Recent trials of patients with atrial fibrillation (AF) have convincingly shown that anticoagulation with non–vitamin K antagonists or so-called novel oral anticoagulants (NOACs) are at least as effective as oral vitamin K antagonism by warfarin in reducing the risk of cardioembolic stroke and systemic embolism, but with a lower risk of serious bleeding events, in particular, intracranial hemorrhage. 1
Trial results have been supported by large registry studies in broad populations. AF frequently occurs in patients with underlying heart disease entities. In fact, up to one-third of all patients with AF have an underlying stable coronary artery disease, undergo coronary stenting, or will develop an acute coronary syndrome necessitating treatment with single- or 2-antiplatelet agents.
The intersection between the 2 disease entities, AF and coronary artery disease, has gained a lot of attention and clinical and scientific interest recently. In patients with a recent acute coronary syndrome or stent procedure with or without AF, dual-antiplatelet therapy has been advocated to protect patients from stent thrombosis and recurrent ischemic events. However, adding anticoagulation for stroke prevention on top of antiplatelet therapy may increase the risk of severe or life-threatening bleeding complications to a degree that the net clinical benefit may get lost.
International clinical guideline committees have therefore recommended the shortest possible duration of triple therapy with dual-antiplatelet therapy in addition to anticoagulation. Recently, new treatment alternatives with apixaban 2 or rivaroxaban 3 in addition to single-antiplatelet therapy with a P2Y12 receptor inhibitor have been shown to be safer than triple therapy with 2 antiplatelet agents in addition to warfarin.
However, no trial so far has had a sufficiently large sample size to evaluate the efficacy of such a strategy. More evidence is needed to understand the optimal combination of anticoagulation and antiplatelet therapy to reduce the risk of cardioembolic stroke, the risk of systemic embolism from AF, and, at the same time, the risk of stent-related and spontaneous coronary ischemic events with an acceptable bleeding risk.
Recently, in patients with stable atherosclerotic disease but without AF, a dual-pathway inhibition with very-low-dose rivaroxaban in combination with low-dose aspirin has been shown to reduce ischemic events with an acceptable bleeding risk in comparison with low-dose aspirin alone. 4
In light of the ongoing clinical and scientific uncertainty about the optimal treatment of patients with AF and also a clinical indication for antiplatelet therapy, and the new interest in dual-pathway inhibition, as well, the meta-analysis by Bennaghmouch et al 5 in this issue of Circulation adds a new piece to the jigsaw puzzle of understanding the complex interaction between atherothrombosis/stent thrombosis in coronary arteries and stroke or systemic embolism resulting from cardiac thromboembolism, necessitating the inhibition of platelet activation, and coagulation, as well.
The authors report the results of a very well conducted meta-analysis including patients with AF on aspirin and randomly assigned to either a vitamin K antagonist or a NOAC. In the 4 major, randomized trials of different NOACs in patients with AF almost 22 000 (30%) patients were noted as receiving aspirin therapy. In these patients, NOAC in comparison with vitamin K antagonist was associated with a consistent benefit with a 20% lower risk for stroke or systemic embolism, 15% lower risk of vascular death, no significant difference in ischemic stroke, but a numerically lower risk of major bleeding and a highly significant 60% lower risk of intracranial hemorrhage.
These data support the guideline recommendations to always favor NOAC over vitamin K antagonist, but, in particular, in patients at high bleeding risk, also including patients with concomitant aspirin therapy. 6 It is interesting to note that the risk of myocardial infarction was numerically higher with NOAC versus vitamin K antagonist, although with a very low absolute risk in both groups. This finding indicates a need for further research on the optimal antithrombotic strategies in patients with AF who have had a recent myocardial infarction or are at high risk for stent thrombosis and recurrent infarction.
As with all meta-analyses, this analysis also has multiple inherent limitations that are appropriately acknowledged by the authors. One of the most important limitations is the lack of information about the duration and net exposure of aspirin during the anticoagulation treatment period.
Other important limitations are the lack information about other platelet inhibitors and if there are differences between different types of NOACs or different doses. It is important to note that the study does not provide any information about whether aspirin in itself is needed or provides any benefit in addition to oral anticoagulation.
Furthermore, the study does not provide any direct insights in the treatment of patients with AF in combination with a recent acute coronary syndrome or stent procedure requiring an initial phase of dual-antiplatelet therapy. It is unlikely that any dedicated large-scale randomized trial will be performed in this patient population, and, thus, a study-level meta-analysis is the best evidence that can be generated.
The present meta-analysis provides us with clinically important information that, in patients with a clinical indication for aspirin because of stable coronary artery disease, anticoagulation with a non–vitamin K antagonist may be more effective than a vitamin K antagonist with warfarin and have a lower risk of intracranial bleeding.