• Treatment decisions should be made on an individual basis; the person's bleeding risk, likelihood of compliance with treatment, and preferred options should always be fully assessed before starting treatment.
• Offer people with atrial fibrillation (AF):
• At low risk of stroke — aspirin.
• At high risk of stroke — warfarin.
• At moderate risk of stroke — either aspirin or warfarin.
• If uncertain about stroke risk, start aspirin whilst awaiting specialist assessment.
• CKS do not recommend the use of clopidogrel or a combination of aspirin and clopidogrel for AF in primary care.
• Risk factors have a cumulative effect on stroke risk; this should be considered when discussing treatment options. For example, if the person is in the moderate stroke risk category but has more than one risk factor for stroke (that is, more than one of hypertension, diabetes, coronary artery disease, or peripheral artery disease), there may be a stronger case for choosing warfarin over aspirin.
• For detailed information on the use of aspirin and warfarin in AF, including contraindications, starting doses and titration, monitoring, and adverse effects, see the CKS topics on Antiplatelet treatment and Anticoagulation - oral.

• The National Institute for Health and Clinical Excellence does not state that a formal bleeding risk assessment is necessary before starting antithrombotic treatment, but it provides a list of risk factors that increase the risk of bleeding with warfarin:
• More than 75 years of age.
• Use of antiplatelet drugs (such as aspirin or clopidogrel).
• Use of nonsteroidal anti-inflammatory drugs.
• Use of multiple other drugs (polypharmacy).
• Uncontrolled hypertension.
• History of bleeding (for example bleeding peptic ulcer or cerebral haemorrhage).
• History of previously poorly controlled anticoagulation therapy.
• For more information on starting warfarin, see the CKS topic on Anticoagulation - oral.

These recommendations are based on the National Institute for Health and Clinical Excellence (NICE) guideline Atrial fibrillation: national clinical guideline for management in primary and secondary care [National Collaborating Centre for Chronic Conditions, 2006].

Warfarin versus aspirin

• Good evidence indicates that warfarin is more effective than aspirin for preventing stroke or vascular events in people with atrial fibrillation (AF) but is associated with a higher risk of intracranial and extracranial bleeding.
• For people without prior stroke or transient ischaemic attack (TIA), a Cochrane systematic review (eight randomized controlled trials [RCTs]) found that warfarin reduced the risk of stroke and other major vascular events in people with AF by about one-third compared with antiplatelets [Aguilar et al, 2007].
• For people who have had a stroke or TIA, a Cochrane systematic review (two RCTs) found that warfarin reduced the risk of stroke by a half, and the risk of any vascular event by one-third, compared with antiplatelets [Saxena and Koudstaal, 2004b].
• For people 75 years of age or older, an open-label RCT (973 participants) found that warfarin (target international normalized ratio 2.0–3.0) reduced the risk of stroke by half and the risk of vascular events by two-thirds, compared with aspirin 75 mg daily [Mant et al, 2007b].

Clopidogrel alone

• NICE do not recommend the use of clopidogrel in AF. Clopidogrel is not licenced for AF and there is no trial evidence to support its use. However, a number of CKS expert reviewers do state that they would use clopidogrel as an alternative treatment option in people with a true aspirin allergy.

Combining antithrombotic treatment

• NICE states that if warfarin is offered, aspirin should not be taken concomitantly to prevent stroke, as it provides no additional benefit and may increase the risk of bleeding.
• CKS does not recommend the use of aspirin plus clopidogrel in primary care to reduce the risk of stroke in AF.
• There is evidence that warfarin significantly reduces the risk of stroke, myocardial infarction, and vascular events compared with clopidogrel plus aspirin [Connolly et al, 2006]. Warfarin did not increase the risk of bleeding compared with clopidogrel plus aspirin.
• A recently published trial found evidence that combined treatment with clopidogrel plus aspirin reduces the risk of major vascular events compared with aspirin alone [ACTIVE Investigators et al, 2009]. However, this trial also found evidence that combined treatment with clopidogrel plus aspirin was associated with a two-fold increased risk of major bleeding.