Author + information
- Received July 2, 2009
- Revision received November 16, 2009
- Accepted November 19, 2009
- Published online June 1, 2010.
- Jurriën M. ten Berg, MD, PhD⁎,⁎ (, )
- Arnoud W.J. van 't Hof, MD, PhD†,
- Thorsten Dill, MD, PhD‡,
- Ton Heestermans, MD†,
- Jochem W. van Werkum, MD, PhD⁎,
- Arend Mosterd, MD, PhD§,#,⁎⁎,
- Gert van Houwelingen, MD∥,
- Petra C. Koopmans, MSc¶,
- Pieter R. Stella, MD, PhD⁎⁎,
- Eric Boersma, MSc, PhD††,
- Christian Hamm, MD, PhD‡,
- On-TIME 2 Study Group
- ↵⁎Reprint requests and correspondence:
Dr. Jurriën M. ten Berg, Department of Cardiology, St. Antonius Hospital, Koekoekslaan 1, 3435 CM Nieuwegein, the Netherlands
Objectives The purpose of this trial was to study the effect of a high bolus dose (HBD) of tirofiban on clinical outcome in patients undergoing primary percutaneous coronary intervention for ST-segment elevation myocardial infarction (STEMI).
Background The On-TIME 2 (Ongoing Tirofiban In Myocardial infarction Evaluation 2) placebo-controlled, double-blind, randomized trial showed that early administration of HBD tirofiban in the ambulance improves ST-segment resolution in patients with STEMI undergoing primary percutaneous coronary intervention. The effect of early tirofiban treatment on clinical outcome is unclear.
Methods The On-TIME 2 trial consisted of 2 phases: an open-label phase, followed by a double-blind, placebo-controlled phase. STEMI patients were randomized to either HBD tirofiban or no tirofiban (phase 1) or placebo (phase 2) in addition to aspirin, heparin, and high-dose clopidogrel. The protocol pre-specified a pooled analysis of the 2 study phases to assess the incidence of major adverse cardiac events at the 30-day follow-up and on total mortality at the 1-year follow-up.
Results During a 3-year period, 1,398 patients were randomized, 414 in phase 1 and 984 in phase 2. Major adverse cardiac events at 30 days were significantly reduced (5.8% vs. 8.6%, p = 0.043). There was a strong trend toward a decrease in mortality (2.2% vs. 4.1%, p = 0.051) in patients who were randomized to tirofiban pre-treatment, which was maintained during the 1-year follow-up (3.7% vs. 5.8%, p = 0.08). No clinically relevant difference in bleeding was observed.
Conclusions Early, pre-hospital initiation of HBD tirofiban, in addition to high-dose clopidogrel, improves the clinical outcome after primary percutaneous coronary intervention in patients with STEMI. (Ongoing 2b/3a inhibition In Myocardial infarction Evaluation; ISRCTN06195297)
Early and complete restoration of blood flow in the infarct-related vessel in patients with ST-segment elevation myocardial infarction (STEMI) is of utmost importance to improve the chance of survival and the clinical outcome (1–3). Primary percutaneous coronary intervention (pPCI) has been shown to be the most effective treatment modality for restoring the blood flow in the infarct-related vessel (4,5). Transporting patients to an intervention center, however, inevitably leads to a delay in initiation of the pPCI and thus prolongs ischemic time. The early initiation of antithrombotic treatment for STEMI patients before pPCI potentially opens the occluded vessel during transportation and reduces periprocedural thrombotic complications (6). Glycoprotein (GP) IIb/IIIa receptor blockers are very effective inhibitors of platelet aggregation, even when administered in addition to aspirin and clopidogrel (7) and have been shown to reduce thrombotic complications in patients undergoing percutaneous coronary intervention (PCI) (8). In our randomized, double-blind, placebo-controlled On-TIME 2 (Ongoing Tirofiban In Myocardial infarction Evaluation 2) trial, we recently demonstrated that the GP IIb/IIIa receptor blocker tirofiban, when given in the ambulance, resulted in an improvement in ST-segment resolution as a marker for myocardial perfusion in patients with STEMI undergoing pPCI (9). However, this trial did not have sufficient power to study the effect of a high bolus dose (HBD) of tirofiban on clinical outcome. In this article, we present a pre-specified pooled analysis of the placebo-controlled study phase and the open-label study phase to investigate the effect of early administration of HBD tirofiban on clinical outcome. This larger sample size also enabled us to study subgroups of patients who would most likely benefit from early administration of HBD tirofiban.
The On-TIME 2 trial was a multicenter, prospective, randomized clinical trial. The study consisted of 2 phases: a randomized, open-label phase (phase 1) and a placebo-controlled, randomized, double-blind phase (phase 2). The rationale and design of the study were described previously (10). In brief, the study population consisted of patients with STEMI who were candidates for pPCI treatment. Eligible patients were men and women 21 to 85 years of age with symptoms of acute myocardial infarction (MI) for >30 min but <24 h, and an ST-segment elevation of >1 mV in 2 adjacent electrocardiogram leads. Exclusion criteria were known severe renal dysfunction, therapy-resistant cardiogenic shock, persistent severe hypertension, and an increased risk of bleeding. Also excluded were patients with a left bundle branch block and patients with a life expectancy of <1 year.
Written informed consent was obtained by an intensive care nurse in the ambulance or, in a minority of the patients, by a physician in the referral center. The study protocol was approved by all local ethics committees involved.
Patients were randomly assigned to pre-hospital treatment with tirofiban (25-μg/kg bolus and 0.15-μg/kg/min maintenance infusion) or no tirofiban (phase 1), or placebo (phase 2). Each participating ambulance or referral center was supplied with sealed study kits in blocks of 4, containing either open-label tirofiban or saline solution vials (phase 1) or blinded study medication (phase 2). The study flowchart is presented in Figure 1.In the ambulance or referral center, all patients also received a bolus of unfractionated heparin (5,000 IU) intravenously, together with aspirin 500 mg (Aspegic) intravenously and an oral 600-mg loading dose of clopidogrel. Before pPCI, an additional 2,500 IE unfractionated heparin was only administered if the activated clotting time was <200 s.
Coronary angiography and PCI were performed according to each institution's guidelines and standards. During or after pPCI, bail-out tirofiban could be administered for the following indications: decrement in Thrombolysis In Myocardial Infarction (TIMI) flow grade (TIMI flow grades of 0 to 2 or slow reflow), dissection with decreased flow, distal embolization, side branch closure, abrupt closure of the culprit vessel, clinical instability, and prolonged ischemia. Bail-out tirofiban was administered in a HBD (25-μg/kg bolus). In the double-blind phase of the trial, blinded bail-out vials were included in the medication box to maintain blinding of the initial treatment assignment to avoid double-dosing. When a bailout vial was administered, the study infusion was replaced by open-label tirofiban. In the open-label phase, the doses of bail-out tirofiban and the indications for administration were similar to those in the double-blind phase.
The key primary end point of this pooled analysis was the incidence of major adverse cardiac events (MACE), as defined by the composite of death, recurrent MI, or urgent target vessel revascularization at the 30-day follow-up. The secondary end point was total mortality at the 1-year follow-up. A blinded independent Clinical Endpoint Committee adjudicated all clinical end points. Death was defined as all-cause mortality. Recurrent MI was defined as a new increase of creatine kinase, myocardial bound ≥3 times the upper limit of normal present in 2 separate blood samples and accompanied by chest pain and/or changes on the electrocardiogram. Early recurrent MI was defined as a decrease in creatine kinase, myocardial bound 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 3 times the upper limit of normal. Urgent target vessel revascularization during the hospitalization period was defined as a new episode of ischemic signs or symptoms at rest, with documentation of a new ST-segment shift ≥0.05 mV (0.5 mm) on a 12-lead electrocardiogram that necessitated an unplanned coronary intervention or a coronary artery bypass graft of the infarct-related vessel.
The safety end points included the occurrence of a hemorrhage, transfusions, stroke, thrombocytopenia, and serious adverse events. Bleeding was defined according to the TIMI criteria (11). Major bleeding was defined as clinical overt signs of hemorrhage associated with a decrease in hemoglobin of >5 g/dl (or when hemoglobin assessment is not available, a decrease in hematocrit of >15%). For patients undergoing coronary artery bypass graft surgery, the rate of surgical re-exploration for bleeding and the post-operative volume of blood loss were also evaluated.
Follow-up information was derived from visits to the outpatient clinic or from telephone contact at the 30-day follow-up and at the 1-year follow-up.
The pooled analysis of the 2 study phases was pre-specified in the final study protocol and statistical plan. The effect of pre-hospital HBD tirofiban was evaluated in pre-specified subgroups of patients. In addition, a post hoc analysis was performed for patients who underwent pPCI. For continuous variables, such as the time between the onset of symptoms and diagnosis, subgroup analyses were performed, with the median values as cutoff.
All analyses were performed according to the intention-to-treat principle. Categorical variables were analyzed with the chi-square test or the Fisher exact test, ordinal categorical variables with the chi-square test for linear trend, and continuous variables with the Mann-Whitney Utest. We used the Breslow-Day test to investigate homogeneity of odds ratios to determine whether the null hypothesis of equal odds ratios for different strata should be rejected. Time-to-event analysis was performed to describe survival free from MACEs at 30 days and death after 1 year from the date of randomization. Survival was represented by Kaplan-Meier curves. Log-rank tests were used to test the significance of difference in survival between tirofiban and placebo/no tirofiban. SPSS for Windows, version 15 (SPSS Inc., Chicago, Illinois) was used for all the analyses.
During a 3-year period, 1,398 patients were randomized: 414 patients in the open-label phase (phase 1, June 2004 to June 2006, 2 participating centers) and 984 patients in the double-blind phase (phase 2, June 2006 to November 2007, 24 participating centers). A flowchart of the study is presented in Figure 1. There were no relevant differences in baseline clinical or angiographic characteristics between the patients in the open-label phase and the patients in the double-blind phase (Table 1).Patients in the open-label group had slightly higher mean diastolic blood pressure, but a similar systolic blood pressure; they more often had a drug-eluting stent placed (33.2% vs. 23.6%, p = 0.001), and the mean stent length was slightly longer. Due to the difference in study design, a difference in the use of bail-out tirofiban was found between the study phases: 41.8% in the open-label phase versus 24.2% in the double-blind phase (19.9% in the tirofiban group and 28.5% in the placebo group). In both study phases, early HBD tirofiban treatment decreased the extent of residual ST-segment deviation and improved ST-segment resolution 1 h after PCI as well as clinical outcome at 30 days and 1 year compared with placebo/no tirofiban treatment (Table 2).After pooling the open-label and double-blind study phases on an individual basis, 709 patients were randomized to HBD tirofiban and 689 patients to placebo/no tirofiban infusion. There were also no relevant differences in baseline clinical or angiographic characteristics between the patients allocated to HBD tirofiban treatment and the patients allocated to placebo/no tirofiban infusion (Table 1). Ninety-six percent of the patients were included and randomized after pre-hospital infarct diagnosis in the ambulance. The median time of diagnosis was 77 min (interquartile range [IQR] 45 to 150 min) after the onset of symptoms. The time intervals are shown in Figure 2.The median duration of transport to the nearest PCI center was 33 min (IQR 19 to 49 min). Acute angiography was performed in 98.5% of the patients at a median time interval of 20 min (IQR 10 to 33 min) after arrival at the PCI center. The median total ischemic time was 166 min (IQR 128 to 246 min), whereas the median duration of treatment with tirofiban or placebo/no tirofiban infusion until angiography was 55 min (IQR 44 to 70 min). There were 81 patients (5.8%) who had a false-positive STEMI diagnosis, evenly distributed between the tirofiban and placebo/no tirofiban infusion groups. Five patients died before angiography could be performed, and 16 patients did not undergo angiography because of a clearly false-positive diagnosis on arrival at the hospital. The angiographic characteristics are shown in Table 1. Initial TIMI flow grade 3 of the infarct-related artery was present in 22.5% of patients in the tirofiban group versus 19.4% in the placebo/no tirofiban infusion group (p = 0.177). Eighty-seven percent of the patients received PCI, after which distal embolization was present in 5.2% of the tirofiban group and in 7.4% of the placebo/no tirofiban infusion group (p = 0.170).
The 30-day follow-up data were complete for 1,339 patients (96%). Fifty patients (3.6%) were unwilling to give written informed consent even though they had given oral informed consent in the ambulance, and 9 patients (0.6%) were lost to follow-up. All clinical outcome- and safety-associated parameters are described in Table 3.The incidence of the composite primary end point was 5.8% in the tirofiban group compared with 8.6% in the placebo/no tirofiban infusion group (odds ratio [OR]: 0.65; 95% confidence interval [CI]: 0.43 to 0.99, p = 0.043). Total mortality was 2.2% in the tirofiban group compared with 4.1% in the placebo/no tirofiban infusion group (OR: 0.53; 95% CI: 0.28 to 1.01, p = 0.051), and the incidence of major bleeding was 3.4% in the tirofiban group and 2.9% in the placebo/no tirofiban infusion group (OR: 1.19; 95% CI: 0.64 to 2.21, p = 0.58). Stroke occurred significantly less often in patients pre-treated tirofiban (0.3% vs. 1.4%, OR: 0.215; 95% CI: 0.046 to 0.999, p = 0.03). Kaplan-Meier curves for survival free from MACE are shown in Figure 3.
Figure 4shows the OR and 95% CI for the incidence of MACE in each of the pre-specified subgroups of patients and in the patients who underwent pPCI (post-hoc analysis). In patients who presented early after the onset of symptoms (time from symptom onset until diagnosis ≤75 min, which is the median in the follow-up group), initial TIMI flow grade 3 of the infarct-related vessel was present in 27.4% the tirofiban group compared with 19.4% in the placebo/no tirofiban group (OR: 1.57; 95% CI: 1.07 to 2.30). In this group of early presenters, the incidence of MACE was 4.3% in the tirofiban group compared with 8.1% in the placebo/no tirofiban group (OR: 0.51; 95% CI: 0.26 to 0.99). The Breslow-Day test for time from symptom onset until diagnosis (≤75 min or >75 min) was not significant (p = 0.216). In patients who underwent pPCI (n = 1,203, 86%), the incidence of MACEs was 4.5% in the tirofiban group compared with 8.5% in the placebo/no tirofiban group (OR: 0.50; 95% CI: 0.31 to 0.82). The Breslow-Day test for pPCI was significant (p = 0.035).
The 1-year follow-up data were available for 1,326 of 1,398 patients (94.8%): 94.0% from the open-label phase and 95.2% from the double-blind phase. For a total of 72 patients (5.2%), there were no 1-year follow-up data for the following reasons: decision of physician (n = 1), lost to follow-up (n = 11), no written informed consent (n = 43) or withdrawal of informed consent (n = 17). The mean follow-up duration was 361 ± 76 days. Total mortality was 3.7% in the tirofiban group compared with 5.8% in the placebo/no tirofiban group (OR: 0.63; 95% CI: 0.38 to 1.06, p = 0.08) (Fig. 5A).Total mortality was lower in the pre-specified subgroup of patients who received HBD tirofiban within 75 min after the onset of symptoms (1.8% vs. 4.3%, p = 0.059), whereas the effect on long-term mortality was less evident for patients pre-treated >75 min after the onset of symptoms (5.7% vs. 6.2%, p = 0.803). Post hoc analysis in the subgroup of patients who underwent pPCI showed that 1-year mortality was significantly lower in the HBD tirofiban group than in the placebo/no tirofiban group (2.4% vs. 5.5%, p = 0.007) (Fig. 5B).
This pooled analysis of the 2 phases of the On-TIME 2 trial showed that early initiation of HBD tirofiban, in addition to aspirin, heparin, and high-dose clopidogrel, improved the clinical outcome at the 30-day follow-up, with no significant increase in major bleeding. The strong trend toward a lower mortality rate was maintained at the 1-year follow-up. The effect was more evident in patients who received HBD tirofiban early after the onset of symptoms and in patients who underwent pPCI.
In the randomized, double-blind, placebo-controlled On-TIME 2 trial, we demonstrated that the early administration of HBD tirofiban in the ambulance resulted in a significantly better ST-segment resolution, both before and after pPCI (9). This represents a better myocardial perfusion with a GP IIb/IIIa blocker. With the pooled analysis of the 2 study phases, we were also able to show an effect of pre-hospital HBD tirofiban on clinical outcome. The worse ST-segment resolution in the placebo or no HBD tirofiban arm might be a reflection of insufficient inhibition of platelet aggregation with aspirin and clopidogrel alone, which was shown in our previous work that found the limited inhibition of platelet aggregation during the first hours, when platelet activation is highest, in patients with STEMI (12).
The On-TIME 2 trial shows that especially mortality and urgent PCI were reduced by early initiation of tirofiban. The reduction in mortality is probably related to the improved myocardial reperfusion. Urgent target vessel revascularization was reduced mainly through a lower rate of stent thrombosis, which again may affect mortality (13).
The beneficial effects of tirofiban on MACEs after 30 days and on mortality were stronger for patients with a brief (≤75 min) duration of symptoms before diagnosis and for patients who underwent pPCI. The number needed to treat (NNT) in the subgroup of early presenters was 26, and in the subgroup of patients with a longer period from symptom onset until diagnosis (>75 min), it was 113. The fact that the early presenters, in particular, have more benefit from early tirofiban treatment is probably due to the lytic effects of GP IIb/IIIa inhibitors on fresh thrombi. Patients who received tirofiban within 75 min after symptom onset had a significantly higher initial TIMI flow grade 3 (27.4% vs. 19.4%, p = 0.02). Therefore, every effort should be made to administer the medication as soon as possible in the ambulance at the first contact with the patient. This pre-hospital administration of medication differs from the procedure in most other trials in which medication was started in the hospital. In our trial, the median time from the onset of symptoms until the initiation of the medication was only 90 min (Fig. 2). This compares favorably with other STEMI trials, such as the BRAVE-3 (Bavarian Reperfusion AlternatiVes Evaluation) study and the FINESSE (Facilitated Intervention With Enhanced Reperfusion Speed to Stop Events) trial, in which GP IIb/IIIa blockers were started a mean of 200 and 165 min, respectively, after the onset of symptoms (14,15). In the BRAVE-3 trial, the GP IIb/IIIa blocker abciximab did not improve myocardial salvage, which probably demonstrates that after a mean of 200 min, organized thrombi do not lyse after the administration of GP IIb/IIIa blockers (14). The results of the ADMIRAL (Abciximab Before Direct Angioplasty and Stenting in Myocardial Infarction Regarding Acute and Long-term Follow-up) study, in which patients received abciximab in the ambulance well in advance of arrival at the catheterization laboratory, are consistent with our findings in showing a greater benefit for patients who received early treatment after the onset of symptoms (16). The observational report of a well-organized STEMI hub-and-spoke network shows a clear advantage of early administration of GP IIb/IIIa blockers in a spoke center emergency department or during ambulance transfer (n = 380; pain to GP IIb/IIIa agents: mean 126 min) compared with delayed (in the catheterization laboratory) administration (n = 744; pain to GP IIb/IIIa agents: mean 169 min) (17). Our findings are in agreement with the results of thrombolytic therapy for patients with STEMI, the efficacy of which is also highest within the so-called golden hour after the onset of symptoms.
In the present study, we pooled the results from the double-blind study phase with 414 patients from the open-label phase who were randomized to HBD tirofiban or no tirofiban in the ambulance. The purpose of this open-label phase was to gain experience in recruiting patients for a randomized clinical trial in the ambulance, without the involvement of a physician. Although the treatment in the open-label phase was not blinded, it is important to note that the study protocol for the 2 phases only differed with regard to trial design: open-label versus blinded. The pooled analysis was pre-specified in the final protocol and statistical plan for the study, and there was no difference in baseline characteristics or in the effect of HBD tirofiban on ST-segment resolution between the 2 phases of the study (Table 2). The fact that the results of the open-label study phase are consistent with the results from the double-blind phase may be interpreted as a proof of concept.
Early pre-hospital administration of a HBD of tirofiban, in addition to aspirin, high-dose clopidogrel and unfractionated heparin, improves the clinical outcome after pPCI in patients with acute STEMI, with no increased risk of major bleeding. The results emphasize the importance of pre-hospital infarct diagnosis in the ambulance and subsequent initiation of potent antithrombotic therapy very early after the onset of symptoms.
For a list of the members of the On-TIME 2 study group, please see the online version of this article.
Members of the On-TIME 2 study group are listed in the Online Appendix. The On-TIME 2 study was partly funded by an unrestricted grant from Merck & Co.and Iroko. Dr. van 't Hof has received speaker fees from Merck, Sanofi-Aventis, and Schering-Plough. Dr. van Werkum has received speaker fees from Siemens and Accumetrics, and is a consultant for The Medicines Company. Dr. Hamm has received advisory board/speaker fees from Merck, Iroko, Lilly, GlaxoSmithKline, Sanofi-Aventis, The Medicines Company, Roche, Iroko, and Abbott. The first 2 authors contributed equally to this work.
- Abbreviations and Acronyms
- confidence interval
- high bolus dose
- interquartile range
- major adverse cardiac events
- myocardial infarction
- odds ratio
- percutaneous coronary intervention
- primary percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- Thrombolysis In Myocardial Infarction
- Received July 2, 2009.
- Revision received November 16, 2009.
- Accepted November 19, 2009.
- American College of Cardiology Foundation
- Simes R.J.,
- Topol E.J.,
- Holmes D.R. Jr..,
- et al.,
- GUSTO-I Investigators
- Goto S.,
- Tamura N.,
- Ishida H.
- Gurbel P.A.,
- Bliden K.P.,
- Zaman K.A.,
- Yoho J.A.,
- Hayes K.M.,
- Tantry U.S.
- Antoniucci D.,
- Rodriguez A.,
- Hempel A.,
- et al.
- van 't Hof A.W.J.,
- ten Berg J.,
- Heestermans T.,
- et al.,
- Ongoing Tirofiban In Myocardial infarction Evaluation (On-TIME) 2 Study Group
- van 't Hof A.W.J.,
- Hamm C.,
- Rasoul S.,
- Guptha S.,
- Paolini J.F.,
- ten Berg J.M.,
- On-TIME 2 Investigators
- Rao A.K.,
- Pratt C.,
- Berke A.,
- et al.
- Mehilli J.,
- Kastrati A.,
- Schulz S.,
- et al.,
- Bavarian Reperfusion Alternatives Evaluation-3 (BRAVE-3) Study Investigators
- Ortolani P.,
- Marzocchi A.,
- Marrozzini C.,
- et al.