Author + information
- ↵*Reprint requests and correspondence: Dr. Francine K. Welty, Division of Cardiology, Beth Israeal–Deaconess Medical Center, One Autumn Street, 5th Floor, Boston, Massachusetts 02215
In the prethrombolytic era, in-hospital mortality of patients with non–Q wave myocardial infarction (MI) was only one-half that of patients with Q wave MI (see discussion in reference 1). Despite this more favorable early outcome, patients with non–Q wave MI have a higher incidence of recurrent myocardial ischemia and reinfarction. Consequently, long-term rates of reinfarction and mortality after non–Q wave MI have been similar to or worse than those after Q wave MI (1). In 1987, the American College of Cardiology (ACC) and the American Heart Association (AHA) recommended that all patients with non–Q wave MI undergo coronary angiography, although no randomized trial had tested the approach (2). The rationale for this recommendation was that patients with non–ST segment elevation MI were thought to have a high grade residual stenosis in a vessel supplying viable myocardium, which may result in continued risk for recurrent ischemic events. Early coronary angiography would demonstrate whether critical stenoses are present; however, it does not provide functional information as to the hemodynamic significance of any residual stenosis. Thus, in the absence of clinical ischemic symptoms, it is reasonable to question whether routine angiography alone will provide information essential in determining the benefit of revascularization with respect to reinfarction and death.
In 1994, the Thrombolysis in Myocardial Infarction (TIMI) IIIB trial results suggested that routine catheterization in non–Q wave MI may not improve outcome. In TIMI-IIIB, 1,473 patients with unstable angina or non–Q wave MI were randomized to receive either fibrinolysis or placebo and were then separately randomized to undergo either early angiography and revascularization within 18 to 48 h after randomization or conservative therapy, with angiography being performed only in the presence of demonstrable ischemia (3). There was no difference in end point events (death, MI or refractory angina); therefore, in 1996, the ACC/AHA guidelines no longer advocated coronary angiography in all patients presenting with non–Q wave MI (4). However, limitations of TIMI-IIIB include a relatively late time to revascularization (mean 36 h after presentation). In addition, most of the TIMI-IIIB patients had unstable angina for which the risk of death is less than that for acute MI. Finally, this study was performed before utilization of intracoronary stents as well as glycoprotein (GP) IIb/IIIa inhibitors. Further, cost analysis of TIMI-IIIB has shown that patients in the conservative group had more readmissions to the hospital and were treated with more medications; therefore, an early invasive strategy may actually be more cost-effective.
Since the ACC/AHA guidelines were revised in 1996, two additional randomized trials have reported their results: one favored a noninvasive strategy and one favored an invasive strategy. In the Veterans Affairs Non-Q-wave Infarction Strategies in Hospital (VANQWISH) trial, patients with non–Q wave MI who were randomized to a routine invasive approach, including catheterization and revascularization, had a higher rate of nonfatal MI or death than those who were randomized to a more conservative strategy (with intervention for patients with ischemia before discharge) at hospital discharge (36 vs. 15 events, p = 0.004), at one month after discharge (48 vs. 26 events, p = 0.012) and at one year after discharge (111 vs. 85 events, p = 0.05) (5). A major reason for this was a higher mortality rate of 11.6% for patients undergoing coronary artery bypass graft surgery in the early invasive group, as compared with 3.4% in the conservative group. In addition, catheterizations in the invasive strategy group were performed electively after hospital admission, with a mean time from randomization to revascularization of 8 days, which may be too late to salvage threatened myocardium, and may have contributed to the lack of a beneficial effect with revascularization in VANQWISH. In contrast to the short-term survival, there was no difference in long-term survival at the end of 1,000 days; therefore, initial conservative management did not improve long-termsurvival.
The first randomized trial to show an advantage for the invasive strategy was the Fast Revascularization during InStability in Coronary artery disease (FRISC II) trial. This was a multicenter, randomized trial conducted in Scandinavia in which 2,457 patients with unstable angina or non–Q wave MI were treated with aspirin, beta-blockers, nitrates and low molecular weight heparin (Fragmin) and then randomized within 48 h to an early invasive strategy with revascularization within seven days or a noninvasive evaluation with revascularization if clinically indicated (6). Those in the early revascularization group had a 21% reduction in the primary end point—death and MI—as compared with those in the conservative group at six months (9.4% vs. 12.1%, risk ratio 0.78, 95% confidence interval 0.62 to 0.98, p = 0.031). Male subjects had a significant reduction in mortality alone (1.5% vs. 3.2%, p = 0.03). Symptoms of angina and readmission to the hospital were 50% lower with the invasive strategy.
The current study (7)attempts, in a retrospective fashion, to analyze the benefits of very early routine catheterization and revascularization (within 6 h), versus a conservative approach, in patients with non–ST segment elevation MI in the Myocardial Infarction Triage and Intervention (MITI) registry. The stated objective is to compare short- and long-term outcome after early invasive or conservative strategies in the treatment of non–ST segment elevation acute MI; however, the authors actually compare outcomes in patients treated at hospitals that favor an early invasive approach with those in patients treated at hospitals that do not favor an early invasive approach (7). The early invasive approach was defined as performing catheterization within 6 h of hospital admission in at least 25% of patients with chest pain and nondiagnostic electrocardiographic (ECG) changes. There were two hospitals that used the early invasive approach (n = 308) and 17 that favored the conservative approach (n = 1,327). Of patients admitted to conservative hospitals, 52% underwent catheterization during admission, but only 8% of these during the first 6 h, as compared with 59% of those in hospitals using the invasive strategy. Univariate analyses showed that both 30-day and four-year mortality were significantly lower in patients treated at the hospitals using the early invasive strategy. After multivariate analysis, 30-day mortality was no longer significantly different; however, long-term mortality remained significantly lower. The authors’ conclusion is that an early invasive strategy in patients with acute non–ST segment elevation MI and nondiagnostic ECG changes is associated with lower long-term mortality. However, because the study compared outcomes based on admission to an invasive versus conservative hospital—not on whether catheterization or revascularization was actually done—the true conclusion is that admission to a hospital where physicians favored an early invasive approach was associated with lower long-term mortality.
In any observational, retrospective study, there are always a number of confounding variables that could account for any observed differences. In the current study, those patients admitted to hospitals using a conservative approach had a lower socioeconomic status, were older, were less predominantly Caucasian and had a higher comorbidity, which included an 89% higher rate of previous congestive heart failure (CHF), a 33% higher rate of previous angina, a 50% higher rate of previous MI, a 46% higher rate of new CHF, a 31% higher rate of diabetes and an 18.5% higher rate of current smoking. All of these factors are associated with a worse short- and long-term outcome after MI. Factors associated with higher hospital mortality in the current study included increased age and new-onset heart failure, both of which were more prevalent in the conservative hospitals. After adjustment for these factors, 30-day mortality was no longer significantly different between the two types of hospitals. Thus, differences in baseline characteristics may account for the better short-term outcome in the hospitals favoring an invasive approach rather than the use of early catheterization.
In a subgroup analysis of only those patients admitted to the hospitals favoring an early invasive strategy, patients undergoing early angiography had significantly lower in-hospital mortality than those who did not. This was the only analysis that compared an invasive and conservative strategy. After adjustment, there was a trend that was no longer significantly different. Although the authors do not show us the differences in baseline characteristics between these two groups, the lack of association after adjustment suggests that differences in baseline characteristics were more important contributors to differences in outcome, rather than the use of early catheterization. The authors comment that the low in-hospital mortality observed in this subgroup analysis implies that myocardial salvage may play an important role. However, the lack of mortality difference after adjustment does not support this. In addition, it should be noted that these patients underwent early catheterization but not necessarily early revascularization. Because there is no apparent reason why early angiography alone will salvage myocardium and thus improve outcome, the most important subgroup analysis would be a comparison of those who underwent early angioplasty (45% underwent angioplasty within 6 h) as compared with those who did not of patients admitted to hospitals using the invasive approach. These data are not presented.
In their analyses of long-term mortality, the authors observed that older age, blue-collar employment, CHF during their hospital stay, previous CHF and previous MI were all associated with increased long-term mortality. All of these factors were more prevalent in those patients admitted to the more conservative hospitals; however, after adjustment, those admitted to hospitals using the early invasive approach still had significantly better long-term survival. A limitation of the current study is that adjustment was made only for differences in characteristics that were significantly different. Patients admitted to the conservative hospitals had higher rates of diabetes and cigarette smoking and were more likely to be female and nonwhite. All of these are associated with a worse outcome, and failure to adjust for them may bias the results toward a better outcome in patients admitted to hospitals using an invasive strategy.
Despite control for measured factors, unmeasured factors may still contribute to confounding and may affect the results in an observational study, as the authors acknowledge. Other factors include differences in the choice of medical treatments for CHF—for example, angiotensin-converting enzyme inhibitors—and differences in the use of hydroxymethyl glutaryl coenzyme A (HMG CoA) reductase inhibitor (statin) drugs between the two hospitals. Because statin drugs have been shown to improve survival after MI (8), differential use may have contributed to outcome differences. Residual confounding may also be due to differences in institutional characteristics and processes of care. For example, do invasive hospitals treat a larger volume of patients with MI? Were patients admitted to hospitals using an invasive approach more likely to have a cardiologist taking care of them than patients at the conservative hospitals? Did invasive hospitals have housestaff, whereas conservative hospitals did not? Did conservative hospitals with on-site catheterization laboratories offer percutaneous coronary angioplasty and coronary artery bypass graft surgery? Was revascularization more complete in those admitted to hospitals favoring an invasive approach?
Although the current study is limited by its observational, retrospective design, the authors raise the question of whether there may be benefit with regard to myocardial salvage, in addition to any benefit in prevention of reinfarction and death, with very early catheterization and revascularization in patients with non–ST segment elevation MI. Traditional thought has been that if the infarct-related artery is patent in non–ST segment elevation MI, benefit from revascularization is due to prevention of recurrent MI in the setting of a critical residual stenosis. However, it may be that early revascularization resulting in a widely patent infarct-related artery may also favorably affect ventricular remodeling after non–Q wave MI and thus have a beneficial effect on outcome.
The current study, TIMI-IIIB, VANQWISH and FRISC-II were all performed in the prestent and pre–GP IIb/IIIa inhibitor era. Since then, the use of GP IIb/IIIa inhibitors has been shown to have an incremental benefit beyond that of heparin and aspirin in reducing adverse outcomes in patients with non–ST segment elevation MI. In the Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrelin Therapy (PURSUIT) trial, 10,948 patients with unstable angina or non–Q wave MI were randomized to receive eptifibatide or placebo in addition to heparin and aspirin within 24 h of presentation. Those randomized to eptifibatide had a 10% reduction in the combined incidence of the primary end point of death or MI at 30 days (14.2% vs. 15.7% in the placebo group, p = 0.042) (9). In the Platelet Receptor inhibition for Ischemic Syndrome Management in Patients Limited by Unstable Signs and Symptoms (PRISM-PLUS) study, 1,915 patients with unstable angina or non–Q wave MI were randomized to tirofiban plus heparin or heparin alone. All patients received aspirin if not contraindicated. Coronary angiography and angioplasty were performed when indicated after 48 h; 75% underwent revascularization. At seven days, those receiving tirofiban and heparin had a 28% reduction in death, MI or revascularization (12.9% vs. 17.9%) as compared with those receiving heparin alone (10). The Evaluation of IIb/IIIa Platelet Inhibitor for Stenting (EPISTENT) trial evaluated the potential synergy between coronary stents and GP IIb/IIIa inhibition and observed that the addition of abciximab improved outcomes as compared with stenting without abciximab in Q wave MI and both large and small non–Q wave MI (11). Therefore, the use of both stents and GP IIb/IIIa inhibitors in a randomized trial is important.
Treat Angina with Aggrastat and determine Cost of Therapy with an Invasive or Conservative Strategy (TACTICS, or Thrombolysis in Myocardial Infarction trial [TIMI]-18) is an ongoing trial investigating the incremental benefit of an early invasive approach using stents versus a conservative approach beyond that of aspirin, heparin and GP IIb/IIIa inhibitors in patients with unstable angina or non–Q wave MI (12). In this trial, 1,720 patients will receive aspirin, heparin and tirofiban and then be randomized to either catheterization with revascularization (if indicated) within 48 h or a conservative approach. The primary end point will be the combined incidence of death, MI and readmission to the hospital. Even this study design may not be optimal for three reasons.
First, it includes both unstable angina and non–Q wave MI. As noted earlier, the death rate is lower in patients with unstable angina, and this may affect the outcome.
Second, the 48-h time frame may be too long to maximally salvage myocardium. Results from thrombolytic trials have shown that mortality reduction is greatest when reperfusion is achieved within 6 h of onset and, at the most, 12 h; therefore, one would predict greater benefit with achievement of a widely patent infarct-related artery within 6 h rather than within 48 h. The diagnosis of non–ST segment elevation MI requires a positive creatine kinase, which may not be available for at least 12 h (especially if the first creatine kinase measurement is negative); therefore, a study of confirmed non–ST segment elevation MI probably could not be done within a 6-h time frame from presentation; however, a 12-h time frame may be feasible.
Third, the study design did not incorporate the uniform utilization of lipid-lowering agents, but rather recommended additional medical therapy to include cholesterol-lowering agents along with beta-blockers, nitrates and calcium antagonists (12). In the Cholesterol and Recurrent Events (CARE) trial, 4,159 patients with a history of MI and a mean low density lipoprotein cholesterol level of 139 mg/dl were randomized to pravastatin, 40 mg/day, or placebo. At a mean follow-up of five years, those receiving pravastatin had a 24% reduction in fatal coronary events or nonfatal MI, a 23% reduction in coronary angioplasty, a 26% reduction in coronary artery bypass graft surgery and a 31% reduction in stroke (8). These reductions are more significant than those seen in some intervention trials; thus, the ideal trial would also include the use of lipid-lowering agents to reduce low density lipoprotein cholesterol to comparable levels in both groups. Beneficial effects with statin drugs may include mechanisms other than lipid lowering. Aggressive lipid lowering has been shown to decrease ischemia, presumably by improving endothelial function (13). Because impaired endothelial function may cause vasoconstriction and thus contribute to ischemia in the setting of a ruptured plaque in non–ST segment elevation MI, one would predict that aggressive, early lipid lowering would ameliorate vasoconstriction and thus improve outcome.
The Antiplatelet and Statin Therapy in Acute Coronary Syndromes trial (Aggrastat to Zocor, or A to Z trial) is designed to determine whether the addition of low molecular weight heparin provides incremental benefit beyond tirofiban and aspirin in patients with unstable angina and non–Q wave MI and whether immediate treatment with an HMG CoA reductase inhibitor improves outcome, as compared with a four-month trial of diet and placebo before initiation of an HMG Co-A reductase inhibitor.
This is a rapidly evolving field, and those designing new post-MI trials should also consider the results of several recent randomized dietary trials that have shown reductions in totalmortality in post-MI patients. In the Lyon Diet Heart Study, 605 patients with a first MI were randomized to either a Mediterranean diet containing 31% fat and 8% saturated fat and that was high in alpha-linolenic acid (an omega-3 polyunsaturated fatty acid) or the National Cholesterol Education Program step I diet. At four-year follow-up, total mortality was reduced 56% (odds ratio [OR] 0.44, 95% confidence interval [CI] 0.21 to 0.94, p = 0.03), and cardiac death and nonfatal MI were reduced 72% (OR 0.28, 95% CI 0.15 to 0.53, p = 0.0001) (14). These changes occurred without significant changes in lipid levels. Postulated beneficial effects include stabilization of membranes, prevention of lethal cardiac arrhythmias and platelet inhibition. Omega-3 fatty acids (eicosapentanoic acid) from fish have also been shown to be important in prevention of mortality after MI. In the Death and Myocardial Infarction: Diet and Reinfarction Trial (DART), those men who had an MI and were randomized to two to three portions of fatty fish intake per week had a significant (29%) reduction in two-year all-cause mortality compared with those randomized to the other two groups with a reduction in fat intake—one with an increase in the ratio of polyunsaturated to saturated fat intake and the other with an increase in cereal fiber intake (15). In the Gruppo Italiano per lo Studio della Sopravvivenza nell’ Infarto miocardico (GISSI) Prevenzione trial, post-MI patients randomized to 1 g of omega-3 fish oil capsules daily had a 15% reduction in the combined end point of total death, nonfatal MI and nonfatal stroke (p = 0.023); a 20% reduction in cardiovascular death, nonfatal MI and nonfatal stroke (p = 0.008); and a 45% reduction in sudden death (OR 0.55, 95% CI 0.40 to 0.76) (16). On the basis of these results, a National Institutes of Health–funded randomized trial—the Fatty Acid Arrhythmia Trial—is currently examining the effect of omega-3 polyunsaturated fatty acids in the form of fish oil capsules on arrhythmias in patients with implanted cardioverter-defibrillators. In addition, the Nutrition Committee of the AHA is issuing new guidelines recommending the ingestion of at least two fatty fish meals per week or daily fish oil capsules. Until a randomized trial that also incorporates lipid-lowering drugs and omega-3 fatty acids is performed, we may not have the complete answer as to what is the optimal management strategy for non–ST segment elevation MI.
↵∗ Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology
- Welty F.K.,
- Mittleman M.A.,
- Lewis S.M.,
- Healy R.W.,
- Shubrooks S.J. Jr.,
- Muller J.E.
- ↵Guidelines for coronary angiography: a report of the American College of Cardiology/American Heart Association Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures (Subcommittee on Coronary Angiography). J Am Coll Cardiol 1987;10:935–50.
- TIMI IIIB Investigators
- Ryan T.J.,
- Anderson J.L.,
- Antman E.M.,
- et al.
- MITI Investigators,
- Scull G.S.,
- Martin J.S.,
- Weaver W.D.,
- Every N.R.
- Cannon C.P.,
- Weintraub W.S.,
- Demopoulos L.A.,
- et al.
- Andrews T.C.,
- Raby K.,
- Barry J.,
- et al.
- de Lorgeril M.,
- Salen P.,
- Martin J.-L.,
- Monjaud I.,
- Delaye J.,
- Mamelle N.