Risk of Triple Antithrombic Therapy

Abstract & Commentary

By Michael H. Crawford, MD, Editor

Source: Nilolsky E, et al. Outcomes of patients treated with triple antithrombic therapy after primary percutaneous coronary intervention for ST-elevation myocardial infarction. Am J Cardiol 2012;109:831-838.

In the setting of acute ST segment elevation myocardial infarction (STEMI), warfarin is indicated for atrial fibrillation, a large dyskinetic or aneurysmal area of the left ventricle, or mural thrombus. When treated with primary percutaneous coronary intervention (PCI), patients may require dual antiplatelet therapy as well. Treatment with three antithrombic agents will increase the bleeding risk, but the overall benefit is unclear. Thus, these investigators assessed the outcomes of such patients in the Harmonizing Outcomes and Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trial. This trial involved more than 3600 STEMI patients who were randomized to bivaliruden vs heparin plus a glycoprotein IIb/IIIa inhibitor, most of whom had PCI and were then treated with antiplatelet drugs and warfarin per physician discretion. There was also a randomization between a drug-eluting stent and a bare metal stent. Outcomes over 1 year of follow-up were compared between those who received dual antiplatelet therapy and those who received triple therapy during the initial hospitalization. There were two primary endpoints: major bleeding and net adverse clinical events (MACE plus major bleeding). Triple therapy was prescribed in 3.8% and dual antiplatelet therapy in 96.2% of the 3320 patients who underwent PCI. Warfarin indications included reduced ejection fraction (EF) with a large akinetic region in 24%; atrial fibrillation in 24%; a mural thrombus in 23%; a left ventricular aneurysm in 10%; and deep venous thrombosis in 2%. Despite the higher risk profile of the triple therapy group, they had similar short- and long-term ischemic related outcomes, but had a significantly increased rate of major bleeding during the initial hospitalization (17% vs 6.5%, P < 0.001) requiring discontinuation of warfarin in 14% of those on it. Triple therapy continued to predict major bleeding at 1 year (odds ratio 2.16, 95% confidence interval, 1.22-3.83) but was not associated with 1-year mortality. The authors concluded that the risks of triple therapy should be considered before selecting a drug-eluting stent vs a bare metal stent in patients with STEMI undergoing PCI.

Commentary

It is no surprise that triple antithrombic therapy resulted in more major and minor bleeding, but the three-fold magnitude of risk increase is concerning. Also, the observed increase in stroke risk on triple therapy may have been due to the need to stop warfarin in 14% due to major bleeding. The dual need to reduce stent thrombosis and prevent systemic emboli in some patients treated with PCI for STEMI is a dilemma. Dual antiplatelet therapy is superior for the former and warfarin for the latter, so dropping to any two-drug regimen would result in inferior therapy. One approach may be to tailor stent choice depending on the likelihood of needing triple therapy. If warfarin therapy seems necessary, a bare metal stent could be chosen to limit the duration of dual antiplatelet therapy. However, most major bleeding events were seen in the first 30 days when dual antiplatelet therapy would be needed for a bare metal stent. Thus, this approach may not reduce major bleeding much and would result in inferior therapy for the STEMI.

Some may take comfort in the low incidence of triple therapy patients (4%), but in a non-trial setting it is likely to be higher. Consider that anyone already on warfarin was excluded from the trial. Given the frequency of atrial fibrillation in an older STEMI population, the real need for warfarin may be more like 15-20%. Clearly, we need safer therapies to prevent stent thrombosis and systemic emboli. Whether some of the newer antiplatelet and oral anticoagulant drugs will fill this bill remains to be seen. In the meantime, careful patient selection for acute MI therapies is necessary.