Wcaappf

M

TEns1tTC

Cr

S

0d

Efficacy and Safety of Triple Antiplatelet Therapy Withand Without Concomitant Anticoagulation During ElectivePercutaneous Coronary Intervention (the REMOVE Trial)

Rafael Valencia, MD*, Matthew J. Price, MD, Neil Sawhney, MD, Steve S. Lee, MD,Garrett B. Wong, MD, Raghava R. Gollapudi, MD, Michelle Banares, Richard A. Schatz, MD,

and Paul S. Teirstein, MD

Adjunctive glycoprotein IIb/IIIa inhibition decreases ischemic events after percutaneouscoronary intervention (PCI) but is associated with increased bleeding. We hypothesizedthat maximal antiplatelet therapy with aspirin, a thienopyridine, and a glycoproteinIIb/IIIa inhibitor without unfractionated heparin (UFH) would result in fewer bleedingcomplications and maintain efficacy in elective PCI. A total of 159 patients undergoingelective PCI were randomized to intraprocedural eptifibatide alone or eptifibatide plusUFH. Patients received aspirin 325 mg and clopidogrel 300 mg before the procedure. Theprimary end point was the Landefeld bleeding index. Secondary end points included thecomposite clinical outcome of in-hospital death, myocardial infarction, urgent target vesselrevascularization, and Thrombolysis In Myocardial Infarction major bleeding, and a com-posite bleeding outcome of major, minor, and nuisance bleeding. The Landefeld bleedingindex was significantly lower in the eptifibatide-only group compared with the eptifibatide-plus-UFH group (3.0 vs 3.9, p � 0.03). There was no significant difference in the compositeclinical end point between groups (eptifibatide only 17% vs eptifibatide plus UFH 15%, p �0.7). There was a trend toward a decrease in the composite bleeding end point in theeptifibatide-only compared with the eptifibatide-plus-UFH group (43% vs 56%, p � 0.10).In conclusion, during elective PCI, a strategy of aggressive antiplatelet therapy usingaspirin, clopidogrel, and eptifibatide without anticoagulant therapy appears to decreasebleeding complications. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:

1099–1102)

Wp

tapat�dphtdckoTltqrd

m

e hypothesized that in patients undergoing elective per-utaneous coronary intervention (PCI), a pharmacologicpproach of triple antiplatelet therapy alone consisting ofspirin, clopidogrel, and eptifibatide would result in similarrocedural success and less bleeding than the standard ap-roach of triple antiplatelet therapy with concomitant un-ractionated heparin (UFH).

ethods

he Reduction in Major and Minor Adverse Events Withptifibatide-based Pharmacotherapy in Percutaneous Coro-ary Intervention (REMOVE) trial was a single-center, pro-pective, randomized-controlled trial that enrolled a total of59 patients with coronary artery disease undergoing elec-ive PCI performed between June 2004 and August 2005.he protocol was reviewed and approved by the Scrippslinic (La Jolla, California) institutional review board.

Department of Cardiovascular Diseases, Scripps Clinic, La Jolla,alifornia. Manuscript received January 25, 2007; revised manuscript

eceived and accepted April 24, 2007.This work was supported by an unrestricted educational grant from

chering-Plough, Kenilworth, New Jersey.*Corresponding author: Tel: 858-554-9905; fax: 858-554-6883.

pE-mail address: valencia.rafael@scrippshealth.org (R. Valencia).

002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.oi:10.1016/j.amjcard.2007.04.057

ritten informed consent was obtained from all partici-ants.

Entry criteria included men and nonpregnant women 18o 85 years of age who were scheduled for PCI and wereble to provide informed consent. Exclusion criteria wereatients with a history of bleeding diathesis or evidence ofctive abnormal bleeding within the previous 30 days, pa-ients with severe hypertension (systolic blood pressure

200 mm Hg or diastolic blood pressure �110 mm Hg)espite antihypertensive therapy, major surgery within thereceding 6 weeks, history of stroke within 30 days or anyistory of hemorrhagic stroke, current or planned adminis-ration of another parenteral glycoprotein IIb/IIIa inhibitor,ependency on renal dialysis, known hypersensitivity to anyomponent of eptifibatide or heparin, increased creatineinase-MB isoenzyme �2 times the upper limit of normalr increased troponin, unstable angina with dynamic ST or-wave changes within 48 hours of enrollment, and a target

esion that was totally occluded. During the study, the pro-ocol was amended to also exclude bifurcation lesions re-uiring treatment with “crush” or “kissing” stents. Patientsandomized before this amendment were included in theata analysis.

All patients received aspirin 325 mg and clopidogrel 300g �1 hour before the procedure. After the procedure,

atients were treated with aspirin 325 mg/day and clopi-

www.AJConline.org

dwea2idcthpbt

mgc(ooaetc

ttAJcWcu

whgbopchsptpvbNamasp

suegr

tv

aPiItuvappstrqfieeov

dnscctcu

R

Aaltp

TD

V

AMWDSHHRHPPPHCP

1100 The American Journal of Cardiology (www.AJConline.org)

ogrel 75 mg/day. Patients were randomized to treatmentith eptifibatide without adjunctive anticoagulant use or to

ptifibatide with adjunctive UFH. Eptifibatide was given as180-�g/kg intravenous bolus before PCI, followed by

-�g/kg/min infusion, with a second bolus of 180 �g/kgntravenously given 10 minutes later. The infusion rate wasecreased to 1 �g/kg/min in subjects with an estimatedreatine clearance �50 ml/min. The infusion was continuedhrough the intervention until 18 hours after PCI or untilospital discharge, whichever came first. In the eptifibatide-lus-UFH arm, UFH was administered as an intravenousolus dosed at 50 U/kg and adjusted for an activated clottingime of 200 to 250 seconds.

At the discretion of the investigator, UFH could be ad-inistered to patients assigned to the eptifibatide-only

roup on a provisional “bailout” basis in the setting oflinical or angiographic evidence of an ischemic eventThrombolysis In Myocardial Infarction [TIMI] flow gradesf 0 to 2 or slow reflow, dissection with decreased flow, newr suspected thrombus, distal embolization, side branch orbrupt vessel closure, clinical instability, or prolonged isch-mia). Provisional heparin was administered as an initial 50-o 60-U/kg intravenous bolus and adjusted for an activatedlotting time of 200 to 250 seconds.

If femoral anatomy was suitable, sheaths were removed athe completion of the PCI using Food and Drug Administra-ion–approved vascular closure devices (Perclose ProGlide,bbott Medical, Abbott Park, Illinois; Angio-Seal STS, St.

ude Medical, St. Paul, Minnesota). Manual or pneumaticompression (FemoStop II Plus, Radi Medical Systems,

ilmington, Massachusetts) was used once the activatedlotting time was �180 seconds if a closure device was notsed at the end of the procedure.

Blood samples for creatine kinase and its MB isoenzymeere drawn before PCI and 6 to 8, 12 to 14, and 18 to 24ours after PCI or before discharge if �24 hours. Hemo-lobin, hematocrit, and platelet counts were measured ataseline, at completion of PCI, and 18 to 24 hours after PCIr before hospital discharge if �24 hours. The primary endoint was the Landefeld bleeding index, defined as a de-rease in hematocrit (grams per deciliter) over the first 24ours, adjusted for whole blood or red blood cell transfu-ions ([{units transfused � 3} � baseline hematocrit) �ostprocedure hematocrit).1 Secondary end points includedhe composite ischemic and bleeding outcomes of in-hos-ital death, myocardial infarction, urgent target vessel re-ascularization, and TIMI major bleeding, and a compositeleeding outcome of major, minor, and nuisance bleeding.uisance bleeding was defined as (1) any bleeding that was

ssociated with significant ecchymosis, oozing, and/or he-atoma formation not meeting TIMI major or minor criteria

nd (2) pseudoaneurysm formation documented by ultra-onography. Thrombocytopenia was defined as an absolutelatelet count �150,000/mm3 and graded as mild to severe.

Periprocedural myocardial infarction was defined as 2erial measurements of creatine kinase-MB �3 times thepper limit of normal or development of new abnormallectrocardiographic Q waves in �2 contiguous leads. Ur-ent target revascularization was defined as any episode of

ecurrent ischemic discomfort that prompted revasculariza- l

ion (PCI or coronary artery bypass grafting) of the targetessel during the index hospitalization.

This was a pilot study of the safety of PCI using triplentiplatelet therapy without concomitant anticoagulation.revious studies have demonstrated major and minor bleed-

ng rates of approximately 4% for UFH plus glycoproteinIb/IIIa inhibitors,2,3 whereas a relatively large observa-ional study suggested bleeding may be decreased to �1.0%sing triple antiplatelet therapy.4 With 80% power and an �alue of 0.05 (2-tailed), the projected sample size to detectdecrease in bleeding from 4% to 0.5% would be 582

atients (281 patients per study arm). In this single-centerilot study we prospectively sought to enroll 150 patients toerve as guidance for a larger, multicenter trial. Nine pa-ients who met the angiographic criteria for study entry wereandomized and treated with study medication but subse-uently did not undergo coronary intervention because ofndings on intravascular ultrasound (3 eptifibatide only; 6ptifibatide � UFH). Additional patients were enrolled tonsure a minimum PCI sample of 150 subjects. Thus, a totalf 159 patients were randomized, 9 undergoing only intra-ascular ultrasound and 150 undergoing PCI.

Continuous data are expressed as means � SDs or me-ians with first and third quartiles if the distribution was notormal. Categorical variables were compared using chi-quare test or Fisher’s exact test when any expected cellount was �5 for a 2 � 2 table. Continuous variables wereompared using the 2-sample t test or Wilcoxon rank-sumest when applicable. Differences were considered statisti-ally significant at p �0.05. Analyses were performed bysing SAS version 8 (SAS Institute, Cary, North Carolina).

esults

total of 159 patients were enrolled. Demographic char-cteristics for the entire cohort are listed in Table 1. Base-ine lesion and procedural characteristics for the 150 pa-ients with PCI are listed in Table 2. Indication for therocedure was stable angina in 70%. The average number of

able 1emographic and clinical characteristics at admission

ariable Eptifibatide(n � 82)

Eptifibatide � UFH(n � 77)

pValue

ge (yrs) 65 � 10 65 � 9 NSen 71 (87%) 58 (75%) 0.070eight (kg) 91 � 16.6 88 � 16.6 NSiabetes mellitus 20 (24%) 23 (30%) NSmokers (within previous yr) 8 (10%) 10 (13%) NSypertension 66 (80%) 67 (87%) NSyperlipidemia 67 (82%) 63 (82%) NSenal insufficiency 11 (14%) 14 (18%) NSistory of stroke 2 (2%) 6 (8%) NSrevious myocardial infarction 22 (28%) 17 (22%) NSrevious coronary bypass 22 (27%) 22 (29%) NSrevious coronary intervention 54 (66%) 50 (65%) NSematocrit (%) 42.3 � 3.9 41.9 � 4.3 NSreatinine (mg/dl) 1.1 � 0.2 1.2 � 0.3 NSlatelets (thousands/mm3) 228 � 72 230 � 54 NS

Data presented as mean � SD or number of patients (percentage).

esions treated per patient was 1.75 � 0.8. Treatment groups

wplB

eicpip1

Pi(ima1t

odcadtri

D

Adtwscbppspdds

stfalpppTtii

taat

sh

TB

V

I

L

BT

RPVL

TI

V

L

B

AH

T

TI

C

1101Coronary Artery Disease/Aggressive Antiplatelet Therapy Without Anticoagulation During PCI

ere well matched with regard to baseline clinical androcedural characteristics, except for significantly greateresion complexity in the eptifibatide-only group (lesion type2/C in 47% vs 39%, p � 0.008).

Bleeding outcomes are presented in Table 3. The primarynd point of the study, the Landefeld bleeding severityndex, was significantly lower in the eptifibatide-only groupompared with the eptifibatide-plus-UFH group (3.0 vs 3.9,� 0.030). The composite bleeding end point was numer-

cally lower in the eptifibatide-alone group (43% vs 56%,� 0.10) as was the rate of hematoma formation (6% vs

4% of patients, p � 0.13).In-hospital ischemic outcomes are listed in Table 4.

eriprocedural myocardial infarction was numerically greatern the eptifibatide-only versus the eptifibatide-plus-UFH group17% vs 13%), but this difference was not statistically signif-cant (p � 0.5). The composite clinical end point of death,yocardial infarction, urgent target vessel revascularization,

nd major bleed was similar between the 2 groups (17% vs5%, p � 0.7). In situ stent thrombosis occurred in 1 patient in

able 2aseline angiographic and lesion characteristics

ariable Eptifibatide(n � 152)*

Eptifibatide � UFH(n � 119)*

pValue

ndex coronary vessel NSRight 44 (29%) 25 (21%)Left main 5 (3%) 3 (3%)Left anterior descending 56 (37%) 51 (43%)Left circumflex 40 (26%) 32 (27%)Saphenous vein graft 7 (5%) 7 (6%)esion type NSDe novo 128 (84%) 100 (84%)Restenotic 24 (16%) 19 (16%)ifurcation 42 (28%) 28 (23%) NSIMI grade �3 flow before

procedure9 (5%) 5 (5%) NS

eference vessel diameter (mm) 2.9 � 0.5 2.9 � 0.4 NSreintervention stenosis (%) 81 � 12 80 � 14 NSisible thrombus 3 (2%) 1 (�1%) NSesion type B2/C 72 (47%) 45 (39%) 0.008

Data presented as mean � SD or number of patients (percentage).* Total number of lesions present in final population.

able 3n-hospital bleeding outcomes

ariable Eptifibatide(n � 79)

Eptifibatide � UFH(n � 71)

pValue

andefeld bleeding index,mean � SD

3.0 � 2.7 3.9 � 2.6 0.030

leedingComposite 34 (43%) 40 (56%) 0.104

Major 0 (0%) 1 (1%) NSMinor 1 (1%) 1 (1%) NSNuisance 33 (42%) 38 (54%) 0.150

fter procedureematocrit (%)Mean � SD 39.5 � 3.7 38.2 � 4.5 0.057Mean before/after 2.8 3.7hrombocytopenia 15 (19%) 7 (10%) 0.115

he eptifibatide-only arm who underwent bifurcation stenting q

f the left anterior descending and first diagonal using 2 kissingrug-eluting stents (prompting amendment of the study proto-ol). This was treated with bailout UFH, abciximab infusion,nd balloon dilatation and resulted in a non–Q-wave myocar-ial infarction. Bailout heparin was used in 1 other patient inhe eptifibatide-only arm due to TIMI grade 2 flow after dis-uption of a left main plaque by a coronary wire, also resultingn non–Q-wave myocardial infarction.

iscussion

lthough adjunctive use of glycoprotein IIb/IIIa inhibitionuring PCI decreases ischemic events, bleeding complica-ions are increased.5 In this prospective, randomized study,e demonstrate that in patients undergoing elective PCI, a

trategy of stand-alone antiplatelet therapy with aspirin,lopidogrel, and eptifibatide in the absence of an antithrom-in decreased periprocedural bleeding complications com-ared with similar therapy in combination with UFH. Ourrimary end point, the Landefeld bleeding index, has beenhown to be an objective and useful measurement oferiprocedural bleeding in PCI trials.6 In addition to theecrease in the Landefeld index, there was a trend toward aecrease in clinically meaningful bleeding complications,uch as hematoma formation.

The beneficial impact on bleeding we observed is con-istent with previous, nonrandomized studies.7 In a prospec-ive registry, Denardo et al4 found that elective PCI per-ormed with eptifibatide, clopidogrel, and aspirin in thebsence of antithrombin therapy was associated with aow rate of in-hospital major and minor bleeding com-lications (0.2% and 0.6%, respectively). Furthermore, inatients randomized to eptifibatide in the Enhanced Sup-ression of the Platelet IIb/IIIa Receptor with Integrilinherapy (ESPRIT) trial, there was no association be-

ween decreasing activated clotting times and rate ofschemic events, whereas major bleeding increased withncreasing activated clotting times.2,8

Inherent disadvantages of UFH, such as potential activa-ion of platelets and its association with immune-mediatednd nonimmune-mediated thrombocytopenia and increasedggregability, make its elimination from PCI protocols at-ractive.9–11

Furthermore, decreases in even minor bleeding may haveignificant clinical and economic benefits. Recent reportsave highlighted the adverse impact of bleeding on subse-

able 4n-hospital clinical outcomes

Eptifibatide(n � 79)

Eptifibatide � UFH(n � 71)

pValue

omposite of death, myocardialinfarction, urgent targetvessel revascularization andmajor bleed

13 (17%) 10 (15%) 0.720

Death 0 (0%) 0 (0%) NSMyocardial infarction 13 (17%) 9 (13%) 0.541Urgent target vessel

revascularization0 (0%) 0 (0%) NS

uent death and myocardial infarction.12,13 A meta-analysis

ottEEtc

tispsCTif

fse1ewst

sciwiaddmtshpe

1

1

1

1

1

1

1

1

1

1

2

1102 The American Journal of Cardiology (www.AJConline.org)

f �26,000 patients enrolled in acute coronary syndromerials found an increase in mortality and myocardial infarc-ion with increased bleeding severity.14 In the Randomizedvaluation in PCI Linking Angiomax to Reduced Clinicalvents 2 (REPLACE-2) trial, minor bleeding (as defined by

he trial) was a major contributor to increased hospitalost.15

In the present study, although removing UFH decreasedhe incidence of bleeding, there appeared to be an increasen ischemic outcomes (although this was not statisticallyignificant). This might be reflective of the complex studyopulation; our trial had no specific angiographic exclu-ion criteria, 44% of lesions were classified as type B2 or, and nearly 1/4 of target lesions involved bifurcations.he recently reported Nordic Bifurcation Study16 found

ncreased biomarkers in 18% of patients undergoing bi-urcation stenting.

There was 1 occurrence of possible in situ thrombusormation in the eptifibatide-only group during bifurcationtenting, leading to a protocol change that excluded furthernrollment of patients with bifurcation lesions. However, inreport, the incidence of in situ thrombosis during drug-

luting stent implantation using standard anticoagulationas 0.7%, with 3 of the 5 cases occurring during bifurcation

tenting and none receiving glycoprotein IIb/IIIa inhibi-ion.17

The REMOVE study was a single-center, open-labeltudy with a small sample. The purpose of this trial was toollect pilot data to plan a definitive trial. A potential lim-tation includes the intraprocedural use of heparinized flush,hich was not controlled for. Given the small amount typ-

cally used during cases, it is not likely to have significantlyffected activated clotting times. We used a 300-mg loadingose of clopidogrel �1 hour before the procedure. Detailedata regarding when the clopidogrel loading dose was ad-inistered was not collected and presents a potential limi-

ation. Recent trials since this study was designed haveuggested that a 600-mg loading dose �2 and possibly 4ours before PCI may provide further benefit.18–20 It isossible that such an approach might have improved isch-mic end points in the eptifibatide-only group.

1. Landefeld CS, Anderson PA, Goodnough LT, Moir TW, Hom DL,Rosenblatt MW, Goldman L. The bleeding severity index: validationand comparison to other methods for classifying bleeding complica-tions of medical therapy. J Clin Epidemiol 1989;42:711–718.

2. ESPRIT Investigators. Enhanced Suppression of the Platelet IIb/IIIaReceptor with Integrilin Therapy. Novel dosing regimen of eptifibatidein planned coronary stent implantation (ESPRIT): a randomised, pla-cebo-controlled trial. Lancet 2000;356:2037–2044.

3. Lincoff AM, Bittl JA, Harrington RA, Feit F, Kleiman NS, JackmanJD, Sarembock IJ, Cohen DJ, Spriggs D, Ebrahimi R, et al, for theREPLACE-2 Investigators. Bivalirudin and provisional glycopro-tein IIb/IIIa blockade compared with heparin and planned glycop-rotein IIb/IIIa blockade during percutaneous coronary intervention:REPLACE-2 randomized trial. JAMA 2003;289:853– 863.

4. Denardo SJ, Davis KE, Tcheng JE. Elective percutaneous coronaryintervention using broad-spectrum antiplatelet therapy (eptifibatide,clopidogrel, and aspirin) alone, without scheduled unfractionated hep-

arin or other antithrombin therapy. Am Heart J 2005;149:138–144.

5. Cote AV, Berger PB, Holmes DR Jr, Scott CG, Bell MR. Hemorrhagicand vascular complications after percutaneous coronary interventionwith adjunctive abciximab. Mayo Clin Proc 2001;76:890–896.

6. Bhatt DL, Lee BI, Casterella PJ, Pulsipher M, Rogers M, Cohen M,Corrigan VE, Ryan TJ Jr, Breall JA, Moses JW, et al, for the CoronaryRevascularization Using Integrilin and Single bolus Enoxaparin Study.Safety of concomitant therapy with eptifibatide and enoxaparin inpatients undergoing percutaneous coronary intervention: results of theCoronary Revascularization Using Integrilin and Single bolus Enox-aparin Study. J Am Coll Cardiol 2003;41:20–25.

7. Denardo SJ, Davis KE, Reid PR, Tcheng JE. Efficacy and safety ofminimal dose (�1,000 units) unfractionated heparin with abciximab inpercutaneous coronary intervention. Am J Cardiol 2003;91:1–5.

8. Tolleson TR, O’Shea JC, Bittl JA, Hillegass WB, Williams KA, LevineG, Harrington RA, Tcheng JE. Relationship between heparin anticoagu-lation and clinical outcomes in coronary stent intervention: observationsfrom the ESPRIT trial. J Am Coll Cardiol 2003;41:386–393.

9. Xiao Z, Theroux P. Platelet activation with unfractionated heparin attherapeutic concentrations and comparisons with a low-molecular-weight heparin and with a direct thrombin inhibitor. Circulation 1998;97:251–256.

0. Brieger DB, Mak KH, Kottke-Marchant K, Topol EJ. Heparin-inducedthrombocytopenia. J Am Coll Cardiol 1998;31:1449–1459.

1. Davoren A, Aster RH. Heparin-induced thrombocytopenia and throm-bosis. Am J Hematol 2006;81:36–44.

2. Kinnaird TD, Stabile E, Mintz GS, Lee CW, Canos DA, Gevorkian N,Pinnow EE, Kent KM, Pichard AD, Satler LF, et al. Incidence, pre-dictors, and prognostic implications of bleeding and blood transfusionfollowing percutaneous coronary interventions. Am J Cardiol 2003;92:930–935.

3. Attubato MJ, Feit F, Bittl JA, Chew D, Lincoff AM. Major hemor-rhage is an independent predictor of 1-year mortality following per-cutaneous coronary intervention: an analysis from REPLACE-2. Am JCardiol 2004;946(suppl 1):39E.

4. Rao SV, O’Grady K, Pieper KS, Granger CB, Newby LK, Van deWerf F, Mahaffey KW, Califf RM, Harrington RA. Impact of bleedingseverity on clinical outcomes among patients with acute coronarysyndromes. Am J Cardiol 2005;96:1200–1206.

5. Cohen DJ, Lincoff AM, Lavelle TA, Chen HL, Bakhai A, Berezin RH,Jackman D, Sarembock IJ, Topol EJ. Economic evaluation of biva-lirudin with provisional glycoprotein IIB/IIIA inhibition versus hepa-rin with routine glycoprotein IIB/IIIA inhibition for percutaneouscoronary intervention: results from the REPLACE-2 trial. J Am CollCardiol 2004;44:1792–1800.

6. Steigen TK, Maeng M, Wiseth R, Erglis A, Kumsars I, Narbute I,Gunnes P, Mannsverk J, Meyerdierks O, Rotevatn S, et al, for theNordic PCI Study Group. Randomized study on simple versus com-plex stenting of coronary artery bifurcation lesions: the Nordic bifur-cation study. Circulation 2006;114:1955–1961.

7. Chieffo A, Bonizzoni E, Orlic D, Stankovic G, Rogacka R, Airoldi F,Mikhail GW, Montorfano M, Michev I, Carlino M, Colombo A.Intraprocedural stent thrombosis during implantation of sirolimus-eluting stents. Circulation 2004;109:2732–2736.

8. Kandzari DE, Berger PB, Kastrati A, Steinhubl SR, Mehilli J, DotzerF, Ten Berg JM, Neumann FJ, Bollwein H, Dirschinger J, Schomig A,for the ISAR-REACT Study Investigators. Influence of treatment du-ration with a 600-mg dose of clopidogrel before percutaneous coronaryrevascularization. J Am Coll Cardiol 2004;44:2133–2136.

9. Patti G, Colonna G, Pasceri V, Pepe LL, Montinaro A, Di Sciascio G.Randomized trial of high loading dose of clopidogrel for reduction ofperiprocedural myocardial infarction in patients undergoing coronaryintervention: results from the ARMYDA-2 (Antiplatelet therapy forReduction of MYocardial Damage during Angioplasty) study. Circu-lation 2005;111:2099–2106.

0. Price MJ, Coleman JL, Steinhubl SR, Wong GB, Cannon CP, TeirsteinPS. Onset and offset of platelet inhibition after high-dose clopidogrelloading and standard daily therapy measured by a point-of-care assay

in healthy volunteers. Am J Cardiol 2006;98:681–684.