Home

West Hertfordshire Cardiology logo

West Herts Hospitals NHS Trust logo

Cardiac Risk Assessment: Anticoagulation in AF

Objective, systematic assessment is useful in selecting which patients with non-rheumatic permanent AF may benefit from anticoagulation with Warfarin.

General risk factors and the patient's preference should be taken into account.

 

Compared to placebo treatment in AF, Warfarin (target INR 2.5, range 2.0-3.0) reduces relative stroke risk by 68%.

Aspirin (75-300 mg/day) reduces relative stroke risk by 20%.

 

But Warfarin increases annual absolute risk of major haemorrhage by 2%, so only patients at high initial risk of stroke (>6% absolute annual risk) are likely to achieve greater benefit from Warfarin than from Aspirin.

 

A simple, validated risk assessment tool is the CHADS2 score 1.

If CHADS2 score >3 then risk of stroke >6%, so Warfarin is usually indicated.

If CHADS2 score =3 then risk of stroke =6%, so Warfarin may be indicated

If CHADS2 score <3 then risk of stroke <=4%, so Warfarin is usually not indicated, unless recent Stroke or TIA.

 

CHADS2 SCORE
Risk factor Yes No
Congestive Heart Failure (past or present) 1 point
Hypertension (>160/90, past or present) 1 point
Age >=75 1 point
Diabetes 1 point
Stroke or TIA (past or present) 2 points

Risk Score  :

   points

Annual risk of Stroke (rate per 100 pt-years):

    %   Range

Comment :
 

 

 

The Scottish Intercollegiate Guidelines Network (SIGN) has published a comprehensive analysis of anticoagulation in AF 2, from which the following tables and comments have been developed.

 

SIGN - Antithrombotic Therapy Guidelines SIGN 1999 (No 36)
  Annual risk of Stroke in non-valvular AF
Group UnRx Aspirin Warfarin NNT
Very High Risk Previous TIA/Stroke 12% 10% 5% 18
High Risk Age>65 + Risk Factor(s) 5-8% 4-6% 2-3% 42
Mod Risk Age>65 + No Risk Factor(s) 3-5% 2-4% 1-2% 71
Age<65 + Risk Factor(s)
Low Risk Age<65 + No Risk Factors 1-2% 1% 0.5% 142
           
BUT Annual risk of major bleed 0.1% 1% 2% NNH=100
  NNT = Number needed to treat with Warfarin instead of Aspirin for 1 year to prevent 1 stroke

NNH = Number needed to harm by treating with Warfarin instead of Aspirin for 1 year (causing a major haemorrhage)

 

 

An alternative way to express risk to patient when discussing whether to anticoagulate

 

Annual risk of NOT having a Stroke in non-valvular AF

Net absolute benefit of Warfarin rather than Aspirin

(reduced stroke but more major haemorrhage)
Group UnRx Aspirin Warfarin
Very High Risk Previous TIA/Stroke 88% 90% 95% 3%
High Risk Age>65 + Risk Factor(s) 92% 94% 97% 1%
Mod Risk Age>65 + No Risk Factor(s) 95% 96% 98% 0
Age<65 + Risk Factor(s)
Low Risk Age<65 + No Risk Factors 98% 99% 99.5% -1.5%

 

 

SIGN recommendations (with grade of evidence)

  • Patients with AF but without additional Risk factors require no antithrombotic prophylaxis unless there are other indications for aspirin.(A)

  • Patients with one or more risk factors should be considered for warfarin therapy in preference to aspirin.(A)

  • Warfarin prophylaxis should also be considered in patients with atrial fibrillation and heart valve disease or prostheses, thyrotoxicosis, intracardiac thrombus, or non-cerebral thromboembolism. (C)

  • The decision to use warfarin or not should be based on discussion of the balance of risk and benefit with each individual, including assessment of compliance.(B)

  • To minimise the risk of intracranial bleeding in patients on warfarin, hypertension should be controlled, compliance assessed, and the risks and benefits of warfarin reviewed annually, especially in those aged over 75 years.(Good Practice)

  • Cardioversion to restore sinus rhythm should be considered in selected patients, because it may avoid the need for long term warfarin.(C)

 

 

General Risk Factors

  • Previous TIA/Stroke

  • Age >65 years (Note that CHADS2 uses Age >=75 years)

  • Hypertension (>160/90 or on treatment)

  • Diabetes

  • Clinical Heart Failure or known LV dysfunction

  • Echo LV dysfunction

  • Echo Mitral valve ring calcification

  • Thyroid disease
     

  • Note that Echo LA>5cm is NOT very predictive, so Echo is usually NOT immediately essential in deciding if Warfarin is required.

 

 

Other Evidence-based notes

  • Adding Dipyridamole (Persantin) 200mg bd to Aspirin does not reduce the risk of recurrent TIA/Stroke 5. Consider Warfarin instead…

  • Patients allergic to Aspirin should receive Clopidogrel 75mg od instead (see Clopidogrel guidelines)

  • If previous history of GI bleed, peptic ulcer or GORD, use Proton Pump Inhibitor (rather than H2 antagonist) with Aspirin (or with Clopidogrel if Aspirin allergic)

  • Echocardiography is NOT helpful in diagnosing cause of TIA/Stroke in the absence of clinical evidence of cardiac pathology and risk factors

  • Echocardiography is NOT helpful in deciding need for anticoagulation after thromboembolic TIA/Stroke if patient is in AF: Warfarin is indicated!

 

 

 

References
1
CHADS		 Gage BF, Waterman AD, Shannon W, et al
Validation of clinical classification schemes for predicting stroke.  Results from the National Registry of Atrial Fibrillation.
JAMA 2001;285:2864-70
2
SIGN		 Antithrombotic therapy
SIGN 1999;publication 36:18-22
3 ACC/AHA/ESC  Guidelines for the Management of Patients With Atrial Fibrillation
JACC 2001;38:1266i-lxx
4 David A Fitzmaurice, Andrew D Blann, Gregory Y H Lip
ABC of antithrombotic therapy.  Bleeding risks of antithrombotic therapy
BMJ 2002;325:828-31
5 Antithrombotic Triallists Collaboration
Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients
BMJ 2002;324:71-86 (for dipyridamole data see Fig 5 and page 82)
Page designed by Dr John Bayliss

Page last updated:  22 Dec 2003