Rescue Percutaneous Coronary Intervention: A Review

The primary goal of therapy for acute myocardial infarction (AMI) is rapid, complete and sustained restoration of infarct-related artery (IRA) blood flow and myocardial perfusion, with a consequent positive impact on the patient’s outcome. Both fibrinolytic and mechanical restoration of anterograde coronary blood flow in patients suffering AMI have shown to improve left ventricular function, reduce infarct size and reduce mortality.^1–10 The benefits of myocardial reperfusion, including the prevention of infarct expansion and improvement of electrical stability, are amplified when vessel patency can be achieved quickly after the onset of symptoms, particularly in the first two hours. Current thrombolytic regimens appear to have reached a ceiling of efficacy, with normal coronary blood flow being restored at 90 minutes after therapy in only 50–60% of patients.¹¹ However, full myocardial perfusion occurs in only about 30–45% of patients at best. In about one-third of all cases, reocclusion and/or reinfarction are seen within 3 months after the index event. Although the newer fibrinolytic agents such as TNK, t-PA12 and n-PA¹³ have theoretical advantages over previous agents, clinical outcomes with these agents have not proven to be superior compared with the front-loaded regimen of t-PA. While primary angioplasty may lead to the restoration of flow of the infarct-related vessel in about 85–95% of patients,^14–18 the use of this technique is associated with a delay of at least 30–60 minutes. In addition, many hospitals do not have facilities or infrastructure to administer this form of therapy in an effective and expeditious manner. For these reasons, primary angioplasty is currently not a practical approach for a large proportion of patients with AMI. A strategy of initial pharmacologic fibrinolysis followed by immediate transfer to a facility with a catheterization laboratory for rescue angioplasty in the event of thrombolytic failure has now become common practice. This monograph will focus on the current status of rescue percutaneous coronary intervention (PCI), the rationale and clinical experience of GP IIb/IIIa antagonists as adjunctive therapy for PCI in AMI. It will also discuss the safety issues regarding the administration of these agents in patients who have received full-dose thombolytic therapy, and finally, will review the available data on the use of GP IIb/IIIa antagonists in the setting of rescue PCI. Rescue Percutaneous Coronary Intervention Initial observational data suggested that the rate of procedural success of rescue angioplasty was lower and associated with a higher incidence of complications compared with primary percutaneous transluminal coronary angioplasty (PTCA).¹⁹ Until recently, two small randomized trials had provided data suggesting a mortality benefit of rescue angioplasty in patients with totally occluded epicardial vessels (Figure 1).^20–22 In the randomized trial of rescue angioplasty versus a conservative approach for failed fibrinolysis in ST-segment elevation myocardial infarction: the Middlesbrough Early Revascularization to Limit INfarction (MERLIN) Trial,²³ a total of 307 patients with STEMI and failed fibrinolysis were randomized to emergency coronary angiography with or without rescue PCI, or conservative treatment. Thirty-day all-cause mortality was similar in the rescue and conservative groups [9.8% versus 11%; p = 0.7, risk difference (RD) = 1.2%, 95% confidence interval (CI) = 5.8–8.3]. The composite secondary end point of death/re-infarction/stroke/subsequent revascularization/heart failure occurred less frequently in the rescue group (37.3% versus 50%; p = 0.02, RD 12.7%, 95% CI 1.6–23.5), driven by less subsequent revascularization (6.5% versus 20.1%; p p = 0.3, RD 3.2%, 95% CI = 3.3–9.9; and 24.2% versus 29.2%; p = 0.3, RD 5.7%, 95% CI = 4.3–15.6, respectively). Stroke and need for transfusion were more common in the rescue group (4.6% versus 0.6%; p = 0.03, RD 3.9%, 95% CI = 0.5–8.6, and 11.1% versus 1.3%; p p 24 all consecutive patients who underwent PCI for myocardial infarction (MI) within n = 840) or a primary (n = 8,531) procedure. Rescue patients were significantly younger males with anterior wall infarctions associated with LV dysfunction, but had less multivessel disease compared with those treated with primary intervention. Coronary stents were implanted at similar rates (56.9% versus 54.9%; p = 0.283). Procedural success rates were lower for rescue cases (88.1% versus 91.2%; p p = 0.034), compared with the primary intervention group; target vessel revascularization (TVR) (less than or equal to 0.5%), emergency bypass surgery (CABG) (less than or equal to 0.3%) and reinfarction (less than or equal to 2.6%) rates were similar for both strategies. Multivariate analysis identified the rescue procedure as a predictor of inhospital death [OR (95% CI) = 1.60 (1.17–2.19); p = 0.003]. The authors concluded that patients who underwent a rescue coronary intervention had higher inhospital death rates compared with those who underwent a primary coronary intervention. The above studies suggest a modest benefit of rescue PCI over medical therapy in the setting of AMI. However, the applicability of the results of the aforementioned studies is limited by major advances in interventional techniques and adjunctive pharmacotherapy [e.g., strict and lower activated clotting time-titrated heparin dosing, advent of direct thrombin inhibitors and drug-eluting stents, newer ADP receptor antagonists, intravenous glycoprotein (GP) IIb/IIIa antagonists] since the time of their performance. Rationale for the use of GP IIb/IIIa antagonists in AMI. Platelets play a pivotal role in the pathogenesis of acute coronary syndromes, including MI.^25,26 The rupture or erosion of a lipid-rich atherosclerotic lesion exposes the highly thrombogenic core contained within the plaque and its subendothelial components to arterial blood flow. The ensuing platelet adhesion and activation lead to changes in platelet shape, degranulation and aggregation. During this process, platelets release several proaggregatory and prothrombotic mediators. In addition, several potent antifibrinolytic components such as plasminogen activator inhibitor-1 (PAI-1) and alpha-2-antiplasmin are also released. Conformational changes in the GP IIb/IIIa receptor occur during the activation process, making this integrin receptive to ligand binding. Platelet aggregation at the rupture site leads to the growth of occlusive thrombi that can cause myocardial ischemia, necrosis and subsequent complications.²⁷ The generation of thrombin and activation of platelets at the site of vascular injury frequently limits thrombolysis and may, in fact, be enhanced as a consequence of this form of therapy. The paradoxical state that may ensue creates a resistance to thrombolytic therapy and predisposes the patient to “failed thrombolysis”, in which reperfusion of the ischemic myocardium supplied by the affected artery is not attained, with a consequent negative impact on the clinical outcome of the patient.²⁸ Consistent with the critical role platelets play in thrombosis, abciximab has been shown to reduce thrombus mass, to limit thrombin generation in vitro²⁹ and in vivo,³⁰ and to block the binding of activated factor XIII to platelets, thereby decreasing the stability of fibrin crosslinks.³¹ By abrogating the thrombophilic contribution of platelets during thrombosis, endogenous fibrinolytic activity may be sufficient to lyse the clot when it is no longer antagonized by the aforementioned antifibrinolytic properties conferred by platelets.^35,36 In addition to reducing the bulk of the thrombus at the site of vascular injury in the large epicardial vessel, GP IIb/IIIa antagonists may protect the microvasculature supplied by the affected artery. The effect of GP IIb/IIIa blockade in enhancing microvascular perfusion and preventing microembolization of thrombus debris or platelet aggregates probably has a profound impact on the clinical outcome of patients, and may partly explain the remarkable impact these drugs have in reducing ischemic complications. GP IIb/IIIa antagonists as adjuncts to PCI in AMI. Angioplasty is a powerful stimulus for platelet activation.³⁸ Platelet deposition has been shown to occur within minutes after balloon deflation in the treated segment of the vessel, with detachment and subsequent embolization of thrombus fragments.³⁹ Platelet activation is particularly relevant when angioplasty is performed in patients with AMI. After an initial transient deactivation of circulating platelets, the level of platelet activation increases in patients being treated with primary angioplasty.⁴⁰ An intravascular ultrasound study showed that smaller luminal diameters, greater plaque area and especially the presence of a disrupted plaque and thrombus, were predictors of abrupt vessel closure after angiographically successful primary angioplasty.⁴¹ In patients receiving primary angioplasty, the use of platelet GP IIb/IIIa antagonists is initially aimed at the immediate reduction in the risk of acute thrombotic complications. Early recurrent ischemia after primary angioplasty is related mainly to thrombus formation, intimal dissection or a combination of the two in the dilated lesion. Platelet GP IIb/IIIa inhibitors may reduce subacute or threatened occlusion by blocking platelet aggregation and by enhancing endogenous fibrinolysis. There is now a considerable amount of evidence supporting the use of GP IIb/IIIa antagonists during primary PCI for the treatment of AMI (Figures 2 and 3). Initial data in this setting are primarily derived from small subgroup analysis of patients suffering an AMI who were enrolled in the Evaluation of 7E3 for the Prevention of Ischemic Complications (EPIC)⁴² and Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis (RESTORE) trials.⁴³ Both studies pointed to a benefit of GP IIb/IIIa antagonists as compared to placebo. Since then, several clinical trials specifically addressing this issue have been performed. The ReoPro in AMI Primary PTCA Organization and Randomized Trial (RAPPORT)⁴⁴ was a multicenter, placebo-controlled, double blind trial in which 483 patients with AMI presenting within 12 hours of symptom onset and eligible for angioplasty were randomly assigned to receive either abciximab or placebo. The use of stents was strongly discouraged. Although the primary endpoint of death, re-infarction or repeat TVR at 6 months did not reach a statistical difference between both groups, at 30 days this same composite endpoint was reduced from 12% in the placebo group to 4.6% in the abciximab group (p = 0.005). Major bleeding and transfusions were significantly more frequent in the abciximab group. Nevertheless, it is important to note that intracranial bleeding did not occur in either the abciximab or the placebo groups. In the Intracoronary Stenting and Antithrombotic Regimen-2 (ISAR-2) Trial, 401 patients undergoing stenting within 48 hours after the onset of AMI were randomly assigned to receive abciximab with reduced-dose heparin or a standard dose of heparin alone. Although no differences were noted in the primary outcome of angiographic restenosis between both groups, the secondary outcome, which included the composite clinical endpoint of death, re-infarction and TVR, was reached in 5% of the abciximab group versus 10.5% of the control group (p = 0.038). In the Abciximab before Direct angioplasty and stenting in Myocardial Infarction Regarding Acute and Long-term follow-up (ADMIRAL) Trial, 300 patients suffering an AMI underwent primary coronary angioplasty and stenting and were randomized to receive either abciximab or placebo in addition to weight-adjusted heparin. The composite outcome of death, re-infarction and urgent TVR at 30 days occurred in 7.3% of patients assigned to abciximab versus 14.7% of patients assigned to receive placebo (p 45 The 30-day composite outcome of death, re-infarction or urgent TVR did not reveal a significant benefit of abciximab in patients undergoing stenting (5.1% versus 5.7% with and without abciximab, respectively), quite the contrary to what had been expected. Nevertheless, when outcomes more reliably reflecting thrombotic or ischemic complications were analyzed separately, abciximab was shown to confer a significant benefit to patients. For subacute thrombosis at 30 days, abciximab reduced the rate of events from 1.7% to 0.6% among patients undergoing primary angioplasty (p = 0.07), and from 1% to 0% among patients undergoing stenting (p = 0.03). Abciximab also reduced the rate of ischemic TVR at 30 days both, in patients undergoing primary PTCA (3.2% versus 5.4%; p p = 0.04). The benefit of abciximab in reducing the rate of ischemic TVR was maintained at 6 months. A possible explanation as to why this trial did not show a significant reduction in the risk of reaching the triple composite which had also been used as an endpoint in the ISAR-2 and ADMIRAL trials may be that the CADILLAC population was at a lower risk for ischemic complications compared to patients enrolled in the previous trials. Furthermore, the overall event rate in CADILLAC was low, making the ascertainment of some of the endpoints used in the other trials more difficult. Schomig and colleagues reported the results of the 140-patient Stent versus Thrombolysis for Occluded Coronary Arteries in Patients with Acute Myocardial Infarction (Myocardial) study.⁴⁶ Patients presenting within 12 hours after the onset of symptoms were randomly assigned to receive either stenting plus abciximab or intravenous rt-PA. The primary endpoint was the degree of myocardial salvage as assessed by scintigraphic study with technetium Tc 99m sestamibi. Despite having similar initial perfusion defects, the final size of the infarction was significantly smaller among patients who received stenting plus abciximab, as compared to patients who received rt-PA (14.3% versus 19.4% of the left ventricle; p = 0.02). At 6 months, the cumulative incidence of death, reinfarction or stroke was 8.5% in the stent plus abciximab group, and 23.2% in the rt-PA group. There are data suggesting that GP IIb/IIIa antagonists, specifically abciximab, not only exert their beneficial effects in patients with AMI by inhibiting platelet aggregation and rethrombosis, but also by disaggregating platelet-rich thrombi that may still be lodged in the epicardial vessel or which may have embolized to the distal vascular beds, thus impeding microvascular perfusion in spite of apparently patent coronary arteries. In both the SPEED⁴⁷ and TIMI-14⁴⁸ studies, approximately 30% of patients who had been randomly assigned to receive monotherapy with abciximab as the sole reperfusion therapy reached TIMI 3 flow at 60–90 minutes, a figure strikingly similar to that seen with streptokinase in the earlier fibrinolytic trials.⁶ In a complimentary fashion, Neumann and colleagues reported the results of their elegant study, the Munich Trial, in which peak coronary flow velocities, LV wall motion and ejection fraction were measured in 200 patients presenting with AMI who underwent revascularization within 48 hours after the onset of symptoms, and who were randomized to receive either abciximab with low-dose heparin, or no abciximab with standard-dose heparin.⁴⁹ It was observed that although most patients had optimal blood flow of the epicardial artery as assessed by the TIMI flow score, those assigned to receive abciximab had better microvascular perfusion as evidenced by higher scores of the mentioned parameters. Safety of GP IIb/IIIa inhibition after full-dose thrombolytic therapy. Data regarding the safety of administering GP IIb/IIIa antagonists to patients with AMI in whom thrombolytic therapy has already been initiated primarily derive from three trials. In the Thrombolysis and Angioplasty in Myocardial Infarction-8 (TAMI-8) study, 60 patients received incremental doses of m7E3, a precursor of abciximab, up to 15 hours after receiving t-PA (infused over 3 hours).⁵⁰ Ten patients treated only with rt-PA served as controls. All patients received aspirin and heparin. A consistent m7E3 dose-dependent decrease in platelet aggregation was observed and a relationship between receptor occupancy and extent of platelet inhibition was established. Major bleeding events were more common in the control group and were frequently associated with CABG. Although this study was not specifically designed to examine the angiographic efficacy of m7E3 combined with rt-PA, coronary artery patency was observed more frequently in the m7E3-treated patients. Additional data pointing to a reasonable safety profile of administering GP IIb/IIIa antagonists in patients treated with full-dose fibrinolytics derive from the IMPACT-AMI and PARADIGM trials. The Integrilin to Manage Platelet Aggregation to prevent Coronary Thrombosis (IMPACT-AMI) Trial represents the first time a GP IIb/IIIa antagonist was concomitantly administered with a fibrinolytic agent.⁵¹ One-hundred thirty-two patients receiving rt-PA for AMI were randomized to various doses of eptifibatide or to placebo. Due to the sample size, no differences in clinical outcomes could be noted. However, TIMI flow grade 3 at 90 minutes was more common in patients treated with the highest dose of eptifibatide compared with placebo (66% versus 39% in the placebo group; p = 0.006). No differences were observed in major bleeding complication rates, including intracranial hemorrhage (ICH). It is worth noting that the dosages of eptifibatide were later found to be suboptimal, achieving only 50–60% platelet inhibition. The Platelet Aggregation Receptor Antagonist Dose Investigation for reperfusion Gain in Myocardial infarction (PARADIGM) Trial was the third study to evaluate GP IIb/IIIa antagonists as adjuncts to full-dose fibrinolytic agents.⁵² Three-hundred fifty-two patients presenting within 12 hours of AMI symptom onset were randomized to receive escalating doses of lamifiban or placebo in addition to aspirin, heparin and SK or rt-PA. Lamifiban was associated with improved myocardial reperfusion as measured by early resolution of ST-segment elevation. However, no difference in the composite clinical outcomes was noted. Lamifiban was associated with a slightly higher rate of bleeding events compared with placebo. One of the main concerns when considering the administration of a GP IIb/IIIa antagonist to a patient who has received full-dose lytics is bleeding — in particular, ICH.53 The review of three randomized trials shows that GP IIb/IIIa inhibitors on their own do not increase the risk of ICH.⁵⁴ Furthermore, the available safety data on the combination of full-dose fibrinolytics and GP IIb/IIIa antagonists reveals a reasonable safety price. A point of caution, however, must be made regarding the safety of administering GP IIb/IIIa antagonists in patients who have received streptokinase. An unacceptably high rate of major bleeding has been clearly observed in patients randomly assigned to receive a GP IIb/IIIa antagonist combined with this fibrinolytic agent.^48,55 Therefore, it is unlikely that the combination of these two agents, even in the setting of failed thrombolysis, will have greater potential benefits compared to the associated risks. Although none of these trials specifically addressed safety issues of GP IIb/IIIa antagonist use during rescue PCI after full-dose fibrinolytics, they are applicable with regard to the safety of concomitant administration of these agents, with a significant number of patients proceeding to coronary angiography. GP IIb/IIIa inhibition during rescue PCI after full-dose fibrinolysis. Despite the remarkable benefits that GP IIb/IIIa antagonists have shown in the PCI arena, together with a strong biological rationale for their use in the setting of failed thrombolysis — all at an apparently reasonable safety price⁵⁶ — there are no large-scale prospective trials specifically answering the question of whether the use of GP IIb/IIIa antagonists during rescue angioplasty is preferable. However, there are four retrospective analyses which attempt to address this issue. In the EPIC Trial, 2,099 patients undergoing PCI randomly received chimeric 7E3 Fab (c7E3) as a bolus, a bolus plus a 12-hour infusion, or placebo. Sixty-four patients underwent PTCA for AMI; 42 patients underwent direct primary PTCA for AMI, and 22 patients underwent rescue angioplasty after failed thrombolysis. The primary composite endpoint comprised death, reinfarction, repeat intervention or CABG. Outcomes were assessed at 30 days and 6 months. Baseline characteristics were similar in direct and rescue PTCA patients. Pooling the two groups, the bolus and infusion of c7E3 reduced the primary composite endpoint by 83% (26.1% placebo versus 4.5% c7E3 bolus and infusion; p = 0.06). At 6 months, ischemic events were reduced from 47.8% with placebo to 4.5% with c7E3 bolus and infusion (p = 0.002), particularly reinfarction (p = 0.05) and repeat revascularization (p = 0.002). Major bleeding was increased with c7E3 (24% versus 13%; p = 0.28), although it was later recognized that the fixed and nonweight-adjusted dose of heparin used was inappropriate. It was concluded that adjunctive c7E3 therapy during direct and rescue PTCA reduced acute ischemic events and clinical restenosis in the EPIC Trial.⁵⁷ A retrospective analysis performed in 103 consecutive patients who presented with AMI and underwent PTCA with adjunctive abciximab therapy at the Albert Einstein Medical Center were divided into 3 groups: Group A, rescue PTCA within 15 hours of thrombolytic therapy (n = 22); Group B, fibrinolytic therapy followed by elective PTCA more than 15 hours after the initiation of fibrinolysis (n = 36); and Group C, primary PTCA without prior fibrinolysis (n = 45).58 All patients received abciximab as an adjunct to the interventional procedure at the standard dose. Patients who received fibrinolytics (Groups A and B) received alteplase in the recommend “front-loaded” regimen (bolus dose of 15 mg, 0.75 mg/kg body weight over a 30-minute period, not to exceed 50 mg, and 0.5 mg/kg up to 35 mg over the next 60 minutes). Stents were used significantly more frequently during rescue PTCA (64%) than during primary PTCA (31%; p = 0.02). The incidence of major, minor or insignificant bleeding was significantly higher in those who received alteplase and rescue PTCA (Group A) than in those who received alteplase and underwent elective PTCA (Group B; p = 0.006), or in those who had primary PTCA (Group C; p = 0.01). There were no significant differences in major bleeding complications between the alteplase and elective PTCA groups versus the primary PTCA group. In this trial, a significant increase in major bleeding complications was noted when abciximab was used in conjunction with rescue PTCA within 15 hours (mean 4.1 ± 3.3 hours) of failed thrombolytic therapy. It was suggested that the use of higher heparin doses may have led to an increased bleeding risk. In a multicenter retrospective cohort of 147 consecutive patients who underwent PTCA within 48 hours after full-dose thrombolysis for AMI, Jong and colleagues tried to examine the risk and predictors of bleeding complications.⁵⁹ Bleeding events (major, minor, nuisance) from the onset of AMI to discharge were compared between those who received abciximab (n = 57) and those who did not (n = 90). Baseline clinical characteristics were similar between the 2 groups. Despite lower doses of procedural heparin, the incidence of non-CABG-related major and minor bleeding was higher in the abciximab group than in the controls (63% versus 39%; p = 0.004). Although the risk of major bleeding was four-fold with abciximab (12% versus 3%; p = 0.04), only 1 intracranial and 1 fatal bleeding event occurred. Multivariable regression identified abciximab therapy as the most powerful independent predictor of combined major and minor bleeding, with a hazard risk ratio of 1.9 (p = 0.04). The authors concluded that in the setting of rescue or urgent PTCA within 48 hours after full-dose thrombolytic therapy after AMI, major, and particularly minor, bleeding were frequently encountered. To date, the largest subgroup analysis yielding data on the efficacy of rescue angioplasty with GP IIb/IIIa antagonists derives from the Global Use of Strategies To Open occluded coronary arteries (GUSTO III) Trial.⁶⁰ Of more than 15,000 patients randomized to receive reteplase or alteplase for AMI, 392 patients underwent angioplasty for failed reperfusion at a median of 3.5 hours after thrombolytic therapy had been administered. Eighty-three of these patients received procedural abciximab as adjunctive therapy, while 309 did not. Although patients given abciximab were more frequently in Killip Classes III or IV, the 30-day mortality rate tended to be lower in this group compared to patients who did not receive any abciximab (3.6% versus 9.7%; p = 0.076), even more so after adjustment for baseline differences (p = 0.042). The composite of death, stroke or reinfarction among patients receiving abciximab occurred less often in patients who had been randomized to reteplase as opposed to alteplase, suggesting a greater synergism between these two drugs. Severe bleeding was more common in the abciximab-treated patients (3.6% versus 1.0%; p = 0.08), despite less heparin use. No ICH events occurred with abciximab. In conclusion, this retrospective analysis pointed to a benefit in 30-day mortality among patients undergoing rescue PTCA who received procedural abciximab, with only a minimal increase in serious bleeding complications. Conclusions Early PCI after clinically failed thrombolysis improves AMI-related artery patency, may salvage myocardium and is successful in up to 90% of cases. Despite these potential benefits, early PCI, if unsuccessful, carries a high mortality rate. Therapy with GP IIb/IIIa antagonists has clearly demonstrated remarkable effects in a wide range of clinical scenarios, especially in the interventional arena, with a low rate of bleeding complications. Recent trials examining the effects of abciximab in primary PCI (with or without stenting) in AMI have shown a risk reduction of about 50% in the incidence of ischemic complications. In addition, it has been shown that the combination or co-administration of GP IIb/IIIa antagonists with fibrinolytic agents improves and accelerates reperfusion at a reasonable safety price. Therefore, in conjunction with a strong biological rationale, it would seem logical to administer GP IIb/IIIa antagonists as adjuncts to rescue angioplasty in the face of failed thrombolysis. Although at present there are some nonrandomized data suggesting that there may also exist a synergism between GP IIb/IIIa antagonists and rescue PCI with an acceptable safety profile, there is not enough evidence to delineate the adjunctive chemotherapy to rescue procedures. Until further data become available, this review, together with others, should assist physicians in balancing the risks and benefits when prescribing GP IIb/IIIa antagonists before or during PTCA to patients who have received full-dose thrombolytic therapy and yet failed to reperfuse.

1. Simoons ML, Serruys PW, van den Brand M, et al. Early thrombolysis in acute myocardial infarction: Limitation of infarct size and improved survival. J Am Coll Cardiol 1986;7:717–728. 2. Schroder R, Neuhaus K-L, Leizorovicz A, et al. A prospective placebo-controlled double-blind multicenter trial of intravenous streptokinase in acute myocardial infarction (ISAM): Long-term mortality and morbidity. J Am Coll Cardiol 1987;9:197–203. 3. Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico (GISSI). Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lancet 1986;1:397–402. 4. Wilcox RG, von der Lippe G, Olsson CG, et al. Trial of tissue plasminogen activator for mortality reduction in acute myocardial infarction: Anglo-Scandinavian Study of Early Thrombolysis (ASSET). Lancet 1988;2:525–530. 5. AIMS Trial Study Group. Effect of intravenous APSAC on mortality after acute myocardial infarction: Preliminary report of a placebo-controlled clinical trial. Lancet 1988;1:545–549. 6. ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988;2:349–360. 7. O'Keefe JH Jr, Rutherford BD, McConahay DR, et al. Early and late results of coronary angioplasty without antecedent thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1989;64:1221–1230. 8. Rothbaum DA, Linnemeier TJ, Landin RJ, et al. Emergency percutaneous transluminal coronary angioplasty in acute myocardial infarction: A 3-year experience. J Am Coll Cardiol 1987;10:264–272. 9. Miller PF, Brodie BR, Weintraub RA, et al. Emergency coronary angioplasty for acute myocardial infarction: Results from a community hospital. Arch Intern Med 1987;147:1565–1570. 10. Lieu TA, Gurley RJ, Lundstrom RJ, Parmley WW. Primary angioplasty and thrombolysis for acute myocardial infarction: An evidence summary. J Am Coll Cardiol 1996;27:737–750. 11. GUSTO Angiographic Investigators. The effects of tissue plasminogen activator, streptokinase, or both, on coronary-artery patency, ventricular function and survival after acute myocardial infarction. N Engl J Med 1993;329:1615–1622. 12. Assessment of the Safety and Efficacy of a New Thrombolytic (ASSENT-2) Investigators. Single-bolus tenecteplase compared with front-loaded alteplase in acute myocardial infarction: The ASSENT-2 double-blind randomised trial. Lancet 1999;354:716–722. 13. Coulter SA, McCabe CH, Giugliano RP, et al. Dosing errors and outcomes in patients receiving single bolus compared with bolus + infusion thrombolytic regimens: An InTIME-II study. Circulation 1999;100(Suppl I):I–791. 14. Grines CL, Browne KF, Marco J. A comparison of immediate angioplasty with thrombolytic therapy for acute myocardial infarction. N Engl J Med 1993;328:673–679. 15. Zijlstra F, de Boe MJ, Hoorntje JA. Comparison of immediate coronary angioplasty with intravenous streptokinase in acute myocardial inf