Articles. Methods Study design and patients The Ongoing Tirofiban in Myocardial Evaluation (On-TIME) 2 trial was a prospective, multicentre, placebo-

Transcription

1 Prehospital initiation of tirofiban in patients with ST-elevation myocardial infarction undergoing primary angioplasty (On-TIME 2): a multicentre, double-blind, randomised controlled trial Arnoud W J van t Hof, Jurriën ten Berg, Ton Heestermans, Thorsten Dill, Reinhard C Funck, Wouter van Werkum, Jan-Henk E Dambrink, Harry Suryapranata, Gert van Houwelingen, Jan Paul Ottervanger, Pieter Stella, Evangelos Giannitsis, Christian Hamm, on behalf of the Ongoing In Myocardial infarction Evaluation (On-TIME) 2 study group* Summary Background The most effective magnitude and timing of antiplatelet therapy is important in patients with acute ST-elevation myocardial infarction (STEMI). We investigated whether the results of primary coronary angioplasty (PCI) can be improved by the early administration of the glycoprotein IIb/IIIa blocker tirofiban at first medical contact in the ambulance or referral centre. Methods We undertook a double-blind, randomised, placebo-controlled trial in 24 centres in the Netherlands, Germany, and Belgium. Between June 29, 2006, and Nov 13, 2007, 984 patients with STEMI who were candidates to undergo PCI were randomly assigned to either high-bolus dose tirofiban (n=491) or placebo (N=493) in addition to aspirin (500 mg), heparin (5000 IU), and clopidogrel (600 mg). Randomisation was by blinded sealed kits with study drug, in blocks of four. The primary endpoint was the extent of residual ST-segment deviation 1 h after PCI. Analysis was by intention to treat. The trial is registered, number ISRCTN Findings 936 (95%) patients were randomly assigned to treatment after a prehospital diagnosis of myocardial infarction in the ambulance. Median time from onset of symptoms to diagnosis was 76 min (IQR ). Mean residual ST deviation before PCI (10 9 mm [SD 9 2] vs 12 1 mm [9 4], p=0 028) and 1 h after PCI (3 6 mm [4 6] vs 4 8 mm [6 3], p=0 003) was significantly lower in patients pretreated with high-bolus dose tirofiban than in those assigned to placebo. The rate of major bleeding did not differ significantly between the two groups (19 [4%] vs 14 [3%]; p=0 36). Interpretation Our finding that routine prehospital initiation of high-bolus dose tirofiban improved ST-segment resolution and clinical outcome after PCI, emphasises that further platelet aggregation inhibition besides high-dose clopidogrel is mandated in patients with STEMI undergoing PCI. Funding Merck (USA). Introduction Early and complete restoration of blood flow in the infarct-related vessel after acute myocardial infarction is associated with improved survival and clinical outcomes. 1 3 Both short-term and long-term outcomes are better after primary percutaneous coronary angioplasty (PCI) than after fibrinolytic therapy for acute myocardial infarction. 4 6 The benefits of PCI have mostly been attributed to accelerated and improved myocardial reperfusion. 3 Platelet activation and aggregation have a crucial role in the cascade of events leading to myocardial infarction, and in the thrombotic complications that can accompany PCI. Insufficient inhibition of platelet aggregation at the time of PCI correlates with increased likelihood of major cardiovascular adverse events after such a procedure. 7 Antiplatelet therapy, therefore, is an important component of medical therapy for patients with acute myocardial infarction. Triple antiplatelet treatment including aspirin, a thienopyridine, and a glycoprotein IIb/IIIa blocker is recommended for high-risk patients with non-st-segment elevation acute coronary syndromes. 8 Although many studies have reported beneficial results of routine treatment with a glycoprotein IIb/IIIa blocker shortly before PCI, 9,10 the additional value of early prehospital IIb/IIIa blockade in addition to dual antiplatelet therapy, including high-dose clopidogrel in patients with acute myocardial infarction, has not been established. Similarly, the most effective timing of administration of antiplatelet and antithrombotic drugs in this setting is still being investigated. 11,12 We investigated whether the results of PCI can further be improved by the early administration of the glycoprotein IIb/IIIa blocker tirofiban at first medical contact in the ambulance or referral centre. Methods Study design and patients The Ongoing in Myocardial Evaluation (On-TIME) 2 trial was a prospective, multicentre, placebo- Lancet 2008; 372: See Comment page 509 *Members listed in webappendix Isala Klinieken, Department of Cardiology, Zwolle, Netherlands (A W J van t Hof MD, T Heestermans MD, J-H E Dambrink MD, H Suryapranata MD, J P Ottervanger MD); St Antonius Ziekenhuis, Nieuwegein, Netherlands (J ten Berg MD, W van Werkum MD); Kerckhoff-Klinik GmbH, Bad Nauheim, Germany (T Dill MD, Prof C Hamm MD); Diagram BV, Zwolle, Netherlands, (H Suryapranata MD); Medisch Spectrum Twente, Enschede, Netherlands (G van Houwelingen MD); Utrecht Medisch Centrum, Utrecht, Netherlands (P Stella MD); Philipps Universität, Marburg, Germany (R C Funck MD); and Universitätsklinikum Heidelberg, Heidelberg, Germany (E Giannitsis MD) Correspondence to: Arnoud W J van t Hof, Isala Klinieken, Locatie Weezenlanden, Department of Cardiology, Groot Wezenland 20, 8011 JW Zwolle, Netherlands v.r.c.derks@isala.nl Vol 372 August 16,

2 984 patients enrolled 984 randomly assigned 491 assigned to pretreatment with tirofiban 493 assigned to pretreatment with placebo 9 did not undergo angiography 4 died 5 clear false positive diagnoses 5 did not undergo angiography because of a clear false positive diagnosis 482 underwent angiography 488 underwent angiography 14 CABG indicated (11 assessed for primary endpoint) 46 treated conservatively (31 assessed for primary endpoint) 13 CABG indicated (9 assessed for primary endpoint 36 treated conservatively (29 assessed for primary endpoint) 422 underwent PCI 439 underwent PCI 40 could not be assessed for primary endpoint (but included in clinical follow-up) 4 died on day 0 9 false positive diagnoses 1 resuscitated 2 urgent surgery 2 early transportation back to referral hospital 15 missing ECG 7 ECG not analysable 38 could not be assessed for primary endpoint (but included in clinical follow-up) 4 died on day 0 7 false positive diagnoses 3 urgent surgery 1 early transportation back to referral hospital 15 missing ECG 8 ECG not analysable 451 assessed for primary endpoint 455 assessed for primary endpoint 14 refused written consent or withdrew consent 4 lost to folllow-up 13 refused written consent or withdrew 3 lost to follow-up 473 had clinical follow-up 477 had clinical follow-up Figure 1: Trial profile PCI=percutaneous coronary intervention. CABG=coronary artery bypass grafting. controlled, randomised, clinical trial. The rationale and design of the study has been described previously. 13 Study enrolment was from June 29, 2006, to Nov 13, 2007, at 24 participating centres in three countries (the Netherlands, Germany, and Belgium). The study population consisted of patients with ST-elevation myocardial infarction (STEMI) who were candidates to undergo primary PCI. Eligible patients were men and women aged years with symptoms of acute myocardial infarction of more than 30 min but less than 24 h and ST-segment elevation of more than 1 mv in two adjacent electrocardiograph (ECG) leads. Exclusion criteria were known severe renal dysfunction (glomerular filtration rate <30 ml/min or serum creatinine >200 mmol/l [>2 5 mg/dl]), therapy resistant cardiogenic shock (systolic blood pressure 80 mm Hg for >30 min), persistent severe hypertension (systolic pressure >180 mm Hg or diastolic pressure >110 mm Hg), or a contraindication to anticoagulation or increased risk of bleeding. We also excluded patients with a left bundle branch block, pregnant women or women who were breastfeeding, and patients with a life expectancy of less than 1 year Vol 372 August 16, 2008

3 Either the physician or intensive-care nurse obtained written informed consent in the ambulance or referring hospital. The study protocol was approved by all local ethics committees involved. Procedures Patients were randomly assigned to prehospital treatment with tirofiban (25 μg/kg bolus and 0 15 μg/kg/min maintenance infusion for 18 h) or placebo (bolus plus infusion) by blinded sealed kits with study drug. All staff and study personnel were blinded to treatment. Kits were distributed among the ambulance services or referring centres in blocks of four (two containing active drug and two placebo). In the ambulance or referring centre, all patients also received a bolus of 5000 IU of unfractionated heparin intravenously (Heparine Choay, Sanofi-Aventis, Paris, France) together with aspirin 500 mg intravenously (Aspegic, Sanofi-Aventis, Maassluis, the Netherlands) and a 600 mg loading dose of clopidogrel orally (Plavix, Sanofi-Winthrop Industry, Quetigny, France). Before PCI, additional unfractionated heparin was only given if the activated clotting time was less than 200 seconds. Coronary angiography and PCI were done according to each institution s guidelines and standards. Before, during, or after PCI, bail-out tirofiban could be given for the following indications: decrease in thrombolysis in myocardial infarction (TIMI) flow grade (TIMI flow grades of 0 2 or slow reflow), dissection with decreased flow, distal embolisation, side-branch closure, abrupt closure of the culprit vessel, clinical instability, and prolonged ischaemia. Bail-out tirofiban was also given as the high-bolus dose (25 μg/kg bolus). We included blinded bail-out bolus vials in the medication kit to maintain blinding of the initial treatment assignment and to match with the content of the primary infusion. When a bail-out bolus vial was given, the study infusion was replaced by open-label tirofiban maintenance infusion. The primary efficacy endpoint was the extent of residual ST-segment deviation at 1 h after PCI, as previously described. 14 The sum of ST-segment deviation in all 12 leads was measured 20 ms after the end of the QRS complex with a caliper. We calculated both residual ST segment-deviation on the single ECG after PCI and ST-segment deviation resolution from paired ECGs. We divided patients into four groups of residual ST-segment deviation (0 mm: normalised ST segment and no residual ST-segment deviation; 1 3 mm; 4 6 mm; and >6 mm) and into three groups of ST-segment deviation resolution (complete: >70% resolution; partial: >30% but <70% resolution; and no resolution: <30% resolution), as described previously. 14,15 The key secondary endpoint was the composite of death, recurrent myocardial infarction, urgent target vessel revascularisation (TVR), or blinded bail-out use of tirofiban at 30 days. Other secondary endpoints were TIMI grade 3 flow of the infarct related vessel at initial angiography and distal embolisation after PCI. An independent core lab (Diagram BV, Zwolle, Netherlands) analysed all ECGs and angiograms. A blinded, independent clinical endpoint committee adjudicated all clinical endpoints apart from death. Death was defined as all-cause mortality. Recurrent myocardial infarction was defined as a new increase of creatine kinase MB that was three or more times the upper limit of normal, present in two separate blood samples, and accompanied by chest pain or ECG changes. Early recurrent myocardial infarction was defined as a decrease in creatine kinase MB of at least 50% of the upper limit of normal from a previous peak concentration to a valley, followed by a new increase with a value above the sum of the preceding valley and three times the upper limit of normal. Urgent TVR during admission to hospital was defined as a new (n=493) (n=491) Age (years) 62 0 (12 0) 61 6 (11 8) Men 370/493 (75 1%) 376/491 (76 6%) BMI (kg/m 2 ) 26 7 (3 9) 26 8 (3 5) Current smoker 240/488 (49 2%) 219/488 (44 9%) Anterior infarct location 185/441 (42 0%) 182/426 (42 7%) Diabetes 54/493 (11 0%) 59/490 (12 0%) Hypertension 170/493 (34 5%) 166/491 (33 8%) Hypercholesterolaemia 124/492 (25 2%) 140/490 (28 6%) Family history 192/487 (39 4%) 184/483 (38 1%) Previous MI 37/492 (7 5%) 46/489 (9 4%) Previous PCI 37/493 (7 5%) 50/491 (10 2%) Previous CABG 9/493 (1 8%) 8/491 (1 6%) Previous CVA 11/493 (2 2%) 6/490 (1 2%) Inclusion in ambulance 472/493 (95 7%) 464/491 (94 5%) Killip class >1 65/485 (13 4%) 54/480 (11 3%) Systolic BP (mm Hg) (25 0) (23 7) Diastolic BP (mm Hg) 77 0 (15 1) 77 0 (14 7) Heart rate (beats/min) 75 3 (18 6) 74 9 (16 4) Cumulative ST-deviation diagnostic ECG (mm) 14 3 (9 1) 14 5 (9 1) False positive diagnosis 29/493 (5 9%) 31/491 (6 3%) Ischaemic time (min) 167 ( ) 165 ( ) Time SO to diagnosis (min) 79 (45 156) 72 (44 141) Angiography done 488/493 (99 0%) 482/491 (98 2%) Three-vessel disease 85/488 (17 4%) 84/482 (17 4%) IRV LAD 185/457 (40 5%) 193/450 (42 9%) RCX 53/457 (11 6%) 51/450 (11 3%) RCA 213/457 (46 6%) 202/450 (44 9%) Graft 4/457 (0 9%) 3/450 (0 7%) Left main 2/457 (0 4%) 1/450 (0 2%) Data are n/n (%), mean (SD), or median (IQR). BMI=body-mass index. MI=myocardial infarction. PCI=percutaneous coronary intervention. CABG=coronary artery bypass grafting. CVA=cerebrovascular attack. BP=blood pressure. SO=symptom onset. IRV=infarct-related vessel. LAD=left anterior descending artery. RCX=ramus circumflex artery. RCA=right coronary artery. Table 1: Baseline and angiographic characteristics Vol 372 August 16,

4 episode of ischaemic signs or symptoms at rest with documentation of a new ST-segment shift of 0 05 mv (0 5 mm) or greater on a 12-lead ECG that needed an unplanned coronary intervention or coronary artery bypass grafting. The safety endpoints of interest included the rates of haemorrhage, transfusions, stroke, thrombocytopenia, and serious adverse events. An independent Data Safety Monitoring Committee was responsible for identification of safety issues and suggestion of any recommendation regarding modification or termination of the study to the Executive Committee. Bleeding was assessed with the TIMI criteria. 16 We defined major bleeding as clinical overt signs of haemorrhage associated with a drop in haemoglobin by more than 50 g/l (or, when haemoglobin assessment was not available, a decrease in haematocrit of >15%). For patients undergoing coronary artery bypass graft surgery, we also assessed the rate of surgical re-exploration for bleeding and the postoperative volume of blood loss. Follow-up information was derived from outpatient clinic visits or via contact by telephone at 30 days. Statistical analysis Sample size calculation was based on the assumption that residual ST-segment deviation of more than 3 mm at 1 h after PCI was present in 50% of unselected patients with myocardial infarction who had not received glycoprotein IIb/IIIa blocker treatment in addition to aspirin, heparin, and clopidogrel. 14 Treatment with the high-bolus dose of tirofiban was assumed to decrease the incidence of residual ST deviation of more than 3 mm by 20% (relative), from 50% to 40%. On the basis of these assumptions, with 80% power and α=0 05, 814 patients p value Cumulated ST deviation before angiography 12 1 (9 4)(n=409) 10 9 (9 2) (n=380) (at PCI centre [mm]) ST-resolution diagnosis before angiography Complete resolution 61/396 (15 4%) 76/367 (20 7%) Partial resolution 74/396 (18 7%) 72/367 (19 6%) No resolution 261/396 (65 9%) 219/367 (59 7%) Cumulated ST deviation 1 h after angiography/pci (mm) 4 8 (6 3)(n=455) 3 6 (4 6) (n=451) Residual ST deviation >3 mm 1 h after angiography/pci 205/455 (45 1%) 172/451 (38 1%) Residual ST deviation 1 h after angiography/pci Normalised ST segment 132/455(29 0%) 160/451 (35 5%) 1 3 mm 118/455 (25 9%) 119/451 (26 4%) 4 6 mm 92/455 (20 2%) 85/451 (18 8%) >6 mm 113/455 (24 8%) 87/451 (19 3%) ST resolution diagnosis 1 h after angiography/pci Complete 264/440 (60 0%) 286/436 (65 6%) Partial 104/440 (23 6%) 100/436 (22 9%) No 72/440 (16 4%) 50/436 (11 5%) Data are mean (SD) or n/n (%). PCI=percutaneous coronary intervention. Table 2: ECG parameters on arrival at PCI centre, and after angiography or PCI (407 in each group) were needed to show superiority of tirofiban pretreatment. To account for incomplete or uninterpretable ECG data and false positive ECG diagnoses for about 15% of patients, we calculated that 958 patients needed to be randomly assigned. All analyses were done according to the intentionto-treat principle. We analysed the primary and key secondary efficacy variable with the χ² test or Fisher s exact test. Additionally, we used the χ² test for trend to analyse the percentages of patients in each of the four prespecified groups of residual ST-segment deviation and in each of the three groups of ST-segment resolution, TIMI flow, or myocardial blush grade. All p values were two-sided. Time-to-event analysis was done to describe survival free from major adverse cardiac events from the date when the patient was admitted to the hospital with SPSS for Windows (version 15.0). Survival was represented by Kaplan-Meier curves. This study is registered, number ISRCTN Role of the funding source The study was investigator initiated. The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. Results Figure 1 shows the trial profile. 491 patients were randomly assigned to tirofiban and 493 to placebo. Baseline clinical and angiographic characteristics of the patients were much the same between the groups (table 1). 936 (95%) patients were included and randomly assigned after prehospital diagnosis of myocardial infarction in the ambulance at a median of 75 min (IQR ) after the onset of symptoms. Median time from symptom onset to diagnosis in the 48 (5%) patients who were referred from a non-pci centre was 120 min (IQR ).The median transportation distance to the nearest PCI centre was 25 km (range 1 113) and the median time from start of antithrombotic pretreatment (heparin/aspirin/clopidogrel) and study drug until angiography was 55 min (IQR 43 70). 60 (6%) patients had a false positive diagnosis of myocardial infarction and were evenly distributed between the groups. On arrival at the PCI centre, a second ECG was made before angiography at a median of 31 min (IQR 19 44) after the start of the study drug. The total amount of residual ST-segment deviation was significantly lower in patients who were pretreated with tirofiban than in those assigned to placebo (10 9 mm [SD 9 2] vs 12 1 mm [9 4]; p=0 028). Furthermore, ST-segment resolution was improved before angiography in patients who were pretreated with tirofiban (table 2). Acute angiography was done in 970 of 984 (99%) patients. Four patients (all Vol 372 August 16, 2008

5 in the tirofiban group) died before angiography could be undertaken and ten did not undergo angiography because of a clear false positive diagnosis on arrival at the hospital (five in the placebo group and five in the tirofiban group). Table 3 shows angiographic outcomes. At initial angiography, a completely occluded infarctrelated vessel (TIMI 0 flow) was present in 185 of 443 (42%) patients in the tirofiban group versus 227 of 456 (50%) in the placebo group (p=0 013). However, initial TIMI 3 flow of the infarct-related artery did not differ significantly between the groups (table 3). PCI was undertaken in almost all patients (table 3). Blinded bail-out tirofiban was administered in 97 (20%) of 488 patients in the tirofiban group and 140 (29%) of 492 in the placebo group (p=0 002). Distal embolisation was present in fewer patients pretreated with tirofiban than in those assigned to placebo (table 3). Distal embolisation or no reflow, which was defined as TIMI flow of 0 or 1 after PCI, occurred less often after pretreatment with tirofiban (table 3). No difference in final TIMI 3 flow was present between the groups. An interpretable ECG min after angiography or PCI was available in 906 (92%) patients (455 [92%] in the placebo group and 451 [92%] in the tirofiban group). The percentage of patients with more than 3 mm residual ST deviation was lower in the tirofiban group than in the placebo group (table 2). Further analysis showed that residual ST deviation was less in all four prespecified subgroups for the patients pretreated with tirofiban than for those given placebo (table 2, figure 2). Furthermore, ST-segment resolution was significantly better in the tirofiban group (table 2, figure 2). Figure 3 shows the risk ratios for the incidence of the primary endpoint in each of the prespecified subgroups of patients. 30-day follow-up was complete for almost all patients (figure 1). 27 patients refused written informed consent after having given oral informed consent in the ambulance and seven were lost to follow-up (figure 1). Table 4 shows all clinical outcomes and safety parameters. The combined rate of death, recurrent myocardial infarction, urgent TVR, or blinded bail-out use of tirofiban was 26 0% in the tirofiban group compared with 32 9% in the placebo group (p=0 020), with much fewer patients needing bail-out study drug in the tirofiban group (table 4). The extent of residual ST deviation 1 h after PCI (5 5 mm [SD 7 2] for bail-out vs 3 7 mm [4 8] for no bail-out; p=0 005) and the rate of major adverse cardiac events (12 2% [29/237] vs 5 6% [40/709]; p=0 001) were significantly related to the need for bail-out study drug. Figure 4 shows Kaplan-Meier survival free from the key secondary endpoint. Total mortality was lower in patients who were pretreated with tirofiban than in those allocated to placebo or no tirofiban (p=0 144; table 4). The rate of major bleeding did not differ significantly between the two groups (table 4). (n=493) (n=491) Discussion This study shows that in patients with STEMI, routine prehospital initiation of a glycoprotein IIb/IIIa blocker in addition to aspirin, heparin, and high-dose clopidogrel, improves ST-segment resolution before and after PCI. This finding was associated with an improved outcome free from adverse angiographic or clinical events at 30-day follow-up, without a significant increase in major bleeding. The routine administration of glycoprotein IIb/IIIa blockers in patients with STEMI is a class IIa indication according to the guidelines of the American and European Societies of Cardiology. 17,18 However, large registry studies show that in real-world practice, glycoprotein IIb/IIIa blockers are given to only 25 30% of patients with STEMI, often for bail-out situations. 19,20 Therefore the placebo group in our study with provisional use of high-bolus dose tirofiban in 29% of patients is a good representation of routine daily practice and is not undertreated. This study shows that further optimisation of platelet aggregation p value TIMI flow at initial angiography TIMI 0 227/456 (49 8%) 185/443 (41 8%) TIMI 1 50/456 (11 0%) 64/443 (14 4%) TIMI 2 125/456 (19 7%) 138/443 (22 1%) TIMI 3 89/456 (19 5%) 96/443 (21 7%) ACT (seconds) 180 (64) 181 (62) Extra UFH (2500 IU) 353/485 (72 8%) 345/477 (72 3%) PCI done 439/493 (89 0%) 422/488 (86 5%) PCI immediately after angiography 437/439 (99 5%) 418/422 (99 1%) Stent placement 392/437 (89 7%) 378/418 (90 4%) Drug-eluting stent 85/392 (21 7%) 97/378 (25 7%) Bare-metal stent 310/392 (79 1%) 286/378 (75 7%) Additional devices 103/488 (21 1%) 84/482 (17 4%) Rescue trombosuction 75/488 (15 4%) 58/482 (12 0%) IABP 26/488 (5 3%) 19/482 (3 9%) Temporary pacemaker 2/488 (0 4%) 6/482 (1 2%) Other 5/488 (1 0%) 8/482 (1 7%) TIMI flow after PCI TIMI 0/1 17/436 (3 9%) 11/419 (2 6%) TIMI 2 15/436 (3 4%) 31/419 (7 4%) TIMI 3 404/436 (92 7%) 377/419 (90 0%) Distal embolisation after PCI 28/360 (7 8%) 15/333 (4 5%) TIMI flow 0/1 or distal embolisation after PCI 45/436 (10 3%) 26/419 (6 2%) Myocardial blush grade after PCI MBG 0/1 86/411 (20 9%) 101/394 (25 6%) MBG 2 143/411 (34 8%) 121/394 (30 7%) MBG 3 182/411 (44 3%) 172/394 (43 7%) Data are n/n (%) or mean (SD). TIMI=thrombolysis in myocardial infarction. ACT=activated clotting time. UFH=unfractionated heparin. PCI=percutaneous coronary intervention. IABP=intra-aortic balloon pump. MBG=myocardial blush grade. Table 3: Angiographic outcomes Vol 372 August 16,

6 A Number of patients (%) B Number of patients (%) % 18 8% 26 4% 35 5% 11 5% 22 9% 65 6% 24 8% 20 2% 25 9% 29 0% 16 4% 23 6% 60 0% Normalised 1 3 mm 4 6 mm >6 mm N=906 Complete Partial No N=876 Figure 2: Myocardial reperfusion data for electrocardiography, according to treatment group The percentages of patients are shown according to the degree of residual ST-segment deviation (A) or ST-segment resolution (B) on the electrocardiogram between 30 min and 90 min after angiography or percutaneous coronary intervention (PCI). inhibition with early use of a glycoprotein IIb/IIIa blocker before PCI is beneficial. The IIb/IIIa blocker should not be restricted to patients with complications after PCI, even in a setting in which high-dose clopidogrel is given well in advance of PCI. Previous studies showed that the maximum antiplatelet effect of 600 mg clopidogrel is achieved only after 2 4 h, and is less effective in patients with STEMI than in those with stable coronary artery disease. 21,22 We used the high-bolus dose (25 μg/kg) of tirofiban in our study. Previous studies showed that the 10 μg/kg bolus dose did not achieve early sufficient platelet aggregation inhibition after administration. 23 Recent studies suggest that the high-bolus dose of tirofiban results in optimum platelet aggregation in a large percentage of patients very early after PCI. 24,25 The aspirate, which was obtained with thrombosuction before angioplasty in patients with acute STEMI, contained platelet-rich thrombi in most patients, emphasising the importance of early initiation of effective antiplatelet treatment in these patients. 26 Much of the efficacy of tirofiban in this study might be related to the short time between the onset of symptoms of acute myocardial infarction and the administration of study drug. The median time from the onset of symptoms until the start of study drug was only 76 min. In the Facilitated Intervention with Enhanced Reperfusion Speed to Stop Events (FINESSE) study, 27 the investigators recorded no benefit of early versus late administration of abciximab. In the Bavarian Reperfusion AlternatiVes Evaluation (BRAVE 3) study, abciximab did not reduce infarction size in patients who were already pretreated with 600 mg of clopidogrel. 28 However, abciximab was started fairly late (>200 min) after the onset of symptoms in both studies. The Abciximab before Direct Angioplasty and stenting in Myocardial Infarction Regarding Acute and Long-term follow-up (ADMIRAL) study 29 already showed that most of the benefit of this drug was in patients who received the IIb/IIIa blocker in the ambulance, well in advance of arrival at the PCI centre. 29 Subgroup analysis in our study also showed that the largest effect was seen when patients received tirofiban shortly after the onset of symptoms (figure 3). Prehospital diagnosis and immediate triage to the nearest PCI centre is well known to reduce time to treatment and improve outcome. 30,31 In our study, patients were diagnosed and immediately treated in the ambulance by either a physician or a dedicated well-trained paramedic. Diagnosis was correct in 94% of patients with computerised ECG algorithms only, without the need for transmission of the ECG to the hospital. This prehospital diagnosis of infarction and treatment in the ambulance enables the early initiation of antithrombotic and antiplatelet treatment well in advance of arrival at the PCI centre and early after the start of symptoms. Furthermore, PCI was undertaken within 90 min after diagnosis of myocardial infarction in almost all patients and total ischaemic time was 166 min, which is close to the goal of 120 min according to the latest STEMI guideline of the American Heart Association and American College of Cardiology. 32 Surprisingly, pretreatment with tirofiban improved ST-segment resolution after PCI without a significant improvement of final TIMI flow or myocardial blush. This discrepancy between electrocardiographic and angio graphic assessment of myocardial reperfusion has been noted previously. The Zwolle Myocardial Infarction Study Group has shown that only 65% of patients with blush grade 3 have complete ST-segment resolution. 33 Further more, Poli and colleagues 34 noted that 19% of patients at intermediate risk of adverse events have Vol 372 August 16, 2008

7 n/n n/n RR (95% CI) RR (95% CI) p value All patients Sex Female Male Age (years) 61 3 >61 3 Diabetes No Yes Location Anterior MI Non-anterior MI Killip >1 No Yes TIMI flow grade before PCI Time start study drug to balloon inflation or angiography (min) 55 98/218 >55 72/221 Time symptom onset to diagnosis (min) 76 > /451 30/ /344 84/228 88/ /397 22/53 98/168 54/ /396 25/49 135/329 23/91 81/234 86/ /455 47/ /342 99/ / /404 30/51 106/170 78/ /392 29/56 168/343 18/83 104/226 93/213 89/ / ( ) 0 67 ( ) 0 89 ( ) 0 87 ( ) 0 82 ( ) 0 87 ( ) 0 71 ( ) 0 94 ( ) 0 72 ( ) 0 83 ( ) 0 99 ( ) 0 84 ( ) 1 17 ( ) 0 98 ( ) 0 75 ( ) 0 82 ( ) 0 90 ( ) better better Figure 3: Risk ratios for the primary endpoint (extent of residual ST segment deviation 1 h after PCI), according to prespecified clinical or angiographic subgroups Data are reported for the patients who had clinical electrocardiographical data at baseline. MI=myocardial infarction. TIMI=thrombolysis in myocardial infarction. PCI=percutaneous coronary intervention. residual ST-segment elevation despite normal myocardial blush. 34 Presumably, both parameters indicate myocardial reperfusion differently; however, each has shown independent prognostic value. 35,36 The reduction in PCI complications, together with improved initial TIMI flow and less distal embolisation or no reflow, probably bring about the better ST-segment resolution 1 h after PCI. Our study shows that a high-bolus dose of tirofiban improved ST-segment elevation resolution during ambulance transport, before PCI. After PCI, the difference in ST-segment resolution between the groups further increased a finding which is in contrast with that from studies on facilitated PCI. The ASsessment of the Safety and Efficacy of a New Treatment strategy with percutaneous coronary intervention (ASSENT) 4 study 37 showed a benefit in ST-segment resolution before PCI; however, this benefit totally disappeared after PCI and resulted in a worse outcome in patients who were pretreated with full-dose tenecteplase. 37 Only a few strategies have been able to improve ST-segment resolution after PCI: routine thrombosuction of intracoronary thrombus or debris 26 and earlier reperfusion. 38 Both these strategies again noted a clear relation between ST resolution and mortality. The improved ST resolution might also be the reason for the trend towards better survival in our study. Whether this improved resolution results in a mortality benefit at long-term follow-up is not yet known. We will monitor clinical outcome until 1 year after randomisation. Triple antiplatelet therapy, with high-dose tirofiban in addition to high-dose clopidogrel (600 mg) and aspirin pretreatment was not associated with an increased risk of major bleeding. This finding might be related to a very careful heparin-dose protocol with routine measurements of activated clotting time at the catheterisation laboratory. Extra heparin (2500 IU) was only given when the activated clotting time was less than 200 seconds. A closure device (Angioseal, St Jude Medical, St Paul, MN, USA) was used in 70% of patients. Furthermore, low molecular-weight heparin after PCI was discouraged except for patients with large anterior myocardial infarction. With the results of the recently published Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trial, 39 one might argue that the unfractionated heparin and glycoprotein IIb/IIIa blocker would be better replaced by bivalirudin to reduce bleeding. 39 However, in that trial, the high rate of major bleeding in the glycoprotein IIb/IIIa group might partly be due to a less strict Vol 372 August 16,

8 (n=477) (n=473) p value Clinical outcome Death/recurrent MI/urgent TVR or thrombotic bail-out 157/477 (32 9%) 123/473 (26 0%) Death/recurrent MI or urgent TVR 39/477 (8 2%) 33/473 (7 0%) Death 19/477 (4 0%) 11/473 (2 3%) Recurrent MI 14/477 (2 9%) 13/473 (2 7%) Urgent TVR 20/477 (4 2%) 18/473 (3 8%) Urgent PCI 19/477 (4 0%) 11/473 (2 3%) Urgent CABG 1/477 (0 2%) 7/473 (1 5%) Thrombotic bail-out 140/492 (28 5%) 97/488 (19 9%) TIMI flow grade 0 2 or slow reflow 45/492 (9 1%) 29/488 (5 9%) Dissection 6/492 (1 2%) 5/488 (1 0%) Distal embolisation 58/492 (11 8%) 44/488 (9 0%) Side-branch closure 4/492 (0 8%) 3/488 (0 6%) Abrupt closure of culprit vessel 11/492 (2 2%) 1/488 (0 2%) Clinical instability 15/492 (3 0%) 13/488 (2 7%) Prolonged ischaemia 4/492 (0 8%) 4/488 (0 8%) Safety Major bleeding 14/477 (2 9%) 19/473 (4 0%) Major CABG related 7/477 (1 5%) 10/473 (2 1%) Major non-cabg related 7/477 (1 5%) 9/473 (1 9%) Minor bleeding 21/477 (4 4%) 29/473 (6 1%) Minor CABG related 8/477 (1 7%) 7/473 (1 5%) Minor non-cabg related 13/477 (2 7%) 22/473 (4 7%) Stroke within 30 days 7/477 (1 5%) 1/473 (0 2%) Thrombocytopenia (in-hospital) 9/492 (1 8%) 10/489 (2 0%) Net clinical outcome* 56/477 (11 7%) 44/473 (9 3%) Data are n/n (%). MI=myocardial infarction. TVR=target vessel revascularisation. PCI=percutaneous coronary intervention. CABG=coronary artery bypass grafting. TIMI=thrombolysis in myocardial infarction.*the combined incidence of death, recurrent myocardial infarction, urgent TVR, stroke, or major bleeding. Table 4: Clinical outcome and safety parameters Event-free survival (%) Number at risk Time (days) p= Figure 4: Kaplan-Meier event-free survival showing survival free from death, recurrent myocardial infarction, urgent target vessel revascularisation, or blinded bail-out use of study drug heparin protocol (heparin was given both before [76%] and during [99%] the procedure, with target activated clotting time around 250 seconds). Additionally, the glycoprotein blocker was given fairly late after symptom onset (>2 h) thereby losing the opportunity for improvement of ST resolution. Therefore, a trial comparing the two drugs when given in the ambulance shortly after symptom onset might be needed before switching to bivalirudin in patients undergoing primary PCI for STEMI. Our trial was not powered on a difference in clinical outcome between groups. However, we noted a better clinical outcome in the tirofiban group than in the placebo group, with lower overall mortality and less urgent repeat PCI. The clinical significance of the need for bail-out study drug could be questioned. The strong relation, however, between major adverse cardiac events and the need for bail-out therapy lends supports to the trend towards improved clinical outcome for early treatment with tirofiban. As part of preparation for the double-blind On-TIME 2 trial, two participating centres in the Netherlands randomly assigned 414 patients in the ambulance to high-bolus dose tirofiban or no tirofiban (open label) with the same inclusion and exclusion criteria and concomitant treatment. By pooling the data of these 414 open-label study patients with the 984 patients included in the double-blind study (N=1398), the rate of major adverse cardiac events was significantly reduced by early prehospital initiation of high-bolus dose tirofiban (5 8% vs 8 6%; p=0 043), with reduced mortality (2 2% vs 4 1%; p=0 051) and no increased risk of major bleeding (3 4% vs 2 9%; p=0 581; unpublished data). Although combination of the results from two studies with different design (open label vs double blind) has statistical limitations, the results further lend support to our main study findings. Limited information is available about the number of eligible patients screened for inclusion. The percentage of patients randomly assigned is estimated to be around 50 80% of screened patients. A screening or exclusion log is difficult to organise in an ambulance setting. Most patients were excluded because they did not meet the inclusion or exclusion criteria or because the ECG algorhythm in the ambulance failed to give a definite diagnosis of myocardial infarction. Therefore, a limitation of the trial is that the study results might not apply for patients in cardiogenic shock, renal insufficiency, or the very elderly population (>85 years). Furthermore, the study does not address whether initiation of the IIb/IIIa blocker, other than in the ambulance or at a referral centre, is beneficial. Although many studies showed a beneficial effect of routine IIb/IIIa blocker initiation before PCI in the catheterisation laboratory, the results of the BRAVE 3 study suggest less benefit of initiation in the catheterisation laboratory when patients have been pretreated with high-dose clopidogrel Vol 372 August 16, 2008

9 Contributors AWJvtH, JtB, and CH were responsible for conception and design of the study. AWJvtH, JtB, and CH did the analysis and interpreted the analysis in collaboration with TH, WvW, JPO, and HS. TD was responsible for study coordination in Germany. RCF, GvH, and PS coordinated the study and were responsible for patient recruitment in Marburg, Enschede, and Utrecht. AWJvtH wrote the first draft of the report. All authors critically revised the report for important intellectual content and approved the final version of the manuscript. AWJvtH obtained funding from Merck. Conflict of interest statement AWJvtH has received speaker fees from Merck, Sanofi Aventis, and Schering Plough. CH has received advisory board/speaker fees from Merck, Iroko, Lilly, GlaxoSmithKline, Sanofi Aventis, the Medicines Company, Roche, and Abbott. All other authors declare that they have no conflict of interest. Acknowledgments The study was partly funded by an unrestricted grant from Merck and Co, Whitehouse Station, NJ, USA. We thank all ambulance personnel who were involved in the trial for their cooperation and recruitment of patients. References 1 Simes RJ, Topol EJ, Holmes DR Jr, et al, for the Gusto-I Investigators. Link between the angiographic substudy and mortality outcomes in a large randomized trial of myocardial reperfusion. Importance of early and complete infarct artery reperfusion. GUSTO-I Investigators. Circulation 1995; 91: Anderson JL, Karagounis LA, Califf RM. Meta-analysis of five reported studies on the relation of early coronary patency grades with mortality and outcomes after acute myocardial infarction. Am J Cardiol 1996; 78: Zeymer U, Schroder R, Machnig T, Neuhaus KL. Primary percutaneous transluminal coronary angioplasty accelerates early myocardial reperfusion compared to thrombolytic therapy in patients with acute myocardial infarction. Am Heart J 2003; 146: Zijlstra F, de Boer MJ, Hoorntje JCA, Reiffers S, Reiber JH, Suryapranata H. A comparison of immediate coronary angioplasty with intravenous streptokinase in acute myocardial infarction. N Engl J Med 1993; 328: Keeley EC, Boura JA, Grines CL. Comparison of primary and facilitated percutaneous coronary interventions for ST-elevation myocardial infarction: quantitative review of randomised trials. Lancet 2006; 367: Stenestrand U, Lindback J, Wallentin L. Long-term outcome of primary percutaneous coronary intervention vs prehospital and in-hospital thrombolysis for patients with ST-elevation myocardial infarction. JAMA 2006; 296: Matetzky S, Shenkman B, Guetta V, et al. Clopidogrel resistance is associated with increased risk of recurrent atherothrombotic events in patients with acute myocardial infarction. Circulation 2004; 109: Bassand JP, Hamm CW, Ardissino D, et al. Task force for diagnosis and treatment of non ST elevation acute coronary syndromes. Eur Heart J 2007; 28: Montalescot G, Antoniucci D, Kastrati A, et al. Abciximab in primary coronary stenting of ST-elevation myocardial infarction: a European meta-analysis on individual patients data with long-term follow-up. Eur Heart J 2007; 28: Topol EJ, Neumann FJ, Montalescot G. A preferred reperfusion strategy for acute myocardial infarction. J Am Coll Cardiol 2003; 42: Montalescot G, Borentain M, Payot L, Collet JP, Thomas D. Early vs late administration of glycoprotein IIb/IIIa blockers in primary percutaneous coronary intervention of acute ST-segment elevation myocardial infarction: a meta-analysis. JAMA 2004; 292: De Luca G, Gibson M, Bellandi F, et al. Early Glycoprotein IIb-IIIa inhibitors in Primary angioplasty (EGYPT) cooperation. An individual patients data meta-analysis. Heart 2008; published online June 5. DOI: /hrt van t Hof AWJ, Hamm C, Rasoul S, Guptha S, Paolini JF, ten Berg JM, on behalf of the On-TIME 2 investigators. Ongoing tirofiban in myocardial infarction evaluation (On-TIME) 2 trial: rationale and study design. EuroInterv 2007; 3: De Luca G, Maas AC, Suryapranata H, et al. Prognostic significance of residual cumulative ST-segment deviation after mechanical reperfusion in patients with ST-segment elevation myocardial infarction. Am Heart J 2005; 150: van t Hof AWJ, Liem AL, de Boer MJ, Zijlstra F, on behalf of the Zwolle Myocardial Infarction Study Group. Clinical value of 12-lead electrocardiogram after successful reperfusion therapy for acute myocardial infarction. Lancet 1997; 350: Bovill EG, Terrin ML, Stump DC, et al. Hemorrhagic events during therapy with recombinant tissue-type plasminogen activator, heparin, and aspirin for acute myocardial infarction. Results of the Thrombolysis in Myocardial Infarction (TIMI), Phase II Trial. Ann Intern Med 1991; 115: Van de Werf F, Ardissino D, Betriu A, et al. Management of acute myocardial infarction in patients presenting with ST-segment elevation. The Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology. Eur Heart J 2003; 24: Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction; a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of patients with acute myocardial infarction). J Am Coll Cardiol 2004; 44: E Budaj A, Brieger D, Steg PG, et al; GRACE Investigators. Global patterns of use of antithrombotic and antiplatelet therapies in patients with acute coronary syndromes: insights from the Global Registry of Acute Coronary Events (GRACE). Am Heart J 2003; 146: Mandelzweig L, Battler A, Boyko V, et al; Euro Heart Survey Investigators. The second Euro Heart Survey on acute coronary syndromes: characteristics, treatment, and outcome of patients with ACS in Europe and the Mediterranean Basin in Eur Heart J 2006; 27: Hochholzer W, Trenk D, Frundi D, et al. Time dependence of platelet inhibition after a 600-mg loading dose of clopidogrel in a large, unselected cohort of candidates for percutaneous coronary intervention. Circulation 2005; 111: Heestermans AA, van Werkum JW, Taubert D, et al. Impaired bioavailability of clopidogrel in patients with a ST-segment elevation myocardial infarction. Thromb Res 2008; published online April 30. DOI: /j.thromres Kereiakes DJ, Broderick TM, Roth EM, et al. Time course, magnitude, and consistency of platelet inhibition by abciximab, tirofiban, or eptifibatide in patients with unstable angina pectoris undergoing percutaneous coronary intervention. Am J Cardiol 1999; 84: Schneider DJ, Herrmann HC, Lakkis N, et al. Enhanced early inhibition of platelet aggregation with an increased bolus of tirofiban. Am J Cardiol 2002; 90: Ernst NM, Suryapranata H, Miedema K, et al. Achieved platelet aggregation inhibition after different antiplatelet regimens during percutaneous coronary intervention for ST-segment elevation myocardial infarction. J Am Coll Cardiol 2004; 44: Svilaas T, Vlaar PJ, van der Horst IC, et al. Thrombus aspiration during primary percutaneous coronary intervention. N Engl J Med 2008; 358: Ellis SG, Tendera M, de Belder MA, et al, for the FINESSE Investigators. Facilitated PCI in patients with ST-elevation myocardial infarction. N Engl J Med 2008; 358: Mehili J. Results of the Bavarian Reperfusion Alternatives Evaluation (BRAVE) 3 Study. Presented at American College of Cardiology 2008 Scientific Sessions/i2 Summit-SCAI Annual Meeting; Chicago, USA; March 30, 2008 (abstr number ). 29 Montalescot G, Barragan P, Wittenberg O, et al; for the ADMIRAL Investigators. Platelet glycoprotein IIb/IIIa inhibition with coronary stenting for acute myocardial infarction. N Engl J Med 2001; 344: Vol 372 August 16,

10 30 Carstensen S, Nelson GC, Hansen PS, et al. Field triage to primary angioplasty combined with emergency department bypass reduces treatment delays and is associated with improved outcome. Eur Heart J 2007; 28: van t Hof AWJ, Rasoul S, van de Wetering H, et al. On-TIME study group. Feasibility and benefit of prehospital diagnosis, triage, and therapy by paramedics only in patients who are candidates for primary angioplasty for acute myocardial infarction. Am Heart J 2006; 151: 1255.e Antman EM, Hand M, Armstrong PW, et al Focused Update of the ACC/AHA 2004 Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines: developed in collaboration with the Canadian Cardiovascular Society endorsed by the American Academy of Family Physicians: 2007 Writing Group to Review New Evidence and Update the ACC/AHA 2004 Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction, Writing on Behalf of the 2004 Writing Committee. Circulation 2008; 117: van t Hof AWJ, Liem A, Suryapranata H, Hoorntje JCA, de Boer MJ, Zijlstra F. Angiographic assessment of myocardial reperfusion in patients treated with primary angioplasty for acute myocardial infarction: myocardial blush grade. Zwolle Myocardial Infarction Study Group. Circulation 1998; 97: Poli A, Fetiveau R, Vandoni P, et al. Integrated analysis of myocardial blush and ST-segment elevation recovery after successful primary angioplasty: real-time grading of microvascular reperfusion and prediction of early and late recovery of left ventricular function. Circulation 2002; 106: Angeja BG, Gunda M, Murphy SA, et al. TIMI myocardial perfusion grade and ST segment resolution: association with infarct size as assessed by single photon emission computed tomography imaging. Circulation 2002; 105: De Luca G, Suryapranata H, de Boer MJ, et al, for the Ongoing In Myocardial Infarction Evaluation (On-TIME) study group. Combination of electrocardiographic and angiographic markers of reperfusion in the prediction of infarct size in patients with ST-segment elevation myocardial infarction undergoing successful primary angioplasty. Int J Cardiol 2007; 117: McDonald MA, Fu Y, Zeymer U, et al, for the ASSENT-4 PCI Investigators. Adverse outcomes in fibrinolytic-based facilitated percutaneous coronary intervention: insights from the ASSENT-4 PCI electrocardiographic substudy. Eur Heart J 2008; 29: De Luca G, van t Hof AWJ, de Boer MJ, et al. Time-to-treatment significantly affects the extent of ST-segment resolution and myocardial blush in patients with acute myocardial infarction treated by primary angioplasty. Eur Heart J 2004; 25: Stone GW, Witzenbichler B, Guagliumi G, et al; HORIZONS-AMI Trial Investigators. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med 2008; 358: Vol 372 August 16, 2008