What is the best approach to anticoagulation for patients with nonvalvular atrial fibrillation who also take low-dose aspirin?

The balance of benefits and harms favors novel oral anticoagulants (NOACs) over warfarin for patients with nonvalvular atrial fibrillation (NVAF) who require anticoagulation and are already taking low-dose aspirin. It is worth noting that for low-risk patients with NVAF, aspirin alone is an option. With regard to the choice of NOAC, since these patients were largely taking aspirin as secondary prevention for cardiovascular disease, edoxaban was least likely to increase the risk of myocardial infarction. (LOE = 1a)

Bennaghmouch N, de Veer AJWM, Bode K, et al. Efficacy and safety of the use of non-vitamin K antagonist oral anticoagulants in patients with nonvalvular atrial fibrillation and concomitant aspirin therapy a meta-analysis of randomized trials. Circulation 2018;137(11):1117-1129.  [PMID:29101289]

Meta-analysis (randomized controlled trials)

Foundation

Outpatient (any)

Many patients with atrial fibrillation have a separate indication for aspirin. This meta-analysis performed a thorough search of several databases and identified 4 randomized trials with a total of 21,722 patients who had NVAF and were taking antiplatelet therapy (most commonly low-dose aspirin) for cardiovascular prevention. Each of the studies randomized patients to receive either warfarin or a NOAC such as edoxaban, apixaban, rivaroxaban, or dabigatran. The studies had between 1.8 and 2.8 years of follow-up. The mean age of included patients was between 70 years and 72 years, approximately one-third were women, and between 10% and 55% had experienced a previous stroke. After performing a random effects meta-analysis, the authors found that patients randomized to receive NOACs were less likely to experience a stroke or systemic embolism (hazard ratio [HR] 0.78; 95% CI 0.67 - 0.91) or vascular death (HR 0.85; 0.76 - 0.93) than those randomized to receive warfarin. There was a trend toward a higher risk of myocardial infarction in the NOAC group, primarily driven by the one dabigatran trial (HR 1.2; 0.97 - 1.4), but a trend toward fewer major hemorrhages with NOACs (HR 0.83; 0.69 - 1.01). Patients randomized to receive a NOAC were significantly less likely to experience an intracranial hemorrhage (HR 0.38; 0.26 - 0.56). There was minimal to moderate heterogeneity for most outcomes, although the measure used (I2) is unreliable with only 4 studies