Tirofiban and Sirolimus-Eluting Stent vs Abciximab and Bare-Metal Stent for Acute Myocardial Infarction: Title and subTitle BreakA Randomized Trial FREE

Bare-metal stenting, by decreasing the rate for target-vessel revascularization (TVR),^{1 - 2} has been endorsed as a class I indication in the treatment of acute ST-segment elevation myocardial infarction (STEMI).³ Abciximab, by virtue of its salutary effect on tissue perfusion and coronary artery patency, has also been recommended as part of a reasonable treatment strategy.^{3 - 4}

Sirolimus-eluting stents greatly reduce the need for TVR compared with bare-metal stents and thus have the potential to further improve long-term clinical outcome after primary percutaneous coronary intervention (PCI).⁵ However, the lack of randomized trials to assess the safety and long-term efficacy of sirolimus-eluting stent implantation in patients with STEMI, in conjunction with the financial consequences of replacing bare-metal stenting with sirolimus-eluting stenting, currently limit use of sirolimus-eluting stents in this setting.⁶ Furthermore, in the first tertiary referral center to adopt unrestricted use of sirolimus-eluting stents, a concomitant decrease in the use of abciximab has been reported.⁷ This may indicate a reluctance to use glycoprotein (Gp) IIb/IIIa inhibitors and drug-eluting stents simultaneously for economic reasons.

Current clinical guidelines specifically recommend abciximab during primary PCI,³ and recent additional evidence strongly supports the value of Gp IIb/IIIa inhibitors in this setting.⁸ Replacing abciximab with tirofiban, administered as a single high-dose bolus (SHDB) regimen, is a promising strategy that could preserve financial resources. At current European market prices (approximately 580 € [$742] for tirofiban vs approximately 1900 € [$2432] for abciximab), this would absorb the additional cost of sirolimus-eluting stenting vs bare-metal stenting (European list price: approximately 1800 € [$2304] for sirolimus-eluting vs approximately 600 € [$768] for bare-metal stenting).

Thus, in an attempt to further improve outcome after myocardial infarction (MI) without affecting overall costs, we compared a strategy of tirofiban-supported infarct artery sirolimus-eluting stent implantation with a current preferred strategy for STEMI treatment in terms of angiographic and clinical outcome.⁴

The Single High Dose Bolus Tirofiban and Sirolimus Eluting Stent vs Abciximab and Bare Metal Stent in Myocardial Infarction (STRATEGY) trial was a prospective, single-center, randomized trial of 2 strategies: SHDB tirofiban plus sirolimus-eluting stent implantation vs standard abciximab regimen plus bare-metal stent implantation, in patients undergoing primary angioplasty for acute STEMI. The design of the study has been described in detail.⁹

Individuals eligible for enrollment were consecutive patients presenting with STEMI who fulfilled the following inclusion criteria: (1) chest pain for more than 30 minutes with ST-segment elevation of 1 mm or more in 2 or more contiguous electrocardiograph leads or with presumably new left bundle-branch block and (2) admission either within 12 hours of symptom onset or between 12 and 24 hours with evidence of continuing ischemia. Exclusion criteria included administration of fibrinolytic agents in the previous 30 days, history of bleeding diathesis or allergy to the study drugs, major surgery within 15 days, and active bleeding or previous stroke in the last 6 months. The University of Ferrara institutional review board approved the protocol in January 2003, and all patients gave written informed consent.

Open-label randomization was performed in the coronary care unit by the treating physician immediately after eligibility criteria were met. A 1:1 computer-generated random sequence supplied by an academic statistician, without blocking or stratification, was used. Sealed envelopes indicated the treatment group to which the patients were assigned: SHDB tirofiban (Aggrastat; Merck, West Point, Pa [Europe]/Guilford Pharmaceuticals, Baltimore, Md [United States]) or abciximab (ReoPro; Centocor, Malvern, Pa). The study drug was started in the coronary care unit and subsequently patients were transferred to the cardiac catheterization laboratory. Per protocol, patients allocated to the tirofiban group received sirolimus-eluting stenting (Cypher; Cordis Corp, Miami Lakes, Fla). Crossover to bare-metal stenting was allowed only when sirolimus-eluting stent implantation failed or when it was impossible to match sirolimus-eluting stent diameter with coronary reference diameter (sirolimus-eluting stents were only available in diameters from 2.25-3.00 mm during the enrollment period). Patients allocated to the abciximab group received bare-metal stenting: the first choice was a Sonic stent (Cordis Corp), but any commercially available bare-metal stent could be implanted.

Tirofiban was given as a bolus of 25 μg/kg over 3 minutes, followed by an infusion of 0.15 μg/kg per minute for 18 to 24 hours. Abciximab was administered as a bolus of 0.25 mg/kg over 3 minutes, followed by a 12-hour infusion of 0.125 μg/kg per minute. All patients received aspirin (160-325 mg orally as a loading dose and then 80-125 mg/d orally indefinitely) and clopidogrel (300 mg orally as a loading dose and then 75 mg/d for at least 3 months). Heparin (50 U/kg) was administered before the procedure, with additional boluses administered to achieve and maintain an activated clotting time of at least 200 seconds.

Cumulative ST-segment elevation, evaluated in all leads with any ST-segment elevation of 1 mm or greater, was measured to the nearest 0.5 mm at 60 milliseconds after the J point.

In a subset of 70 randomly selected patients, a platelet aggregation assay was conducted before and 10 minutes after the bolus dose of Gp IIb/IIIa inhibitors using a platelet function analyzer (PFA-100; Dade Behring Inc, Deerfield, Ill), as previously described.¹⁰ Closure time, ie, the time necessary to form a hemostatic platelet plug and to stop blood flow when the sample is aspirated at high shear rate (5000-6000 seconds⁻¹) through a collagen- and adenosine diphosphate–coated capillary (147 μm diameter), inversely reflects platelet reactivity in the sample evaluated.

Thrombolysis in Myocardial Infarction (TIMI) grade 3 coronary flow in the treated vessel and a residual stenosis less than 30% were the criteria used to define a successful PCI. Quantitative angiographic analyses, using a validated edge-detection system (CAAS II; Pie Medical, Maastricht, the Netherlands), were performed by an experienced cardiologist who was unaware of treatment assignment. Acute luminal gain was defined as the minimal luminal diameter (MLD) immediately after the procedure minus the MLD at baseline. Late loss was defined as the MLD immediately after the procedure minus the MLD at follow-up. The target lesion was defined as the stented segment plus the 5-mm segments immediately proximal and distal to the stent(s).

The primary end point was freedom, at 8 months after randomization, from death, nonfatal MI, stroke, and binary restenosis. The data for all primary end point events were reviewed by an independent adjudication committee whose members were unaware of treatment assignment. Nonfatal MI was defined as recurrent chest pain with ST-segment or T-wave changes and recurrent elevation of cardiac enzyme levels. Patients were considered eligible for angiographic follow-up if protocol-mandated PCI had been attempted. Binary restenosis, dichotomized for in-stent and in-segment (5 mm proximal and distal to the stent margins) was defined as a greater than 50% diameter stenosis demonstrated by quantitative coronary angiography.

Secondary end points included freedom, at day 30 and month 8, from major cardiac or cerebrovascular adverse events defined as the composite of death, reinfarction, stroke, and repeat TVR. Urgent TVR was defined as repeat PCI or coronary artery bypass graft surgery performed within 24 hours of severe recurrent ischemic symptoms. As a safety analysis, the end points of major or minor bleeding along with severe or mild thrombocytopenia were defined according to the criteria of the TIMI trial.¹¹

Sample size was calculated on the assumption that the incidence of death, reinfarction, stroke, and binary restenosis at 8 months would be 13% in the tirofiban plus sirolimus-eluting stent group and 30% in the abciximab plus bare-metal stent group. To detect this effect size with 80% power and a type I error (α) of .05, taking into account an expected 10% rate of dropouts, required 160 patients.

All analyses were conducted according to the intention-to-treat principle. Discrete data were summarized as frequencies, whereas continuous data were expressed as median and interquartile range (IQR). The Fisher exact test (categorical variables) and Mann-Whitney test (continuous variables) were used to analyze differences between the 2 study groups. Event-free survival curves were generated by the Kaplan-Meier method, and survival between groups was compared using the log-rank test. Hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated using a Cox proportional hazards model. The proportional hazards assumption was confirmed by an academic statistician (E.B.). A 2-sided P value less than .05 was considered significant for all tests. All analyses were performed using STATISTICA version 6.1 (StatSoft Inc, Tulsa, Okla).

Between March 6, 2003, and April 23, 2004, 175 patients with STEMI were randomized. A trial flow diagram is shown in Figure 1. Baseline demographics and angiographic features (Table 1 and Table 2) were well matched between the 2 randomized cohorts. A single patient in either group had a time from symptom onset to intervention of greater than 12 hours. Median time from Gp IIb/IIIa inhibitor bolus to intervention was 33 (IQR, 22-42) minutes in the tirofiban plus sirolimus-eluting stent group vs 35 (IQR, 23-41) minutes in the abciximab plus bare-metal stent group (P = .73). Total median duration of study drug infusion was 20 (IQR, 14-28) hours in the tirofiban plus sirolimus-eluting stent group vs 12 (IQR, 11-13) hours in the abciximab plus bare-metal stent group (P<.001).

Based on initial angiographic findings, 3 patients (3%) in the tirofiban plus sirolimus-eluting stent group and 5 patients (6%) in the abciximab plus bare-metal stent group did not receive PCI (Figure 1). In 5 patients (2 tirofiban plus sirolimus-eluting stent, 3 abciximab plus bare-metal stent), no significant lesion was present in the culprit vessel on angiography, and anterograde flow was normal. In 2 patients (1 per group) the small size of the occluded vessel precluded intervention. Finally, 1 patient in the abciximab plus bare-metal stent group was found to have a type I aortic dissection with occlusion of the right coronary artery. Where angioplasty was attempted, procedural success did not differ significantly between the tirofiban plus sirolimus-eluting stent group (82/84 patients [98%]) and the abciximab plus bare-metal stent (79/83 patients [95%]) group (P = .91). Nominal stent diameter was smaller in the tirofiban plus sirolimus-eluting stent group, reflecting the limited available range of sirolimus-eluting stent diameters.

Overall, 74 patients (85%) in the tirofiban plus sirolimus-eluting stent group and 77 (88%) in the abciximab plus bare-metal stent group received the protocol-mandated treatment combination (Figure 1). The extent of ST-segment resolution (Table 2) and the TIMI flow grade postintervention (Table 2) did not differ between treatment groups, nor did quantitative angiographic parameters at baseline or immediately after intervention (Table 3).

Median platelet aggregation, evaluated in 70 patients as closure time, was similar at baseline in the tirofiban plus sirolimus-eluting stent group (77 [IQR, 68-86] seconds) and the abciximab plus bare-metal stent (74 [IQR, 69-83] seconds) group (P = .71). Ten minutes after the start of drug infusion, 30 of 35 patients (86%) receiving abciximab vs 31 of 35 (89%) treated with tirofiban reached full platelet inhibition (closure time, >300 seconds). In the remaining 9 patients, closure time was 298 (IQR, 297-299) seconds in the 4 who received tirofiban and 295 (IQR, 292-296) seconds in the 5 who received abciximab (P = .07).

There were no significant differences between the tirofiban plus sirolimus-eluting stent group and the abciximab plus bare-metal stent group in the incidence of major adverse cardiac or cerebrovascular events during the first 30 days (Table 4). Five fatal events occurred: 2 (1 due to left ventricular free wall rupture, another to cardiogenic shock) in the tirofiban plus sirolimus-eluting stent group and 3 (as a consequence of aortic dissection, cardiogenic shock, and left ventricular free wall rupture) in the abciximab plus bare-metal stent group. Among patients receiving abciximab, 2 had reinfarction due to angiographically confirmed stent thrombosis requiring urgent TVR. In 1 patient with diabetes, a sirolimus-eluting stent had been implanted as a protocol violation. In the tirofiban plus sirolimus-eluting stent group, 1 patient with diffuse 3-vessel disease and cardiogenic shock at entry satisfied the criteria for reinfarction while mechanically ventilated 3 days after the index procedure. Coronary angiography performed thereafter excluded new changes in coronary anatomy. Finally, in 1 patient per group, urgent TVR was required due to residual infarct-related artery dissection threatening vessel closure. No cerebrovascular accident occurred. The incidence of any thrombocytopenia was significantly lower in the tirofiban plus sirolimus-eluting stent group vs the abciximab plus bare-metal stent group (Table 4).

Complete follow-up information up to 240 days was available for all patients. Eight patients in the tirofiban plus sirolimus-eluting stent group and 8 in the abciximab plus bare-metal stent group refused to undergo angiographic follow-up performed after a median of 209 days (range, 148-267). In 1 additional patient per group, follow-up coronary angiography was not performed due to a severe allergic reaction to contrast media during the index procedure and transfusion-dependent chronic anemia. None of these patients experienced major adverse cardiac or cerebrovascular events; all were asymptomatic and none had inducible myocardial ischemia on stress testing. Their baseline characteristics did not differ with respect to patients who underwent follow-up angiography.

Cumulatively, 14 of 74 (19%; 95% CI, 10%-28%) patients in the tirofiban plus sirolimus-eluting stent group and 37 of 74 (50%; 95% CI, 44%-56%) in the abciximab plus bare-metal stent group fulfilled the criteria for the 8-month primary end point (death, MI, stroke, or binary restenosis) (HR, 0.33; 95% CI, 0.18-0.60; P<.001 [P<.001 by Fischer exact test]).

The cumulative incidence of major adverse cardiac or cerebrovascular events (death, reinfarction, stroke, or TVR) was significantly lower in the tirofiban plus sirolimus-eluting stent group vs the abciximab plus bare-metal stent group (18% vs 32%, respectively; HR, 0.53; 95% CI, 0.28-0.92; P = .04) (Figure 2). Overall, the composite of death or reinfarction was similar in the tirofiban plus sirolimus-eluting stent group (13%) vs the abciximab plus bare-metal stent group (17%) (HR, 0.71; 95% CI, 0.34-1.5; P = .39) (Figure 2), while there was a significant reduction in the need for TVR (7% vs 20%, respectively; HR, 0.30; 95% CI, 0.12-0.77; P = .01) in the tirofiban plus sirolimus-eluting stent group.

The results of quantitative coronary analyses are shown in Table 3. Cumulative distribution curves (paired-lesion analysis) for the difference of MLD before and immediately after the intervention and at follow-up in the 2 groups are shown in Figure 3. In-lesion binary restenosis—assessed in 67 and 66 patients in the tirofiban plus sirolimus-eluting stent and abciximab plus bare-metal stent groups (87% and 88% of those eligible, respectively)—was present in 6 (9%; 95% CI, 2%-16%) and 24 (36%; 95% CI, 26%-46%) patients, respectively (P = .002).

In the United States, the Food and Drug Administration approved the first drug-eluting stent (the sirolimus-eluting stent) in April 2003. Since then, penetration of drug-eluting stenting has rapidly increased, replacing bare-metal stenting in more than 80% of PCI procedures.¹² This rapid transition has been facilitated by the presence of specific reimbursement for drug-eluting stenting. However, evidence for the efficacy and safety of drug-eluting stents has been obtained only for highly selected patient or lesion subgroups,^{13 - 15} and concerns regarding the safety and economic consequences of liberal use of these devices have also been raised.^{12 ,15 - 20}

In Europe, sirolimus-eluting stents became available 1 year previous (April 2002). Nevertheless, overall use of drug-eluting stents remains limited, with penetration rates as low as 6% in some countries.⁶ This may be related to their high cost in conjunction with the almost universal lack of specific reimbursement. Thus, their short-term costs are currently playing a major role in clinical decision making.

To the best of our knowledge, no randomized trial has specifically examined the safety and efficacy of drug-eluting stents in the setting of primary intervention for STEMI. Abciximab and drug-eluting stent implantation have potentially complementary effects on death or MI and the need for reintervention in the setting of acute MI,^{5 ,8 ,21 - 23} providing the rationale for testing their combination in this setting.²⁴ However, both therapies are expensive. The average cost of combining abciximab treatment with a single sirolimus-eluting stent exceeds 50% of current reimbursement for acute MI in some European countries.²⁵ Thus, financial pressure may lead to competitive use between drug-eluting stents and abciximab on the basis of their high cost.⁷

Our results demonstrate that a lower-cost strategy of tirofiban-supported sirolimus-eluting stent implantation during primary PCI was safe and resulted in improved clinical (ie, decreased need for TVR) and angiographic (ie, decreased restenosis) outcomes with respect to a strategy of abciximab-supported bare-metal stent implantation. Moreover, such a strategy was associated with an improved safety profile with respect to thrombocytopenia.

Two specific aspects of our study protocol deserve emphasis. First, all consecutive patients with acute MI referred to a single referral center for intended primary intervention were prospectively enrolled provided they had no contraindication to the use of Gp IIb/IIIa inhibitors, thus substantially reducing any potential bias related to clinical presentation or angiographic findings. Second, Gp IIb/IIIa inhibitors were started at admission at a median interval of more than 30 minutes before coronary intervention. In a recent trial of patients with STEMI referred for intended bare-metal stenting, a similar approach to abciximab administration showed a benefit on total and cardiovascular mortality.⁸ Furthermore, a recent meta-analysis supports the hypothesis that early administration of Gp IIb/IIIa inhibitors to patients with STEMI is associated with consistent trends toward a reduction in mortality and other major adverse cardiac events.²⁶

In a previously reported head-to-head comparison between abciximab and tirofiban in patients undergoing PCI, abciximab was superior to tirofiban with respect to the prespecified combined end point.²⁷ The superiority of abciximab was driven by a higher rate of periprocedural MI in the tirofiban group, suggesting inadequate early platelet inhibition with the bolus regimen used (10 μg/kg). Subsequent dose-ranging studies showed that increasing the tirofiban bolus dose from 10 to 25 μg/kg was necessary to obtain an optimal level of platelet inhibition,²⁸ and initial clinical trials with the high bolus appear promising.^{29 - 30} In the present study, evaluation of platelet function before and soon after administration of the Gp IIb/IIIa inhibitor showed that in the primary PCI setting, SHDB tirofiban resulted in a degree of early platelet inhibition similar to that observed with abciximab. Similarly, TIMI flow patterns and ST-segment resolution—both surrogates of long-term mortality that have been consistently improved by abciximab treatment even in relatively small studies^{21 - 22} —did not differ between treatment groups. A large-scale noninferiority trial comparing a SHDB tirofiban regimen with abciximab is currently enrolling patients. The results of this trial will complement and extend our present findings.

Our study has several important limitations. First, it was underpowered to assess the effect of tirofiban-supported sirolimus-eluting stent implantation on the rate of major adverse cardiac or cerebrovascular events, implying that the benefit of treatment observed here was greater than expected but also that larger studies are needed to confirm our preliminary observations. Second, in keeping with the design of our investigation, no conclusion can be drawn regarding the relative contribution of a specific Gp IIb/IIIa inhibitor or stent type with respect to the other. This would have required a factorial study design. Third, despite careful avoidance of the “oculo-stenotic” reflex, we cannot rule out the possibility that protocol-mandated angiographic follow-up has not increased the magnitude of clinical benefit of study treatment. Finally, a formal cost-effectiveness analysis, which would have complemented our clinical and angiographic results, was not performed in this trial.

In conclusion, our study provides proof of concept for a new treatment strategy in STEMI that incorporates unrestricted use of sirolimus-eluting stenting but results in no (European market) or only a modest (US market) increase in medical expenditure.