ARISTOTLE: Major Bleeding in Patients With Atrial Fibrillation Receiving Apixaban or Warfarin

Editor's Note: Commentary based on Hylek EM, Held C Alexander JH, et al. Major bleeding in patients with atrial fibrillation receiving apixaban or warfarin: The ARISTOTLE Trial (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation): Predictors, Characteristics, and Clinical Outcomes. J Am Coll Cardiol 2014;63:2141-7.


The ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation) trial randomized patients with atrial fibrillation and at least one risk factor for stroke to receive warfarin or apixaban, an oral factor Xa inhibitor. Apixaban was superior to warfarin with regard to the primary outcome of ischemic or hemorrhagic stroke or systemic embolism, and assignment to the apixaban arm was associated with a 31% reduction in the risk of major hemorrhage.1 In this new analysis from Hylek et al.,2 the authors seek to further characterize major bleeding events in the ARISTOTLE trial with respect to severity, predictors, anatomic location, and components of the major bleeding definition.


In the ARISTOTLE trial, patients received either warfarin with a target international normalized ratio (INR) of 2-3 or apixaban at a dose of 5 mg twice a day. Apixaban was administered at a lower dose of 2.5 mg twice a day to participants ≥80 years of age, weighing ≤60 kg, or with a serum creatinine ≥1.5 mg/dL. Major bleeding was defined in accordance with the criteria from the International Society on Thrombosis and Haemostasis (ISTH). All patients who received at least one dose of study drug were included in the analysis by Hylek et al. Additional bleeding parameters analyzed included anatomic location, decrease of hemoglobin of at least 2 g/dL, hospitalization because of bleeding, red cell transfusion, medical or surgical consultation, hemodynamic compromise, change in antithrombotic therapy, and 30-day mortality following ISTH major hemorrhage.


The ARISTOTLE trial enrolled 18,201 patients with an on-treatment safety population of 18,140 and a mean follow-up of 20.5 months. Major hemorrhage occurred in 4.3% of patients with a rate of 2.13% per year in the apixaban group and 3.09% per year in the warfarin group (hazard ratio [HR] 0.69; p <0.001). The following factors were associated with increased risk for ISTH major hemorrhage: older age, prior myocardial infarction, prior hemorrhage, impaired renal function, fall within the previous year, lower weight, prior stroke or transient ischemic attack (TIA), diabetes, hypertension, and the use of aspirin, clopidogrel, non-steroidal anti-inflammatory drugs, statins, or antacids. There were fewer bleeding events among females.

As compared to warfarin, apixaban was associated with fewer soft tissue hematomas (HR 0.46, 95% confidence interval [CI]: 0.29-0.74), fewer trauma-related hemorrhages (HR 0.60, 95% CI: 0.40-0.91), and fewer gastrointestinal hemorrhages, though the latter did not reach statistical significance (HR 0.89, 95% CI: 0.70-1.14). Apixaban was associated with fewer intracranial hemorrhages (HR 0.42, 95% CI: 0.30-0.58), as was previously reported with the ARISTOTLE trial.

Among patients with major extra-cranial hemorrhage, there were fewer adverse consequences among patients assigned to apixaban as measured by number of hospitalizations, medical or surgical interventions, transfusions, and changes in antithrombotic therapy. Each of these individual measures reached statistical significance for difference between the groups in the Hylek et al. analysis. The 30-day mortality rate following a major bleeding event was significantly lower in the apixaban group (HR 0.50, 95% CI: 0.33-0.74).

An analysis of subgroup by treatment interactions found that patients with renal dysfunction and low body weight had an even greater degree of reduced bleeding with apixaban as compared to warfarin. Among patients with diabetes, the reduction in bleeding events with apixaban was less pronounced.


Rates of hemorrhage and subsequent complications are lower among patient with non-valvular atrial fibrillation treated with apixaban as compared to warfarin.


While the target-specific oral anticoagulants are no longer "novel," with the approval of the first occurring nearly four years ago,3 their appropriate role in clinical practice is still being defined. Critical to appropriate patient selection is a nuanced understanding of the bleeding risk associated with each agent. As such, Hylek et al.'s analysis provides important confirmation and additional insight into the previously reported favorable hemorrhage profile of apixaban as compared to warfarin in the ARISTOTLE trial. The authors report a 31% reduction in first major ISTH hemorrhage, fewer intracranial hemorrhages, fewer adverse consequences of extra-cranial hemorrhage, and a 50% reduction in 30-day mortality following a major hemorrhage with apixaban. In addition, apixaban was associated with fewer less serious hemorrhages, such as soft tissue hematomas and hemorrhages related to trauma. These hemorrhages, while less likely to be fatal, are important to avoid since patients who experience bleeding are more likely to interrupt or stop anticoagulant therapy. Such interruption in anticoagulation may then lead to serious thromboembolic events, some of which may be fatal.

Importantly, these associations remained consistent in subgroup analysis. In fact, apixaban had an even greater safety profile among patients whom most clinicians would consider at higher risk from apixaban – those with low body weight and impaired renal function. The authors also acknowledge the finding of a lower hemorrhage rate among women, which deserves further investigation. Also of interest was the numerically fewer GI hemorrhages with apixaban – a finding that distinguishes apixaban from dabigatran 150 mg BID (the currently approved dose in the U.S.) and rivaroxaban.

This analysis is limited to data from a single trial and ongoing post-market surveillance will provide critical additional safety information. Comparisons among the target-specific agents are also needed. Even so, this report from Hylek et al. contributes importantly to the understanding of apixaban's safety profile relative to warfarin.


  1. Granger CB, Alexander JH, McMurry JJV, et al. Apixaban versus warfarin in patients with atrial fibrillation. NEJM. 2011;365:981-92.
  2. Hylek EM, Held C Alexander JH, et al. Major bleeding in patients with atrial fibrillation receiving apixaban or warfarin: The ARISTOTLE Trial (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation): Predictors, Characteristics, and Clinical Outcomes. J Am Coll Cardiol 2014;63:2141-7.
  3. U.S. Food and Drug Administration. FDA Approves Pradaxa to Prevent Stroke in People With Atrial Fibrillation (FDA Website). 2010. Available at: Accessed 08/06/2013.

Keywords: Anticoagulants, Atrial Fibrillation, Body Weight, Double-Blind Method, International Normalized Ratio, Myocardial Infarction, Renal Insufficiency, Stroke

< Back to Listings