Comparison between non-vitamin K antagonist oral anticoagulants versus warfarin for the risk of dementia in Asian patients with non-valvular atrial fibrillation

Abstract Background Atrial fibrillation (AF) is a risk factor for dementia and oral anticoagulant (OAC) use is associated with a decreased risk of dementia in patients with AF. Objective We sought to find whether the risk of dementia would be different between patients treated with non-vitamin K antagonist oral anticoagulants (NOACs) and those treated with warfarin. Methods Using the Korean National Health Insurance database from January 2014 to December 2017, we identified OAC naïve non-valvular AF patients aged 40 years or older. For the comparison, warfarin and NOAC groups were balanced using the inverse probability of treatment weighting method (IPTW). The primary outcome was incident dementia. Patients were censored at the occurrence of dementia, discontinuation of index treatment, or the end of the study period (December 31, 2018). Results Among 72,846 of total study patients, 25,948 were treated with warfarin and 46,898 were treated with NOAC (17,193 with rivaroxaban, 9,882 with dabigatran, 11,992 with apixaban, and 7,831 with edoxaban). Crude incidence of dementia was 4.65 per 100 person-years in warfarin group and 4.98 per 100 person years in NOAC group. Baseline characteristics were well-balanced after IPTW (mean age 72 years, 93% CHA2DS2-VASc ≥2, 58% men). Compared to warfarin, NOAC showed a comparable risk of dementia (hazard ratio [HR] 0.944, 95% confidence interval [CI] 0.934–1.059), vascular dementia (HR 0.921, 95% CI 0.814–1.043) and Alzheimer dementia (HR 1.101, 95% CI 1.019–1.191). When comparing individual NOACs with warfarin, edoxaban was associated with a lower risk of dementia (HR 0.830, 95% CI 0.740–0.931). Rivaroxaban (HR 0.767, 95% CI 0.648–0.907) and edoxaban (HR 0.786, 95% CI 0.620–0.996) were associated with a lower risk of vascular dementia than warfarin. In subgroup analyses, NOAC was associated with a lower incidence of dementia than warfarin particularly in patients with prior stroke (HR 0.891, 95% CI 0.820–0.968) and in patients aged 65–74 years (HR 0.815, 95% CI 0.709–0.936). Conclusion In this large Asian population with AF, NOAC showed a comparable result with warfarin on the risk of dementia. Edoxaban was associated with a lower risk of dementia. NOAC appeared more beneficial especially in those with AF who had a history of prior stroke and aged 65–74 years. Funding Acknowledgement Type of funding source: None