Early Stent Thrombosis and Mortality After Primary...

1

Early ST within 30 days after primary percutaneous coro-nary intervention (PCI) in patients with ST-segment–

elevation myocardial infarction (STEMI) is a serious complication, resulting in increased mortality.1–4 Three of

4 randomized trials have demonstrated an increased risk of acute (≤24 hours) ST with bivalirudin compared with heparin with either routine or provisional glycoprotein IIb/IIIa receptor inhibitor (GPI) use after primary PCI in

Background—Early stent thrombosis (ST) within 30 days after primary percutaneous coronary intervention in ST-segment–elevation myocardial infarction is a serious event. We sought to determine the predictors of and risk of mortality after early ST according to procedural antithrombotic therapy.

Methods and Results—In a patient-level pooled analysis from the Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) and European Ambulance Acute Coronary Syndrome Angiography (EUROMAX) trials, we examined 30-day outcomes in 4935 patients undergoing primary percutaneous coronary intervention with stent implantation at 188 international sites, randomized to either bivalirudin or heparin±a glycoprotein IIb/IIIa inhibitor (GPI). Early ST occurred in 100 patients (2.0%), 20 of whom (20.0%) died. Bivalirudin was associated with higher rates of early ST compared with heparin±GPI (2.5% versus 1.6%, P=0.04), because of more acute (≤24 h) ST (1.5% versus 0.2%, P<0.0001), with the risk limited to the first 4 hours after percutaneous coronary intervention. The rates of subacute (1–30 days) ST were similar with bivalirudin and heparin±GPI (1.0% versus 1.4%, P=0.24). Among patients with early ST, mortality within 30 days occurred in 4 of 60 (6.7%) bivalirudin-treated patients compared with 16 of 40 (40.0%) heparin±GPI–treated patients (adjusted hazard ratio, 0.12; 95% CI, 0.04–0.39; P=0.0004 and adjusted hazard ratio, 0.122; 95% CI, 0.04–0.39; P=0. 0004). Thus, 30-day mortality attributable to early ST occurred in 4 of 2479 (0.2%) bivalirudin-treated patients versus 16 of 2456 (0.7%) heparin±GPI–treated patients (P=0.007).

Conclusions—In the present large-scale pooled analysis from 2 multicenter randomized trials, early ST was more frequent in patients treated with bivalirudin compared with heparin±GPI because of increased ST within 4 hours after primary percutaneous coronary intervention. However, the mortality attributable to early ST was significantly lower after bivalirudin than after heparin±GPI.

Clinical Trial Registration—URL: http://www.clinicaltrials.gov. Unique identifiers: NCT00433966 (HORIZONS-AMI) and NCT01087723 (EUROMAX). (Circ Cardiovasc Interv. 2016;9:e003272. DOI: 10.1161/CIRCINTERVENTIONS.115.003272.)

Key Words: mortality ◼ myocardial infarction ◼ pharmacology ◼ stent ◼ thrombosis

© 2016 American Heart Association, Inc.

Circ Cardiovasc Interv is available at http://circinterventions.ahajournals.org DOI: 10.1161/CIRCINTERVENTIONS.115.003272

Received October 21, 2015; accepted March 28, 2016.From the Division of Cardiology, Mount Sinai Medical Center, New York, NY (G.D.D., M.M.S., R.M.); Clinical Trials Center, Cardiovascular

Research Foundation, New York, NY (G.D.D., R.M., G.W.S.); Department of Cardiology, Rigshospitalet, Copenhagen, Denmark (M.M.S., P.C.); French Alliance for Cardiovascular Trials, Département Hospitalo-Universitaire Fibrosis, Inflammation, Remodelling, Université Paris-Diderot, Paris, France (P.G.S.); INSERM U-1148, Paris, France (P.G.S.); Department of Cardiology, Hôpital Bichat, Paris, France (P.G.S.); National Heart and Lung Institute, Imperial College, Royal Brompton Hospital, London, United Kingdom (P.G.S.); Department of Medicine, Nykoebing Falster Hospital, Denmark (P.C.); Department of Cardiology, Isala Klinieken, Zwolle, The Netherlands (A.v.H.); The Medicines Company, Parsippany, NJ (J.P., D.B., E.N.D.); Division of Cardiology, Columbia University Medical Center, New York, NY (G.W.S.); and Department of Cardiology, Zealand University Hospital, Denmark (M.M.S.).

The Data Supplement is available at http://circinterventions.ahajournals.org/lookup/suppl/doi:10.1161/CIRCINTERVENTIONS.115.003272/-/DC1.Correspondence to George Dangas, MD, PhD, Cardiovascular Institute, Box 1030, Mount Sinai Medical Center, New York, NY 10029. E-mail George.

[email protected]

Early Stent Thrombosis and Mortality After Primary Percutaneous Coronary Intervention

in ST-Segment–Elevation Myocardial InfarctionA Patient-Level Analysis of 2 Randomized Trials

George D. Dangas, MD, PhD; Mikkel M. Schoos, MD, PhD; Philippe Gabriel Steg, MD; Roxana Mehran, MD; Peter Clemmensen, MD, DMSc; Arnoud van ‘t Hof, MD, PhD;

Jayne Prats, PhD; Debra Bernstein, PhD; Efthymios N. Deliargyris, MD; Gregg W. Stone, MD

Coronary Interventions

by guest on May 3, 2018

http://circinterventions.ahajournals.org/D

ownloaded from

by guest on M

ay 3, 2018http://circinterventions.ahajournals.org/

Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from

mailto:[email protected]

mailto:[email protected]

http://circinterventions.ahajournals.org/








2 Dangas et al Stent Thrombosis and Mortality After Primary PCI

STEMI, with no significant difference in the rates of sub-acute (1–30 days) ST.1–4 In the Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) and in a patient-level pooled analysis from HORIZONS-AMI and European Ambulance Acute Coronary Syndrome Angiography (EUROMAX) trials, despite a greater risk of acute and early ST with bivalirudin compared with heparin±GPI, cardiac mortality within 30 days was consistently reduced with bivalirudin.3,5 In addition, this pooled analysis from the HORIZONS and EUROMAX trials5 indicated a differential mortality risk after acute ST depend-ing on antithrombotic regimen. In this context, the extent to which the increased propensity for early ST associated with bivalirudin translates into adverse outcomes compared with heparin±GPI has not been examined in detail. We therefore sought to determine the independent predictors of acute, sub-acute, and early ST and to evaluate whether the risk of mortal-ity differed according to antithrombotic therapy use in patients who develop acute, subacute, and early ST after primary PCI in the HORIZONS-AMI and EUROMAX trials.

Methods

Patient PopulationThis study was drawn from a prespecified pooled analysis of patient-level data from the HORIZONS-AMI and EUROMAX trials.2,3 As previously described,2,3,5 both trials were international, randomized, open-label studies, randomizing patients with STEMI undergoing primary PCI 1:1 to bivalirudin or heparin with either routine of pro-visional GPI use. In brief, HORIZONS-AMI enrolled 3602 patients between March 2005 and May 2007 from 123 centers in 11 countries, among whom 3202 had a stent implanted. EUROMAX enrolled 2198 patients between March 2010 and June 2013 from 65 European sites, among whom 1733 had a stent implanted. Each study was approved by an institutional review committee at each participating center and

all subjects gave written informed consent. The present pooled study population is thus comprised of 4935 STEMI patients at risk for ST after primary PCI, including 2479 and 2456 patients who were ran-domized to bivalirudin and heparin±GPI, respectively. The baseline clinical and angiographic characteristics of patients randomized to the 2 treatment groups were well matched (Tables I and II in the Data Supplement).

TreatmentsRandomization and study drugs were initiated at the tertiary hos-pital in HORIZONS-AMI and in the transporting ambulance in EUROMAX. In both trials, patients assigned to bivalirudin received a bolus of 0.75 mg/kg, followed by an infusion of 1.75 mg/kg per hour during the PCI procedure (adjusted for renal insufficiency). In HORIZONS-AMI, the infusion was terminated immediately after successful PCI, but could be continued in case of suboptimal results after PCI. The EUROMAX protocol specified that the infusion could be routinely continued for ≤4 hours at 0.25 or 1.75 mg/kg per hour, at the investigator’s discretion. Bailout GPI use was allowed in both trials for refractory thrombotic complications during PCI.

Patients in the control arm of HORIZONS-AMI received an intra-venous heparin bolus of 60 IU/kg, with subsequent boluses targeted to an activated clotting time of 200 to 250 s, and routine use of a GPI (ei-ther abciximab or double-bolus eptifibatide). In the EUROMAX trial, patients assigned to the heparin group received either unfractionated heparin (recommended doses of 100 IU/kg without a GPI or 60 IU/kg with a GPI) or an intravenous bolus of enoxaparin (0.5 mg/kg). GPI use was left to physician preference and was categorized as ei-ther routine (started before PCI) or provisional (bailout) for refractory thrombotic complications. GPI were administered at the approved doses and regimens—abciximab or eptifibatide in HORIZONS-AMI, and abciximab, eptifibatide or tirofiban in EUROMAX. In both tri-als, patients received aspirin and a loading dose of an oral platelet P2Y12 receptor antagonist (clopidogrel only in HORIZONS-AMI, and clopidogrel, prasugrel, or ticagrelor in EUROMAX) as early as possible after first medical contact. Dual antiplatelet therapy was con-tinued for at least 1 year in both studies. Both trials were powered for 30-day primary end points. At the present time, follow-up is complete to 3 years in HORIZONS-AMI and 30 days in EUROMAX. For the present analysis, we therefore considered outcomes through 30 days.

End Points and StatisticsThe end points of interest in the present trial were 30-day all-cause mortality and ST (acute, subacute, and early). ST was defined ac-cording to the definite or probable Academic Research Consortium criteria.6 All acute ST events occurred after the primary PCI proce-dure had been completed. Events in both trials were adjudicated by an independent committee blinded to randomization allocation.

Continuous variables are presented as mean (±SD) and were com-pared by ANOVA. Categorical variables are presented as counts and percentages and were analyzed with the Cochran–Mantel–Haenszel test controlling for study. ST and mortality rates were also calculated by Kaplan–Meier methodology and compared with log-rank tests. Predictors of 30-day ST were determined by multivariable logistic regression. Models were built in a stepwise, backwards and forwards fashion, with significance levels of 0.15 and 0.10 required to allow a variable to enter and stay in the model. The following covariates were considered: age >65 years, sex, bivalirudin versus heparin±GPI, clop-idogrel versus ticagrelor/prasugrel loading dose, calculated baseline creatinine clearance ≤60 mL/min, diabetes mellitus, previous PCI, stent type, Killip class II–IV, preprocedural thrombolysis in myocar-dial infarction (TIMI) flow 0/1, and postprocedural TIMI flow 0/1. Heterogeneity between trials was examined with the Breslow–Day test. Among patients with early ST, the propensity-adjusted risk of subsequent mortality within 30 days was determined for patients treated with bivalirudin versus heparin±GPI in a Cox multivariable model, with ST treated as a time-dependent variable; all differing baseline covariates were included in the propensity model. Among variables included in the propensity score were aged >65 years, sex, clopidogrel versus ticagrelor/prasugrel loading dose, calculated

WHAT IS KNOWN

•Early ST within 30 days after primary PCI is a seri-ous complication, resulting in increased mortality.

•Three of 4 randomized trials have demonstrated an increased risk of acute (≤24 hours) ST with bivali-rudin compared with heparin with either routine or provisional GPI use after primary PCI, with no significant difference in the rates of subacute (1–30 days) ST.

WHAT THE STUDY ADDS

•The increased hazard of ST in bivalirudin-treated patients was confined to the first 4 hours after PCI; thereafter the rates of ST were similar with bivaliru-din and heparin±GPI through 30-day follow-up.

•Although more patients developed early ST after bivalirudin compared with heparin±GPI, 30-day mortality was considerably less frequent when early ST occurred in bivalirudin-treated patients. As a result, fewer patients treated with bivalirudin compared with heparin±GPI died within 30 days because of ST.



ownloaded from



baseline creatinine clearance ≤60 mL/min, diabetes mellitus, previ-ous PCI, stent type, Killip class II–IV, preprocedural TIMI flow 0/1, postprocedural TIMI flow 0/1, anemia, platelet count, current smoker, access site, hyperlipidemia, hypertension, previous coronary artery bypass graft, single-vessel disease, and previous myocardial infarc-tion. All analyses are performed on the intention-to-treat population excluding the patients who did not receive a stent and a 2-sided P value of <0.05 was considered statistically significant. All analyses were performed with SAS statistical software (SAS Institute Inc, Cary, NC).

ResultsIncidence of ST and Subsequent MortalityEarly ST developed in 100 (2.0%) of 4935 STEMI patients with stents implanted during primary PCI, including 85 (85.0%) definite and 15 (15.0%) probable ST events. Within 30 days of study enrollment, 20 patients (20.0%) with early ST died. All deaths within 30 days were adjudicated as cardiac deaths.

Predictors of STAs shown in Table 1, patients who developed early ST were more likely to have undergone a previous PCI, less likely to have hyperlipidemia and previous myocardial infarction, pre-sented with greater Killip class at baseline, had higher base-line platelet counts and were more likely to have baseline TIMI flow 0/1 at index angiography.

As shown in Figure 1, patients randomized to bivalirudin compared with heparin±GPI had a higher rate of early ST (60 versus 40 events, 2.5% versus 1.6%, P=0.04) because of a higher incidence of acute ST (1.5% versus 0.2%, P<0.0001), with similar rates of subacute ST (1.0% versus 1.4%, P=0.24). The higher rate of acute ST with bivalirudin was confined to the first 4 hours after PCI (1.1% versus 0.04%, P<0.0001), whereas there was no significant difference in acute ST rates between 4 and 24 hours with bivalirudin versus heparin±GPI (0.2% versus 0.4%, P=0.28; Figure 2). Among patients who developed early ST, those treated with bivalirudin compared with heparin±GPI were younger, had less heart failure (Killip class II–IV), and less baseline anemia, but were more likely to have baseline TIMI 0/1 flow (Table 2).

By multivariable analysis, the independent predictors of early ST were baseline TIMI grade flow 0/1 and Killip class ≥2 at presentation (Table 3). A trend toward increased early ST with bivalirudin was present, which did not reach statistical sig-nificance. Randomization to bivalirudin and baseline TIMI flow 0/1 were independently associated with acute ST, whereas only Killip class ≥2 was independently predictive of subacute ST.

Clinical Events After STDespite the increased rates of early ST with bivalirudin, 30-day mortality after early ST was significantly less frequent in bivali-rudin-treated compared with heparin±GPI–treated patients (Figure 3; Table 4). Mortality within 30 days occurred in 4 of 60 (6.7%) bivalirudin-treated patients with early ST compared with 16 of 40 (40.0%) heparin±GPI–treated patients with early ST (relative risk, 0.19; 95% CI, 0.07–0.52; P=0.0002). By propensity-adjusted Cox modeling, treatment with bivaliru-din compared with heparin±GPI was an independent predictor

Table 1. Baseline Demographics and Procedural Characteristics Stratified by the Occurrence of Stent Thrombosis

Patients With Stent

Thrombosis (n=100)

Patients Without Stent Thrombosis (n=4835) P Value

Age, y 59.7±11.1 61.0±11.9 0.31

Women 29 (29.0) 1082 (22.4) 0.12

Diabetes mellitus 18 (18.0) 714/4833 (14.8) 0.15

Hypertension 48 (48.0) 2369/4833 (49) 0.20

Hyperlipidemia 40 (40.0) 1972/4833 (40.8) 0.007

Current smoker 47 (47.5) 2240/4823 (46.4) 0.45

Previous myocardial infarction

8 (8.0) 433/4833 (9.0) 0.02

Previous PCI 12 (12.0) 431/4832 (8.9) 0.01

Previous CABG 1 (1.0) 100/4833 (2.1) 0.76

Killip class ≥2 16 (16.0) 368/4687 (7.9) 0.005

Baseline platelet count, cells/mm3×103

272.9±92.6 251.4±71.3 0.005

Baseline hemoglobin, g/dL

14.5±1.72 14.5±1.6 0.99

Anemia 10 (10.0) 514/4509 (11.4) 0.84

Baseline creatinine clearance, mL/min

94.9±37.9 94.2±35.7 0.74

≤60 mL/min 15/96 (15.6) 678/4467 (15.2) 0.96

Region

United States 19 (19.0) 664 (13.7) 0.47

Europe 65 (65.0) 3533 (73.1) 0.51

Rest of world 16 (16.0) 638 (13.2) 0.94

Aspirin use 100 (100.0) 4827/4833 (99.9) 0.71

P2Y12 inhibitor loading dose, any

97 (98.0) 4760/4833 (98.5) 0.95

Clopidogrel 84/99 (84.8) 3915/4797 (81.6) 0.26

Ticlopidine 1/99 (1.0) 13/4796 (0.3) 0.22

Prasugrel or ticagrelor

12/99 (12.1) 840/4799 (17.5) 0.21

P2Y12 inhibitor maintenance dose, any

96/99 (97.0) 4746/4830 (98.3) 0.22

Clopidogrel 83/98 (84.7) 3734/4779 (78.1) 0.61

Ticlopidine 1/98 (1.0) 46/4780 (1) 0.95

Prasugrel or ticagrelor

13/98 (13.1) 997/4783 (20.8) 0.07

Any glycoprotein IIb/IIIa inhibitor

57 (57.0) 2345/4829 (48.6) 0.16

Radial artery access site

11 (11.0) 988/4834 (20.6) 0.16

(Continued )



ownloaded from



for reduced 30-day mortality after early ST (adjusted hazard ratio, 0.12; 95% CI, 0.04–0.39; P=0.0004). The lower mortal-ity in bivalirudin-treated patients was consistent for both acute and subacute ST (Figure 4). Thus, 30-day mortality attribut-able to early ST occurred in 4 of 2479 (0.2%) bivalirudin-treated patients versus 16 of 2456 (0.7%) heparin±GPI–treated patients (P=0.007). Among patients with early ST, there were no significant differences between the 2 treatment groups in the 30-day rates of reinfarction, ischemia-driven revasculariza-tion, stroke, or major bleeding (Table 4).

A list with all deaths related to early ST is provided in Table III in the Data Supplement with listings of important patient characteristics and their propensity scores.

DiscussionIn this patient-level pooled analysis from 2 large-scale, mul-ticenter trials in which 4935 patients with STEMI undergo-ing primary PCI with stent implantation were randomized to

anticoagulation with bivalirudin versus heparin±GPI, ST occur-ring within 30 days was more frequent in bivalirudin-treated patients. This increased hazard was confined to the first 4 hours after PCI; thereafter the rates of ST were similar with bivali-rudin and heparin±GPI through 30-day follow-up. However, although more patients developed early ST after bivalirudin compared with heparin±GPI (60 versus 40 events), 30-day mor-tality was considerably less frequent when early ST occurred in bivalirudin-treated patients. As a result, fewer patients treated with bivalirudin compared with heparin±GPI, respectively, died within 30 days because of ST (4 versus 16 deaths).

Our findings of a higher hyperacute ST hazard confined to the first 4 hours after PCI with bivalirudin compared with heparin±GPI confirms previous observations.1–3 The mecha-nism of this risk is presumably lack of sufficient antithrombin activity in the early hours after stent implantation, before oral antiplatelet agent effectiveness.7,8 This risk may be amplified given the fact that bivalirudin possesses inherent antiplatelet activity, blocking both collagen-induced and thrombin-induced platelet activation.9 The risk of acute ST with immediate post-procedural bivalirudin discontinuation may be reduced with the use of heparin in the emergency room, without increasing bleeding.10 Prolongation of the post-PCI infusion of bivaliru-din at the low 0.25 mg/kg per hour dose does not reduce the risk of acute ST as shown in the EUROMAX trial.2 However, a recent randomized trial in China and a subgroup analysis from EUROMAX indicated that prolongation of the infusion at the full PCI dose of 1.75 mg/kg per hour seems effective in miti-gating the excess risk of acute ST risk.4,11 A recent trial in Italy randomized patients with acute coronary syndrome to bivaliru-din infusion during the PCI procedure only or to a prolonged infusion regimen at either the low dose of 0.25 mg/kg per hour or the PCI dose of 1.75 mg/kg per hour for at least 4 hours after the end of the procedure. The primary results did not show a difference in the composite outcome of ischemic and bleeding events, including ST, between the long or short infusion arms. However, an exploratory analysis published in the supplemental appendix of this trial revealed that within the prolonged infu-sion arm the 1.75 mg/kg per hour dose fully mitigated the risk of acute ST, whereas the 0.25 mg/kg per hour dose did not,12 which is consistent with the findings of the aforementioned trial in China and the subgroup analysis from EUROMAX.4,11

Additional independent predictors of early ST were Killip class ≥2 at presentation and baseline TIMI grade flow 0/1, representing subgroups of particular concern, although all patients with STEMI in whom stents are implanted should be considered high-risk for ST.

The central finding of this study is that the increased risk of acute ST with bivalirudin did not translate into greater mor-tality within 30 days. Conversely, in patients in whom acute or subacute ST occurred, mortality was significantly lower if this event occurred in a patient treated with bivalirudin rather than heparin±GPI (by ≈88% in a multivariable analysis which accounted for differences in age and other comorbidities in patients with ST treated with the 2 regimens). Thus, despite the fact that there were 20 more early ST events with bivaliru-din than heparin±GPI (an absolute 0.9% increase in early ST), there were in fact 12 fewer deaths because of early ST with bivalirudin (an absolute 0.5% decrease in mortality).5

Single-vessel disease

41 (41.0) 2456/4831 (50.8) 0.10

Infarct coronary artery treated with primary PCI

Left main 1 (1.0) 36/4828 (0.7) 0.72

Left anterior descending

44 (44.0) 2126/4828 (44.0) 0.97

Left circumflex 17 (17.0) 745/4828 (15.4) 0.76

Right 46 (46.0) 2149/4828 (44.5) 0.79

Drug-eluting stent implanted

63 (63.0) 3308/4835 (68.4) 0.15

Thrombectomy 8/99 (8.1) 564/4802 (11.7) 0.98

TIMI flow, pre-PCI

0/1 79/99 (79.8) 3168/4807 (65.9) 0.007

2 9/99 (9.1) 793/4807 (16.5) 0.05

3 11/99 (11.1) 839/4807 (17.5) 0.15

TIMI flow, post-PCI

0/1 3/99 (3.0) 56/4806 (1.2) 0.11

2 7/99 (7.1) 243/4806 (5.1) 0.51

3 89/99 (89.9) 4503/4806 (93.7) 0.20

Medications at discharge

ACE inhibitor or ARB

69/86 (80.2) 3722/4786 (77.8) 0.79

Aspirin 83/86 (96.5) 4675/4786 (97.7) 0.34

β-blocker 80/86 (93.0) 4398/4786 (91.9) 0.74

P2Y12 inhibitor 84/86 (97.7) 4689/4786 (98.0) 0.61

Statin 81/86 (94.2) 4563/4786 (95.3) 0.56

ACE indicates angiotensin-converting enzyme; ARB, angiotensin receptor blocker; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; and TIMI, thrombolysis in myocardial infarction.

Table 1. Continued

Patients With Stent

Thrombosis (n=100)

Patients Without Stent Thrombosis (n=4835) P Value



ownloaded from



The mechanism of observed difference in mortality when early ST occurs after bivalirudin than after heparin±GPI use in primary PCI requires further study. Among patients with early ST, there were no significant differences between the 2 treatment groups in the 30-day rates of reinfarction, ischemia-driven revascularization, stroke or major bleeding, which might otherwise explain the lower mortality in the bivaliru-din arm. These findings are, however, consistent with a recent report demonstrating that the cardiac mortality reduction with

bivalirudin compared with heparin±GPI in the HORIZONS-AMI trial was in large part because of thrombocytopenia and other nonhematologic mechanisms.13 In this study, thrombo-cytopenia did not vary significantly in patients with and with-out ST–associated mortality.

Patients who developed early ST after bivalirudin were younger and had less baseline heart failure and anemia than in patients who developed early ST after heparin±GPI, which we adjusted for. Although we cannot exclude the possibility

Figure 1. Time-to-event curves for acute (≤24 hours) and subacute (1–30 days) stent thrombosis (landmark analysis). GPI indicates glyco-protein IIb/IIIa inhibitor.

Figure 2. Time-to-event curves for acute stent thrombosis (within 24 hours of percutaneous coronary intervention) with a 4-hour landmark analysis. GPI indicates glycoprotein IIb/IIIa inhibitor.



ownloaded from



that unmeasured confounders related to the differences in base-line demographics may affect the results, several interesting explanatory mechanisms deserve consideration. In addition to the lack of effective antithrombin and antiplatelet activity early, after bivalirudin discontinuation, genetic polymorphisms

Table 3. Independent Predictors of ST

Odds Ratio (95% CI) P Value

Predictors of acute ST (≤24 h)

Bivalirudin (vs heparin±GPI) 6.89 (2.69–17.62) <0.0001

Baseline TIMI flow 0/1 2.41 (1.06–5.47) 0.04

Clopidogrel vs prasugrel or ticagrelor

0.55 (0.27–1.12) 0.10

Predictors of subacute ST (1–30 d)

Diabetes mellitus 1.82 (0.99–3.33) 0.051

Killip class ≥2 3.15 (1.67–5.93) 0.0004


Predictors of early ST (0–30 d)

Bivalirudin (vs heparin±GPI) 1.45 (0.96–2.19) 0.07

Killip class ≥2 2.31 (1.33–3.99) 0.003


CI indicates confidence interval; GPI, glycoprotein IIb/IIIa inhibitor; ST, stent thrombosis; and TIMI, thrombolysis in myocardial infarction.

Table 2. Baseline Demographic and Procedural Characteristics in Patients With Early Stent Thrombosis Stratified by Antithrombotic Regimen

Patients With Stent Thrombosis (n=100)

P ValueBivalirudin

(n=60)Heparin+GPI

(n=40)

Age, y 57.3±10.6 63.4±10.9 0.002

Women 16 (26.7) 13 (32.5) 0.51

Weight, kg 84.8±16.9 78.4±15.0 0.055

Diabetes mellitus 8 (13.3) 10 (25.0) 0.59

Hypertension 25 (41.7) 23 (57.5) 0.60

Hyperlipidemia 22 (36.7) 18 (45.0) 0.74

Current smoker 28 (46.7) 19 (48.7) 0.28

Previous myocardial infarction

6 (10.0) 2 (5.0) 0.10

Previous PCI 6 (10.0) 6 (15.0) 0.09

Previous CABG 1 (1.7) 0 (0.0) 0.37

Killip class ≥2 6 (10.3) 10 (25.0) 0.06

Baseline platelet count, cells/mm3×103

266.6±93.0 282±92.3 0.54

Baseline hemoglobin, g/dL 14.8±1.49 14.0±1.96 0.01

Anemia 3/59 (5.1) 7/37 (18.9) 0.02

Baseline creatinine clearance, mL/min

98.4±35.9 89.5±40.6 0.32

Creatinine clearance ≤60 mL/min

6/58 (10.3) 9/38 (23.7) 0.13

Region

United States 13 (21.7) 6 (15.0) 0.21

Europe 39 (65.0) 26 (65.0) 0.51

Rest of world 8 (13.3) 8 (20.0) 0.60

Aspirin use 60 (100.0) 40 (100.0) N/A

P2Y12 inhibitor loading dose, any

57/59 (96.6) 40 (100) 0.21

Clopidogrel 47/59 (79.7) 37 (92.5) 0.24

Ticlopidine 1/59 (1.7) 0 (0.0) 0.26

Prasugrel or ticagrelor 9/59 (15.3) 3 (7.5) 0.35

P2Y12 inhibitor maintenance dose, any

57/59 (96.6) 39 (97.5) 0.85

Clopidogrel 46/58 (79.7) 37 (92.5) 0.20

Ticlopidine 1/58 (1.7) 0 (0.0) 0.37

Prasugrel or ticagrelor 11/58 (19.0) 2 (5.0) 0.07

Any GPI 18 (30) 39 (97.5) <0.0001

Radial artery access site 8 (13.3) 3 (7.5) 0.85

Single-vessel disease 23 (38.3) 18 (45) 0.48


Left main 1 (1.7) 0 (0.0) 0.55

Left anterior descending

23 (38.3) 21 (52.5) 0.16

(Continued )

Left circumflex 7 (11.7) 10 (25.0) 0.17

Right 30 (50) 16 (40.0) 0.41

Thrombectomy 6 (10.0) 2/39 (5.1) 0.81

TIMI flow, pre-PCI

0/1 49 (81.7) 30/39 (76.9) 0.45

2 6 (10.0) 3/39 (7.7) 0.81

3 5 (8.3) 6/39 (15.4) 0.24

TIMI flow, post-PCI

0/1 2 (3.3) 1/39 (2.6) 0.70

2 4 (6.7) 3/39 (7.7) 0.94

3 54 (90.0) 35/39 (89.7) 0.78


ACE inhibitor or ARB 44/56 (78.6) 25 (83.3) 0.53

Aspirin 56/56 (100.0) 27 (90.0) 0.025

β-blocker 52/56 (92.9) 28 (93.3) 0.88

P2Y12 inhibitor 55/56 (98.2) 29 (96.7) 0.73

Statin 54/56 (96.4) 27 (90.0) 0.24

ACE indicates angiotensin-converting enzyme; ARB, angiotensin receptor blocker; CABG, coronary artery bypass graft; GPI, glycoprotein IIb/IIIa inhibitor; PCI, percutaneous coronary intervention; and TIMI, thrombolysis in myocardial infarction.

Table 2. Continued

Patients With Stent Thrombosis (n=100)

P ValueBivalirudin

(n=60)Heparin+GPI

(n=40)



ownloaded from



may contribute to the ST risk in bivalirudin-treated patients. For example, Collet et al14 demonstrated a strong association between the presence of the CYP2C19*2 allelic variant and recurrent thrombotic coronary events in young clopidogrel-treated patients (age <45 years). The nature of the thrombus might also be different in patients treated with bivalirudin compared with heparin±GPI. Thrombi in STEMI are mainly composed of fibrin with platelets, erythrocytes, cholesterol crystals, and leukocytes, with the fibrin and leukocyte com-ponents increasing and the platelet component decreasing with greater ischemic time.15,16 In a recent study using optical coherence tomography, patients undergoing PCI and treated with bivalirudin compared with heparin had similar in-stent thrombus volume and length, but less platelet-rich white thrombus.17 Augmented platelet–leukocyte aggregation in vivo has also been reported from the blood of patients treated with heparin+GPI compared with bivalirudin.18 These data are fur-thermore consistent with the greater frequency of immunologi-cally mediated thrombocytopenia with heparin+GPI compared with bivalirudin,13 which could contribute to platelet clumping

and ST that may be refractory to treatment in patients already receiving GPI. Conversely, the platelet–leukocyte depleted red thrombus that may be more frequent with bivalirudin is known to be more susceptible to endogenous or pharmacological fibri-nolysis,19 and thus may be more responsive to repeat PCI with potent P2Y12 inhibitors±bailout GPI with lesser impact on clinical outcomes.

Finally, bivalirudin may have additional attributes which possibly contribute to survival in the event of ST. Previous studies have shown direct thrombin inhibitors to reduce inflam-mation,20,21 apoptosis,21,22 and myocardial infarct size,20 and thereby potentially improve myocardial function21 and survival. Infarct size tended to be reduced in bivalirudin-treated patients in the HORIZONS-AMI magnetic resonance substudy,23 and greater coronary flow reserve was noted in patients treated with bivalirudin compared with heparin+GPI in A Randomized Trial to Evaluate the Relative Protection Against Post-PCI Microvascular Dysfunction and Post-PCI Ischemia Among Anti-Platelet and Anti-Thrombotic Agents–Thrombolysis In Myocardial Infarction-30 (PROTECT-TIMI 30) trial.24

Figure 3. Kaplan–Meier estimates of 30-day mortality in patients who developed stent thrombosis (ST) according to the timing of the ST event and antithrombin regimen randomization. Note: 1 bivalirudin-treated patient had both an acute and subacute ST. CI indicates confidence interval; GPI, glyco-protein IIb/IIIa inhibitor; and RR, relative risk.

Table 4. Adverse Events Within 30 Days in Patients With Early Stent Thrombosis Stratified by Antithrombotic Regimen

Bivalirudin (n=60) Heparin+GPI (n=40) Relative Risk (95% CI) P Value

Death 4 (6.7) 16 (40.0) 0.19 (0.07–0.52) 0.0002

Cardiac death 4 (6.7) 16 (40.0) 0.19 (0.07–0.52) 0.0002

Reinfarction 36 (60) 26 (65) 0.91 (0.67–1.25) 0.57

Ischemia-driven revascularization 53 (88.3) 28 (70.0) 1.22 (0.98–1.50) 0.056

Stroke 2 (3.3) 1 (2.5) 1.58 (0.15–16.71) 0.70

Major bleeding 12 (20) 8 (20.0) 1.15 (0.52–2.53) 0.74

Death or major bleeding* 15 (25.0) 22 (55.0) 0.52 (0.31–0.85) 0.008

Death, reinfarction or major bleeding* 44 (73.3) 37 (92.5) 0.82 (0.68–0.98) 0.03

Thrombocytopenia 3 (5.2) 4 (10) 0.56 (0.14–2.21) 0.41

CI indicates confidence interval; and GPI, glycoprotein IIb/IIIa inhibitor.*Not related to coronary artery bypass graft surgery.



ownloaded from



LimitationsDespite adjustment for covariates known to be predictive of ST and early mortality after primary PCI, we cannot claim causal-ity about our observation that mortality was lower in bivali-rudin-treated compared with heparin GPI–treated patients with early ST. Although the ST and mortality events were adjudicated blinded to treatment and the differences in mor-tality were statistically robust, unmeasured cofounders may call into question the causal nature of the observed relation-ships. More definitive conclusions warrant additional study. Most control group patients were treated with heparin+GPI, whether the results would have been similar with heparin alone is unknown.

Angiograms of the ST episodes and subsequent assess-ments of ST-segment resolution, post-PCI biomarkers, left ventricular ejection fraction, and infarct size were not avail-able for the present analysis, and thus we cannot state whether the procedural treatments of early ST episodes were more suc-cessful in the bivalirudin group, which might have contributed to their improved prognosis. Finally, follow-up of the present analysis was limited to 30 days (the current duration of fol-low-up from EUROMAX). However, ST and mortality rates tended to be greater in patients treated with heparin±GPI com-pared with bivalirudin in the HORIZONS-AMI trial between 30 days and 3 years,25 and it is thus unlikely that the lower absolute mortality after ST observed with bivalirudin in this study would be reversed with further follow-up.

ConclusionsIn the present patient-level pooled analysis from 2 large-scale, randomized, multicenter trials, the risk of ST within 30 days was higher among patients treated with bivalirudin compared with heparin±GPI because of a greater hazard of ST within the first 4 hours after PCI. However, mortality after early ST was

substantially less in patients randomized to bivalirudin com-pared with heparin±GPI, and as a result, among all 4935 ran-domized patients bivalirudin resulted in fewer deaths within 30 days because of ST.

Sources of FundingThe HORIZONS-AMI trial was sponsored by the Cardiovascular Research Foundation, with unrestricted grant support from Boston Scientific Corporation and The Medicines Company. The Medicines Company funded the EUROMAX trial.

DisclosuresDr Dangas received speaker honoraria from The Medicines Company. Dr Steg received speaker and consultant honoraria from The Medicines Company, Eli Lilly, and Daiichi-Sankyo. Dr Mehran received speaker honoraria from The Medicines Company and is an advisory board member of The Medicines Company, Boston Scientific, Eli Lilly, and Daiichi-Sankyo. Dr Clemmensen received research grant and con-sultant honoraria from The Medicines Company, Eli Lilly, Daiichi-Sankyo, AstraZeneca, and Bayer. Dr van ‘t Hof received speakers honoraria from The Medicines Company and Daiichi-Sankyo. Drs Prats, Bernstein, and Deliargyris are full-time employees of The Medicines Company. Dr Stone received speaker and consultant hono-raria from Boston Scientific, Eli Lilly, and Daiichi-Sankyo. The other authors report no conflicts.

References 1. Shahzad A, Kemp I, Mars C, Wilson K, Roome C, Cooper R, Andron

M, Appleby C, Fisher M, Khand A, Kunadian B, Mills JD, Morris JL, Morrison WL, Munir S, Palmer ND, Perry RA, Ramsdale DR, Velavan P, Stables RH; HEAT-PPCI Trial Investigators. Unfractionated heparin versus bivalirudin in primary percutaneous coronary intervention (HEAT-PPCI): an open-label, single centre, randomised controlled trial. Lancet. 2014;384:1849–1858. doi: 10.1016/S0140-6736(14)60924-7.

2. Steg PG, van ‘t Hof A, Hamm CW, Clemmensen P, Lapostolle F, Coste P, Ten Berg J, Van Grunsven P, Eggink GJ, Nibbe L, Zeymer U, Campo dell’ Orto M, Nef H, Steinmetz J, Soulat L, Huber K, Deliargyris EN, Bernstein D, Schuette D, Prats J, Clayton T, Pocock S, Hamon M, Goldstein P;

Figure 4. Time-to-event curves for mortality after acute and subacute stent thrombosis (ST). P values are for the comparison of ST rates in the acute and subacute periods for patients treated with bivalirudin vs heparin±glycoprotein IIb/IIIa inhibitor (GPI).



ownloaded from



EUROMAX Investigators. Bivalirudin started during emergency trans-port for primary PCI. N Engl J Med. 2013;369:2207–2217. doi: 10.1056/NEJMoa1311096.

3. Stone GW, Witzenbichler B, Guagliumi G, Peruga JZ, Brodie BR, Dudek D, Kornowski R, Hartmann F, Gersh BJ, Pocock SJ, Dangas G, Wong SC, Kirtane AJ, Parise H, Mehran R; HORIZONS-AMI Trial Investigators. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med. 2008;358:2218–2230. doi: 10.1056/NEJMoa0708191.

4. Han Y, Guo J, Zheng Y, Zang H, Su X, Wang Y, Chen S, Jiang T, Yang P, Chen J, Jiang D, Jing Q, Liang Z, Liu H, Zhao X, Li J, Li Y, Xu B, Stone GW; BRIGHT Investigators. Bivalirudin vs heparin with or with-out tirofiban during primary percutaneous coronary intervention in acute myocardial infarction: the BRIGHT randomized clinical trial. JAMA. 2015;313:1336–1346. doi: 10.1001/jama.2015.2323.

5. Stone GW, Mehran R, Goldstein P, Witzenbichler B, Van’t Hof A, Guagliumi G, Hamm CW, Généreux P, Clemmensen P, Pocock SJ, Gersh BJ, Bernstein D, Deliargyris EN, Steg PG. Bivalirudin versus heparin with or without glycoprotein IIb/IIIa inhibitors in patients with STEMI under-going primary percutaneous coronary intervention: pooled patient-level analysis from the HORIZONS-AMI and EUROMAX trials. J Am Coll Cardiol. 2015;65:27–38. doi: 10.1016/j.jacc.2014.10.029.

6. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW; Academic Research Consortium. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation. 2007;115:2344–2351. doi: 10.1161/CIRCULATIONAHA.106.685313.

7. Alexopoulos D, Xanthopoulou I, Gkizas V, Kassimis G, Theodoropoulos KC, Makris G, Koutsogiannis N, Damelou A, Tsigkas G, Davlouros P, Hahalis G. Randomized assessment of ticagrelor versus prasug-rel antiplatelet effects in patients with ST-segment-elevation myocar-dial infarction. Circ Cardiovasc Interv. 2012;5:797–804. doi: 10.1161/CIRCINTERVENTIONS.112.972323.

8. Parodi G, Valenti R, Bellandi B, Migliorini A, Marcucci R, Comito V, Carrabba N, Santini A, Gensini GF, Abbate R, Antoniucci D. Comparison of prasugrel and ticagrelor loading doses in ST-segment elevation myo-cardial infarction patients: RAPID (Rapid Activity of Platelet Inhibitor Drugs) primary PCI study. J Am Coll Cardiol. 2013;61:1601–1606. doi: 10.1016/j.jacc.2013.01.024.

9. Kimmelstiel C, Zhang P, Kapur NK, Weintraub A, Krishnamurthy B, Castaneda V, Covic L, Kuliopulos A. Bivalirudin is a dual inhibitor of thrombin and collagen-dependent platelet activation in patients un-dergoing percutaneous coronary intervention. Circ Cardiovasc Interv. 2011;4:171–179. doi: 10.1161/CIRCINTERVENTIONS.110.959098.

10. Dangas GD, Mehran R, Nikolsky E, Claessen BE, Lansky AJ, Brodie BR, Witzenbichler B, Guagliumi G, Peruga JZ, Dudek D, Möckel M, Caixeta A, Parise H, White H, Stone GW; HORIZONS-AMI Trial Investigators. Effect of switching antithrombin agents for primary angioplasty in acute myocardial infarction: the HORIZONS-SWITCH analysis. J Am Coll Cardiol. 2011;57:2309–2316. doi: 10.1016/j.jacc.2011.01.038.

11. Clemmensen P, Wiberg S, Van’t Hof A, Deliargyris EN, Coste P, Ten Berg J, Cavallini C, Hamon M, Dudek D, Zeymer U, Tabone X, Kristensen SD, Bernstein D, Anthopoulos P, Prats J, Steg PG. Acute stent throm-bosis after primary percutaneous coronary intervention: insights from the EUROMAX trial (European Ambulance Acute Coronary Syndrome Angiography). JACC Cardiovasc Interv. 2015;8(1 pt B):214–220. doi: 10.1016/j.jcin.2014.11.002.

12. Valgimigli M, Frigoli E, Leonardi S, Rothenbühler M, Gagnor A, Calabrò P, Garducci S, Rubartelli P, Briguori C, Andò G, Repetto A, Limbruno U, Garbo R, Sganzerla P, Russo F, Lupi A, Cortese B, Ausiello A, Ierna S, Esposito G, Presbitero P, Santarelli A, Sardella G, Varbella F, Tresoldi S, de Cesare N, Rigattieri S, Zingarelli A, Tosi P, van ‘t Hof A, Boccuzzi G, Omerovic E, Sabaté M, Heg D, Jüni P, Vranckx P; MATRIX Investigators. Bivalirudin or unfractionated heparin in acute coronary syndromes. N Engl J Med. 2015;373:997–1009. doi: 10.1056/NEJMoa1507854.

13. Stone GW, Clayton T, Deliargyris EN, Prats J, Mehran R, Pocock SJ. Reduction in cardiac mortality with bivalirudin in patients with and without

major bleeding: The HORIZONS-AMI trial (Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction). J Am Coll Cardiol. 2014;63:15–20. doi: 10.1016/j.jacc.2013.09.027.

14. Collet JP, Hulot JS, Pena A, Villard E, Esteve JB, Silvain J, Payot L, Brugier D, Cayla G, Beygui F, Bensimon G, Funck-Brentano C, Montalescot G. Cytochrome P450 2C19 polymorphism in young patients treated with clopidogrel after myocardial infarc-tion: a cohort study. Lancet. 2009;373:309–317. doi: 10.1016/S0140-6736(08)61845-0.

15. Silvain J, Collet JP, Nagaswami C, Beygui F, Edmondson KE, Bellemain-Appaix A, Cayla G, Pena A, Brugier D, Barthelemy O, Montalescot G, Weisel JW. Composition of coronary thrombus in acute myocardial infarction. J Am Coll Cardiol. 2011;57:1359–1367. doi: 10.1016/j.jacc.2010.09.077.

16. Ramaiola I, Padró T, Peña E, Juan-Babot O, Cubedo J, Martin-Yuste V, Sabate M, Badimon L. Changes in thrombus composition and profilin-1 release in acute myocardial infarction. Eur Heart J. 2015;36:965–975. doi: 10.1093/eurheartj/ehu356.

17. Vergallo R, Joye R, Barlis P, Jia H, Tian J, Soeda T, Minami Y, Hu S, Dauerman HL, Toma C, Chan J, Lee H, Biasucci LM, Crea F, Jang IK. Bivalirudin versus unfractionated heparin for residual thrombus bur-den: a frequency-domain optical coherence tomography study. Catheter Cardiovasc Interv. 2015;85:575–582. doi: 10.1002/ccd.25631.

18. Keating FK, Dauerman HL, Whitaker DA, Sobel BE, Schneider DJ. Increased expression of platelet P-selectin and formation of platelet-leukocyte aggregates in blood from patients treated with unfraction-ated heparin plus eptifibatide compared with bivalirudin. Thromb Res. 2006;118:361–369. doi: 10.1016/j.thromres.2005.07.020.

19. Jang IK, Gold HK, Ziskind AA, Fallon JT, Holt RE, Leinbach RC, May JW, Collen D. Differential sensitivity of erythrocyte-rich and platelet-rich arterial thrombi to lysis with recombinant tissue-type plasminogen activator. A possible explanation for resistance to coronary thrombolysis. Circulation. 1989;79:920–928.

20. Erlich JH, Boyle EM, Labriola J, Kovacich JC, Santucci RA, Fearns C, Morgan EN, Yun W, Luther T, Kojikawa O, Martin TR, Pohlman TH, Verrier ED, Mackman N. Inhibition of the tissue factor-thrombin pathway limits infarct size after myocardial ischemia-reperfusion injury by reduc-ing inflammation. Am J Pathol. 2000;157:1849–1862.

21. Jormalainen M, Vento AE, Lukkarinen H, Kääpä P, Kytö V, Lauronen J, Paavonen T, Suojaranta-Ylinen R, Petäjä J. Inhibition of thrombin during re-perfusion improves immediate postischemic myocardial function and modu-lates apoptosis in a porcine model of cardiopulmonary bypass. J Cardiothorac Vasc Anesth. 2007;21:224–231. doi: 10.1053/j.jvca.2006.02.028.

22. Mirabet M, Garcia-Dorado D, Ruiz-Meana M, Barrabés JA, Soler-Soler J. Thrombin increases cardiomyocyte acute cell death after ischemia and reperfusion. J Mol Cell Cardiol. 2005;39:277–283. doi: 10.1016/j.yjmcc.2005.03.005.

23. Wöhrle J, Merkle N, Kunze M, Cristea E, Mehran R, Rottbauer W, Stone GW. Effect of bivalirudin compared with unfractionated heparin plus abciximab on infarct size and myocardial recovery after primary percu-taneous coronary intervention: the HORIZONS-AMI CMRI substudy. Catheter Cardiovasc Interv. 2012;79:1083–1089. doi: 10.1002/ccd.23179.

24. Gibson CM, Morrow DA, Murphy SA, Palabrica TM, Jennings LK, Stone PH, Lui HH, Bulle T, Lakkis N, Kovach R, Cohen DJ, Fish P, McCabe CH, Braunwald E; TIMI Study Group. A randomized trial to evaluate the relative protection against post-percutaneous coronary intervention micro-vascular dysfunction, ischemia, and inflammation among antiplatelet and antithrombotic agents: the PROTECT-TIMI-30 trial. J Am Coll Cardiol. 2006;47:2364–2373. doi: 10.1016/j.jacc.2005.12.077.

25. Stone GW, Witzenbichler B, Guagliumi G, Peruga JZ, Brodie BR, Dudek D, Kornowski R, Hartmann F, Gersh BJ, Pocock SJ, Dangas G, Wong SC, Fahy M, Parise H, Mehran R; HORIZONS-AMI Trial Investigators. Heparin plus a glycoprotein IIb/IIIa inhibitor versus bivalirudin monother-apy and paclitaxel-eluting stents versus bare-metal stents in acute myocar-dial infarction (HORIZONS-AMI): final 3-year results from a multicentre, randomised controlled trial. Lancet. 2011;377:2193–2204. doi: 10.1016/S0140-6736(11)60764-2.



ownloaded from


Gregg W. StoneClemmensen, Arnoud van 't Hof, Jayne Prats, Debra Bernstein, Efthymios N. Deliargyris and

George D. Dangas, Mikkel M. Schoos, Philippe Gabriel Steg, Roxana Mehran, Peter2 Randomized Trials

Elevation Myocardial Infarction: A Patient-Level Analysis of−Intervention in ST-Segment Early Stent Thrombosis and Mortality After Primary Percutaneous Coronary

Print ISSN: 1941-7640. Online ISSN: 1941-7632 Copyright © 2016 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleCirculation: Cardiovascular Interventions

doi: 10.1161/CIRCINTERVENTIONS.115.0032722016;9:e003272Circ Cardiovasc Interv.

http://circinterventions.ahajournals.org/content/9/5/e003272World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circinterventions.ahajournals.org/content/suppl/2016/05/10/CIRCINTERVENTIONS.115.003272.DC1Data Supplement (unedited) at:

http://circinterventions.ahajournals.org//subscriptions/

is online at: Circulation: Cardiovascular Interventions Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Answer

Permissions and Rights Question andunder Services. Further information about this process is available in thepermission is being requested is located, click Request Permissions in the middle column of the Web pageClearance Center, not the Editorial Office. Once the online version of the published article for which

can be obtained via RightsLink, a service of the CopyrightCirculation: Cardiovascular Interventionsin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:



ownloaded from

http://circinterventions.ahajournals.org/content/9/5/e003272

http://circinterventions.ahajournals.org/content/suppl/2016/05/10/CIRCINTERVENTIONS.115.003272.DC1

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circinterventions.ahajournals.org//subscriptions/


Supplemetnal Material

Supplemental Table 1. Baseline Characteristics (Intent-to-Treat Stented Population)

Variable Bivalirudin (n = 2479)

Heparin ± GPI (n = 2456)

Overall (N = 4935)

P value*

Age, y 60.7±11.9 61.3±11.8 61.0±11.9 0.10

Age >65 881/2479 (35.5) 895/2456 (36.4) 1776/4935 (36.0) 0.51

Weight, kg 81.25±15.24 81.33±15.20 81.29±15.22 0.84

Women 558/2479 (22.5) 553/2456 (22.5) 1111/4935 (22.5) 0.99

Diabetes 341/2479 (13.8) 391/2454 (15.9) 732/4933 (14.8) 0.33

Hypertension 1173/2479 (47.3) 1244/2454 (50.7) 2417/4933 (49.0) 0.29

Hyperlipidemia 1012/2479 (40.8) 1000/2454 (40.7) 2012/4933 (40.8) 0.87

Current smoker 1167/2472 (47.2) 1120/2450 (45.7) 2287/4922 (46.5) 0.60

Prior myocardial infarction 191/2479 (7.7) 250/2454 (10.2) 441/4933 (8.9) 0.40

Prior percutaneous coronary intervention 215/2478 (8.7) 228/2454 (9.3) 443/4932 (9.0) 0.19

Prior coronary artery bypass grafting 51/2479 (2.1) 50/2454 (2.0) 101/4933 (2.0) 0.65

Killip class ≥2 201/2408 (8.3) 183/2377 (7.7) 384/4785 (8.0) 0.41

Baseline platelet count, cells/mm3 x103 253.93±73.15 249.69±70.59 251.83±71.91 0.04

Baseline hemoglobin, g/dL 14.50±1.63 14.47±1.56 14.48±1.60 0.52

Anemia 255/2325 (11.0) 269/2280 (11.8) 524/4605 (11.4) 0.37

Baseline creatinine clearance, mL/min 95.18±35.12 94.71±35.59 94.94±35.35 0.68

≤60 mL/min 327/2297 (14.2) 366/2266 (16.2) 693/4563 (15.2) 0.07

>60 mL/min 1970/2297 (85.8) 1900/2266 (83.8) 3870/4563 (84.8) 0.07

Region

United States 347/2479 (14.0) 336/2456 (13.7) 683/4935 (13.8) 0.77

Europe 1805/2479 (72.8) 1793/2456 (73.0) 3598/4935 (72.9) 0.92

Rest of world 327/2479 (13.2) 327/2456 (13.3) 654/4935 (13.3) 0.86

Aspirin use 2476/2479 (99.9) 2451/2454 (99.9) 4927/4933 (99.9) 0.99

P2Y12 inhibitor loading dose, any 2442/2476 (98.6) 2415/2456 (98.3) 4857/4932 (98.5) 0.40

Clopidogrel 2006/2462 (81.5) 1993/2434 (81.9) 3999/4896 (81.7) 0.59

Ticlopidine 5/2461 (0.2) 9/2434 (0.4) 14/4895 (0.3) 0.28

Prasugrel 258/2462 (10.5) 242/2436 (9.9) 500/4898 (10.2) 0.45

Ticagrelor 176/2462 (7.1) 176/2436 (7.2) 352/4898 (7.2) 0.94

P2Y12 inhibitor maintenance dose, any 2432/2475 (98.3) 2410/2454 (98.2) 4842/4929 (98.2) 0.89

Clopidogrel 1914/2449 (78.2) 1903/2428 (78.4) 3817/4877 (78.3) 0.62

Ticlopidine 26/2449 (1.1) 21/2429 (0.9) 47/4878 (1.0) 0.49

Prasugrel 284/2450 (11.6) 260/2431 (10.7) 544/4881 (11.1) 0.20

Ticagrelor 227/2450 (9.3) 239/2431 (9.8) 466/4881 (9.5) 0.52

Glycoprotein IIb/IIIa inhibitor use 220/2475 (8.9) 2182/2454 (88.9) — <0.0001

*P value from ANOVA or Cochran-Mantel-Haenszel test controlling for study

Supplemental Table 2. Procedural Characteristics (Intent-to-Treat Stented Population)

Variable Bivalirudin (n = 2479)

Heparin ± GPI (n = 2456)

Overall (N = 4935)

P value*

Femoral artery access site 1956/2479 (78.9) 1959/2455 (79.8) 3915/4934 (79.3) 0.33

Radial artery access site 517/2479 (20.9) 492/2455 (20.0) 1009/4934 (20.4) 0.37

Single-vessel disease 1263/2478 (51.0) 1234/2453 (50.3) 2497/4931 (50.6) 0.63

Left main disease 112/2478 (4.5) 88/2454 (3.6) 200/4932 (4.1) 0.09


Left main 19/2476 (0.8) 18/2452 (0.7) 37/4928 (0.8) 0.89

Left anterior descending 1066/2476 (43.1) 1104/2452 (45.0) 2170/4928 (44.0) 0.16

Left circumflex 385/2476 (15.5) 377/2452 (15.4) 762/4928 (15.5) 0.87

Right 1116/2476 (45.1) 1079/2452 (44.0) 2195/4928 (44.5) 0.45

Bypass grafting (venous or arterial)

20/2476 (0.8) 22/2452 (0.9) 42/4928 (0.9) 0.73

Balloon angioplasty only 33/2459 (1.3) 27/2440 (1.1) 60/4899 (1.2) 0.46

Drug-eluting stent implanted 1695/2479 (68.4) 1676/2456 (68.2) 3371/4935 (68.3) 0.93

Thrombectomy 291/2465 (11.8) 281/2436 (11.5) 572/4901 (11.7) 0.66

TIMI flow, pre-PCI

0/1 1661/2464 (67.4) 1586/2442 (64.9) 3247/4906 (66.2) 0.07

2 395/2464 (16.0) 407/2442 (16.7) 802/4906 (16.3) 0.55

3 404/2464 (16.4) 446/2442 (18.3) 850/4906 (17.3) 0.09

TIMI flow, post-PCI

0/1 29/2463 (1.2) 30/2442 (1.2) 59/4905 (1.2) 0.87

2 135/2463 (5.5) 115/2442 (4.7) 250/4905 (5.1) 0.22

3 2296/2463 (93.2) 2296/2442 (94.0) 4592/4905 (93.6) 0.26


ACE inhibitor or ARB 1892/2455 (77.1) 1899/2417 (78.6) 3791/4872 (77.8) 0.18

Aspirin 2407/2455 (98.0) 2351/2417 (97.3) 4758/4872 (97.7) 0.08

Beta-blocker 2263/2455 (92.2) 2215/2416 (91.7) 4478/4871 (91.9) 0.53

P2Y12 inhibitor 2407/2455 (98.0) 2366/2417 (97.9) 4773/4872 (98.0) 0.73

Statin 2335/2455 (95.1) 2309/2417 (95.5) 4644/4872 (95.3) 0.48

*P value from ANOVA or Cochran-Mantel-Haenszel test controlling for study

ACE indicates angiotensin converting enzyme; ARB, angiotensin receptor blocker; PCI, percutaneous coronary intervention; TIMI: Thrombolysis In Myocardial Infarction.

Supplemental Table 3. List of stent thrombosis subjects who died.

Age Female Study Time of ST

(days) Study Time of ST (hrs)

Study Time of death (days)

History of

Diabetes Mellitus

Prior MI

SVD CrCl ≤60

mL/min

IRA Killip Class

TIMI Flow Before

PCI

Treatment Group

Clopidogrel Loading

Dose

Clopidogrel Maintenance

Dose

Propensity score

57 0 6 144 11 0 0 0 0 LCX I 0/1 HEP + GPI 1 1 0,02824 68 1 2 48 3 1 0 0 1 RCA I 0/1 HEP + GPI 1 1 0,7464 71 0 14 336 14 1 0 1 0 LAD I 3 HEP + GPI 1 1 0,54941 67 1 24 576 24 0 0 0 0 LAD I 0/1 HEP + GPI 1 1 0,59595 64 0 2 48 2 1 0 0 1 RCA II 0/1 HEP + GPI 1 1 0,3476 54 0 6 144 6 0 0 1 1 LCX I 0/1 HEP + GPI 1 1 0,68887 78 1 6 144 7 0 0 0 0 LAD, RCA II 0/1 HEP + GPI 1 1 0,52634 67 1 17 408 17 0 0 0 0 LAD, RCA I 0/1 HEP + GPI 1 1 0,41176 76 0 8 192 8 0 0 1 0 LAD I 0/1 HEP + GPI 1 1 0,29907 71 0 18 432 18 1 0 0 1 LAD II 0/1 HEP + GPI 1 1 0,89877 68 0 3 72 4 0 0 0 0 LAD, LCX I 0/1 HEP + GPI 1 1 0,91525 68 1 2 48 2 0 0 1 0 LAD I 0/1 HEP + GPI 1 1 0,66093 70 0 7 168 14 0 0 1 LAD I 0/1 HEP + GPI 1 1 0,80866 69 0 13 312 13 1 0 1 0 RCA II 0/1 HEP + GPI 1 1 0,33782 69 1 0 10 1 0 0 1 1 RCA I 0/1 HEP + GPI 1 0 0,54568 81 1 17 408 29 1 0 0 1 LAD III 0/1 HEP + GPI 1 1 0,91921 55 0 5 120 5 0 1 0 0 LAD I 0/1 BIVAL 1 1 0,30179 84 0 4 96 4 0 0 0 1 RCA I 2 BIVAL 1 1 0,59913 54 0 0 4 2 0 0 0 0 LAD I 0/1 BIVAL 1 1 0,2517 45 0 25 600 25 1 1 0 1 RCA I 2 BIVAL 1 1 0,51681

SVD: Single Vessel Disease, BIVAL: Bivalirudin, HEP + GPI: Ufractionated heparin and Glycoprotein IIb/IIIa Inhibitor IRA: Infarct Related Artery, MI: Myocardial Infarction, ST: Stent Thrombosis, PCI: Percutaneous Coronary Intervention, TIMI: Thrombolysis in Myocardial Infarction, LAD: Left Anterior Descending Artery, RCA: Right Coronary Artery, LCX: Left Circumflex

Early Stent Thrombosis and Mortality After Primary...

Documents

Transcript of Early Stent Thrombosis and Mortality After Primary...