Can the CHA2DS2-VA schema be used to decide on anticoagulant therapy in Aboriginal and other Australians with non-valvular atrial fibrillation?

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

The Australasian guidelines recommend use of the CHA2DS2-VA schema to stratify ischaemic stroke risk in patients with non-valvular atrial fibrillation (N-VAF) and determine risk thresholds for recommending oral anticoagulant (OAC) therapy. However, the CHA2DS2-VA score has not been validated in a representative Australian population cohort with N-VAF, including in Aboriginal people who are known to have a higher age-adjusted stroke risk than other Australians. In a retrospective data-linkage study of 49 114 patients aged 24–84 years with N-VAF, 40.0% women and 2.5% Aboriginal, we found that patients with a CHA2DS2-VA score >2 had high annual stroke rates (>2%) that would justify OAC therapy. This occurred regardless of Aboriginal status. Non-Aboriginal patients with a CHA2DS2-VA score of 0 had a mean annual stroke rate of 0.4%, and hence were not likely to benefit from antithrombotic therapy. However, Aboriginal patients with a zero CHA2DS2-VA score had a significantly higher annual stroke rate of 0.9%, and could potentially obtain net clinical benefit from anticoagulation, primarily with the safer non-vitamin K antagonist OAC. We conclude that clinicians can confidently use the CHA2DS2-VA score to make decisions regarding anticoagulation in accordance with stroke risk in patients with N-VAF, except in Aboriginal people in whom the risk score was unable to identify those at truly low risk of stroke.

Original languageEnglish
Pages (from-to)600-603
Number of pages4
JournalInternal Medicine Journal
Volume51
Issue number4
DOIs
Publication statusPublished - Apr 2021

Fingerprint

Dive into the research topics of 'Can the CHA2DS2-VA schema be used to decide on anticoagulant therapy in Aboriginal and other Australians with non-valvular atrial fibrillation?'. Together they form a unique fingerprint.

Cite this