Oral anticoagulants for atrial fibrillation and acute coronary syndrome with or without stenting

Introduction The optimal antithrombotic treatment for patients with atrial fibrillation and acute coronary syndrome (ACS), whether or not treated with revascularization, is a complex and debated issue. Clinical practice guidelines are based on expert opinion – lacking strong evidence based on clinical trials – and recommend a personalized treatment balancing the atherothrombotic, thromboembolic and bleeding risks. The presence of atrial fibrillation in the setting of ACS is not infrequent. Patients with atrial fibrillation have a 34% incidence of coronary artery disease, and up to 20% of these patients at some time will require a revascularization procedure, usually by means of angioplasty (percutaneous coronary intervention, PCI) with stents. After PCI, dual antiplatelet therapy (DAPT) with aspirin and a PY12 receptor inhibitor has been shown to be better than aspirin alone, or warfarin, or aspirin and warfarin,1,2 in preventing the often fatal complication of stent thrombosis. On the other hand, albeit DAPT is the standard of care for at least 1 year after ACS, the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events study demonstrated that this treatment was clearly inferior to anticoagulation with a vitamin K antagonist (VKA) in preventing ischemic stroke and systemic thromboembolic episodes.3 Accordingly, anticoagulation still remains the gold standard for thromboembolic prevention in patients with atrial fibrillation. The different approaches to the antithrombotic treatment in the setting of ACS and atrial fibrillation stem from the different pathophysiologic mechanism behind stent thrombosis in ACS and ischemic stroke in atrial fibrillation: stent thrombosis is due mainly to platelet activation when blood faces the foreign surface of the stent (primary hemostasis); on the other hand, in atrial fibrillation, the activation of coagulation is mostly due to blood stagnation. When a patient presents with both clinical situations, the concomitant use of DAPT and anticoagulation has long been considered necessary to achieve adequate atherothrombotic and thromboembolic protection. On the contrary, this combination therapy is burdened with a significant increase in major bleeding, reaching 7% per year.4 We will review new treatment avenues provided by the recently introduced nonvitamin K antagonists oral anticoagulants (new oral anticoagulants, NOACs) when there is the need for both DAPT and oral anticoagulation. We will also question whether the presence of a coronary stent requires ‘triple therapy’, or if the NOACs could offer new solutions. Acute coronary syndrome and atrial fibrillation ACS carries an increased intrinsic risk of recurrent ischemia and cardiovascular death as compared with stable coronary artery disease. Currently, despite best medical treatment, the risk of recurrent infarction, stroke and death is about 10% during the 1st year. The implant of a coronary stent during PCI – regarded as the best option for coronary revascularization during the acute phase of an ACS – entails a peculiar thrombotic risk because of the possibility of stent thrombosis. When this dreaded complication occurs, there is a 70% risk of myocardial infarction (MI), and up to a 20% risk of death, depending on the implant site, vessel dimension, length and type of stent, the technique of implant and the overall clinical context. When atrial fibrillation and ACS with or without stent implant coexist, a 2014 European Society of Cardiology (ESC) consensus document5 recommends ‘triple therapy’ with oral anticoagulants, aspirin and clopidogrel for 1–6 months according to the bleeding risk (HAS-BLED ≥3 or ≤2, respectively; recommendation class IIa, level of evidence C). In a subsequent consensus document, the European Heart Rhythm Association6 released an updated, but independent version of the 2014 recommendations, in which the introduction of NOACs simplified the decisional algorithm: regardless of stent implant. The patient with atrial fibrillation and ACS should have triple therapy with new or standard oral anticoagulant for the first 6 months after the event. There are some factors, beside the assessment of the bleeding risk, that could determine a shortening or a prolongation of this reference duration of triple therapy. In particular, a protracted triple therapy is recommended when the atherothrombotic risk is high, the GRACE score is at least 118, or in the presence of a recurrent MI, left main, proximal left anterior descending or proximal bifurcation stenting and when first-generation drug-eluting stents (DES) are used.6 For patients with atrial fibrillation and anticoagulant therapy presenting with a history of stent thrombosis, the consensus has been to favor the antithrombotic treatment with DAPT and NOACs over the containment of the risk of bleeding related to this treatment. Accordingly, in this contest, triple therapy should be continued for a longer period of time.5,6 At variance from the previous two documents, the 2015 ESC guidelines for ACS without ST elevation [non ST elevation myocardial infarction - acute coronary syndrome (NSTE-ACS)]7 and the 2016 Atrial Fibrillation Guidelines8 provide different therapeutic recommendation for the antithrombotic treatment in patients with atrial fibrillation and ACS, depending on whether the patient was treated with PCI or conservatively. The recommendation for antithrombotic treatment of patients receiving stenting are essentially the same as previous guidelines, but they differ for the patient with ACS not receiving revascularization, in whom the oral anticoagulation is associated with a single antiplatelet agent (aspirin or clopidogrel) from the time of the acute event (recommendation class IIa, level of evidence C).7,8 There is a substantial agreement among guidelines and consensus documents as to the continuation of antivitamin K/NOACs and single antiplatelet drug until – at most – the 12th month after the ACS and, after such time, recommending a monotherapy with oral anticoagulants. The only class I recommendation (level of evidence B) in this setting is continuation of monotherapy with the oral anticoagulant after the 12th month from the acute event. In the secondary postischemic prevention, in fact, warfarin has been shown to be superior to aspirin,9 and the association of warfarin and aspirin does not afford any certain additional benefit as compared with the same levels of anticoagulation of warfarin alone, but certainly increases the risk of bleeding. All the consensus documents5–8 consider only clopidogrelas as the P2Y12 inhibitor to use in this context, as the more powerful inhibitors prasugrel or ticagrelor would entail an unacceptably higher risk of bleeding. Empiric or evidence-base medicine? Class IIa recommendation of the ESC Consensus documents, the 2014 Guidelines for myocardial revascularization, the 2015 Guidelines for NSTE-ACS7 and the 2016 atrial fibrillation Guidelines8 all have a level of evidence C, meaning that they are based on expert opinion and/or small or retrospective studies and registries. Although there are no trials favoring triple therapy, the efficacy of a strategy limiting treatment to oral anticoagulant and a single antiplatelet drug still remains to be proven, even in selected patients such as those not treated with stents and those at lower atherothrombotic risk.10 An analysis of the Danish National Registry comparing double and triple therapy did not show statistically significant differences in terms of incidence of MI or cardiac deaths. After 90 days, although bleeding events were more frequent after triple therapy [hazard ratio 1.47; 95% confidence interval (CI), 1.04–2.08], but after 1 year, the two treatments were equivalent (hazard ratio 1.36; 95% CI, 0.95–1.95). The occurrence of major bleeding episodes were also similar.11 The possibility of limiting combination therapy to only one antiplatelet drug has been tested by studies not large enough to conclusively determine efficacy, albeit this regimen gained popularity. Accordingly, and relaying essentially on ‘opinions’, aspirin has been the drug most commonly withdrawn from the antithrombotic ‘cocktail’. Considering the combination of clopidogrel and a VKA, the What is the Optimal antiplatElet and anticoagulant therapy in patients with oral anticoagulant and coronary StenTing (WOEST) trial12 enrolled 570 patients on oral anticoagulants requiring a PCI, either for stable disease or after ACS, who were randomized either to clopidogrel and a VKA or triple therapy. After 1-year follow-up, the clopidogrel and VKA arm of the study had a significantly lower incidence of bleeding episodes (19.4 vs. 44.4%, P < 0.0001), no increase in thrombotic complications and even a decrease in the ischemic endpoint.12 To curb the enthusiasm brought by the WOEST results, some important points should be considered:13 The better safety of the double therapy was the result reduction of Thrombolysis in MI (TIMI) minimal and minor bleeding, Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) small/moderate bleeding and Bleeding Academic Research Consortium (BARC) 1/2/3a bleeding. TIMI major bleeding episodes, GUSTO severe and BARC 3b/3c bleeding episodes were similar between the two arms of the study.12,13 The reduction of the efficacy endpoint, a composite of death, MI, target vessel revascularization, stent thrombosis and cerebrovascular stroke [31 events in the clopidogrel-VKA arm (11.1%) vs. 50 events (17.6%) in the triple therapy arm; hazard ratio 0.60, 95% CI, 0.38–0.94, P = 0.025], appeared to be the result of reduction of all-cause mortality (hazard ratio 0.39; 95% CI, 0.16–0.93; P = 0.027), itself the result of reduction of noncardiovascular deaths, more likely chance-related rather than an indirect effect of the reduction of bleeding episodes. Only a minority of patients in the WOEST study experienced an ACS; the vast majority had an elective stenting procedure. Double therapy was compared in these patients with unjustified prolongation of the triple therapy (12 months also inpatients after elective PCI). The relevance of the observed reduction of minor bleeding episodes with the double therapy in the WOEST study should not, however, be underestimated, in that even minor bleeding episodes have a significant prognostic implication, albeit indirect, as they may increase the ischemic events once they prompt the suspension of the antithrombotic treatment. These data are consistent with those of the Danish National Registry, in which 12 165 patients with atrial fibrillation admitted to hospital for ACS or elective PCI, the combination clopidogrel-VKA did not result in more recurrent coronary events as compared with the triple therapy (hazard ratio 0.69, 95% CI 0.48–1.00).14 The AVIATOR Registry as well, studying 859 patients with atrial fibrillation and ACS, showed fewer bleeding events without a difference in the occurrence of ischemic events in patients in whom aspirin was withheld as compared with triple therapy.15 Stent implant: impact on the thrombotic risk From a pathophysiologic perspective, it is possible that the presence and type of stent could critically affect the need and duration of DAPT in association with the oral anticoagulant. After the elective placement of a bare metal stent (BMS), guidelines recommend at least 4 weeks of DAPT, the time needed for stent strut endothelization. The first-generation DES required at least 3 months of DAPT if the eluting drug was a limus (inhibitors of mammalian target of rapamycin), and 6 months when the eluting drug was paclitaxel. Recent data indicate that with the latter generation DES, protracting DAPT was not as essential as after first-generation or second-generation DES.16,17 Even though the duration of DAPT after the newer generation stents is still debated, some of these stents were clinically approved for a 3-month duration of therapy after elective PCI, not in the context of ACS. Accordingly, first-generation and second-generation stents are used sparingly in patients on oral anticoagulants. Nevertheless, recommendations on the type of stent are still not conclusive. Recent data suggest that the risk of stent thrombosis during the 1st year after the implant is similar for new-generation DES and BMS,18 likewise similar is the incidence of adverse events in patients who suspended DAPT.19 Retrospective data from the Anti-Xa Therapy to Lower Cardiovascular Events in Addition to Standard Therapy in Subjects with Acute Coronary Syndrome study showed a similar incidence of ischemic events and all-cause mortality in patients receiving anticoagulants after PCI with BMS or DES; the occurrence of major bleeding events during the follow-up was higher after DES.20 The latest ESC guidelines for myocardial revascularization recommend new-generation DES in patients at low risk for bleeding (HAS-BLED score ≤2), whereas for patients at high risk (HAS-BLED >3), the choice between BMS and DES should be made on an individual basis. The Duration of triple therapy in patients requiring oral anticoagulant after drug-eluting stent implantation (ISAR-TRIPLE) study21 compared 6 months’ vs. 6 weeks’ duration of combination DAPT and a VKA in 614 patients undergoing PCI with DES. The composite primary endpoint of death, MI, stent thrombosis, stroke and major bleeding episodes was not different amongst the two regimens. Furthermore, the two groups were similar as to the ischemic combined endpoint (cardiac death, MI, stent thrombosis and ischemic stroke), as well as to the secondary endpoint of major bleeding episodes according to the TIMI classification. Accordingly, as far as the net clinical benefit is concerned, a 6-week course of triple therapy seems not better than 6 months, thus emphasizing the need for individualized assessment of the ischemic and bleeding risk.21 Albeit the present recommendation to continue triple therapy for 3–6 months, considering also the type of stent, applies to the majority of patients, it still remains to be determined whether a shorter course would be as well tolerated and effective, particularly considering the third-generation limus DES.22 Recent data regarding new-generation stents have provided a new, more limited dimension to the risk of stent thrombosis as a deciding factor for long-term DAPT. In situations of high bleeding risk, such as the need for a VKA in patients with atrial fibrillation, it is the atherothrombotic risk following an ACS rather than the risk of stent thrombosis that drives the need for a course of triple therapy. Triple therapy with nonvitamin K antagonist oral anticoagulants rather than with vitamin K antagonists: the best net clinical benefit presently available? The net clinical benefit of NOACs is now well known, with an efficacy profile at least as good as VKAs, but with a better safety, particularly because of a lower rate of intracranial hemorrhage. The NOACs as an alternative to VKAs should thus provide, as part of the triple therapy, an adequate antiischemic and antithrombotic protection, but with lower risk of bleeding, especially intracranial bleeding. Data on the direct comparison between NOACs and VKAs in patients with atrial fibrillation undergoing PCI are not yet available, nonetheless some recommendations could be derived from post-hoc analyses. The Randomized Evaluation of Long-term anticoagulant therapY (RE-LY) trial23 showed that the use of single or double antiplatelet therapy did not affect the efficacy and safety profile of two different doses of dabigatran vs. warfarin. A post-hoc analysis of the Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE)24 study confirmed the net clinical benefit of apixaban vs. warfarin, regardless of the concurrent use of aspirin. The Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation25 study, comparing warfarin with edoxaban, also confirmed that outcome measures were not changed by the use of aspirin.26 The NOACs were tested in combination with DAPT in secondary postischemic prevention. In this setting, dabigatran27 and apixaban28 were given at the same doses effective in patients with atrial fibrillation, demonstrating, in the respective trials, an increase in bleeding events without a clear antiischemic benefit. On the other hand, in patients with ACS already taking DAPT, rivaroxaban, at lower dosage than the one used in atrial fibrillation [2.5 mg bis in die (b.i.d.) and 5 mg b.i.d. compared with the 20/15 mg as a single dose used in atrial fibrillation] was associated with significantly less all-cause mortality and an additional protective effect on stent thrombosis.20,29 The current information could suggest two main investigative avenues: on the one hand, the evaluation of the efficacy and safety of an ‘ATLAS-ACS-like’ strategy, that is the concomitant use of a low-dose NOAC with DAPT. On the other hand, using a ‘WOEST-like’ strategy, consisting of a single antiplatelet drug and a NOAC, could provide enough antithrombotic effect as to prevent adverse ischemic events, including stent thrombosis and thromboembolism, with a clear safety advantage. These two strategies were recently tested in the An Open-label, Randomized, Controlled, Multicenter Study Exploring Two Treatment Strategies of Rivaroxaban and a Dose-Adjusted Oral Vitamin K Antagonist Treatment Strategy in Subjects with Atrial Fibrillation Who Undergo Percutaneous Coronary Intervention (PIONEER-AF) study with rivaroxaban.30 Here both these ‘new’ antithrombotic regimens (rivaroxaban 15 mg once a day and clopidogrel; andrivaroxaban 2.5 mg b.i.d. and aspirin and clopidogrel) resulted in fewer bleeding episodes than the classical triple therapy with warfarin, aspirin and clopidogrel, maintaining the same efficacy. On the contrary, the study did not have sufficient statistical power as to provide definitive efficacy conclusions. Also, the observation that the incidence of major adverse cardiac events (major adverse cardiovascular events: cardiovascular death, MI and stroke) was similar in the three groups cannot be considered definitive. The same limitation applies also to stent thrombosis (<1% in each arm of the study), similar in the three groups (Table 1). Accordingly, it is still uncertain whether these regimens should represent the new standard of treatment. Neither the recently published PIONEER-AF study,30 nor, in all likelihood, the still ongoing trials with dabigatran, apixaban, andedoxaban (see 26,29 for comprehensive reviews) are large enough as to provide definitive conclusions concerning efficacy.Table 1: Trials with oral antiplatelets drugs (antivitamin K or nonvitamin K antagonists oral anticoagulants) in patients with atrial fibrillation undergoing percutaneous coronary interventionConclusion The advent of NOACs provided new and varied opportunities for their use with concomitant antiplatelet drugs in ‘difficult’ situations. NOACs as part of triple therapy should provide a better safety profile than traditional VKAs. Furthermore, with the new-generation DES stent, thrombosis is rare, and mostly limited to the 1st postimplant month. Rivaroxaban is currently the only NOAC for which objective data derived from a prospective randomized study are available. Although the clinical implications of the PIONEER-AF study are still uncertain, the regimen with rivaroxaban 15 mg and clopidogrel can reasonably be recommended to the majority of patients with ACS and atrial fibrillation, particularly when the risk of adverse cardiac events and stent thrombosis is not high. As alternatives, and more in tune with the classic recommendation, a triple therapy with rivaroxaban 20 mg or even 15 mg once daily and DAPT (not addressed in the PIONEER-AF study), or other NOACs with less bleeding risk than warfarin, and DAPT, such as dabigatran 110 mg b.i.d. (tested in the RE-LY study23); apixaban 5 mg b.i.d. (tested in the ARISTOTLE study24); and edoxaban 30 mg/die (tested in the ENGAGE AF-TIMI 48 study25) are all viable options. It is hoped that the still ongoing studies such as the Evaluation of Dual Therapy with Dabigatran vs. Triple Therapy with Warfarin in Patients with AF that Undergo a PCI with Stenting (RE-DUAL PCI) with dabigatran,31 the Apixaban Versus Warfarin in Patients with AF and ACS or PCI (AUGUSTUS) with apixaban32 (Table 1), and the planned trial with edoxaban, could provide definitive answers on the optimal association regimens of anticoagulants (mainly NOACs) and antiplatelet drugs (old and new P2Y12 inhibitors) in patients with ACS and atrial fibrillation. Acknowledgements Conflicts of interest There are no conflicts of interest.