Randomized, Blinded Trial Comparing Fondaparinux With Unfractionated Heparin in Patients Undergoing Contemporary Percutaneous Coronary Intervention
Arixtra Study in Percutaneous Coronary Intervention: A Randomized Evaluation (ASPIRE) Pilot Trial
Background— Factor Xa plays a central role in the generation of thrombin, making it a novel target for treatment of arterial thrombosis. Fondaparinux is a synthetic factor Xa inhibitor that has been shown to be superior to standard therapies for the prevention of venous thrombosis. We performed a randomized trial to determine the safety and feasibility of fondaparinux in the percutaneous coronary intervention (PCI) setting.
Methods and Results— A total of 350 patients undergoing elective or urgent PCI were randomized in a blinded manner to receive unfractionated heparin (UFH), 2.5 mg fondaparinux IV, or 5.0 mg fondaparinux IV. Randomization was stratified for planned or no planned use of glycoprotein (GP) IIb/IIIa antagonists. The primary safety outcome was total bleeding, which was a combination of major and minor bleeding events. The incidence of total bleeding was 7.7% in the UFH group and 6.4% in the combined fondaparinux groups (hazard ratio, 0.81; 95% confidence interval, 0.35 to 1.84; P=0.61). Bleeding was less common in the 2.5-mg fondaparinux group compared with the 5-mg fondaparinux group (3.4% versus 9.6%, P=0.06). The composite efficacy outcome of all-cause mortality, myocardial infarction, urgent revascularization, or need for a bailout GPIIb/IIIa antagonist was 6.0% in the UFH group and 6.0% in the fondaparinux group, with no significant difference in efficacy among the fondaparinux doses compared with UFH. Coagulation marker analysis at 6 and 12 hours after PCI demonstrated that fondaparinux was superior to UFH in inducing a sustained reduction in markers of thrombin generation, as measured by prothrombin fragment F1.2 (P=0.02).
Conclusions— In this pilot study of patients undergoing contemporary PCI, factor Xa inhibition with the synthetic anticoagulant fondaparinux in doses of 2.5 and 5.0 mg was comparable to UFH for clinical safety and efficacy outcomes. These data form the basis for further evaluation of fondaparinux in arterial thrombosis.
Received July 19, 2004; revision received December 13, 2004; accepted December 21, 2004.
Unfractionated heparin (UFH) has traditionally been used to prevent complications in patients undergoing percutaneous coronary intervention (PCI).1,2 UFH exerts its anticoagulant effect by catalyzing the inhibition of thrombin and factor Xa by antithrombin. UFH has several important limitations, including its unpredictable anticoagulant effect necessitating careful laboratory monitoring, high protein binding, and inactivation by platelet factor 4.3 To overcome some of these limitations, newer agents with more predictable anticoagulant effect and with greater anti-factor Xa activity are being evaluated. Factor Xa occupies a pivotal role in the clotting cascade because it is the final common pathway linking the intrinsic and extrinsic systems leading to the generation of thrombin, thus making it an attractive target for the development of new anticoagulant agents.3
Fondaparinux is a synthetic, pure factor Xa inhibitor that is structurally similar to the antithrombin binding site of heparin and low-molecular-weight heparin (LMWH). In contrast to heparins, fondaparinux selectively binds to antithrombin and causes rapid and predictable inhibition of factor Xa. Fondaparinux has a half-life of 15 hours, with linear pharmacokinetics and low interindividual and intraindividual variability, thus obviating the need for laboratory monitoring.4 Fondaparinux has been evaluated extensively for the prevention of venous thrombosis in patients undergoing orthopedic surgery, where it was shown to be consistently more effective than LMWH.5–9 For the treatment of deep vein thrombosis or pulmonary embolus, fondaparinux was at least as effective as enoxaparin or UFH, respectively.10,11 In arterial thrombosis, an early study of patients undergoing PCI treated with 12 mg fondaparinux demonstrated feasibility with few complications.10 Subsequent dose-finding phase II trials in patients with acute coronary syndromes (ACS) have suggested that doses as low as 2.5 mg may be optimal from both a safety and efficacy perspective.11,12
To explore the feasibility and safety of fondaparinux at doses of 2.5 and 5.0 mg given intravenously in patients undergoing contemporary urgent or elective PCI, we conducted a pilot phase II, randomized, multicenter, blinded trial of these 2 doses compared with UFH.
The Arixtra Study in Percutaneous coronary Intervention: a Randomized Evaluation (ASPIRE) Trial was a randomized, blinded, multicenter phase II trial comparing 2 intravenous doses of fondaparinux (2.5 and 5.0 mg) with UFH. Twenty-two centers from Canada, France, and the United States participated in the trial (see online-only Data Supplement for names of centers).
Patients >21 years of age and able to provide informed consent were eligible if they were undergoing urgent or elective PCI for ACS (including primary PCI for ST-segment-elevation myocardial infarction [STEMI]) or stable angina. Exclusion criteria included an activated clotting time >200 seconds immediately before PCI, use of LMWH within 6 hours of PCI, currently receiving an oral anticoagulant with an international normalized ratio >1.8, thrombolytic therapy for STEMI in the previous 24 hours (ie, rescue PCI), active internal bleeding or a history of hemorrhagic diathesis, pregnancy or intent of pregnancy, absolute contraindication to anticoagulation, participation in other clinical research studies involving the evaluation of other investigational drugs or devices within 30 days of enrollment, or prior participation in this trial.
Patients were recruited between June and November 2003. The ethics review board at each institution approved the study, and all patients gave written, informed consent. The study was organized and coordinated and all the data were managed and analyzed by the Canadian Cardiovascular Collaboration Project Office, McMaster University, Hamilton, Canada. A steering committee consisting of national coordinators oversaw the study. The data were periodically reviewed by an independent data and safety monitoring board. An independent events adjudication committee adjudicated all primary outcome events (see the online-only Data Supplement for names of committee members).
We randomly assigned patients to receive fondaparinux (2.5 or 5.0 mg IV) or UFH. Randomization was stratified for planned use of a glycoprotein (GP) IIb/IIIa antagonist, which was given as per local practice. The dose of UFH was 100 U/kg without a GPIIb/IIIa antagonist or 65 U/kg with a GPIIb/IIIa antagonist. Permuted block randomization was used and was performed by using sealed envelopes on NCR paper, ensuring adequate concealment of treatment allocation. Before commencement of the trial, each center identified an unblinded anticoagulation coordinator, whose role it was to open the randomization envelope and prepare the study drug, which was identical in appearance for all treatment groups. All other personnel were blinded to treatment allocation. The activated clotting time and activated partial thromboplastin time were not routinely monitored because fondaparinux has no effect on these tests, and the doses of UFH selected were those routinely used. Markers of periprocedural myonecrosis (creatine kinase [CK], CK-MB, and troponin) and an ECG were determined systematically in all patients before PCI and at 6 to 8 hours and 12 to 14 hours after PCI. Vascular access site sheaths were removed immediately after the procedure when a vascular closure device was used or when a radial artery approach was used. Sheaths were otherwise removed 6 hours after the procedure.
An additional blood sample was taken at baseline (before PCI) and at 6 and 12 hours after PCI from 233 patients. All samples were immediately processed, stored at −70°C, and shipped in dry ice to the Canadian Cardiovascular Collaboration Project Office. Samples were analyzed in the Coagulation Core Laboratory at the Henderson Research Centre, Hamilton, Canada. Baseline characteristics and outcomes in patients providing blood samples were similar to those of the overall population. Plasma was analyzed for prothrombin fragment 1.2 (F1.2), thrombin-antithrombin complex (TAT), P-selectin, d-dimer, factor VIIa, and anti-factor Xa levels. F1.2 and TAT were assayed with commercial ELISA kits from Dade Behring. Factor VIIa was measured with a commercially available kit (Staclot VIIa-rTF, Diagnostica Stago). P-selectin antigen was measured with a commercial assay from R&D Systems. d-dimer was measured with an automated microlatex assay (MDA d-dimer, BioMerieux). Anti-factor Xa levels were measured by an amidolytic assay with the Stachrom heparin kit (Diagnostica Stago) according to the method of Teien et al.13 Fondaparinux, which has a specific activity of 700 anti-factor Xa units per milligram, was used as the standard in this assay.
The primary analysis was related to safety and consisted of the composite of major and minor bleeding (total bleeding) events 48 hours after PCI. Major bleeding was defined as clinically overt bleeding with one of the following criteria: fatal, symptomatic intracranial hemorrhage, retroperitoneal hemorrhage, intraocular hemorrhage, or a fall in hemoglobin of ≥3.0 g/dL, with each blood transfusion unit counting for 1.0 g/dL of hemoglobin, or transfusion of ≥2 U of blood. Minor bleeding was defined as any other clinically overt bleed not meeting the definition for major bleeding. The principal efficacy measure was a composite of clinical outcomes, including death, MI, urgent revascularization, and bailout use of a GPIIb/IIIa antagonist. Diagnosis of myocardial (re)infarction after PCI was made on the basis of a CK-MB or a CK value >3 times the upper limit of normal or an increase by at least 50% over the previous valley level; new, significant Q waves of ≥0.04 second duration in 2 or more contiguous leads; new, persistent ST-segment elevation; or new left bundle-branch block. In patients presenting with STEMI, myocardial reinfarction was defined as recurrent chest discomfort associated with new ST-segment elevation or reelevation of CK or CK-MB to >50% of the previous valley level. Urgent revascularization was defined as ischemic symptoms that resulted in either urgent repeated PCI or coronary artery bypass graft surgery. Need for a bailout GPIIb/IIIa inhibitor was defined as an urgent need for a GPIIb/IIIa inhibitor because of an acute (or suspected acute) complication of PCI. This outcome event applied only to those patients who did not receive a planned upfront GPIIb/IIIa inhibitor. Abrupt closure was defined as TIMI 0 or 1 flow in a target vessel that had TIMI 2 or 3 flow at any time from enrollment to hospital discharge, documented by angiography. An additional objective of the study was to assess the impact of fondaparinux on coagulation markers. An independent clinical events committee that was blinded to treatment allocation adjudicated all major safety and efficacy outcomes.
The purpose of this pilot study was to assess the feasibility of using fondaparinux during PCI and to explore the risk (either bleeding or periprocedural clinical complications) to guide the conduct of much larger studies. We acknowledged a priori that smaller differences among the groups would not be detectable given the modest sample size. With 300 patients and a composite rate of major and minor bleeding in the UFH group of ≈6%, the study would have 80% power to detect a relative risk of 2.95 at a 2-tailed α (type I error) of 5%. The study eventually randomized 350 patients, which provided 80% power to detect a relative risk of 2.77 in total bleeding, with an event rate in the UFH group of 7.7% and a 2-tailed α of 5%.
All analyses were based on the intention-to-treat principle. The primary safety and efficacy analyses used the time-to-first event with Cox proportional-hazards models stratified by planned or no planned GPIIb/IIIa inhibitor use. The point estimate of the hazard ratio (HR) of fondaparinux (both doses combined) versus UFH and its associated 2-sided 95% confidence interval (CI) was calculated. Comparisons of both fondaparinux dosages separately against UFH as well as against each other were also done. Separate analyses were conducted to compare treatment groups in patients receiving GPIIb/IIIa inhibitors and those not receiving these agents. Because of the modest sample size, there were some imbalances in baseline characteristics. Because the study was randomized, these imbalances were, by definition, due to chance alone. All of these characteristics were entered into a Cox proportional-hazards model to determine whether they were associated with the primary safety and efficacy outcomes, regardless of treatment allocation. Only those baseline characteristics that were associated with the outcomes were chosen as covariates in the models. These covariates were age, sex, and ST-segment depression.
Coagulation data are presented as geometric means and were analyzed by ANOVA on logarithmically transformed measures. The F test was used to determine the significance of treatment effect. Adjustment for baseline coagulation factor levels was performed by ANCOVA.
Overall, we randomized 350 patients: 117 to UFH, 118 to 2.5 mg fondaparinux, and 115 to 5.0 mg fondaparinux. No patient was lost to follow-up. The most common reason for undergoing PCI was after presentation with an ACS (79%). Primary PCI for STEMI was performed in 0.9% of UFH patients and 1.7% in each of the fondaparinux groups.
Baseline characteristics were generally well matched between the groups, with the exception of a few chance imbalances in some key variables (Table 1). Patients in the fondaparinux groups were older than the UFH group. Compared with UFH, there were more patients in the 5.0-mg fondaparinux group with prior coronary artery bypass graft surgery (8.5% for UFH versus 12.2% for fondaparinux), intervention on saphenous vein grafts (1.2% versus 3.2%), TIMI 0 flow/total occlusions (6.1% versus 7.7%), major ST-segment depression (0.9% versus 2.6%), and preprocedural angiographic thrombus (8.5% versus 10.3%). Patients in the 2.5-mg fondaparinux group more frequently had prior stroke (2.6% for UFH versus 4.2% for fondaparinux), heart failure (2.4% versus 9.3%), diabetes (24.8% versus 26.3%), left main PCI (0% versus 1.2%), and saphenous vein graft PCI (1.2% versus 3.0%) compared with those in the UFH group. The 2.5 mg fondaparinux group also had a greater proportion of women (21% for UFH versus 25% for fondaparinux). Given these differences in baseline characteristics, adjusted HRs for main outcomes are presented. Concomitant medications were similar among the groups and are shown in Table 2.
There were fewer patients in the fondaparinux groups who experienced total bleeding compared with the UFH group (6.4% versus 7.7%; HR, 0.81; 95% CI, 0.35 to 1.84; P=0.61; Figure 1). When adjusted for baseline characteristic imbalances that were associated with this outcome (age, sex, and ST-segment depression), there was a trend toward a 22% relative risk reduction in bleeding (HR, 0.78; 95% CI, 0.34 to 1.80; P=0.56). There was no significant difference in bleeding between the 2 doses of fondaparinux and UFH (Table 3); however, there was a nonsignificant trend toward a lower rate of bleeding with the lower dose of fondaparinux compared with the higher dose (3.4% versus 9.6%; relative risk, 0.33; 95% CI, 0.10 to 1.04; P=0.06). The results were similar in patients who received a planned intravenous GPIIb/IIIa antagonist and those with no planned intravenous GPIIb/IIIa antagonist (Table 3). Major bleeding was low overall, with no significant differences among the groups. Components of the bleeding composite, including need for blood transfusion ≥2 U and intracranial hemorrhage, also were infrequent. There were nonsignificantly fewer vascular site complications, including large hematomas and pseudoaneurysms, in the fondaparinux groups compared with UFH (Table 3).
The predefined composite outcome to assess efficacy (all-cause mortality, MI, urgent revascularization, and need for bailout GPIIb/IIIa antagonist) was similar between the combined fondaparinux and UFH groups (6.0% versus 6.0%; HR, 1.02; 95% CI, 0.41 to 2.52; P=0.97; Figure 1). When adjusted for baseline characteristic imbalances that were associated with this outcome (age, sex, and ST-segment depression), there was a nonsignificant 17% relative risk reduction in the composite efficacy outcome (HR, 0.83; 95% CI, 0.33 to 2.09; P=0.69). There was no obvious dose response with respect to efficacy between the 2 fondaparinux groups. Individual components of the efficacy composite outcome were also similar among the groups, with no significant differences in the rate of events among those receiving and those not receiving a planned GPIIb/IIIa antagonist (Table 3).
Procedural success rates were high and similar across the groups: 96.3% for UFH, 96.5% for 2.5 mg fondaparinux, and 98.4% for 5.0 mg fondaparinux (Table 4). There was no significant difference among the groups in the proportion of patients with side-branch closure, dissection, or persistent residual stenosis. There was 1 case of abrupt closure and 1 case of reported angiographic thrombus in the UFH group, 3 and 6 cases respectively in the 2.5-mg fondaparinux group, and 0 and 5 cases respectively in the 5.0-mg fondaparinux group (P=NS). Despite this, the 2.5-mg fondaparinux group tended to have fewer clinical events, including periprocedural MI (7 in the UFH group versus 4 in the 2.5-mg fondaparinux group).
Fondaparinux produced a greater reduction in prothrombin F1.2, a marker of thrombin generation, at 6 and 12 hours after PCI, than did UFH (Table 5, Figure 2). There was no difference in P-selectin levels between the fondaparinux and UFH groups at either time point. There was a small increase in d-dimer levels in the 2.5-mg fondaparinux group at 6 hours but not at 12 hours. There was no increase in d-dimer levels in the UFH or 5.0-mg fondaparinux groups. There was a greater reduction in the levels of factor VIIa observed in both fondaparinux groups at 6 and 12 hours compared with UFH (Table 5). Mean anti-factor Xa levels (with fondaparinux as a standard) at 6 hours after PCI were 0.21 and 0.38 mg/L for the 2.5- and the 5.0-mg doses, respectively.
The ASPIRE results suggest that fondaparinux in doses of 2.5 and 5.0 mg was comparable to UFH for clinical safety and efficacy outcomes when given with or without planned use of GPIIb/IIIa antagonists. Fondaparinux also produced greater and more sustained suppression of thrombin generation after PCI than did UFH.
The primary objective of the study was to determine the safety of fondaparinux as assessed by rates of total bleeding between the groups. Overall, the combined doses of fondaparinux demonstrated a nonsignificantly lower bleeding rate and fewer vascular access site complications compared with UFH. Although bleeding overall in the study was relatively infrequent, the 2.5-mg fondaparinux group was associated with the lowest rate of bleeding. Our study was performed on a background of aggressive antiplatelet therapy, including GPIIb/IIIa antagonists in 58%, preprocedural clopidogrel in >87%, and aspirin in >96% of patients, all of which are effective in patients undergoing PCI but also have the potential to increase bleeding.14–17 Despite the use of these multiple antithrombotic therapies, our trial suggests that a large increase in bleeding is unlikely with the use of fondaparinux compared with a weight-adjusted dose of UFH.
The predefined efficacy composite outcome was similar between the UFH and fondaparinux groups. Unlike bleeding, there did not appear to be a dose-response relation between efficacy and the 2 doses of fondaparinux. The lower dose of fondaparinux seemed to yield the best point estimate in terms of efficacy, although the overall number of events was small. Similar findings were observed in the PENTUA (Pentasaccharide in Unstable Angina) study evaluating 4 doses of fondaparinux ranging from 2.5 up to 12 mg in patients with non-ST-segment elevation ACS (NSTEACS), wherein the 2.5-mg dose appeared to be associated with the lowest event rates.12 Also, in the PENTALYSE (synthetic PENTasaccharide as an Adjunct to fibrinoLYsis in ST-Elevation acute myocardial infarction) study evaluating 3 doses of fondaparinux in alteplase-treated STEMI patients (4, 8, and 12 mg), there did not appear be a clear dose response with respect to efficacy.11 A trial of 5 doses of fondaparinux for the prevention of venous thrombosis in patients undergoing hip replacement suggested that a 2.5-mg dose would yield the most favorable efficacy-safety profile.5 Data from those studies taken in context with the results of ASPIRE suggest that the 2.5-mg dose is the most optimal dose to study in the phase III ACS trials of fondaparinux.
The angiographic data largely support the clinical efficacy data in the trial, with high procedural success rates (96.3% for UFH, 96.5% for 2.5 mg fondaparinux, and 98.4% for 5.0 mg fondaparinux). There were no significant differences in the rates of abrupt closure, persistent residual stenosis, major side-branch occlusion, or need for a bailout GPIIb/IIIa inhibitor because of thrombotic complications between each of the fondaparinux groups and UFH. Angiographic complications occurred with similar frequency in the groups receiving and not receiving a planned GPIIb/IIIa antagonist. Suspected angiographic thrombus was reported more frequently in the fondaparinux groups, but this finding should be interpreted with some caution, as it is often confused with dissection after PCI18–22 (none of which were reported in ASPIRE). The sensitivity for correctly characterizing an intraluminal filling defect on angiography (thrombus, dissection, or both) is only 19% to 37%,16,17 suggesting that in most cases, it is misclassified. Whether the rates of angiographic thrombus are real, owing to the play of chance or to the unexpectedly low rate in the UFH group, can only be clarified in larger trials, which are currently ongoing. The fact that procedural success rates were similar in the 3 groups, with no apparent excess in clinical events or rates of abrupt closure, is reassuring.
The coagulation factor analysis demonstrated that fondaparinux was superior to UFH in inducing a sustained, dose-dependent reduction in thrombin generation up to 12 hours after PCI without resulting in an increase in bleeding and with fewer vascular access site complications. This reduction in markers of thrombin generation may reflect the longer half-life of fondaparinux compared with UFH and could potentially have benefit in reducing the rate of thrombotic events after PCI, particularly in complex or high-risk procedures. On the other hand, we did note a small increase in d-dimer levels in the 2.5-mg fondaparinux group at the 6-hour time point but not in the 5-mg fondaparinux or UFH group. The clinical significance of this finding remains elusive, as the 2.5-mg fondaparinux group tended to have the lowest rates of primary efficacy outcome. At present, there is no international standard for measurement of anti-factor Xa with fondaparinux like those that are available for UFH and LMWH. Therefore, the anti-factor Xa levels achieved with fondaparinux are not directly comparable with these other agents.
The main limitation of our pilot study was the modest sample size, which can only evaluate large differences in event rates. Nevertheless, the study was designed as an exploratory trial to assess any major hazard associated with fondaparinux and to demonstrate its feasibility in the setting of PCI before the conduct of larger studies. In this respect, the goals of the study were met. More definitive data will emerge from the Michelangelo OASIS (Organization to Assess Strategies for Ischemic Syndromes) 5 and 6 trials, evaluating fondaparinux in NSTEACS and STEMI settings, respectively. In those trials, a large proportion of patients are undergoing PCI (including primary PCI in OASIS 6) while on fondaparinux.
In this pilot, randomized comparison of fondaparinux with UFH in patients undergoing contemporary PCI, intravenous doses of 2.5 and 5.0 mg appear promising, with no excess in bleeding complications or clinical events compared with UFH. These data form the basis for further evaluation of fondaparinux in a larger number of patients in this setting.
Dr Mehta was supported by a Canadian Institutes of Health Research new investigator award. We would like to acknowledge and thank all of the site principal investigators and research coordinators who recruited patients into this trial. Their names and affiliations can be found in the online-only Data Supplement.
↵*The online-only Data Supplement, which contains information about the participating clinical centers, committees, and sponsors of ASPIRE, can be found with this article at http://www.circulationaha.org.
Smith SC Jr, Dove JT, Jacobs AK, Kennedy JW, Kereiakes K, Kern MJ, Kuntz RE, Popma JJ, Schaff HV, Williams DO. ACC/AHA guidelines for percutaneous coronary intervention: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1993 Guidelines for Percutaneous Transluminal Coronary Angioplasty). J Am Coll Cardiol. 2001; 37: 2239i–2239lxvi.
Bauer KA. Fondaparinux sodium: a selective inhibitor of factor Xa. Am J Health Syst Pharm. 2001; 58: S14–S17.
Lassen MR, Bauer KA, Eriksson BI, Turpie AG; European Pentasaccharide Elective Surgery Study (EPHESUS) Steering Committee. Postoperative fondaparinux versus preoperative enoxaparin for prevention of venous thromboembolism in elective hip-replacement surgery: a randomised double-blind comparison. Lancet. 2002; 359: 1715–1720.
Turpie AG, Bauer KA, Eriksson BI, Lassen MR; PENTATHALON 2000 Study Steering Committee. Postoperative fondaparinux versus postoperative enoxaparin for prevention of venous thromboembolism after elective hip-replacement surgery: a randomised double-blind trial. Lancet. 2002; 359: 1721–1726.
Vuillemenot A, Schiele F, Meneveau N, Claudel S, Donat F, Fontecave S, Cariou R, Samama MM, Bassand JP. Efficacy of a synthetic pentasaccharide, a pure factor Xa inhibitor, as an antithrombotic agent: a pilot study in the setting of coronary angioplasty. Thromb Haemost. 1999; 81: 214–220.
Coussement PK, Bassand JP, Convens C, Vrolix M, Boland J, Grollier G, Michels R, Vahanian A, Vanderheyden M, Rupprecht HJ, Van de Werf F; PENTALYSE Investigators. A synthetic factor-Xa inhibitor ORG31540/SR9017A) as an adjunct to fibrinolysis in acute myocardial infarction: the PENTALYSE study. Eur Heart J. 2001; 22: 1716–1724.
Simoons ML, Bobbink IW, Boland J. Gardien M, Klootwijk P, Lensing AW, Ruzyllo W, Umans VA, Vahanian A, Van De Werf F, Zeymer U, PENTUA Investigators. A dose-finding study of fondaparinux in patients with non ST-segment elevation acute coronary syndromes: the Pentasaccharide in Unstable Angina (PENTUA) study. J Am Coll Cardiol. 2004; 43: 2183–2190.
Mehta SR, Yusuf S, Peters RJG, Bertrand ME, Lewis BS, Natarajan MK, Malmberg K, Rupprecht H-J, Zhao F, Chrolavicius S, Copland I, Fox KAA, for the Clopidogrel in Unstable angina to prevent Recurrent Events trial (CURE) Investigators. Effects of pretreatment with clopidogrel and aspirin followed by long-term therapy in patients undergoing percutaneous coronary intervention: the PCI-CURE study. Lancet. 2001; 358: 527–533.
Steinhubl SR, Berger PB, Mann JT3rd, Fry ET, DeLago A, Wilmer C, Topol EJ; CREDO Investigators. Clopidogrel for the Reduction of Events During Observation Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: a randomized controlled trial. JAMA. 2002; 288: 2411–2420.
Kong DF, Califf RM, Miller DP, Moliterno DJ, White HD, Harrington RA, Tcheng JE, Lincoff AM, Hasselblad V, Topol EJ. Clinical outcomes of therapeutic agents that block the platelet glycoprotein IIb/IIIa integrin in ischemic heart disease. Circulation. 1998; 98: 2829–2835.
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.
White CJ, Ramee SR, Collins TJ, Escobar AE, Karsan A, Shaw D, Jain SP, Bass TA, Henser RR, Teirstein PS, Bonan R, Walter PD, Smalling RW. Coronary thrombi increase PTCA risk: angioscopy as a clinical tool. Circulation. 1996; 93: 253–258.