CHA₂DS₂-VASc — Curie.MD Calc

Stroke risk in atrial fibrillation.

Pair this assessment

Add HAS-BLED

Reuse entered values for a separate bleeding-risk assessment.

Applies to / doesn’t apply to3

Applies to: Non-valvular atrial fibrillation only. Excludes moderate-to-severe rheumatic mitral stenosis and mechanical heart valves.

Does not apply if:

Hypertrophic cardiomyopathy.
Cardiac amyloidosis.
Valvular atrial fibrillation — moderate-to-severe rheumatic mitral stenosis or a mechanical heart valve — which falls outside “non-valvular AF”.

CHA₂DS₂-VASc inputs

Age

Sex

Congestive heart failure

Clinical heart failure, or objective evidence of moderate-to-severe LV dysfunction. Reduced ejection fraction is not required.

Hypertension history

Resting BP >140/90 mmHg on at least two occasions, or on antihypertensive treatment. History, not current control.

Diabetes mellitus

Fasting glucose ≥126 mg/dL, or on hypoglycaemic treatment.

Prior stroke

Any prior ischaemic stroke.

TIA, peripheral or pulmonary embolism

Asked separately from stroke because HAS-BLED’s criterion counts none of these. Scores the same +2 and does not add to it.

Vascular disease

Coronary artery disease — prior MI, angina, prior PCI, or CABG — or peripheral artery disease.

What this score estimates

Estimates annual ischaemic stroke risk in patients with non-valvular atrial fibrillation who are not taking an anticoagulant. The score runs 0–9; the point for female sex is included in the total and the anticoagulation threshold differs by sex rather than the score being adjusted.

Pearls & pitfalls9

Discrimination is modest — C-statistic approximately 0.6–0.7. This is a screening tier, not a precise individual estimate.
The derivation cohort was small (n=1,084) with 31% lost to follow-up at one year. Credibility rests on the later validations (Olesen n=73,538; Friberg n=182,678).
The displayed rates come from Friberg’s non-anticoagulated subgroup (n=90,490), not its full cohort — none of those patients were taking an anticoagulant, which is what makes the figures describe untreated risk.
The rate at a given score varies across populations even for the same endpoint. At a score of 2, the 2023 ACC/AHA guideline reports annual ischaemic-stroke rates across 15 validation cohorts of below 1% in four cohorts, 1–2% in six, and above 2% in five (guideline §8, Figure 9); the ~2.2% per year shown here is the point estimate from Friberg’s non-anticoagulated cohort.
Anticoagulation decisions should not turn on whether the atrial fibrillation is paroxysmal or permanent: the risk is similar across patterns.

Risk factors the score does not count:

Obesity
Poorly controlled hypertension
eGFR below 45, or proteinuria
Enlarged left atrium

What the guidelines say2

2023 ACC/AHA AF guideline

Keeps the point for female sex and applies a sex-specific threshold, at a score of 2 in men and 3 in women.

View guideline source ↗

2024 ESC AF guideline

Advocates the CHA₂DS₂-VA formulation, which removes the sex criterion and scores 0–8.

Guideline positions are stated as published. Curie.MD Calc does not publish treatment direction.

Where this comes from3

How it’s scored

Age<65 0 · 65–74 +1 · ≥75 +2

SexMale 0 · Female +1

Congestive heart failureNo 0 · Yes +1Clinical heart failure, or objective evidence of moderate-to-severe LV dysfunction. Reduced ejection fraction is not required.

Hypertension historyNo 0 · Yes +1Resting BP >140/90 mmHg on at least two occasions, or on antihypertensive treatment. History, not current control.

Diabetes mellitusNo 0 · Yes +1Fasting glucose ≥126 mg/dL, or on hypoglycaemic treatment.

Prior strokeNo 0 · Yes +2Any prior ischaemic stroke.

TIA, peripheral or pulmonary embolismNo 0 · Yes +2Asked separately from stroke because HAS-BLED’s criterion counts none of these. Scores the same +2 and does not add to it.

Vascular diseaseNo 0 · Yes +1Coronary artery disease — prior MI, angina, prior PCI, or CABG — or peripheral artery disease.

What was observed

0~0.2 ischaemic strokes per 100 patient-years in non-anticoagulated patients · Friberg 2012 cohort

1~0.6 ischaemic strokes per 100 patient-years in non-anticoagulated patients · Friberg 2012 cohort

2~2.2 ischaemic strokes per 100 patient-years in non-anticoagulated patients · Friberg 2012 cohort

3~3.2 ischaemic strokes per 100 patient-years in non-anticoagulated patients · Friberg 2012 cohort

4~4.8 ischaemic strokes per 100 patient-years in non-anticoagulated patients · Friberg 2012 cohort

5~7.2 ischaemic strokes per 100 patient-years in non-anticoagulated patients · Friberg 2012 cohort

6~9.7 ischaemic strokes per 100 patient-years in non-anticoagulated patients · Friberg 2012 cohort

≥7~10.8–12.2 ischaemic strokes per 100 patient-years in non-anticoagulated patients (scores ≥7) · Friberg 2012 cohort

References

Lip GYH, Nieuwlaat R, Pisters R, et al. Chest. 2010;137(2):263-72. View source ↗
Friberg L, Rosenqvist M, Lip GYH. Eur Heart J. 2012;33(12):1500-10. View source ↗
2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation. View source ↗

Live result

CHA₂DS₂-VASc

0 of 8 answeredAnswer the remaining 8 to see the score

Not applicable in hypertrophic cardiomyopathy or cardiac amyloidosis — both current guidelines recommend against basing stroke-prevention decisions on this score in those patients.

Document your decision

Answer the inputs to see the score