OUP user menu

Does ‘SWITCHing’ from heparin to bivalirudin monotherapy result in a bleeding benefit?

Zoë Astroulakis , Jonathan M. Hill
DOI: http://dx.doi.org/10.1093/eurheartj/sup006 C13-C18 First published online: 22 May 2009


Bleeding is the commonest non-ischaemic complication associated with percutaneous coronary intervention (PCI), leading to a substantial increase in morbidity and mortality rates. Until recently, this was accepted as an inherent risk associated with the use of antithrombotic and antiplatelet agents in the management of acute coronary syndromes (ACS). However, recent trials have shown that the direct thrombin inhibitor bivalirudin can be used safely and effectively at the time of PCI, significantly reducing rates of bleeding with equivalent rates of ischaemic complications. This article reviews the data from recent randomized controlled trials comparing the use of unfractionated/low molecular weight heparin plus glycoprotein (GP) IIb/IIIa agents with bivalirudin and presents the evidence suggesting that making the switch from heparin to bivalirudin monotherapy at the time of PCI does indeed confer a “bleeding benefit”.

  • Acute coronary syndromes
  • Percutaneous coronary intervention
  • Bleeding
  • Heparin
  • Bivalirudin


Bleeding is the commonest non-ischaemic complication observed in the management of patients with acute coronary syndromes (ACS) and has long been accepted as an inherent consequence of antithrombotic and antiplatelet treatments. However, it is becoming increasingly evident that post-procedural bleeding has a deleterious impact on morbidity and mortality outcomes. As a result, the prevention of bleeding has become as important as the prevention of ischaemic events. The European Society of Cardiology (ESC) guidelines for the management of patients with non-ST-segment elevation myocardial infarction (NSTEMI) now include the recommendation that the assessment of bleeding risk should form an important part of the management of such patients.1 In order to improve outcomes, due consideration should be given to the method of reperfusion, choice of antithrombotic therapy, and individual patient factors.

There are several factors that are associated with a higher risk of bleeding including individual patient characteristics such as older age, reduced renal function, female gender, and low body weight,2 as well as factors associated with the treatment of ACS including the use of high doses of antithrombotic agents, length of antithrombotic treatment, and the use of combinations of antithrombotic drugs.3 There are also a number of concerns and issues regarding the concept of switching between different antithrombotic drugs.

Switching antithrombins: perceptions, issues, and concerns

Most patients presenting with ACS are treated with the low-molecular-weight heparin (LMWH) enoxaparin or unfractionated heparin (UFH), started either in the emergency department or on the wards of transferring hospitals. In the Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA Guidelines (CRUSADE) National Quality Improvement Initiative, 87% of patients received either LMWH or UFH in conjunction with glycoprotein (GP) IIb/IIIa inhibitor therapy within 24 h of admission, before proceeding to early coronary angiography.4 This trial showed that the use of LMWH in conjunction with a GPIIb/IIIa inhibitor led to improved outcomes and similar bleeding risks, when compared with the use of UFH. In other recent clinical trials including the Superior Yield of the New Strategy of Enoxaparin, Revascularization, and Glycoprotein IIb/IIIa Inhibitors (SYNERGY)5 and the Fifth Organization to Assess Strategies for Ischemic Syndromes (OASIS-5)6 trials, 72 and 50% of patients, respectively, received antithrombin treatment prior to randomization and proceeding to early percutaneous coronary intervention (PCI).

Conventionally, however, PCI has taken place under the cover of UFH, and there has been an understandable reluctance to change therapy once one antithrombotic agent has been initiated, as previous data had shown switching between heparins to be associated with an increase in both ischaemic and bleeding events;5 the conclusion being that monotherapy is better.

In the SYNERGY trial, patients who switched from UFH to the LMWH enoxaparin, or vice versa, following randomization, had an increased 30 day death/myocardial infarction (MI) and blood transfusion rate when compared with those continued on monotherapy.5 However, since these were post-randomization crossovers, association or causality cannot be easily defined.

A subgroup analysis of the impact of pre-randomization antithrombin therapy on outcomes in the SYNERGY trial demonstrated that those patients who received enoxaparin monotherapy experienced fewer deaths or MI than those who received UFH monotherapy with no increase in TIMI (Thrombolysis In Myocardial Infarction) bleeding.7 However, there was a significant increase in GUSTO (Global Utilization of Streptokinase and t-PA for Occluded Arteries) severe bleeding with enoxaparin monotherapy vs. UFH.


More recently, bivalirudin has provided an alternative to the use of heparin at the time of PCI. From a pharmacological perspective, bivalirudin differs from the heparins in its mode of action, and therefore the findings of heparin switching trials may not apply in the context of bivalirudin usage in PCI. Bivalirudin is a direct thrombin inhibitor, which acts independently of the cofactor antithrombin-III and has more predictable pharmacokinetics than the heparins, therefore providing more reliable anticoagulation.8 It inhibits both clot-bound and free circulating thrombin with similar effectiveness and has been demonstrated to inhibit the activation of platelets by thrombin at very low concentrations.8 It is easy to use, with a short half-life (25 min) allowing for rapid post-procedural sheath removal, and can be quickly and effectively administered immediately upon arrival at hospital, or even during ambulance transit.

Bivalirudin has been shown to provide effective ischaemic protection, at least equivalent to the use of UFH and GP IIb/IIIa inhibitors, and also significantly reduces bleeding. Significantly reduced bleeding rates, alongside comparable short and long-term mortality rates, have been seen in a host of trials across the spectrum of ACS including the Bivalirudin Angioplasty Trial (BAT),9 the Randomized Evaluation of PCI Linking Angiomax to Reduced Clinical Events-2 (REPLACE-2),10 the Intracoronary Stenting and Antithrombotic Regimen–Rapid Early Action for Coronary Treatment 3 (ISAR-REACT 3)11 trials in patients undergoing elective and non-elective PCI, and in the Acute Catheterization and Urgent Intervention Triage strategY (ACUITY)12 trial of patients with unstable angina (UA)/NSTEMI. In the Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trial in patients with STEMI, bivalirudin was shown to reduce rates of both short- and long-term mortality.13

Switching to bivalirudin in unstable angina/non-ST-segment elevation myocardial infarction patients

The BAT trial was a double-blind randomized trial comparing bivalirudin vs. UFH in 4312 patients undergoing PCI for new onset, crescendo or rest angina, or the development of angina within 2 weeks of an MI.9 Overall, treatment with bivalirudin resulted in a 65% reduction in haemorrhage (3.5 vs. 9.7%, P < 0.001) as well as a 22% reduction in the composite of death, MI, and revascularization at 7 days (6.2 vs. 7.9%, P = 0.039) compared with UFH. Favourable ischaemic and bleeding results for bivalirudin were observed for all risk strata.

Bittl14 assessed the impact of switching antithrombin drugs in the BAT trial in a protocol pre-specified analysis of 1006 patients undergoing PCI within 1 h of UFH therapy. A switch to bivalirudin compared with UFH monotherapy resulted in a significant reduction not only in hemorrhage (11.9 vs. 4.1%, P < 0.001) but also in the composite of death, MI, and revascularization at 7 days (9.5 vs. 6.3%, P = 0.04) (Figure 1). These results were seen in both the high-risk post-MI patients, as well as in the non-MI cohort.

Figure 1

Benefit of a switch to bivalirudin vs. consistent unfractionated heparin therapy in the BAT.

The REPLACE-2 trial examined over 6000 patients undergoing urgent or elective PCI.10 Patients were randomized to receive either bivalirudin with provisional GP IIb/IIIa inhibition or UFH and planned GP IIb/IIIa inhibition. Both groups received daily aspirin and a thienopyridine for a minimum of 30 days post-PCI. Within the bivalirudin group, 7.2% received provisional GP IIb/IIIa therapy. The results showed a non-significant difference in the 30 day incidence of death/MI/urgent repeat revascularization, whereas there was a significant reduction in major bleeding rates associated with bivalirudin use (2.4 vs. 4.1%; P < 0.001).10

In the SWITCH trial, the safety of switching from enoxaparin to bivalirudin in 91 non-ST elevation ACS patients who had received ≥1 dose of enoxaparin in the 12 h preceding PCI was examined.15 Patients were divided into three groups according to the number of hours from last dose of enoxaparin to PCI (Group 1: 0–4 h, Group 2: 4–8 h, and Group 3: 8–12 h). The overall risk of major bleeding (7.7%) was the same regardless of the time between the last enoxaparin injection and administration of bivalirudin, confirming the safety of switching antithrombotic therapy prior to PCI without the need for rheological testing.

In the ACUITY trial, 13 819 patients with moderate-to high-risk NSTEMI were randomized to treatment with either UFH or enoxaparin plus planned GP IIb/IIIa inhibition, bivalirudin with provisional use of GP IIb/IIIa inhibition (used in 9% of patients), or to bivalirudin monotherapy.12 There was no statistically significant difference in the 30 day composite ischaemia endpoint (death, MI, urgent revascularization) between the UFH/enoxaparin plus GP IIb/IIIa inhibitor arm vs. the bivalirudin monotherapy arm (7.3 vs. 7.8%, P = 0.32). However, there was a significant reduction in 30 day net clinical outcome (death, MI, unplanned revascularization, and major bleeding) with bivalirudin (11.7 vs. 10.1%, P = 0.015), primarily driven by a significant reduction in major bleeding (5.7 vs. 3.0%, P < 0.0001).12

A retrospective analysis of the REPLACE-2 trial data set out to examine the safety of UFH or LMWH administered to patients prior to their study medication. Bleeding rates amongst patients treated with and without antithrombin therapy in the 48 h before study treatment were analyzed.16 In patients receiving bivalirudin, pretreatment with heparin was not associated with an increased risk of bleeding. In contrast, pretreatment with heparin in those patients receiving UFH and planned GP IIb/IIIa inhibition at the time of PCI led to a significant increase in the composite of major and non-major bleeding and blood transfusions.

White et al.17 published a paper exploring the safety of switching antithrombins in the ACUITY trial. In patients who had received prior antithrombin therapy, a switch to bivalirudin resulted in better 30 day outcomes than if the patient had remained on UFH or enoxaparin monotherapy plus a GP IIb/IIIa inhibitor (Figure 2, top). A switch to bivalirudin from UFH plus GP IIb/IIIa inhibitors vs. monotherapy resulted in similar rates of composite ischaemia (7.4 vs. 7.7%, P = 0.80), and significantly reduced major bleeding (2.8 vs. 6.3%, P < 0.01) with a resultant decrease in net clinical outcomes (9.4 vs. 12.5%, P = 0.01).17 Among patients naïve to antithrombin therapy, composite ischaemia was not statistically different from bivalirudin vs. UFH plus GP IIb/IIIa inhibitors (6.2 vs. 5.2%, P = 0.38), although again there was a significant reduction in major bleeding (2.5 vs. 4.7%, P = 0.01) and net clinical outcome (8.0 vs. 9.2%, P = 0.01) with bivalirudin (Figure 2, bottom). A similar pattern of results was observed for the comparison of enoxaparin monotherapy plus GP IIb/IIIa inhibitors vs. a switch to bivalirudin in patients who were naïve to antithrombin therapy (Figure 3).17

Figure 2

Thirty day primary outcomes with consistent unfractionated heparin plus GP IIb/IIIa inhibitors vs. switch to bivalirudin in patients who received prior antithrombin (top) and those naïve to antithrombin therapy (bottom) in the ACUITY trial.

Figure 3

Thirty day primary outcomes with consistent enoxaparin plus GP IIb/IIIa inhibitors vs. switch to bivalirudin in patients who received prior antithrombin (top) and those naïve to antithrombin therapy (bottom) in the ACUITY trial.

In ‘high-risk’ patients, defined as those with elevated cardiac biomarkers or ECG changes at baseline, and in the other patients undergoing PCI, composite ischaemia was similar with a switch to bivalirudin vs. UFH/enoxaparin monotherapy plus GP IIb/IIIa inhibitors, and yet the rates of major bleeding and net clinical outcome were significantly lower (Figure 4).17 At 1 year, there was no significant difference in mortality seen between the PCI groups.18,19

Figure 4

Thirty day and 1 year mortality with consistent unfractionated heparin/enoxaparin plus GP IIb/IIIa inhibitors therapy vs. switch the bivalirudin in patients undergoing percutaneous coronary intervention overall and high-risk percutaneous coronary intervention patients.

Switching to bivalirudin in ST-segment elevation myocardial infarction patients

The HORIZONS-AMI trial evaluated 3602 patients with STEMI undergoing primary PCI randomized to either bivalirudin with provisional use of GP IIb/IIIa inhibitors (in ∼7% of patients) or UFH plus planned use of GP IIb/IIIa inhibitors.13 In this trial, pre-procedure UFH had been administered to 65.8% of bivalirudin patients and 76.3% of UFH plus GP IIb/IIIa inhibitors patients. This practice reflects the standard of care across Europe and was anticipated in the protocol with a pre-specified stratification. A switch to bivalirudin resulted in significantly reduced rates of 30 day major bleeding irrespective of UFH pre-treatment (Figure 5).13 Major adverse cardiac events (MACE; death, MI, unplanned revascularization, and stroke) in patients who were pre-treated with UFH vs. those who were not pre-treated with UFH, did not reach statistical significance (P = 0.08), although a numerical excess of 30 day MACE was apparent when there was no UFH pre-treatment, a finding that may reflect an advantage associated with early or pre-hospital anti-thrombin administration.20

Figure 5

Thirty day MACE and major bleeding in the HORIZONS-AMI trial in patients with and without unfractionated heparin pre-treatment.


The prevention of bleeding has become fundamental to the decision-making process regarding the treatment of patients with ACS, and the careful selection of antithrombin drugs which reduce the risk of bleeding, while maintaining ischaemic complications low, is recommended. The majority of patients with ACS are treated with enoxaparin or UFH and to date there has been a reluctance on the part of clinicians to switch therapy once an agent has been initiated as data had shown that switching between heparins to be associated with an increase in both ischaemic and bleeding events. However, data from numerous recent randomized trials examining the spectrum of ACS patients support switching from heparin to bivalirudin with a reduction in bleeding rates without an increase in ischaemic complications; switching does indeed lead to a bleeding benefit.

Conflict of interest: none declared.


View Abstract