OUP user menu

Early and aggressive treatment of patients with ST-segment elevation myocardial infarction: deciphering recent clinical trials and the timing of optimal platelet inhibition

Arnoud W.J. van ‘t Hof
DOI: http://dx.doi.org/10.1093/eurheartj/suq022 D24-D35 First published online: 2 September 2010


In patients with ST-segment elevation myocardial infarction (STEMI), early and effective revascularization is vital to an improved outcome. This not only involves enhancing blood flow through the infarct-related epicardial vessel, but also the restoring of perfusion to the myocardial tissue as it has been strongly linked to clinical outcome. Platelets hold a pivotal position in the distal embolization in patients with STEMI, particularly those who are managed with primary percutaneous coronary intervention, and therefore provide robust rationale to include intense inhibition of their management. Recent clinical studies focusing on STEMI and subanalyses from larger ACS trials reflecting contemporary practice including newer and higher-dose oral agents along with GP IIb/IIIa inhibitors, have painted a mixed picture as to the optimal treatment regimen. Together with dosing, timing of administration could be critical in determining the proper strategy as earlier administration may facilitate more rapid perfusion and ultimately reduce adverse events.

  • ST-segment elevation myocardial infarction
  • Platelet
  • Antiplatelet therapy
  • Distal embolization
  • Thienopyridines
  • GP IIb/IIIa inhibitors

Reperfusion in ST-segment elevation myocardial infarction

Prompt restoration of blood flow and continued perfusion to infarcted myocardial tissue is the ultimate goal of the acute management for patients with ST-segment elevation myocardial infarction (STEMI).13 This requires a multifaceted approach. The first critical component, opening the infarct-related artery, may be accomplished by means of either mechanical [primary percutaneous coronary intervention (PCI)] or pharmacological (fibrinolysis) revascularization. The results of several studies have established the significant benefit of primary PCI when compared with fibrinolysis, in STEMI.48

Timing in primary percutaneous coronary intervention

Still, while primary PCI is the more effective reperfusion strategy in STEMI, timing is vital to its benefit. Subsequent studies have proven that delays to the catheterization laboratory in excess of 90 min from hospital presentation have a significant impact on indicators of myocardial perfusion and clinical outcomes.911 Results from a 1072-patient study conducted by De Luca et al.9 noted that even when adjusting for baseline confounding factors, the time from symptom onset to the first balloon inflation was significantly associated with impaired ST-segment resolution {adjusted odds ratio (OR), 1.01 [95% confidence interval (CI), 1.01–1.02]; P < 0.001} and myocardial blush [adjusted OR, 1.01 (95% CI, 1.01–1.02); P < 0.0001] and according to an analysis of 2173 patients enrolled in the Global Registry of Acute Coronary Events (GRACE) registry who underwent primary PCI, after multivariable adjustment, longer treatment delays were associated with a higher 6-month mortality (P < 0.001) with an increase by 0.18% for every 10 min of delay in door-to-balloon time between 90 and 150 min.11

Acknowledging the proven benefit of primary PCI but that a considerable delay from symptom onset to revascularization has a significant impact on myocardial salvage and clinical outcome, the guidelines put forth by the European Society of Cardiology (ESC) specify primary PCI as the ideal reperfusion strategy, patients with STEMI, though, state that it should ideally be implemented within 120 min of first medical contact.12

Distal embolization in ST-segment elevation myocardial infarction

The other component involves the use of pharmacological therapy to deal with the thrombotic process through the use of agents which exhibit antithrombin and antiplatelet properties. This is particularly important from the perspective of microvascular embolization, which is quite frequent in STEMI, particularly among those who undergo primary PCI. It has been established that often when the lumen of the occluded coronary vessel has been completely re-opened, microvascular perfusion remains impaired.1315 Moreover, microvascular embolization has been determined to be a strong indicator of poor outcome in patients with STEMI who undergo primary PCI. A substudy of 178 patients with STEMI managed with primary PCI noted a significant increase in long-term mortality in patients with distal embolization (44 vs. 9%, P < 0.001).16 Thus, tissue-level reperfusion is an important requirement for attaining the best possible outcome.

Distal embolization and the platelet

Platelet aggregation has been considered an important factor in microcirculatory dysfunction in STEMI. Thrombi consisting mostly of platelets tend to be smaller and are more likely to embolize distally.17 Besides their critical contribution to microcirculatory thrombotic occlusion, platelets may incite vasospasm through the release of serotonin, thromboxane A2, and free radicals. The complementary actions of these two platelet-driven mechanisms results in impaired myocardial perfusion and ischaemia of the impacted area.18

The degree of platelet activity has furthermore been linked with the extent of myocardial damage and clinical outcome in patients with STEMI.1921 Platelets may in fact comprise the majority of thrombus in STEMI as they have been found to make up the main constituent of thrombus aspirated from the coronary arteries of patients with STEMI.17

Antiplatelet therapy in ST-segment elevation myocardial infarction patients undergoing primary percutaneous coronary intervention

A role for antiplatelet therapy in patients with STEMI managed with primary PCI is quite apparent based upon the scientific and clinical evidence. Nevertheless, the exact regimen and combination of agents have not been clearly defined as results from clinical trials varied with respect to the efficacy and safety of available and pending antiplatelet medications.

There are three vital questions which should be discussed. The first is the type, dose, and timing of oral antiplatelet therapy. Another is the extent of benefit which is afforded by treatment with an intravenous glycoprotein IIb/IIIa inhibitor (GPI) in patients who are treated with high-dose oral antiplatelet therapy. The third is clarification of the type and timing of GPI therapy as well as patients most likely to benefit, based upon the most recent data.

Type, dose, and timing of oral antiplatelet therapy

For the management of STEMI patients treated with primary PCI, the writing groups of both the ESC and the American College of Cardiology (ACC)/American Heart Association (AHA) advocate pre-treatment with a clopidogrel loading dose, at least 300 and 600 mg if additional platelet inhibition is desired.15,22 Yet, the basis for these recommendations involves trials of NSTE-ACS or elective PCI as well as those comprised of STEMI patients not managed with primary PCI. For example, two large-scale trials ClOpidogrel and Metoprolol in Myocardial Infarction Trial (COMMIT) and he Clopidogrel as Adjunctive Reperfusion Therapy (CLARITY)-TIMI 28 evaluated clopidogrel in patients with STEMI, although neither enrolled patients who had undergone primary PCI.23,24 Furthermore, the analysis of the main outcome results of CLARITY noted that clopidogrel's benefit was due primarily to a reduction in the incidence of infarct artery occlusion rather than early tissue-level reperfusion.25

More recently, evidence which supports clopidogrel's role in patients managed with primary PCI exists only in the form of data from registries and subgroup analyses. A small prospective registry evaluated the efficacy of clopidogrel pre-treatment on both angiographic and clinical outcomes in patients treated with primary PCI. It was observed that pre-treatment with a clopidogrel loading dose was associated with improved myocardial perfusion [OR of 2.2 for TIMI Myocardial Perfusion grade 3 (1.2–3.9; P = 0.01)] and a significant reduction in the incidence of reinfarction at 30 days (0 vs. 3.2%, respectively, P= 0.04).26 A 1-year mortality benefit was observed in the Acute Coronary Syndromes (ACOS) registry.27 In the subgroup of 2707 patients managed with primary PCI, mortality was significantly lower in patients treated with aspirin (ASA) plus clopidogrel when compared with ASA alone (OR, 0.38; 95% CI, 0.23–0.62), although neither the loading dose nor the timing of clopidogrel was specified in the study results.

A subgroup analysis of CURRENT-OASIS 7, a 2 × 2 factorial, randomized trial was designed to determine whether treatment with clopidogrel using a double (600 mg loading dose/150 mg QD for 7 days then 75 mg/day) or standard (300 mg loading dose/75 mg QD) dose regimen provides greater benefit when compared with standard dosing, when administered on top of ASA.28 Out of 17 232 patients, 6346 presented with STEMI. Although this was not exclusively a primary PCI population, a significant majority had undergone the procedure. By 30 days, there was a 37% significant reduction in the risk of MI in patients treated with double-dose clopidogrel [1.9 vs. 1.2%; hazard ratio (HR), 0.63 (95% CI, 0.41–0.94)]. No mortality benefit was observed.29

A relatively recent development has been the acknowledgement of prasugrel as another option for thienopyridine treatment in patients with STEMI.30 This too is based on a subgroup analysis of the TRITON-TIMI 38 trial which included patients with STEMI; however, not all patients were managed with primary PCI. Briefly, the TRITON-TIMI 38 trial compared prasugrel (60 mg loading dose/10 mg daily maintenance dose) with clopidogrel (300 mg loading dose/75 mg daily maintenance dose) in patient with acute coronary syndrome (ACS).31 Twenty-six per cent of the patients (n = 3534) had the diagnosis of STEMI with 69% of them undergoing primary PCI.32 Investigators observed that in the overall STEMI subgroup, when compared with those treated with clopidogrel, prasugrel-treated patients experienced a significant reduction in the primary endpoint of death, MI, or stroke at 30 days (9.5 vs. 6.5%, P = 0.017) which was maintained out to 15 months (12.4 vs. 10.0%, P = 0.0221). This benefit was only observed in patients with an anterior MI. In the subgroup that underwent primary PCI however, treatment with prasugrel provided no significant benefit. The 30-day primary endpoint was not significantly different between clopidogrel and prasugrel [8.2 vs. 6.6%; HR, 0.80 (95% CI, 0.60–1.08); P = 0.1440] or 15 months [11.6 vs. 10.2%; HR, 0.87 (95% CI, 0.68–1.11); P = 0.2662]. Inconsistent with the intent-to-treat population in TRITON-TIMI 38, the incidence of TIMI major bleeding in the STEMI cohort observed at 30 days and 15 months was not statistically different between the two treatment groups, although patients in the prasugrel group experienced a significant increase in coronary artery bypass graft-related TIMI major bleeding.

Ticagrelor is a novel agent referred to as a cyclo-pentyl-triazolo-pyrimidines. Unlike clopidogrel and prasugrel, ticagrelor is a non-thienopyridine and exhibits a more rapid onset of action holding the promise of obviating the need for pre-treatment. Ticagrelor is also reversible; therefore, platelet activity returns within a few hours when compared with days for clopidogrel and prasugrel.33 To determine the effect of ticagrelor in patients with STEMI treated with primary PCI, a subgroup analysis was conducted by Steg et al.34 of the 8430 patients (38% of study population) in the PLATelet inhibition and patient Outcomes (PLATO) trial who had a final diagnosis of STEMI and were scheduled for primary PCI. Essentially, PLATO was designed to compare ticagrelor with clopidogrel in patients with ACS.35 At the time of PCI, patients were administered either clopidogrel using a 300 mg loading dose (if patient had not received for prior 5 days) and 75 mg QD maintenance or ticagrelor at a 180 mg loading dose and 90 mg QD.

In the STEMI subgroup, there was a significant reduction with ticagrelor in the rate of the primary endpoint (death from vascular causes, MI, or stroke) at 12 months (11 vs. 9.3%, P = 0.02), mainly due to a significant reduction in MI (6.1 vs. 4.7%, P = 0.01). PLATO was not powered to detect differences in mortality. Stent thrombosis was significantly reduced in patients randomized to ticagrelor. This included ARC-defined definite (2.5 vs. 1.6%, P = 0.01), probable or definite (3.6 vs. 2.5%, P = 0.01), and possible, probable, or definite (4.4 vs. 3.2%, P = 0.02).

Protocol-defined (PLATO) major bleeding was not significantly different between clopidogrel and ticagrelor (9.3 vs. 9.0%, respectively, P = NS). There were also no significant differences between clopidogrel and ticagrelor in the rates of protocol-defined (PLATO) life-threatening or major bleeding (4.9 vs. 4.5%, respectively, P = NS), TIMI major bleeding (6.4 vs. 6.0%, P = NS), fatal bleeding (0.1 vs. 0.3%, respectively, P = NS), or transfusion (any blood product) (8.9 vs. 8.9%, P = NS).

Neither TRITON-TIMI 38 nor PLATO employed a clopidogrel regimen which included a loading dose higher than 300 mg, and in both, the vast majority of patients were administered the loading dose during PCI. It may be hypothesized that the results of these trials may have been different if higher loading doses were used, given the results demonstrated in the primary PCI subgroup in the CURRENT-OASIS 7 trial with high-dose clopidogrel. Additionally, earlier pre-loading might also have had an impact. Studies have demonstrated that even when using a 600 mg loading dose of clopidogrel, it takes 3–4 h before its peak effect is observed.36 Further clarity is necessary concerning the comparative effectiveness of prasugrel and ticagrelor alongside higher doses of clopidogrel.

Role for triple antiplatelet therapy in primary percutaneous coronary intervention

The debate ensues whether GPIs should be administered in patients with STEMI undergoing primary PCI in patients for whom high-dose oral agents such as thienopyridines are used. Furthermore, there is the issue of timing—pre-angiogram or pre-PCI and drug choice. Using a powerful, rapid-acting antiplatelet agent should make clinical sense, given the relationship between elevated platelet activity and myocardial perfusion and poor outcome in STEMI, not to mention the brief time window for revascularization. This is particularly valid in the presence of oral antiplatelet therapy. Although a prasugrel loading dose of 60 mg tends to work considerably more rapidly than that for clopidogrel, 1–1.5 h is still required for peak platelet inhibition.37 As the goal is to open the culprit vessel as quickly as possible, there is often not enough time to allow for the oral antiplatelet therapy to achieve optimal effect.

Data supporting a principle role for GPIs in the management of STEMI patients undergoing PCI have historically been quite robust, although much of it was derived from studies concerning abciximab and were prior to the era of dual antiplatelet use.3840

Later studies evaluating these agents as part of more contemporary regimens have been less consistent.4143 In addition, the matter of timing with respect to drug initiation (pre-angiogram or pre-PCI), thienopyridine loading, as well as the period between symptom onset and admission has come into play.

Results of a recent meta-analysis by De Luca et al.41 suggest that small-molecule agents such as tirofiban and eptifibatide may not be dissimilar from abciximab in patients being managed with primary PCI for STEMI. A total of 2197 patients were included in the analysis comprised of trials of abciximab, eptifibatide, and tirofiban (using the high-dose bolus regimen). There was no significant difference with abciximab in the rate of ST-segment resolution (67.8 vs. 68.2%, P = 0.66), 30-day mortality (2.2 vs. 2.0%, P = 0.66), or reinfarction (1.2 vs. 1.2%, P = 0.88). Major bleeding complications were also not significantly different (1.3 vs. 1.9%, P = 0.27).

This was echoed in The Multicentre Evaluation of Single High-Dose Bolus Tirofiban vs. Abciximab With Sirolimus-Eluting Stent or Bare Metal Stent/ in Acute Myocardial Infarction (MULTISTRATEGY) trial. MULTISTRATEGY was the single largest trial which compared a small-molecule GPI with abciximab in patients with STEMI managed with primary PCI.44 An open-label, 2 × 2 factorial trial of 745 patients presenting with STEMI or new left bundle branch block at multiple European sites, MULTISTRATEGY, evaluated the effect of tirofiban using the 25 µg/kg bolus regimen and abciximab in combination with either sirolimus-eluting stent or bare metal stent. GPI therapy was initiated at first medical contact in all patients with treatment initiation occurring between 27 and 60 min prior to the first balloon inflation. All patients were pre-treated with 300 mg of clopidogrel.

Tirofiban demonstrated non-inferiority to abciximab with respect to the primary endpoint of at least 50% ST-segment elevation resolution at 90 min post-intervention. This occurred in 85.3% of the patients who had received tirofiban and in 83.6% who received abciximab [relative risk, 1.020 (97.5% CI, 0.958–1.086); P < 0.001 for non-inferiority]. There was no significant difference in major adverse cardiac events (MACE) observed between tirofiban and abciximab at 30 days (4.0 vs. 4.3%, respectively, P = 0.85) and at 8 months (9.9 vs. 12.4%, respectively, P = 0.30; Figure 1).

Figure 1

MULTISTRATEGY: rate of MACE at 30 days and 8 months by drug treatment. MACE = death, reinfarction, and TVR.

Additionally, there were no significant differences in the rate of stent thrombosis at 30 days and 8 months between the tirofiban and abciximab groups (30 days: 1.9 vs. 1.3%, P = 0.56; 8 months: 2.7 vs. 2.7%, P = 0.99). However, thrombocytopenia (platelet count ≤50 000/mm3) was more often observed in the abciximab-treated patients (tirofiban: 0.5% vs. abciximab: 2.4%; P = 0.03). Such increases in thrombocytopenia are directionally consistent with other trials of abciximab in patients with ACS.

It should be recognized that a direct correlation between thrombocytopenia and an increased risk of major bleeding and clinical events in patients with ACS (UA/NSTEMI and STEMI) has been determined through results from post hoc analyses and a large multinational registry.4548 A post hoc analysis from the CADILLAC trial revealed that patients with STEMI managed with primary PCI who developed in-hospital thrombocytopenia were at a significantly higher risk for major bleeding, transfusion, and 30-day mortality than patients without thrombocytopenia.46 Similarly, an analysis of 52 647 patients enrolled in the GRACE registry demonstrated that STEMI patients who develop thrombocytopenia are at a higher likelihood of major bleeding, recurrent infarction, stroke, and in-hospital mortality.47

Early use of GP IIb/IIIa inhibitors in ST-segment elevation myocardial infarction

Prior studies have suggested that initiation of a GPI during the period between first medical contact and angiography may be clinically beneficial in STEMI.39,49 Such benefit may be related to early myocardial salvage produced by platelet inhibition within the microvasculature.16,17 However, results have been inconsistent; therefore, there is no clear consensus with respect to the value offered through early administration. Still, it should be considered that a number of factors have played significant roles in the diversity of these results. First and foremost, trials have compared early initiation of GPIs with planned PCI administration as well as with no therapy (including option of provisional use at time of intervention) and endpoints have been measured at various times in relation to angiography and PCI (Tables 1 and 2).39,42,43,5062 Not surprisingly, when compared with planned administration, early use has commonly demonstrated significant improvement in angiographic measures of epicardial and tissue-level perfusion when assessed prior to angiography, as evident in the ReoPro-Bridging and ERAMI trials. However, there was little benefit noted when clinical parameters served as the primary measure of efficacy, as demonstrated in FINESSE.52 Another aspect which should be factored in is the practice of high-dose clopidogrel pre-treatment. Although in the vast majority of trials comparing early with PCI administration, clopidogrel (if administered) was not used until immediately post-intervention, patients in the AGIR-2 study were indeed preloaded.50 Contrary to findings from the other trials, in AGIR-2, there was no significant benefit of early GPI administration with respect to pre-PCI measures of coronary flow. It may be hypothesized that in this particular study population, the antiplatelet effects induced by clopidogrel were sufficient for pre-procedural perfusion with no added value from abciximab.

View this table:
Table 1

Trials of early vs. late GP IIb/IIIa inhibitors in ST-segment elevation myocardial infarction

StudyStudy designSymptom onset to randomization (h)Clopidogrel loadingPrimary endpointResults
APEX-AMI (subgroup analysis)Retrospective analysis of pre-sheath GPI (abciximab, eptifibatide, tirofiban) vs. PCI GPI≤6Not consistentlyMortality at 90 days Composite of death, centrally adjudicated CHF or cardiogenic shock at 90 daysMortality: none, 5.5%; pre-sheath use, 3.2%; PCI use, 4.8% Adjusted HR, 0.68 (95% CI, 0.48–0.95), P = 0.025Composite of death/CHF/shock: none, 11.6%; pre-sheath use, 8.4%; PCI use, 10%
Adjusted HR, 0.81 (95% CI, 0.65–1.00), P = 0.054
AGIR-2Pre-hospital vs. cath lab tirofiban (25 µg/kg dose bolus regimen)<12600 mgTIMI 2–3 flow at initial angiographyPre-hospital, 44.2%; cath lab, 39.7%; P = 0.42
FINESSEEarly abciximab vs. PCI abciximab≤6?Death, ventricular fibrillation cardiogenic shock, and CHF at 90 daysEarly, 10.5%; PCI, 10.7%; P = 0.86
On-TIMEPre-hospital vs. PCI tirofiban (10 µg/kg dose bolus regimen)≤6NoTIMI flow grade 3 of the infarct-related vessel at initial angiographyPre-hospital, 19%; PCI, 15%; P = 0.22
ReoPro-BRIDGINGEarly abciximab vs. PCI abciximab≤6NoTIMI flow grade, cTFC at initial angiographyTIMI 3 flow: early, 29% vs. PCI, 7%; P = 0.042
cTFC: early, 58.4 ± 32.7; PCI, 78.9 ± 28.4; P = 0.018
RELAX-AMIEarly vs. PCI abciximab≤6NoInitial TIMI flow grade, cTFC, MBG; LVF recoveryTIMI flow grade 3: early, 24%; PCI, 10%; P = 0.01
cTFC: early, 78 ± 30; PCI, 92 ± 21; P = 0.001
MBG 2 or 3: early, 15%; PCI, 6%; P = 0.02
EF: early, 8 ± 7%; PCI, 6 ± 7%; P = 0.02
Mean gain wall motion score index: early, 0.4 ± 0.3; PCI, 0.3 ± 0.3; P = 0.03
ERAMIEarly vs. PCI abciximab≤12?Angiographic endpoints at initial angiographyPrior to PCI, no significant differences were observed between the two groups regarding the angiographic endpoints or ST-segment resolution
Zorman et al.Early vs. PCI abciximab≤12NoEarly (60 min) ST-segment resolution; TIMI 3 flow at initial angiographyNeed to obtain data
Cutlip et al.Early vs. late or no tirofiban (10 µg/kg bolus followed by 0.15 µg/kg/min infusion)≤12NoTIMI flow at initial angiographyEarly, 39%; PCI, 27%; P> 0.20
Emre et al.Early vs. PCI tirofiban (10 µg/kg bolus followed by 0.15 µg/kg/min infusion)<6NoMyocardial salvageFinal infarct size: early, 11.8 ± 5.2%; PCI, 22.4 ± 6.2%; P = 0.01
Salvage index: early, 0.68 ± 0.22; PCI, 0.44 ± 0.18; P = 0.003
INTAMIEarly vs. PCI or no eptifibatide<12NoTIMI 3 flow at initial angiographyEarly, 34%; PCI or no, 10%; P = 0.01
TITAN-TIMI 34Early vs. PCI eptifibatide<6NocTFC at initial angiographyEarly, 77.5 ± 32.2; PCI, 84.3 ± 30.7; P = 0.049
  • GPI, GP IIb/IIIa inhibitor; CHF, congestive heart failure; cTFC, corrected TIMI frame count; HR, hazard ratio; MBG, myocardial blush grade; LVF, left ventricular function; EF, ejection fraction; PCI, percutaneous coronary intervention.

View this table:
Table 2

Trials of early vs. no GP IIb/IIIa inhibitor in ST-segment elevation myocardial infarction

StudyStudy designSymptom onset to randomization (h)Clopidogrel loadingPrimary endpointResults
DEBATEREarly abciximab vs. no/provisional abciximab≤6600 mgTarget vessel failure at 30 daysAbciximab, 2%; placebo, 8%; P = 0.001
BRAVE-3Early vs. no/provisional abciximab≤24600 mgFinal infarct size at 5–7 daysMean % ± SD of left ventricle: abciximab, 15.7 ± 17.2%; placebo, 16.6 ± 18.6%; P = 0.47
On-TIME 2Pre-hospital vs. no/provisional tirofiban (25 µg/kg dose bolus regimen)≤6600 mgResidual ST segment deviation (>3 mm) 1 h after PCITirofiban, 43.3%; placebo, 36.0%; P = 0.02
ADMIRALEarly vs. no/provisional abciximab<12NoDeath, reinfarction, or urgent revascularization of the target vessel at 30 daysAbciximab, 6.0%; placebo, 14.6%; P = 0.01
  • PCI, percutaneous coronary intervention.

Results have been quite varied among the trials which evaluated early GPIs with provisional use at the time of PCI. Most notable are the results of the 800-patient Bavarian Reperfusion AlternatiVes Evaluation (BRAVE)-3 trial which was designed primarily to compare the effect of abciximab on the final infarct size when compared with placebo, in patients with STEMI being managed with primary PCI and pre-treated with clopidogrel 600 mg.43At 5–7 days, when compared with placebo, the addition of abciximab did not result in a significant reduction in the primary endpoint of final infarct size (16.6 vs. 15.7%, P = 0.47) nor clinical endpoints at 30 days. At 1 year, there was also no significant difference in the rate of death, MI, stroke, or urgent target vessel revascularization (uTVR) between the two treatment groups (25.7 vs. 23%, P = 0.46).63 Moreover, there was a non-significant increase with abciximab in the rate of the combined endpoint of death, MI, or stroke (6.0 vs. 9.2%, P = NS), which was driven primarily by a non-significant increase in the incidence of all-cause mortality (4.0 vs. 6.8%, P = 0.08). It should be acknowledged that BRAVE-3 enrolled patients who had been experiencing symptoms for as long as 24 h, which is considerably longer than other contemporary STEMI trials. The onset of symptoms to drug administration may indeed have a correlation with mortality, although the reason for this is unknown. To determine whether this may have impacted the results, the primary endpoint was analysed by the time interval from symptom onset to hospital admission and from study drug administration to primary PCI. Although no significant differences in infarct size were observed upon this interval-based analysis (Figure 2), there was an indication that abciximab might be most effective in patients presenting within 3 h from symptom onset.

Figure 2

Mean difference in infarct size between treatment groups in various subgroups. Reproduced with permission from Mehilli et al.43

With respect to small-molecule agents, results have also been rather mixed. The Safety and Efficacy Study of Integrilin Facilitated PCI vs. Primary PCI in ST Elevation Myocardial Infarction (ASSIST) trial was designed to ascertain whether eptifibatide administered pre-angiography could provide any additional clinical value in patients with STEMI who had been experiencing symptoms for <12 h, being managed with primary PCI, who had been preloaded with 600 mg of clopidogrel.42 Clopidogrel was administered at a median of 75 min prior to PCI. ASSIST demonstrated no significant difference at 30 days between placebo and eptifibatide in the frequency of death, reinfarction, or recurrent severe ischaemia (5.5 vs. 6.5%, P = 0.69) nor in mortality (2 vs. 3.5%, P = 0.54). Although these results are consistent with that of BRAVE-3, in ASSIST, patients had only been experiencing symptoms for a comparatively short time (mean time from symptom onset to first balloon inflation was 195 min).

Results of the safety analysis demonstrated a significant increase in the composite of TIMI major or minor bleeding with eptifibatide when compared with the placebo group (14.6 vs. 22.4%, P = 0.04). Non-significant differences in the rates of TIMI major and minor bleeding were demonstrated.

The Ongoing Tirofiban in Myocardial Evaluation (On-TIME) 2, a large-scale, multi-centred, prospective, randomized clinical trial, was designed to assess the benefit of pre-hospital tirofiban administration (using 25 µg/kg bolus regimen) on top of dual antiplatelet therapy with high-dose clopidogrel, in patients with STEMI planned for primary PCI.61,64,65 Ninety-five per cent of patients were randomized after pre-hospital infarct diagnosis in the ambulance. On-TIME 2 was conducted in two phases, open-label and double-blind, placebo controlled.

A total of 1398 patients STEMI (n = 414 in open-label phase; n = 984 in double-blind phase) were randomly assigned to receive pre-treatment with either tirofiban or placebo prior to a primary PCI.65 In addition to receiving a bolus of 5000 IU of unfractioned heparin and ASA 500 mg intravenously, along with 600 mg loading dose of clopidogrel orally, subjects were assigned randomly to pre-hospital treatment of tirofiban (25 µg/kg bolus and 0.15 µg/kg/min maintenance infusion for 24 h) or placebo (bolus plus infusion). Results of the double-blind phase noted that at 1 h post-PCI, the rate of residual ST-segment deviation (>3 mm) was significantly reduced in the tirofiban group (43.3 vs. 36.0%, P = 0.02). The amount of residual ST-segment deviation was as well significantly reduced at 1 h in the tirofiban-treated patients [3.6 (SD 4.6) vs. 4.8 (6.3) mm, P = 0.003], and this effect was evident even prior to angiography [10.9 (9.2) vs. 12.1 (9.4) mm, P = 0.028].61 At 30 days, the results of the pooled analysis demonstrated that the combined rate of death, recurrent MI, or uTVR (MACE) at 30 days was significantly lower in the tirofiban arm (8.6 vs. 5.8%, P = 0.043). There was a strong trend towards a significant decrease in mortality with tirofiban with respect to all-cause mortality (4.1 vs. 2.2%, P = 0.051). At 1-year follow-up, the mortality difference was maintained (5.8 vs. 3.7%, P = 0.078).65

Pre-specified subgroup analyses were conducted with respect to 30-day outcomes in On-TIME 2. Patients who presented early after symptom onset (median time ≤75 min) demonstrated a significant reduction in MACE with tirofiban therapy [8.1 vs. 4.3%; OR, 0.51 (95% CI, 026–0.99)].65 This is consistent with BRAVE-3 in which time from symptom onset to admission was a potential factor in the results for abciximab in a similar patient population.43

In patients who underwent primary PCI (n = 1203, 86%), as illustrated in Figure 3, there was a significant reduction in MACE at 30 days [8.5 vs. 4.5%; OR, 0.50 (95% CI, 0.31–0.82)]. Clinical benefit was sustained at 1 year as results of a post-hoc analysis indicated a significantly lower mortality rate associated with tirofiban therapy (5.5 vs. 2.4%, P = 0.007).65

Figure 3

On-TIME 2: clinical outcomes at 30 days and 1 year in patients who underwent primary percutaneous coronary intervention (n = 1213). P-value for major adverse cardiac events is based on the Breslow–Day. *Adjusted.

The safety analysis revealed no significant differences in the incidence of major bleeding between the tirofiban (3.4%) and placebo (2.9%) arms (OR, 1.19; 95% CI, 0.64–2.21, P = 0.58), although the occurrence of stroke was significantly lower in the tirofiban-treated patients (0.3 vs. 1.4%, P = 0.031).

A comparison of On-TIME 2 with BRAVE-3 and ASSIST exposes the most noticeable differences being the timeframes between symptom onset and PCI and clopidogrel loading. In On-TIME 2, the large majority of patients were randomized within 6 h of symptom onset, whereas in ASSIST and BRAVE-3, patients were allowed enrolment as late as 12 and 24 h, respectively. As discussed earlier, there was a trend towards benefit with abciximab in BRAVE-3 in patients who were enrolled <180 min from the onset of symptoms, and the incidence of MACE was significantly reduced in the subgroup of patients who were diagnosed within 75 min of diagnosis, whereas there was no benefit in those diagnosed beyond that time.

There appears a marked difference between the trials with respect to the timing of clopidogrel in relation to the administration of the GPI. In both BRAVE-3 and ASSIST, the GPI was administered much later than the thienopyridine, often in the catheterization laboratory, whereas in On-TIME 2, both clopidogrel and tirofiban were given at the same time at a median of 75 min after the onset of symptoms in the ambulance. In addition, in On-TIME 2, the median time from clopidogrel pre-treatment to PCI was 55 min, which was considerably shorter than that which occurred in either BRAVE-3 (73 min) or ASSIST (75 min). It is probable that in patients treated with thienopyridines, that early administration of GPIs may offer benefit only in those who do not receive them early enough to provide the desired antiplatelet effect. Finally, risk level may have also impacted the value offered by GPI therapy as supported by a meta-analysis conducted by De Luca et al.,66 comprising 16 randomized trials of GPI therapy in STEMI (n = 10 085). Results noted a particular advantage afforded by GPIs in high-risk patients with respect to 30-day mortality (P = 0.008). In BRAVE-3, only low- to intermediate-risk patients were enrolled.43

Guideline recommendations for GP IIb/IIIa inhibitors

At the present time, the writing groups of the ESC and the ACC/AHA advocate that the use of GPI is reasonable in STEMI patients who undergo primary PCI, when initiated just prior to the procedure.12,30 Since abciximab is the most-studied GPI in patients with STEMI relative to either tirofiban or eptifibatide, it has been assigned a higher classification than either of the small-molecule agents. However, the results of contemporary trials such as MULTISTRATEGY have resulted in a recent update in the ACC/AHA's position, thereby recognizing the benefit of tirofiban provided to patients with STEMI managed with primary PCI30 and the fact of tirofiban as being comparable to that of abciximab.

With respect to pre-angiography GPI therapy, both groups are currently uncertain of its usefulness, citing the failure of the Facilitated PCI in Patients With ST-Elevation Myocardial Infarction (FINESSE) trial and noting a lack of a significant difference with respect to the clinical endpoint (death, recurrent MI, or uTVR) at 30 days. It should be kept in mind that neither group makes any reference to the long-term mortality benefit demonstrated with early tirofiban in On-TIME 2, specifically in patients who underwent primary PCI, nor any of the factors that might have led to the heterogeneity of results in trials of early therapy. In addition, it is not an accurate generalization of the effectiveness of early administration as FINESSE compared early routine GPI on top of high-dose clopidogrel vs. late GPI.52


In patients with STEMI managed with primary PCI, the goal of rapid reperfusion combined with the central role of the platelet underscores the need for intensive antiplatelet therapy timed effectively in order to achieve optimal effect. Although trials such as CURRENT-OASIS 7 and TRITON-TIMI 38 have demonstrated the efficacy of dual antiplatelet therapy including high-dose clopidogrel and prasugrel in patients with STEMI, these regimens may not be loaded early enough to sufficiently inhibit platelet activity, given limited time available between diagnosis and primary PCI. Although ticagrelor appears promising, it has not yet been proven in comparison with high-dose clopidogrel.

Triple antiplatelet therapy including GPIs initiated just prior to PCI is currently recommended by the ESC and ACC/AHA guidelines for managing patients with STEMI. Nevertheless, these guideline authors are ambivalent on the role of pre-angiography initiation for these agents. Although it is an accurate assessment that the results of various trials have been less than consistent, the trend has been in support of early vs. late (pre-PCI) administration. The 1398-patient On-TIME 2 trial confirms the benefit of pre-hospital, very early tirofiban, particularly in those who undergo primary PCI, even when pre-treated with high-dose clopidogrel. When contrasted with the results of BRAVE-3 and ASSIST, it may be assumed that timing of the GPI and clopidogrel as well as the level of risk play a significant role in the benefit offered by these drugs. Still, given that in clinical practice, it is not certain that thienopyridines will be administered early enough, it is reasonable to treat all high-risk patients planned for primary PCI, with GPIs prior to angiography.

Conflict of interest: A.W.J.H. has received speakers fee from Merck, Sanofi-Aventis, and Schering-Plough. The On-TIME 2 study was partly funded by an unrestricted grant from Merck & Co. and Iroko.


Supported by an unrestricted grant from Iroko Cardio, LLC.


View Abstract