Clinical article · Specialties, Medicines and results

Stroke prevention in atrial fibrillation: scoring risk and choosing an anticoagulant

Written for
Hospital physicians, cardiology and medical registrars, and residents
Reading time
8 minutes
Published
Updated

Written by the Sana Health team for clinicians. It summarises published guidance, is not a guideline and has not been peer reviewed. Check current local protocols before acting.

Key points

  • The anticoagulation decision rests on the absolute risk of stroke weighed against bleeding, not on whether the atrial fibrillation is paroxysmal or persistent.
  • The 2024 ESC guideline uses CHA2DS2-VA, which drops the point for female sex; US guidance uses CHA2DS2-VASc with sex-specific thresholds.
  • A high bleeding score is a prompt to correct modifiable risk factors and review more often, not a reason on its own to withhold anticoagulation.
  • Direct oral anticoagulants are preferred to warfarin in eligible patients, but not with a mechanical valve or moderate to severe mitral stenosis.
  • DOAC doses follow each drug's label criteria; off-label low doses, aspirin as a substitute and stopping after cardioversion are common errors.

The decision in outline

Atrial fibrillation raises the risk of stroke, heart failure, dementia and death [12]. Anticoagulation prevents most of the excess stroke risk, so the decision for each patient turns on the absolute risk of stroke and thromboembolism, the risk of bleeding, and the net clinical benefit between them [4]. The ESC 2024 guideline [1], the 2023 ACC/AHA/ACCP/HRS guideline [2] and the CHEST guideline [3] agree on the framework, although they score risk slightly differently.

The pattern of atrial fibrillation does not change the decision: paroxysmal, persistent and permanent atrial fibrillation are treated alike for stroke prevention, and a return to sinus rhythm after cardioversion or ablation is not on its own a reason to stop anticoagulation.

Scoring stroke risk

LetterRisk factorPoints
CCongestive heart failure or left ventricular dysfunction1
HHypertension1
A2Age 75 or over2
DDiabetes mellitus1
S2Previous stroke, transient ischaemic attack or thromboembolism2
VVascular disease: previous myocardial infarction, peripheral arterial disease or aortic plaque1
AAge 65 to 741
ScSex category: female (CHA2DS2-VASc only)1

The ESC 2024 guideline dropped the sex category and uses CHA2DS2-VA: anticoagulation is recommended at a score of 2 or more and should be considered at 1 [1]. The ACC/AHA guideline keeps CHA2DS2-VASc, recommends anticoagulation when the annual risk is 2% or more (a score of 2 or more in men, 3 or more in women), and considers it reasonable at a score of 1 in men or 2 in women [2]. In practice the two approaches reach similar decisions.

Scores are a floor, not a ceiling. They were derived in populations that underrepresented some groups, and other features, such as a recent embolic event or active cancer, can raise risk beyond the score.

Using a bleeding score properly

HAS-BLED counts uncontrolled hypertension, abnormal renal or liver function, previous stroke, bleeding history or predisposition, labile INR, age over 65, and antiplatelet or anti-inflammatory drugs or alcohol excess [4]. Its main value is to bring the modifiable items to attention. A high score calls for blood pressure control, stopping unnecessary antiplatelet drugs and NSAIDs, addressing alcohol, and closer follow-up; it is not, on its own, a reason to withhold anticoagulation [1, 3, 4].

Choosing the anticoagulant

In a meta-analysis of the four pivotal trials, direct oral anticoagulants reduced stroke or systemic embolism, intracranial haemorrhage and all-cause mortality compared with warfarin, with more gastrointestinal bleeding [5]. Patient-level analyses have examined whether the benefit holds across age and sex [6], and in ARISTOTLE apixaban was compared with warfarin in 18,201 patients with atrial fibrillation and at least one additional stroke risk factor [7]. Guidelines therefore prefer a DOAC in eligible patients [1, 2].

  • Warfarin, not a DOAC, is used with a mechanical heart valve or moderate to severe mitral stenosis.
  • DOACs are generally avoided in antiphospholipid syndrome, particularly when triple-positive.
  • Severe renal impairment, interacting drugs and extremes of body weight need a drug-specific decision.
  • Patient preference, adherence, cost and access, and the availability of a reversal agent all enter the choice.

Each DOAC has its own dose-reduction criteria, and they are not interchangeable. Apixaban, for example, is reduced when two of three apply: age 80 or over, weight 60 kg or less, and serum creatinine of 133 micromol/L (1.5 mg/dL) or more. Creatinine clearance by Cockcroft-Gault, as used in the trials, is the usual renal measure for dosing, and criteria differ between US and European labels.

Older patients, falls and frailty

Older patients carry the highest stroke risk and the highest bleeding risk. Syncope and fall-related injury are the commonest reasons anticoagulation is not prescribed in people aged 75 or over [9], yet for most patients the stroke risk outweighs the risk of a serious bleed after a fall. In ELDERCARE-AF, Japanese patients aged 80 or over who were not candidates for standard doses were randomised to edoxaban 15 mg daily or placebo, and the low dose reduced stroke or systemic embolism [8]. That regimen is specific to the trial population and is not a licence for low doses in general.

Atrial fibrillation found on a device, and appendage closure

Short episodes detected by a pacemaker or implanted monitor carry a lower stroke risk than clinical atrial fibrillation. A meta-analysis of the NOAH-AFNET 6 and ARTESiA trials examined anticoagulation in device-detected atrial fibrillation, and the decision is individualised by burden and risk [10]. Left atrial appendage closure is an alternative for selected patients who cannot take long-term anticoagulation, and continues to be tested against best medical care in patients at high risk of both stroke and bleeding [11].

Traps

  • Prescribing a reduced DOAC dose when the patient does not meet the label criteria.
  • Using aspirin instead of an anticoagulant; antiplatelet monotherapy does not provide adequate stroke prevention.
  • Stopping anticoagulation after cardioversion or ablation because the rhythm is now normal.
  • Failing to recheck renal function, weight and interacting drugs as the patient ages.
  • Withholding anticoagulation because of a high HAS-BLED score rather than treating its modifiable causes.
  • Using a DOAC with a mechanical valve.

How Sana Health helps

The Cardiology engine calculates CHA2DS2-VASc, or CHA2DS2-VA where that is the local standard, and HAS-BLED from the details entered, showing each point it counted and which bleeding risk factors can be modified. It then checks for reasons a DOAC is unsuitable, gives the dose that fits the patient's age, weight and creatinine clearance, and notes interacting drugs. US product information for the named anticoagulant is attached from DailyMed, and the guidelines and trials it draws on are listed from PubMed.

Questions

What CHA2DS2-VASc score needs anticoagulation?

Under US guidance, anticoagulation is recommended at a score of 2 or more in men and 3 or more in women, and is reasonable at 1 in men or 2 in women. The ESC uses CHA2DS2-VA without the sex point: recommended at 2 or more, considered at 1.

Should a patient who falls be anticoagulated for atrial fibrillation?

Usually yes. For most patients the reduction in stroke risk outweighs the risk of a serious bleed after a fall. Falls should prompt a falls assessment and a review of contributing drugs rather than automatic withdrawal of anticoagulation.

Is HAS-BLED a reason not to anticoagulate?

No. A high HAS-BLED score identifies patients who need closer follow-up and correction of modifiable bleeding risk factors, such as uncontrolled blood pressure or concurrent NSAID use. It is not on its own a reason to withhold anticoagulation.

References

  1. 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS) (opens in a new tab)Eur Heart J, 2024
  2. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines (opens in a new tab)Circulation, 2024
  3. Antithrombotic Therapy for Atrial Fibrillation: CHEST Guideline and Expert Panel Report (opens in a new tab)Chest, 2018
  4. Stroke and bleeding risk assessment in atrial fibrillation: when, how, and why? (opens in a new tab)Eur Heart J, 2013
  5. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials (opens in a new tab)Lancet, 2014
  6. Direct Oral Anticoagulants Versus Warfarin in Patients With Atrial Fibrillation: Patient-Level Network Meta-Analyses of Randomized Clinical Trials With Interaction Testing by Age and Sex (opens in a new tab)Circulation, 2022
  7. Apixaban versus warfarin in patients with atrial fibrillation (opens in a new tab)N Engl J Med, 2011
  8. Low-Dose Edoxaban in Very Elderly Patients with Atrial Fibrillation (opens in a new tab)N Engl J Med, 2020
  9. Management of Atrial Fibrillation in Patients 75 Years and Older: JACC State-of-the-Art Review (opens in a new tab)J Am Coll Cardiol, 2022
  10. Direct Oral Anticoagulants for Stroke Prevention in Patients With Device-Detected Atrial Fibrillation: A Study-Level Meta-Analysis of the NOAH-AFNET 6 and ARTESiA Trials (opens in a new tab)Circulation, 2024
  11. Left Atrial Appendage Closure or Medical Therapy in Atrial Fibrillation (opens in a new tab)N Engl J Med, 2026
  12. Atrial Fibrillation: A Review (opens in a new tab)JAMA, 2025