Treatment discontinuations with new oral agents for long-term anticoagulation: insights from a meta-analysis of 18 randomized trials including 101,801 patients

Mayo Clin Proc. 2014 Jul;89(7):896-907. doi: 10.1016/j.mayocp.2014.01.030.


Objective: To systematically examine discontinuation rates with new US Food and Drug Administration-approved oral anticoagulants (NOACs) in patients with various indications for long-term anticoagulation.

Patients and methods: Poor adherence to medications is considered a potential and frequent cause of treatment failure. We searched the PubMed, Cochrane Central Register of Controlled Trials, EMBASE, EBSCO, Web of Science, and CINAHL databases for articles published from January 1, 2001, through September 15, 2013. The following Medical Subject Heading terms and/or keywords were used for our database searches: rivaroxaban, dabigatran, apixaban, new oral anticoagulants, oral thrombin inhibitors, and oral factor Xa inhibitors. Articles in English that focused on randomized controlled trials (RCTs) comparing NOACs (apixaban, dabigatran, and rivaroxaban) with conventional therapy or placebo were abstracted. Independent extraction of relevant data was performed by 2 authors. The primary end point of interest was discontinuation due to all causes. Other end points of interest were discontinuation due to adverse events, consent withdrawal, and nonadherence.

Results: Eighteen RCTs including a total of 101,801 patients were included for analysis. Total study drug discontinuation rates were not statistically different with NOACs in comparison to pharmacologically active comparators for treatment of venous thromboembolism/pulmonary embolism (risk ratio [RR], 0.91; 95% CI, 0.74-1.13; P=.40) and for NOACs in comparison to warfarin and aspirin for prevention of stroke in patients with atrial fibrillation (RR, 1.01; 95% CI, 0.87-1.17; P=.92). In contrast, in acute coronary syndromes, total study drug discontinuation with NOACs was significantly higher than with placebo (RR, 1.40; 95% CI, 1.07-1.83; P=.01). Overall discontinuations were comparable to those with active comparators.

Conclusion: Study drug discontinuations with NOACs were not significantly different from those with conventional drugs in treatment of venous thromboembolism/pulmonary embolism and prevention of stroke in patients with atrial fibrillation but were worse in acute coronary syndromes as noted in evidence from contemporary RCTs.

Publication types

  • Meta-Analysis

MeSH terms

  • Acute Coronary Syndrome / drug therapy*
  • Administration, Oral
  • Anticoagulants / therapeutic use*
  • Benzimidazoles / therapeutic use
  • Dabigatran
  • Humans
  • Medication Adherence / statistics & numerical data*
  • Morpholines / therapeutic use
  • Pulmonary Embolism / drug therapy*
  • Pyrazoles / therapeutic use
  • Pyridones / therapeutic use
  • Randomized Controlled Trials as Topic
  • Rivaroxaban
  • Stroke / prevention & control*
  • Thiophenes / therapeutic use
  • Venous Thromboembolism / drug therapy*
  • beta-Alanine / analogs & derivatives
  • beta-Alanine / therapeutic use


  • Anticoagulants
  • Benzimidazoles
  • Morpholines
  • Pyrazoles
  • Pyridones
  • Thiophenes
  • beta-Alanine
  • apixaban
  • Rivaroxaban
  • Dabigatran