Predictors and Impact of Major Hemorrhage on Mortality Following Percutaneous Coronary Intervention from the REPLACE-2 Trial

Predictors and Impact of Major Hemorrhage on Mortality Following Percutaneous Coronary Intervention from the REPLACE-2 Trial Frederick Feit, MDa,*, Michele D. Voeltz, MDb, Michael J. Attubato, MDa, A. Michael Lincoff, MDc, Derek P. Chew, MBBSd, John A. Bittl, MDe, Eric J. Topol, MDf, and Steven V. Manoukian, MDb Patients undergoing percutaneous coronary intervention (PCI) have a significant risk of hemorrhagic complications. Predictors of major hemorrhage and its relation to mortality in PCI are not well defined. Baseline and periprocedural predictors of major hemorrhage and its impact on mortality in patients undergoing elective or urgent PCI randomly assigned to heparin plus planned glycoprotein IIb/IIIa inhibitor (GPI) versus bivalirudin plus provisional GPIs in the REPLACE-2 Trial were determined. Of 6,001 patients, 3.2% experienced a major hemorrhage. Independent baseline predictors of major hemorrhage included advanced age, female gender, impaired creatinine clearance, and anemia. Independent periprocedural predictors of major hemorrhage included treatment with heparin plus GPI, increased procedural duration, provisional use of GPI, increased time to sheath removal, length of intensive care unit stay, and use of an intra-aortic balloon pump (all p <0.05). Mortality rates were higher in patients with than without major hemorrhage at 30 days (5.1% vs 0.2%), 6 months (6.7% vs 1.0%), and 1 year (8.7% vs 1.9%; p <0.001 for all). Furthermore, major hemorrhage was an independent predictor of 1-year mortality (odds ratio 2.66, 95% confidence interval 1.44 to 4.92, p ⴝ 0.002). In conclusion, in patients undergoing elective or urgent PCI, major hemorrhage was an independent predictor of 1-year mortality. A number of baseline and periprocedural factors independently predicted major hemorrhage, including treatment with heparin plus GPI. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:1364 –1369)

Advances in antithrombin and antiplatelet therapy for patients undergoing percutaneous coronary intervention (PCI) have been associated with a decrease in major adverse cardiac events.1–5 Traditionally, this decrease in ischemic complications was accompanied by an increase in periprocedural hemorrhage, particularly with use of glycoprotein IIb/IIIa inhibitors (GPI).1,2 However, the direct thrombin inhibitor bivalirudin decreased the risk of hemorrhage while maintaining anti-ischemic efficacy.3,4,6 Retrospective and registry data indicates that hemorrhage is associated with mortality in patients undergoing PCI, emphasizing the potential importance of minimizing bleeding, as well as ischemic events.5,7 The Randomized Evaluation in PCI Linking Angiomax to Reduced Clinical Events-2 (REPLACE-2) a

Department of Medicine, New York University School of Medicine, New York, New York; bDepartment of Medicine, Emory University School of Medicine, Atlanta, Georgia; cDepartment of Medicine, Cleveland Clinic Foundation, Cleveland, Ohio; dDepartment of Cardiovascular Medicine, Flinders University, SA, Australia; eOcala Heart Institute, Munroe Regional Medical Center, Ocala, Florida; and fDivision of Cardiovascular Diseases, Scripps Clinic, La Jolla, California. Manuscript received March 20, 2007; revised manuscript received and accepted June 6, 2007. REPLACE-2 was supported by a grant from The Medicines Company, Parsippany, New Jersey. *Corresponding author: Tel.: 212-263-5656; fax: 212-263-8534. E-mail address: [email protected] (F. Feit). 0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2007.06.026

Trial showed that treatment with bivalirudin plus provisional GPI resulted in similar ischemic outcomes at 30 days, similar mortality at 1 year, and a significant decrease in incidence of hemorrhagic complications compared with heparin plus GPI in patients undergoing elective or urgent PCI.3,4 We analyzed the REPLACE-2 trial to determine baseline and periprocedural predictors of major hemorrhage and its impact on mortality. Methods Study protocol: The methods of REPLACE-2 have been reported.3 In this double-blind active-controlled trial, patients undergoing elective or urgent PCI were randomly assigned to 1 of 2 antithrombotic strategies: heparin (65U/kg bolus, maximum 7,000 U) plus abciximab (0.25mg/kg bolus, 0.125-␮g/kg/min [maximum 10 ␮g/min] infusion for 12 hours) or eptifibatide (2 boluses of 180 ␮g/kg 10 minutes apart, 2.0-␮g/kg/min infusion for 18 hours; heparin plus GPI group) versus bivalirudin (0.75-mg/kg bolus, 1.75-mg/kg/hour infusion for procedure duration) plus provisional GPI for prespecified angiographic or clinical criteria (bivalirudin group). Provisional GPIs were used in 7.2% of patients.3 Activated clotting time (ACT) was measured in a blinded fashion at study drug administration, 5 minutes after the initial bolus, and end of the procedure. In patients with 5-minute ACTs ⬍225 seconds, an additional bolus of bivalirudin (0.3 mg/kg) or heparin (20 U/kg) was administered to the bivalirudin or heparin plus GPI groups, www.AJConline.org

Coronary Artery Disease/Hemorrhagic Complications of Percutaneous Coronary Intervention

1365

Table 1 Characteristics of patients by major hemorrhage Variable

Baseline patient characteristics Mean age (yrs) Age ⱖ75 yrs Women Mean weight (kg) Body mass index ⬍25 kg/m2 Mean creatinine clearance (ml/min) Previous myocardial infarction Previous PCI Previous coronary artery bypass grafting Previous angina pectoris History of hypertension History of congestive heart failure History of diabetes mellitus Acute coronary syndrome Anemia (hemoglobin⬍13 g/dl, men; ⬍12 g/dl, women) Previous thienopyridine Periprocedural factors No. of lesions attempted ⱖ3 Randomization to heparin plus GPI group Provisional GPI use Bolus 2 for ACT ⬍225 s) Closure device Time to sheath pull ⬎6 h Length of procedure ⬎1 h Intra-aortic balloon pump

Major Hemorrhage

p Value

Yes (n ⫽ 195; 3.2%)

No (n ⫽ 5,806; 96.8%)

66.8 58/195 (29.7%) 74/195 (37.9%) 84.4 51/195 (26.2%) 80.5 66/188 (35.1%) 51/192 (26.6%) 43/194 (22.2%) 154/189 (81.5%) 136/194 (70.1%) 20/193 (10.4%) 51/195 (26.2%) 49/186 (26.3%) 68/194 (35.1%) 157/195 (80.5%)

62.5 869/5,806 (15%) 1,463/5,806 (25.2%) 87.5 1,118/5,804 (19.3%) 94.9 2,118/5,704 (37.1%) 2,037/5,782 (35.2%) 1,059/5,802 (18.3%) 4,388/5,743 (76.4%) 3,868/5,785 (66.9%) 395/5,760 (6.9%) 1,573/5,796 (27.1%) 1,302/5,717 (22.8%) 1303/5,675 (23.0%) 4,998/5,805 (86.1%)

⬍0.0001 ⬍0.0001 ⬍0.0001 0.019 0.017 ⬍0.0001 NS 0.013 NS NS NS NS NS NS ⬍0.0001 0.027

21/186 (11.3%) 124/195 (63.6%) 25/195 (12.8%) 18/174 (10.3%) 45/194 (23.2%) 56/183 (30.6%) 25/162 (15.4%) 11/176 (6.3%)

462/5,722 (8.1%) 2,884/5,806 (49.7%) 349/5,806 (6.0%) 391/5,389 (7.3%) 13,55/5,801 (23.4%) 869/5,695 (15.3%) 373/5,629 (6.6%) 24/5,431 (0.4%)

NS 0.0001 0.0001 NS NS ⬍0.0001 ⬍0.0001 ⬍0.0001

Data expressed as number/total number (percentage) unless otherwise noted.

respectively. All patients received aspirin. Clopidogrel (300 mg orally) was recommended 2 to 12 hours before PCI, followed by 75 mg/day for ⱖ30 days. Vital status was assessed at 30 days, 6 months, and 1 year. Major hemorrhage was defined as intracranial, intraocular, or retroperitoneal; overt blood loss with hemoglobin decrease ⬎3 g/dl; any hemoglobin decrease ⬎4 g/dl; or transfusion of ⱖ2 U of blood products. Minor hemorrhage was defined as overt bleeding that did not meet these criteria. Unless otherwise specified, the terms major and minor hemorrhage were according to REPLACE-2 protocol-defined criteria. Hemorrhagic events were also classified using Thrombolysis In Myocardial Infarction (TIMI) criteria.8 Anemia was defined using World Health Organization criteria (hemoglobin ⬍13 g/dl in men and ⬍12 g/dl in women).9 Statistical analysis: Discrete variables and betweengroup comparisons were presented as frequencies and percentages. Baseline predictors of major hemorrhage were identified using multivariable analysis based on the intentto-treat population. Baseline variables assessed in the logistic regression model included demographic features, clinical history, and clinical and laboratory parameters, including creatinine clearance and concomitant medications. Multivariable analysis of predictors of major hemorrhage was then performed, including such periprocedural factors as treatment received, procedural characteristics, 5-minute ACT, intra-aortic balloon pump use, and intensive care unit stay,

in addition to the baseline variables mentioned. Stepwise selection was used to identify potential predictors, which were entered into the model at p ⬍0.20 and retained at p ⬍0.05. The relation between major hemorrhage and 1-year mortality was assessed using multivariable logistic regression modeling. Variables in this model included demographics, medical history, creatinine clearance, procedural complications (30-day myocardial infarction, urgent revascularization, and major and minor hemorrhage), provisional GPIs, time to sheath removal, and treatment group. Data from the models are presented as odds ratios with confidence intervals. Results Baseline and periprocedural predictors of major hemorrhage: Of 6,001 patients from REPLACE-2, a total of 195 (3.2%) experienced a major hemorrhage. Univariable analysis indicated that a number of baseline and periprocedural factors were associated with major hemorrhage (Table 1). Multivariable analysis identified baseline predictors of major hemorrhage that included age ⱖ75 years, female gender, previous angina, anemia, and impaired creatinine clearance. A second multivariable analysis using periprocedural and baseline data showed that randomization to the heparin plus GPI group, provisional GPI use, procedure ⬎1 hour, sheath removal ⬎6 hours, length of intensive care unit stay, and use of an intra-aortic balloon

1366

The American Journal of Cardiology (www.AJConline.org)

Table 2 Independent predictors of major hemorrhage Factors Baseline predictors Age ⱖ75 yrs Gender (women vs men) Previous angina pectoris Creatinine clearance* No previous PCI Anemia Periprocedural predictors Treatment group (heparin plus GPI vs bivalirudin) Provisional GPI received Procedure duration ⬎1 h Time to sheath removal ⬎6 h Intensive care unit stay† Intra-aortic balloon pump No previous thienopyridine

Table 3 Independent predictors of 1-year mortality

Odds Ratio (95% CI)

p Value

1.48 (1.01–2.18) 1.54 (1.12–2.10) 1.59 (1.08–2.35) 1.01 (1.00–1.01) 1.59 (1.14–2.23) 1.40 (1.02–1.94)

0.045 0.007 0.02 0.006 0.007 0.040

1.97 (1.37–2.84)

0.0003

2.68 (1.59–4.51) 2.05 (1.22–3.45) 1.61 (1.06–2.45) 1.25 (1.18–1.32) 8.71 (3.43–22.07) 1.66 (1.07–2.59)

0.0002 0.007 0.024 ⬍0.0001 ⬍0.0001 0.023

* Expressed as a continuous variable, odds ratio represents risk of major hemorrhage with each 1 ml/min decrease in creatinine clearance. † Expressed as a continuous variable, odds ratio represents risk of major hemorrhage with each 1-day increase in intensive care unit stay. CI ⫽ confidence interval.

Figure 1. Mortality was significantly higher in patients with compared with those without major hemorrhage.

pump were independent periprocedural predictors of major hemorrhage (Table 2). Baseline and periprocedural predictors of mortality: Mortality was significantly higher at 30 days (5.1% vs 0.2%; p ⬍0.001), 6 months (6.7% vs 1.0%; p ⬍0.001), and 1 year (8.7% vs 1.9%; p ⬍0.0001) in patients with major hemorrhage compared with those without hemorrhage (Figure 1). Multivariable analysis of baseline and periprocedural factors showed that major hemorrhage was a powerful independent predictor of 1-year mortality (odds ratio 2.66, 95% confidence interval 1.44 to 4.92; p ⫽ 0.002). Other independent predictors of 1-year mortality included age ⱖ75 years, preprocedural anemia, lower body mass index, preprocedural left ventricular ejection fraction ⱕ50%, congestive heart failure, previous angina, periprocedural myocardial infarction (within 30 days), and urgent repeat revascularization (within 30 days; Table 3). Impact of antithrombotic regimen on major hemorrhage and mortality: Patients in the heparin plus GPI group were significantly more likely to experience major

Predictor Age ⱖ75 yrs Preprocedural anemia Body mass index 20–25 (vs ⬎25) kg/m2 Preprocedural LV ejection fraction ⱕ50% Congestive heart failure Previous angina pectoris Major hemorrhage Myocardial infarction (30-d) Urgent repeat revascularization (30-d)

Odds Ratio (95% CI)

p Value

2.28 (1.51–3.46) 2.12 (1.43–3.13) 1.64 (1.08–2.50)

0.0001 0.0002 0.007

2.15 (1.44–3.21)

0.0002

3.58 (2.27–5.65) 2.16 (1.25–3.75) 2.66 (1.44–4.92) 2.46 (1.44–4.20) 3.30 (1.36–8.00)

⬍0.0001 0.006 0.002 0.001 0.008

CI ⫽ confidence interval; LV ⫽ left ventricular.

(4.1% vs 2.4%; p ⬍0.001) and minor (25.6% vs 13.4%; p ⬍0.0001) hemorrhage than those in the bivalirudin group despite lower 5-minute median ACTs (317 vs 358 seconds; p ⬍0.001). Furthermore, there was no significant relation between major hemorrhage and ACT by quartile in either treatment group (Figure 2). Although TIMI major bleeding was similar in the heparin plus GPI and bivalirudin groups (0.9% vs 0.6%; p ⫽ 0.30), TIMI minor bleeding occurred more frequently in the heparin plus GPI group (3.0% vs 1.3%; p ⬍0.0001). Rates of specific hemorrhagic complications, such as gastrointestinal bleeding, access-site bleeding, organ hemorrhage, and transfusion, were significantly higher in the heparin plus GPI group (Table 4). In the REPLACE-2 Trial, 19 patients died within the first 30 days, 10 of whom experienced major hemorrhage. Of these 10 patients, 9 were randomly assigned to the heparin plus GPI group, and the 1 patient randomly assigned to the bivalirudin group received a provisional GPI. In 6 of these 10 patients, the major hemorrhage was causative of death (hemopericardium in 4 patients, intracranial hemorrhage in 2 patients; all occurred early on day 1). All 6 of these patients were in the heparin plus GPI group. Discussion This comprehensive analysis of major hemorrhagic complications associated with elective or urgent PCI in the REPLACE-2 Trial indicated that (1) several baseline and periprocedural factors were independent predictors of major hemorrhage; (2) major hemorrhage was a powerful independent predictor of 1-year mortality; and (3) randomization to the heparin plus GPI group was an independent predictor of major hemorrhage. This was the first prospective randomized trial to compare 2 active antithrombotic regimens in which periprocedural hemorrhage was shown to be an independent predictor of 1-year mortality. In the ⬎6,000 patients in the REPLACE-2 Trial, a number of baseline characteristics independently predicted major hemorrhage, including advanced age, female gender, impaired creatinine clearance, previous angina, and anemia. Furthermore, such periprocedural factors as randomization to the heparin plus GPI group, provisional GPI use, time to sheath removal, procedure time, and use of an intra-aortic

Coronary Artery Disease/Hemorrhagic Complications of Percutaneous Coronary Intervention

1367

Figure 2. There was no significant relation between rate of major hemorrhage and ACT quartile in either treatment group. Rates of major hemorrhage were lower for patients in the bivalirudin group compared with the heparin plus GPI group. Table 4 Hemorrhagic events by treatment group Hemorrhagic Event

Major hemorrhage Any major organ hemorrhage* Retroperitoneal Intracranial Gastrointestinal Vascular access Minor hemorrhage TIMI criteria† Major hemorrhage Minor hemorrhage Any transfusion Thrombocytopenia Platelets ⬍100,000 and decrease ⬎25% Platelets ⬍50,000

Heparin ⫹ GPI (n ⫽ 3,008)

Bivalirudin (n ⫽ 2,993)

Odds Ratio (95% CI)

p Value

n/N

%

n/N

%

124/3,008 37/3,008 16/3,008 2/3,008 18/3,008 74/3,008 771/3,008

4.1% 1.2% 0.5% 0.07% 0.6% 2.5% 25.6%

71/2,993 12/2,993 7/2,993 1/2,993 4/2,993 25/2,993 400/2,993

2.4% 0.4% 0.2% 0.03% 0.1% 0.8% 13.4%

0.57 (0.42–0.76) 0.32 (0.17–0.62) 0.44 (0.18–1.07) 0.50 (0.05–5.54) 0.22 (0.08–0.66) 0.33 (0.21–0.53) 0.45 (0.39–0.51)

0.0001 0.0004 0.06 1.00 0.004 ⬍0.0001 ⬍0.0001

26/3,008 91/3,008 76/3,008

0.9% 3.0% 2.5%

19/2,993 39/2,993 50/2,993

0.6% 1.3% 1.7%

0.73 (0.41–1.33) 0.42 (0.29–0.62) 0.66 (0.46–0.94)

0.30 ⬍0.0001 0.02

51/2,927 20/2,929

1.7% 0.7%

20/2,921 7/2,921

0.7% 0.2%

0.39 (0.23–0.65) 0.35 (0.15–0.83)

0.0002 0.0124

* Defined as gastrointestinal, genitourinary, intraocular, and retroperitoneal hemorrhage. Prespecified secondary analysis. CI ⫽ confidence interval.

†

balloon pump were independent predictors of major hemorrhage. These findings expand those from a retrospective database of 10,974 patients undergoing PCI and a registry of 24,045 patients with acute coronary syndromes from the Global Registry of Acute Coronary Events (GRACE).5,7 Furthermore, in contrast to our evaluation, which compared heparin plus GPI with bivalirudin, neither of those reports assessed hemorrhagic risk comparing 2 similarly efficacious anti-ischemic strategies. In addition, only 31% of patients in the GRACE registry underwent PCI. Therefore, the present study specifically identified baseline and periprocedural factors that could be used to predict the risk of major hemorrhage in patients undergoing elective or urgent PCI.

Historically, decreases in ischemic events achieved using potent antithrombin and antiplatelet strategies have generally been accompanied by an increase in hemorrhage.1,2,10,11 Recent evidence, including the present analysis, supports the concept that major hemorrhage may have a far more significant and sometimes delayed effect on mortality than previously appreciated.7,12 Hemodynamic effects of acute blood loss may result in cardiovascular and noncardiovascular complications. These sequlae may have a greater adverse effect on clinical outcomes in patients with baseline anemia, as suggested by a recent analysis of this population from REPLACE-2.13 Hemorrhage often necessitates cessation of antithrombotic therapy, which may lead to an in-

1368

The American Journal of Cardiology (www.AJConline.org)

Figure 3. Rates of TIMI-defined hemorrhage in the ESPRIT, EPISTENT, and REPLACE-2 trials.

crease in ischemic events. In addition, transfusions in patients with acute coronary syndromes, many of whom underwent PCI, independently predicts mortality.14 Hemorrhage also had significant economic impact. A cost-analysis of REPLACE-2 indicated that major and minor hemorrhage was associated with an increased cost of $6,300 and $396 per event, respectively.15 When the frequency of complications was considered, both major and minor hemorrhage were among the 3 most costly outcomes on a per-patient basis. This was the first large-scale randomized trial to compare hemorrhagic event rates associated with bivalirudin versus heparin plus GPI in patients undergoing elective or urgent PCI. The lower rate of hemorrhage with bivalirudin likely was due to its pharmacologic characteristics, which provide many theoretical advantages over heparin. Bivalirudin is a specific bivalent direct thrombin inhibitor.16 When bound to circulating or clot-bound thrombin, bivalirudin is slowly cleaved, freeing the active site of the thrombin molecule, allowing return of thrombin’s hemostatic functions. This naturally reversing inhibition accounts for the short half-life of the drug.17 The propensity of heparin to bind nonspecifically to plasma proteins results in poorly predictable bioavailability and variable dose response.17–19 Heparin-induced platelet activation may explain the need for concomitant use of GPIs, which amplify the risk of hemorrhage and thrombocytopenia.20,21 Comparing rates of major hemorrhage across trials is difficult because no single standardized widely accepted definition exists. Early PCI trials used the TIMI criteria, which required a ⱖ5 g/dl decrease in hemoglobin or intracranial hemorrhage for major bleeding.8 The TIMI definition was developed to quantify hemorrhagic complications in patients receiving thrombolytic therapy for ST-elevation myocardial infarction. A recent analysis by Chew et al22 concluded that TIMI major hemorrhage was not ideally suited to patients undergoing elective or urgent PCI because it may under-represent the impact of hemorrhage on mortality. In 1995, Bittl et al6 developed a contemporary PCIspecific definition for major hemorrhage, nearly identical to

that used in the REPLACE-2 trial. Although a standardized definition of major hemorrhage is still being debated, a 3 g/dl loss of hemoglobin appears to be a relevant measure of clinically significant hemorrhage in patients undergoing elective or urgent PCI. In REPLACE-2, major hemorrhage, based predominantly on this criterion, was predictive of 1-year mortality. In patients treated with heparin plus GPI, TIMI-defined hemorrhage rates in REPLACE-2 were similar to those in the Evaluation of Platelet IIb/IIIa Inhibitor for Stenting (EPISTENT) and Enhanced Suppression of the Platelet IIb/IIIa Receptor with Integrilin Therapy (ESPRIT) trials (Figure 3).2,10 This finding is not surprising given the similarities in heparin dosing in these 3 trials, which resulted in median ACTs for patients who received heparin plus GPI in REPLACE-2, EPISTENT, and ESPRIT of 317 seconds (5-minute), 314 seconds (peak ACT, stent plus abciximab group), and 273 seconds (peak ACT), respectively.2,3,10 Interestingly, in REPLACE-2, patients in the bivalirudin group had significantly higher mean ACTs than those in the heparin plus GPI group.3 Furthermore, there was no significant trend for major hemorrhage by ACT quartile in either treatment group. A recent meta-analysis of 8,369 patients from 4 large randomized PCI trials showed only a modest correlation between ACT levels and hemorrhagic complications, most notably minor bleeding.11 In the contemporary era of PCI, ACT levels did not correlate reliably with major hemorrhage.11 This analysis indicated that major hemorrhage, much like periprocedural myocardial infarction, was an independent predictor of 1-year mortality in patients undergoing elective or urgent PCI. Recognition of the predictors of hemorrhage and the importance of hemorrhagic and ischemic complications is essential to optimize outcomes in patients undergoing PCI. 1. The EPIC Investigators. Use of a monoclonal antibody directed against the platelet glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty. N Engl J Med 1994;330:956 –961.

Coronary Artery Disease/Hemorrhagic Complications of Percutaneous Coronary Intervention 2. ESPRIT Investigators. Enhanced Suppression of the Platelet IIb/IIIa Receptor with Integrilin Therapy. Novel dosing regimen of eptifibatide in planned coronary stent implantation (ESPRIT): a randomised, placebo-controlled trial. Lancet 2000;356:2037–2044. 3. Lincoff AM, Bittl JA, Harrington RA, Feit F, Kleiman NS, Jackman JD, Sarembock IJ, Cohen DJ, Spriggs D, Ebrahimi R, et al; REPLACE-2 Investigators. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention: REPLACE-2 randomized trial. JAMA 2003;289:853– 863. 4. Lincoff AM, Kleiman NS, Kereiakes DJ, Feit F, Bittl JA, Jackman JD, Sarembock IJ, Cohen DJ, Spriggs D, Ebrahimi R, et al; REPLACE-2 Investigators. Long-term efficacy of bivalirudin and provisional glycoprotein IIb/IIIa blockade vs heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary revascularization: REPLACE-2 randomized trial. JAMA 2004;292:696 –703. 5. Moscucci M, Fox KA, Cannon CP, Klein W, Lopez-Sendon J, Montalescot G, White K, Goldberg RJ. Predictors of major bleeding in acute coronary syndromes: the Global Registry of Acute Coronary Events (GRACE). Eur Heart J 2003;24:1815–1823. 6. Bittl JA, Strony J, Brinker JA, Ahmed WH, Meckel CR, Chaitman BR, Maraganore J, Deutsch E, Adelman B. Treatment with bivalirudin (Hirulog) as compared with heparin during coronary angioplasty for unstable or postinfarction angina. Hirulog Angioplasty Study Investigators. N Engl J Med 1995;333:764 –769. 7. Kinnaird TD, Stabile E, Mintz GS, Lee CW, Canos DA, Gevorkian N, Pinnow EE, Kent KM, Pichard AD, Satler LF, et al. Incidence, predictors, and prognostic implications of bleeding and blood transfusion following percutaneous coronary interventions. Am J Cardiol 2003; 92:930 –935. 8. Rao AK, Pratt C, Berke A, Jaffe A, Ockene I, Schreiber TL, Bell WR, Knatterud G, Robertson TL, Terrin ML. Thrombolysis In Myocardial Infarction (TIMI) Trial—phase I: hemorrhagic manifestations and changes in plasma fibrinogen and the fibrinolytic system in patients treated with recombinant tissue plasminogen activator and streptokinase. J Am Coll Cardiol 1988;11:1–11. 9. World Health Organization. Nutritional anaemias: report of a WHO scientific group. World Health Organ Tech Rep Ser 1968;405:5–37. 10. The Evaluation of Platelet IIb/IIIa Inhibitor for Stenting (EPISTENT) Investigators. Randomised placebo-controlled and balloon-angioplasty-controlled trial to assess safety of coronary stenting with use of platelet glycoprotein-IIb/IIIa blockade. Lancet 1998;352:87–92. 11. Brener SJ, Moliterno DJ, Lincoff AM, Steinhubl SR, Wolski KE, Topol EJ. Relationship between activated clotting time and ischemic or hemorrhagic complications: analysis of 4 recent randomized clinical

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

1369

trials of percutaneous coronary intervention. Circulation 2004; 110:994 –998. Omoigui NA, Califf RM, Pieper K, Keeler G, O’Hanesian MA, Berdan LG, Mark DB, Talley JD, Topol EJ. Peripheral vascular complications in the Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT-I). J Am Coll Cardiol 1995;26:922–930. Voeltz MD, Patel AD, Feit F, Fazel R, Lincoff AM, Manoukian SV. Effect of anemia on hemmorrhagic complications and mortality in patients undergoing percutaneous coronary intervention. Am J Cardiol 2007;99:1513–1517. Rao SV, Jollis JG, Harrington RA, Granger CB, Newby LK, Armstrong PW, Moliterno DJ, Lindblad L, Pieper K, Topol EJ, Stamler JS, Califf RM. Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA 2004;292:1555–1562. Cohen DJ, Lincoff AM, Lavelle TA, Chen HL, Bakhai A, Berezin RH, Jackman D, Sarembock IJ, Topol EJ. Economic evaluation of bivalirudin with provisional glycoprotein IIb/IIIa inhibition versus heparin with routine glycoprotein IIb/IIIa inhibition for percutaneous coronary intervention: results from the REPLACE-2 trial. J Am Coll Cardiol 2004;44:1792–1800. Carswell CI, Plosker GL. Bivalirudin: a review of its potential place in the management of acute coronary syndromes. Drugs 2002;62: 841– 870. Moen MD, Keating GM, Wellington K. Bivalirudin: a review of its use in patients undergoing percutaneous coronary intervention. Drugs 2005;65:1869 –1891. Eitzman DT, Chi L, Saggin L, Schwartz RS, Lucchesi BR, Fay WP. Heparin neutralization by platelet-rich thrombi: role of platelet factor 4. Circulation 1994;89:1523–1529. Weitz JI, Hudoba M, Massel D, Maraganore J, Hirsh J. Clot-bound thrombin is protected from inhibition by heparin-antithrombin III but is susceptible to inactivation by antithrombin III-independent inhibitors. J Clin Invest 1990;86:385–391. Sobel M, Fish WR, Toma N, Luo S, Bird K, Mori K, Kusumoto S, Blystone SD, Suda Y. Heparin modulates integrin function in human platelets. J Vasc Surg 2001;33:587–594. Xiao Z, Theroux P. Platelet activation with unfractionated heparin at therapeutic concentrations and comparisons with a low-molecularweight heparin and with a direct thrombin inhibitor. Circulation 1998; 97:251–256. Chew DP, Bhatt DL, Lincoff AM, Wolski K, Topol EJ. Clinical end point definitions after percutaneous coronary intervention and their relationship to late mortality: an assessment by attributable risk. Heart 2006;92:945–950.