Correspondence

Cardiac Troponins in Patients with Chest Pain

N Engl J Med 1998; 338:1314-1316April 30, 1998

Article

To the Editor:

Hamm and colleagues present impressive results concerning the operating characteristics of troponin T and troponin I in the triage of patients with acute chest pain (Dec. 4 issue).1 I am concerned that their conclusion that “negative test results are associated with low risk and allow rapid and safe discharge of patients with an episode of acute chest pain from the emergency room” may be overstated and could easily be taken out of context and misinterpreted. The authors base their statement on the findings that only 1.1 percent of the patients with negative troponin T results and 0.3 percent of the patients with negative troponin I results died or had nonfatal myocardial infarctions in the 30 days following presentation. First, I question whether these specificities are adequate when one is deciding on an admission threshold. The high frequency of patients presenting with acute chest pain and the confidence intervals around these percentages may demand even lower numbers. Second, included in these percentages were 245 patients who had unstable angina and 201 who were admitted with negative troponin T and troponin I results. Although the natural history of this subgroup of patients is unknown, some of these patients may have had adverse cardiac outcomes had they not been hospitalized and received subsequent care.

Although troponin T and troponin I, along with electrocardiography, improve our ability to rule in or rule out high-risk coronary syndromes, until we know the natural history of unstable angina in patients with negative troponin tests, the tests' true utility in emergency-room triage may be limited to the subgroup with clinically stable angina and atypical or noncardiac chest pain. The authors' disclaimer that the “tests cannot replace the clinical evaluation of the patient with chest pain” must be reemphasized.

(The opinions expressed in this letter are the author's and do not represent the official policies of the Agency for Health Care Policy and Research or of the U.S. Department of Health and Human Services.)

Daniel B. Stryer, M.D.
Agency for Health Care Policy and Research, Rockville, MD 20852

1 References

1
Hamm CW, Goldmann BU, Heeschen C, Kreymann G, Berger J, Meinertz T. Emergency room triage of patients with acute chest pain by means of rapid testing for cardiac troponin T or troponin I. N Engl J Med 1997;337:1648-1653
Full Text | Web of Science | Medline

To the Editor:

Although the study by Hamm et al. shows that troponin T and troponin I are predictors of death and myocardial infarction in patients who present to the emergency department with chest pain and are managed conventionally, their conclusion that negative test results “allow rapid and safe discharge . . . from the emergency room” is not supported.

The majority of these patients were hospitalized. Although we are given no information about their hospital course, presumably most were put on restricted activity and treated with various drugs, such as anticoagulants, antiplatelet agents, beta-blockers, and nitrates. Presumably, many of them underwent angiography or noninvasive coronary evaluation and some went on to mechanical revascularization. To say that the patients who were thus managed and subsequently had low event rates would have done as well if discharged directly from the emergency department is a leap of faith that we should not be ready to take.

Daniel A. Waxman, M.D.
Jacobi Hospital, Bronx, NY 10461

To the Editor:

In 19881 and 19912 we found that elevated levels of cardiac myofibrillar proteins in the blood of patients with chest pain were indicative of a high risk of cardiac events even when acute myocardial infarction was ruled out. The study by Hamm and coworkers is a further confirmation of the prognostic importance of the troponins.

In addition, Hamm and coworkers compare the diagnostic performance of two whole-blood assays — the troponin T assay (Boehringer Mannheim) and the troponin I assay (Spectral Corporation). We believe that the design of this comparison creates several problems that may necessitate a reevaluation of the results and conclusions of the study.

Although both assays were developed to test whole-blood specimens, only the troponin T assay was used in this way. By contrast, troponin I was tested with precentrifuged plasma samples. The use of plasma instead of whole blood improves the analytical performance of the rapid-assay systems. Thus, the test protocol will favor the diagnostic performance of the troponin I assay.

The troponin T assay was evaluated under “in field” conditions — i.e., various physicians on call evaluated the test-strip device during the day and at night. By contrast, the troponin I assay was performed by well-trained laboratory personnel during daytime working hours. Taking into account the great interindividual variability of visual assessment of test results at the detection limit, this difference not only introduces a major bias against the troponin T assay but also casts doubt on the practical relevance of the troponin I results.

The risk stratification and therapeutic decisions were based on the results of the troponin T assay but not the troponin I assay. As a consequence, patients with positive troponin T results may have been treated more aggressively to reduce the complication rate. An unbiased comparison would have required that all physicians be blinded to the results of both assays.

The authors report on six patients with “false” positive troponin T results who had chest pain and end-stage renal disease. Since coronary artery disease is the leading cause of death in end-stage renal disease and the patients suffered from anginal pain, they may well have had “true” positive results.

More than a year ago, we improved the sensitivity and specificity of the troponin T bedside assay used by Hamm.3 Thus, the comparison is not relevant given the currently available assay.

This study therefore supports our belief4 that myofibrillar proteins are superior markers for the risk stratification of patients with chest pain. However, in comparative analyses of two test systems it is required either to select equal test conditions or, when this is not possible, to state the limitations of the study.

Hugo A. Katus, M.D.
Norbert Frey, M.D.
Margit Müller-Bardorff, M.D.
University of Lübeck, D-23538 Lübeck, Germany

4 References

1
Katus HA, Diederich KW, Hoberg E, Kubler W. Circulating cardiac myosin light chains in patients with angina at rest: identification of a high risk subgroup. J Am Coll Cardiol 1988;11:487-493
CrossRef | Web of Science | Medline

2
Katus HA, Remppis A, Neumann FJ, et al. Diagnostic efficiency of troponin T measurements in acute myocardial infarction. Circulation 1991;83:902-912
Web of Science | Medline

3
Muller-Bardorff M, Rauscher T, Kampmann M, et al. Quantitative assessment of the bedside troponin T assay with an automated Reader. Circulation 1997;96:Suppl I:I-273 abstract.

4
Katus HA, Kübler W. Detection of myocardial cell damage in patients with unstable angina by serodiagnostic tools. In: Bleifeld W, Braunwald E, Hamm C, eds. Unstable angina. New York: Springer-Verlag, 1992:29-41.

Author/Editor Response

The authors reply:

To the Editor: We welcome the comments of Dr. Stryer and Dr. Waxman concerning the limitations of troponins in critical decision making about patients with acute chest pain. The introduction of a new, superior marker carries the risk that previous standards will be too easily disregarded. Our view is best reflected in our discussion, in which we stated that “bedside tests for troponin T and troponin I result in more accurate diagnoses than do previous, more time-consuming methods and allow safer and more rapid decision making for most patients with acute chest pain.” The additional information provided by troponin measurements does not, however, exempt the treating physician from carefully evaluating patients with chest pain.

We also share the concern regarding the possible effect on the event rate of hospital admissions and treatment of patients with negative troponin T tests. Conclusive evidence from other studies indicates, however, that cardiac risk in troponin-negative patients is low.1-3 In our series, 49 percent of the patients without troponins and with normal echocardiograms were admitted to the hospital, of whom only 5.6 percent underwent coronary angiography and none had percutaneous transluminal coronary angioplasty or bypass surgery. There was no difference in the base-line characteristics of the patients with negative tests and normal electrocardiograms between those who were directly discharged and those who were admitted. In the Fragmin during Instability in Coronary Artery Disease trial, the event rate in patients without troponin T was low, and the benefit of low-molecular-weight heparin was shown only for patients positive for troponin T.4 Accordingly, a major effect of hospitalization in troponin-negative patients cannot be completely excluded but appears unlikely. A trial design that randomly assigns troponin-negative patients to discharge or hospitalization may answer this question but does not appear feasible.

Our study was not designed to compare troponin T with troponin I as suggested by Katus et al. Since the analytical developments are still ongoing, any statement would only be short-lived and would detract from our key intention, which was to demonstrate the unique potential of both troponins as diagnostic markers. The improved troponin T bedside test was developed during our study. On the basis of our quantitative controls we would expect at the new cutoff level (0.10 ng per milliliter) 22 more positive troponin T test results, predicting two more cardiac events (one death and one infarction) and resulting in an event rate of 0.8 percent.

We share the view of Katus et al. that the reading of the test results is subject to error. The reading has to be done exactly at the indicated time under good light. We therefore trained our assistants before they were certified to read the test results. To detect any bias due to false readings, we made quantitative control measurements part of our protocol, which showed excellent agreement with the rapid tests.

Christian W. Hamm, M.D.
Christopher Heeschen, M.D.
Thomas Meinertz, M.D.
University Hospital Eppendorf, D-20246 Hamburg, Germany

4 References

1
Lindahl B, Venge P, Wallentin L. Relation between troponin T and the risk of subsequent cardiac events in unstable coronary artery disease. Circulation 1996;93:1651-1657
Web of Science | Medline

2
Antman EM, Tanasijevic MJ, Thompson B, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335:1342-1349
Full Text | Web of Science | Medline

3
Luscher MS, Thygesen K, Ravkilde J, Heickendorff L. Applicability of cardiac troponin T and I for early risk stratification in unstable coronary artery disease. Circulation 1997;96:2578-2585
Web of Science | Medline

4
Lindahl B, Venge P, Wallentin L. Troponin T identifies patients with unstable coronary artery disease who benefit from long-term antithrombotic protection. J Am Coll Cardiol 1997;29:43-48
CrossRef | Web of Science | Medline

Citing Articles (1)