Treatment and outcomes of patients with evolving myocardial infarction

Download PDF
2 downloads 90 Views 81KB Size Report
Comment
According to the American College of Cardiology/American ..... Papuchis G, Mautner B, Corbalan R, Radley D, Braunwald E. The TIMI risk score for unstable ...
Clinical research

European Heart Journal (2007) 28, 1079–1084 doi:10.1093/eurheartj/ehm016

Coronary heart disease

Treatment and outcomes of patients with evolving myocardial infarction: experiences from the SYNERGY trial Chadwick D. Miller1*, Anindita Banerjee2, Kenneth W. Mahaffey3, Michael C. Kontos4, Gregory Fermann5, Charles V. Pollack Jr.6, Elliott Antman7, Philip Aylward8, Shaun G. Goodman9, Renato Santos10, James J. Ferguson11, Robert M. Califf3, and James W. Hoekstra1 1

Department of Emergency Medicine, Wake Forest University School of Medicine, Winston-Salem, NC, USA; 2Department of Statistics, North Carolina State University, Raleigh, NC, USA; 3Duke Clinical Research Institute, Durham, NC, USA; 4VCU Medical Center, Virginia Commonwealth University, Richmond, VA, USA; 5Department of Emergency Medicine, University of Cincinnati, Cincinnati, OH, USA; 6Department of Emergency Medicine, Pennsylvania Hospital, Philadelphia, PA, USA; 7 Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital, Boston, MA, USA; 8Flinders Medical Centre, Bedford Park, South Australia, Australia; 9Canadian Heart Research Centre and Terrence Donnelly Heart Centre, Division of Cardiology, St Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada; 10Section of Cardiology, Wake Forest University School of Medicine, Winston-Salem, NC, USA; and 11Texas Heart Institute, Houston, TX, USA Received 3 November 2006; revised 17 January 2007; accepted 15 February 2007; online publish-ahead-of-print 3 April 2007

See page 1043 for the editorial comment on this article (doi:10.1093/eurheartj/ehm083)

KEYWORDS Evolving myocardial infarction; Troponin; Initial treatment; Outcomes

Aims Patients with myocardial infarction (MI) presenting immediately after symptom onset may be treated less aggressively due to their non-elevated troponin status. We compared the initial treatment and clinical outcomes of patients presenting with evolving MI (EMI) with those presenting with MI. Methods and results This study analysed data from the Superior Yield of the New strategy of Enoxaparin, Revascularisation, and Glycoprotein IIb/IIIa inhibitors (SYNERGY) trial, which enrolled patients meeting at least two of the following: age60 years, elevated cardiac biomarkers, or ST-segment changes. Patients were stratified by troponin results obtained within 12 h of presentation: EMI [initial troponin (2), second troponin (þ)], MI [initial troponin (þ)], and no MI at enrolment [first and second troponin (2)]. Comparisons were made using Wilcoxon rank-sum and x2 tests. Of the 8309 patients with complete data, 5503 (66%) had MI, 1686 (20%) had EMI, and 1120 (13%) had no MI. Treatment patterns prior to enrolment were similar among EMI and MI patients [aspirin (88 vs. 86%), beta-blockers (62 vs. 61%), heparin (83 vs. 81%), and glycoprotein IIb/IIIa inhibitors (23 vs. 24%)]. Similar rates of percutaneous coronary intervention (48 vs. 50%) and coronary artery bypass grafting (21 vs. 22%) were seen after enrolment. Patients presenting with MI had a higher rate of death or recurrent MI compared with patients with EMI [16 vs. 13%, adjusted OR 1.22 (95% CI 1.04, 1.44)]. Conclusion Initial treatment patterns were similar among patients with EMI and MI in the SYNERGY trial. Patients with EMI had lower rates of death or re-infarction at 30 days compared with patients presenting with positive troponin results.

Introduction Emergency department (ED) patients with chest pain and presumed acute coronary syndrome (ACS) are stratified according to risk using information derived from the history, physical examination, electrocardiogram (ECG), and cardiac biomarkers. According to the American College of Cardiology/American Heart Association (ACC/AHA) guidelines, high-risk criteria for adverse outcomes in patients with ACS include positive biomarkers, ST-segment deviation, advanced age, arrhythmias, haemodynamic instability, congestive heart failure (CHF), ongoing rest pain, and an

* Corresponding author. Tel: þ1 336 716 1740; fax: þ1 336 716 5438. E-mail address: [email protected]

accelerated pattern of angina. Patients meeting high-risk criteria should receive both aggressive pharmacological therapy and an early invasive management strategy.1 Of the high-risk ACS criteria, positive biomarkers—particularly elevated troponin—are especially significant. Elevated troponin values signify that myocardial necrosis has occurred, offer an objective determinant that ACS is present,2 and correlate with mortality.1,3,4 Furthermore, clinical trials indicate that patients with elevated serum troponin levels benefit most from aggressive treatment.5–7 However, troponin values do not immediately become abnormal after the onset of myocardial infarction (MI). Cell membrane rupture must occur before cardiac biomarkers are released, making detection in the blood possible; this often requires 6–8 h. Thus, patients presenting

& The European Society of Cardiology 2007. All rights reserved. For Permissions, please e-mail: [email protected]

1080

C.D. Miller et al.

within 6–8 h of symptom onset initially may have negative cardiac biomarkers, delaying early recognition of MI and the use of aggressive therapy. The initial care and outcomes of patients presenting with MI and initially negative cardiac biomarkers have not been well described. We analysed data from the Superior Yield of the New strategy of Enoxaparin, Revascularisation, and Glycoprotein IIb/IIIa inhibitors (SYNERGY) trial8 to compare the treatment patterns and outcomes of patients with evolving MI (EMI) to patients with MI.

included treatment with aspirin, beta-receptor antagonists, enoxaparin, UFH, clopidogrel, and GP IIb/IIIa inhibitors.

Methods

Statistical methods

Study design

Data are presented as medians with 25th and 75th inter-quartiles for continuous variables and as frequencies with percentages for discrete variables. Statistical comparisons were made among the EMI and MI groups. All modelling also included the no-MI group. Univariate analyses were performed using Wilcoxon rank-sum test for ordinal measures and the Pearson’s x2 test for dichotomous measures. These same tests were used when comparing the two groups for the procedural factors. Comparisons between the two MI groups for the efficacy endpoints were performed using logistic regression modelling techniques. Multivariable models had been developed previously for mortality within 30 days of randomization, re-infarction within 30 days of randomization, and death or MI within 30 days of randomization. Characteristics found to be statistically significant in predicting death alone were smoking status, age, weight, region of enrolment, creatinine, rales, pulse, t-wave inversion, treatment, and an interaction term of treatment and current smoking. To predict MI by 30 days post-randomization, age, sex, height, race, region of enrolment, diabetes, smoking status, pulse, creatinine, and prior MI were found to be statistically significant. Lastly, age, sex, region of enrolment, height, race, diabetes, smoking status, pulse, creatinine, and prior MI were significant predictors of death or MI within 30 days of randomization. EMI, MI, and no-MI classifications were then added to these models to calculate adjusted results. Two-sided tests were run using a P-value of 0.05 to determine the statistical significance. As this is a hypothesis-generating manuscript, no adjustments were made to the pair-wise comparisons. These results are to be used to identify areas for further analyses and not as confirmatory results. Statistical software was used for the analyses (SAS Institute, v. 8.2, Cary, NC, USA).

This investigation is a secondary analysis of data collected during the SYNERGY trial. Enrolling centres obtained institutional review board or Ethics Committee approval for patient enrolment, and all patients provided written consent for participation. The institutional review board of Wake Forest University School of Medicine has approved this analysis.

Study setting and population Details of the SYNERGY trial have been previously reported in detail.8,9 In brief, patients presenting to one of 467 participating sites in 12 countries with ischaemic symptoms of at least 10 min duration within the previous 24 h were enrolled if they met two of three inclusion criteria: age60 years, troponin or creatinine kinase-MB elevation above the upper limit of normal for the local laboratory, or definitive ST-segment changes on 12-lead ECG. Patients in this trial were randomized to receive unfractionated heparin (UFH) or enoxaparin. The use of ACC/AHA guidelines-recommended therapies and procedures, including aspirin, glycoprotein (GP) IIb/IIIa inhibitors, and early cardiac catheterization, were encouraged. Patients were followed for 1 year to determine the outcomes.

Study protocol This analysis evaluated data from patients with two troponin assay results available within 12 h of presentation or one positive troponin result drawn within 12 h of presentation. Subjects who underwent percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) surgery within 12 h of presentation and did not have a prior positive troponin, who only had one normal troponin result within 12 h of presentation, and who had incomplete biomarker data were excluded. Included patients were divided into three groups. (1) Evolving MI. Patients were considered to have an EMI if their initial troponin result was below the institutional limit for normal and was followed by an additional assay that was above the institutional limit for normal within 12 h of hospital arrival; also, if PCI or CABG was performed, it was after the positive troponin. (2) MI at presentation. Patients were considered to have an MI at presentation if their initial troponin assay was above the institutional upper limit of normal and was obtained within 12 h of hospital arrival; if PCI or CABG was performed, it was after the positive troponin. (3) No MI at presentation. Patients were considered to have no MI at presentation if at least two troponin results were available within 12 h of hospital arrival and all were negative; these patients also did not undergo PCI or CABG within 12 h of hospital arrival.

Measurements and outcome measures Initial medical therapy Medical treatments administered between the time of hospital arrival and enrolment were recorded on the case report form. These

Clinical outcomes The primary efficacy outcome for this study was 30-day death or recurrent MI as described previously.8,9 Patients with MI at enrolment were not considered in the composite outcome to have recurrent infarction unless the definition of recurrent infarction was met. Secondary endpoints included the use of catheterization and revascularization procedures, as well as the time to each of these procedures.

Results The SYNERGY trial included data from 9978 patients enrolled at 467 centres in 12 countries. Patients were excluded from our analysis due to missing troponin values and times within 12 h (n ¼ 1540), missing revascularization information (n ¼ 2), PCI or CABG prior to a positive troponin (n ¼ 78), and revascularization occurring prior to the negative troponin value within 12 h in the no-MI group (n ¼ 46), leaving 8312 (83%) satisfying our inclusion criteria. Excluded patients were statistically similar to included patients with respect to prior MI, prior CABG, prior CHF, and the outcome measures of death, MI, and their composite within 30 days. Excluded patients were significantly more likely to have ST-segment depression (60 vs. 54%) and be enrolled based on age and ECG changes (37 vs. 12%). Of 8312 included patients, 5597 (66%) had MI at enrolment, 1707 (20%) had an EMI, and 1008 (13%) had no MI at enrolment. Baseline characteristics of these groups are shown in Table 1. Compared with MI patients, EMI patients were similar in most demographic categories including age.60 years (both 74%), prior MI (28 vs. 27%), diabetes

Treatment patterns in patients with EMI

1081

Table 1 Demographics

Age.60, n/N (%) Female sex, n/N (%) White race, n/N (%) Medical history Prior MI, n/N (%) Prior CABG, n/N (%) Prior CHF, n/N (%) Prior hypertension, n/N (%) Prior stroke, n/N (%) Hypercholesterolaemia, n/N (%) Diabetes, n/N (%) Tobacco use, n/N (%) Prior angina, n/N (%) ST-depression on initial ECG, n/N (%) Enrolment criteria Age.60, (þ) ECG, (þ) biomarkersa, n/N (%) Age.60, (2) ECG, (þ) biomarkersa, n/N (%) Age.60, (þ) ECG, (2) biomarkersa, n/N (%) Age,60, (þ) ECG, (þ) biomarkersa, n/N (%) Time from symptom onset to hospital arrival (h), N, median (Q1, Q3) Time from hospital arrival to enrolment (h), N, median (Q1, Q3) TIMI risk score, N, median (Q1, Q3) PURSUIT risk score, N, median (Q1, Q3)

Overall N ¼ 8312

MI N ¼ 5597

EMI N ¼ 1707

No MI N ¼ 1008

6256/8312 (75.3) 2757/8312 (33.2) 7170/8312 (86.3)

4127/5597 (73.7) 1818/5597 (32.5) 4921/5597 (86.1)

1262/1707 (73.9) 530/1707 (31.0) 1477/1707 (86.5)

867/1008 (86.0) 409/1008 (40.6) 872/1008 (86.5)

2311/8286 (27.9) 1372/8307 (16.5) 778/8311 (9.4) 5597/8312 (67.3) 416/8312 (5.0) 4851/8262 (58.7) 2414/8312 (29.0) 4961/8307 (59.7) 3684/8311 (44.3) 4477/8298 (54.0)

1502/5582 (26.9) 827/5593 (14.8) 540/5596 (9.6) 3747/5597 (66.9) 286/5597 (5.1) 3320/5561 (57.9) 1624/5597 (29.0) 3396/5593 (60.7) 2422/5597 (43.3) 2882/5585 (51.6)

481/1701 (28.3) 313/1706 (18.3) 143/1707 (8.4) 1106/1707 (64.8) 71/1707 (4.2) 1001/1698 (59.0) 465/1707 (27.2) 1023/1707 (59.9) 704/1706 (41.3) 894/1706 (52.4)

328/1003 (32.7) 232/1008 (23.0) 95/1008 (9.4) 744/1008 (73.8) 59/1008 (5.9) 630/1003 (62.8) 325/1008 (32.2) 542/1007 (53.8) 558/1008 (55.4) 701/1007 (69.6)

3797/8108 (46.8)

2918/5485 (53.2)

738/1672 (44.1)

141/951 (14.8)

1729/8108 (21.3)

1218/5485 (22.2)

390/1672 (23.3)

121/951 (12.7)

955/8108 (11.8)

154/5485 (2.8)

176/1672 (10.5)

625/951 (65.7)

1627/8108 (20.1)

1195/5485 (21.8)

368/1672 (22.0)

64/951 (6.7)

7050, 3.1 (1.4, 7.1)

4841, 4.0 (1.7, 8.9)

1428, 1.7 (1.0, 3.1)

781, 2.6 (1.2, 6.0)

6796, 8.2 (3.6, 15.2)

4674, 6.7 (3.2, 14.0)

1373, 13.2 (8.0, 18.3)

749, 6.6 (2.9, 14.6)

8312, 4 (3, 5) 7862, 14 (13, 15)

5297, 4 (3, 5) 5296, 14 (13, 15)

1707, 4 (3, 5) 1626, 13 (12, 14)

1008, 4 (3, 5) 940, 13 (12, 14)

CABG, coronary artery bypass grafting; CHF, congestive heart failure; ECG, electrocardiogram; EMI, evolving myocardial infarction; MI, myocardial infarction; PURSUIT, Platelet glycoprotein IIb/IIIa in Unstable angina: Receptor Suppression Using Integrilin Therapy; TIMI, Thrombolysis In Myocardial Infarction. a Includes elevated creatine kinase-MB and/or troponin results.

mellitus (27 vs. 29%), and ST-segment depression on presenting ECG (both 52%). Enrolment criteria varied by group. EMI patients were less likely than MI patients to be enrolled based upon age.60, ECG changes, and elevated biomarkers (troponin or creatine kinase-MB). However, EMI patients were more likely to be enrolled based upon age.60 and ECG changes with non-elevated biomarkers. EMI patients presented sooner after symptom onset than did MI patients (median 1.7 vs. 4.0 h) yet were enrolled later after presentation (median 13.2 vs. 6.7 h). The subset of EMI patients enrolled based upon age.60 and ECG changes had the shortest time from symptom onset to enrolment [median 5.8 h (Q1 3.4 h, Q3 10.2 h)]. Medical therapies provided are displayed in Table 2. Prior to enrolment, similar usage patterns for pharmacological therapies were seen among EMI and MI patients, including aspirin (88 vs. 86%), beta-blockers (62 vs. 61%), heparin [either low-molecular-weight heparin or unfractionated (80 vs. 77%)], clopidogrel (26 vs. 27%), and GP IIb/IIIa inhibitors (23 vs. 24%). Medical therapies and in-hospital procedures during the baseline hospitalization were similar among EMI and MI groups (Table 3). Clinical outcomes of 30-day death, re-infarction, and the composite of these measures are shown in Table 4. Among groups, MI patients had the highest incidence of the composite outcome (15.9% MI vs. 12.9% EMI, unadjusted OR 1.27,

95% CI 1.09, 1.49) and remained statistically different from the EMI group after adjustment [adjusted OR 1.22 (95% CI 1.04, 1.43)]. This finding was driven by a higher rate of both death and re-infarction among MI patients. Patients with EMI had better 30-day composite outcomes in both SYNERGY treatment assignments—enoxaparin [MI 15.4% vs. EMI 13.8%, OR 1.10 (95% CI 0.88, 1.38)] and heparin [MI 16.3% vs. EMI 11.9%, OR 1.36 (95% CI 1.07, 1.72)].

Discussion The incorporation of troponin assays into clinical practice has changed the triage of patients with chest pain. Elevated troponin values correlate with mortality1,3,4 and help identify a subset of patients who will benefit from the most aggressive therapies.5–7 However, a prominent limitation to these tests is the time it takes for troponin to become released in the serum—6–8 h after MI. Fortunately, other features besides cardiac biomarkers have proved useful in the risk stratification of patients with chest pain. The ACC/AHA guidelines1 list 11 criteria that place patients at high risk for short-term death or infarction, only one of which is related to cardiac biomarkers. The extent to which these other elements—such as medical history, physical examination findings, and ECG results—are used to guide treatment decisions is largely unknown.

1082

C.D. Miller et al.

Table 2 Medical treatment during hospitalization Treatment Aspirin Prior to enrolment, n/N (%) During baseline hospitalization, n/N (%) Clopidogrel Prior to enrolment, n/N (%) During baseline hospitalization, n/N (%) Beta-blocker Prior to enrolment, n/N (%) During baseline hospitalization, n/N (%) Heparin (LMWH or unfractionated) Prior to enrolment, n/N (%) GP IIb/IIIa inhibitor Prior to enrolment, n/N (%) During baseline hospitalization: peri-procedural, n/N (%)

Total

MI

EMI

No MI

7117/8312 (85.6) 7915/8312 (95.2)

4783/5597 (85.5) 5353/5597 (95.6)

1497/1707 (87.8) 1623/1707 (95.1)

837/1008 (83.0) 939/1008 (93.2)

2195/8312 (26.4) 5300/8312 (63.8)

1501/5597 (26.8) 3657/5597 (65.3)

446/1707 (26.1) 1105/1707 (64.7)

248/1008 (24.6) 538/1008 (53.4)

5042/8312 (60.7) 7200/8312 (86.6)

3394/5597 (60.6) 4895/5597 (87.5)

1058/1707 (62.0) 1481/1707 (86.8)

590/1008 (58.5) 824/1008 (81.7)

6310/8312 (75.9)

4306/5597 (76.9)

1364/1707 (79.9)

640/1008 (63.5)

1835/8287 (22.1) 2674/3908 (68.4)

1313/5582 (23.5) 1913/2776 (68.9)

398/1704 (23.4) 558/816 (68.4)

124/1001 (12.4) 203/316 (64.2)

EMI, evolving myocardial infarction; GP, glycoprotein; LMWH, low-molecular-weight heparin.

Table 3 In-hospital procedures Catheterization

Total

MI

EMI

No MI

In-hospital post-rand diagnostic catheterization, n/N (%) Hours from enrolment to first catheterization, N, median (Q1, Q3) In-hospital revascularization In-hospital CABG, n/N (%) Hours from enrolment to CABG, N, median (Q1, Q3) PCI during baseline and post-rand hospitalization, n/N (%) Hours from enrolment to first PCI, N, median (Q1, Q3)

7664/8312 (92.2)

5190/5597 (92.7)

1588/1707 (93.0)

886/1008 (87.9)

7617, 21.8 (7.0, 43.9)

5164, 21.3 (6.3, 43.3)

1556, 23.1 (7.7, 45.9)

877, 22.5 (14.2, 42.9)

1594/8311 (19.2) 1590, 90.5 (44.7, 166.6)

1131/5597 (20.2) 1131, 89.9 (44.7, 164.2)

332/1706 (19.5) 329, 93.4 (44.1, 187.8)

131/1008 (13.0) 130, 90.5 (44.1, 163.3)

3908/8312 (47.0)

2776/5597 (49.6)

816/1707 (47.8)

316/1008 (31.3)

3906, 22.9 (7.5, 49.1)

2776, 22.2 (6.6, 48.4)

815, 24.1 (7.4, 49.7)

315, 25.9 (18.6, 52.9)

CABG, coronary artery bypass grafting; EMI, evolving myocardial infarction; MI, myocardial infarction; PCI, percutaneous coronary intervention.

This analysis examines the treatments administered to patients with EMI prior to enrolment in the SYNERGY clinical trial. The results demonstrate that the initial treatment patterns of patients presenting with EMI are similar to those presenting with MI. In fact, only minor differences in use were seen among EMI and MI groups for all guideline-recommended therapies, including aspirin, heparin, beta-blockers, clopidogrel, and GP IIb/IIIa inhibitors. These aggressive treatment patterns suggest that clinicians used and acted upon other high-risk features besides troponin values. These results also demonstrate that subjects with EMI have lower death and re-infarction rates compared with patients presenting with MI. Examining this finding further, EMI patients arrived earlier than MI patients (1.7 vs. 4.0 h) and both groups received similar treatments. Thus, EMI patients received aggressive, guideline-recommended care earlier in their course of infarction compared with those presenting with MI. It is likely that this early hospitalization and aggressive therapy translates into a decreased rate of adverse events.

Analysis of the Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation of the ACC/AHA Guidelines (CRUSADE) registry by Roe et al.10 suggests that treatment patterns among EMI and MI patients are similar in the first 24 h. In this analysis, subjects with initially negative troponin values that subsequently turned positive received treatments in the first 24 h similar to that received by subjects with initially positive troponin results. The present analysis from the SYNERGY trial builds upon the CRUSADE results by examining therapy earlier in the hospital course and suggests that the similarities in care begin long before the 24 h mark. As such, the care patterns described in the SYNERGY trial are more likely to reflect therapies ordered or administered in the ED. Risk stratification of patients with chest pain utilizes information available at one moment in time to estimate the probability of cardiac disease and the likelihood of adverse cardiac events. Using these estimates, clinicians can adjust treatment regimens according to risk. Once riskstratified, the EMI patients in the SYNERGY trial were a highrisk cohort, and physicians were correct in treating them as

Treatment patterns in patients with EMI

1083

Table 4 Adjusted clinical outcomes of evolving myocardial infarction vs. myocardial infarction patients Outcomes

Total

MI

Composite 30-day death or re-infarction Total, n/N (%) 1197/8308 (14.4) 888/5597 (15.9) Enoxaparin, n/N (%) 590/4131 (14.3) 428/2773 (15.4) UFH, n/N (%) 607/4180 (14.5) 460/2824 (16.3) 30-day death Total, n/N (%) 268/8312 (3.2) 209/5597 (3.7) Enoxaparin, n/N (%) 141/4131 (3.4) 107/2773 (3.9) UFH, n/N (%) 127/4181 (3.0) 102/2824 (3.6) 30-day re-infarction Total, n/N (%) 1017/8312 (12.2) 745/5597 (13.3) Enoxaparin, n/N (%) 492/4131 (11.9) 353/2772 (12.7) UFH, n/N (%) 525/4181 (12.6) 392/2824 (13.9)

EMI

No MI

MI vs. EMI adjusted OR (95% CI)

220/1706 (12.9) 121/877 (13.8) 99/829 (11.9)

89/1008 (8.8) 41/481 (8.5) 48/527 (9.1)

1.22 (1.04, 1.43) 1.10 (0.88, 1.38) 1.36 (1.07, 1.72)

42/1707 (2.5) 28/877 (3.2) 14/830 (1.7)

17/1008 (1.7) 6/481 (1.3) 11/527 (2.1)

1.48 (1.04, 2.11) 1.18 (0.76, 1.84) 2.17 (1.19, 3.96)

194/1707 (11.4) 102/877 (11.6) 92/830 (13.9)

78/1008 (7.7) 37/481 (7.7) 41/527 (7.8)

1.16 (0.98, 1.38) 1.09 (0.86, 1.39) 1.22 (0.96, 1.55)

CI, confidence interval; EMI, evolving myocardial infarction; MI, myocardial infarction; OR, odds ratio.

such despite their negative troponin status. Among riskstratification systems, age, elevated cardiac biomarkers, and ischaemic ECG findings consistently have been found to identify a high-risk cohort of patients among those who present with possible MI. Most of our EMI population had at least two of those criteria, with many having all three. Boersma et al.11 developed the ‘Platelet GP IIb/IIIa in Unstable angina: Receptor Suppression Using Integrilin Therapy’ (PURSUIT) risk score. Elements of the PURSUIT score include vital sign abnormalities, age, sex, Canadian Cardiovascular Society angina score, ST-segment depression, and heart failure on presentation. The PURSUIT risk score has the added advantage of distinguishing between patients presenting with unstable angina and MI and providing a different risk assessment for each. In this analysis, the group of patients with EMI had a median PURSUIT risk score of 13, correlating to almost a 20% 30-day rate of death or re-infarction. Similarly, the Thrombolysis In Myocardial Infarction (TIMI) risk score is a validated method for assigning risk of death, re-infarction, or urgent revascularization in patients with unstable angina and non– ST-segment elevation MI.12,13 Variables directly predictive of adverse outcomes include age, ST-segment changes, known coronary artery disease (CAD), risk factors for CAD, recent use of aspirin, and elevated cardiac biomarkers. The patients with EMI in this analysis had a median TIMI risk score of 4, correlating to a 14-day adverse event rate of 20% in patients from the TIMI 11b study12 and a 30-day rate of 42% in patients prospectively evaluated by Bartholomew et al.13 Finally, the ACC/AHA guidelines for management of patients with unstable angina and non– ST-segment elevation MI1 recommend using the criteria by Braunwald14 to risk stratify. Under this risk-stratification system, 77% of our EMI patients would have met ACC/AHA high-risk criteria based upon ECG findings alone.

Limitations Our data were collected as part of the SYNERGY trial, and, thus, the demographics of our groups closely resemble the inclusion criteria. The inclusion criteria for SYNERGY were designed to evaluate patients with high-risk unstable angina or non–ST-segment elevation MI. These strict criteria excluded many patients with EMI due to negative cardiac

biomarkers at the time of initial presentation. Our sample of EMI patients likely represents a unique subset of higherrisk patients—those who met the other two inclusion criteria of age60 or ST-segment changes at the time of enrolment. Additionally, patients with EMI had a longer elapsed time from presentation to enrolment compared with those presenting with MI. During that time, sicker patients could have been excluded, which could have affected the event rates. As such, the proportion and prognosis of patients with EMI in the general population cannot be ascertained from this study. In this clinical trial, high-risk patients received aggressive pharmacological care. However, the care patterns seen in this well-controlled clinical trial may not necessarily reflect those seen in real-world practice. In fact, in realworld practice, a disconnection between risk stratification and therapy for high-risk ACS patients exists.15 Data from the CRUSADE registry demonstrate that higher-risk patients—based on age, sex, chronic renal failure, diabetes, or risk score—do not necessarily receive more aggressive therapy.16,17 These patients are often excluded from clinical trials such as SYNERGY, and, as such, the results of this analysis cannot be used to measure compliance. Some patients enrolled in the SYNERGY trial could not be included in our analysis, predominantly due to missing biomarker data prior to enrolment. The recording of preenrolment troponin values was not mandatory and was routinely performed at some sites but not others. Although included and excluded patients seemed mostly similar with respect to demographics, the excluded group was much more likely to be enrolled based upon age and ECG criteria and have negative biomarkers at the time of enrolment. No differences were seen among these groups with respect to outcome events. It is possible that exclusion of these patients introduces selection bias. However, we feel this is less likely, and it more likely represents a characteristic of the site as opposed to a particular patient subgroup. However, the precise impact on our analysis is unknown. As a secondary analysis of a completed trial, it is not possible to adjust for all confounding variables. Although our clinical outcome differences are statistically significant, they should be prospectively validated and can only be hypothesis-generating at this point.

1084

Conclusions The results of this analysis demonstrate that patients with EMI are treated with fundamental and advanced therapies for ACS at a similar rate when compared with patients who are initially troponin-positive. These findings suggest that clinicians incorporate other elements of risk stratification when deciding on treatment regimens for troponinnegative patients. Patients with EMI demonstrated lower rates of death and re-infarction at 30 days compared with those who presented troponin-positive.

Acknowledgements The SYNERGY trial was funded by sanofi-aventis. The authors would like to thank Amanda Stebbins of the Duke Clinical Research Institute (DCRI) for her statistical assistance with this manuscript. They would also like to acknowledge the editorial assistance of Amanda McMillan of the DCRI. Conflict of interest: The authors have the following conflicts of interest to disclose. K.W.M., E.A., P.A., S.G.G., J.J.F., and R.M.C. were supported by Research Grants, sanofi-aventis. K.W.M., M.K., C.V.P., P.A., S.G.G., J.J.F., and J.W.H. were supported by Speaker’s Bureau, sanofi-aventis. K.W.M., M.K., C.V.P., P.A., S.G.G., J.J.F., and J.W.H. were supported by Consulting or Honoraria, sanofi-aventis.

References 1. Braunwald E, Antman EM, Beasley JW, Califf RM, Cheitlin MD, Hochman JS, Jones RH, Kereiakes D, Kupersmith J, Levin TN, Pepine CJ, Schaeffer JW, Smith EE III, Steward DE, Theroux P, Gibbons RJ, Alpert JS, Faxon DP, Fuster V, Gregoratos G, Hiratzka LF, Jacobs AK, Smith SC Jr ; American College of Cardiology; American Heart Association. Committee on the Management of Patients With Unstable Angina.. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction—summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366–1374. 2. The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. Eur Heart J 2000;21:1502–1513. 3. Antman EM, Tanasijevic MJ, Thompson B, Schactman M, McCabe CH, Cannon CP, Fischer GA, Fung AY, Thompson C, Wybenga D, Braunwald E. 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. 4. Green GB, Beaudreau RW, Chan DW, DeLong D, Kelley CA, Kelen GD. Use of troponin T and creatine kinase-MB subunit levels for risk stratification of emergency department patients with possible myocardial ischemia. Ann Emerg Med 1998;31:19–29.

C.D. Miller et al. 5. Boersma E, Harrington RA, Moliterno DJ, White H, Theroux P, Van de Werf F, de Torbal A, Armstrong PW, Wallentin LC, Wilcox RG, Simes J, Califf RM, Topol EJ, Simoons ML. Platelet glycoprotein IIb/IIIa inhibitors in acute coronary syndromes: a meta-analysis of all major randomised clinical trials. Lancet 2002;359:189–198. 6. 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. 7. Heeschen C, Hamm CW, Goldmann B, Deu A, Langenbrink L, White HD. Troponin concentrations for stratification of patients with acute coronary syndromes in relation to therapeutic efficacy of tirofiban. Lancet 1999; 354:1757–1762. 8. SYNERGY Trial Investigators. Enoxaparin vs. unfractionated heparin in high-risk patients with non–ST-segment elevation acute coronary syndromes managed with an intended early invasive strategy: primary results of the SYNERGY randomized trial. JAMA 2004;292: 45–54. 9. The SYNERGY Executive Committee trial: study design and rationale. Am Heart J 2002;143:952–960. 10. Roe MT, Peterson ED, Pollack CV Jr, Newby LK, Li Y, Christenson RH, Peacock WF, Fesmire FM, Diercks D, Kirk JD, Smith SC Jr, Ohman EM, Gibler WB ; CRUSADE Investigators. Influence of timing of troponin elevation on clinical outcomes and use of evidence-based therapies for patients with non–ST-segment elevation acute coronary syndromes. Ann Emerg Med 2005;45:355–362. 11. Boersma E, Pieper KS, Steyerberg EW, Wilcox RG, Chang WC, Lee KL, Akkerhuis KM, Harrington RA, Deckers JW, Armstrong PW, Lincoff AM, Califf RM, Topol EJ, Simoons ML. Predictors of outcome in patients with acute coronary syndromes without persistent ST-segment elevation: results from an international trial of 9461 patients. Circulation 2000; 101:2557–2567. 12. Antman EM, Cohen M, Bernink PJ, McCabe CH, Horacek T, Papuchis G, Mautner B, Corbalan R, Radley D, Braunwald E. The TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284: 835–842. 13. Bartholomew BA, Sheps DS, Monroe S, McGorray S, Smith K, Pepine CJ. A population-based evaluation of the thrombolysis in myocardial infarction risk score for unstable angina and non-ST elevation myocardial infarction. Clin Cardiol 2004;27:74–78. 14. Braunwald E, Jones RH, Mark DB, Brown J, Brown L, Cheitlin MD, Concannon CA, Cowan M, Edwards C, Fuster V, Goldman L, Green LA, Grines CL, Lytle BW, McCauley KM, Mushlin AI, Rose GC, Smith EE III, Swain JA, Topol EJ, Willerson JT. Diagnosing and managing unstable angina. Agency for Health Care Policy and Research. Circulation 1994; 90:613–622. 15. Giugliano RP, Braunwald E. The year in non-ST-segment elevation acute coronary syndromes. J Am Coll Cardiol 2005;46:906–919. 16. Hoekstra JW, Roe MT, Peterson ED, Menon V, Mulgund J, Pollack CV, Miller C, Palabrica T, Harrington RA, Ohman EM, Gibler WB. Early glycoprotein IIb/IIIa inhibitor use for non–ST-segment elevation acute coronary syndrome: patient selection and associated treatment patterns. Acad Emerg Med 2005;12:431–438. 17. Ohman EM, Roe MT, Smith SC Jr, Brindis RG, Christenson RH, Harrington RA, Gibler WB, Peterson ED; CRUSADE Investigators. Care of non-ST-segment elevation patients: insights from the CRUSADE national quality improvement initiative. Am Heart J 2004; 148(Suppl. 5): S34–S39.