Use of Lepirudin During Percutaneous Vascular Interventions in Patients with Heparin-Induced Thrombocytopenia

Percutaneous coronary intervention is a common procedure, with over 750,000 interventions performed in the United States per year. Balloon inflation and stent deployment cause endothelial injury and plaque disruption, leading to platelet activation, thrombin generation and an inflammatory response. Aspirin, clopidogrel, platelet glycoprotein (GP) IIb/IIIa receptor inhibitors and unfractionated or low-molecular weight heparin are routinely used to counteract the effects of thrombin and the stimulated platelet population. The GP IIb/IIIa inhibitors block the final common pathway of platelet activation and the heparins in conjunction with antithrombin III attenuate the action of thrombin. Both the GP IIb/IIIa inhibitors and the heparins can lead to thrombocytopenia. Heparin-induced thrombocytopenia (HIT) has been associated with clinical thrombosis. HIT is a well-recognized clinical syndrome, defined as a 50% decrease in platelet level from baseline or an absolute platelet nadir of less than 100,000/µl.1–3 Laboratory diagnosis can be confirmed by either activation assays (serotonin release assays) or antigen assays (enzyme-linked immunosorbent assay for the identification of heparin-platelet factor 4 complexes).2–5 HIT generally occurs more than 5 days after heparin administration, but can occur almost immediately if there has been prior heparin exposure.6 The pathophysiology of HIT is explained by an immune process whereby heparin and platelet factor 4 form complexes on the platelet surface, acting as an antigen which stimulates antibody formation (HIT antibody).7–10 These heparin and platelet factor 4 complexes provide a site for antibody binding to the Fc receptor on the platelet membrane, which results in platelet activation and prothrombotic microparticle release. The activated platelets and microparticles create an intense prothrombotic state. The antiplatelet factor 4/heparin IgG antibodies may also activate endothelial cells, which localizes thrombosis to sites of endothelial injury.11 Thrombosis occurs clinically in over 50% of HIT patients,4,12–13 but has been demonstrated to occur subclinically in nearly 85% of orthopedic patients.14 Thrombotic saphenous vein graft occlusion has been reported in 66% of vein grafts for cardiac patients who develop post-operative HIT.15 HIT patients with thrombosis are identified as having heparin-induced thrombocytopenia and thrombosis syndrome (HITTS) and carry a reported mortality of up to 30%.7 The mortality and morbidity associated with HITTS can be attributed to venous thrombosis, arterial thrombosis or a combination of venous and arterial thromboses.4,7,12,13,16 HIT patients who require percutaneous intervention pose a unique clinical problem with regard to thrombin inhibition. The standard antithrombin agent used for percutaneous interventions is heparin, but heparin cannot be given to HIT patients. Hirudin is a thrombin inhibitor that binds directly and irreversibly to the catalytic site on thrombin. Hirudin is structurally and immunologically unrelated to heparin and therefore does not promote thrombosis in HIT patients. The HELVETICA study,17 designed to look at restenosis in non-HIT patients when comparing heparin to hirudin during angioplasty, showed equality at the predetermined angiographic restenosis endpoint of 7 months, but did demonstrate the superiority of hirudin at 96 hours when comparing major adverse cardiac events (MACE: 11% in heparin arm versus 6.7% in combined hirudin arms; p = 0.023). The dose of hirudin in the HELVETICA study was 40 mg intravenous bolus followed by a 0.2 mg/kg/hour hirudin drip. There was no difference in the incidence of major or minor bleeding between the 2 groups. The results of HELVETICA suggest that hirudin is at least as efficacious as heparin for anticoagulation during PTCA. Clinical evidence also exists that demonstrates the efficacy of hirudin for anticoagulation in HIT patients. Two studies18,19 that used lepirudin, a recombinant hirudin, to treat HIT patients demonstrated a decreased incidence of the combined endpoints of limb amputations, new thromboembolic complications and death in hirudin-treated HITTS patients when comparing hirudin with historic controls. These data led to the recent approval of hirudin for anticoagulation of HIT patients in the United States. During the time of this study, hirudin became the only non-heparin anticoagulant approved for clinical use in HIT or non-HIT patients. Therefore, hirudin was the only potential non-heparin anticoagulant available for percutaneous coronary intervention (PCI) in HIT patients. Furthermore, standard of care for coronary interventions during the time of this study included adjuvant use of GP IIb/IIIa inhibitors in non-HIT patients. Our rationale for anticoagulant therapy during PCI of HIT patients was to substitute lepirudin for heparin. The highly prothrombotic and proinflammatory state, in part created by the activated platelet of HIT, warranted use of adjuvant antiplatelet therapy with GP IIb/IIIa inhibitors whenever possible during percutaneous intervention. In this study, we report our experience with lepirudin, often administered in combination with a GP IIb/IIIa inhibitor, used in HIT patients undergoing percutaneous intervention. Patients were assigned to receive between 0.1 and 0.8 mg/kg intravenous bolus of lepirudin alone or in combination with tirofiban (Aggrastat®, Merck & Company, West Point, Pennsylvania), abciximab (Reopro®, Eli Lilly and Company, Indianapolis, Indiana) or eptifibatide (Integrilin®, COR Therapeutics, Inc., South San Francisco, California). MACE (death, non-fatal myocardial infarction [MI], stroke and target vessel revascularization) were recorded for each patient. Bleeding complications are also reported. Methods Eligibility criteria. To be eligible for the study, patients needed to have vascular disease requiring percutaneous intervention and a clinical or laboratory diagnosis of HIT with or without thrombosis. Informed written consent was given prior to diagnostic angiography with potential percutaneous intervention. Patients undergoing both coronary (including balloon angioplasty, stent deployment, rotoblation and Angiojet) and peripheral (including balloon angioplasty, stent deployment, inferior vena cava filter and thrombolytic therapy) interventions were eligible. Patients were excluded if there was a contraindication to lepirudin or anticoagulant therapy, were less than 21 years of age, had a bleeding diathesis or did not give written informed consent. Study protocol. Activated clotting time (ACT), activated prothrombin time (aPTT), protime (PT), total platelet count and hemoglobin were drawn prior to the administration of the drugs and prior to the performance of the intervention. Patients were not treated until platelet counts recovered to greater than 50,000/µl. The aPTT and PT had to be less than 1.5 times normal before initiation of percutaneous intervention. All ACTs were performed on the HemoTec ACT machine (Medtronic Inc., Parker, Colorado). A goal ACT of greater than 250 seconds was reached prior to proceeding with the intervention if lepirudin and a GP IIb/IIIa inhibitor were used and greater than 300 seconds if lepirudin alone was used. Periprocedural clopidogrel and aspirin were administered at the discretion of the attending physician. All coronary stents were treated with standard doses of aspirin and clopidogrel. The selection of the platelet GP IIb/IIIa inhibitor was determined by the attending interventional cardiologist and administered at full dose. Eptifibatide was administered as two 180 µg/kg boluses, ten minutes apart, and a 2 µg/kg/minute infusion for 18 hours.21 Tirofiban dosing included an initial 10 µg/kg bolus followed by an infusion of 0.15 µg/kg/minute for 18 hours.22 Patients were then assigned to 0.1–0.8 mg/kg of lepirudin given as an intravenous bolus. Lower doses (0.1–0.2 mg/kg) were used for peripheral interventions, whereas higher doses (0.4–0.8 mg/kg) were given for coronary interventions. The ACT was then used as a guide for additional dosing. Both GP IIb/IIIa inhibitor and lepirudin boluses were given prior to the beginning of the intervention. Study endpoints: Efficacy. The primary endpoints were angiographic and clinical success of the percutaneous intervention. Angiographic success was defined as a post-angioplasty minimum luminal diameter of less than 50% or a post-stent minimum luminal diameter of less than 20%. Clinical success was defined as angiographic success and freedom from MACE (death, non-fatal MI, stroke and target vessel revascularization). Non-fatal MI was defined as an increase in the serum creatine kinase MB fraction level to 3 times the upper limit of normal or new significant Q-waves in 2 or more leads at 24 hours post-procedure.20 Study endpoints: Safety. All patients underwent routine surveillance by the cardiac catheterization team at 24 hours post-procedure and at discharge. The primary safety endpoint related to hirudin use was major bleeding (defined as hemorrhage requiring transfusion greater than 2 units of packed red blood cells), intracranial hemorrhage or retroperitoneal hemorrhage. Minor bleeding and both major and minor groin complications were also monitored. Minor bleeding was defined as bleeding requiring transfusion of 2 or less units of packed red blood cells (PRBCs). A major groin complication was defined as a hematoma resulting in unexplained blood loss of greater than 3 grams of hemoglobin or a hematoma requiring surgical treatment or that prolonged hospitalization. Minor hematomas were hematomas not meeting criteria for major hematomas. Results A total of 25 patients were enrolled in the study between May 1999 and October 2000. ACTs and clinical data were collected at the time of the procedure. Data analysis was performed after 25 patients had been treated. Baseline characteristics of the patients are shown in Table 1, which shows a relatively ill patient population, as evidenced by 68% of the patients with unstable angina, 60% of patients with diabetes mellitus, 56% of patients with prior coronary artery bypass and 56% of patients with peripheral vascular disease. These clinical features suggest a high-risk population. All patients had a clinical diagnosis of HIT and 11 patients had HITTS. Twenty of the patients had laboratory confirmation of HIT with either an activation or antigen assay. The remaining 5 patients had platelet activation assays, which were negative (sensitivity with single testing %). Table 2 shows the baseline coagulation studies and post-procedure platelet counts for the patients enrolled. The aPTT time, protime and platelet counts were all normal at the time of enrollment. The patients underwent a total of 36 interventions (27 percutaneous transluminal coronary interventions, four peripheral interventions, one intravascular ultrasound, one inferior vena cava filter, one Angiojet thrombectomy and 1 selective thrombolysis). Adjuvant GP IIb/IIIa inhibitors were administered to 21 patients (13 abciximab, 7 tirofiban and 1 eptifibatide). The PCI group included 7 PCIs of the right coronary artery system, six PCIs of the circumflex/obtuse marginal system, nine PCIs of the left anterior descending system, four PCIs of saphenous vein grafts and 1 PCI of a left internal mammary artery bypass graft. Seven of the coronary interventions involved rotational atherectomy therapy prior to stent deployment. Stents were deployed in 17/27 PCIs (63%). The percutaneous transluminal angioplasty group involved interventions on 2 renal arteries, one common femoral artery and 1 popliteal artery. Endovascular stents were placed in both renal arteries and the common femoral artery. Angioplasty alone was used to treat the popliteal artery. Lepirudin bolus dosing in preparation for intervention ranged from 0.1 mg/kg to 0.8 mg/kg, with the most common dose being 0.6 mg/kg (7 patients received 0.1/0.2 mg/kg, twelve received 0.4 mg/kg, fifteen received 0.6 mg/kg and 2 received 0.8/0.9 mg/kg). No lepirudin drips were administered following initial bolus dosing. The dose response relationship between the bolus dose of lepirudin and the initial post-bolus ACT level achieved is shown in Figure 1. Both lepirudin bolus dosing at 0.4 mg/kg and 0.6 mg/kg demonstrated consistent initial post-bolus ACT values above 250 seconds. The individual ACT values for the varying lepirudin doses as observed after the initial bolus and over time are displayed in Figure 2. Only the 0.6 mg/kg dosing consistently provided adequate ACT levels (greater than 300 seconds) for interventions performed in the absence of GP IIb/IIIa inhibitors at 60 minutes. In both figures, patients receiving lepirudin with and without GP IIb/IIIa inhibitors were included due to the low numbers in each group. Efficacy. Angiographic success was achieved in all coronary and peripheral interventions. In addition, the patients undergoing IVUS, Angiojet, thrombolysis and inferior vena cava filter placement had successful procedures. Clinical success was achieved in 92% of the patients. Procedure-related MACE occurred in 1 patient. In-hospital cardiac mortality was 4%. In-hospital non-cardiac mortality was also 4%. The cardiac mortality was observed in a patient with unstable angina and peripheral vascular disease (PVD) who underwent PTCA of the left anterior descending (LAD) coronary artery. Lepirudin 0.4 mg/kg and tirofiban were used for anticoagulation in addition to 325 mg of aspirin and 300 mg of clopidogrel. The initial ACT was 239 seconds. The patient was stable for 8 hours, but prior to completion of the tirofiban infusion developed a retroperitoneal bleed and subsequent refractory hypotension and death. The non-cardiac mortality had end-stage renal disease (ESRD) and had a MI 16 days prior to angioplasty. The patient had severe 3-vessel coronary artery disease and cardiomyopathy with an ejection fraction of 30% as well as severe PVD. The patient underwent PTCA of the LAD and right coronary artery (RCA) while being supported by an intra-aortic balloon pump and mechanical ventilation. Lepirudin, at a dose of 0.2 mg/kg, was used in addition to abciximab for anticoagulation. Aspirin at 325 mg and clopidogrel 300 mg load were also given. The initial ACT after bolus was 200 seconds. Following percutaneous transluminal coronary angioplasty, the patient was hemodynamically stable and the intra-aortic balloon pump was removed with no alterations in electrocardiogram or cardiac enzymes. Subsequent to the intervention but during the same hospitalization, the patient developed sepsis and multi-organ system failure due to pneumonia and expired. Both mortalities required transfusion of greater than 4 units of packed RBCs. However, only the patient who developed the retroperitoneal bleed was deemed a procedure-related bleed. No patient developed non-fatal MI. No patient had a cerebral vascular accident. No patient required target vessel revascularization in the first 30 days. No patient required emergent or urgent coronary artery bypass grafting. Safety. Major bleeding occurred in 2 patients. The first patient required 8 units of PRBCs and 12 units of platelets for a retroperitoneal bleed. The second patient, who died of sepsis, had disseminated intravascular coagulation as a complication of sepsis and required 6 units of packed RBCs prior to death. No patients had an intracranial hemorrhage. Three patients had minor bleeding requiring a total of 4 units of PRBCs. Two patients had groin complications of moderate sized hematomas, neither of which required transfusion or surgical intervention. Discussion Heparin-induced thrombocytopenia with (HITTS) or without (HIT) thrombosis is a complex pathophysiologic event. The acuity for this event is high, with a reported mortality near 30% in HITTS patients. Heparin abstinence is usually mandatory, which is problematic in patients undergoing both arterial and venous peripheral interventions. An alternative for this patient population is to administer direct thrombin inhibitors to counteract the thrombin arm of the coagulation pathway. Multiple studies have demonstrated comparable efficacy between heparin and hirudin (lepirudin) in patients with acute coronary syndromes.23–25 The OASIS-2 investigators23 demonstrated a trend toward fewer patients with unstable angina or suspected acute MI without ST elevation suffering cardiovascular death or new MI in the lepirudin group compared to the heparin group. The incidence of major bleeding requiring transfusion was significantly higher in the lepirudin group, although there was no difference in the occurrence of life-threatening bleeds. In GUSTO IIb, patients with both ST-segment elevation and non-ST segment elevation were randomized to either 72 hours of heparin or hirudin with a predetermined endpoint of death or nonfatal MI. Recombinant hirudin showed a significant benefit at 24 hours, but this dissipated to only a trend at 30 days.24 In TIMI-9B,25 patients with acute MI receiving thrombolytic therapy were randomized to either heparin or hirudin. There was no significant difference between the 2 groups when comparing the primary endpoint of death, recurrent nonfatal MI or development of severe congestive heart failure or cardiogenic shock. There was also no significant difference in the rate of major hemorrhage. HELVETICA, a study comparing hirudin and heparin for restenosis, showed a significant reduction in major cardiac events at 96 hours in the hirudin group, but no difference in the occurrence of major bleeding or in outcomes at 7 months. These studies consistently show an equivalence between hirudin and heparin for the treatment of various acute manifestations of CAD. However, patients with HIT were not specifically evaluated and GP IIb/IIIa inhibitors were not co-administered for either the ACS indication or during PCI. Minor and major bleeding also are issues that remain problematic with direct thrombin inhibitors. The cause of bleeding is difficult to ascertain because multiple agents are used in the contemporary management of CAD patients. In this series, patients with HIT undergoing percutaneous intervention were treated with lepirudin, a recombinant hirudin. Most patients undergoing PCI also received GP IIb/IIIa inhibitors, which will contribute to bleeding. In the CURE trial,26 clopidogrel and aspirin were used and were associated with a dose-dependent increase in the incidence of major bleeding. The administration of lepirudin in patients with HIT seems to be efficacious, with a clinical success of 92% in the patients undergoing percutaneous intervention and angiographic success in 100%. No patient required target vessel revascularization in the first 30 days and no emergent or urgent coronary bypass surgery was required in any patient. Both patients with major bleeding had initial ACT levels than were not supertherapeutic and both received GP IIb/IIIa inhibitors in addition to 300 mg of clopidogrel and 325 mg of aspirin. These observations suggest that coadministration of lepirudin and GP IIb/IIIa inhibitors may predispose the patient to hemorrhage. Recent data with the hirudin analogue bivalirudin has demonstrated a significant reduction in bleeding complications when using a direct thrombin antagonist and “provisional” GP IIb/IIIa inhibitor use without a compromise in efficacy.27 Perhaps this strategy could also be applied to HIT patients undergoing PI with lepirudin. Further data are needed to assess the value of ACT for prediction of bleeding, but our experience suggests that a “low” ACT on lepirudin may not be predictive of freedom from bleeding risk. Figure 1 shows a bar graft of 0.2, 0.4 and 0.6 mg/kg dosing. One can see a dose response relationship between lepirudin dose and ACT level. The 0.6 mg/kg dose produces an average post-bolus ACT of approximately 350 seconds, which is adequate for angioplasty procedures in the absence of GP IIb/IIIa inhibitors. Dosing strategies in the range of 0.2–0.4 mg/kg appear to provide ACT values of approximately 250 seconds and should provide adequate anticoagulation in conjunction with coronary interventions performed with adjuvant GP IIb/IIIa inhibitors. The relationship between lepirudin dosing and ACT level is displayed in Figure 2. Lepirudin bolus dosing at 0.4 mg/kg or 0.6 mg/kg demonstrated consistent ACTs above 250 seconds. However, only the 0.6 mg/kg dosing provided adequate anticoagulation throughout the first 60 minutes. Prolonged procedures, which last longer than 60 minutes, should prompt periodic ACT evaluation. Doses under 0.4 mg/kg do not consistently provide ACT levels greater than 250 seconds and therefore would not be expected to provide adequate anticoagulation for coronary intervention in the absence of GP IIb/IIIa inhibition. Conclusion. The direct thrombin inhibitor lepirudin is efficacious as an anticoagulant in patients with HIT undergoing percutaneous intervention. Dosing strategies probably should include bolus dosing of 0.6 mg/kg in cases where adjuvant GP IIb/IIIa inhibitors are not planned for the intervention. If concomitant GP IIb/IIIa inhibition is planned, a lower dose of lepirudin should be considered. However, because of the potential increased risk of bleeding when an adjuvant GP IIb/IIIa inhibitor is combined with lepirudin, caution should be used when lepirudin and GP IIb/IIIa inhibitors are coadministered. A strategy of “provisional” GP IIb/IIIa inhibitor use as an adjuvant therapy to lepirudin where GP IIb/IIIa inhibitors are reserved for use only when angiographic or clinical evidence of thrombotic complications arise on lepirudin with a “therapeutic” ACT may be a more prudent approach for patients treated with longer acting, irreversible agents like lepirudin.

1. Lewis BE, Walenga JM, Wallis DE. Anticoagulation with Novastan argatroban) in patients with heparin-induced thrombocytopenia and thrombosis syndrome. Semin Thromb Hemostasis 1997;23:197–202. 2. Wallis DE, Lewis BE, Walenga JM. Failure of current strategies in the prevention of thrombosis in patients with heparin-induced thrombocytopenia: A clinician’s perspective. Semin Thromb Hemostasis 1999;25(Suppl):3–7. 3. Walenga JM, Jeske WP, Fasanella AR, et al. Laboratory diagnosis of heparin-induced thrombocytopenia. Clin Applied Thrombosis/Hemostasis 1999;5(Suppl):S21–S27. 4. Walls JT, Curtis JJ, Silver D, Boley TM. Heparin-induced thrombocytopenia in patients who undergo open heart surgery. Surgery 1990;108:686–693. 5. Izban KF, Lietz HW, Hoppensteadt DA, et al. Comparison of two PF4/heparin ELISA assays for the laboratory diagnosis of heparin-induced thrombocytopenia. Semin Thromb Hemostasis 1999;25(Suppl):51–56. 6. Greinacher A, Eichler P, Lubenow N, et al. Heparin-induced thrombocytopenia with thromboembolic complications: Meta-analysis of 2 prospective trials to assess the value of parenteral treatment with lepirudin and its therapeutic aPTT range. Blood 2000;96:846–851. 7. Lewis BE, Ferguson JJ, Grassman ED, et al. Successful coronary interventions performed with argatroban anticoagulation in patients with heparin-induced thrombocytopenia and thrombosis syndrome. J Invas Cardiol 1996;8:410–417. 8. Kelton JG, Smith JW, Warkentin TE, et al. Immunoglobulin G from patients with heparin-induced thrombocytopenia binds to a complex of heparin and platelet factor 4. Blood 1994;83:3232–3239. 9. Greinacher A, Poetzsch B, Amiral J, et al. Heparin-associated thrombocytopenia: Isolation of the antibody and characterization of a multimolecular PF4-heparin complex as the major antigen. Thromb Haem 1994;71:247–251. 10. Visentin GP, Ford SE, Scott JP, Aster RH. Antibodies from patients with heparin-induced thrombocytopenia/thrombosis are specific for platelet factor 4 complexes with heparin or heparin bound endothelial cells. J Clin Invest 1994;93:81–88. 11. Blank M, Shoenfeld Y, Tavor S, et al. Antiplatelet factor 4/heparin antibodies from patients with heparin-induced thrombocytopenia provoke direct activation of microvascular endothelial cells. Internat Immunol 2002;14:121–129. 12. Lewis BE, Grassman ED, Wrona L, Rangel Y. Novastan anticoagulation during renal stent implant in a patient with heparin-induced thrombocytopenia. Blood Coag Fibrinolysis 1997;8:54–58. 13. King DE, Kelton JG. Heparin-induced thrombocytopenia. Ann Intern Med 1984;100:535–540. 14. Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N Engl J Med 1995;332:1330–1335. 15. Liu JC, Lewis BE, Steen LH, et al. Patency of coronary artery bypass grafts in patients with heparin-induced thrombocytopenia. Am J Cardiol 2002;89:979–981. 16. Magnani HN. Heparin-induced thrombocytopenia (HIT): An overview of 230 patients treated with Organon (ORG 10172). Thromb Haemostasis 1993;70:554–561. 17. Serruys PW, Herrman JPR, Simon R, et al. A comparison of hirudin with heparin in the prevention of restenosis after coronary angioplasty. N Engl J Med 1995;333:757–763. 18. Greinacher A, Volpel H, Janssens U, et al. Recombinant hirudin (lepirudin) provides effective and safe anticoagulation in patients with heparin-induced thrombocytopenia type II: A prospective study. Circulation 1999;99:73–80. 19. Greinacher A, Volpel H, Janssens U, et al. Lepirudin (recombinant hirudin) for parenteral anticoagulation in patients with heparin-induced thrombocytopenia. Circulation 1999;100:587–593. 20. Chesebro JH, Knatterud G, Roberts R, et al. Thrombolysis in Myocardial Infarction (TIMI) Trial, Phase I: A comparison between intravenous tissue-plasminogen activator and intravenous steptokinase. Circulation 1987;76:142–154. 21. Organisation to Assess Strategies for Ischemic Syndromes (OASIS-2) Investigators. Effects of recombinant hirudin (lepirudin) compared with heparin on death, myocardial infarction, refractory angina, and revascularization procedures in patients with acute myocardial ischaemia without ST elevation: A randomized trial. Lancet 1999;353:429–438. 22. The ESPRIT Investigators. Novel dosing regimen of eptifibatide in planned coronary stent implantation (ESPRIT): A randomized, placebo-controlled trial. Lancet 2000;356:2037–2044. 23. The RESTORE Investigators. Effects of platelet glycoprotein IIb/IIIa blockade with tirofiban on adverse cardiac events in patients with unstable angina or acute myocardial infarction undergoing angioplasty. Circulation 1997;96:1445–1453. 24. Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIb Investigators. A comparison of recombinant hirudin with heparin for the treatment of acute coronary syndromes. N Engl J Med 1996;335:775–782. 25. Antman EM, et al. Hirudin in acute myocardial infarction. Thrombolysis and thrombin inhibition in Myocardial Infarction (TIMI) 9B trial. Circulation 1996;94:911–921. 26. Peters RJG, Zao F, Lewis BS, et al., for the CURE Investigators. Aspirin dose and bleeding events in the CURE study. Eur Heart J 2002;4(Suppl):510. 27. Lincoff AM, Bittl JA, Harrington RA, et al. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention. The REPLACE-2 Randomized Trial. JAMA 2003;289:853–863.