RESUMO
INTRODUCTION: The risk of liver injury in patients with atrial fibrillation (AF) using nonvitamin K antagonist oral anticoagulants (NOACs) has not been previously examined using liver function tests as the primary outcome in the real-world setting. This study assessed the association between NOACs (dabigatran, rivaroxaban, and apixaban) and warfarin and the risk of liver injury, as defined by laboratory tests. METHODS: Patients newly diagnosed with AF and prescribed NOACs or warfarin between 2010 and 2016, identified using the Hong Kong Clinical Database and Reporting System, were matched on age, sex, health status scores, comorbidities, and medications by propensity score on a 1:1 ratio. Risk of liver injury, defined as laboratory test values >3 times the upper limit of normal of alanine aminotransferase or aspartate aminotransferase and >2 times the upper limit of normal of total bilirubin, was compared between NOAC and warfarin users using Cox proportional hazards regression. RESULTS: After propensity score matching, 13,698 patients were included, of which 141 (2.1%) NOAC users and 232 (3.4%) warfarin users developed liver injury. The hazard ratio (HR) for NOAC vs warfarin users was 0.71 (95% confidence interval: 0.58-0.89). When comparing individual NOACs, only dabigatran (hazard ratio: 0.63; 95% confidence interval: 0.48-0.82) was associated with a lower risk of liver injury. DISCUSSION: Among patients with AF, NOACs as a group, and dabigatran alone were associated with a significantly lower risk of laboratory-based liver injury when compared with warfarin. However, liver injury occurs more frequently in real-world practice than in NOAC randomized controlled trials.