By reasonable estimates, antithrombotic therapy is the most common acquired bleeding disorder. Aspirin is a common yet potent antiplatelet agent used for secondary cardiovascular disease (CVD) prevention and now less commonly for primary prophylaxis. As an over-the-counter (seemingly benign) medication, use of aspirin (for CVD or analgesia) may be underdocumented, under-reported, and perhaps overlooked when patients report medications used to their physicians. Aspirin use is so ingrained in the history of CVD prevention that it sometimes feels impossible to stop aspirin in a patient with known arterial disease, even when current data demonstrate lack of benefit or even harm by continuing it. In patients in whom aspirin has been stopped appropriately, well-intentioned clinical “best practice advisories” built into electronic medical records may inappropriately encourage providers to restart aspirin in patients with stable CVD.
The AFIRE study investigated the risks and benefits of antiplatelet agents in anticoagulated patients.1 The study was a multicenter (294 centers), open-label, randomized controlled trial completed between 2015 and 2017 in patients with nonvalvular atrial fibrillation (CHADS2 score ≥1) with stable coronary artery disease (CAD), defined as a history of percutaneous coronary intervention or coronary artery bypass grafting more than a year before, or angiographically confirmed CAD (>50% stenosis) not requiring intervention. Patients were randomized to rivaroxaban monotherapy (10 mg once daily for creatinine clearance 15-50 or 15 mg once daily for creatinine clearance ≥ 50; dosing approved in Japan) or combination therapy with rivaroxaban and an antiplatelet agent (aspirin or P2Y12 inhibitor). The primary outcome in this post hoc secondary analysis was total number of first and subsequent events (bleeding, thrombotic events, and death) and was adjudicated by an independent committee.
Researchers included 2,215 of the 2,240 enrolled patients in the modified intention-to-treat analysis. Baseline characteristics and comorbidities were well balanced between treatment groups. Among patients in the combination therapy group with rivaroxaban, 70 percent were on aspirin and 27 percent were on a P2Y12 inhibitor. The total event rate for rivaroxaban monotherapy was 12.2 percent compared to 19.2 percent for combination therapy. Rivaroxaban monotherapy lowered the risk for first (relative risk, 0.69; 95% CI, 0.55-0.87) and subsequent events (relative risk, 0.46; 95% CI, 0.29-0.74) compared to combination therapy. Mortality was also lower with monotherapy (3.7% vs. 6.6%). As might be expected, the incidence of thrombotic events was higher than bleeding and death in patients with monotherapy, and bleeding was higher than thrombotic events with combination therapy. However, in a time-adjusted multivariable Cox regression analysis, monotherapy with rivaroxaban was associated with a lower risk of death when adjusting for first nonfatal bleeding (hazard ratio, 0.57) or first nonfatal thrombotic event (hazard ratio, 0.61). Even among patients with a first nonfatal thrombotic event, the majority of subsequent fatal events were bleeding (75% for rivaroxaban monotherapy and 62% in combination group). The authors concluded that in patients with stable CAD on anticoagulation for nonvalvular atrial fibrillation, rivaroxaban monotherapy was superior to combination therapy.
In Brief
This secondary post hoc analysis of the AFIRE study adds to the strength of the literature for antiplatelet therapy reduction or discontinuation in anticoagulated patients with stable CAD. The improved safety of a single antiplatelet therapy over dual antiplatelet therapy after percutaneous coronary intervention has previously been demonstrated in the PIONEER AF-PCI trial with rivaroxaban, the RE-DUAL PCI trial with dabigatran, the ENTRUST-AF PCI trial with edoxaban, and the AUGUSTUS trial with apixaban. The European Society of Cardiology guidelines, developed with the special contribution of the European Heart Rhythm Association,2 recommends patients requiring anticoagulation for atrial fibrillation to use mono or dual antiplatelet therapy for six to 12 months after percutaneous coronary intervention, and then monotherapy with anticoagulation only (irrespective of the stent type) provided that there were no recurrent ischemic events in the interim.
Although the specific patient population and rivaroxaban dose in the AFIRE study may limit the generalizability, the findings are consistent with the previous literature and in fact align with current guidelines in Europe and the United States. It is unclear whether the results of this study can be extrapolated to a patient population with atherosclerotic vascular disease and venous thromboembolism (with or without associated malignancy) who are initiating anticoagulation. Although limited, similar data do exist in the setting of anticoagulation and antiplatelet therapy in patients with venous thromboembolism.3 When hematologists contribute to the care of bleeding patients on combination antithrombotic therapies or initiate anticoagulation, we must think beyond “stop the clot” and “fear the bleed” and its associated morbidity.
Competing Interests
Dr. Houghton and Dr. Wysokinski indicated no relevant conflicts of interest.
