Author + information
- Received June 19, 2007
- Revision received August 24, 2007
- Accepted September 23, 2007
- Published online February 5, 2008.
- Hitinder S. Gurm, MD⁎,1,
- Dean E. Smith, MS, PhD⁎,
- J. Stewart Collins, MD⁎,
- David Share, MD, MPH†,
- Arthur Riba, MD‡,
- Andrew J. Carter, MD§,
- Thomas LaLonde, MD∥,
- Eva Kline-Rogers, RN, MS⁎,
- Michael O’Donnell, MD¶,
- Hameem Changezi, MD#,
- Marcel Zughaib, MD⁎⁎,
- Robert Safian, MD††,
- Mauro Moscucci, MD⁎,2,⁎ (, )
- Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2)
- ↵⁎Reprint requests and correspondence:
Dr. Mauro Moscucci, University of Michigan Health System, University Hospital, TC B1-226, 1500 East Medical Center Drive, Ann Arbor, Michigan 48103-0311.
Objectives This study sought to assess whether the use of eptifibatide instead of abciximab is associated with a difference in outcomes of patients undergoing primary percutaneous coronary intervention (PCI) for ST-segment elevation myocardial infarction (STEMI).
Background Pooled data from randomized controlled trials suggest that the use of abciximab may be associated with a survival advantage in patients undergoing primary PCI for acute STEMI. However, a large proportion of patients in the community are treated with eptifibatide, an agent that shares some but not all pharmacological properties with abciximab.
Methods We evaluated the outcomes of 3,541 patients who underwent primary PCI for STEMI from October 2002 to July 2006 in a large regional consortium and who were treated with abciximab (n = 729) or with eptifibatide (n = 2,812).
Results There was no difference in the incidence of in-hospital death (4.1% with abciximab vs. 3.5% with eptifibatide, p = 0.39), recurrent myocardial infarction (0.8% vs. 1.2%, p = 0.42), or stroke/transient ischemic attack (0.7% vs. 0.6%, p = 0.80). There was no difference in the need for blood transfusion (12.4% vs. 11.7%, p = 0.61), whereas there was a greater incidence of gastrointestinal bleeding with abciximab (4.8% vs. 2.8%, p = 0.01). In parsimonious risk-adjusted models, no significant difference between abciximab and eptifibatide was observed with respect to any of the outcomes measures.
Conclusions Currently, eptifibatide is used as the adjunct antiplatelet agent in the majority of patients undergoing primary PCI. There is no apparent difference in early outcomes of patients treated with eptifibatide compared with patients treated with abciximab.
Primary percutaneous coronary intervention (PCI) is the preferred reperfusion agent in ST-segment elevation myocardial infarction (STEMI) (1). Data from randomized controlled trials suggest that the adjunctive use of abciximab may be associated with a survival advantage in patients undergoing primary PCI (2). Although the current guidelines support the use of abciximab in patients undergoing primary PCI, there is significant variability in the choice of antiplatelet therapy in real-world clinical practice for various reasons (3). First, the magnitude of survival benefit associated with abciximab is still controversial (4). Second, eptifibatide or tirofiban, agents that share some but not all pharmacological properties with abciximab, are often less expensive and more widely available in many hospitals. Although the only direct comparison between tirofiban and abciximab suggested that abciximab may be a superior agent, that comparison did not include STEMI patients (5). Given the widespread use of small molecule glycoprotein (GP) IIb/IIIa inhibitors in PCI, it is important to evaluate how often these agents are used in patients undergoing primary PCI and whether their use is associated with any adverse clinical impact. Therefore, we evaluated the outcomes of patients undergoing primary PCI in a regional consortium to analyze the impact of different GP IIb/IIIa inhibitors on clinical outcomes of STEMI patients.
The study cohort for our analysis included patients undergoing primary PCI in a large regional registry of contemporary PCI. The details of the registry and of the data collection process have been described elsewhere (6–10). Briefly, procedural data on all patients undergoing elective and nonelective PCI at the participating hospitals are collected using standardized data collection forms. Baseline data include clinical, demographic, procedural, and angiographic characteristics as well as medications used before, during, and after the procedure, and in-hospital outcomes. All data elements have been prospectively defined, and the protocol was approved by the local institutional review board at each institution. The data were collected by a dedicated staff member and forwarded to the coordinating center. Medical records of all patients undergoing coronary artery bypass grafting (CABG) or of patients who died in the hospital were reviewed to ensure data accuracy. A further 2% of cases were randomly selected for audit.
The study population for this analysis included patients who underwent primary PCI for STEMI between October 2002 and July 2006 and who were treated with either abciximab or eptifibatide. We excluded patients treated after 12 h of symptom onset, those transferred from another institution, those undergoing rescue PCI after failed thrombolysis, and patients treated with heparin alone or tirofiban.
All procedures were performed using standard coronary intervention technique. The choice of adjunct therapy was at the discretion of the operating physician. The primary end point for this analysis was in-hospital death. Other end points evaluated included post-procedure transfusion and in-hospital major adverse cardiovascular events defined as the composite of death, myocardial infarction, stroke, all CABG, or target lesion revascularization. In-hospital death was defined as death for either cardiac or noncardiac cause; blood transfusion was defined as any transfusion of blood product regardless of the number of units transfused. Vascular complication was defined as any of the following: pseudoaneurysm, arteriovenous fistula, femoral neuropathy, retroperitoneal hematoma, hematoma at the access site requiring transfusion/prolonged hospital stay or causing a decrease in hemoglobin >3.0 g/dl, or any access site complication requiring surgical repair.
Continuous variables are expressed as mean ± standard deviation, and discrete variables are expressed as frequency counts and percentages. The differences in discrete variables between groups were evaluated by the chi-square test and Fisher exact test. Continuous variables were analyzed using the t test and Wilcoxon rank sum test as needed. Univariate and multivariate logistic regression modeling was used to calculate unadjusted and adjusted odds of periprocedural events in association with abciximab use.
To further adjust for the nonrandomized use of abciximab and for a possible selection bias in this cohort, a predictive model that adjusted for the propensity to receive abciximab was also developed (11). The probability or a propensity score of receiving abciximab was calculated using a nonparsimonious logistic regression model. The variables included in the model were age, gender, prior history of stroke, peripheral vascular disease, hypertension, diabetes, congestive heart failure, prior myocardial infarction, renal failure requiring dialysis, prior gastrointestinal bleeding, atrial fibrillation, prior revascularization, chronic obstructive airway disease, serum creatinine and hemoglobin, extent of coronary artery disease, presence of thrombus or calcification, pre-procedural medication use, left ventricular ejection fraction, and use of intra-aortic balloon pump before the intervention. The propensity score was then included as an additional explanatory variable in the final models. Furthermore, we used Greedy matching techniques to select patients treated with eptifibatide as counterparts to patients treated with abciximab by choosing the patient with the nearest propensity score (12). In-hospital outcome was then compared within this propensity-matched cohort. Random effect models were fitted to control for clustering and variation by hospital. We also calculated the observed and predicted mortality rates by the GP IIb/IIIa agent used and calculated the standardized mortality rates as previously described (9).
A total of 4,128 patients underwent primary PCI for STEMI during the study period. Patients who were treated with tirofiban (n = 4) or without a GP IIb/IIIa inhibitor (n = 583) were excluded from the analysis. Thus, the study cohort included 3,541 patients. The majority of patients (n = 2,812) were treated with eptifibatide, whereas the remainder (n = 729) were treated with abciximab. The proportion of patients treated with abciximab versus eptifibatide did not show any significant temporal trends (Fig. 1). There was wide variation in the proportion of patients treated with abciximab versus eptifibatide in the participating hospitals (Fig. 2).
Baseline demographic and clinical characteristics of the population are shown in Table 1. Patients treated with abciximab were more likely to have pre-existing renal failure requiring dialysis and to have angiographically evident thrombus. There was a slightly greater amount of contrast used in this group compared with those treated with eptifibatide, and they were more likely to undergo multivessel interventions. Patients treated with abciximab were more likely to receive pre-procedural clopidogrel and less likely to receive pre-procedural beta-blockers. There was no difference in the other baseline demographics, comorbidities, adjunct medications, or procedural variables.
Clinical outcomes before hospital discharge are shown in Table 2. There was no difference in the angiographic outcome, need for emergency bypass surgery, stroke, or in-hospital mortality. There was no difference in vascular complications and need for transfusion, although the incidence of gastrointestinal bleeding was higher among patients treated with abciximab (4.8% vs. 2.8%).
We evaluated the unadjusted mortality rate in multiple subgroups; elderly patients, women, patients in cardiogenic shock, patients with cardiac arrest before presentation, and patients with left anterior descending artery occlusion. No subgroups were identified in which 1 drug seemed to be superior to the other (data not shown).
There were no differences in any of the outcome variables after multivariate or propensity adjustment (Fig. 3). The model used for developing the propensity score had a c-statistic of 0.6. The observed mortality in patients treated with either abciximab or eptifibatide was lower than in patients treated without a GP IIb/IIIa inhibitor, and the standardized mortality rates between the 2 drug groups were remarkably similar (Fig. 4). Furthermore, in a propensity-matched cohort in which each patient treated with abciximab was matched to a similar patient treated with eptifibatide, no difference in any of the outcomes was seen (Table 3).
In our study, we evaluated the use of GP IIb/IIIa inhibitors in patients undergoing PCI for STEMI in current clinical practice. Our data suggest that eptifibatide is the most commonly used GP IIb/IIIa inhibitor, and that the choice of the GP IIb/IIIa inhibitor does not seem to influence short-term outcomes.
Although mechanical reperfusion has emerged as the therapy of choice for acute STEMI, the optimal pharmacological regimen for these patients has not been well studied. In contrast to studies of fibrinolytic agents, most trials evaluating the safety and efficacy of GP IIb/IIIa inhibitors in primary PCI for STEMI have been small (combined number of patients enrolled <5,000) and underpowered to detect a mortality advantage. Within this small body of data, the majority of studies have evaluated the role of abciximab when compared with placebo. In the meta-analysis performed by De Luca et al. (2), compared with placebo, the use of abciximab was associated with a significant reduction in 30-day mortality (2.4% vs. 3.4%, odds ratio 0.68, p = 0.047) and long-term mortality (4.4% vs. 6.2%). There was a commensurate reduction in re-infarction, whereas no difference in bleeding events was noted. Recently, Montalescot et al. (13) reported a patient-level meta-analysis evaluating the impact of abciximab on clinical outcomes of patients undergoing stenting for primary PCI. Compared with placebo, the use of abciximab was associated with a 37% reduction in the composite hazard of death or reinfarction at 3 years (hazard ratio 0.63, 95% confidence interval 0.45 to 0.88, p = 0.008).
The experience with small-molecule GP IIb/IIIa inhibitors is more limited and is restricted to smaller trials of tirofiban and 1 trial of eptifibatide. These trials were designed to compare upfront versus in-laboratory administration of GP IIb/IIIa inhibitor and suggested better angiographic outcomes with an early administration approach (13). A small angiographic study compared high-dose tirofiban with abciximab in 100 patients undergoing primary PCI. There was no difference in angiographic success, epicardial reperfusion, or left ventricular recovery between the 2 groups (14). Although these data can be used to support potentially similar efficacy of tirofiban in primary PCI, none of the trials have evaluated the impact of the therapeutic switch on hard clinical end points. Thus, given the lack of placebo-controlled trials, only indirect inferences can be drawn about the safety and efficacy of small-molecule GP IIb/IIIa inhibitors in primary PCI.
In the only large-scale randomized trial evaluating the relative efficacy of tirofiban and abciximab in a general patient population, there was a reduction in ischemic events in patients treated with abciximab that was predominantly seen in patients with acute coronary syndromes (5). However, patients with acute STEMI were excluded from this trial.
Although purists would argue that abciximab is the only agent that should be used in patients undergoing primary PCI for STEMI, market forces work against that. The wholesale acquisition cost of treating a 70-kg man with a standard dose of eptifibatide is $885.48, compared with $1,980.42 for abciximab (based on a 12-h infusion of abciximab and an 18-h infusion of eptifibatide) (15). This cost difference is the likely reason that many hospital formularies stock eptifibatide in preference to abciximab.
Our findings are thus important for several reasons. First, they provide an assessment of the current use of GP IIb/IIIa inhibitors in contemporary clinical PCI practice for STEMI. Second, we were unable to detect any difference in outcomes with the use of either agent. It can be estimated that given the 30-day outcomes of patients undergoing primary PCI, a trial comparing eptifibatide and abciximab would need to enroll over 10,000 patients to show the noninferiority of one agent when compared with the other. The lack of major differences in observational data, the cost of a trial of that size, and market realities suggest that such a trial is unlikely to be performed.
Our findings are based on observational data that are not centrally adjudicated. Although the data were analyzed using risk adjustment and propensity analysis, we cannot exclude that we were unable to adjust for other unknown factors that may influence outcome. Prior observational studies have detected differences in long-term but not in short-term outcomes of patients treated with GP IIb/IIIa inhibitors versus those treated with heparin only (16). It is possible that the enhanced microcirculatory improvement achieved with these drugs impacts long-term survival. Our study cannot detect differences in long-term outcomes because these follow-up data are not collected in the registry.
Eptifibatide is the most commonly used GP IIb/IIIa inhibitor in patients undergoing primary PCI for acute STEMI in our registry. Within the limits of the observational nature of our study, there is no evidence to suggest that the use of eptifibatide in lieu of abciximab is associated with worse short-term outcomes in this clinical setting. However, this finding and the question of whether differences in use of these agents affect longer-term outcomes should be examined in larger prospective studies.
↵1 Dr. Gurm has received a one-time honoraria from The Medicines Company and he is a consultant for CardioMems and Icon Interventional Systems.
↵2 Dr. Moscucci has received a research grant from Blue Cross Blue Shield of Michigan; is on the Speakers’ Bureau for Pfizer; is a Study Investigator for Centocor; and has received honoraria from The Medicines Company.
- Abbreviations and Acronyms
- coronary artery bypass grafting
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- Received June 19, 2007.
- Revision received August 24, 2007.
- Accepted September 23, 2007.
- American College of Cardiology Foundation
- Antman E.M.,
- Anbe D.T.,
- Armstrong P.W.,
- et al.
- Moscucci M.,
- Kline-Rogers E.,
- Share D.,
- et al.
- Moscucci M.,
- Eagle K.A.,
- Share D.,
- et al.
- Moscucci M.,
- Rogers E.K.,
- Montoye C.,
- et al.
- Moscucci M.,
- O’Connor G.T.,
- Ellis S.G.,
- et al.
- Parson L.
- ↵(2007) Red Book (Thomson Healthcare, Montvale NJ).