Oral anticoagulant use and incident dementia in atrial fibrillation by baseline LIBRA score: a population-based cohort study using the 45 and up study.

backgroundOral anticoagulants (OACs), particularly direct oral anticoagulants (DOACs), may lower dementia risk, but existing studies are constrained by incomplete dementia ascertainment, short follow-up, and lack of evaluation of individual agents or baseline lifestyle-related risk profiles. We examined the association of OAC use and incident dementia in atrial fibrillation (AF) and its variation by LIfestyle for BRAin Health (LIBRA) score.

methodsIn this population-based cohort using linked 45 and Up Study data, participants with a first AF diagnosis between 1 January 2010 and 30 June 2022 were included, excluding those with prior AF or dementia (screened 2004-2009). OAC exposure was ascertained from pharmaceutical benefits scheme dispensing records and classified as warfarin or DOACs (dabigatran, rivaroxaban, and apixaban). A six-month landmark period following AF diagnosis was used to define exposure: participants initiating any OAC during this period were classified as OAC users, whereas those without OAC dispensing were classified as non-users. Participants were followed from AF diagnosis until dementia diagnosis, death, or 30 June 2023. Propensity score (PS) based inverse probability of treatment weighting and matching were applied to balance baseline characteristics. Cox proportional hazards models estimated adjusted hazard ratios (HRs) for incident dementia and dementia-related mortality, accounting for demographic, clinical, and lifestyle-related risk factors. Competing-risk analyses were performed to account for mortality.

resultsAmong 21,671 AF patients (mean age 74.2 years, 45.9% women) with a mean follow-up of 6.0 (3.3) years, OAC use was associated with a lower risk of incident dementia compared with non-use (HR, 0.77; 95% CI, 0.70-0.85). DOAC use was associated with lower dementia risk (HR, 0.63; 95% CI, 0.56-0.70), whereas warfarin use was associated with higher dementia risk (HR, 1.31; 95% CI, 1.16-1.49). Individual DOAC agents were each associated with reduced dementia risk compared with non-use. In head-to-head comparisons, apixaban was associated with a lower dementia risk than dabigatran (HR, 0.79; 95% CI, 0.64-0.96). Findings were consistent across PS-matched, competing-risk, and high-adherence sensitivity analyses. Associations were similar in women and men (P for interaction = 0.528) and were stronger among participants aged ≥ 75 years (HR, 0.66; 95% CI, 0.60-0.74; P for interaction < 0.001). However, there was no statistically significant evidence that the association differed across LIBRA strata (P for interaction = 0.699) or ORBIT bleeding risk categories (P for interaction = 0.069), suggesting no clear effect modification by either score. Use of DOACs was also associated with lower dementia-related mortality (HR, 0.53; 95% CI, 0.44-0.63).

conclusionsIn people with AF, use of DOACs was associated with a reduced risk of incident dementia and dementia-related mortality, independent of baseline LIBRA score. These findings suggest potential cognitive benefits of DOAC therapy across diverse baseline dementia risk profiles.