Non-vitamin K oral Anticoagulants and Risk of Fractures

A Systematic Review and Meta-analysis

Pajaree Mongkhon; Laura Fanning; Kirstie H.T.W. Wong; Kenneth K.C. Man; Ian C.K. Wong; Wallis C.Y. Lau


Europace. 2021;23(1):39-48. 

In This Article

Abstract and Introduction


Aims: Comparative fracture risk for non-vitamin K antagonist oral anticoagulants (NOACs) and vitamin K antagonists (VKAs) among patients with atrial fibrillation (AF) remains unclear. This study aimed to provide summary relative risk (RR) estimates for associations between NOACs vs. VKAs and fracture risk.

Methods and Results: PubMed, EMBASE, and Cochrane Library were searched from 2010 to 26 May 2020. Observational studies investigating the association between NOACs vs. VKAs and fracture risk in patients with AF were included. The adjusted effect estimates were pooled using the DerSimonian–Laird random effects models. The Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) and the Meta-analysis of Observational Studies in Epidemiological (MOOSE) guidelines were followed. Five observational studies comprising 269 922 patients and 4289 fractures were included. Non-vitamin K antagonist oral anticoagulants use was associated with a lower risk of any fractures compared to VKAs use, with moderate heterogeneity [pooled RR = 0.83, 95% confidence interval (CI): 0.75–0.92, P < 0.001, I 2 = 73.0%]. When comparing individual NOAC to VKAs, a statistically significant lower risk of any fractures was found for rivaroxaban (pooled RR = 0.79, 95% CI: 0.71–0.88, P < 0.001, I 2 = 55.2%) and apixaban (pooled RR = 0.75, 95% CI: 0.60–0.92, P = 0.007, I 2 = 54.5%), but not dabigatran (pooled RR = 0.87, 95% CI: 0.74–1.01, P = 0.061, I 2 = 74.6%). No differences were observed in all head-to-head comparisons between NOACs.

Conclusion: This large meta-analysis suggests that NOACs use was associated with a lower risk of fractures compared with VKAs. Fracture risks were similar between NOACs. These findings may help inform the optimal anticoagulant choice for patients with AF at high risk of fracture.


Atrial fibrillation (AF) is a common aged-related cardiac arrythmia associated with increased risk of cardioembolic stroke.[1,2] Disability and premature death due to stroke of AF can be mitigated through the use of vitamin K antagonists (VKAs), including warfarin, or non-vitamin K oral anticoagulants (NOACs). For decades, VKAs were the only anticoagulants available for stroke prevention in AF, until the approval of the NOACs, including dabigatran, rivaroxaban, apixaban, and edoxaban since 2010. Non-vitamin K antagonist oral anticoagulants demonstrated at least non-inferiority compared with warfarin for the prevention of stroke or systemic embolism[3–6] and have been associated with lower risk of bleeding complications in both the landmark randomized controlled trials (RCTs),[3–6] and in post-marketing observational studies.[7,8]

Atrial fibrillation is a condition of ageing, and ageing is inherently linked to an increased risk of falls, therefore, osteoporotic fracture is an important clinical concern in older people. Warfarin, through its effect on vitamin K, decreases osteocalcin content in bone.[9] Ultimately, links between warfarin use, low bone mineral density, and increased risk of osteoporotic fracture have been established.[10–12] Non-vitamin K antagonist oral anticoagulants are not known to impair bone quality, but uncertainty remains regarding the comparative risk of fracture with NOACs vs. warfarin. Observational studies[13,14] and also a meta-analysis of adverse fracture events reported in RCTs[15] demonstrated lower fracture risk with NOAC treatment compared to warfarin. However, a meta-analysis of observational studies[16] has raised questions regarding fracture risk in patients using NOACs vs. warfarin. This meta-analysis of four studies demonstrated no difference in fracture risk for NOACs vs. warfarin. However, the meta-analysis has been criticized for inappropriate definition of the reference group and inconsistent use of adjusted and unadjusted data in the analyses.[17] Given the uncertainty of these findings, coupled with the availability of new data,[14,18–20] the objective of this study was to systematically investigate the osteoporotic fracture risk with NOACs vs. VKAs in patients with AF and to summarize the data using meta-analysis.