This article is more than 5 years old. Information may no longer be current.
Enoxaparin in Primary PCI: Insights from the ATOLL Trial
The randomized ATOLL trial is an academic trial led by the ACTION Research group that aimed to show the superiority of enoxaparin used as an IV drug when compared with the classic anticoagulant used by interventionalists, namely unfractionated heparin.
The IV administration of a single dose of enoxaparin (0.5mg/kg) had previously been shown to have an excellent pharmacodynamic profile. In particular, fast onset of administration as the peak of effect was obtained in less than 10 minutes to reach the anticoagulation target in more than 95% of patients with a controlled effect that lasted only 2 hours. These favorable pharmacodynamic properties did show a 48% reduction of major bleeding in elective PCI performed through the femoral approach in the STEEPLE trial, with no effect on ischemic endpoints in this low-risk population.
Johanne Silvain
Gilles Montalescot
The ATOLL — or acute STEMI treated with primary angioplasty and IV enoxaparin or unfractionated heparin (UFH) to lower ischemic and bleeding events at short- and long-term follow-up — trial had an ambitious design. In short, it intended to show that a more optimal anticoagulation with enoxaparin, a drug that also limits platelet activation or the von Willebrand factor release observed with UFH, could provide a reduction of an ischemic endpoint on top of a reduction of major bleeding events. This is important to understand the choice of the primary endpoint of this trial, which combined both ischemic and major bleeding endpoints.
Patients with STEMI (n=910) were randomly assigned in the pre-hospital setting or directly in the cath lab if they did not receive any previous anticoagulation, with a protocol that forbid crossover to a different anticoagulant than the one allocated by randomization. Therefore this study was a pure head-to-head comparison of anticoagulant therapy, with use of glycoprotein IIb/IIIa inhibitors in both arms if necessary, two important characteristics absent from the HORIZONS trial. Another important specificity of this European trial is the predominant use of the radial approach, which was used in two-thirds of STEMI patients.
Intention-to-Treat Analysis
In the intention-to-treat analysis, enoxaparin compared with UFH reduced the primary composite endpoint of 30-day death, complication of MI, procedural failure and major bleeding in patients undergoing primary PCI for STEMI by 17%, although this fell just short of statistical significance (P=.06). To analyze the results of this study properly, we should look at the ischemic and bleeding components separately.
Indeed, all ischemic endpoints were significantly reduced with IV enoxaparin, which provided a 41% reduction (RR=0.59; 95% CI, 0.38-0.91) of the classic ischemic endpoint of death, recurrent MI/ACS or urgent revascularization. However, the expected reduction in major bleeding with enoxaparin did not occur, most likely due to predominant use of radial access, which limited the difference of bleeding previously observed in the STEEPLE trial.
Most importantly, we observed a trend toward a 40% mortality reduction (RR=0.60; 95% CI, 0.33-1.07) and a significant reduction of the combination of death and resuscitated cardiac arrest (P=.049) with IV enoxaparin. This effect on mortality appeared, this time, to be in opposition to bivalirudin (Angiomax, The Medicines Company) in the HORIZONS trial and fondaparinux (Arixtra, GlaxoSmithKline) in the OASIS-6 trial, driven by the reduction in the rate of ischemic endpoints rather than a reduction in bleeding events.
Per-Protocol Analysis
At the European Society of Cardiology Congress 2012 in Munich, we presented the results of the pre-specified per-protocol analysis of the ATOLL trial. This analysis was performed by excluding patients that received more than one heparin (protocol violation). This analysis was conducted according to the ICH guidelines using the same statistical analysis as the one performed for the intent-to-treat analysis. Of the 910 patients enrolled in the trial, 87% (n=797) fulfilled the criteria for the per-protocol analysis.
In this analysis, enoxaparin was superior to UFH to prevent all types of ischemic endpoints in the study, including a significant 33% reduction of the primary endpoint (RR=0.77; 95% CI, 0.62-0.95). Similar superiority of enoxaparin was observed for major bleeding (RR=0.46; 95% CI, 0.21-1.01) and CV mortality (RR=0.36; 95% CI, 0.18-0.74).
Overview from Meta-Analyses
Even if the secondary endpoints analysis or the per-protocol analysis results seems valid, ATOLL can be criticized for its lack of power for hard endpoints comparison, an unfortunate but frequent characteristic of academic trials when compared with industry-designed trials. However, this lack of power was overcome with the publication of two meta-analyses, including one from our group and one from a different research group.
The first study, which was published in the Journal of Thrombosis and Haemostasis in 2011 by Navarese and colleagues, focused on primary PCI, showing that low–molecular-weight heparin (in most cases enoxaparin) was superior in reducing mortality (RR=0.51; 95% CI 0.41-0.64) and major bleeding (RR=0.68; 95% CI 0.49-0.94) when regrouping the data of 16,286 patients.
The second meta-analysis, which we published in the British Medical Journal in 2012, pooled more than 30,000 patients treated with enoxaparin only and extended these results to all types of PCI, with a global reduction in mortality of 34% (RR=0.66; 95% CI, 0.57-0.76) independently of the clinical presentation. The benefit was driven largely by the effect obtained in the 10,243 STEMI patients treated with primary PCI, with a reduction in mortality of 48% (RR=0.52; 95% CI, 0.42-0.64) through a combined effect on the ischemic endpoints — recurrent MI or complication of MI (RR=0.76; 95% CI 0.60-0.96) and bleeding endpoints (RR=0.72; 95% CI, 0.56-0.93). In our opinion, these meta-analyses confirm and reinforce the ATOLL findings, with an improved safety and efficacy globally with enoxaparin compared with UFH, especially when you choose the intravenous administration of a single bolus of enoxaparin.
ESC Recommendations
Taking into account the results of the three recent trials on STEMI patients treated with primary PCI — OASIS-6 PCI subgroup with fondaparinux, HORIZONS with bivalirudin and ATOLL with enoxaparin — the question remains as to which is the preferred anticoagulation choice for our patients in primary PCI. This question was addressed by the ESC working group on STEMI. The underlying question was the following: Should we prefer a reduction in the risk of major bleeding, translating into a reduction of mortality, or should we use drugs that decrease the risk of ischemic endpoints (and bleeding in femoral primary PCI), with also a benefit on mortality?
The ESC guidelines updated the levels of recommendation for anticoagulants in primary PCI and logically attributed a Class IIIb (“not recommended for primary PCI”) to fondaparinux. Bivalirudin was also logically granted a recommendation Class Ib with the positive results of the large HORIZONS trial (“recommended over UFH”). However, they gave a mixed message concerning enoxaparin and UFH. Not only does the level of evidence for UFH in the current guidelines remain poor (level C), but there is also now at least two other options that look superior (bivalirudin and enoxaparin) to UFH with a direct impact on mortality that could have logically impacted the actual level or class of recommendation of UFH, which remains in Class Ic with the indication that “it must be used in patients not receiving bivalirudin or enoxaparin.” Enoxaparin was granted a Class IIb recommendation with the indication that “it may be preferred over UFH,” as the ATOLL randomized trial lacked the power to show significant differences in its primary endpoint, although both meta-analyses, which included randomized trials and registries, showed a reduction in mortality. Our opinion is that this drug could have been granted a better level of recommendation, especially compared with UFH for which evidence does not rely on randomized trials, registries or meta-analyses.
Conclusion
Intravenous enoxaparin, used as a single bolus, has some major advantages: the drug is widely available worldwide, inexpensive, well known by all cardiologists, can be used with or without glycoprotein IIb/IIIa inhibitors, and has a pharmacodynamic profile perfectly adapted to primary PCI and excellent clinical results in primary PCI in terms of efficacy and safety. The ischemic benefit remains consistent regardless of the arterial access and results also show reduction in bleeding risk, especially if using the femoral approach. While waiting for a trial comparing enoxaparin with bivalirudin in primary PCI, the main conclusion is that we are moving away from UFH in this procedure.
Disclosure: Montalescot reports receiving grant support from Abbott Vascular, Boston Scientific, Cordis, Eli Lilly, Fédération Française de Cardiologie, Fondation de France, Guerbet Medical, INSERM, ITC Edison, Medtronic, Pfizer, Sanofi-Aventis, Société Française de Cardiologie and Stago; consulting or board and lecture fees from AstraZeneca, Bayer, Boehringer Ingelheim, Cardiovascular Research Foundation, Cleveland Clinic Research Foundation, Daiichi Sankyo, Duke Institute, Eli Lilly, Europa, Lead-up, GlaxoSmithKline, Institut de Cardiologie de Montréal, Menarini, Nanospheres, Novartis, Pfizer, Portola, Sanofi-Aventis, The Medicines Company and the TIMI Study Group. Silvain reports receiving research grants from Brahms, Daiichi Sankyo, Eli Lilly, INSERM, Fédération Française de Cardiologie, Sanofi-Aventis and Société Française de Cardiologie; consulting fees from Daiichi Sankyo and Eli Lilly; and speakers’ honoraria from AstraZeneca, Daiichi Sankyo, Eli Lilly and Servier.