Is Female Sex a Truly Independent Risk Factor for Stroke in Atrial Fibrillation? Insights from a Large European Cohort

Highlights

• Women with atrial fibrillation (AF) are significantly less likely to be prescribed oral anticoagulants (OACs) compared to men (OR 0.79), despite a higher baseline burden of comorbidities.
• Although crude systemic thromboembolism (STE) rates are higher in women, female sex is not a statistically significant independent risk factor after adjusting for age, comorbidities, and the competing risk of death.
• The CHA2DS2-VASc and CHA2DS2-VA scores demonstrate similar overall discrimination, but the CHA2DS2-VA score (which excludes female sex) shows worse clinical reclassification performance.

Background: The Sex Paradox in Atrial Fibrillation

The role of female sex as an independent risk factor for stroke and systemic thromboembolism (STE) in patients with atrial fibrillation (AF) has been a subject of intense debate for decades. While the CHA2DS2-VASc score—the gold standard in European and international guidelines—includes female sex as a ‘risk modifier’ (assigning one point), some contemporary guidelines, such as those from the American Heart Association (AHA) and American College of Cardiology (ACC), have moved toward the CHA2DS2-VA score, effectively removing sex as a weighted variable.

The clinical dilemma lies in whether women are inherently at higher risk due to biological factors, or if the observed increased risk in previous studies was merely a reflection of women being older and having more comorbidities at the time of diagnosis. Furthermore, evidence suggests that women are often undertreated with oral anticoagulants (OACs), which may contribute to higher residual risk. This study by Mei et al., published in European Heart Journal – Quality of Care and Clinical Outcomes, seeks to clarify these disparities within a large, contemporary European cohort.

Study Design and Methodology

Researchers analyzed data from a prospective European cohort comprising 10,080 patients with AF. The study population had a mean age of 70.1 years, with women representing 41.8% of the cohort. The primary objectives were threefold: to assess sex-specific differences in OAC prescription rates, to evaluate the residual risk of STE in patients already treated with OACs, and to compare the predictive performance of the CHA2DS2-VASc and CHA2DS2-VA scores.

The inclusion criteria focused on patients with a CHA2DS2-VA score of 1 or higher. Statistical methods included multivariable logistic regression to assess OAC prescription patterns and Fine-Gray sub-distribution hazard models to account for the competing risk of death when evaluating STE outcomes. The predictive performance of the scoring systems was compared using the Area Under the Curve (AUC) and Net Reclassification Index (NRI).

Key Findings: Prescription Gaps and Residual Risk

Disparities in Anticoagulation Prescription

The study found a significant sex gap in the initiation of stroke prevention therapy. Even after adjusting for clinical variables, women were 21% less likely than men to receive OACs (OR 0.79, 95% CI: 0.69-0.90). This finding is particularly concerning given that the women in this cohort presented with a higher burden of cardiovascular comorbidities than their male counterparts. This suggests a persistent bias or a perceived lower risk-to-benefit ratio in women that is not supported by current clinical evidence.

Thromboembolic Outcomes and the Role of Sex

In terms of outcomes, the crude incidence rate (IR) of STE among OAC-treated patients was higher in women (1.33 per 100 person-years) compared to men (0.94 per 100 person-years). However, the narrative changed significantly upon deeper statistical adjustment. When the researchers accounted for age, hypertension, heart failure, and the competing risk of mortality, the association between female sex and STE became non-significant (sHR 1.24, 95% CI 0.89-1.74, P = 0.210).

This suggests that female sex may not be an independent driver of thromboembolism but rather a marker for other risk factors that are more prevalent or more severe in the female AF population. In other words, the ‘residual risk’ observed in women is largely explained by their clinical profile rather than biological sex alone.

Comparing Risk Scores: CHA2DS2-VASc vs. CHA2DS2-VA

One of the study’s most critical contributions is the head-to-head comparison of the two dominant risk stratification tools. Both scores showed similar, albeit modest, discriminatory power (AUC 0.603 for CHA2DS2-VA vs. 0.605 for CHA2DS2-VASc, P = 0.665). However, the CHA2DS2-VA score demonstrated a negative net reclassification index (-0.088, 95% CI -0.164 to -0.001). This indicates that removing the sex component led to a worse reclassification of patients, potentially misidentifying the risk levels of a significant portion of the female population.

Expert Commentary: Clinical Implications

The findings of this study reinforce the complexity of stroke prevention in women. While female sex might not be a primary biological driver of thrombosis, it remains an essential clinical marker. The fact that women are still less likely to receive OACs—despite having more comorbidities—highlights a critical area for quality improvement in cardiology.

From a mechanistic perspective, the higher crude risk in women may be linked to differences in vascular biology, hormonal influences on coagulation, or even differences in the quality of anticoagulation control (e.g., Time in Therapeutic Range for those on Vitamin K Antagonists). However, this study suggests that if we treat the underlying comorbidities effectively and ensure equitable OAC prescription, the ‘sex gap’ in stroke risk may be bridged.

The failure of the CHA2DS2-VA score to outperform the original CHA2DS2-VASc in reclassification suggests that clinicians should remain cautious about oversimplifying risk models. The ‘S’ in CHA2DS2-VASc serves as a vital reminder to consider the holistic risk profile of the female patient.

Conclusion

In this large European cohort, the increased residual risk of thromboembolism in women with AF was non-significant after rigorous adjustment for clinical confounders and mortality. The study identifies a significant under-prescription of OACs in women, which remains a primary target for clinical intervention. Furthermore, the CHA2DS2-VASc score remains a robust tool, showing superior reclassification compared to the sex-neutral CHA2DS2-VA score. Future guidelines should continue to emphasize the importance of sex-specific considerations while striving for equitable treatment standards across all AF populations.

References

Mei DA, Romiti GF, Vitolo M, Imberti JF, Corica B, Mantovani M, Bonini N, Marin F, Diemberger I, Dan GA, Potpara T, Proietti M, Lip GYH, Boriani G. Atrial fibrillation and female sex: use of oral anticoagulants in a large European cohort and residual risk of thromboembolism and stroke. Eur Heart J Qual Care Clin Outcomes. 2025 Dec 19;11(8):1329-1339. doi: 10.1093/ehjqcco/qcaf075. PMID: 40755396.