A 68-year-old woman with atrial fibrillation has treated hypertension. She has no heart failure, diabetes, prior stroke, TIA or embolism, and no vascular disease.
Her CHA₂DS₂-VASc is 3: one point for age 65 to 74, one for hypertension and one for female sex. Her CHA₂DS₂-VA is 2: the same score without the sex point.
The numbers differ, yet under every major guideline she falls in the top group. Guidelines that kept CHA₂DS₂-VASc already set their thresholds one point higher for women, so the two changes cancel out. The difference is real for another patient: a woman aged 65 to 74 with no other factor, read against a threshold shared by both sexes. This guide sets the frameworks side by side, then covers the 2024 definitions and what the rates measure.
Illustrative case: synthetic, and any criterion not mentioned is absent.
At a glance
- CHA₂DS₂-VA is CHA₂DS₂-VASc without the point for female sex, as defined in the 2024 ESC guideline. It runs 0 to 8. With the other items scored the same way, a woman's CHA₂DS₂-VA is one point lower than her CHA₂DS₂-VASc and a man's is the same.
- Where a guideline uses sex-specific CHA₂DS₂-VASc thresholds, no woman changes group. Under a single threshold for both sexes, as in NICE NG196, a woman aged 65 to 74 with no other factor sits at 2 on CHA₂DS₂-VASc and at 1 on CHA₂DS₂-VA.
- The ESC 2024 version also rewrote four definitions. Pulmonary embolism no longer counts, and coronary disease seen only on imaging now does.
What CHA₂DS₂-VA measures
CHA₂DS₂-VA counts thromboembolic risk factors in adults with atrial fibrillation or atrial flutter. It has no derivation study of its own. Its weights are those of CHA₂DS₂-VASc, derived by Lip and colleagues in 2010 from the Euro Heart Survey (1,084 patients), with the sex point removed.1 The 2024 ESC guideline proposes it "in the absence of other locally validated alternatives".2
The ESC describes female sex as "an age-dependent stroke risk modifier rather than a risk factor".2 It adds that a sex criterion complicates practice and omits people who are non-binary or transgender.2 The modifier argument rests partly on Danish registry data from 239,671 patients not on anticoagulation at baseline: one-year absolute risk of ischaemic stroke or systemic embolism was 0.5% at a CHA₂DS₂-VA of 0 in both sexes, while women had higher risk than men at scores above 1.3
The score has no endpoint or time window of its own; those come from the cohort whose rates sit beside it, here FinACAF, a Finnish registry of patients aged 20 and over with newly diagnosed atrial fibrillation or flutter.4
Source: Lip 20101; ESC 2024 §6.1.12; Nielsen 20183; Teppo 20244.
Criteria and points
| Letter | Criterion | Points | Definition used in Curie Calc |
|---|---|---|---|
| C | Chronic heart failure | 1 | Symptoms and signs of heart failure at any ejection fraction (HFpEF, HFmrEF or HFrEF), or asymptomatic LVEF ≤40%. |
| H | Hypertension | 1 | Resting blood pressure >140/90 mmHg on at least two occasions, or current antihypertensive treatment. |
| A₂ | Age ≥75 | 2 | |
| D | Diabetes mellitus | 1 | Type 1 or type 2 diabetes by currently accepted diagnostic criteria, or treatment with glucose-lowering therapy. |
| S₂ | Prior stroke, TIA or arterial thromboembolism | 2 | The ESC wording is arterial: pulmonary embolism is not listed. |
| V | Vascular disease | 1 | Coronary artery disease (prior MI, angina, coronary revascularization, or significant CAD on angiography or cardiac imaging), or peripheral vascular disease (intermittent claudication, prior revascularization for PVD, intervention on the abdominal aorta, or complex aortic plaque on imaging: mobile, ulcerated, pedunculated, or ≥4 mm thick). |
| A | Age 65–74 | 1 |
The maximum is 8. The ESC defines thresholds, not named risk tiers, so Curie Calc shows a rate for each score and no tier label.
Source: ESC 2024, Table 10.2
What dropping sex changes: the frameworks side by side
The arithmetic is exact when the other items are scored the same way. For a woman, CHA₂DS₂-VA = CHA₂DS₂-VASc − 1; for a man the two are identical.2 (The ESC 2024 definitions can also move a score, as the next section shows.)
Two patients make the point. The opening case is a woman aged 68 with hypertension (CHA₂DS₂-VASc 3, CHA₂DS₂-VA 2). The second is the same woman without hypertension (CHA₂DS₂-VASc 2, CHA₂DS₂-VA 1).
| Framework | Body and year | Score used | Thresholds as the guideline states them | Opening case | Same woman, no other factor |
|---|---|---|---|---|---|
| ESC | 20242 | CHA₂DS₂-VA | ≥2 recommended as an indicator of elevated thromboembolic risk for anticoagulation decisions; 1 should be considered one; no statement at 0 | 2: in the ≥2 group | 1: in the score-1 group |
| ESC | 20205 | CHA₂DS₂-VASc | Anticoagulation recommended at ≥2 in men, ≥3 in women; considered at 1 in men, 2 in women | 3: in the ≥3 (women) group | 2: in the 2 (women) group |
| ACC/AHA/ACCP/HRS | 20236 | CHA₂DS₂-VASc | Recommended at an estimated risk ≥2% per year, "eg" ≥2 in men, ≥3 in women; reasonable at 1% to <2%, equivalent to 1 in men, 2 in women | 3: in the ≥3 (women) group | 2: in the 2 (women) group |
| NICE NG196 | 2021, unchanged since7 | CHA₂DS₂-VASc | Offer at ≥2, the same for both sexes; consider at 1 for men | 3: in the ≥2 group | 2: in the ≥2 group |
| APHRS | 20268 | CHA₂DS₂-VASc | Recommended at ≥2 in males, ≥3 in females; should be considered at 1 in males, 2 in females | 3: in the ≥3 (females) group | 2: in the 2 (females) group |
For the opening case, every framework places her in its top group. For the second woman, four frameworks place her in the same group as a man with a single risk factor, because the sex-specific thresholds already sat one point higher for women. Only under NICE's single threshold does she sit in the top group, with a CHA₂DS₂-VASc of 2. ESC 2024's thresholds are also the same for both sexes, but they apply to CHA₂DS₂-VA, which has no sex point to cross them. The framework travels with the number: a result reported as "2" is clear only when it names the score.
Where guidelines now stand. The ESC itself changed position. Its 2020 guideline discussed CHA₂DS₂-VA and cautioned that "not considering the sex component would underestimate stroke risk in women".5 Its 2024 guideline adopted it.2 Thailand, Germany and Brazil adopted CHA₂DS₂-VA in 2025.91011 The 2026 APHRS guideline kept CHA₂DS₂-VASc, writing that routinely omitting female sex "may not be appropriate in the Asia-Pacific context".8 It cites Asian cohorts in which women had higher rates of stroke or systemic embolism than men at CHA₂DS₂-VA scores of 2 or more, a difference not seen at lower scores.8
The evidence moved over time. In FinACAF, CHA₂DS₂-VASc reclassified better in 2007–08; by 2017–18 the category-based net reclassification index favoured CHA₂DS₂-VA (0.070). The authors call the differences between the scores marginal.4
Details worth checking: same name, new definitions
The ESC 2024 version did more than drop a letter. Its Table 10 rewrote four of the six remaining definitions: heart failure, diabetes, the prior-event item and vascular disease.2 Curie's CHA₂DS₂-VASc calculator uses the 2010 definitions as reproduced in the 2023 ACC/AHA guideline (Table 10), with complex aortic plaque added, because that is the score it calculates.6 Each score is defined by its own source, and each calculator follows its own. Read the CHA₂DS₂-VASc guide.
| Item | Curie's CHA₂DS₂-VASc (2010 wording) | Curie's CHA₂DS₂-VA (ESC 2024) |
|---|---|---|
| Heart failure | Clinical heart failure, or objective evidence of moderate-to-severe LV dysfunction | Symptoms and signs at any ejection fraction, or asymptomatic LVEF ≤40% |
| Diabetes | Fasting glucose ≥126 mg/dL, or hypoglycaemic treatment | Type 1 or 2 by currently accepted criteria, or glucose-lowering therapy |
| Prior event (2 points) | Ischaemic stroke; or TIA, peripheral or pulmonary embolism | Stroke, TIA or arterial thromboembolism |
| Vascular disease | Prior MI, angina, PCI or CABG; peripheral artery disease; complex aortic plaque | Adds significant CAD on angiography or cardiac imaging |
Does a prior pulmonary embolism count?
Not in CHA₂DS₂-VA. The ESC row reads "arterial thromboembolism", so a venous event is not listed.2 The 2010 definitions, as the ACC/AHA guideline reproduces them, named pulmonary embolism.6
Does coronary disease seen only on imaging count?
In CHA₂DS₂-VA, yes. The ESC added "significant CAD on angiography or cardiac imaging" to the vascular item.2 The 2010 list names MI, angina and revascularisation.6
Take a synthetic 70-year-old man with a pulmonary embolism two years ago and significant coronary disease on CT coronary angiography, with no MI, angina or revascularisation; all other criteria are absent. Under the 2010 wording his CHA₂DS₂-VASc is 3 (age 1, embolism 2). Under ESC 2024 his CHA₂DS₂-VA is 2 (age 1, vascular 1). He is a man, so sex plays no part: the whole difference is definitions, and the two changes run in opposite directions.
Does a prior haemorrhagic stroke count?
The source doesn't settle it. The ESC wording says "prior stroke" without the word ischaemic.2 The 2010 definitions said ischaemic stroke.6 FinACAF, whose rates the calculator shows, coded ischaemic and unspecified stroke and TIA, with no haemorrhagic stroke codes.4 Curie Calc does not take a position on it.
Does every "CHA₂DS₂-VA" use these definitions?
No. Brazil's 2025 table keeps "left ventricular dysfunction", systemic embolism and a shorter vascular list.11 Thailand's 2025 text describes "thromboembolism or ischemic stroke".9 The name is shared; the wording behind it is not always the ESC's.
Reading the result: what the FinACAF rates are
Curie Calc shows ischaemic strokes per 100 patient-years in the first year after AF diagnosis, in patients not on anticoagulation, from the 2017–18 period of FinACAF.4 That is a rate per unit of person-time, not a percentage of patients per year. Follow-up began 14 days after diagnosis and stopped at the first anticoagulant purchase.4 The 2017–18 period covered 34,559 patients, 11,460 patient-years and 596 strokes.4
| CHA₂DS₂-VA | 2017–18, shown in Curie Calc (95% CI) | 2007–08, same cohort |
|---|---|---|
| 0 | 0.5 (0.2–1.0) | 0.3 |
| 1 | 1.5 (1.0–2.1) | 0.9 |
| 2 | 2.9 (2.2–3.8) | 3.4 |
| 3 | 5.1 (4.2–6.2) | 5.8 |
| 4 | 7.3 (6.0–8.8) | 9.8 |
| 5 | 11.2 (9.3–13.4) | 20.9 |
| 6 | 15.5 (12.4–19.2) | 25.1 |
| 7 | 14.7 (10.1–20.8) | 22.6 |
| 8 | 19.5 (10.1–34.0) | 9.4 |
Teppo 2024 (FinACAF), Table 2: ischaemic stroke per 100 patient-years, first year after AF diagnosis, anticoagulant-naïve patients in Finland.4 Group averages, not individual forecasts.
Three things to know about this table:
- Rates at higher scores fell between periods. At a score of 5 the rate roughly halved, while the rates at 0 and 1 were slightly higher in 2017–18. The authors suggest better detection of comorbidities may have moved lower-risk patients into higher score groups.4 Anticoagulation ended follow-up for 68.2% of patients in 2017–18 against 20.2% in 2007–08, which, the authors note, affects event rates less than cumulative incidence.4
- The tail is imprecise. The rate at 7 sits below 6, and the interval at 8 runs from 10.1 to 34.0.4
- The scores behind the rates were coded from registry diagnoses, not the ESC 2024 definitions. The cohort's prior-event item counted stroke or TIA only.4
The same patient can show two different rates. For the opening case, Curie's CHA₂DS₂-VASc calculator shows about 3.2 ischaemic strokes per 100 patient-years at her score of 3, from the never-anticoagulated subgroup of a Swedish cohort diagnosed between 2005 and 2008 and followed for an average of 1.5 years12; this calculator shows 2.9 at her CHA₂DS₂-VA of 2, from Finland in 2017–18.4 Neither corrects the other. A rate is readable only with its endpoint, window and cohort.
The 2024 ESC guideline states that the absolute risk at which to start anticoagulation in an individual "cannot be estimated from population-level studies".2 Discrimination is modest for all clinical scores, with c-statistics of about 0.65.4
Worked examples
Synthetic cases. Any criterion not mentioned is absent. Each result was computed by the calculator's code. Rates are FinACAF 2017–18: ischaemic strokes per 100 patient-years in the first year after AF diagnosis, untreated patients, group averages.4
| Case | Calculation | Total | Learning point |
|---|---|---|---|
| 77-year-old woman, no other factor | Age ≥75: 2 | 2 (rate 2.9) | Age 75 or over reaches 2 on its own. Her CHA₂DS₂-VASc would be 3. |
| 59-year-old man, no symptoms, LVEF 38% on echocardiography | Heart failure: 1 | 1 (rate 1.5) | Asymptomatic LVEF ≤40% scores under the ESC 2024 wording. |
| 70-year-old woman, prior TIA, type 2 diabetes, hypertension | Age 1 + TIA 2 + diabetes 1 + hypertension 1 | 5 (rate 11.2) | A TIA scores the same 2 points as a stroke. |
What the score doesn't tell you
The 2024 ESC guideline addresses thromboembolic risk in hypertrophic cardiomyopathy and cardiac amyloidosis regardless of the score, so CHA₂DS₂-VA does not apply there.2 Curie Calc also lists a mechanical heart valve and moderate-to-severe mitral stenosis. These patients are addressed separately by AF guidelines.
It leaves out factors the ESC lists as associated with risk: cancer, chronic kidney disease, ethnicity, troponin and BNP, and in some groups an enlarged left atrium, hyperlipidaemia, smoking and obesity.2 By design it omits sex; in Danish data, female sex raised risk in patients with two or more other factors.3 The weights were inherited, not refitted. Bleeding risk is a separate question, which HAS-BLED addresses. Read the HAS-BLED guide
Frequently asked questions
Is CHA₂DS₂-VASc still used?
What is the difference between CHA₂DS₂-VA and CHA₂DS₂-VASc?
CHA₂DS₂-VA drops the female-sex point, so it runs 0 to 8; with the other items scored alike, a woman scores one point lower. The ESC 2024 version also rewrote the heart failure, diabetes, embolism (arterial only) and vascular items.2
Why was female sex removed?
What does a CHA₂DS₂-VA score of 1 mean?
One risk factor other than sex, such as age 65 to 74 or hypertension. The 2024 ESC guideline says a score of 1 should be considered an indicator of elevated thromboembolic risk for anticoagulation decisions.2 In FinACAF 2017–18, untreated patients scoring 1 had 1.5 ischaemic strokes per 100 patient-years in the first year, a group average.4
Sources
- Lip GYH, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. 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(2):263-72. (opens in new tab)
- Van Gelder IC, Rienstra M, Bunting KV, et al. 2024 ESC guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J. 2024;45(36):3314-3414. (opens in new tab)
- Nielsen PB, Skjøth F, Overvad TF, Larsen TB, Lip GYH. Female sex is a risk modifier rather than a risk factor for stroke in atrial fibrillation: should we use a CHA₂DS₂-VA score rather than CHA₂DS₂-VASc? Circulation. 2018;137(8):832-840. (opens in new tab)
- Teppo K, Lip GYH, Airaksinen KEJ, et al. Comparing CHA₂DS₂-VA and CHA₂DS₂-VASc scores for stroke risk stratification in patients with atrial fibrillation: a temporal trends analysis from the retrospective Finnish AntiCoagulation in Atrial Fibrillation (FinACAF) cohort. Lancet Reg Health Eur. 2024;43:100967. (opens in new tab)
- Hindricks G, Potpara T, Dagres N, et al. 2020 ESC guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J. 2021;42(5):373-498. (opens in new tab)
- Joglar JA, Chung MK, et al. 2023 ACC/AHA/ACCP/HRS guideline for the diagnosis and management of atrial fibrillation. J Am Coll Cardiol. 2024;83(1):109-279. (opens in new tab)
- National Institute for Health and Care Excellence. Atrial fibrillation: diagnosis and management (NG196), recommendations 1.6.3, 1.6.4 and 1.6.7. Last updated 30 June 2021. (opens in new tab)
- Chao TF, Lip GYH, Chan YH, et al. 2026 Asia Pacific Heart Rhythm Society (APHRS) guidelines (Part II) on stroke prevention in atrial fibrillation. J Arrhythm. 2026;42(5):e70491. (opens in new tab)
- Ngarmukos T, Methavigul K, Rungpradubvong V, et al. 2025 Thai guideline for the diagnosis and management of atrial fibrillation. Asian Biomed (Res Rev News). 2025;19(5):220-265. (opens in new tab)
- Willems S, Wegner F, Eckardt L, et al. Clinical practice guideline: preventive measures and treatment options for atrial fibrillation. Dtsch Arztebl Int. 2025;122(16):439-444. (opens in new tab)
- Cintra FD, Pisani CF, Rezende AGDS, et al. Brazilian atrial fibrillation guidelines - 2025. Arq Bras Cardiol. 2025;122(9):e20250618. (opens in new tab)
- Friberg L, Rosenqvist M, Lip GYH. Evaluation of risk stratification schemes for ischaemic stroke and bleeding in 182 678 patients with atrial fibrillation: the Swedish Atrial Fibrillation cohort study. Eur Heart J. 2012;33(12):1500-10. (opens in new tab)
This article is educational reference material for healthcare professionals. It does not provide medical advice, diagnosis, or treatment recommendations, and it does not decide anticoagulation for any individual patient. The underlying Curie Calc CHA₂DS₂-VA calculator carries independent clinical review; this article is drawn from that reviewed source and the cited guidelines.