By Jeffrey Zimmet, MD, PhD

Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center

Dr. Zimmet reports no financial relationships relevant to this field of study.

SYNOPSIS: In this trial of post-percutaneous coronary intervention patients with atrial fibrillation, dual antithrombotic therapy with dabigatran and a P2Y12 inhibitor showed lower rates of bleeding but similar ischemic and thrombotic outcomes compared to a triple therapy regimen with warfarin.

SOURCE: Cannon CP, Bhatt DL, Oldgren J, et al. Dual antithrombotic therapy with dabigatran after PCI in atrial fibrillation. N Engl J Med 2017 Aug 27. doi: 10.1056/NEJMoa1708454. [Epub ahead of print].

Balancing antithrombotic therapy with bleeding risk is a major issue in cardiovascular medicine, but is an especially thorny problem in patients with atrial fibrillation (AF) who undergo coronary stenting. In these patients, anticoagulation therapy is indicated for the prevention of stroke, and a combination of aspirin with a P2Y12 inhibitor is prescribed for prevention of stent thrombosis and other ischemic events. The combination of these agents, generally known as triple therapy, has been demonstrated to substantially increase the risk of serious bleeding compared to individual components of this treatment. Yet, U.S. guidelines still recommend this course of treatment. The strategy of dropping aspirin from the mix was first studied in the 573-patient WOEST trial, published in 2013, where it was reported to reduce the risk of bleeding without an increase in ischemic complications. The authors of the PIONEER AF-PCI trial, who published their results in late 2016, subsequently studied a dual antithrombotic approach with the direct oral anticoagulant rivaroxaban in more than 2,000 post-percutaneous coronary intervention (PCI) AF patients. In this study, the safety data appeared promising, with a reduction in bleeding events without evidence of decreased efficacy.

The RE-DUAL PCI trial, the results of which were presented recently at the 2017 European Society of Cardiology meeting, examined this question with the oral direct thrombin inhibitor dabigatran. In what is now the largest such study to date, 2,725 patients with AF undergoing coronary stenting were assigned randomly to either triple therapy with warfarin, aspirin, and a P2Y12 inhibitor, or to dual therapy with one of two doses of dabigatran plus a P2Y12 inhibitor in a paired fashion with triple therapy. Notably, only the 150 mg dose of dabigatran is approved in the United States for treatment of patients with non-valvular AF; the 110 mg dose is an option in most countries. For patients in the triple therapy group, aspirin was dropped after one month in those receiving bare metal stents, and after three months in drug-eluting stent recipients. Similar to PIONEER, the primary endpoint here concerned bleeding alone — time to first major or clinically relevant bleeding event, as defined by the International Society on Thrombosis and Haemostasis (ISTH). Ischemic and thromboembolic events were analyzed as a secondary endpoint using a noninferiority design, as a composite of all-cause death, myocardial infarction, stroke, systemic embolism, or unplanned revascularization.

Among patients in the trial, approximately 50% underwent PCI for acute coronary syndrome, with the remainder presenting with stable coronary disease. More than 80% received only drug-eluting stents, and the majority received clopidogrel as the P2Y12 inhibitor, with ticagrelor used in only 12%. Of the patients receiving warfarin, the mean time in the therapeutic international normalized ratio range was 64%, which closely mirrors prior trials.

At a mean follow-up of 14 months, a qualifying bleeding event was recorded in 15.4% of the 110 mg dual-therapy group and 26.9% of the triple-therapy group with warfarin (95% confidence interval [CI], 0.42-0.63; P < 0.001), and 20.2% in the 150 mg dual-therapy group compared to 25.7% in the corresponding triple-therapy group (hazard ratio [HR], 0.72; 95% CI, 0.58-0.88; P < 0.001). Rates of major bleeding alone, even when defined by the more stringent Thrombolysis in Myocardial Infarction (TIMI) criteria, also were significantly lower in both dual-therapy groups: 1.4% in the 110 mg dual-therapy group and 2.1% in the 150 mg dual-therapy group, compared to 3.8% and 3.9% in the corresponding triple-therapy groups.

The composite secondary efficacy endpoint of thromboembolic events, death, or unplanned revascularization was similar between the dual- and triple-therapy groups: 13.7% in the two dual-therapy groups combined compared to 13.4% in the triple-therapy group (HR, 1.04; 95% CI, 0.84-1.29; P = 0.005 for noninferiority). Rates of stent thrombosis were low overall, occurring in 15 patients in the 110 mg dual-therapy group compared to eight in the triple-therapy group (P = 0.15), and in seven patients in the 150 mg dual-therapy group compared to seven in the corresponding triple-therapy group (P = 0.98). The authors concluded that dual antithrombotic therapy with either of the two doses of dabigatran conferred lower bleeding risk compared to triple therapy with warfarin, and was noninferior regarding the risk of ischemic and thromboembolic events.


The main results of the RE-DUAL PCI trial should come as no surprise. After all, this study was sponsored by the company that manufactures dabigatran, and was designed as a safety trial. The bleeding definition used as part of the primary endpoint, ISTH clinically relevant bleeding, is one of the more inclusive measures available and virtually guaranteed a large number of events. The “dual-therapy” groups included only the two dosing regimens with dabigatran, with no “WOEST-like” arm with warfarin. The triple-therapy arm was destined to lose out regarding bleeding, with full-dose warfarin acting as a strawman of sorts. So why should clinicians pay attention to these results, and what can clinicians learn that can inform general practice?

First, this is a confirmation and reminder that triple therapy is associated with strikingly high rates of clinically evident bleeding. Fully one-quarter of patients in the triple therapy group exhibited some clinically evident bleeding, and nearly 4% experienced major bleeding, according to the TIMI definition. RE-DUAL used an abbreviated version of triple therapy that has been common outside the United States: stopping aspirin after one month for bare metal stents and after three months for drug-eluting stents. So, in effect, this trial compared dual therapy using a direct oral anticoagulant with three months of triple therapy (in the majority who received drug-eluting stents) followed by dual therapy with warfarin, and still showed a significant difference in bleeding hazard. More difficult is the interpretation of the ischemic and thrombotic events in the trial, which were relegated to secondary endpoints. The authors of the trial enrolled fewer patients than originally planned, and the authors noted that they had limited power to examine efficacy. It was certainly underpowered to study low-frequency events, including the feared complication of stent thrombosis. Yet, as in the similarly underpowered PIONEER and WOEST trials, there was not even a signal or trend suggesting increased risk with a direct oral anticoagulant-based dual-therapy regimen. This leaves room for individualization of therapy, balancing a patient’s specific risk profiles for thrombotic vs. bleeding outcomes. There remain patients with elevated thrombotic risk (based on clinical presentation and stent anatomy) for whom triple therapy still may be the favored approach. However, for most patients with AF after PCI, a dual antithrombotic regimen should be considered a viable option.