Antithrombotic therapy for patients with atrial fibrillation undergoing percutaneous coronary intervention: balance best with double antithrombotic therapy

This editorial refers to ‘Edoxaban in atrial fibrillation patients with percutaneous coronary intervention by acute or chronic coronary syndrome presentation: a pre-specified analysis of the ENTRUST-AF PCI trial’†, by P. Vranckx et al., on page 4497. Management of acute coronary syndrome (ACS) and/or percutaneous coronary intervention (PCI) in patients with concomitant atrial fibrillation (AF) requires balancing competing risks of bleeding and ischaemic events. These patients have indications for dual antiplatelet therapy [DAPT (aspirin and a P2Y12 inhibitor)] and oral anticoagulation (OAC), but the major bleeding rate in patients treated with all three agents is unacceptably high.1 Four randomized controlled trials (RCTs) in patients with AF and ACS and/or undergoing PCI compared double antithrombotic therapy [non-vitamin K antagonist (VKA) OAC (NOAC) plus a P2Y12 inhibitor] with triple antithrombotic therapy (OAC plus DAPT).2–5 In a network meta-analysis of these studies, double antithrombotic therapy was associated with a lower risk of major bleeding than triple therapy, without a significant increase in ischaemic events.6 These trials enrolled patients undergoing PCI for both ACS and chronic coronary syndromes (CCS). Patients undergoing PCI for ACS are at higher risk of recurrent ischaemic events, including stent thrombosis, than those undergoing PCI for CCS,7 but are not at higher risk for post-discharge bleeding events.8 Thus, the regimen that strikes the most appropriate balance in patients with AF undergoing PCI may differ based on the indication for PCI. In a subgroup analysis of ENTRUST-AF PCI in this issue of the European Heart Journal, Vranckx et al. evaluated the interaction between indication for PCI (ACS vs. CCS) and a regimen of edoxaban plus P2Y12 inhibitor on bleeding and ischaemic events compared with triple therapy with VKA plus DAPT.9ENTRUST-AF PCI was an open-label clinical trial that randomized 1506 patients with AF who had undergone PCI within 5 days to either full-dose edoxaban and a P2Y12 inhibitor or VKA plus a P2Y12 inhibitor and aspirin.6 The edoxaban regimen was non-inferior to the VKA regimen on the primary endpoint, International Society on Thrombosis and Haemostasis (ISTH) major or clinically relevant non-major (CRNM) bleeding, and there was no significant difference in ischaemic events. The indication for PCI was ACS in 777 patients (51.6%), and there was no interaction between indication for PCI and antithrombotic regimen for the primary bleeding and ischaemic endpoints. The lack of interaction between PCI indication and the safety and effectiveness of NOAC plus a P2Y12 inhibitor as compared with VKA plus DAPT observed in this analysis parallels similar analyses from the three other trials evaluating antithrombotic therapies in patients with AF undergoing PCI (Figure 1).3,10,11 In RE-DUAL PCI and AUGUSTUS, formal analyses examining the interaction between indication for PCI (ACS vs. CCS) and the safety and effectiveness of double compared with triple antithrombotic therapy were non-significant. No formal interaction analysis was performed in PIONEER AF-PCI, but the effect of NOAC-based double antithrombotic therapy on bleeding and ischaemic outcomes was similar for patients with ACS and CCS. In all four studies, the reduction in bleeding with double as compared with triple antithrombotic therapy was similar in patients undergoing PCI for ACS or CCS. Estimates of the effect of double compared with triple antithrombotic therapy on ischaemic events were broad, but there was no significant increase in patients with ACS or CCS. Major bleeding and ischaemic events in patients with ACS and CCS treated with triple or double antithrombotic therapy.* Rates of major or CRNM bleeding and composite ischaemic events are shown for patients undergoing PCI for ACS and CCS by treatment arm (OAC + P2Y12 inhibitor and VKA + DAPT) in major trials evaluating antithrombotic regimens in patients with AF undergoing PCI. *Raw percentages and odds ratios are displayed, as not all trials reported cumulative incidence and hazard ratios for both subgroups. Data are from the 150 mg dabigatran twice daily + P2Y12 inhibitor arm of RE-DUAL PCI. From AUGUSTUS, data are derived from the aspirin vs. placebo half of the factorial design. Major bleeding and ischaemic events in patients with ACS and CCS treated with triple or double antithrombotic therapy.* Rates of major or CRNM bleeding and composite ischaemic events are shown for patients undergoing PCI for ACS and CCS by treatment arm (OAC + P2Y12 inhibitor and VKA + DAPT) in major trials evaluating antithrombotic regimens in patients with AF undergoing PCI. *Raw percentages and odds ratios are displayed, as not all trials reported cumulative incidence and hazard ratios for both subgroups. Data are from the 150 mg dabigatran twice daily + P2Y12 inhibitor arm of RE-DUAL PCI. From AUGUSTUS, data are derived from the aspirin vs. placebo half of the factorial design. These results show that double antithrombotic therapy with an NOAC and P2Y12 inhibitor provides the best balance of protection from ischaemic events while limiting bleeding in patients with AF undergoing PCI, regardless of the indication for PCI. Though this subgroup analysis of ENTRUST-AF PCI did not show a difference in the incidence of bleeding between double and triple antithrombotic therapy in the CCS subgroup, the P-value for interaction was non-significant and conclusions about the relative safety of edoxaban-based double therapy in patients with CCS based on an underpowered subgroup analysis cannot be made. Similarly, the point estimate suggesting a higher rate of ischaemic events with the edoxaban-based regimen in patients with ACS in ENTRUST-AF PCI should be interpreted with caution, as the P-value for interaction was not significant. This analysis of ENTRUST-AF PCI does not identify patients with PCI for ACS as a subgroup that benefits from triple antithrombotic therapy, but there may yet be a role for short-term triple antithrombotic therapy in some patients with AF undergoing PCI. Stent thrombosis is most common within the first 30 days after PCI, and use of triple antithrombotic therapy during this time results in a relatively equal trade-off between bleeding and ischaemic events among patients with AF undergoing PCI.12,13 Pooling data from the major clinical trials of patients with AF undergoing PCI may identify subgroups of patients that benefit more from 30-day triple antithrombotic therapy, and this strategy could be tested in RCTs specifically enrolling these patients. Finally, other combinations of drugs, including DAPT without OAC for the initial period after ACS or PCI to minimize bleeding risk, still need to be tested in RCTs. For now, in the 12 months after PCI, the current evidence is clear that patients with AF should be treated with NOAC plus a P2Y12 inhibitor, regardless of indication for PCI. Conflict of interest: A.C.F. reports a career development grant and honoraria from the American Heart Association, and grant funding from Boston Scientific. R.D.L. reports Institutional research grants and consulting fees from Bristol-Myers Squibb, Pfizer, GlaxoSmithKline, Medtronic PLC, and Sanofi, and consulting fees from Amgen, Bayer, and Boehringer Ingelheim. The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.