特殊 肝素与鱼精蛋白VS比伐

Comparison of Bivalirudin and Unfractionated Heparin Plus Protamine in Patients With Coronary Heart Disease Undergoing Percutaneous Coronary Intervention(from the Antithrombotic Regimens aNd Outcome[ARNO]Trial)Guido Parodi,MD,PhD,Angela Migliorini,MD,Renato Valenti,MD,Benedetta Bellandi,MD, Umberto Signorini,MD,Guia Moschi,MD,Piergiovanni Buonamici,MD,Giampaolo Cerisano,MD,and David Antoniucci,MD*Previous studies have compared bivalirudin and unfractionated heparin(UFH)plus theroutine use of glycoprotein IIb/IIIa inhibitors.They have demonstrated that bivalirudin candecrease bleeding complications without a significant increase in ischemic complications,resulting in a better net clinical outcome,as defined by the efficacy(ischemic complications)or safety(bleeding complications)end point.The aim of the present study was to comparebivalirudin and UFH plus protamine in patients undergoing elective percutaneous coro-nary intervention and pretreated with clopidogrel and aspirin.We randomly assigned850patients with stable or unstable coronary artery disease to bivalirudin or UFH followed byprotamine at the end of the percutaneous coronary intervention.The primary end pointwas in-hospital major bleeding.The main secondary end points were the1-month com-posite of death,myocardial infarction,unplanned target vessel revascularization;and the1-month net clinical outcome.The rate of major bleeding(primary end point)was0.5%inpatients randomized to bivalirudin and2.1%in patients randomized to UFH(p؍0.033).At30days,the rate of major bleeding was0.9%in the bivalirudin arm and2.8%in the UFHarm(p؍0.043).The composite of death,myocardial infarction,and target vessel revas-cularization rate and the net clinical outcome rate was2.8%and6.4%(p؍0.014)and3.3%and7.8%(p؍0.004),respectively,in the bivalirudin and UFH arms.In conclusion,inpercutaneous coronary intervention patients pretreated with clopidogrel and aspirin,biva-lirudin was associated with less major bleeding and fewer ischemic complications and abetter net clinical outcome than UFH.©2010Elsevier Inc.All rights reserved.(Am JCardiol2010;105:1053–1059)Compared to unfractionated heparin(UFH),bivalirudin has2major advantages in terms of bleeding risk,a highly predictable anticoagulant effect at a standardized dose and a short duration of the anticoagulant effect(drug plasma half-life25minutes).The latter can be canceled by the neutral-ization of heparin by protamine at the end of the proce-dure.1,2The aim of the present randomized study was to determine whether bivalirudin remains superior to UFH plus protamine in patients undergoing percutaneous coro-nary intervention(PCI)who have been pretreated with as-pirin and clopidogrel.MethodsAll patients undergoing PCI and pretreated with aspirin(325mg)and a600-mg loading dose of clopidogrelՆ6hours before PCI were considered eligible for enrollment.The exclusion criteria were acute ST-segment elevationmyocardial infarction;PCI for chronic total occlusion;renalinsufficiency(creatinine clearance rateϽ30ml/min or se-rum creatinineϾ3mg/dl)or dependence on renal dialysis;co-morbid conditions with a life expectancy ofϽ1year;active bleeding,bleeding diathesis,or recent major surgery(Ͻ15days);gastrointestinal or genitourinary bleedingwithin the previous6weeks;pretreatment with UFH orlow-molecular-weight heparin or bivalirudin before PCI;uncontrolled hypertensionϾ180/110mm Hg unresponsiveto therapy;relevant hematologic abnormalities(hemoglobin Ͻ10g/dl or platelet countϽ100ϫ109/L);allergy to the study medications;a history of heparin-induced thrombo-cytopenia;and ageϽ18years.The institutional ethics com-mittee approved the study,and all patients provided writteninformed consent.Allocation to treatment was done using acomputer-generated sequence.The patients were assignedequally in an open-label fashion to UFH or bivalirudin,which were started immediately before PCI.UFH was ad-Division of Cardiology,Careggi Hospital,Florence,Italy.Manuscriptreceived September30,2009;revised manuscript received and acceptedDecember1,2009.This study was supported by the investigators,and no company pro-vided anyfinancial support or drugs free of charge;we have no conflicts ofinterest to declare.*Corresponding author:Tel:(39)055-794-7966;fax:(39)055-794-9303.E-mail address:david.antoniucci@virgilio.it(D.Antoniucci).0002-9149/10/$–see front matter©2010Elsevier Inc.All rights doi:10.1016/j.amjcard.2009.12.001ministered as an intravenous bolus of 100IU/kg body weight;additional boluses were administered to achieve an activated clotting time of 250to 300seconds during the entire procedure.At the end of the procedure,protamine sulfate was administered at a dose of 0.5mg/100IU of UFH used.1,2Bivalirudin was administered as an intravenous bolus dose of 0.75mg/kg,followed by infusion of 1.75mg/kg/hour for the duration of the procedure.The use of glycoprotein IIb/IIIa inhibitors (GPIs)was at discretion of the operator,and abciximab was the only GPI allowed.All patients underwent femoral sheath removal immediately after PCI,and closure devices were used routinely.After PCI,dual antiplatelet treatment,including clopidogrel 75mg/day and aspirin 325mg/day was recommended for Ն1year in all patients.The primary end point of the present study was in-hospital major bleeding,defined according to the Random-ized Evaluation of PCI Linking Angiomax to Reduced Clin-ical Events (REPLACE)2trial criteria.3The diagnosis of major bleeding required the presence of intracranial,in-traocular,or retroperitoneal hemorrhage,clinically overt blood loss resulting in a decrease in hemoglobin of Ͼ3g/dl,any decrease in hemoglobin of Ͼ4g/dl,or the transfusion of Ն2units of packed red blood cells or whole blood.Other prespecified end points were (1)the 30-day composite of ischemia (death from any cause,myocardial infarction,un-planned revascularization for ischemia);(2)the 30-day net clinical outcome,defined as the 30-day composite ischemia or 30-day major bleeding;(3)30-day minor bleeding;(4)30-day entry site vascular complications;(5)6-month mor-tality;(6)the 6-month composite of death and myocardial infarction;and (7)6-month unplanned target vessel revas-cularization.Definite and probable stent thrombosis were defined according the Academic Research Consortium cri-teria.4The myocardial infarction definition included the following criteria:electrocardiographic changes consistent with myocardial infarction or cardiac biomarker elevation (creatine kinase-MB or troponin I at one measurement 3times greater than the upper normal limit)5or cardiac bi-omarker re-elevation in patients with pre-PCI values greater than the upper normal limit (Ն50%more than the previous nadir with documentation that the cardiac biomarker levels were decreasing before PCI).Minor bleeding was defined according to the REPLACE 2criteria as clinical overt bleeding that did not meet the criteria for major bleeding.3Entry site vascular complications included all complications requiring percutaneous or surgical intervention or blood transfusion.All events were adjudicated by an event adju-dication committee whose members were unaware of the treatment assignments.The study was designed on the basis of the superiority principle:the primary end point rate for patients allocated to the UFH group would be greater than for patients allocated to bivalirudin.We assumed that the primary end point would occur in 6%of the UFH group and in 2%of the bivalirudin group,for a 66%risk reduction with bivalirudin.The planned enrollment of 850patients provided 91%power for detecting this reduction at an ␣level of 0.05.Categorical variables are expressed as frequencies and pro-portions and were compared using the chi-square test or Fisher’s exact test,as appropriate.Continuous data are sum-marized as the mean ϮSD and were compared using Student’s t test.The Kaplan-Meier method was used to estimate the 6-month event rates,and comparisons between the 2treatment groups were performed using the log-rank statistic.Univariate and multivariate logistic regression analyses were performed to evaluate the independent con-tribution of the clinical,angiographic,and procedural vari-ables to the 30-day end points.Variables with p Ͻ0.10were entered into the multivariate model.The interactiontermsFigure 1.Study flow chart.1054The American Journal of Cardiology ()for subgroup analysis(abciximab use vs no abciximab use) were tested in a logistic regression model.All analyses were performed in a blinded manner regarding the randomly assigned treatment and on an intention-to-treat basis.An␣level of0.05was considered statistically significant.All tests were2sided.The analyses were performed using the Statistical Package for Social Sciences,version11.5(SPSS, Chicago,Illinois).This trial was registered with Clinical Trials gov(num-ber NCT00448461).ResultsFrom October2006to July2008,850patients were enrolled in the present study(Figure1).Table1lists the baseline characteristics.The2groups were well-matched in all baseline characteristics.Among patients with unstable angina,a Thrombolysis In Myocar-dial Infarction risk score ofՆ5was revealed in19%of cases.Table2lists the procedural characteristics.Abcix-imab was used in179patients(21.1%),and its use was more frequent in the UFH arm than in the bivalirudin arm (27.6%vs14.6%,pϭ0.0001).The abciximab-treated pa-tients more frequently had a greater risk profile than to the patients who did not receive the drug(Table3).Among theTable1Baseline characteristicsVariable Bivalirudin(nϭ425)Heparin(nϭ425)p ValueAge(years)68.7Ϯ10.669.1Ϯ10.60.544 Men329(77%)319(75%)0.420 Hypertension249(59%)254(61%)0.489 Diabetes mellitus91(21%)92(22%)0.818 Total cholesterolϾ200mg/dl204(48%)210(51%)0.492 Current smoker76(18%)70(17%)0.675 Body mass index(kg/m2)26.50Ϯ3.0726.13Ϯ3.390.957 Previous myocardial infarction172(41%)158(38%)0.443 Myocardial infarctionՅ30days50(12%)38(9%)0.213 Previous percutaneous coronaryintervention234(55%)233(56%)0.780 Previous coronary surgery42(10%)32(8%)0.267 Serum creatinine(mg/dl)0.99Ϯ0.62 1.02Ϯ0.760.529 Indication for procedureStable angina pectoris190(45%)171(40%)0.291 Unstable angina pectoris117(27%)112(26%)0.844 Positive stress result or other118(28%)142(33%)0.074 Increased baseline troponin I62(15%)79(19%)0.109 Left ventricular ejectionfraction(%)48Ϯ1246Ϯ120.125 Multivessel coronary disease224(53%)242(58%)0.118 Data are expressed as meanϮSD or n(%).Table2Procedural characteristicsVariable Bivalirudin(nϭ425)Heparin(nϭ425)p ValueRotational atherectomy7(2%)10(2%)0.435 Intra-aortic balloon pump3(1%)2(1%)0.654 Stenting360(85%)373(88%)0.196 Bare-metal stent38(9%)45(11%)0.419 Drug-eluting stent322(76%)328(77%)0.628 Abciximab62(15%)117(28%)0.0001 Activated clotting time(seconds)327Ϯ36297Ϯ410.0001 Fluoroscopic time(minutes)10.6Ϯ10.310.7Ϯ9.10.779 Femoral arterial access418(98%)415(98%)0.462 Femoral hemostasis0.516 Closure device380(91%)374(91%)Manual compression45(9%)51(10%)Treated coronary vessels(n)595589Vessel location0.277 Left main28(5%)31(5%)Left anterior descending259(43%)257(44%)Right134(23%)148(25%)Left circumflex156(26%)141(24%)Venous bypass graft18(3%)12(2%)Data are expressed as meanϮSD or n(%).Table3Baseline characteristics stratified by abciximab useVariable Abciximab p ValueNo(nϭ671)Yes(nϭ179)Age(years)69Ϯ1269Ϯ110.985 Diabetes mellitus137(21%)46(26%)0.162 Previous myocardial infarction281(43%)48(27%)0.0001 Unstable angina pectoris149(23%)80(45%)0.0001 Raised baseline troponin I88(14%)53(30%)0.0001 Multivessel coronary disease349(53%)116(65%)0.004 Left main intervention27(4%)20(11%)0.0001 Data are presented as meanϮSD or n of patients(%).Table4One-month clinical outcomesVariable Bivalirudin(nϭ425)Heparin(nϭ425)p ValueMajor bleedingIn-hospital2(0.5%)9(2.1%)0.033 1Month4(0.9%)12(2.8%)0.043 Composite end point*12(2.8%)27(6.4%)0.014 Death1(0.2%)6(1.4%)0.069†Myocardial infarction11(2.4%)20(4.5%)0.098 Q-wave11Non–Q-wave1019Definite stent thrombosis2(0.5%)1(0.2%)0.563 Acute10Subacute11Target vessel revascularization2(0.4%)3(0.7%)0.686†Net clinical outcome14(3.3%)33(7.8%)0.004 Minor bleeding10(2.4%)10(2.4%)0.997 Vascular access complications5(1.2%)8(1.9%)0.399 Abciximab-treated patients(nϭ179)62117Major bleeding1(1.6%)4(3.4%)0.660†Composite end point*4(6.5%)7(6.0%)0.901 Net clinical outcome5(8.1%)11(9.4%)0.765 Non–abciximab-treated patients(nϭ671)363308Major bleeding3(0.8%)8(2.6%)0.072 Composite end point*8(2.2%)21(6.8%)0.003 Net clinical outcome9(2.5%)23(7.5%)0.003 Data are expressed as n(%).*Death,myocardial infarction,target vessel revascularization.†Fisher’s exact test.1055Coronary Artery Disease/ARNO Trialpatients who received abciximab,no differences were found between those randomized to UFH or bivalirudin.Most patients in both arms had immediate successful entry site hemostasis using closure devices.No patient in the UFH arm experienced negative reactions,including no allergic reactions,after protamine administration.The in-hospital bleeding and 30-day clinical outcomes are listed in Table 4.The 1-month follow-up rate was 100%.The rate of major bleeding (primary end point)was 0.5%in the bivalirudin arm and 2.1%in the UFH arm (p ϭ0.033).In-hospital major bleeding events included 1case of peri-cardial bleeding and 1case of gastrointestinal bleeding in the bivalirudin arm and 1case each of fatal rupture of an abdominal aortic aneurysm,retroperitoneal hematoma,and pericardial bleeding and 6blood transfusions in the UFH arm.At 30days,the major bleeding rate had increased to 0.9%in the bivalirudin arm and 2.8%in the UFH arm (p ϭ0.043).The difference between the 2arms did not result from the more frequent use of abciximab in the UFH arm because among the patients who did not receive GPI treat-ment,the major bleeding rate was more than threefold greater in the UFH arm than in the bivalirudin arm (2.6%and 0.8%,respectively).No major bleeding occurred from the vascular access site after sheath removal.No patient with major bleeding required discontinuation of clopidogrel treatment.In both arms,1/2of the patients with major bleeding had had an adverse ischemic event at 30days of follow-up.The rate of the 30-day composite ischemic end point was greater in the UFH arm than in the bivalirudin arm.The difference between the 2arms was driven mainly by death and periprocedural non–Q-wave myocardial infarction.The only death in the bivalirudin arm occurred in 1patient with distal left main disease and chronic occlusion of the right coronary artery.That patient died from refractory cardiac arrest during the PCI procedure.In the UFH arm,6patients died.Death was due to fatal bleeding in 1patient,conges-tive heart failure in 1patient (Killip class 3on admission,with successful multivessel PCI of venous graft and native vessel lesions),and probable stent thrombosis in 4patients (4sudden deaths after discharge).Three definite stent thromboses occurred.Acute stent thrombosis occurred in 1patient randomized to bivalirudin.Subacute stent thrombo-sis occurred in 1patient in the bivalirudin arm (5days after PCI)and 1patient in the UFH arm (10days after PCI).All 3definite stent thromboses were complicated by myocardial infarction despite successful emergency repeat PCI.No differences were found between the 2groups in minor bleeding and vascular entry site complications.The net clinical outcome was better in the bivalirudin arm than in the UFH arm.On multivariable analysis,the independent predictors of 30-day ischemic events were in-hospital major bleeding (odds ratio 22.0,95%confidence interval 7.5to 65.0,p ϭ0.0001),multivessel disease (odds ratio 2.4,95%confi-18015012090603020181614121086420Days after randomizationI n c i d e n c e (%)HeparinBivalirudinp= 0.008Randomization Number at risk Bivalirudin 425414413412411410408Heparin425400395394393393391Six-Month Major Bleeding18015012090603020181614121086420Days after randomizationI n c i d e n c e (%)HeparinBivalirudinp= 0.021Six-Month Composite Ischemic End PointRandomization Number at risk Bivalirudin 425406404397393392387Heparin42539138838338137837218015012090603020181614121086420Days after randomizationI n c i d e n c e (%)HeparinBivalirudinp= 0.003Six-Month Net Clinical OutcomeRandomization Number at risk Bivalirudin 425404402395391390384Heparin425386380374372368362ABCFigure 2.Kaplan-Meier analysis of major bleeding (A),composite of death,myocardial infarction,and target vessel revascularization (B),and net clinical outcome (C)for patients randomized to bivalirudin or UFH plus protamine.1056The American Journal of Cardiology ()dence interval1.1to5.3,pϭ0.031),and age(odds ratio 1.04,95%confidence interval1.00to1.07,pϭ0.049).The effect of abciximab did not differ according to ran-domization(pϭNS for all outcome measures).The6-month clinical outcomes are summarized in Figure 2and Table5.The follow-up rate was98.5%.The differ-ences between the2arms in the primary and secondary end point were maintained at6months,with a major bleeding rate of4.5%in the UFH arm and1.4%in the bivalirudin arm(pϭ0.008)and a composite ischemic rate of12.4% and7.6%(pϭ0.021),respectively.The net clinical out-come was better in the bivalirudin arm than in the UFH arm (8.4%and14.8%,respectively;pϭ0.003).DiscussionThe present study compared bivalirudin and UFH plus protamine in patients undergoing PCI and pretreated with aspirin and a600-mg loading dose of clopidogrel.Our results have shown that bivalirudin is superior to UFH in terms of both major bleeding complications and ischemic complications and provides a better net clinical outcome at 30days.Also,this benefit was maintained at6months.The only trial comparing the2anticoagulant agents, UFH and bivalirudin,in patients undergoing elective PCI in the modern stent era without the use of GPIs was the Intracoronary Stenting and Antithrombotic Regimen:Rapid Early Action for Coronary Treatment(ISAR-REACT)-3 trial,which excluded patients with increased troponin lev-els.The REPLACE-2and Acute Catheterization and Urgent Intervention Triage strategY(ACUITY)-PCI trials com-pared2different antithrombotic strategies with the routine use of GPIs in patients randomized to UFH.3,6,7The Anti-thrombotic Regimens aNd Outcome(ARNO)trial,in con-trast to these studies,directly compared the2anticoagulant agents,heparin and bivalirudin,allowed the liberal use of a unique GPI(abciximab),included neutralization of UFH with protamine to reduce bleeding complications,and used, as the only eligibility criterion,PCI in patients pretreated with600mg of clopidogrelՆ6hours before the interven-tion,whatever the risk of the procedure.Major bleeding complications occurred at a lower than expected rate in both arms and were less than the rates reported by the REPLACE-2trial,the ACUITY-PCI sub-study,and the ISAR-REACT-3trial.3,6,7The REPLACE-2 trial,which included mostly patients with stable angina or low-risk unstable angina,reported a rate of major bleeding in the bivalirudin-alone arm of2.4%and4.1%in the hep-arin plus GPI arm.3In the ACUITY-PCI substudy,which included patients with acute coronary syndromes(69%of patients had a Thrombolysis In Myocardial Infarction risk score ofϽ5),the rate of major hemorrhagic complications according to the REPLACE-2definition was3%in the bivalirudin-alone arm and4%in the heparin plus GPI arm.7 The ISAR-REACT-3trial reported major bleeding rates similar to those reported by the REPLACE2trial(3.1%in the bivalirudin arm and4.6%in the UFH arm).6 The routine use of closure devices in the ARNO study could explain,in part,the low rate of major bleeding.The use of closure devices achieved successful immediate he-mostasis after the procedure in most patients,resulting in no major bleeding from vascular access in either arm.In con-trast,vascular access hemorrhage accounted for60%and 35%of the major bleeding reported by the REPLACE-2 trial in the heparin and bivalirudin arms,respectively.3In that study,closure devices were used in only23%of pa-tients,and delayed sheath removal was an independent predictor of major bleeding.8The rate of major bleeding complicating vascular access hemorrhage was not reported by the ACUITY-PCI substudy or the ISAR-REACT-3trial. In the latter study,closure devices were used in only10%of the patients.6The neutralization of UFH with protamine at the end of PCI could explain the relatively low rate of bleeding in the patients randomized to heparin.Nevertheless,the in-hospi-tal major bleeding rate was more than fourfold greater in the UFH arm than in to the bivalirudin arm,suggesting that the superiority of bivalirudin in terms of bleeding hazard is not related to its short half-life.Thisfinding is even more relevant considering that the more frequent use in the UFH arm of abciximab did not have a signif-icant effect on the revealed difference in the major bleed-ing rate between the2arms.With regard to the secondary efficacy composite end point of death,myocardial infarction,and unplanned revas-cularization,in the present study,fewer ischemic complica-tions occurred in the bivalirudin arm than in the UFH arm. Thisfinding could be considered unexpected for2reasons. First,previous studies showed that bivalirudin was not as-sociated with a decrease in the hazard of30-day ischemic events compared with heparin with or without a GPI.Sec-ond,in the present trial,the patients randomized to UFH more frequently received abciximab,which has a strong protective effect against ischemic complications.9–12How-ever,in the REPLACE-2trial and the heparin plus GPI arm of the ACUITY trial,bivalirudin resulted in ischemic rates comparable to those in the control arm,despite the use of a GPI in all patients randomized to UFH,which is known to reduce adverse ischemic event.Bivalirudin also did not reduce the incidence of ischemia in the ISAR-REACT-3 trial.However,the lack of a protective effect against isch-emic complications might have resulted from the much greater dose of heparin used in that study than was used in the ARNO trial,as well as the use of protamine in theTable5Six-month clinical outcomesVariable Bivalirudin(nϭ419)Heparin(nϭ418)p ValueMajor bleeding6(1.4%)19(4.5%)0.008 Composite end point*32(7.6%)52(12.4%)0.021 Death5(1.2%)10(2.4%)0.193 Myocardial infarction14(3.3%)24(5.7%)0.095 Q-wave33Non–Q-wave1121Peak creatine-kinase(U/L)864Ϯ806953Ϯ14290.832 Death and myocardial infarction18(4.2%)31(7.3%)0.055 Target vessel revascularization17(4.1%)24(5.7%)0.259 Net clinical outcome35(8.4%)62(14.8%)0.003 Data are expressed as number(%).*Death,myocardial infarction,target vessel revascularization.1057Coronary Artery Disease/ARNO TrialARNO trial,which further shortened any potential protec-tive effect of heparin.Moreover,in contrast to the ISAR-REACT3trial,the ARNO trial also included troponin-positive patients,who are at a greater risk of ischemic complications.Similar to the incidence of major bleeding complications,in the present study,the number of ischemic events was lower than those reported by the REPLACE2, ACUITY-PCI substudy,and ISAR-REACT3trials.3,6,7It cannot be excluded that the low rate of ischemic events in both arms was the result of the liberal use of abciximab, which was far more frequent than in previous studies.We could not definitely exclude a possible relation be-tween the excess number of deaths in the protamine arm and the use of the drug.However,previous randomized studies have shown the safety of protamine administration imme-diately after PCI.1,2Moreover,it is unlikely that in the present study the excess number of ischemic complications in the UFH arm was related to protamine administration. First,no thrombotic event occurred in thefirst24hours after the procedure.Second,the only definite stent thrombosis occurred10days after PCI.Third,the4probable stent thromboses(sudden deaths)occurred several days after hos-pital discharge.Fourth,none of the periprocedural myocar-dial infarctions were associated with ST-segment elevation or abrupt closure of the target vessel.Fifth,it is unlikely that neutralization of UFH by protamine at the end of the procedure would result in an ischemic event in the sub-sequent days.Sixth,a prothrombotic rebound effect of protamine has not been proved.In contrast,in pharma-cologic terms,the rebound effect of protamine would mean a recurrent state of anticoagulation owing to the late release of UFH bounded to albumin,and this is a well-known phenomenon in cardiac surgery.A recent study has shown that protamine reduces plasma thrombin generation and that this effect is mediated through the inhibition of factor V activation.13Finally,the ischemic complication rate in the UFH arm of the ARNO trial was lower than the ischemic complication rates reported in previous studies comparing bivalirudin and UFH.3,6,7 The differing results of the cited studies cannot be easily explained,because the comparison of the present study with the previous3trials was difficult owing to the differences in criteria used for enrollment and in the assigned treatments. The hypothesis that our study cohort of patients was at a lower risk of periprocedural ischemic events than the pop-ulations of the other trials seems unlikely.Advanced age is a strong predictor of a poor outcome,and our trial popula-tion was several years older than the REPLACE2and ACUITY populations.3,7The ISAR-REACT-3population was2years younger than the population of the present study and those with increased baseline troponin levels were ex-cluded from the trial.6A possible explanation of the small number of ischemic events in the present study is the very low rate of major bleeding.One half of the patients with major bleeding experienced an ischemic event,and multi-variate analysis showed major bleeding to be an indepen-dent predictor of the1-month ischemic events.Thesefind-ings are consistent with those reported from the REPLACE 2and ACUITY trials,which have shown major bleeding to be a strong independent predictor of cardiac ischemic events at30days.8,14It has been hypothesized that in patients with bleeding,the ischemic event could have been precipitated by platelet activation secondary to the bleeding or by the need for antithrombotic treatment discontinuation.15 In addition to the very low bleeding rate in the bivaliru-din group,the difference in the ischemic event rate between the bivalirudin and UFH groups could also have resulted because bivalirudin administration during PCI in patients pretreated with600mg clopidogrel provides additional in vitro suppression of platelet inhibition.In contrast,UFH does not have significant influence on in vitro platelet ag-gregation.16Because of the fewer major bleeding events and isch-emic events in patients randomized to bivalirudin,the 1-month net clinical outcome was much better in the bivalirudin arm than in the UFH arm,and this benefit was maintained at6months.The open-label design might have introduced a potential bias in the decision to administer abciximab before PCI. However,the more frequent use of abciximab in the UFH arm did not result in a lower rate of ischemic events,and the difference in the primary end point between the2study arms was not driven by the associated increase in the bleed-ing rate of the abciximab-treated patients.1.Briguori C,Di Mario C,De Gregorio J,Sheiban I,Vaghetti M,Colombo A.Administration of protamine after coronary stent deploy-ment.Am Heart J1999;138:64–68.2.Thuesen L,Andersen HR,Botker HE,Kristensen SD,Krusell LR,Lassen JF.In laboratory femoral sheath removal after heparin reversal by protamine after percutaneous coronary intervention.EuroInterv 2005;1:66–70.3.Lincoff AM,Bittl JA,Harrington RA,Feit F,Kleiman NS,Jackman JD,Sarembock IJ,Cohen DJ,Spriggs D,Ebrahimi R,Keren G,Carr J,Cohen EA,Betriu A,Desmet W,Kereiakes DJ,Rutsch W,Wilcox RG,de Feyter PJ,Vahanian A,Topol EJ;REPLACE-2investigators.Bivalirudin and provisional glycoprotein blockade compared with heparin and planned glycoprotein blockade during percutaneous coronary intervention: REPLACE-2randomized trial.JAMA2003;289:853–863.4.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 Consor-tium.Clinical end points in coronary stent trials:a case for standard-ized definitions.Circulation2007;115:2344–2351.5.Thygesen K,Alpert JS,White HD,Jaffe AS,Apple FS,Galvani M,Katus HA,Newby LK,Ravkilde J,Chaitman B,Clemmensen PM, Dellborg M,Hod H,Porela P,Underwood R,Bax JJ,Beller GA, Bonow R,Van der Wall EE,Bassand JP,Wijns W,Ferguson TB,Steg PG,Uretsky BF,Williams DO,Armstrong PW,Antman EM,Fox KA, Hamm CW,Ohman EM,Simoons ML,Poole-Wilson PA,Gurfinkel EP,Lopez-Sendon JL,Pais P,Mendis S,Zhu JR,Wallentin LC, Fernández-Avilés F,Fox KM,Parkhomenko AN,Priori SG,Tendera M,Voipio-Pulkki LM,Vahanian A,Camm AJ,De Caterina R,Dean V,Dickstein K,Filippatos G,Funck-Brentano C,Hellemans I,Kris-tensen SD,McGregor K,Sechtem U,Silber S,Tendera M,Widimsky P,Zamorano JL,Morais J,Brener S,Harrington R,Morrow D,Lim M, Martinez-Rios MA,Steinhubl S,Levine GN,Gibler WB,Goff D, Tubaro M,Dudek D,Al-Attar N;Joint ESC/ACC/AHA/WHF Task Force for the Redefinition of Myocardial Infarction.Universal defini-tion of myocardial infarction.Circulation2007;116:2634–2653.6.Kastrati A,Neumann F-J,Mehilli J,Byrne RA,Iijima R,Büttner HJ,Khattab AA,Schulz S,Blankenship JC,Pache J,Minners J,Seyfarth M,Graf I,Skelding KA,Dirschinger J,Richardt G,Berger PB, Schömig A;ISAR-REACT3Trial Investigators.Bivalirudin versus unfractionated heparin during percutaneous coronary intervention.N Engl J Med2008;359:688–696.7.Stone GW,White HD,Ohman EM,Bertrand ME,Lincoff AM,McLaurin BT,Cox DA,Pocock SJ,Ware JH,Feit F,Colombo A, Manoukian SV,Lansky AJ,Mehran R,Moses JW;Acute Catheter-ization and Urgent Intervention Triage strategy(ACUITY)Trial In-1058The American Journal of Cardiology()。