CHA₂DS₂-VA Score

Clinical background

The CHA₂DS₂-VA score was introduced in the 2024 ESC guidelines on atrial fibrillation as a sex-independent replacement for CHA₂DS₂-VASc [1]. The background is twofold: sex as an independent stroke risk factor has been questioned in the modern literature, and the fact that the female sex point is a risk modifier that changes with age rather than a fixed factor makes it difficult to handle in a static scoring system. There was also a wish to simplify decision-making and to avoid excluding patients who identify as non-binary or transgender or who are receiving gender-affirming hormone treatment [2].

The tool serves the same decision as its predecessor: when is stroke risk low enough that anticoagulation is not warranted, and when is it high enough that treatment should be recommended? Like CHA₂DS₂-VASc, the score is primarily a tool for identifying low-risk patients reliably, not for finding high-risk patients.

Calculating the CHA₂DS₂-VA score

The score is calculated as follows:

$$\text{CHA}_2\text{DS}_2\text{-VA} = C + H + A_2 + D + S_2 + V + A_1$$

where the variables are scored as:

The maximum score is 8. Sex is not scored. The score therefore corresponds to CHA₂DS₂-VASc minus the female sex point, but is applied irrespective of sex to all patients with non-valvular atrial fibrillation.

The derivation is not a traditional cohort study but a reinterpretation of existing data. The 2024 ESC guidelines recommend CHA₂DS₂-VA with level of evidence C, and state explicitly that the score should be used in the absence of locally validated alternatives [1]. The empirical basis for omitting sex comes chiefly from the Finnish FinACAF cohort, comprising 144,879 anticoagulant-naive patients with new-onset atrial fibrillation between 2007 and 2018, and from the global GLORIA-AF registry with 21,260 patients enrolled between 2014 and 2016 [3,4].

Interpretation in practice

The 2024 ESC guidelines set the thresholds as follows [1]:

At a score of 0 the annual stroke risk is low enough that no net benefit from anticoagulation is expected, provided there is no other indication. At a score of 1 the benefit is weighed against the bleeding risk individually, with particular attention to patients with a high bleeding risk or limited life expectancy. At a score of 2 or higher the stroke risk is high enough for anticoagulation to be recommended irrespective of sex.

It is important to note that the guidelines state that anticoagulation is recommended at a score of 2 or higher, and may be considered at a score of 1, irrespective of sex [1]. This is a simplification compared with CHA₂DS₂-VASc, in which the female sex point raises the threshold for women with a low score but no other risk factors.

Validation and performance

The largest external validation comes from the FinACAF cohort: 144,879 anticoagulant-naive patients with new-onset atrial fibrillation in Finland between 2007 and 2018, of whom 2.7 per cent (3,936 patients) suffered an ischaemic stroke during one year of follow-up [3]. In the earlier period (2007 to 2008), CHA₂DS₂-VASc was superior to CHA₂DS₂-VA by both continuous and categorical NRI, but the difference attenuated over time. By the end of the study period (2017 to 2018) the categorical NRI had become significant in favour of CHA₂DS₂-VA, and the IDI was also significantly in favour of the sex-independent score. The c-statistic was comparable between the two scores throughout the period.

In the GLORIA-AF registry, a global prospective multicentre cohort of 21,260 patients with atrial fibrillation (mean age 70.2 years, 44.9 per cent women), the AUC for thromboembolism was 0.641 (95 per cent CI 0.585 to 0.697) for CHA₂DS₂-VA versus 0.636 (95 per cent CI 0.580 to 0.692) for CHA₂DS₂-VASc, P = 0.593 [4]. For ischaemic stroke the corresponding AUCs were 0.660 (95 per cent CI 0.582 to 0.739) versus 0.646 (95 per cent CI 0.568 to 0.725), P = 0.847. The predictive properties were therefore statistically equivalent. A significant interaction between sex and age was observed for the risk of thromboembolism, with a tendency towards higher risk in younger women, suggesting that sex may have an age-dependent role that a fixed point does not fully capture.

A prospective German stroke cohort (714 patients, 161 of them with atrial fibrillation) has, however, shown a potential weakness [5]. Among 81 women with atrial fibrillation-related ischaemic stroke, 13.6 per cent had a premorbid CHA₂DS₂-VA score of 1 or lower, which placed them below the ESC anticoagulation threshold despite their having had a stroke. These women predominantly had cardioembolic strokes with moderate to severe neurological impairment. Women had significantly greater stroke severity at presentation (median NIHSS 12 versus 8, P < 0.01), and sex was independently associated with more severe stroke after matching for age and comorbidity (aOR 1.54, 95 per cent CI 1.03 to 2.29). The study is small and from a single centre, but it illustrates that the score may underestimate risk in a minority of women.

Limitations

CHA₂DS₂-VA applies only to non-valvular atrial fibrillation. With a mechanical valve prosthesis, significant mitral stenosis or other valvular heart disease carrying a high embolic risk, anticoagulation is indicated whatever the score, and the score must not be used to justify withholding treatment.

The level of evidence for CHA₂DS₂-VA in the ESC guidelines is C, the weakest level, and rests on observational data rather than randomised trials [1]. The ESC states explicitly that the score should be used in the absence of locally validated alternatives, which means that national or regional validations of CHA₂DS₂-VASc may justify retaining the traditional score.

The German stroke cohort shows that a minority of women with a low CHA₂DS₂-VA score may nonetheless carry a substantial embolic risk [5]. This accords with the observation from GLORIA-AF of a trend towards an interaction between sex and age, in which younger women may have a proportionally higher thromboembolic risk than a sex-neutral score captures [4]. The FinACAF data do, however, show that this sex difference has attenuated as the overall stroke risk in atrial fibrillation has fallen, and that in contemporary cohorts CHA₂DS₂-VA performs marginally better than CHA₂DS₂-VASc [3].

The risk is not static. The Norwegian AFNOR study, comprising 40,782 patients with incident atrial fibrillation and an initially low or intermediate CHA₂DS₂-VA score, found that the score increased in 50 per cent of patients after a median of 1.7 years of follow-up [6]. The proportion with a new risk factor was 19 per cent at 6 months, 25 per cent at 1 year and 40 per cent at 3 years. The authors propose that patients aged ≥55 years be reassessed after 6 months and younger patients after 1 year, and routinely on reaching the ages of 65 and 75, since the age steps in the score are the commonest reason for an increase.

A Catalan primary care cohort of 3,370 patients with new-onset atrial fibrillation found that CHA₂DS₂-VA reduced the sex-based differences in risk stratification, and that only 3.2 per cent of women were reclassified as very low risk [7]. The clinical benefit of switching to the sex-independent score is therefore marginal for the majority of patients, but for those women who score exactly 1 on CHA₂DS₂-VASc by virtue of the sex point, the change means that they may now not receive anticoagulation where it would previously have been considered.

References

1. Van Gelder IC, Rienstra M, Bunting KV, et al. 2024 ESC Guidelines for the management of atrial fibrillation. Eur Heart J 2024;45(36):3314–3414. PMID: 39210723
2. Lip GYH, Teppo K, Nielsen PB. CHA₂DS₂-VASc or a non-sex score (CHA₂DS₂-VA) for stroke risk prediction in atrial fibrillation: contemporary insights and clinical implications. Eur Heart J 2024;45(36):3718–3720. PMID: 39217500
3. 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. PMID: 39171253
4. Lam SHM, Romiti GF, Corica B, et al. Stroke risk stratifications according to CHA₂DS₂-VASc vs. CHA₂DS₂-VA in patients with atrial fibrillation: insights from the GLORIA-AF registry. Eur Heart J Cardiovasc Pharmacother 2025;11(5):433–440. PMID: 40289614
5. Boettger P, Sedighi J, Piayda K, et al. Clinical implications of the recalibrated CHA₂DS₂-VA score for women after ischemic stroke: a prospective cohort study. Acta Neurol Belg 2025;125(6):1653–1662. PMID: 41082160
6. Anjum M, Ariansen I, Myrstad M, et al. Timing of stroke risk reassessment in atrial fibrillation patients with a CHA₂DS₂-VA score of 0 or 1: the Norwegian AFNOR study. Europace 2025;27(10):euaf145. PMID: 40704643
7. Clua-Espuny JL, Panisello-Tafalla A, Lucas-Noll J, et al. Stroke Risk Stratification in Incident Atrial Fibrillation: A Sex-Specific Evaluation of CHA₂DS₂-VA and CHA₂DS₂-VASc. J Cardiovasc Dev Dis 2025;12(7):259. PMID: 40710784