Author + information
- Received September 12, 2007
- Revision received December 13, 2007
- Accepted December 17, 2007
- Published online May 6, 2008.
- Harvey D. White, DSc, FACC⁎,⁎ (, )
- Derek P. Chew, MBBS, MPH†,
- James W. Hoekstra, MD‡,
- Chadwick D. Miller, MD‡,
- Charles V. Pollack Jr, MD§,
- Frederick Feit, MD∥,
- A. Michael Lincoff, MD, FACC¶,
- Michel Bertrand, MD, FACC#,
- Stuart Pocock, PhD⁎⁎,
- James Ware, PhD††,
- E. Magnus Ohman, MD, FACC‡‡,
- Roxana Mehran, MD, FACC§§ and
- Gregg W. Stone, MD, FACC∥∥
- ↵⁎Reprint requests and correspondence:
Dr. Harvey D. White, Green Lane Cardiovascular Service, Auckland City Hospital, Private Bag 92024, Auckland 1030, New Zealand.
Objectives The aim of this study was to compare outcomes in patients receiving consistent unfractionated heparin (UFH)/enoxaparin (ENOX) therapy and in those switched at randomization to bivalirudin monotherapy.
Background Crossover between UFH and ENOX has been associated with increased adverse outcomes in patients with acute coronary syndromes. The ACUITY (Acute Catheterization and Urgent Intervention Triage strategY) trial demonstrated superior net clinical outcomes with similar rates of ischemia and significantly less major bleeding with bivalirudin monotherapy compared with UFH/ENOX plus a glycoprotein (GP) IIb/IIIa inhibitor. It is unknown if these results would be preserved in patients switched from UFH/ENOX to bivalirudin monotherapy.
Methods We compared composite ischemia, major bleeding, and net clinical outcomes at 30 days in patients receiving consistent UFH/ENOX therapy and in those switched at randomization from pre-treatment with UFH/ENOX to bivalirudin monotherapy. We also compared outcomes in patients naive to antithrombin therapy who were randomized to UFH/ENOX or bivalirudin monotherapy.
Results Two thousand one hundred thirty-seven patients received consistent UFH/ENOX (UFH n = 1,294, ENOX n = 843), and 2,078 patients pre-treated with UFH/ENOX were switched to bivalirudin. Patients switching to bivalirudin had similar rates of ischemia (6.9% vs. 7.4%, p = 0.52), less major bleeding (2.8% vs. 5.8%, p < 0.01), and improved net clinical outcomes (9.2% vs. 11.9%, p < 0.01) than those on consistent UFH/ENOX plus a GP IIb/IIIa inhibitor. Patients naive to antithrombin therapy who were administered bivalirudin (n = 1,427) had similar rates of ischemia (6.2% vs. 5.5%, p = 0.47), less major bleeding (2.5% vs. 4.9%, p < 0.001), and similar net clinical outcomes (8.0% vs. 9.4%, p = 0.17) compared with naive patients administered UFH/ENOX plus a GP IIb/IIIa inhibitor (n = 1,462).
Conclusions Switching from UFH/ENOX to bivalirudin monotherapy results in comparable ischemic outcomes and an approximately 50% reduction in major bleeding compared with consistent UFH/ENOX plus a GP IIb/IIIa inhibitor. Patients naive to antithrombin therapy administered bivalirudin monotherapy had a significant reduction in bleeding and similar rates of ischemia compared with naive patients initiated with UFH or ENOX plus a GP IIb/IIIa inhibitor.
Optimal management for moderate- and high-risk patients with non–ST-segment elevation acute coronary syndromes (NSTE-ACS) includes an invasive strategy with intensive antithrombin therapy (1). Most patients with acute coronary syndromes are currently treated with either unfractionated heparin (UFH) or enoxaparin (ENOX) initiated either in the emergency department or in a transferring hospital before coronary intervention. In recent clinical trials, such as the SYNERGY (Superior Yield of the New Strategy of ENOX, Revascularization and Glycoprotein IIb/IIIa Inhibitors) trial (2) and the OASIS-5 (Organization to Assess Strategies for Ischemic Syndromes) trial (3), 72% and 50% of patients, respectively, received clinician-selected antithrombin treatment before randomization. It is of note, however, that transitions from an upstream heparin to an alternate heparin have been associated with an increase in adverse clinical outcomes. In the SYNERGY trial, patients who crossed over after randomization from UFH to ENOX or vice versa had increased rates of death/myocardial infarction (MI) within 30 days compared with those who did not cross over (22.0% vs. 14.2%, UFH; 17.4% vs. 13.5%, ENOX). The rate of transfusion doubled with crossover (35% vs. 15.1%, UFH; 30.2% vs. 15.3%, ENOX) (4). However, because these crossovers occurred after randomization, association or causality cannot be defined. Patients may have crossed over due to ischemia or bleeding, or, alternatively, crossing over may have caused an ischemic or bleeding event (2). Consistent therapy, however, was associated with lower rates of death/MI (5). Consistent antithrombin therapy has been recommended in guidelines while changing therapy has been discouraged (6).
The ACUITY (Acute Catheterization and Urgent Intervention Triage strategY) trial (7) demonstrated that, in patients with moderate- and high-risk NSTE-ACS, treatment with bivalirudin monotherapy resulted in similar rates of composite ischemia, a 47% relative (2.7% absolute) reduction in major bleeding, and improved net clinical outcomes compared with treatment with UFH/ENOX plus a glycoprotein (GP) IIb/IIIa inhibitor. In light of recent data indicating an association between bleeding and mortality (8,9), these results suggest that bivalirudin is an attractive alternative for patients with NSTE-ACS (8). It is unknown, however, if the results observed in the overall trial would be preserved in the subgroup of patients pre-treated with either UFH or ENOX before randomization and then switched to bivalirudin monotherapy. In the present analysis of the ACUITY trial, we sought to determine whether patients could be safely switched from UFH or ENOX to bivalirudin.
In the ACUITY trial, 13,819 patients with moderate- and high-risk NSTE-ACS were randomized to receive either UFH or ENOX plus GP IIb/IIIa inhibitors or bivalirudin monotherapy with provisional use of GP IIb/IIIa inhibitors for ischemic complications (7). Entry criteria have been previously described in detail (10). In brief, patients ≥18 years of age with symptoms of unstable angina lasting ≥10 min within the preceding 24 h were eligible for enrollment if 1 or more of the following criteria were met: new ST-segment depression or transient elevation ≥1 mm; troponin I, troponin T, or creatine kinase-MB elevation; known coronary artery disease; or all 4 other Thrombolysis In Myocardial Infarction unstable angina risk criteria (11) positive. Enrollment of patients treated with UFH or a single dose of low-molecular-weight heparin before randomization was permitted. Major exclusion criteria included acute ST-segment elevation MI or shock; bleeding diathesis or major bleed within 2 weeks; or thrombocytopenia or calculated creatinine clearance <30 ml/min. The study was approved by the institutional review board or ethics committee at each participating center, and all patients signed written, informed consent.
Randomization and study protocol
Patients were assigned by a primary randomization to 1 of 3 antithrombin regimens started before angiography: heparin (either unfractionated or ENOX at site discretion) plus glycoprotein (GP) IIb/IIIa inhibitors, bivalirudin (intravenous bolus of 0.1 mg/kg and an infusion of 0.25 mg/kg/h, with an additional bolus 0.5 mg/kg and an increased infusion rate of 1.75 mg/kg/h if the patient continued on to percutaneous coronary intervention [PCI]) plus GP IIb/IIIa inhibitors, or bivalirudin monotherapy, with GP IIb/IIIa inhibitors reserved for ischemic complications. The heparin dosing regimens have been previously described (10). Patients assigned to heparin plus a GP IIb/IIIa inhibitor or bivalirudin plus a GP IIb/IIIa inhibitor were randomized again in a 2 × 2 factorial design to upstream GP IIb/IIIa inhibitor initiation in all patients immediately after randomization versus deferred GP IIb/IIIa inhibitor initiation for selective use starting in the catheterization laboratory in patients undergoing PCI.
Angiography was planned, by protocol, in all patients within 72 h after randomization. Patients then underwent either PCI, coronary artery bypass grafting (CABG), or medical management per physician discretion. Aspirin 300 to 325 mg orally or 250 to 500 mg intravenously was administered daily during the index hospitalization, and 75 to 325 mg daily was prescribed indefinitely after discharge. The choice to administer clopidogrel as well as its initial dosing and timing were left to investigator discretion per local standards, though a 300-mg or greater loading dose was required in all cases no later than 2 h after PCI. Clopidogrel 75 mg daily was strongly recommended for 1 year in all patients with coronary artery disease.
Initially, all patients in the control arm were assigned to ENOX only, but after release of the SYNERGY trial results the protocol was amended so that the control arm included both UFH and ENOX (12). Investigators at each site were required to prospectively declare and utilize a single preferred heparin agent (UFH or ENOX) for all patients unless a patient had been previously started on the alternate antithrombin agent, in which case the principle of maintaining consistent therapy applied. Switching between UFH and ENOX was not permitted. Patients receiving UFH before randomization who were switched to bivalirudin at randomization were required to wait 30 min before administration of bivalirudin. Patients receiving a single dose of ENOX who were switched to bivalirudin were required to wait 8 h before administration of bivalirudin.
Three primary end points at 30 days were analyzed: composite ischemia (death, MI, or unplanned revascularization for ischemia), non-CABG major bleeding, and net clinical outcome composed of composite ischemia or non-CABG major bleeding. Noncoronary artery bypass grafting major bleeding was defined as intracranial bleeding or intraocular bleeding, retroperitoneal bleeding, access site bleed resulting in a hematoma ≥5 cm or requiring intervention/surgery, Hgb decrease by ≥3 g/dl with an overt source or by ≥4 g/dl without an overt source, any blood product transfusion, or reoperation for bleeding. All primary end points were adjudicated by a blinded clinical events committee.
To evaluate the safety of switching to bivalirudin after pre-treatment with UFH or ENOX before randomization, we compared ischemic and bleeding outcomes in patients maintained on consistent UFH/ENOX with those for patients switched to bivalirudin monotherapy. We defined switching as a protocol-defined activity driven by randomization. Patients were defined as being on consistent therapy if their randomized antithrombin was the same as their pre-randomization antithrombin; patients were defined as having been switched if after receiving either UFH or ENOX before randomization they were subsequently randomized to bivalirudin monotherapy. In a separate exploratory analysis, we evaluated ischemic and bleeding outcomes by treatment assignment in patients who were naive to antithrombin therapy at the time of randomization (Fig. 1). Patients with high-risk features (defined as elevated cardiac biomarkers or electrocardiographic changes at presentation) as well as those triaged to a PCI strategy were also analyzed within the groups of patients receiving prior antithrombin therapy and those naive to antithrombin therapy.
To compare baseline characteristics, the chi-square test was used for categorical variables and the Wilcoxon rank sum test was used for continuous variables. Medians and interquartile ranges (or ranges) are presented for continuous variables. To evaluate potential dependencies of the safety and efficacy of bivalirudin monotherapy relative to UFH/ENOX on whether the patient was receiving pre-randomization antithrombin, interaction tests between prior therapy (receiving or not receiving antithrombin therapy before randomized study drug) and treatment assignment (UFH/ENOX vs. bivalirudin monotherapy) were conducted. Relative risk (RR) was used to make comparisons among antithrombin therapy groups. The corresponding confidence intervals (CIs) and p values were derived using normal approximation. A 2-sided value of α = 0.05 was used for all superiority testing. All analyses were performed using SAS version 9 (SAS Institute, Cary, North Carolina).
Of the 13,819 randomized patients, 4,604 patients who were randomized to bivalirudin plus GP IIb/IIIa inhibitors were excluded. The reason for excluding these patients is that in the overall ACUITY trial bivalirudin with GP IIb/IIIa inhibitors resulted in a similar rate of ischemic events as the UFH/ENOX + GP IIb/IIIa inhibitor arm (7.7% vs. 7.3%) and similar major bleeding rates (5.3% vs. 5.7%). Also excluded from the present analysis were 2,111 patients who either did not receive the randomization drug (n = 324), who were administered the randomization drug for the first time after angiography (n = 135), or who received both UFH and ENOX before or after randomization (n = 1,652). A total of 4,215 patients received prior antithrombin therapy with either UFH or ENOX before randomization. Of these, 2,137 were randomized to receive the same antithrombin plus a GP IIb/IIIa inhibitor (consistent), while 2,078 patients were randomized to receive bivalirudin (switch). There were 2,889 patients naive to antithrombin therapy at randomization, and, of these, 1,462 patients were randomized to UFH/ENOX plus a GP IIb/IIIa inhibitor and 1,427 to bivalirudin monotherapy (Fig. 1). There were 351 patients who potentially may have been excessively dosed because of lack of dose adjustment for renal dysfunction; creatinine clearance <30 ml/min for ENOX and <50 ml/min for eptifibatide.
Patients receiving prior antithrombin therapy
As shown in Table 1, patients randomized to consistent UFH/ENOX therapy were, on median, 1 year older than patients switched to bivalirudin, though more patients switched to bivalirudin had high-risk features (defined as elevated cardiac biomarkers or electrocardiogram changes at presentation); there were no other significant baseline demographic differences. At 30 days, there was no difference in composite ischemia between the 2 groups: 6.9% for patients switched to bivalirudin versus 7.4% for patients remaining on consistent UFH/ENOX (RR 0.93; 95% CI 0.75 to 1.16, p = 0.52). Major bleeding was significantly reduced by 51%: 2.8% for patients switched to bivalirudin versus 5.8% for patients remaining on consistent UFH/ENOX (RR 0.49; 95% CI 0.36 to 0.66, p < 0.01). Transfusions were also lower in the patients switched to bivalirudin versus patients remaining on consistent UFH/ENOX (1.5% vs. 2.6% [RR 0.60; 95% CI 0.39 to 0.92], p = 0.02) (Table 2).
There were 181 patients on antithrombin therapy before randomization who potentially were overdosed; eptifibatide 167, ENOX 12, and 2 both eptifibatide and ENOX. When these patients were excluded from analyses, the analyses for ischemic events were unchanged and major bleeding rates for the consistent UFH or ENOX group were 5.6% versus 2.8% for the bivalirudin group (RR 0.50, 95% CI 0.37 to 69, p < 0.0001).
Net clinical outcomes were significantly improved in patients switched to bivalirudin. In patients defined as high risk and in patients undergoing PCI, composite ischemia was similar but bleeding was significantly lower in patients switched to bivalirudin (Table 2). These analyses were not affected by exclusion of potentially overdosed patients.
Patients naive before antithrombin therapy
As shown in Table 1, patients naive to antithrombin therapy randomized to UFH/ENOX plus a GP IIb/IIIa inhibitor or to bivalirudin had similar baseline characteristics except there were more patients with a history of prior MI and prior PCI in the bivalirudin group. Net clinical outcome and composite ischemia occurred with similar frequency in the 2 groups, while major bleeding was significantly lower with bivalirudin (Table 2, Fig. 2).
In naive antithrombin patients, there were 170 who potentially were excessively dosed; 167 eptifibatide, 1 ENOX, and 2 eptifibatide plus ENOX. The results for ischemic events and net clinical outcomes were unchanged when these patients were excluded from analyses. Major bleeding rates with these excluded patients were 4.5% for the consistent UFH ENOX + GP IIb/IIIa inhibitors group versus 2.5% for the bivalirudin group (RR 0.57, 95% CI 0.38 to 0.85, p = 0.006).
Comparison of outcomes in patients receiving prior antithrombin therapy with outcomes for antithrombin-naive patients
Results of formal interaction testing indicated that there was no interdependency between prior antithrombin therapy, randomized treatment assignment, and outcome. The interaction p values for patients randomized to bivalirudin or UFH/ENOX plus a GP IIb/IIIa inhibitor and prior antithrombin therapy were not significant (composite ischemia, p = 0.34; non-CABG major bleeding, p = 0.80; net clinical outcomes, p = 0.51).
Comparisons of consistent therapy with UFH and consistent therapy with ENOX to switch to bivalirudin
Tables 3 and 4⇓⇓ show clinical outcomes according to type of heparin therapy (UFH or ENOX) in patients on prior antithrombin therapy and in patients naive to antithrombin therapy. In patients switched from either UFH or ENOX to bivalirudin, there were similar rates of composite ischemia and significant reductions in major bleeding; in patients naive to antithrombin therapy, there were similar rates of composite ischemia and significant reductions in major bleeding for those randomized to bivalirudin versus either UFH or ENOX plus a GP IIb/IIIa inhibitor.
This analysis demonstrates that patients with moderate- and high-risk acute coronary syndromes in whom an invasive strategy is planned can be safely switched from UFH or ENOX to bivalirudin monotherapy before angiography. This approach results in an approximate 50% reduction in major bleeding with similar rates of ischemia as compared with consistent therapy with UFH or ENOX plus a GP IIb/IIIa inhibitor. These findings are consistent with the overall results from the ACUITY trial. The demonstration that net clinical outcomes are improved by switching patients pre-treated with UFH or ENOX to bivalirudin is clinically relevant, since moderate and high-risk NSTE-ACS patients are often treated with either UFH or ENOX in the emergency department or at a transferring hospital before treatment with bivalirudin. Also, the recent European Society of Cardiology guidelines for the diagnosis and treatment of NSTE-ACS recommend that for PCI procedures the initial anticoagulant should be maintained during the procedure regardless of whether this treatment is UFH, ENOX, or bivalirudin (6).
The SYNERGY trial demonstrated that crossing over to an alternate heparin is associated with increased bleeding (2). This may be due to the ‘stacking’ of 2 antithrombin therapies on top of one another without a sufficient washout period, although the possibility of a chance spurious finding cannot be excluded. Algorithms for switching from UFH or ENOX to the alternate heparin have not been validated.
The results of the present analysis are consistent with previous data. A recent open-label trial analyzed 91 NSTE-ACS patients who had previously received ≥1 dose of ENOX (1 mg/kg) within 12 h before PCI and were subsequently administered bivalirudin (0.75 mg/kg bolus and 1.75 mg/kg/h) at the time of PCI. Rates of bleeding and ischemic events were similar in groups defined by the interval between the last dose of ENOX and bivalirudin administration (13). In the REPLACE-2 (Randomized Evaluation of PCI Linking Angiomax to Reduced Clinical Events) trial, among patients randomized to receive bivalirudin, there was no difference in major bleeding between patients receiving prior UFH (n = 287) or ENOX (n = 258) and patients not receiving prior antithrombin therapy (n = 2,345). Cumulative distribution curves suggest that the duration of time between discontinuation of UFH or ENOX did not affect bleeding rates (14).
The safety and efficacy of switching from prior UFH or ENOX to bivalirudin may be partly explained by the mechanism of action of the antithrombin agents. Bivalirudin directly targets thrombin, specifically inhibits both clot bound and fluid phase thrombin and has a short half-life of approximately 25 min. Bivalirudin becomes biologically inactive when it is cleaved by thrombin and then dissociates, allowing thrombin to return to normal hemostatic activity (15). In contrast, UFH and ENOX are nonspecific, indirect thrombin inhibitors with relatively longer half-lives. Combining 2 agents such as UFH and ENOX may lead to excessive bleeding due to the additive effect of 2 agents with prolonged anti Xa activity. Adding bivalirudin with its lack of Xa activity to UFH or ENOX would not increase the existing anti-Xa activity, and any incremental anti-IIa effect would be temporary given its short half-life and rapid clearance.
Study strengths and limitations
The strength of the present analysis is that switching took place through protocol-mandated randomization and was not determined by investigator preference or the occurrence of either bleeding or ischemic events. Also, baseline characteristics of patients on consistent UFH or ENOX therapy were well balanced with those of patients switched to bivalirudin monotherapy. However, our analysis was post hoc, and pre-randomization use of antithrombin therapy was not stratified.
In patients with moderate- and high-risk NSTE-ACS, this analysis suggests that switching from either UFH or ENOX to bivalirudin monotherapy before angiography results in a similar rate of composite ischemia compared with consistent treatment with UFH or ENOX plus a GP IIb/IIIa inhibitor. Further, our data suggest that switching to bivalirudin monotherapy may result in an approximate 50% reduction in major bleeding compared with remaining on either UFH or ENOX. Patients naive to antithrombin therapy who were administered bivalirudin monotherapy had a significant reduction in major bleeding with similar rates of composite ischemia compared with patients administered UFH or ENOX plus a GP IIb/IIIa inhibitor.
Dr. White has received consulting fees and lecture fees from The Medicines Company and Sanofi-Aventis and grant support from The Medicines Company, Sanofi-Aventis, Proctor & Gamble, Schering-Plough, Eli Lilly & Co., Alexion, Merck, Neuron Pharmaceuticals, GlaxoSmithKline, Pfizer, Roche, Fournier Laboratories, and Johnson & Johnson. Dr. Chew is on the Speakers' Bureau for Sanofi-Aventis and Commonwealth Serum Laboratory Biotherapeutics, Australia. Dr. Hoekstra is on the Speakers' Bureau for Bristol-Myers Squibb, Sanofi-Aventis, and Schering-Plough, and is a consultant for The Medicines Company, Sanofi-Aventis, and Schering-Plough. Dr. Pollack is a consultant for The Medicines Company, Sanofi-Aventis, Schering-Plough, and Bristol-Myers Squibb; he is also on the Speakers' Bureau for Sanofi-Aventis and Schering-Plough and receives direct research support from GlaxoSmithKline. Dr. Feit is a consultant to The Medicines Company and has stock in Johnson & Johnson, Millennium Pharmaceuticals, and The Medicines Company. Dr. Lincoff has received grant support from The Medicines Company. Dr. Bertrand is on the Advisory Board of Nycomed in Europe. Dr. Pocock has received consulting fees from The Medicines Company. Dr. Ware is a consultant to The Medicines Company, Biogen, InfraReDx, and Schering-Plough. Dr. Ohman has received consulting fees from The Medicines Company, Sanofi-Aventis, Liposcience, Inovise Medical, Response Biomedical, and Savacor, has equity interests in Medtronic and Savacor, has received lecture fees from Schering-Plough, Bristol-Myers Squibb, and Datascope, and has received grant support from Schering-Plough, Bristol-Myers Squibb, and Berlex. Dr. Mehran has received lecture fees from The Medicines Company, Cordis, and Boston Scientific. Dr. Stone has received consulting fees from The Medicines Company.
The authors would like to thank the patients, the many nurses and investigators in the ACUITY trial, and Barbara Semb, Research Secretary, Green Lane Research and Education Fund, for secretarial assistance.
For full author disclosures, please see the end of this paper.
- Abbreviations and Acronyms
- coronary artery bypass grafting
- confidence interval
- myocardial infarction
- non–ST-segment elevation acute coronary syndromes
- percutaneous coronary intervention
- relative risk
- unfractionated heparin
- Received September 12, 2007.
- Revision received December 13, 2007.
- Accepted December 17, 2007.
- American College of Cardiology Foundation
- Braunwald E.,
- Antman E.M.,
- Beasley J.W.,
- et al.
- Ferguson J.J.,
- Califf R.M.,
- Antman E.M.,
- et al.
- Cohen M.,
- Mahaffey K.W.,
- Pieper K.,
- et al.
- Bassand J.P.,
- Hamm C.W.,
- Ardissino D.,
- et al.
- Manoukian S.V.,
- Feit F.,
- Mehran R.,
- et al.
- Eikelboom J.W.,
- Mehta S.R.,
- Anand S.S.,
- Xie C.,
- Fox K.A.,
- Yusuf S.
- White H.D.,
- Kleiman N.S.,
- Mahaffey K.W.,
- et al.
- Waksman R.,
- Wolfram R.M.,
- Torguson R.L.,
- et al.