Author + information
- Received December 30, 1999
- Revision received May 8, 2000
- Accepted June 28, 2000
- Published online November 15, 2000.
- David A Morrow, MDa,* (, )
- Elliott M Antman, MD, FACCa,
- Milenko Tanasijevic, MDb,
- Nader Rifai, PhDc,
- James A de Lemos, MDa,
- Carolyn H McCabe, BSa,
- Christopher P Cannon, MD, FACCa and
- Eugene Braunwald, MD, FACCa
- ↵*Reprint requests and correspondence:
Dr. David A. Morrow, Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115
We sought to evaluate cardiac troponin I (cTnI) for predicting early clinical outcomes and the efficacy of enoxaparin among patients with non–ST segment elevation acute coronary syndrome (ACS) and negative creatine kinase, MB fraction (CK-MB) levels.
Cardiac TnI identifies patients with unstable angina who are at higher risk of death or myocardial infarction (MI) by 30 days. The utility of cTnI for predicting very early clinical events, including recurrent ischemia, and the efficacy of enoxaparin are not yet established.
At baseline and 12 h to 24 h after enrollment in the Thrombolysis in Myocardial Infarction (TIMI)-11B trial, samples were collected for cTnI determination.
Among 359 patients with negative serial CK-MB values, 50.1% had a cTnI result ≥0.1 ng/ml within the first 24 h. Patients with elevated cTnI were at higher risk of death or MI at 48 h (3.9 vs. 0%, p = 0.01) and 14 days (13.9 vs. 2.2%, p < 0.0001). Elevated cTnI also correlated with higher risk of recurrent ischemia requiring urgent revascularization by 48 h (10.0 vs. 1.7%, p = 0.001) and 14 days (20.6 vs. 5.6%, p ≤ 0.0001). Enoxaparin had a greater benefit among patients with elevated vs. normal cTnI (p = 0.03), achieving a 47% reduction in the risk of death, MI or urgent revascularization by 14 days in cTnI-positive patients (p = 0.007).
Elevation of cTnI among patients with non–ST segment elevation ACS and negative levels of CK-MB identifies those at higher risk for very early adverse outcomes, including severe recurrent ischemia. Treatment with enoxaparin reduces the risk associated with elevated cTnI.
Patients with unstable angina are heterogeneous with respect to prognosis and present a particular challenge for early risk assessment and triage (1,2), compared with patients with elevated levels of creatine kinase, MB fraction (CK-MB), who may be recognized as being at higher risk (3). Cardiac-specific troponin I (cTnI) is a highly sensitive and specific marker of myocardial necrosis (4) that has demonstrated prognostic utility among the substantial proportion of patients with non–ST segment elevation acute coronary syndrome (ACS) who have no elevation of CK-MB (5–8). We have shown previously that elevation of cTnI ≥0.4 ng/ml at the time of presentation identifies patients with ischemic symptoms and normal CK-MB who are at higher risk of death by 42 days (5). Other clinical studies have demonstrated that elevated levels of cTnI are associated with a higher incidence of death or myocardial infarction (MI) by 30 days among patients with chest pain suspicious for angina (9) and documented unstable coronary disease (6–8). However, data evaluating the relation between cTnI and the occurrence of refractory angina, as well as the need for revascularization, have been inconclusive (6,7). Furthermore, published data regarding the prognostic utility of cTnI for predicting the risk of clinical events at very early time points (e.g., 48 h) in patients with unstable angina are limited. Finally, current generation cTnI assays offer greater analytic sensitivity than previous versions and have created the need for clinical evaluation of lower cTnI thresholds for risk assessment in ACS (10).
Given increasing economic pressure to constrain the duration of inpatient management and to target more expensive and aggressive therapies, the ability to identify at presentation those patients with ischemic symptoms and negative CK-MB who are at higher risk for early adverse outcomes has taken on particular clinical importance. In addition, data documenting an association between elevation of cardiac troponin T and enhanced efficacy of specific medical therapies (11–14) have demonstrated the potential for the cardiac troponins in guiding early management decisions. Thus, we have evaluated cTnI at a threshold of 0.1 ng/ml, using a current generation assay (Dimension RxL, Dade Behring, Newark, Delaware), for the purpose of predicting early adverse cardiac events, including recurrent ischemia necessitating urgent revascularization by 48 h and 14 days, among patients with unstable angina enrolled in Thrombolysis in Myocardial Infarction (TIMI)-11B Trial. In addition, we tested for any interaction between cTnI results and the efficacy of enoxaparin over unfractionated heparin (UFH) therapy.
Men and women with unstable angina enrolled at any of 63 North or South American centers participating in the biomarker protocol of TIMI-11B were eligible for this substudy. The diagnosis of the presenting ischemic syndrome was made by the local investigator on the basis of World Health Organization criteria for MI, using local serial measurements of CK or CK-MB as defined in the TIMI-11B protocol (15) and collected on the Initial Hospitalization Case Report Form. Patients presenting with an ischemic syndrome without acute MI were diagnosed with unstable angina.
TIMI-11B was a randomized, double-blind, parallel group trial of the low molecular weight heparin, enoxaparin, as compared with UFH, for the treatment of high risk patients presenting with non–ST segment elevation ACS. The trial design and inclusion and exclusion criteria of TIMI-11B have been published previously (15). Participants were required to have symptoms at rest within the previous 24 h and evidence of ischemic heart disease (15). Exclusion criteria included plans for revascularization in the first 24 h, regardless of the initial response to medical therapy, coronary artery bypass graft surgery (CABG) within the previous two months or percutaneous transluminal coronary angioplasty (PTCA) within six months and renal insufficiency (serum creatinine >2.0 mg/dl). The TIMI-11B protocol, including the biomarker substudy, was approved by the institutional Committee on Human Research at each of the participating centers.
During the acute phase of TIMI-11B, patients were treated with either intravenous UFH (n = 190) or weight-adjusted dosing of subcutaneous enoxaparin (n = 169). The median duration of treatment with intravenous UFH and acute-phase subcutaneous enoxaparin in the substudy group was 3.0 and 3.5 days, respectively. Patients eligible for the chronic-term phase (e.g., no bleeding or CABG) were then administered fixed-dose subcutaneous enoxaparin or placebo for an additional 35 days (59% of patients in the enoxaparin group and 52% in the UFH/placebo group), starting at the time of discharge or day 8, whichever occurred first.
Blood sampling and measurement of serum markers
The serum marker protocol specified that on enrollment and 12 h to 24 h later, venous blood samples be drawn and serum stored at −20°C or colder at the enrolling site until shipped to the Core Laboratory at Children’s Hospital Medical Center (Boston, Massachusetts), where they were stored at −80°C until analyzed.
Cardiac TnI results were determined in the Chemistry Laboratory at Brigham and Women’s Hospital (Boston, Massachusetts) on thawed sera using the Dimension RxL cTnI assay (Dade Behring). All operators had no knowledge of patient treatment or clinical outcomes. The day-to-day precision of this assay in the Brigham and Women’s Hospital laboratory, at cTnI concentrations of 0.57 ng/ml, 2.58 ng/ml and 12.22 ng/ml, is reflected by coefficients of variation of 8.8%, 3.1% and 1.5%, respectively. The minimal detectable concentration is 0.04 ng/ml. The 97.5 percentile for 197 healthy blood donors tested in our laboratory was determined to be 0.1 ng/ml. In its “Standards of Laboratory Practice: Recommendations for the Use of Cardiac Markers” the National Academy of Clinical Biochemistry has recommended that the low abnormal decision limit for cardiac troponins in ACS be guided by the 97.5 percentile among normal control subjects (16). Consistent with this recommendation and clinical experience in our laboratory, we used a level of 0.1 ng/ml as the cTnI threshold for analyses in this substudy.
The clinical end point of MI in TIMI-11B was restricted to a new MI occurring after the qualifying ischemic event and included either new diagnostic Q waves or left bundle branch block not present on the enrollment electrocardiogram (ECG) or enzyme evidence of a new infarction (15). Elevation of CK-MB above the upper limit of normal (ULN) and at least 50% over the previous value or, if the CK-MB value was unavailable, elevation of total CK above two times the ULN and ≥25% of the previous value was necessary to meet enzymatic criteria for a new MI. For patients within 24 h of PTCA or CABG, total CK and CK-MB values were required to be ≥50% of the previous value and three and five times the ULN, respectively (15). Participants were determined to have recurrent ischemia requiring urgent revascularization if they had either recurrent angina prompting the performance of coronary revascularization during the same hospital stay or an episode of unstable angina after discharge, resulting in repeat hospital admission and coronary intervention. All clinical end points in TIMI-11B, including recurrent ischemia requiring urgent revascularization, were adjudicated by an independent Clinical Endpoints Committee, which had no knowledge of treatment assignment, incorporating a detailed review of relevant ECGs, serum marker data, angiographic findings and a narrative summary.
Statistical comparison of baseline characteristics and clinical outcomes were performed using the chi-square or Fisher exact test for dichotomous variables and either the Wilcoxon rank-sum or two-sample t test for continuous variables. Multivariate analyses were performed using a logistic regression model incorporating recognized cardiovascular risk factors (age, gender, history of coronary artery disease, diabetes, hypertension, tobacco use, ST segment deviation and presence of congestive heart failure at presentation). Testing for a significant interaction between categoric cTnI results and treatment arm was performed using the Breslow-Day test for homogeneity. Data analyses were performed using SAS statistical software (version 6.12, SAS Institute, Cary, North Carolina). All statistical comparisons were two-tailed, and p < 0.05 was considered to indicate statistical significance.
This substudy of TIMI-11B included 359 patients with unstable coronary disease and negative serial CK-MB measurements who had serum samples obtained at baseline or 12 h to 24 h after enrollment, or both, available for determination of cTnI values in our laboratory. Demographic characteristics of patients eligible for this serum marker substudy showed higher rates of hypertension, previous angina and CABG, but were otherwise representative of those in the overall trial (Table 1), with a profile typical of patients presenting with severe unstable ischemic heart disease, including high rates of previous coronary artery disease (46%) and ST segment changes on presentation (81%). Patients in the substudy with a positive cTnI value (≥0.1 ng/ml) were more likely to be male and were slightly older (Table 1).
Cardiac TnI and clinical outcomes
Three hundred and forty-one patients with negative CK-MB values had pre-enrollment cTnI determinations, with 155 (45.5%) showing elevated baseline cTnI levels. Of the 359 patients with samples from either time point, 180 (50.1%) had a positive cTnI result within the first 24 h after enrollment. There were 13 deaths (3.6%) by 14 days in the substudy group. Twenty-three patients (6.4%) had a MI after the qualifying index event, and 47 (13.1%) had severe recurrent ischemia requiring urgent revascularization.
Patients with an elevated cTnI value during the first 24 h were at significantly increased risk for adverse clinical events by as early as 48 h, as well as by 14 and 43 days (Table 2). An elevated cTnI value identified patients at a 6% higher absolute risk of death and a fourfold (95% confidence interval [CI] 2.5 to 7.7) higher risk of death, nonfatal MI or urgent revascularization (p < 0.0001) by two weeks after presentation. Moreover, patients with an abnormal cTnI value demonstrated a nearly sixfold higher risk of severe ischemia requiring urgent revascularization by 48 h after enrollment (p < 0.001), with a greater than 14% absolute risk difference at both 14 and 43 days (p ≤ 0.0001) (Table 2). Although utilization of the 12- to 24-h measurements added to the strength of the prognostic relation, baseline measurements alone were also highly predictive of each component of the composite end point by 14 days (Fig. 1), as well as severe recurrent ischemia requiring revascularization by 48 h (10.3 vs. 1.6%, p = 0.0005).
In a multivariate model including major clinical risk factors (age, gender, diabetes, tobacco use, history of coronary artery disease, hypertension), as well as the presence of ST segment deviation (depression or transient elevation) on the qualifying ECG, a cTnI level ≥0.1 ng/ml was an independent predictor of adverse clinical outcomes at 14 days (Table 3). Furthermore, baseline elevation of cTnI was independently associated with a higher risk of severe recurrent ischemia necessitating urgent revascularization at 48 h (p = 0.005) and 14 days (p = 0.003).
Cardiac TnI and enoxaparin treatment
Consistent with the superior efficacy of enoxaparin over UFH in patients in the main TIMI-11B trial (15), patients in the cTnI substudy treated with enoxaparin had fewer adverse clinical events (death, MI or urgent revascularization) by 14 days (relative risk 0.65, 95% CI 0.40 to 0.97). Stratification of the substudy group by cTnI results from 0 to 24 h revealed a greater relative benefit of enoxaparin therapy among patients with an abnormal cTnI value (Breslow-Day test, p = 0.03, indicating heterogeneity of the treatment effect of enoxaparin vs. UFH with respect to death, MI or urgent revascularization). Patients with an elevated cTnI value experienced a 17% absolute reduction in the risk of severe recurrent ischemia prompting revascularization and nearly half the risk of death, MI or urgent revascularization by 14 days (Fig. 2).
In this substudy of patients with unstable coronary syndromes and negative serial CK-MB measurements enrolled in TIMI-11B, elevated levels of cTnI (≥0.1 ng/ml) soon after presentation were strongly associated with very early adverse clinical outcomes, including a sixfold higher risk of recurrent ischemia requiring urgent revascularization by as early as 48 h after the initiation of study medication. Notably, an abnormal cTnI value identified patients who were at higher risk for each component of the composite end point (death, MI and recurrent ischemia prompting urgent revascularization) at 14 days. Consistent with previous observations, although elevated baseline levels of cTnI were highly correlated with recurrent cardiac events, repeat determination 12 h to 24 h after presentation added to the prognostic capacity of the marker. Furthermore, cTnI offered prognostic information independent of established clinical risk markers, including age and ST segment deviation on the presenting ECG, and identified higher risk patients who demonstrated particular benefit from enoxaparin versus UFH therapy.
Cardiac TnI and prognosis in unstable angina
A number of previous studies have evaluated the prognostic capacity of cTnI in patients with suspected ischemic syndromes but no elevation of CK-MB. Among 948 patients with non–ST segment elevation ACS and negative CK-MB enrolled in TIMI-IIIB, elevation of baseline cTnI ≥0.4 ng/ml, as measured with an earlier generation assay (Stratus II, Dade Behring), was associated with a threefold higher mortality risk at six weeks (5). Galvani et al. (6) made similar observations regarding the prognostic capacity of cTnI with respect to death and nonfatal MI at 30 days and one year among 91 patients with Braunwald class IIIB unstable angina. Data from the ThRombin Inhibition In Myocardial ischemia (TRIM) investigators (7) demonstrated that cTnI levels ≥2.0 ng/ml (Opus, Behring Diagnostics) were associated with higher 30-day mortality among 309 patients with unstable angina. A meta-analysis including each of these studies showed a cumulative odds ratio of 4.2 (95% CI 2.7 to 6.4) for death or MI over a mean follow-up period of 42 days after presentation with unstable angina and elevated cTnI (8).
Data from TIMI-11B—clinical implications
Our current report expands on previous studies in several areas of clinical importance. Although the predictive capacity of cTnI with respect to death and MI at four to six weeks after presentation with unstable angina is well documented (5–8), data regarding earlier clinical outcomes are limited. In addition, although several previous investigations have directly explored the relation between cTnI levels and the incidence of recurrent ischemia or revascularization, or both, the data from these studies have been inconclusive (6,7,17). Our report addresses each of these areas that impact decisions regarding the early management of patients with non–ST segment elevation ACS and negative CK-MB values, who might otherwise be considered at lower short-term risk for poor outcomes. These data also suggest that angiographic observations of an association between troponin elevation and thrombus burden/lesion complexity (18) may have immediate clinical relevance with respect to the patient’s risk of early recurrent ischemic events.
Cardiac TnI and therapeutic decision-making
Given that economic and safety considerations are likely to prohibit the uniform use of expensive pharmacologic and invasive therapies, the identification of patients with non–ST segment elevation ACS who will benefit most from these interventions has become central to the care of this group of patients. Our report suggests that cTnI levels measured soon after presentation are useful in selecting patients with negative CK-MB values who are at higher risk for early adverse outcomes, including severe recurrent ischemia, and who might therefore be considered candidates for more aggressive medical therapy or an early invasive intervention. Our findings also offer evidence that treatment with enoxaparin, a low molecular weight heparin with demonstrated superiority over UFH (15), reduces the risk of poor outcomes associated with an abnormal cTnI level. As such, our data indicate that cTnI, like cardiac troponin T (11,13), may be useful in targeting new antithrombin and antiplatelet (12,14) therapies.
The TIMI-11B substudy enrolled a high risk group of patients with non–ST segment elevation ACS. One must maintain some caution in extrapolating our data on patients with evidence of ischemic heart disease to more heterogeneous populations with chest symptoms. Furthermore, a number of factors contribute to analytic variability between cTnI methods, as reflected in the diversity of decision limits used in clinical laboratories and reported in the literature (19–21). Thus, although our results provide an evidenced-based guide for the clinical use of a current generation assay (Dimension RxL), one should be wary of generalizing the threshold used in this study to other cTnI assays. Nevertheless, it is interesting to note that in a diverse group of patients with chest pain, a bedside device with a detection limit of 0.1 ng/ml (Spectral Diagnostics; Toronto, Canada) reliably identified patients at higher risk for cardiac death or nonfatal MI over a follow-up of 30 days (9).
Although during the period of this study, few centers participating in TIMI-11B were measuring cardiac troponins in their local laboratories, we cannot exclude the possibility that in few cases, local marker results might have influenced the investigator’s decisions regarding revascularization. However, the end point of severe recurrent ischemia prompting urgent revascularization was carefully adjudicated by an independent Clinical Events Committee and required evidence that both an episode of recurrent ischemia had occurred and the recurrent ischemia had prompted revascularization. Furthermore, given that treatment allocation was blinded, there is no reason to expect systematic bias in the interpretation of any local laboratory measurements with respect to the treatment arm.
This cTnI substudy from TIMI-11B demonstrates a consistent and significant increase in the risk of early adverse cardiovascular outcomes associated with abnormal levels of cTnI among patients with non–ST segment elevation ACS and negative values of CK-MB. The predictive capacity of cTnI is evident as early as 48 h after presentation and is independent of multiple, well-described clinical risk factors, including age and ST segment deviation. Moreover, an elevated cTnI value identifies patients who may derive particular benefit from enoxaparin therapy.
☆ This study was supported by Dade Behring, Newark, Delaware, and Rhone-Poulenc Rorer, Collegeville, Pennsylvania.
- acute coronary syndrome
- coronary artery bypass graft surgery
- confidence interval
- creatine kinase, MB fraction
- cardiac troponin I
- myocardial infarction
- percutaneous transluminal coronary angioplasty
- Thrombolysis in Myocardial Infarction
- unfractionated heparin
- upper limit of normal
- Received December 30, 1999.
- Revision received May 8, 2000.
- Accepted June 28, 2000.
- American College of Cardiology
- Braunwald E
- Theroux P,
- Fuster V
- ↵Braunwald E, Mark D, Jones R, et al. Unstable Angina: Diagnosis and Management. Clinical Practice Guidline Number 10. AHCPR Publication No. 94-0602. Rockville, MD: Agency for Health Care Policy and Research and the National Heart, Lung, and Blood Institute, Public Health Service, U.S. Department of Health and Human Services, 1994:154.
- Bodor G.S,
- Porter S,
- Landt Y,
- Ladenson J.H
- Galvani M,
- Ottani F,
- Ferrini D,
- et al.
- ThRombin Inhibition In Myocardial ischemia (TRIM) Study Group,
- Luscher M.S,
- Thygesen K,
- Ravkilde J,
- Heickendorff L
- Morrow D.A,
- Rifai N,
- Tanasijevic M.J,
- Wybenga D.R,
- de Lemos J.A,
- Antman E.M
- Fragmin in Unstable Coronary Artery Disease (FRISC) Study Group,
- Lindahl B,
- Venge P,
- Wallentin L
- Antman E.M,
- McCabe C.H,
- Gurfinkel E.P,
- et al.
- Wu A,
- Apple F,
- Gibler W,
- Jesse R,
- Warshaw M,
- Valdes R
- GUSTO-IIa Investigators,
- Christenson R.H,
- Duh S.H,
- Newby L.K,
- et al.
- Heeschen C,
- van Den Brand M.J,
- Hamm C.W,
- Simoons M.L
- Wu A.H,
- Feng Y.J,
- Moore R,
- et al.
- Katrukha A.G,
- Bereznikova A.V,
- Esakova T.V,
- et al.
- Apple F.S