Dr Matthew Fay explains how AF-related stroke is potentially avoidable with effective patient identification and assessment, prescribing, and ongoing timely review

Atrial fibrillation should be suspected not only in people with symptoms of palpitations or breathlessness but also in people with cardiovascular comorbidity
Atrial fibrillation can be easily identified with simple pulse-checking and subsequent ECG
Once AF is identified, in the cardiovascularly-stable patient the first priority is to address the stroke risk
Anticoagulation is the recommend form of stroke prevention in people 
with AF
Aspirin monotherapy does not have a role in stroke risk reduction in people with AF
Stroke risk is the same in people with paroxysmal, persistent, permanent, symptomatic, or asymptomatic AF, or atrial flutter
Stroke risk persists even after rhythm-corrective intervention
HAS-BLED scoring should not deny patients access to anticoagulants but be used to assess how their bleeding risk can be reduced
In asymptomatic patients with heart rates below 120 bpm, there is no need for rate-controlling medication
In patients with symptomatic AF, despite rate control, prompt referral to a rhythm specialist is required.

A trial fibrillation (AF) is the commonest sustained arrhythmia, currently stated to be 1.76% in England1 and its prevalence is increasing worldwide.2 It is more common in the older population, with a prevalence of over 10% in people aged over 65 years.1

Although not always symptomatic, AF is associated with atrial thrombus formation leading to cardioembolic stroke. The stroke risk is doubled when other risk factors are taken in to account,3 and the strokes caused by AF are associated with increased levels of disability, morbidity, and mortality.4

Strokes caused by AF could often be avoided if intervention with oral anticoagulants had higher levels of uptake; however, intervention remains poor.2

The recently published NICE guideline, Atrial fibrillation: the management of atrial fibrillation (Clinical Guideline [CG]180, June 2014, available at: www.nice.org.uk/guidance/CG180) outlines a paradigm shift in the care of AF, whereby:5

  • the patient should be central to the choices made
  • the default position should now be to anticoagulate people with AF, except those people without risk factors
  • assessment should be ongoing and those people with symptomatic AF should be rapidly reviewed by specialists in the field of arrhythmias.

This article will articulate the changes to NICE recommendations in CG180 for the management of AF and the challenges, mainly to primary care, that will result.

Diagnosis and assessment

Increasingly, the management of people with AF is being undertaken outside the hospital setting, with many people not needing to see a cardiologist.

Case finding

The routine assessment of the pulse rate and rhythm is a key factor in finding people who have AF.6 Although there was no recommendation in NICE CG180 for screening for AF, the need for case identification was explored. Atrial fibrillation is more commonly found in people with cardiovascular conditions, diabetes, and chronic kidney disease.

Like the previous NICE guideline on AF, CG180 recommends performing a manual pulse palpation in people presenting with any of the following:5

  • breathlessness/dyspnoea
  • palpitations
  • syncope/dizziness
  • chest discomfort
  • stroke/transient ischaemic attack (TIA).

If an irregular pulse is found, an electrocardiogram (ECG) should be performed whether the patient has symptoms or not. If there is a suspicion that someone suffers from paroxysmal AF, then a period of either continuous ECG monitoring or cardiac event monitoring should be undertaken, depending on the frequency of the episodes.5

The indications to perform transthoracic echocardiography (TTE) following an ECG-confirmed diagnosis of AF are explored in the guideline. The need for an echocardiogram to help to understand the aetiology of the dysrhythmia, if underlying valvular or myocardial pathology is suspected, is upheld; do not, however, routinely perform TTE solely for the purpose of further stroke risk stratification in people with AF for whom the need to initiate anticoagulation therapy has already been agreed on appropriate clinical criteria.5

Personalised package of care and information

People with AF need to understand their condition so that they can make informed decisions. This has been encapsulated in the new recommendations in NICE CG180 around personalised packages of care and information: these should cover awareness of stroke risk, symptoms of stroke, and symptom management (see Box 1). This package of care should be shared with the patient and clearly documented in the notes.5

The author has been involved with developing and can recommend the following websites as accessible sources of information for patients and practitioners:

  • www.careaf.org
    • a patient-support programme for people who have been diagnosed with AF
  • www.atrialfibrillation.org.uk
    • provides information, support, and access to established, new, or innovative treatments for AF for patients, carers, and healthcare professionals
  • www.heartofaf.org
    • a central resource primarily for healthcare professionals to access the latest research, source information, share pathways/protocols and to be kept up-to-date on established, new, or innovative treatments and guidelines for AF.

Box 1: Components of a care package for people with atrial fibrillation

  • Stroke awareness and measures to prevent stroke*
  • Rate control
  • Assessment of symptoms for rhythm control
  • Who to contact for advice if needed
  • Psychological support if needed
  • Up-to-date and comprehensive education and information on:
    • cause, effects, and possible complications of AF
    • management of rate and rhythm control
    • anticoagulation
    • practical advice on anticoagulation in line with recommendation 1.3.1 in NICE CG144 on Venous thromboembolic diseases7
    • support networks (e.g. cardiovascular charities).
  • * Stroke awareness includes information on the symptoms and signs of stroke and how AF can lead to a stroke; measures to prevent stroke include anticoagulation for AF
  • AF=atrial fibrillation; CG=clinical guideline

Assessment of stroke and bleeding risks

The essential first steps in the assessment and management of AF are around stroke risk assessment and the assessment of the risk of anticoagulation (see Figure 1).8

A clinical toolkit (referenced in NICE CG180) for assessment and treatment of AF can be found at www.cardiosource.org/Science-And-Quality/Clinical-Tools/Atrial-Fibrillation-Toolkit.aspx9

Stroke risk assessment

There is a paradigm shift in the assessment of stroke risk in AF, whereby all people with AF should now be regarded as requiring anticoagulation, with the exception of those people at very low risk.

To aid in this assessment, there has been a change in recommended stroke risk schema to the CHA2 DS2 -VASc scoring system (see Table 1).10 This is a development on the previous CHADS2 schema,11 in that CHA2 DS2 -VASc can determine low risk with a higher sensitivity.

People are regarded as low risk if they have no risk factors (score 0 in men and score 1 in women),5 and intervention should be offered to all other individuals with AF.

Bleeding risk assessment

The intervention to reduce the risk of AF-related stroke is oral anticoagulation, and so the risks and benefits of intervention need to be assessed. It is recognised that the perception of bleeding risk and benefit of anticoagulation between the clinician and patient is very different, with the clinician being considerably more concerned about the risk of bleeding, and placing less emphasis on the associated reduction of stroke risk, than the patient.12

NICE CG180 recommends the use of the HAS-BLED score (see Table 2).13 A HAS-BLED score of 3 or greater is considered to be 'high' bleeding risk.13 The guideline articulates that the HAS-BLED score is not to be an obstacle to prescribing oral anticoagulation and it highlights those factors that can be modified to reduce the bleeding risk.

Reducing HAS-BLED score and bleeding risk if someone's HAS-BLED bleeding risk score is elevated, then the following measures can be taken to reduce this, demonstrating a reduction in the bleeding risk:

  • if the bleeding risk is higher due to raised blood pressure, then intervention with anti-hypertensives to reduce the blood pressure to <160 mmHg will reduce the HAS-BLED score by 1 point
  • measures to improve liver and renal tests will reduce the number of HAS-BLED points
  • if vitamin K oral anticoagulant intervention is difficult to stabilise, switching to a non-vitamin K oral anticoagulant will resolve the problem, lowering the score
  • discontinuing antiplatelet agents or non-steroidal anti-inflammatory drugs will reduce the score
  • advising the patient to reduce their alcohol consumption to <8 units a week will reduce the score by 1 point.

As the risk of stroke in AF is so high compared with the bleeding risk of anticoagulation, the net clinical benefit to those with a CHA2 DS2 -VASc score of 2 or greater is to intervene, irrespective of bleeding risk.14

Figure 1: Stroke prevention of people with non valvular AF8
Stroke prevention of people with non valvular AF
Table 1: CHA2 DS2 -VASc scoring system10
Risk factorScore
Congestive heart failure/left ventricular dysfunction 1
Hypertension 1
Age ≥75 years 2
Diabetes mellitus 1
Stroke/transient ischaemic attack/thromboembolism 2
Vascular disease* 1
Age 65-74 years 1
Sex category (i.e. female sex) 1
Maximum score 9
  • Maximum score is 9 since age may contribute 0, 1, or 2 points.
  • * Prior myocardial infarction, peripheral artery disease, aortic plaque.
  • Adapted from: Lip G, Nieuwlaat R, Pisters R et al. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest 2010; 137: 263-272. By kind permission.
Table 2: HAS-BLED bleeding risk score13
 Clinical characteristicPoints awarded
H Hypertension (SBP >160 mmHg) 1
A Abnormal renal and liver function (1 point each) 1 or 2
S Stroke 1
B Previous bleeding history or predisposition to bleeding 1
L Labile INRs 1
E Elderly-age >65 years 1
D Drugs* or alcohol abuse 1 or 2
    Maximum 9 points
  • * Includes aspirin and non-steroidal anti-inflammatory agents
  • SBP=systolic blood pressure; INR=international normalised ratio
  • Adapted from: Pisters R, Lane D, Nieuwlaat R et al. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey. Chest 2010; 138: 1093-1100. By kind permission.

Interventions to reduce the risk of stroke

Oral anticoagulants

NICE CG180 emphasises that people at low risk of AF-related stroke do not require intervention. In all other people who do have risk factors, however, intervention with an oral anticoagulant is the recommended treatment option. Anticoagulation should be considered for men who have a CHA2 DS2 -VASc score of 1 or more, and offered to everyone with a score of 2 or more.

The anticoagulant can be a NICE-approved, non-vitamin K novel oral anticoagulant (NOAC), such as apixaban,15 dabigatran,16 or rivaroxaban17 (see NICE CG180 for a summary of indications for prescribing these drugs), or a vitamin K antagonist (VKA), such as warfarin.5

The choice of anticoagulant should be made in consultation with the patient, taking in to account their individual clinical features and preferences.5

Vitamin K antagonists

If someone chooses to receive warfarin therapy, then the quality of this intervention should be regularly assessed. The approach recommended in NICE CG180 for assessing anticoagulation control with VKAs is summarised in Box 2.5


Antiplatelet intervention is no longer recommended in the NICE guideline. If, however, a person chooses not to have oral anticoagulant intervention but is known to have occlusive vascular disease or another indication for aspirin monotherapy, then aspirin should not be discontinued on the grounds of the recommendations in this guideline as there is clearly a role for aspirin in other comorbidities.

Left atrial appendage occlusion

If a person cannot be treated with an anticoagulant, referral to a specialist centre for consideration of a left atrial appendage occlusion may be considered. The left atrial appendage is a small protuberance off the left atrium where much of the thrombus that is responsible for cardioembolic stroke is thought to originate. The occlusion device acts as a plug at the opening, preventing thrombus from leaving the appendage and so reducing the risk of stroke. Implantation of the device carries risk, such as the need for antithrombotic intervention at the time of procedure, risk of transatrial septum puncture to reach the left atrium, and movement of the device.18 The device is currently under 'commissioning through evaluation' procedure in England.19

Box 2: Assessing anticoagulation control with vitamin K antagonists5

  • Calculate the person's TTR at each visit. When calculating TTR:
    • use a validated method of measurement such as the Rosendaal method for computer-assisted dosing or proportion of tests in range for manual dosing
    • exclude measurements taken during the first 6 weeks of treatment
    • calculate TTR over a maintenance period of at least 6 months
  • Reassess anticoagulation for a person with poor anticoagulation control shown by any of the following:
    • two INR values higher than 5 or one INR value higher than 8 within the past 6 months
    • two INR values less than 1.5 within the past 6 months
    • TTR less than 65%
  • When reassessing anticoagulation, take into account and if possible address the following factors that may contribute to poor anticoagulation control:
    • cognitive function
    • adherence to prescribed therapy
    • illness
    • interacting drug therapy
    • lifestyle factors including diet and alcohol consumption
  • If poor anticoagulation control cannot be improved, evaluate the risks and benefits of alternative stroke prevention strategies and discuss these with the person.
  • TTR=time in therapeutic range; INR=international normalised ratio

Review of people with atrial fibrillation

Risk is dynamic and AF risk cannot be assessed once but should be continually reviewed. NICE CG180 recommends that stroke risk and bleeding risk of anticoagulation is done at least annually.5

People not taking an anticoagulant

People not receiving an anticoagulant should have their stroke risk reviewed when they become 65 years of age or if they develop any of the following risk factors:5

  • diabetes
  • heart failure
  • peripheral arterial disease
  • coronary heart disease
  • stroke, TIA, or systemic thromboembolism.

For people not receiving an oral anticoagulant because of bleeding risk or other reason, a review of their stroke and bleeding risk should be undertaken annually and the results and the patient's preferences clearly documented.5

People taking an anticoagulant

For people who are taking an anticoagulant, review the need for and the quality of anticoagulation at least annually, or more often if clinically relevant events that affect anticoagulation or bleeding risk occur (such as the need for co-prescription of interactive medications, development of comorbidities that may affect the anticoagulant, or problems with the patient's adherence to therapy, e.g. through memory loss).

Symptom management and rate and rhythm control

There is currently no evidence that rhythm management in AF gives any prognostic or stroke reduction benefit,20 and so management decisions regarding rhythm management are determined by the patient's symptoms alone.

The guideline group reviewed the current evidence and determined that rate management should be the first-line consideration, except in those people:5

  • whose AF has a reversible cause
  • who have heart failure thought to be primarily caused by AF
  • with new-onset AF
  • with atrial flutter, and whose condition is considered suitable for an ablation strategy to restore sinus rhythm
  • for whom a rhythm-control strategy would be more suitable, based on clinical judgement (e.g. those with structurally normal hearts, symptomatic paroxysmal AF, or the young).

In people who are asymptomatic of AF then no intervention may be required for rate or rhythm management; however, there is one caveat regarding those people who, although asymptomatic, have a persistent tachycardia (defined as a resting heart rate greater than 120 bpm);21 rate control should be instituted in these patients because there is otherwise a significant risk that they may develop a left ventricular systolic dysfunction due to a dilated cardiomyopathy of tachycardia.22

Rate control

Rate control should initially be achieved using a standard beta blocker (not sotalol) or rate-limiting calcium channel blocker.5 Monotherapy should be used first but if this is unsuccessful in improving the patient's symptoms, then combinations can be considered using any two of:5

  • a beta blocker
  • diltiazem (as a branded product)23
  • digoxin.

Digoxin should be considered for monotherapy only in people with non-paroxysmal AF who are predominately sedentary (i.e. not undertaking physical exercise).5

When to refer to a rhythm specialist

In people who remain symptomatic from AF despite rate control, or who are intolerant of the medication, then pharmacological or interventional electrical rhythm management should be considered. There should be an established pathway to ensure there is no undue delay in referring a person with AF to a rhythm specialist when it has been established that rate control is ineffective at achieving symptom relief; the guideline defines this as 4 weeks after the final failed treatment, or no longer than 4 weeks after a recurrence of AF following cardioversion when further specialised management is needed.

Rhythm management

A beta blocker (not sotalol) should initially be used for rhythm management.24 If pharmacological intervention for rhythm management proves unsuccessful, then left atrial ablation is an effective option. This procedure is more effective in paroxysmal AF and when attempted early in the disease progression.25 The risks and benefits should be discussed with the person.5

Challenges to primary care

NICE CG180 on the management of AF brings new challenges to primary care. Many of the traditional treatments (e.g. aspirin, digoxin), and the 'have a go' view of cardioversion, have now been removed from the management pathway. The requirement to obtain a cardiologist's view on stroke prevention for the patient has been removed, reinforcing the role of GPs as the group implementing cardiovascular protection for patients. The unique relationship between those in primary care and the patients positions them best to discuss the risks and benefits of stroke prevention. This may be a new role for many primary care practitioners and there is a need for learning around the safe use of anticoagulants, but when done well the benefits to the patients extend far beyond the mere financial incentives for the healthcare and social care systems.


Atrial fibrillation should not be regarded as a future problem for healthcare professionals as the population continues to age, but one that is already upon us. All AF-related stroke should be considered potentially avoidable with the use of oral anticoagulants.

Despite the overwhelming evidence that oral anticoagulation is effective in reducing the risk of stroke, is an effective and practical intervention for older people,26 and is now augmented with the option of NOACs,27 intervention rates remain poor, with 50% of eligible patients receiving anti platelet agents.1

NICE CG180 has now simplified the approach to stroke-risk reduction for clinicians and the NICE commissioning guide for anticoagulation services is available to support clinical commassigning groups in reviewing, monitoring, and supporting anticoagulation therapy in their local area.31

GP commissioning take home messages for England

written by Dr David Jenner, NHS Alliance GMS contract/PBC Lead

  • This guidance needs to be urgently considered by CCGs as it does change significantly the management of AF
  • There are inconsistencies for GPs though, with QOF still requiring the use of CHADS2 and incentivising the use of anticoagulants OR aspirin, which will remain until changed (probably April 2015)
  • For CCGs, the potential cost pressures of a switch from VKA therapy to NOACs will represent a significant risk to prescribing budgets but some of this cost can be recouped by savings on budgets for warfarin monitoring
  • CCGs should:
    • rapidly identify in local formularies indications for VKA versus NOAC therapy and consider decision support tools for clinicians and patients, as well as minimum monitoring requirements for both NOACs and VKAs
    • agree with local specialist services local care pathways based on this guidance, identifying which patients should be investigated with echocardiography or referred for specialist asessment (e.g. those with suspicion of valvular disease)
    • consider educational events for many health professionals, including pharmacy, in effective case-finding for AF
  • Costs of anticoagulants (per month):
    • NOACs: rivaroxaban £58, dabigatran £65, apixaban £61.50?€"£65.90a
    • warfarin: approximately £1 per month per tablet (plus of course the costs of INR monitoring).a

CCG=clinical commissioning group; AF=atrial fibrillation; QOF=quality and outcomes framework; VKA=vitamin K antagonists; NOAC=novel oral anticoagulant; INR=international normalised ratio


  1. Cowan C, Healicon R, Robson I et al. The use of anticoagulants in the management of atrial fibrillation among general practices in England. Heart. Published online first: 7 February 2014 doi:10.1136/heartjnl- 2012-303472.
  2. Chugh S, Havmoeller R, Narayanan K et al. Worldwide epidemiology of atrial fibrillation: a Global Burden of Disease 2010 Study. Circulation 2014; 129: 837-847.
  3. Fuster V, Rydén L, Cannom D. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation. Circulation 2006; 114 (7): e257-354; and Eur Heart J 2006; 27: 1979-2030.
  4. Savelieva I et al. Stroke in atrial fibrillation: update on pathophysiology, new antithrombotic therapies, and evolution of procedures and devices. Ann Med 2007; 39 (5): 371-391.
  5. NICE. Atrial fibrillation: the management of atrial fibrillation. Clinical Guideline 180. NICE, 2014. Available at: www.nice.org.uk/guidance/CG180.
  6. Mant J, Fitzmaurice D, Richard Hobbs F et al. Accuracy of diagnosing atrial fibrillation on electrocardiogram by primary care practitioners and interpretative diagnostic software: analysis of data from screening for atrial fibrillation in the elderly (SAFE) trial. BMJ 2007; 335 (7616): 380.
  7. NICE. Venous thromboembolic diseases: the management of venous thromboembolic diseases and the role of thrombophilia testing. Clinical Guideline 144. London: NICE, 2012. Available at: www.nice.org.uk/guidance/CG144
  8. National Clinical Guideline Centre. Atrial fibrillation: the management of atrial fibrillation. Clinical guideline. Methods, evidence and recommendations. NCGC, 2014. Available at: www.nice.org.uk/guidance/cg180/resources/cg180-atrial-fibrillation-update-full-guideline3 (accessed 21 August 2014).
  9. American College of Cardiology website. AFib Toolkit. www.cardiosource.org/Science-And-Quality/Clinical-Tools/Atrial-Fibrillation-Toolkit.aspx (accessed 14 August 2014).
  10. Lip G, Nieuwlaat R, Pisters R et al. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest 2010; 137: 263-272.
  11. Gage B, Waterman A, Shannon W et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001; 285: 2864-2870.
  12. Devereaux P, Anderson D, Gardner M et al. Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study; BMJ 2001; 323: 1218.
  13. Pisters R, Lane D, Nieuwlaat R et al. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey. Chest 2010; 138: 1093-1100.
  14. Banerjee A, Lane D, Torp-Pedersen C et al. Net clinical benefit of new oral anticoagulants (dabigatran, rivaroxaban, apixaban) versus no treatment in a 'real world' atrial fibrillation population: A modelling analysis based on a nationwide cohort study. Thromb Haemost 2012; 107 (3): 584-589.
  15. NICE. Apixaban for preventing stroke and systemic embolism in people with nonvalvularatrial fibrillation. Technology Appraisal 275. NICE, 2013. Available at: www.nice.org.uk/guidance/TA275
  16. NICE. Dabigatran etexilate for the prevention of stroke and systemic embolism in atrial fibrillation. Technology Appraisal 249. NICE, 2012. Available at: www.nice.org.uk/guidance/TA249
  17. NICE. Rivaroxaban for the prevention of stroke and systemic embolism in people with atrial fibrillation. Technology Appraisal 256. NICE, 2012. Available at: www.nice.org.uk/guidance/TA256
  18. NICE. Percutaneous occlusion of the left atrial appendage in non-valvular atrial fibrillation for the prevention of thromboembolism. Interventional procedures guidance 349. NICE, 2010. Available at: www.nice.org.uk/guidance/ipg349
  19. NHS England website. Commissioning through evaluation. Left atrial appendage occlusion. Available at: www.england.nhs.uk/ourwork/commissioning/spec-services/npc-crg/comm-eval/ (accessed 15 August 2014).
  20. Wyse D, Waldo A, DiMarco J et al. Atrial fibrillation follow-up investigation of rhythm management (AFFIRM) investigators. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002; 347 (23): 1825-1833
  21. Isabelle C, Van Gelder M, Hessel F et al. for the RACE II Investigators. Lenient versus Strict Rate Control in Patients with Atrial Fibrillation. N Engl J Med 2010; 362: 1363-1373.
  22. McLaran C, Gersh B, Sugrue D et al. Tachycardia induced myocardial dysfunction. A reversible phenomenon? Br Heart J 1985; 53 (3): 323-327.
  23. Royal Pharmaceutical Society Council Advice. Solid oral modified-release preparations. Pharm J 1993: 251: 528.
  24. Freemantle N, Lafuente-Lafuente C, Mitchell S et al. Mixed treatment comparison of dronedarone, amiodarone, sotalol, flecainide, and propafenone, for the management of atrial fibrillation. Europace 2011; 13 (3): 329-345.
  25. Pappone C, Oreto G, Rosanio S et al. Atrial electroanatomic remodeling after circumferential radiofrequency pulmonary vein ablation. Efficacy of an anatomic approach in a large cohort of patients with atrial fibrillation. Circulation 2001; 104: 2539-2544.
  26. Mant J, Hobbs F, Fletcher K et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet 2007; 370 (9586): 493-503.
  27. Connolly S, Ezekowitz M, Yusuf S et al and the RE-LY Steering Committee and Investigators. Dabigatran versus Warfarin in Patients with Atrial Fibrillation. N Engl J Med 2009; 361 (12): 1139-1151.
  28. Patel M, Mahaffey K, Garg J et al and the ROCKET AF Steering Committee for the ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011; 365: 883-891.
  29. Connolly S, Eikelboom J, Joyner C et al for the AVERROES Steering Committee and Investigators. Apixaban in patients with atrial fibrillation. N Engl J Med 2011; 364: 806-817.
  30. Granger C, Alexander J et al for the ARISTOTLE Committees and Investigators. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011; 365 (11): 981-992.
  31. NICE. NICE commissioning guides [CMG49]. Support for commissioning: anticoagulation therapy. NICE, 2013. Available at: www.nice.org.uk/guidance/cmg49/chapter/1-key-issues-in-commissioning-anticoagulation-therapy (accessed 18 August 2014). G