Sex differences in thromboembolic risk and anticoagulant management in patients with non-valvular atrial fibrillation: a systematic review

Background Atrial fibrillation (AF) is a common arrhythmia and a leading cause of thromboembolic events. Oral anticoagulants (OACs) reduce the likelihood of thromboembolism. Women with AF may have an increased risk of thromboembolism but have historically been under-treated with OACs. The existing literature is heterogeneous, and there is uncertainty about whether these disparities persist in the contemporary treatment era of direct oral anticoagulants (DOACs). Purpose To evaluate sex differences in thromboembolic risk and OAC management in adults with non-valvular AF (NVAF). Secondary outcomes were major bleeding and mortality. Methods PubMed, CINAHL, and Scopus were systematically searched from 2004 to January 2026 for studies reporting sex-specific outcomes in adults with NVAF. Record screening and data extraction were performed independently by two reviewers. Risk of bias was assessed using the Newcastle-Ottawa Scale (NOS), and the certainty of the findings for thromboembolic outcomes was evaluated using GRADE. Results Thirty-five observational studies (3,648,207 participants) met the inclusion criteria and were generally of moderate-to-high quality (NOS 6-8). Women experienced a higher thromboembolic risk (5 studies; HR 1.18-1.53, RR 1.21), although 11 studies reported no significant sex differences (HR 0.91-1.14, OR 0.72-1.24, RR 1.05). One study demonstrated an elevated risk for older (age >75 years) women (HR 1.14) compared with younger (age ≤55 years) women (HR 0.75), while another study reported an increased risk during hospital admission (HR 1.48 vs 1.00 at discharge). OAC use was lower in women (7 studies; OR 0.59-0.93, RR 0.90), with variation by anticoagulant type (DOACs vs vitamin K antagonists [VKAs]) and hospitalisation status. The risk of major bleeding was similar between sexes across 9 studies (HR 0.68-1.13; OR 0.63), with a lower risk of intracranial haemorrhage in women (2 studies; HR 0.54–0.78). For gastrointestinal bleeding, one study found a higher risk in women using DOACs (HR 1.48) but no clear sex difference among those receiving VKA (HR 0.82), while another study reported lower risk in women overall (HR 0.81). All-cause mortality was lower in women (5 studies; HR 0.56-0.78; OR 0.56), although there was no significant difference in 7 studies (HR 1.00-1.12). There was no clear difference in cardiovascular mortality between the sexes (3 studies; HR 0.90-1.04; OR 1.04). Conclusion In the contemporary treatment era, women with NVAF are at an increased risk of thromboembolic events and have lower OAC use compared with men. There was less consistency in sex differences for major bleeding and mortality outcomes. These findings confirm the persistence of sex-specific disparities, highlighting the need to optimise anticoagulation management and reduce thromboembolic events in women.Figure 1.PRISMA Flow DiagramFigure 2.