Impact of Selective Infarct-Related Artery Infusion of Tirofiban on Myocardial Reperfusion and Bleeding Complications in Patients With Acute Myocardial Infarction: The SUIT-AMI Trial

Abstract: Background. It is unclear whether selective infarct-related artery (sIRA) administration of glycoprotein IIb/IIIa receptor inhibitors (GPI) may further improve myocardial reperfusion without increasing bleeding in patients with ST-elevation myocardial infarction (STEMI). The aim of this study was to compare the impacts of sIRA infusion with conventional intracoronary (IC) administration of GPI tirofiban on myocardial reperfusion and clinical prognosis in STEMI patients. Methods. A total of 203 consecutive STEMI patients within 12 hours of symptom onset were randomly assigned to receive tirofiban (10 µg/kg body weight) through aspiration catheter (n = 107) or guiding catheter (n = 96) after thrombus aspiration, and following 12 hours of intravenous infusion (0.1 µg/kg/min). The primary endpoint was the corrected Thrombolysis in Myocardial Infarction (TIMI) frame count (CTFC) after the procedure. The secondary endpoint was major adverse cardiac events at 30 days and 6 months. The safety endpoint was in-hospital bleeding, defined according to the TIMI bleeding classification. Results. Similar primary endpoints (CTFC of 28 ± 17 vs 27 ± 15; P=.841) were observed in the sIRA and conventional IC administration groups, respectively. The incidence of major adverse cardiac events (6.5% vs 5.2% at 30 days [P=.688] and 8.4% vs 7.3% at 6 months [P=.762]) and in-hospital major or minor bleeding (9.3% vs 8.3%; P=.800) were comparable in both groups. Conclusions. After thrombus aspiration, sIRA infusion of tirofiban does not improve myocardial reperfusion assessed by CTFC, as well as ischemic or bleeding events in this study. 

J INVASIVE CARDIOL 2013;25(8):376-382

Key words: bleeding, STEMI, tirofiban

_________________________________

In patients with ST-elevation myocardial infarction (STEMI), prompt reperfusion therapy with primary percutaneous coronary intervention (PCI) results in lower rates of reinfarction, stroke, and death than fibrinolytic therapy, which makes PCI the preferred treatment strategy.¹ However, despite restoration of epicardial Thrombolysis in Myocardial Infarction (TIMI) 3 flow after primary PCI, 5%-50% of patients have impaired myocardial reperfusion. Persistent impaired myocardial reperfusion on angiography is associated with more than a doubling of mortality at short- or medium-term follow-up.²

The implementation of adjunctive mechanical and pharmacological therapies during primary PCI, including manual thrombus aspiration³ and intravenous administration of glycoprotein IIb/IIIa receptor inhibitors (GPI), may improve myocardial reperfusion and clinical outcomes in STEMI patients. Recently, observational and prospective studies demonstrated that intracoronary (IC) administration of GPI is associated with better myocardial reperfusion than intravenous administration in STEMI patients,⁴ suggesting that a higher local drug concentration of GPI could further improve clinical outcomes. Furthermore, a small randomized study reported that local drug concentration was even higher with selective infusion of GPI directly to the lesion through a dedicated reperfusion catheter than through a guiding catheter, which resulted in less thrombus burden and improved coronary microcirculation.⁵ Our retrospective study also suggests that selective infarct-related artery (sIRA) infusion of tirofiban through aspiration catheter after aspiration thrombectomy in STEMI patients undergoing primary PCI may be safe and effective.⁶ Thus, we hypothesize that sIRA infusion of GPI through aspiration catheter may further improve myocardial reperfusion without increasing bleeding compared to IC administration through guiding catheter in STEMI patients undergoing primary angioplasty.

Methods

Patients. This study, termed the Impacts of Super-selective Infarct-Related Artery (sIRA) Infusion of Tirofiban on Myocardial Reperfusion and Bleeding Complications in Acute Myocardial Infarction Patients (SUIT-AMI) trial, was a prospective, randomized, open label, double-center trial with blinded assessment of the study endpoints (www.clinicaltrials.gov. NCT01181388). The initial 389 consecutive STEMI patients undergoing primary PCI between January 2010 and April 2011 were screened at two institutions that were both high-volume university hospitals providing 24-hour emergency cardiac care. Of these, a total of 203 patients met the inclusion criteria and were enrolled into this study; the other 186 patients were excluded because of ineligibility (Figure 1). The study was approved by the Medical Ethics Review Committees at these two hospitals. All patients gave informed consent.

Inclusion criteria were: STEMI patients, defined as chest discomfort suggestive of myocardial ischemia for at least 30 minutes before hospital admission; time from symptom onset of <12 hours; an electrocardiogram (ECG) with new ST-segment elevation in 2 or more contiguous leads of ≥0.2 mV in leads V2 to V3 and/or ≥0.2 mV in other leads; age between 18 and 75 years old; and preprocedural TIMI flow ≤ 2. Exclusion criteria were: left bundle branch block; rescue PCI after thrombolytic therapy; presence of cardiogenic shock; left main coronary artery as the infarct-related artery; stent thrombosis; a life expectancy of <6 months; pregnancy; and any contraindications for the use of tirofiban and anticoagulation agents, including active internal bleeding, history of hemorrhagic stroke, ischemic stroke within past 6 months, recent major surgery or trauma, intracranial neoplasm, arteriovenous malformation or aneurysm, bleeding diathesis, severe uncontrolled hypertension, thrombocytopenia, vasculitis, hypertensive or diabetic retinopathy, severe liver or kidney failure.

Study protocol. Patients were pretreated with oral 300 mg aspirin, oral 300-600 mg clopidogrel, and intravenous (IV) 5000 IU unfractionated heparin (UFH). The use of IV vasodilators such as nitroglycerin or nitroprusside was withdrawn before angioplasty. All patients who met the eligibility criteria underwent protocol-specified aspiration thrombectomy. Dual-wire or low-pressure predilation were both acceptable if the aspiration catheter had difficulty accessing the culprit lesion. The choice of thrombus aspiration catheter was at the discretion of the operator, and included Zeek aspiration catheters (Zeon Medical, Inc), Diver CE aspiration catheters (Invatec), or Export aspiration catheters (Medtronic, Inc). After aspiration, all patients received IC administration of 100-300 µg nitroglycerin and were then randomly assigned (by means of computer-generated random numbers) to sIRA infusion of tirofiban (10 µg/kg) through the aspiration catheter over a period of 1 minute, or to conventional IC administration of tirofiban (10 µg/kg) through the guiding catheter directly over a period of 1 minute. All patients were administrated with 12-hour IV infusion of tirofiban (0.1 µg/kg/min) after primary PCI unless clinical situations were contraindicated. The drug used in this study was tirofiban (trade name, Xinweining, manufactured by Grand Pharma). The manufacturer had no economic or scientific involvement in the trial. Additional predilatation or postdilatation and stent implantation after tirofiban administration were at the discretion of the operator. During PCI, additional weight-adjusted UFH was administered. Routine medical therapies after PCI included aspirin, clopidogrel, beta-blockers, statins, and angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, according to current international guidelines.

Endpoints and definitions. The primary endpoint was the corrected TIMI frame count (CTFC) after the procedure, which was obtained by counting the number of cineframes required for dye to reach standardized distal landmarks. In the first frame used for TIMI frame counting, a column of dye touches both borders of the coronary artery and moves forward. In the last frame, dye begins to enter (but does not necessarily fill) a standard distal landmark in the artery. These standard distal landmarks are as follows: in the right coronary artery (RCA), the first branch of the posterolateral artery; in the circumflex system, the most distal branch of the obtuse marginal branch, which includes the culprit lesion in the dye path; and in the left anterior descending (LAD), the distal bifurcation, which is also known as the “moustache,” “pitchfork,” or “whale’s tail.” These frame counts are corrected for the longer length of the LAD by dividing by 1.7 to arrive at the CTFC. The secondary endpoints included major adverse cardiac events (a composite of cardiac death, reinfarction, or target vessel revascularization) at 30 days and 6 months, the incidence of complete ST-segment resolution (STR), and enzymatic infarct size. The safety endpoint was a combination of major or minor in-hospital bleeding, defined according to the TIMI bleeding classification.

Angiographic analysis. Coronary angiograms were analyzed by two interventional cardiologists who were blinded to each other. 

Electrocardiographic analysis. A routine 18-lead ECG was recorded immediately at the time of presentation and at 90 minutes after primary PCI. The magnitude of ST-segment elevation was measured 20 ms from the J point. The STR was assessed by comparing the ST-segment deviation indicated to the infarct-related area after PCI with the deviation at presentation. An STR >70% was defined as “STR,” STR 30%-70% was defined as “partial STR,” and STR <30% was defined as “no STR.” ECG recordings were analyzed by two physicians blinded to clinical data.

Enzymatic infarct size. Infarct size was estimated by serial measurements of cardiac markers, including creatine kinase, creatine kinase-MB, and cardiac troponin I. Blood samples were drawn at baseline and at 6, 12, 18, 24, and 48 hours after PCI. Based on these data, we analyzed peak enzyme release, time to peak release, and the cumulative area under the curve over the first 48 hours (AUC48) of cardiac markers. 

Follow-up. Clinical follow-up was obtained from the hospital records and telephone interviews with the patients and/or their relatives. Mortality was considered cardiac unless an unequivocal non-cardiac cause of death was established. Reinfarction was defined as new elevation of the levels of cardiac markers and/or recurrent symptoms suggestive of ischemia with new ST-segment elevation. If the cardiac troponin concentration was elevated, but stable or decreasing at the time of suspected reinfarction, the diagnosis of reinfarction required a 20% or greater increase of the cardiac troponin value in the second sample.⁷Target vessel revascularization (TVR) was defined as ischemia-driven revascularization of the IRA with PCI or coronary artery bypass grafting.⁸ Clinical events were analyzed by two physicians blinded to clinical data.

Statistical analysis. According to our pilot study, a total of 176 patients were required to achieve 80% power at a 5% significance level (2-sided). For this reason, a minimum of 180 patients (90 patients in each group) can meet the requirement of statistical analysis in this study.

Each categorical variable was expressed as the number and percentage of patients. Continuous variables were expressed as mean ± standard deviation (SD) or median with interquartile range (IQR) for data with normal distribution or non-normal distribution, respectively. Differences between treatment groups were assessed by Fisher’s exact or the Chi-square test for categorical variables and by the Student’s t-test for continuous data with normal distribution or by Mann-Whitney U-test for the medians. Otherwise, the non-parametric Wilcoxon rank-sum test was used. The Kaplan-Meier method was applied for the combined secondary clinical endpoint, and differences were assessed by the Log-rank test. All statistical tests were performed with SPSS software version 16.0 (SPSS, Inc). A 2-tailed P<.05 was considered statistically significant.

Results

Baseline characteristics. There were no significant differences between the sIRA infusion group (n = 107) and the IC administration group (n = 96) in baseline characteristics (Table 1). Also, there was no statistical difference between sIRA group and IC group in the dosage of nitroglycerin administered after thrombus aspiration.

Angiographic endpoint. After aspiration thrombectomy and tirofiban infusion, no patients showed TIMI 0 flow in the two groups, but TIMI grade 1 or 2 was identified in 14 patients (13.1%) in the IC administration group and 9 patients (9.4%) in the sIRA infusion group. CTFCs were 28 ± 17 and 27 ± 15 in the sIRA and IC groups, respectively (P=.841).

Electrocardiographic endpoint. Complete STR was achieved in 47% of the sIRA infusion patients and 43% of the IC administration patients (P=.665). The incidence of no STR in the IC administration patients had a tendency to be higher than in the sIRA infusion patients (34% vs 19%, respectively; P=.052) (Figure 2).

Enzymatic infarct size endpoint. AUC48-assessed infarct size was not statistically different between the sIRA infusion and the IC administration groups: 8.1 ×10⁴ U/L/h (IQR, 4.1-11.1 ×10⁴ U/L/h) vs 7.6 ×10⁴ U/L/h (IQR, 5.2-13.2 ×10⁴ U/L/h) for creatine kinase (P=.575); 5.6 ×10³ ng/mL/h (IQR, 3.7-8.9 ×10³ ng/mL/h) vs 5.8 ×10³ ng/mL/h (IQR, 4.0-10.9 ×10³ ng/mL/h) for creatine kinase-MB (P=.473); and 2.8 ×10³ ng/mL/h (IQR, 1.6-4.7 ×10³ ng/mL/h) vs 2.8 ×10³ ng/mL/h (IQR, 1.3-5.4 ×10³ ng/mL/h) for cardiac troponin I (P=.891). There were no statistically significant differences between the two groups in peak enzyme release and time to peak release (Table 2).

Clinical follow-up. In the IC administration group, a total of 94 patients underwent 30-day and 6-month follow-up. In the sIRA infusion group, 30-day and 6-month follow-up were available in 105 patients and 104 patients, respectively. MACEs were observed in 7 patients (6.5%) in the sIRA infusion group and in 5 patients (5.2%) in the IC administration group at 30 days. The hazard ratio (HR) was 1.27 (95% confidence interval [CI], 0.39-4.16; P=.688). The percentages of cardiac death, reinfarction, and TVR were 0.9%, 4.7%, and 2.8% in the sIRA infusion group, and 1.0%, 3.1%, and 3.1% in the IC administration group, respectively. There were no significant differences between these two groups. 

The incidence of in-hospital major and minor bleeding was low and similar between the IC administration group and the sIRA infusion group (for major bleeding, 1.0% vs 4.7% [P=.164]; for minor bleeding, 7.3% vs 4.7% [P=.433]).

At 6 months, the incidence of MACE was also comparable between these two groups (7.3% vs 8.4%; HR, 0.86; 95% CI, 0.32-2.31; P=.762) (Table 3 and Figure 3).

Primary, secondary, and safety endpoints when stratified by coronary vessel involved. A total of 117 patients had IRAs indicated as left coronary arteries: 105 were LAD arteries and 12 were left circumflex (LCX) arteries. No significant differences between the IC administration group and the sIRA infusion group were observed in baseline characteristics. CTFCs were 25 ± 11 and 30 ± 16 in the IC and sIRA groups, respectively (P=.148). The incidences of MACE and in-hospital bleeding at 30 days and 6 months were comparable between these two groups (Table 4).

Among 86 patients whose IRAs were right coronary arteries, a total of 45 patients were assigned to the IC group and the other 41 patients were assigned to the sIRA group. Baseline characteristics were balanced between the IC administration group and the sIRA infusion group. CTFCs were 30 ± 18 (IC) and 25 ± 18 (sIRA); P=.335. In addition, there was no statistical difference in the incidences of MACE and in-hospital bleeding at 30 days and 6 months (Table 5).

Discussion

To date, there are no randomized clinical trials comparing the safety and efficacy of sIRA infusion of GPI through an aspiration catheter after manual thrombus aspiration in STEMI patients undergoing primary PCI. The SUIT-AMI trial is the first prospective randomized clinical trial to investigate the differences between sIRA infusion and conventional IC administration of tirofiban in the setting of primary angioplasty. Our study indicates that sIRA infusion of tirofiban during primary PCI does not further improve the myocardial reperfusion or clinical outcomes, and the bleeding complication rates between these two groups are comparable.

No-reflow phenomenon, defined as primary PCI that achieves epicardial coronary artery reperfusion but not myocardial reperfusion, occurs in 5%-50% of STEMI patients, and is associated with a worse prognosis during follow-up.⁹ Distal embolization, ischemia-related injury, and reperfusion-related injury are the principal mechanisms of the no-reflow phenomenon.¹⁰ Many therapeutic strategies have been used to treat and prevent no-reflow. Of note, manual thrombus aspiration before stent implantation prevents distal embolization and has been recently shown to improve myocardial reperfusion and clinical outcomes as compared with the standard procedure.¹¹ However, approximately 17% of patients could not achieve myocardial blush grade 2 to 3, which indicates normal myocardial reperfusion, immediately after procedure.⁹ Recently, a pilot study demonstrated that the concentration of pro-inflammatory interleukin-1β distal to the IRA is significantly higher than in the system plasma.¹² It is conceivable that aspiration may not clear away the inflammatory materials, which explains the occurrence of no-reflow despite the use of manual thrombus aspiration. 

Up to one-third of patients have inadequate platelet inhibition, leading to an increased risk of events.^13,14 The intensification of platelet inhibition, including infusion of tirofiban, in poor responders to aspirin, clopidogrel, or both who undergo PCI decreased the rate of periprocedural MI and resulted in a lower rate of MACE at 30 days.¹⁵ Moreover, recent experimental studies have suggested that tirofiban can exert additional anti-inflammatory effects and improve endothelial function when local drug concentration is higher over antiplatelet effect.¹⁶ However, the traditional IV administration of GPI, associated with a lower drug concentration, fails to reach the optimal >80% occupancy of glycoprotein IIb/IIIa receptors, which can induce platelet aggregation <20% of baseline. Compared with IV administration, some randomized clinical trials demonstrate that IC administration of GPI in patients with STEMI undergoing primary PCI is associated with decreased median infarct size.^17,18 However, in most studies, GPI is administrated through the guiding catheter; therefore, the drug often refluxes into the aorta or flows to other areas of the vascular bed rather than the vascular bed of interest. Besides that, slow-flow or no-reflow further reduces drug delivery to the site of interest when administrated through the guiding catheter. Thus, infusion through the lumen of an over-the-wire balloon, an aspiration catheter, or a specialized catheter for local drug delivery may be preferred.¹⁹ The COCTAIL (ClearwayRx System to reduce IC thrombus in patients with acute coronary syndromes according to Optical Coherence Tomography after Abciximab Intracoronary Local infusion) study demonstrated that local IC delivery of abciximab through a dedicated perfusion catheter can reduce thrombus burden, assessed by optical coherence tomography (OCT), and significantly decrease the 1-year MACE incidence.⁵ 

In this study, we failed to demonstrate the benefits of sIRA infusion over conventional IC administration of tirofiban after thrombus aspiration in STEMI patients undergoing primary PCI. Several factors help elucidate our results. First, the patients in our study are at higher risk than those in the COCTAIL study, which enrolled only 38% STEMI patients, leading to the more intensive antithrombotic therapy used in our study. Also, it is unknown whether more intensive antithrombotic therapy may attenuate the benefit of super-selective delivery of GPI in STEMI patients. Second, the bolus dosage of tirofiban in our study (10 µg/kg) seems to be lower than the guideline-recommended dosage, which could result in suboptimal platelet inhibition.²⁰ The On-TIME 2 study demonstrated that 25 µg/kg tirofiban IV bolus could significantly improve the myocardial reperfusion, without increasing bleeding complications.²¹ Third, studies have confirmed that cardiac magnetic resonance (CMR) is a powerful predictor of no-reflow. The INFUSE-AMI study²² compared bolus IC abciximab delivered locally at the infarct lesion site to no abciximab; infarct size at 30 days assessed by CMR was significantly reduced by IC abciximab bolus. Only 27 patients in our study underwent CMR, which precluded further CMR analysis. Fourth, compared to the LAD or LCX, there may be much less difference between sIRA and conventional IC administration of GPI in the RCA. In this study, coronary angiography demonstrated the RCA to be the IRA in 42.4% of patients. The high percentage of RCA as the IRA may influence the final result.

However, effects of adjunctive mechanical (manual thrombus aspiration) and pharmacological therapies (GPI) during primary angioplasty are unclear. Further investigation is required to determine whether high-dose tirofiban bolus may achieve a better outcome.

Conclusion

In STEMI patients undergoing primary PCI with thrombus aspiration, sIRA infusion of tirofiban is not superior to conventional IC administration in improving myocardial reperfusion and short-term ischemic and hemorrhagic events. Larger randomized multicenter trials are required to evaluate whether sIRA infusion of GPI can improve clinical outcomes and reduce clinical adverse events.

References

1. Nielsen PH, Maeng M, Busk M, et al. Primary angioplasty versus fibrinolysis in acute myocardial infarction: long-term follow-up in the Danish acute myocardial infarction 2 trial. Circulation. 2010;121(13):1484-1491.
2. Niccoli G, Burzotta F, Galiuto L, Crea F. Myocardial no-reflow in humans. J Am Coll Cardiol. 2009;54(4):281-292.
3. Noman A, Egred M, Bagnall A, et al. Impact of thrombus aspiration during primary percutaneous coronary intervention on mortality in ST-segment elevation myocardial infarction. Eur Heart J. 2012;33(24):3054-3061.
4. Gurbel PA, Tantry US. Delivery of glycoprotein IIb/IIIa inhibitor therapy for percutaneous coronary intervention: why not take the intracoronary highway? Circulation. 2010;121(6):739-741.
5. Prati F, Capodanno D, Pawlowski T, et al. Local delivery versus intracoronary infusion of abciximab in patients with acute coronary syndromes. JACC Cardiovasc Interv. 2010;3(9):928-934.
6. Yan HB, Li SY, Song L, et al. Thrombus aspiration plus intra-infarct-related artery administration of tirofiban improves myocardial perfusion during primary angioplasty for acute myocardial infarction. Chin Med J (Engl). 2010;123(7):877-883.
7. Thygesen K, Alpert JS, Jaffe AS, et al. Third universal definition of myocardial infarction. Circulation. 2012;126(16):2020-2035.
8. Cutlip DE, Windecker S, Mehran R, et al. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation. 2007;115(17):2344-2351.
9. Svilaas T, Vlaar PJ, van der Horst IC, et al. Thrombus aspiration during primary percutaneous coronary intervention. N Engl J Med. 2008;358(6):557-567.
10. Jaffe R, Dick A, Strauss BH. Prevention and treatment of microvascular obstruction-related myocardial injury and coronary no-reflow following percutaneous coronary intervention: a systematic approach. JACC Cardiovasc Interv. 2010;3(7):695-704.
11. De Luca G, Dudek D, Sardella G, et al. Adjunctive manual thrombectomy improves myocardial perfusion and mortality in patients undergoing primary percutaneous coronary intervention for ST-elevation myocardial infarction: a meta-analysis of randomized trials. Eur Heart J. 2008;29(24):3002-3010.
12. Yan HB, Li WZ, Zhao HJ, et al. Clinical implication and association between local and systemic levels of interleukin-1beta and interleukin-10 in patients with coronary artery disease. Zhonghua Xin Xue Guan Bing Za Zhi. 2011;39(2):142-146.
13. Breet NJ, van Werkum JW, Bouman HJ, et al. Comparison of platelet function tests in predicting clinical outcome in patients undergoing coronary stent implantation. JAMA. 2010;303(8):754-762.
14. Collet JP, Cuisset T, Range G, et al. Bedside monitoring to adjust antiplatelet therapy for coronary stenting. N Engl J Med. 2012;367(22):2100-2109.
15. Valgimigli M, Campo G, de Cesare N, et al. Intensifying platelet inhibition with tirofiban in poor responders to aspirin, clopidogrel, or both agents undergoing elective coronary intervention: results from the double-blind, prospective, randomized Tailoring Treatment with Tirofiban in Patients Showing Resistance to Aspirin and/or Resistance to Clopidogrel study. Circulation. 2009;119(25):3215-3222.
16. Nannizzi-Alaimo L, Alves VL, Phillips DR. Inhibitory effects of glycoprotein IIb/IIIa antagonists and aspirin on the release of soluble CD40 ligand during platelet stimulation. Circulation. 2003;107(8):1123-1128.
17. Gu YL, Kampinga MA, Wieringa WG, et al. Intracoronary versus intravenous administration of abciximab in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention with thrombus aspiration: the comparison of intracoronary versus intravenous abciximab administration during emergency reperfusion of ST-segment elevation myocardial infarction (CICERO) trial. Circulation. 2010;122(25):2709-2717.
18. Thiele H, Schindler K, Friedenberger J, et al. Intracoronary compared with intravenous bolus abciximab application in patients with ST-elevation myocardial infarction undergoing primary percutaneous coronary intervention: the randomized Leipzig immediate percutaneous coronary intervention abciximab IV versus IC in ST-elevation myocardial infarction trial. Circulation. 2008;118(1):49-57.
19. Berger PB, Best PJ. Intracoronary glycoprotein IIb/IIIa inhibitors from questioning the logic to weighing the data. JACC Cardiovasc Interv. 2010;3(9):935-936.
20. Steg PG, James SK, Atar D, et al. ESC guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J. 2012;33(20):2569-2619.
21. Van’t Hof AW, Ten Berg J, Heestermans T, et al. Prehospital initiation of tirofiban in patients with ST-elevation myocardial infarction undergoing primary angioplasty (On-TIME 2): a multicentre, double-blind, randomised controlled trial. Lancet. 2008;372(9638):537-546.
22. Stone GW, Maehara A, Witzenbichler B, et al. Intracoronary abciximab and aspiration thrombectomy in patients with large anterior myocardial infarction: the INFUSE-AMI randomized trial. JAMA. 2012;307(17):1817-1826.