 patients (28.6%) with elevated troponin I values had RV diameters ≤30 mm. Sixteen patients with normal RV diameter and six patients with RV dilation had negative troponin I tests. Five patients had slightly elevated CK values; all of them had positive lung scans, and four had an elevated RV diameter as assessed by echocardiography. In four patients with normal tests at admission, pathologically elevated serum levels were detected up to 6 h later, suggesting a lag period during which troponin I did not appear in the peripheral blood. The clinical data of the study patients are summarized in Table 2.
Using Pearson chi-square tests, positive troponin I tests were significantly associated with RV dilatation (p = 0.009). Although serum D-dimers concentrations above the cut-off level had a significantly higher sensitivity than troponin I tests for identifying patients with pulmonary embolism, D-dimers failed to predict RV dilatation (p = 0.35). Although study patients with elevated serum levels of troponin I had a statistically greater prevalence of RV dilatation, there was no such association between CK-MB levels and ventricular overload (p > 0.70). Elevated serum levels of troponin I were found to be associated with atrial fibrillation (p = 0.009) and S-waves in lead I and aVL (p = 0.03) but not with other electrocardiographic changes such as right axis deviation or RBBB (p > 0.05).
Although study patients with elevated troponin I had lower levels of arterial partial pressures of O2 (mean 56.6 ± 15.7 mm Hg) compared with patients with negative tests (mean 64.1 ± 10.0 mm Hg), this difference was not statistically significant (p = 0.11). However, the number of segmental defects in lung scans was significantly associated with the results from the troponin I measurements. Patients with positive troponin I assays had statistically more segmental defects on lung scans than patients with normal troponin serum levels (p = 0.0002).
Newly developed highly sensitive assays for troponin I allow the detection of troponin I elevations in acute pulmonary embolism
Our main observation in this investigation was that more than one third (39%) of the patients clinically diagnosed as having acute pulmonary embolism presented with elevated troponin I serum concentrations. Furthermore, we demonstrated that in the study sample positive troponin I tests were significantly associated with the occurrence of RV dysfunction. In this prospective study we showed that, in patients with acute pulmonary embolism, the presence of echocardiographic signs of RV dysfunction is statistically related to elevated troponin I concentrations in the systemic circulation.
Although troponin I is a well-known diagnostic marker of myocardial ischemia in unstable angina pectoris resulting from CHD, no systematic investigation so far has evaluated the diagnostic usefulness and applicability of newly developed, highly sensitive troponin I assays for identifying patients at high risk to develop RV overload in case of pulmonary embolism (19). In this study the diagnostic significance of the ACS:180 troponin I values in detecting objective signs of ischemia of the RV was tested using an easily performable automated chemiluminescence system based on a two-site sandwich immunoassay. This newly developed troponin I assay allows the detection of low concentrations of this marker in the systemic circulation, presumably because the amount of myocardial damage increases as pulmonary hypertension rises. In healthy, normal individuals troponin I concentrations are below the cut-off limit, which is 0.15 μg/L for the ACS:180 analyzer (8).
Positive troponin I results are associated with echocardiographic signs of RV dilation
Patients with severe pulmonary embolism are at a high risk of developing increased pulmonary vascular resistance, subsequently leading to RV dilation and hypokinesia (2–7). Echocardiography is a well-established and useful tool for distinguishing fulminant, life-threatening embolism from incidental, clinically unimportant thromboembolism, because it permits the reliable and rapid noninvasive detection of right heart pressure overload (3,4,6,7). The presence of troponin levels above the upper limit of the reference range seems to represent a clinically important determinant of adverse myocardial affection. Thus, measuring troponin I provides additional information for evaluating patients with pulmonary thromboembolism and permits the early identification of patients at increased risk for hemodynamic deterioration.
Numerous clinical studies have demonstrated that patients with pulmonary embolism have a poorer prognosis if they present with increased RV afterload, suggesting that troponin I measurements may be useful in routine clinical practice for the early risk assessment of these patients (2–7). Our data did not allow us to recommend troponin I for the prediction of unfavorable outcomes in patients presenting with pulmonary embolisms, because a follow-up investigation to assess mortality in this study population had not been performed. However, these data indicate that further clinical studies to determine the prognostic value of troponin I assays in this setting are warranted because troponin I, shown in this investigation to identify patients with RV dysfunction, may contribute to the prediction of survival.
Segmental defects in lung scans are related to troponin I elevations
It appeared that major pulmonary embolism, as judged by numerous segmental defects in lung scans and echocardiographic signs of RV dysfunction, may be detected early with the adjunctive knowledge of troponin I values. Because of the high sensitivity of troponin I for the detection of minor myocardial damage, above-normal levels of this cardiac-specific marker may contribute to the early prediction of hemodynamic instability in individuals diagnosed as having pulmonary embolism. Our data suggest that immediate troponin I measurements performed in a routine clinical setting may help to simplify the risk stratification of patients suffering from this potential life-threatening disease. The benefit of troponin I in distinguishing between major and limited courses of pulmonary embolism argues for the inclusion of troponin I assays into the routine evaluation of patients with acute episodes of thromboembolism.
In conclusion, the data from this study demonstrate that troponin tests are highly sensitive for the early detection of minor myocardial-cell injury in pulmonary embolism associated with RV dysfunction. Furthermore, positive troponin I tests appear to directly reflect the impairment of pulmonary blood flow as assessed by a significant association with the number of segmental defects in lung scans. These results are consistent with the hypothesis that troponin I positivity helps to identify patients with major pulmonary embolism. Our study provides evidence that troponin I measurements do not enable us to distinguish specifically between coronary and noncoronary causes of chest pain, because, in some patients presenting with pulmonary embolism, elevated troponin I concentrations above the normal range are observed as well. The use of this biochemical marker in addition to established apparative procedures such as echocardiography, angiography and lung scanning will extend the diagnostic regimen for an early identification of patients with RV afterload stress.
The authors gratefully acknowledge the skillful technical assistance of Ina Martetschläger and Irmina Szymczak. We also thank Bayer Diagnostics (Fernwald, Germany) for providing us with troponin I assay kits.
- coronary heart disease
- creatine kinase
- right bundle branch block
- right ventricular, right ventricle
- Received October 25, 1999.
- Revision received April 11, 2000.
- Accepted June 15, 2000.
- American College of Cardiology
- Egermayer P.,
- Town G.I.,
- Turner J.G.,
- Heaton D.C.,
- Mee A.L.,
- Beard M.E.J.
- Kasper W.,
- Konstantinides S.,
- Geibel A.,
- et al.
- Kasper W.,
- Konstantinides S.,
- Geibel A.,
- Tiede N.,
- Krause T.,
- Just H.
- Konstantinides S.,
- Geibel A.,
- Olschewski M.,
- et al.
- Adams J.E.,
- Bodor G.S.,
- Dávila-Román V.G.,
- et al.
- Galvani M.,
- Ottani F.,
- Ferrini D.,
- et al.
- Kontos M.C.,
- Jesse R.L.,
- Anderson F.P.,
- Schmidt K.L.,
- Ornato J.P.,
- Tatum J.L.
- Lüscher M.S.,
- Thygesen K.,
- Ravkilde J.,
- Heickendorff L.
- Missov E.,
- Calzolari C.,
- Pau B.
- Polanczyk C.A.,
- Lee T.H.,
- Cook E.F.,
- et al.
- Katrukha A.G.,
- Bereznikova A.V.,
- Esakova T.V.,
- et al.