Q J Med 1999; 92: 565-571
© 1999 Association of Physicians
Diagnosis and risk stratification of patients with anginal pain and non-diagnostic electrocardiograms
From the Regional Medical Cardiology Centre, and 1 Department of Clinical Biochemistry, Royal Victoria Hospital, Belfast, UK
Received 21 April 1999
Professor A.A.J. Adgey, Regional Medical Cardiology Centre, Royal Victoria Hospital, Grosvenor Road, Belfast BT12 6BA
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Patients with acute chest pain suggestive of myocardial ischaemia, and normal or non-diagnostic electrocardiograms, form a difficult subgroup for diagnosis and early risk stratification. We prospectively evaluated the role of troponin T (cTnT), troponin I (cTnI), CKMB mass and myoglobin, in the diagnosis and risk stratification of 214 patients with acute chest pain of
24 h and non-diagnostic or normal ECGs admitted directly to the Cardiac Unit of the Royal Victoria Hospital Belfast from the Mobile Coronary Care Unit or the Accident/Emergency Department. This was a single-centre prospective study, and follow-up (3 months) was complete for all patients. Blood was assessed for quantitative cTnT, cTnI, CKMB mass and myoglobin, and qualitative cTnT on admission and at 12 h. Diagnosis of index event and incidence of new cardiac events (death, non-fatal myocardial infarction, revascularization, or readmission for unstable angina) over 3 months were assessed. Based on standard criteria, myocardial infarction occurred in 37/214 (17%), and unstable angina in 72/214 (34%). At 12 h from admission, cardiac troponins had higher sensitivity for the diagnosis of acute coronary syndromes (myocardial infarction and unstable angina) than conventional markers (cTnI 48%, cTnT 38%, CKMB mass 30% or myoglobin 27%). At 3 months, a new cardiac event had occurred in 42/214 (20%). Significantly higher event rates occurred when any of the biochemical markers was elevated, but the statistical significance was highest for patients with elevated cTnI (p<0.0001). Whilst gender, history of ischaemic heart disease (IHD), stress test response, cTnT, cTnI, CKMB mass and myoglobin were univariate predictors, cTnI at 12 h and stress test response were the only two independent significant predictors for a subsequent cardiac event at 3 months. Raised cTnI at 12 h after admission had the highest sensitivity for the diagnosis of acute coronary syndromes, and was independently associated with a 2–3 times increased risk of future cardiac events within 3 months among patients with acute chest pain suggestive of myocardial ischaemia but with normal or non-diagnostic ECGs. |
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
At the time of hospital presentation, of patients with chest pain in whom acute myocardial infarction is subsequently confirmed, approximately 50% show typical changes (ST segment elevation).1 However, of patients presenting to hospital with chest pain suggestive of acute myocardial infarction in whom cardiac enzymes are requested, only 10% show ST elevation.2 About 25% of patients presenting to the hospital with chest pain or other symptoms suggestive of myocardial infarction have bundle branch block, non-specific ST-T changes or no changes in their initial electrocardiograms (ECGs).3 Only 9% of patients with normal or minimal non-specific changes in the ECG ultimately have a proven myocardial infarct.4 Thus many of these patients do not have underlying coronary artery disease, and hospital admission of all these patients represents an inappropriate use of resources. Nevertheless, approximately 6% of patients with chest pain suggestive of acute myocardial infarction are sent home and subsequently found to have infarction,5 and this accounts for the highest dollar losses in emergency department malpractice cases in the US.
Despite various clinical and ECG classifications, acute coronary syndromes lack accurate and reliable non-invasive markers for diagnosis and thus risk stratification. The worst prognosis (death) for patients with acute coronary syndromes is among those with chest pain and ST segment depression on the ECG at rest.6 In addition, dynamic shifts in ST segment or T waves that resolve are important markers of adverse prognosis. These ECG changes are more likely to be recorded during acute chest pain, but it is not always possible (for logistical reasons) to record the ECG during chest pain. Hence new techniques have been developed to diagnose patients with unstable angina due to coronary artery disease, and thus to help risk stratification. Those which seem to be particularly promising include new biochemical markers of myocardial ischaemia, ECG stress testing, and acute perfusion imaging with technetium-99m-labelled sestamibi.
Cardiac troponins are sensitive for the detection of myocardial cell damage even in the presence of normal cardiac enzymes.7 In patients with unstable angina, release of troponin T (cTnT) has been shown in 21–64% of these patients.8,9 In addition, the release of cTnT and troponin I (cTnI) is associated with increased risk (death, myocardial infarction, need for revascularization) as demonstrated in many clinical trials.10–13 Troponins therefore have a role both in diagnosis and prognosis and there is evidence that they have a role in clinical selection of patients for treatment with low-molecular-weight heparins and IIb IIIa inhibitors.14,15
The majority of patients admitted to trials assessing troponin release have acute ischaemic changes on the initial ECG,10,11 but this is not the major burden of chest pain patients in Accident & Emergency. We therefore evaluated simultaneously the diagnostic and prognostic value of four biochemical markers (cTnT, cTnI, CKMB mass and myoglobin) in patients presenting with acute chest pain suggestive of myocardial ischaemia but with non-diagnostic or normal ECGs at initial presentation.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study population consisted of 214 patients admitted directly to the cardiac unit of the Royal Victoria Hospital, Belfast from the Mobile Coronary Care Unit or from the A&E Department between August 1996 and August 1997. These patients had acute chest pain that was considered by the admitting doctor to be sufficiently suggestive of cardiac ischaemia to require hospital admission within 24 h of the onset of symptoms, and non-diagnostic ECG changes (i.e. normal, T wave changes only, bundle branch blocks, chamber hypertrophy, paced rhythm). We excluded patients with: (i) history of myocardial infarction or percutaneous transluminal angioplasty (PTCA) in the preceding 2 weeks; (ii) ECG evidence of acute myocardial infarction/ischaemia defined as ST segment elevation
0.1 mV in two or more contiguous leads or ST depression 0.1 mV 0.08 s after the J point; or (iii) patients presenting after 24 h from the onset of symptoms, unless they had recurrent episodes.
Study protocol
The study protocol was approved by the ethical committee of the Queens University of Belfast. After oral informed consent, 10 ml venous blood was collected for bedside assessment of cTnT (qualitative) and quantitative measurement of cTnT, cTnI, CKMB mass and myoglobin. These tests were repeated 12 h after admission. The qualitative cTnT test was performed in the ward by the doctor or the staff nurse. Serum samples were sent to the laboratory and stored at -20°C and were analysed in batches of 50. The clinical team was blinded to the quantitative results so as not to bias the future treatment decisions.
Clinical data from initial evaluation included the patient's history, physical examination, cardiac enzymes and ECGs. During hospitalization, patients underwent either a stress test (exercise or pharmacological with perfusion imaging) or coronary angiogram, unless another diagnosis became evident within 24 h. Patients were followed until discharge from the hospital, and 3 months after discharge by telephone to record cardiac events. Composite end point was death from cardiac causes, non-fatal myocardial infarction, revascularization procedures (angioplasty or bypass surgery) and further hospitalizations for unstable angina.
Definitions
Acute myocardial infarction was defined as total CK activity more than twice the upper limit of normal within 72 h of admission with elevated CKMB (more than 6%), and/or the presence of evolutionary repolarization changes (T-wave inversion) in a patient with typical symptoms. The diagnosis of unstable angina required an abnormal stress test (0.1 mV ST depression in two or more ECG leads at peak stress, or evidence of reversible ischaemia on perfusion imaging) and/or coronary angiogram (70% stenosis of one or more coronary arteries) in a patient with chest pain at rest in the absence of raised cardiac enzymes.
Analytical techniques
For qualitative determination of cTnT, we used a whole-blood rapid-assay device (Trop T sensitive, Boehringer Mannheim). Blood samples were collected in tubes containing EDTA for bedside testing of cTnT. A positive test was indicated by a coloured line appearing within 20 min (cut-off value >0.1 µg/l). For quantitative assessment, blood was collected in gel tubes and centrifuged, and the plasma was frozen in aliquots and stored for subsequent analysis. Troponin T was determined by an enzyme immunoassay (ES300, Boehringer Mannheim), and values above 0.1 µg/l were taken as clinically significant. Quantitative measurement of cTnI was done on an Opus analyser (Behring). The lower level of detection was 0.5 µg/l, and values above 1.0 µg/l were taken as clinically significant. In all samples, CKMB mass and myoglobin were determined in the same analyser, and abnormal results were 13 µg/l and 92 µg/l, respectively. All biochemical analyses were performed by technicians unaware of the patients' data or the results of qualitative assay of cTnT.
Statistical analysis
All results for continuous variables are expressed as means±SD. The independent-sample t test was used to compare continuous variables between two subgroups. The p values for comparison of categorical variables were generated by the Pearson 
2 test for proportions with appropriate degrees of freedom, and p values <0.05 were taken as significant. All significant variables identified in the univariate analysis were entered into a multiple forward stepwise (likelihood ratio) logistic regression analysis. All calculations were carried out with the SPSS (version 7.1) software package.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The 214 patients (151 men, 63 women) had a mean age of 59±11.6 years. The mean time from onset of symptoms to presentation was 6.6±6.1 h (range 45 minutes–24 h). The admission electrocardiogram showed T inversion only in 65 patients, pathological Q waves suggestive of old infarction in 43 patients, left bundle branch block in eight, right bundle branch block in 10, left ventricular hypertrophy in four, ventricular pacing in two, and no ECG changes in 82 patients. Already-documented ischaemic heart disease, risk factors for ischaemic heart disease, and anti-anginal drugs are shown in Table 1
.
|
Diagnosis for initial admission
Myocardial infarction was retrospectively diagnosed in 37 patients within 72 h of admission (30 patients had CK activity more than twice the upper limit of normal). Among the remaining 177 patients, unstable angina was diagnosed in 72, pericarditis in two, myocarditis in one and unexplained chest pain in 102 patients.
Biochemical markers and clinical diagnosis
See Table 2. Of the 214 patients, 31 (14%) had at least one positive bedside test for cTnT and 49 (23%) patients had at least one blood sample with elevated cTnT quantitative and 55 (26%) patients had elevated cTnI. Of 37 patients with a final diagnosis of myocardial infarction, 28 (76%) had elevated cTnT (quantitative) in one of the samples and 35 (95%) had elevated cTnI. Bedside cTnT was positive in 24/37 (65%) patients with a final diagnosis of myocardial infarction. CKMB mass was elevated in 30 (81%) and myoglobin in 21 (57%) within 12 h of admission. Of note, elevated cTnT (quantitative) and cTnI were present on the admission blood samples in only 13 and 12 patients, respectively.
|
Of the 72 patients with a final diagnosis of unstable angina, 13 (18%) had at least one blood sample with elevated cTnT, and 17 (24%) had elevated cTnI.
Of the remaining eight patients with at least one sample with elevated cTnT, one had myocarditis and seven had unexplained chest pain: of three patients with elevated cTnI, one had myocarditis and two had unexplained chest pain.
Over all 214 patients, the sensitivity for diagnosis of acute coronary syndromes (myocardial infarction and unstable angina) was poor but specificity was very high (91–99%) (Table 3).
|
Follow-up
See Table 4
. All patients were contacted 3 months after the index event. Cardiac events had occurred in 42 (20%) patients. One patient died (cardiac death) and two patients had had myocardial infarctions. Two of these three patients had elevation of all four biochemical markers during the index admission. In the third patient, who developed myocardial infarction 2 months after the index event, all the markers were negative during the index episode, and the stress test response was negative. Nine patients were readmitted with unstable angina, and 30 patients required revascularization (14 CABG and 16 PTCA). Only one patient diagnosed as non-coronary pain had a cardiac event during follow-up (myocardial infarct).
|
Of the 42 patients who had cardiac events during follow-up, 16 (38%) were positive for cTnT (quantitative) and 23 (55%) for cTnI during the index admission. Thus the positive predictive values for a subsequent cardiac event within 3 months are 42% (cTnI) and 33% (cTnT): negative predictive values are 88% and 84%, respectively.
Twenty-three of the 55 (42%) patients with elevated cTnI within 12 h of admission during their index episode had a subsequent cardiac event within 3 months, compared to 19/159 (12%) patients with normal values (p<0.0001). Similar differences were observed for the other markers (Table 4
).
In a univariate analysis of the 214 patients the following factors correlated positively with follow up cardiac events (Table 5): male sex, previously documented IHD, stress test response (n=173), and all four biochemical markers assessed at 12 h from admission. When these univariate predictors were included in the multiple logistic regression analysis, a model was elaborated that identified cTnI at 12 h and the stress test response as independent predictors of cardiac events.
|
For raised cTnI, the odds ratio for cardiac events within 3 months was 5.2 (95% CI 2.5–10.8) and for the stress test response it was 7.2 (95% CI 2.3–22.4).
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Studies have established cTnI and cTnT as significantly more sensitive and specific than CKMB for myocardial necrosis.16–19 In addition, raised cTnT and cTnI are of greater prognostic value for risk stratification in acute coronary syndromes than the conventional markers.10,11 However, the precise indications for the use of these markers over and above the established diagnostic techniques of ECG and conventional markers have yet to be established.
Patients presenting with ischaemic chest pain may be considered to fall into three groups as stratified by the presenting ECG.
- (i) In patients admitted with ST segment elevation myocardial infarct, the measurement of these markers adds only limited information over that provided by ECG and CKMB. In this setting, the ECG remains the cornerstone on which early treatment is based. The role of troponins in this subset is largely to identify a subgroup of patients at high risk for death or recurrent myocardial infarction. The presence of elevated troponins on admission adds independent prognostic information even after adjustment for ECG changes and CKMB levels,10,11 and may identify those with incomplete reperfusion following lysis. However, there is no evidence at present that early risk stratification of this group will lead to improvement in outcome.
(ii) The next category constitutes patients with new ST segment depression or T-wave inversion in the ECG, and includes the diagnostic categories of non-Q myocardial infarction and unstable angina. In these patients, elevated cTnT or cTnI on admission and/or 12–24 h after admission identifies a high-risk group.10,11,20 However, the significance of this is less clear in patients with elevated CKMB. In the GUSTO-IIa substudy, 30-day mortality in patients without ST segment elevation was the same for patients with elevated troponin (9.9%) as for those with elevated CKMB (9.8%), and the incidence of death and/or myocardial infarction was also similar in the two groups.10 Lindahl et al. found that elevated cTnT within 12 h was only of prognostic significance in patients in whom the CKMB mass was below the upper limit of normal.21 ST segment depression during pain is associated with the highest mortality.6 Even in patients with ST depression on the index ECG and a negative troponin test, the short-term cardiac event rate is not low (2.8% of patients with negative troponin T and 1.4% patients with negative troponin I had cardiac events during follow-up).13
(iii) The third and most difficult group in terms of risk stratification is those presenting with non-diagnostic ECGs. In these patients, ECG may be normal, T-wave inversion (old or new), or non-diagnostic due to confounding factors such as bundle branch block, chamber hypertrophy, drug effect, or paced rhythm. Sensitive and rapid assays which detect subtle evidence of myocardial necrosis would be useful in this group. We thus prospectively evaluated the diagnostic and prognostic role of four biochemical markers in this group. Our study confirms that for the diagnosis of acute coronary syndromes, cardiac troponins measured quantitatively have a higher sensitivity than conventional markers (CKMB activity and CKMB mass), and during a period of 3 months, cardiac event rate was significantly higher in patients with an elevated biochemical marker.
Of the various biochemical markers evaluated in the 214 patients, cTnI assessed 12 h from admission had the highest sensitivity (48%) for the diagnosis of acute coronary syndromes, with a specificity of 97%, followed by quantitative cTnT (sensitivity 38%, specificity 91%, Table 3). The higher sensitivity of elevated cTnI compared with cTnT may be related to the different release kinetics and versions of assays available.13 However, as there is no agreed standard for cTnI assay calibration, our findings cannot be directly compared to studies in which cTnI was measured by different assays.
Of 37 patients with acute myocardial infarction, qualitative cTnT at 12 h was positive in 65% (Table 2) and the sensitivity for diagnosis of acute coronary syndromes was only 28% (Table 3). The lower detection rate for this test (qualitative cTnT) in the diagnosis of myocardial infarction compared to previous studies13,22 is probably because the test was performed by medical or nursing staff rather than under optimal laboratory conditions, and the mean concentration of cTnT was lower in these patients. The detection rate using myoglobin for the diagnosis of myocardial infarction was also lower than reported by others.23 The staccato pattern of myoglobin release is well known, and the lower detection rate of this marker in the present study may well be due to blood samples taken only 12 h apart, and late presentation in some patients. Blood samples on admission alone provided low detection rates (Table 2) for all the markers, and hence a single test at the time of admission is inadequate for clinical decision-making. This is in keeping with previous studies.13
It is interesting to note that the presence of Q waves or T-wave inversion on the index ECG was not predictive of future cardiac events. While all four biochemical markers and three clinical variables (male sex, documented IHD and stress test response) were univariate predictors of future cardiac events, only elevated cTnI at 12 h and stress test response retained independent significance after multivariate analysis. Thus it would appear, at least in this group of patients with non-diagnostic ECGs, that measurement of cTnT, CKMB mass or myoglobin in addition to cTnI does not provide further prognostic information over that obtained from measurement of cTnI alone.
In summary, in patients with chest pain and non-diagnostic ECG's at presentation, the newer biochemical markers, particularly cTnI, improve sensitivity for detection of acute coronary syndromes whilst retaining high specificity. In addition, the presence of an elevated cTnI at 12 h is independently associated with a 2–3-fold increased risk of future cardiac events in this otherwise difficult-to-stratify patient group.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Fisch C. The first electrocardiogram in acute myocardial infarction. ACC Curr J Rev 1993; 2:72–3.
2. Collinson PO. Biochemical markers for the millennium. Cardiol News 1997; 1:12–16.
3.
Rawles J. Should patients with suspected acute myocardial infarction without ST elevation be given thrombolytic treatment? Eur Heart J 1997; 18:899–906.
4. Slater DK, Hlatky MA, Mark DB, et al. Outcome in suspected acute myocardial infarction with normal or minimally abnormal admission electrocardiographic findings. Am J Cardiol 1987; 60:766–70.[Web of Science][Medline]
5.
Schor S, Behar S, Modan B, et al. Disposition of presumed coronary patients from an emergency room. A follow up study. JAMA 1976; 236:941–3.
6. Lopez de Sa E, Lopez-Sendon J, Bethencourt A, et al. The PEPA investigators: Prognostic value of ECG changes during chest pain in patients with unstable angina. Results of the Proyecto de Estudio del Pronostico de la Angina (PEPA). J Am Coll Cardiol 1998; 31:79A.
7.
Rottbauer W, Greten T, Mauller-Bardorff M, et al. Troponin T: a diagnostic marker for myocardial infarction and minor cardiac cell damage. Eur Heart J 1996; 17 (Suppl. F):3–8.
8. Johnson PA, Albano MP, Sacks D et al. Troponin T in patients with acute chest pain. J Am Coll Cardiol 1994; 23:412A.
9. Seino Y, Tomita Y, Takano T et al. Early identification of cardiac events with serum troponin T in patients with unstable angina. Lancet 1993; 342:1236–7.[Web of Science][Medline]
10.
Ohman EM, Armstrong PW, Christenson RH et al. Cardiac troponin T levels for risk stratification in acute myocardial ischemia. N Engl J Med 1996; 335:1333–41.
11.
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; 335:1342–9.
12.
Wu AHB, Lane PL. Metaanalysis in clinical chemistry: Validation of cardiac troponin T as a marker for ischemic heart diseases. Clin Chem 1995; 41(B):1228–33.
13.
Hamm CW, Goldmann BU, Heeschen C, et al. 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; 337:1648–53.
14. Lindahl B, Venge P, Wallentin L. Troponin T identifies patients with unstable coronary artery disease who benefit from long-term antithrombotic protection. Fragmin in unstable coronary artery disease (FRISC) study group. J Am Coll Cardiol 1997; 29:43–8.[Abstract]
15. Hamm CW, Heeschen C, Goldmann BU et al. For the capture Investigators: Value of Troponins in predicting therapeutic efficacy of Abciximab in patients with unstable angina. J Am Coll Cardiol 1998; 31:185A.
16.
Katus HA, Remppis A, Neumann FJ et al. Diagnostic efficiency of Troponin T measurements in acute myocardial infarction. Circulation 1991; 83:902–12.
17. Collinson PO, Chandler HA, Stubbs PJ et al. Measurement of serum troponin T, creatine kinase MB isoenzyme, and total creatine kinase following arduous physical training. Ann Clin Biochem 1995; 32:450–3.
18. Bertinchant JP, Larue C, Pernel I, et al. Release kinetics of serum cardiac troponin I in ischemic myocardial injury. Clin Biochem 1996; 29:587–94.[Web of Science][Medline]
19.
Adams JE III, Bodor GS, Davila-Roman VG et al. Cardiac Troponin I. A marker with high specificity for cardiac injury. Circulation 1993; 88:101–6.
20.
Stubbs P, Collinson PO, Moseley D et al. Prospective study of the role of cardiac troponin T in patients admitted with unstable angina. Br Med J 1996; 313:262–4.
21.
Lindahl B, Venge P, Wallentin L for the FRISC Study group: Relation between Troponin T and the risk of subsequent cardiac events in unstable coronary artery disease. Circulation 1996; 93:1651–7.
22.
Antman EM, Grudzien C, Sacks DB. Evaluation of a rapid bedside assay for detection of serum cardiac troponin T. JAMA 1995; 273:1279–82.
23.
Adams JE III, Abendschein DR, Jaffe AS. Biochemical markers of myocardial injury. Is MB creatine kinase the choice for the 1990s? Circulation 1993; 88:750–63.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J.P. Dougan, T.P. Mathew, J.W. Riddell, M.S. Spence, P.G. McGlinchey, G.S. Nesbitt, M. Smye, I.B.A. Menown, and A.A.J. Adgey Suspected angina pectoris: a rapid-access chest pain clinic QJM, December 1, 2001; 94(12): 679 - 686. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.F. Fox Chest pain assessment services: the next steps QJM, December 1, 2001; 94(12): 717 - 718. [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Heidenreich, T. Alloggiamento, K. Melsop, K. M. McDonald, A. S. Go, and M. A. Hlatky The prognostic value of troponin in patients with non-ST elevation acute coronary syndromes: a meta-analysis J. Am. Coll. Cardiol., August 1, 2001; 38(2): 478 - 485. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||