Platelet glycoprotein IIb/IIIa inhibition in acute coronary syndromes

European Heart Journal (00) 3, 1441 1448 doi:10.1053/euhj.00.3160, available online at http://www.idealibrary.com on Platelet glycoprotein IIb/IIIa inhibition in acute coronary syndromes Gradient of benefit related to the revascularization strategy M. Roffi 1, D. P. Chew 1, D. Mukherjee 1, D. L. Bhatt 1, J. A. White 1, D. J. Moliterno 1, C. Heeschen, C. W. Hamm 3, M. A. Robbins 1, N. S. Kleiman 4,P.Théroux 5, H. D. White 6 and E. J. Topol 1 1 Department of Cardiovascular Medicine, The Cleveland Clinic Foundation, Cleveland, OH, U.S.A.; Stanford University School of Medicine, Palo Alto, CA, U.S.A.; 3 Kerckhoff Heart Centre, Bad Nauheim, Germany; 4 Baylor College of Medicine, The Methodist Hospital, Houston, TX, U.S.A.; 5 Montreal Heart Institute, Montreal, Canada; 6 Green Lane Hospital, Auckland, New Zealand Aims To assess the efficacy of platelet glycoprotein IIb/IIIa inhibitors in patients with acute coronary syndromes primarily medically managed. Methods and Results We performed a meta-analysis of the randomized clinical trials of platelet glycoprotein IIb/ IIIa inhibitor therapy in the medical management of non- ST-elevation acute coronary syndromes. Among 9 570 patients, IIb/IIIa integrin blockade was associated with a reduction in death or non-fatal myocardial infarction at 30 days, from 11 5% to 10 7% (odds ratio 0 91, P=0 0). Patients undergoing percutaneous coronary intervention during index hospitalization sustained a greater reduction in ischaemic events (odds ratio 0 8, P=0 01) than patients medically managed (odds ratio 0 95, P=0 7). Among patients undergoing intervention, the benefit was more pronounced if the procedure was performed during glycoprotein IIb/IIIa inhibitor infusion (odds ratio 0 74; P=0 0), than if revascularization was performed after drug discontinuation (odds ratio 0 87, P=0 17). Conclusion This analysis, including the entire large-scale trial experience of intravenous glycoprotein IIb/IIIa inhibitors in patients with acute coronary syndromes primarily medically managed, demonstrates an overall significant, albeit moderate, reduction in 30-day death or myocardial infarction associated with therapy. Although not based on a prospectively defined hypothesis, the findings suggest a gradient of benefit conferred by these agents depending on the revascularization strategy used. (Eur Heart J, 00; 3: 1441 1448, doi:10.1053/euhj.00. 3160) 00 The European Society of Cardiology. Published by Elsevier Science Ltd. All rights reserved. Key Words: Acute coronary syndromes, abciximab, eptifibatide, lamifiban, tirofiban, glycoprotein IIb/IIIa receptor inhibitors. See doi:10.1053/euhj.00.370 for the Editorial comment on this article Introduction Aspirin, and to a lesser extent heparin, have been shown to reduce subsequent myocardial infarction and death in acute coronary syndromes without ST-segment Revision submitted 31 December 001, and accepted January 00. Dr Roffi is supported by a research grant from the Swiss National Science Foundation. Correspondence: Eric J. Topol, MD, Department of Cardiovascular Medicine, Desk F5, The Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195, U.S.A. elevation [1,]. Nevertheless, patients remain at risk for recurrent ischaemic events. Newer strategies involving more potent platelet inhibition have focused on the platelet surface membrane IIb/IIIa integrin, the primary receptor in platelet activation and aggregation [3]. The benefit of these agents as adjunctive treatment for percutaneous coronary intervention has been substantial and consistently observed among trials, which have collectively enrolled over 19 000 patients. The observed reduction in 30-day ischaemic end-points (death, myocardial infarction, and urgent revascularization) ranged from 16% to 56% [4]. Conversely, the efficacy of IIb/IIIa inhibitors in the empiric medical management of 0195-668X/0/$35.00 00 The European Society of Cardiology. Published by Elsevier Science Ltd. All rights reserved.

144 M. Roffi et al. non-st-segment elevation acute coronary syndromes has been less marked [5], and their use has been controversial, particularly following the negative results of GUSTO IV [6]. We performed a meta-analysis of the large-scale randomized trials investigating the use of IIb/IIIa integrin blockers in the medical management of non- ST-segment elevation acute coronary syndromes. The purposes of the study were to better characterize the overall benefit of these agents and to assess whether the reduction in ischaemic end-points associated with therapy was influenced by the revascularization strategy used. Methods Trial selection Six randomized, double-blind, placebo-controlled trials of intravenous platelet glycoprotein IIb/IIIa antagonists evaluating the medical management of acute coronary syndromes in the absence of ST-segment elevation were identified through a MEDLINE search. Records between 1990 and 001 were searched for the words platelet, intravenous, unstable angina, random and inhibit or block, where was a wild card. Trials were included if the total number of patients exceeded 1000 and the duration of clinical follow-up was 30 days. Data from trials presented at major cardiology meetings were also considered. In all trials patients received aspirin. The Platelet Receptor Inhibition in Ischemic Syndrome Management (PRISM) trial [7] randomized 33 patients to either tirofiban (0 6 μg.kg 1 min 1 for 30 min followed by a 0 15 μg.kg 1 min 1 infusion) or unfractionated heparin for 48 h. Invasive assessment was discouraged within the first 48 h of randomization, and the drug was discontinued if percutaneous revascularization was performed. The primary end-point was death, myocardial infarction, or refractory ischaemia at 48 h. Patients were followed for 30 days in a pre-defined exploratory analysis. The Platelet Receptor Inhibition in Ischemic Syndrome Management in Patients Limited by Unstable Signs and Symptoms (PRISM-PLUS) trial [8] randomized 1915 patients to either tirofiban (0 6 μg.kg 1 min 1 for 30 min followed by a 0 15 μg.kg 1 min 1 infusion), tirofiban (0 4 μg.kg 1 min 1 for 30 min followed by a 0 1 μg.kg 1 min 1 infusion) with heparin, or unfractionated heparin alone, for a duration of at least 48 h. The study arm not including heparin (n=345) was discontinued before completion of the trials and was not included in this analysis. Invasive assessment was encouraged, though deferred for the first 48 h. In patients undergoing early percutaneous revascularization, the study drug was continued for additional 1 to 4 h. The primary end-point was the composite of death, myocardial infarction, or refractory ischaemia at 7 days. The Platelet IIb/IIIa Antagonism for the Reduction of Acute Coronary Syndrome Events in a Global Organization Network (PARAGON) A trial [9] randomized 8 patients to lamifiban 300 μg bolus followed by a 1 0 μg min 1 infusion, or 750 μg bolus followed by a 5 0 μg min 1 infusion, or placebo for 3 to 5 days. All patients in the placebo arm received heparin, whereas by factorial design, heparin therapy was randomized among patients receiving lamifiban. Invasive assessment was discouraged. In patients undergoing early percutaneous coronary intervention, the study drug was continued for additional 1 to 4 h. The primary end-point was a composite of death or myocardial infarction at 30 days. The Platelet IIb/IIIa Antagonism for the Reduction of Acute Coronary Syndrome Events in a Global Organization Network (PARAGON) B trial [10] randomized 55 patients to lamifiban (500 μg bolus followed by an infusion of 1 0 to 0 μg min 1 according to estimated renal function) or placebo for 7 h or until hospital discharge. All patients received heparin. Coronary angiography and revascularization were discouraged in the first 4 h and then performed according to local standards of practice. In patients undergoing early percutaneous revascularization, the study infusion was continued for an additional 18 to 48 h. The primary end-point was death, myocardial infarction, or severe recurrent ischaemia at 30 days. Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PUR- SUIT) [11] randomized 10 948 patients to eptifibatide 180 μg.kg 1 bolus and 1 3 μg.kg 1 min 1 infusion, or bolus and 0 μg.kg 1 min 1 infusion, or placebo for up to 7 h. Adjunctive unfractionated heparin was encouraged but not required. The protocol mandated the discontinuation of the lower-dose arm after documentation of an acceptable safety profile of the higher dose in the interim analysis. Accordingly, the patients randomized to either placebo or higher-dose of eptifibatide were included in the analysis (n=9461). Coronary angiography and revascularization were performed at the discretion of the physician. In patients undergoing early percutaneous intervention, the drug was continued for at least 4 h. The primary end-point was a composite of death or myocardial infarction at 30 days. The Global Utilization of Strategies To Open Occluded Coronary Arteries (GUSTO) IV [6] randomized 7800 patients to either abciximab 0 5 μg.kg 1 bolus and 0 15 μg.kg 1 min 1 (max. 10 μg min 1 ) infusion for 4 h, or bolus and infusion for 48 h, or placebo. All patients received either heparin or low-molecular-weight heparin. Early invasive assessment was strongly discouraged. Most of the patients undergoing intervention received either blinded crossover therapy or abciximab on an open-label basis. The primary end-point was death or myocardial infarction at 30 days. For the purpose of this analysis, the 4 h-infusion and 48 h- infusion groups were pooled.

Platelet GPIIb/IIIa inhibition in ACS 1443 All patients PRISM 33 P = 0 11 7 1% 5 8% PRISM-PLUS 1570 P = 0 03 PARAGON A 8 P = 0 71 PARAGON B 55 P = 0 34 PURSUIT 9461 P = 0 04 GUSTO IV 7800 P = 0 36 11 9% 8 7% 1 0% 11 5% 11 4% 10 6% 15 7% 14 % 8 0% 8 7% 9 570 P = 0 0 11 5% 10 7% Breslow-Day P = 0 19 Figure 1 Odds ratio with 95% confidence intervals (CI) and corresponding P-values for treatment effect on overall 30-day death or myocardial infarction (MI). Values to left of 1 0 indicate a benefit of platelet glycoprotein IIb/IIIa inhibition. End-points and statistical analysis Death, and death or non-fatal myocardial infarction, were assessed at 30 days based on an intention-to-treat analysis. The enzyme definitions of myocardial infarction were creatine kinase or creatine kinase-mb greater than the upper limit of normal in PURSUIT; creatine kinase or creatine kinase-mb greater than twice the upper limit of normal in PRISM, PRISM-PLUS, PARAGON A and PARAGON B; and creatine kinase-mb greater than three times the upper limit of normal in GUSTO IV. In addition, PRISM-PLUS, PARAGON B, and PURSUIT defined myocardial infarction post percutaneous intervention as creatine kinase or creatine kinase-mb greater than three times the upper limit of normal. Odds ratios and corresponding 95% confidence intervals were calculated for individual study populations. The Mantel Haenszel statistic (SAS 6.1, SAS Institute Inc.) was used to test the significance of treatment effect within each study. Heterogeneity of the odds ratios across the trials was tested with the Breslow Day statistic. If the resulting P-value was non-significant, individual event rates were weighted and pooled. A Pearson chi-square test was applied to the pooled event rates to assess overall significance of treatment effects. A value of P<0 05 was considered statistically significant. Results The study involved six trials and 9 570 patients. The aggregate analysis showed a significant reduction of death or myocardial infarction at 30 days associated with platelet glycoprotein IIb/IIIa inhibition, from 11 5% to 10 7% (odds ratio 0 91, 95% confidence interval (CI) 0 85 0 99; P=0 0) (Fig. 1). With the exception of GUSTO IV, the benefit was consistent across the trials in the absence of significant heterogeneity (Breslow Day P=0 19), and reached statistical significance independently in PRISM-PLUS (odds ratio 0 70; P=0 03) and PURSUIT (odds ratio 0 89; P=0 04). The reduction in mortality associated with glycoprotein IIb/IIIa inhibition (from 3 7% to 3 4%) was of a similar proportion but did not reach statistical significance (odds ratio 0 90, 95% CI 0 80 1 03, P=0 1). A total of 6337 patients (1%) underwent percutaneous revascularization during index hospitalization. In this setting, glycoprotein IIb/IIIa inhibition was associated with a significant reduction in 30-day death or myocardial infarction, from 1 7% to 10 7% (odds ratio 0 8, 95% CI 0 71 0 96, P=0 01) (Fig. ). The benefit was consistent across the trials in the absence of heterogeneity (Breslow Day P=0 79), but did not reach statistical significance in individual studies. The odds ratios for the composite end-point associated with therapy ranged between 0 61 (P=0 09) in PRISM-PLUS and 0 91 (P=0 63) in GUSTO IV. Mortality at 30 days was reduced from 1 9% to 1 4% (odds ratio 0 73, 95% CI 0 49 1 09, P=0 1). The greatest reduction in death or myocardial infarction was observed among patients undergoing percutaneous revascularization while still receiving the glycoprotein IIb/IIIa inhibitor infusion. Among these 49 patients, the death or myocardial infarction composite was lowered from 13 6% to 10 5% (odds ratio 0 74, 95% CI 0 57 0 96, P=0 0) by IIb/IIIa integrin blockade (Fig. 3). The benefit was observed across the trials in the absence of significant heterogeneity (Breslow Day P=0 78). Within the individual

1444 M. Roffi et al. PCI PRISM 700 P = 0 36 9 1% 7 % PRISM-PLUS 501 P = 0 09 PARAGON A 337 P = 0 18 PARAGON B 136 P = 0 1 PURSUIT 09 P = 0 33 GUSTO IV 164 P = 0 63 13 0% 8 3% 14 3% 9 5% 1 0% 9 9% 14 7% 13 3% 11 4% 10 5% 6337 P = 0 01 1 7% 10 7% Breslow-Day P = 0 79 Figure Odds ratio with 95% confidence intervals (CI) and corresponding P-values for treatment effect on 30-day death or myocardial infarction (MI) in patients undergoing percutaneous coronary intervention (PCI) during index hospitalization. Values to left of 1 0 indicate a benefit of platelet glycoprotein IIb/IIIa inhibition. PCI on study drug PRISM-PLUS 335 P = 0 49 9 7% 7 5% PARAGON A 73 P = 0 45 15 % 7 5% PARAGON B 60 P = 0 04 PURSUIT 1116 P = 0 8 13 3% 8 % 13 % 11 1% GUSTO IV 13 P = 0 84 7 9% 6 3% 49 P = 0 0 13 6% 10 5% Breslow-Day P = 0 78 Figure 3 Odds ratio with 95% confidence intervals (CI) and corresponding P-values for treatment effect on 30-day death or myocardial infarction (MI) in patients undergoing percutaneous coronary intervention (PCI) while on study drug. Patients enrolled in PRISM were excluded from this analysis as protocol mandated drug discontinuation prior to revascularization. Values to left of 1 0 indicate a benefit of platelet glycoprotein IIb/IIIa inhibition. studies the ischaemic end-points reduction associated with therapy reached statistical significance only in PARAGON B (OR 0 58; P=0 04). Patients enrolled in PRISM were excluded from this analysis since protocol mandated drug discontinuation prior to revascularization. The few patients undergoing early percutaneous revascularization in GUSTO IV had high event rates, probably due to the restrictive requirements for early invasive assessment (i.e. recurrent or continuing ischaemia at rest with ECG changes refractory to medical management). Of note, the treatment effect in patients undergoing percutaneous intervention in GUSTO IV could not be adequately assessed because most patients received either blinded

Platelet GPIIb/IIIa inhibition in ACS 1445 PCI after discontinuation of study drug PRISM 700 P = 0 36 9 1% 7 % PRISM-PLUS 166 P = 0 03 PARAGON A 64 P = 0 3 PARAGON B 74 P = 0 91 PURSUIT 1093 P = 0 77 GUSTO IV 1141 P = 0 69 1 1% 9 5% 14 0% 9 9% 10 9% 11 % 16 % 15 6% 9 6% 8 9% 4088 P = 0 17 1 3% 10 9% Breslow-Day P = 0 46 Figure 4 Odds ratio with 95% confidence intervals (CI) and corresponding P-values for treatment effect on 30-day death or myocardial infarction (MI) in patients undergoing percutaneous coronary intervention (PCI) after discontinuation of study drug. Values to left of 1 0 indicate a benefit of platelet glycoprotein IIb/IIIa inhibition. crossover therapy or abciximab on an open-label basis [6]. Patients undergoing percutaneous intervention after drug discontinuation had a moderate event decrease associated with therapy that did not reach statistically significance, from 1 3% to 10 9% (odds ratio 0 87, 95% CI 0 7 1 06, P= 0 17) (Fig. 4). The degree of efficacy ranged from a significant event reduction in PRISM- PLUS (odds ratio 0 39) to a minimal event increase in PARAGON B (odds ratio 1 03). A marginal event reduction was observed among patients treated conservatively (n=0 054). The incidence of death or myocardial infarction at 30 days was 9 7% in the placebo group and 9 3% in patients treated with platelet glycoprotein IIb/IIIa inhibitors (odds ratio 0 95, 95% CI 0 86 1 04, P=0 7) (Fig. 5). Within the individual trials, a significant event reduction associated with therapy was observed in PRISM (odds ratio 0 57) only. The mortality at 30 days was 4 0% in the placebo group and 3 7% in the active treatment group (OR 0 91; 95% CI 0 79 1 06, P=0 3). Discussion This meta-analysis, including the entire large-scale clinical trial experience of intravenous platelet glycoprotein IIb/IIIa inhibitors for the medical management of acute coronary syndromes, demonstrates a significant, albeit moderate (odds ratio 0 91, P=0 0) overall reduction in death or myocardial infarction at 30 days associated with therapy. This translates into eight ischaemic events prevented per 1000 patients treated. Although the trials were primarily designed to test this drug class in medically managed patients, a proportion of patients underwent percutaneous revascularization. Our analysis suggests a gradient of benefit associated with platelet glycoprotein IIb/IIIa inhibition as adjunct treatment for acute coronary syndromes depending upon strategy and timing of revascularization (Fig. 6). Among patients undergoing intervention during the index hospitalization, those randomized to IIb/IIIa integrin blockade experienced a significant reduction in death or myocardial infarction at 30 days, corresponding to 0 events prevented per 1000 patients treated (odds ratio 0 8, P=0 01). The mortality benefit was of even greater proportion, albeit not reaching statistical significance due to lower event rates. When stratified by timing of revascularization, a significant reduction in terms of death or myocardial infarction was observed only in individuals undergoing revascularization while receiving glycoprotein IIb/IIIa inhibitors (odds ratio 0 74, P=0 0). This translates to 31 deaths or myocardial infarctions prevented for 1000 patients treated. If percutaneous intervention was performed after discontinuation of study drug, the benefit was less marked (odds ratio 0 87, P=0 17), corresponding to 14 ischaemic events prevented for 1000 patients treated. In patients solely medically managed, IIb/IIIa integrin blockade was associated with marginal ischaemic event reduction (odds ratio 0 95, P=0 7), equal to four events prevented per 1000 patients treated. The proportion of mortality reduction, as expressed by the odds ratio, was similar as to the combined end-point, but did not reach statistical significance in any of the subgroups addressed.

1446 M. Roffi et al. Medical management PRISM 1999 P = 0 01 6 % 3 6% PRISM-PLUS 719 P = 0 9 PARAGON A 1689 P = 0 97 PARAGON B 3894 P = 0 71 PURSUIT 5919 P = 0 36 GUSTO IV 5834 P = 0 30 10 1% 7 8% 10 4% 10 5% 11 % 10 8% 1 7% 11 9% 6 6% 7 4% 0 054 P = 0 7 9 7% 9 3% Breslow-Day P = 0 1 Figure 5 Odds ratio with 95% confidence intervals (CI) and corresponding P-values for treatment effect on 30-day death or myocardial infarction (MI) in patients medically managed. Values to left of 1 0 indicate a benefit of platelet glycoprotein IIb/IIIa inhibition. Figure 6 Odds ratio with 95% confidence intervals (CI) and corresponding P-values for treatment effect on 30-day death or myocardial infarction (MI) according to revascularization strategy. Values to left of 1 0 indicate a benefit of platelet glycoprotein IIb/IIIa inhibition. While not based on a randomized assessment and derived from a post-hoc analysis, this observation suggests that maximal efficacy of platelet glycoprotein IIb/ IIIa inhibition in unstable patients may be achieved if pharmacological and mechanical stabilization occur simultaneously, whereas the benefit conferred by pharmacological stabilization may decrease if both therapy modalities occur sequentially, and only a modest benefit may be observed if sole aggressive platelet inhibition is pursued. These findings should be considered hypothesis-generating and require independent validation. Ideally, validation would involve a randomized assessment of patients with acute coronary syndromes either medically managed, or started on glycoprotein IIb/IIIa integrin blockade and then undergoing early invasive assessment, or undergoing initial glycoprotein IIb/IIIa-based medical management and then undergoing percutaneous intervention. However, such a trial would be ethically questionable. In fact, the superiority of an early invasive strategy over a conservative treatment in the setting of IIb/IIIa integrin blockade with tirofiban has been convincingly demonstrated in the

Platelet GPIIb/IIIa inhibition in ACS 1447 TACTICS trial [1]. The FRISC II trial [13] documented the enhanced efficacy of an invasive strategy compared to a conservative treatment also in the absence of glycoprotein IIb/IIIa receptor inhibition-based antithrombotic therapy. In addition, the CAPTURE trial [14] showed that pharmacological stabilization with abciximab followed by mandatory revascularization on study drug in high-risk unstable patients led to significant event reduction compared to placebo. And finally, it would be difficult to justify an initial conservative management followed by percutaneous intervention in the absence of potent platelet inhibition. In fact, the majority of the myocardial infarctions occur at the time of revascularization, and platelet glycoprotein IIb/IIIa receptor inhibitors have been consistently shown to reduce periprocedural events [5]. All patients undergoing percutaneous intervention in this analysis had prolonged glycoprotein IIb/IIIa inhibitor infusion prior to revascularization. Therefore, while this analysis supports the concept that the benefit with these agents is closely related to the revascularization procedure, part of the ischaemic event reduction may have occurred prior to percutaneous intervention. Such upstream benefit has been described for CAPTURE, PURSUIT and PRISM PLUS [15]. No conclusion can be drawn from our analysis on the use of these drugs immediately prior to percutaneous intervention, as none of the trials addressed such a strategy. The limited benefit of glycoprotein IIb/IIIa inhibitors observed in this analysis among patients solely medically managed should not lead to the conclusion that these agents are indicated only in patients undergoing percutaneous coronary intervention. In fact, specific subgroups of unstable patients primarily medically managed have been shown to derive significant benefit from IIb/IIIa integrin blockade. Accordingly, an analysis of the diabetic populations enrolled in the same six glycoprotein IIb/IIIa trials demonstrated that the use of these agents was associated with a significant mortality reduction [16].In addition, troponin-positive patients have been shown to derive particular benefit from IIb/IIIa integrin blockade in some [17,18], albeit not all [6], trials. Based on the present study, as well as on the subgroup analyses just mentioned, glycoprotein IIb/IIIa inhibitors should be strongly considered in patients presenting with troponin elevation, in diabetics, and in patients undergoing an early invasive strategy. Our analysis does not allow efficacy comparison among different platelet glycoprotein IIb/IIIa inhibitors due to differences in trial designs and populations enrolled. No conclusive data are available in the absence of direct comparisons within the acute coronary syndrome population. examines the statistical heterogeneity among odds ratios, and therefore provides information about the validity of pooling the results from different trials, failed to demonstrate significant diversity among the analyses. In addition, since the revascularization strategy was not randomized, the ischaemic event reduction associated with platelet glycoprotein IIb/IIIa inhibitors in patients undergoing PCI may have been influenced by selection bias. Therefore, this finding should be considered hypothesis-generating and requires independent validation. In addition, the enzyme definition of myocardial infarction was not uniform across the trials, and we cannot rule out that different cut-offs for enzyme elevation may have influenced the incidence of events and possibly the extent of therapeutic benefit. However, this factor should have been mitigated as the statistical analysis was based on a comparison of treatment groups within each trial. In order to consider only adjudicated events the end-point definitions of the trials were respected. Finally, in PRISM the primary end-points were assessed at 48 h and PRISM-PLUS at 7 days. However, in both trials all 30-day events were adjudicated. Conclusions The aggregate analysis of the entire large-scale clinical trial experience of intravenous platelet glycoprotein IIb/ IIIa inhibitors for the medical management of non-stsegment acute coronary syndromes, demonstrates that the use of these agents is associated with an overall significant, albeit moderate, reduction in death or myocardial infarction at 30 days. Although not based on a prospectively defined hypothesis, our findings suggest a gradient of benefit conferred by these agents depending upon the revascularization strategy used. Accordingly, patients undergoing percutaneous coronary intervention while on platelet glycoprotein IIb/IIIa inhibitor derived a significant benefit, while patients undergoing revascularization after drug discontinuation demonstrated a moderate event reduction that did not reach statistically significance, and only a marginal benefit was observed among patients medically managed. The authors would like to acknowledge Peter M. DiBattiste, MD, and Steven Snapinn, PhD, Merck Research Laboratories, West Point, Pa., for providing the data from PRISM and PRISM- PLUS; Donna Bressan, The Cleveland Clinic Foundation, for editorial assistance; Suzanne Turner, Cardiovascular Graphics Department, The Cleveland Clinic Foundation, for providing assistance with graphics. Limitations Inherent to all meta-analyses, the included trials differed in design, inclusion criteria, therapeutic agents, regimens, and access to percutaneous revascularization, among others. However, the Breslow Day test, which References [1] Collaborative overview of randomised trials of antiplatelet therapy I: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. Antiplatelet Trialists Collaboration. BMJ 1994; 308: 81 106.

1448 M. Roffi et al. [] Oler A, Whooley MA, Oler J, Grady D. Adding heparin to aspirin reduces the incidence of myocardial infarction and death in patients with unstable angina. A meta-analysis. JAMA 1996; 76: 811 5. [3] Topol EJ, Byzova TV, Plow EF. Platelet GPIIb-IIIa blockers. Lancet 1999; 353: 7 31. [4] Chew DP, Moliterno DJ. critical appraisal of platelet glycoprotein IIb/IIIa inhibition. J Am Coll Cardiol 000; 36: 08 35. [5] Lincoff AM, Califf RM, Topol EJ. Platelet glycoprotein IIb/IIIa receptor blockade in coronary artery disease. J Am Coll Cardiol 000; 35: 1103 15. [6] Simoons ML. Effect of glycoprotein IIb/IIIa receptor blocker abciximab on outcome in patients with acute coronary syndromes without early coronary revascularisation: the GUSTO IV-ACS randomised trial. Lancet 001; 357: 1915 4. [7] A comparison of aspirin plus tirofiban with aspirin plus heparin for unstable angina. Platelet Receptor Inhibition in Ischemic Syndrome Management (PRISM) Study Investigators. N Engl J Med 1998; 338: 1498 505. [8] Inhibition of the platelet glycoprotein IIb/IIIa receptor with tirofiban in unstable angina and non-q-wave myocardial infarction. Platelet Receptor Inhibition in Ischemic Syndrome Management in Patients Limited by Unstable Signs and Symptoms (PRISM-PLUS) Study Investigators. N Engl J Med 1998; 338: 1488 97. [9] International, randomized, controlled trial of lamifiban (a platelet glycoprotein IIb/IIIa inhibitor), heparin, or both in unstable angina. The PARAGON Investigators. Platelet IIb/ IIIa Antagonism for the Reduction of Acute coronary syndrome events in a Global Organization Network. Circulation 1998; 97: 386 95. [10] Moliterno DJ. Patient-specific dosing of IIb/IIIa antagonists during acute coronary syndromes: rationale and design of the PARAGON B study. The PARAGON B International Steering Committee. Am Heart J 000; 139: 563 6. [11] Inhibition of platelet glycoprotein IIb/IIIa with eptifibatide in patients with acute coronary syndromes. The PURSUIT Trial Investigators. Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy. N Engl J Med 1998; 339: 436 43. [1] Cannon CP, Weintraub WS, Demopoulos LA et al. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med 001; 344: 1879 87. [13] Invasive compared with non-invasive treatment in unstable coronary-artery disease: FRISC II prospective randomised multicentre study. FRagmin and Fast Revascularisation during InStability in Coronary artery disease Investigators. Lancet 1999; 354: 708 15. [14] Randomised placebo-controlled trial of abciximab before and during coronary intervention in refractory unstable angina: the CAPTURE Study. Lancet 1997; 349: 149 35. [15] Boersma E, Akkerhuis KM, Theroux P, Califf RM, Topol EJ, Simoons ML. Platelet glycoprotein IIb/IIIa receptor inhibition in non-st-elevation acute coronary syndromes: early benefit during medical treatment only, with additional protection during percutaneous coronary intervention. Circulation 1999; 100: 045 8. [16] Roffi M, Chew DP, Mukherjee D et al. Platelet glycoprotein IIb/IIIa inhibitors reduce mortality in diabetic patients with non-st-segment-elevation acute coronary syndromes. Circulation 001; 104: 767 71. [17] 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. PRISM Study Investigators. Platelet Receptor Inhibition in Ischemic Syndrome Management. Lancet 1999; 354: 1757 6. [18] Newby LK, Ohman EM, Christenson RH et al. Benefit of glycoprotein IIb/IIIa inhibition in patients with acute coronary syndromes and troponin t-positive status: the paragon-b troponin T substudy. Circulation 001; 103: 891 6.