brand logo

Am Fam Physician. 2004;70(5):917-918

Searchable Question

Which patients with atrial fibrillation do not need anticoagulation therapy with warfarin?

Evidence-Based Answer

Anticoagulation therapy with warfarin is not indicated for use in patients with non-valvular atrial fibrillation who are at low risk for embolic cerebral vascular events. The classification of “low risk for embolic stroke” is defined as a 1 percent annual risk for stroke or lone atrial fibrillation (i.e., age younger than 65 years without history of hypertension, transient ischemic attack, stroke, coronary heart disease, recent congestive heart failure, or diabetes).

Anticoagulation therapy with warfarin is beneficial in patients with atrial fibrillation who are at moderate or high risk for stroke. Patients with absolute contraindications to anticoagulation therapy (e.g., thrombocytopenia, recent trauma or surgery, hemorrhagic stroke, alcoholism) should not take warfarin.1,2 [Strength of recommendation: A, based on meta-analyses of large randomized controlled trials (RCTs)]

Evidence Summary

A number of meta-analyses of RCTs evaluating prevention of stroke in patients with atrial fibrillation demonstrated the benefits of warfarin therapy.1,2 The Atrial Fibrillation Investigators pooled data from five large RCTs covering 3,691 patient-years. The annual rate of stroke was found to be 4.5 percent among the control groups and 1.4 percent in the warfarin groups. The absolute risk reduction (RR) was 3.1 percent (95 percent confidence interval [CI], 2.3 to 3.6 percent), and the number needed to treat (NNT) was 32.1

A 1999 meta-analysis of 16 RCTs covering 16,785 patient-years showed that the placebo groups had a 6 percent annual stroke rate compared with a 2.2 percent annual stroke rate in the warfarin groups. The absolute RR was 3.8 percent (95 percent CI, 2.9 to 4.3 percent), and the NNT was 26.2

Both meta-analyses reported slightly higher annual rates of major bleeding events among patients taking warfarin (1.3 percent) compared with those taking placebo (1 percent) or aspirin (1 percent).1,2

The meta-analyses showed aspirin therapy had a modest stroke risk reduction compared with placebo. In the first meta-analysis, the absolute RR was 1.6 percent (95 percent CI, 0.2 to 2.6 percent), and the NNT was 62.1. In the second meta-analysis, the absolute RR was 1.7 percent (95 percent CI, 0.2 to 3.0 percent), and the NNT was 59.2

Risk stratification and subgroup analysis of data from the Atrial Fibrillation Investigators showed that warfarin therapy has limited benefit in patients who have atrial fibrillation and no other risk factors. Patients younger than 65 years without other risk factors had an annual risk for stroke of 1 percent whether they were taking warfarin or not. In patients with higher risk for embolic cerebral vascular accidents, the relative RR remained constant, and the absolute RR increased with the annual stroke rate. Stroke risk was reduced from an untreated rate ranging from 4.3 to 12 percent annually to a treated rate of 1.1 to 4 percent annually.3

Patients with absolute contraindications to warfarin anticoagulation therapy (e.g., thrombocytopenia, recent trauma or surgery, hemorrhagic stroke, alcoholism) should not receive warfarin. These patients were systematically excluded from all RCTs of warfarin and atrial fibrillation.

Based on this evidence, warfarin therapy is not indicated for low-risk patients but is beneficial in moderate- and high-risk patients. Defining the exact risk for stroke is the critical task in determining if a patient should take warfarin (see accompanying table).47 As the risk for stroke increases, the NNT decreases. A relatively small but statistically significant risk for hemorrhage with warfarin therapy requires that individual patient characteristics be considered.

Recommendations from Others

The American Academy of Family Physicians, the American College of Physicians,4 the American College of Chest Physicians,5 the American College of Cardiology, the American Heart Association, and the European Society of Cardiology6 have published recommendations consistent with this evidence-based answer.

ModelDefinition
Joint Panel of the American Academy of Family Physicians and the American College of Physicians4 History or presence of no more than one of the following:
  • CHF within 100 days

  • Hypertension

  • Age > 75 years

  • Diabetes

American College of Chest Physicians5 Age < 65 years and all of the following:
  • No hypertension

  • No diabetes

  • No CHD

  • No CHF

American College of Cardiology, American Heart Association, and European Society of Cardiology6 Age < 60 years, CHD but no CHF (LVEF > 0.35) and no hypertension
or
Age < 75 years, but no CHF, hypertension, CHD, or diabetes
Stroke Prevention in Atrial Fibrillation III Study7 Male or female, age < 75 years, and all of the following:
  • No hypertension

  • No recent CHF

  • No severe LV dysfunction

Clinical Commentary

In practice, the decision to anticoagulate with warfarin can be complex. The evidence supports quantifying the patient’s risk for embolic stroke. However, patients at greatest risk for stroke often have a risk for hemorrhage or difficulty complying with warfarin therapy monitoring. In some patients, other antiplatelet and antithrombotic agents such as aspirin and clopidogrel (Plavix) remain a reasonable alternative.

Continue Reading


More in AFP

Copyright © 2004 by the American Academy of Family Physicians.

This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.  See https://www.aafp.org/about/this-site/permissions.html for copyright questions and/or permission requests.