[Skip to Navigation]
Original Investigation
May 25, 2022

Apixaban or Warfarin and Aspirin or Placebo After Acute Coronary Syndrome or Percutaneous Coronary Intervention in Patients With Atrial Fibrillation and Prior Stroke: A Post Hoc Analysis From the AUGUSTUS Trial

Author Affiliations
  • 1INECO Neurociencias Oroño, Fundación INECO, Rosario, Santa Fe, Argentina
  • 2UPMC Heart and Vascular Institute, Harrisburg, Pennsylvania
  • 3Duke Clinical Research Institute, Duke University School of Medicine, Durham, North Carolina
  • 4Canadian VIGOUR Centre, University of Alberta, Edmonton, Alberta, Canada
  • 5Terrence Donnelly Heart Center, St Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada
  • 6Bristol Myers Squibb, Lawrenceville, New Jersey
  • 7Department of Neurology, Duke University School of Medicine, Durham, North Carolina
  • 8Department of Neurosurgery and Brain Repair, University of South Florida, Tampa
  • 9Carol Davila University of Medicine and Pharmacy, University and Emergency Hospital, Bucharest, Romania
  • 10Department of Cardiology, Oslo University Hospital, Oslo, Norway
  • 11Hospital Israelita Albert Einstein, São Paulo, Brazil
  • 12Swiss Cardiovascular Center, Bern, Switzerland
  • 13Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York
  • 14Cardiovascular Research Foundation, New York, New York
JAMA Cardiol. 2022;7(7):682-689. doi:10.1001/jamacardio.2022.1166
Key Points

Question  In patients with atrial fibrillation and acute coronary syndrome and/or percutaneous coronary intervention receiving a P2Y12 inhibitor, does treatment effect vary according to prior stroke, transient ischemic attack (TIA), or thromboembolism (TE)?

Findings  In this randomized clinical trial, a post hoc analysis found that patients with vs without prior stroke/TIA/TE had a higher risk of ischemic and bleeding events. Among those with prior stroke/TIA/TE, patients treated with apixaban had less bleeding, death, or hospitalization than patients treated with vitamin K antagonists; patients treated with aspirin had a higher bleeding risk than those receiving placebo.

Meaning  The safety and efficacy of apixaban compared with vitamin K antagonists were consistent with the overall trial findings, irrespective of history of stroke/TIA/TE.


Importance  Data are limited regarding the risk of cerebrovascular ischemic events and major bleeding in patients with atrial fibrillation (AF) and recent acute coronary syndrome (ACS) and/or percutaneous coronary intervention (PCI).

Objective  Determine the efficacy and safety of apixaban or vitamin K antagonists (VKA) and aspirin or placebo according to prior stroke, transient ischemic attack (TIA), or thromboembolism (TE).

Design, Setting, and Participants  In this prospective, multicenter, 2-by-2 factorial, randomized clinical trial, post hoc parallel analyses were performed to compare randomized treatment regimens according to presence or absence of prior stroke/TIA/TE using Cox proportional hazards models. Patients with AF, recent ACS or PCI, and planned use of P2Y12 inhibitors for 6 months or longer were included; 33 patients with missing data about prior stroke/TIA/TE were excluded.

Interventions  Apixaban (5 mg or 2.5 mg twice daily) or VKA and aspirin or placebo.

Main Outcomes and Measures  Major or clinically relevant nonmajor (CRNM) bleeding.

Results  Of 4581 patients included, 633 (13.8%) had prior stroke/TIA/TE. Patients with vs without prior stroke/TIA/TE were older; had higher CHA2DS2-VASC and HAS-BLED scores; and more frequently had prior bleeding, heart failure, diabetes, and prior oral anticoagulant use. Apixaban was associated with lower rates of major or CRNM bleeding and death or hospitalization than VKA in patients with (hazard ratio [HR], 0.69; 95% CI, 0.46-1.03) and without (HR, 0.68; 95% CI, 0.57-0.82) prior stroke/TIA/TE. Patients without prior stroke/TIA/TE receiving aspirin vs placebo had higher rates of bleeding; this difference appeared less substantial among patients with prior stroke/TIA/TE (P = .01 for interaction). Aspirin was associated with numerically lower rates of death or ischemic events than placebo in patients with (HR, 0.71; 95% CI, 0.42-1.20) and without (HR, 0.93; 95% CI, 0.72-1.21) prior stroke/TIA/TE (not statistically significant).

Conclusions and Relevance  The safety and efficacy of apixaban compared with VKA was consistent with the AUGUSTUS findings, irrespective of prior stroke/TIA/TE. Aspirin increased major or CRNM bleeding, particularly in patients without prior stroke/TIA/TE. Although aspirin may have some benefit in patients with prior stroke, our findings support the use of apixaban and a P2Y12 inhibitor without aspirin for the majority of patients with AF and ACS and/or PCI, regardless of prior stroke/TIA/TE status.

Trial Registration  ClinicalTrials.gov Identifier: NCT02415400

Add or change institution