Viewpoint: a proposal for a simple algorithm for managing oral anticoagulation and antiplatelet therapy in patients with non-valvular atrial fibrillation and coronary stents

Percutaneous coronary intervention (PCI) is not uncommon among patients on oral anticoagulation (OAC) for atrial fibrillation (AF). Yet it poses a conundrum between the demands for combining dual antiplatelet therapy (DAPT) (comprising a P2Y12 receptor blocker and aspirin) and OAC to prevent thrombotic events, and the desire to minimise the bleeding risk associated with stacking antithrombotic agents. Myriad factors come into play in determining the optimal duration, intensity and nature of the possible combinations. Given the difficulty in determining an optimal strategy, there have been many attempts to propose recommendations to help clinicians in this setting, which have recently been reviewed and synthesised.1,2 Interestingly, as the evidence available from randomised trials is extremely limited, the consensus documents published even recently are not consistent and are often highly complex, making them challenging to implement, particularly for the spectrum of physicians involved in the care of these patients. Based on a number of principles delineated in Table 1, we have attempted to provide a simpler antithrombotic regimen, applicable to most patients and situations, including elective PCI for stable coronary artery disease and urgent PCI for acute coronary syndromes (ACSs). Principles guiding the regimen recommendation Principles guiding the regimen recommendation First, the complexity of current recommendations regarding optimal antithrombotic regimens for patients requiring OAC for AF and in whom PCI is undertaken prevents adherence to them by the primary care and specialist physicians involved in the daily care of these patients. Second, an important theme of the consensus document is the need to assess both ischaemic and bleeding risk in each individual and in a dynamic fashion, with periodic reassessment. While we find this a laudable proposal, we will point out that in many patients, ischaemic and bleeding risks evolve in parallel, particularly as age is a major driver of both. When there is either an elevated risk of both or a low risk of both, it becomes difficult to operationalise the recommendations. In addition, no trial so far has utilised an algorithm using both bleeding and ischaemic risk in guiding anticoagulation therapy. We suggest that a possible solution to this conundrum is to adopt a ‘first do no harm’ policy in which every attempt is made to minimise the bleeding risk in all patients (by using the safest agents and combinations for the shortest possible time), and then to use the thromboembolic risk scores (such as the CHA2DS2-VASc) to guide the decision to use anticoagulation. In patients with a CHA2DS2-VASc of 0 or 1, refraining from using anticoagulation and limiting therapy to DAPT is both reasonable and in line with current guidelines. In patients with a CHA2DS2-VASc greater than 1, a combination of anticoagulation and antiplatelet therapy is usually necessary, unless the bleeding risk is so elevated as to contraindicate it. For most patients, this is not the case and therefore the issues are how to choose anticoagulation and antiplatelet therapy and what should be the duration of therapy? Third, while preventing stent thrombosis using oral antiplatelet agents is an important consideration, the frequency and lethality of this condition have decreased recently. Conversely, stroke incurs a major risk of death or permanent disability; therefore, stroke prevention is the first objective of therapy in this setting. Therefore, OAC remains the foundation therapy to which oral antiplatelet therapy needs to be added. Importantly, the data from randomised clinical trials3 and from observational series4 indicate that ‘novel’ OACs are clearly associated with a lower risk of intracranial haemorrhage than vitamin K antagonists, and generally with a similar or lower risk of major bleeding, although there does appear to be a signal of greater risk for gastrointestinal haemorrhage. Therefore, ‘novel’ anticoagulants should be, whenever possible, preferred over vitamin K antagonists. Of course, this recommendation does not apply to patients with advanced chronic kidney disease, in whom vitamin K antagonists remain the preferred anticoagulation option, due to a greater history of use.5 If vitamin K antagonists are used, it appears reasonable to target the low end of the therapeutic range as long as combination therapy with antiplatelet agents is maintained, in order to minimise the risk of bleeding. Conversely, clinical trials and observational studies converge in showing an increased risk of bleeding with the ‘novel’ P2Y12 receptor antagonists, ticagrelor and prasugrel, compared with clopidogrel. Therefore, clopidogrel should be preferred over these new antiplatelet agents when used as part of triple antithrombotic therapy to minimise the risk of bleeding, at least until the optimal regimens of novel antiplatelets and anticoagulants have been clarified from new trials such as RE-DUAL (NCT02164864) and PIONEER AF-PCI.6 With respect to aspirin, there is no evidence that using moderate to high doses is associated with improved efficacy. Conversely, there is evidence from the CURRENT trial that a dose of ⩽100 mg of aspirin is associated with a reduced risk of gastrointestinal bleeding.7 There are solid data to document the increased risk of bleeding related to combining DAPT with OAC8,9 and, therefore, there is a logical basis to shorten the duration of such triple therapy as much as possible to the combination of a single antiplatelet agent with OAC. In this setting, existing data suggest that P2Y12 inhibition is the most important target to prevent stent thrombosis, rather than inhibition of the cyclooxygenase pathway with aspirin. In the early bare metal stent (BMS) era, it was not until ticlopidine was added to aspirin that rates of stent thrombosis were low enough to adopt stents into routine practice. In addition, the pioneering WOEST trial showed that dropping aspirin from the combination antithrombotic regimen early after BMS or drug-eluting stent (DES) placement was associated with a marked reduction in major bleeding rates, without any excess in acute thrombotic events or stent thrombosis rates and, in fact, lower rates of thrombotic events.10 In the WOEST trial, all-cause mortality was actually lower with combination OAC and clopidogrel after stenting than with the ‘conventional’ triple therapy combining OAC, aspirin and clopidogrel (2.6% vs. 6.4%, P=0.027), although this modest-sized trial was not powered for either mortality or cardiovascular events and this may represent a chance finding. Note, however, that the WOEST trial enrolled over two-thirds of non-ACS patients and that the combination of aspirin and oral P2Y12 inhibition remains the standard of care for such patients.11–15 Another recent trial, ISAR-TRIPLE, compared the net clinical outcome (combining ischaemic and bleeding events) with 6 weeks versus 6 months of clopidogrel as part of triple therapy, combining DAPT and OAC. ISAR-TRIPLE did not find a reduction in the net clinical outcome with 6 weeks versus 6 months of clopidogrel, although there was an increase in more sensitive measures of bleeding with the longer duration of triple therapy.16,17 These findings are consistent with the observation that the risk of stent thrombosis is greatest in the first days and weeks following placement and rapidly decreases to a steady but low rate of event accrual, suggesting that minimising the period of combined triple therapy to 4 or 6 weeks is probably not only safe but wise. Conversely, there is accumulating evidence demonstrating that with the newer generations of DES, the risk of late stent thrombosis has diminished to be lower than that with BMS,18,19 and the required duration of antiplatelet therapy is lower than previously thought.20 The DAPT trial21 has recently shown that in patients receiving DES, 30 months of clopidogrel was superior to 12 months, when added to aspirin, in preventing both stent thrombosis and major adverse cerebral and cardiovascular events. However, the DAPT trial, like all recent trials testing various durations of antiplatelet therapy after stenting, excluded patients with an indication for OAC, and focused on the duration of clopidogrel rather than the use of aspirin. Importantly, there was an absolute risk of approximately 1% of stent thrombosis at the time of clopidogrel discontinuation, regardless of whether this occurred ‘early’ (at 12 months) or ‘late’ (30 months), suggesting that the risk of thrombotic events related to the discontinuation of clopidogrel was (at least partly) delayed rather than abolished with a longer treatment duration. Also, the greatest benefit of extending the duration of DAPT appears to be in patients with ACSs.22,23 Therefore, it seems possible to offer a simple, unique, three-stage regimen to prevent stroke and stent thrombosis in AF patients receiving stents (Figure 1, Table 2). Schematic representation of the proposed algorithm NOAC: non-vitamin K antagonist oral anticoagulant; PCI: percutaneous coronary intervention; PPI: proton pump inhibitor. Single simplified antithrombotic regimen for bare-metal and second-generation drug-eluting stents placed in patients requiring anticoagulation for non-valvular atrial fibrillation The duration of stage II can be modulated from 3 to 12 months depending on the patient and procedural characteristics. NOAC: non-vitamin K antagonist oral anticoagulant; PCI: percutaneous coronary intervention; PPI: proton pump inhibitor. Single simplified antithrombotic regimen for bare-metal and second-generation drug-eluting stents placed in patients requiring anticoagulation for non-valvular atrial fibrillation The duration of stage II can be modulated from 3 to 12 months depending on the patient and procedural characteristics. NOAC: non-vitamin K antagonist oral anticoagulant; PCI: percutaneous coronary intervention; PPI: proton pump inhibitor. Stage I: At the time of PCI and up to 1 month afterwards, combine aspirin, clopidogrel and a non-vitamin K antagonist oral anticoagulant (NOAC). This ensures that the procedural thrombotic risk is reduced to a minimum, even if the patient is undergoing PCI in the context of an ACS. If the oral anticoagulant chosen has a lower dose range approved for clinical use (e.g., dabigatran 110 mg twice a day or edoxaban 30 mg), the lower range should be considered. While receiving triple therapy, patients should be treated with a proton pump inhibitor to reduce the risk of gastrointestinal bleeding.17 Stage II: From 1-month follow-up to 6 months post PCI, use the combination of clopidogrel and NOAC. The duration of this phase of double therapy combining anticoagulation and antiplatelet therapy may be modulated from 3 to 12 months according to the patient and procedural characteristics (e.g., a post ACS patient or a patient with high-risk features for stent thrombosis, such as placement of multiple stents of a low diameter would be a candidate for more protracted therapy, whereas a patient treated for stable coronary artery disease with a single second-generation DES of a large diameter may be a good candidate for a shorter duration). Stage III: Use NOAC alone for an indefinite duration. For stage III, there is ongoing uncertainty regarding the benefits and risks of adding an antiplatelet agent as chronic therapy in patients with AF receiving OAC. The addition of an antiplatelet is certain to increase bleeding risk and its benefits in terms of the prevention of ischaemic events are unproved. Major variations exist in routine clinical care, often driven by local practice. In many countries, aspirin is routinely added to OAC, whereas physicians in other countries only use OAC. Given this uncertainty, and pending the availability of robust clinical trial data, it seems reasonable to adopt a ‘first do no harm’ approach, restricting the use of antiplatelet agents and preferring OAC alone. This recommendation may need to be revised if new evidence accrues in this setting. There are obvious and important limitations to this proposal: first, it is no more evidence-based than the many existing consensus documents. Nor does it claim in any way to be a guideline or represent anything else than the opinion of its authors. Second, it does not apply to patients requiring OAC with vitamin K antagonists because of prosthetic heart valves or other indications outside of non-valvular AF, nor is it really applicable to patients who have had a stent placed and subsequently develop an indication for anticoagulation. Third, with clopidogrel being used as a single antiplatelet agent after PCI, there may be a concern that poor responders to clopidogrel may be insufficiently protected against stent thrombosis. However, there is often a disconnect between high persistent platelet reactivity on clopidogrel and actual clinical events. Finally, this proposal has not been tested formally and, in fact, this is unlikely to occur given the enrolment difficulties of testing various antithrombotic strategies in this setting. However, we believe that, pending the availability of hard evidence from large ongoing trials (such as RE-DUAL and PIONEER AF-PCI),6 which are unlikely to be released for several years, it provides a framework that has the merits of simplicity and safety for helping clinicians in this complex setting. It provides a single algorithm, easy to remember, applicable to a broad range of patients (stable or unstable) and stent types (both BMS and DES). As the evidence base available to guide clinical decisions in this setting is limited, should we not at least have a simple option? Dr Ph Gabriel Steg discloses the following relationships: research grant (to INSERM U1148) from Sanofi and Servier; speaking or consulting fees from Amarin, AstraZeneca, Bayer, Boehringer-Ingelheim, Bristol-Myers-Squibb, CSL-Behring, Daiichi-Sankyo, GlaxoSmithKline, Janssen, Lilly, Novartis, Pfizer, Regeneron, Roche, Sanofi, Servier, The Medicines Company; and owns stocks in Aterovax. Dr Deepak L Bhatt discloses the following relationships: advisory board: Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology, Regado Biosciences; board of directors: Boston VA Research Institute, Society of Cardiovascular Patient Care; chair: American Heart Association Get With The Guidelines Steering Committee; data monitoring committees: Duke Clinical Research Institute, Harvard Clinical Research Institute, Mayo Clinic, Population Health Research Institute; honoraria: American College of Cardiology (senior associate editor, Clinical Trials and News, ACC.org), Belvoir Publications (editor in chief, Harvard Heart Letter), Duke Clinical Research Institute (clinical trial steering committees), Harvard Clinical Research Institute (clinical trial steering committee, including RE-DUAL), HMP Communications (editor in chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (associate editor), Population Health Research Institute (clinical trial steering committee, including COMPASS), Slack Publications (chief medical editor, Cardiology Today’s Intervention), WebMD (CME steering committees); other: Clinical Cardiology (deputy editor); research funding: Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Forest Laboratories, Ischemix, Medtronic, Pfizer, Roche, Sanofi Aventis, St Jude Medical, The Medicines Company; trustee: American College of Cardiology; unfunded research: FlowCo, PLx Pharma, Takeda. Sophie Rushton-Smith (Medlink Healthcare Communications Ltd.) provided editorial assistance with formatting and manuscript preparation and was funded by INSERM U1148.