Risk of Adverse Events in Anticoagulated Patients With Atrial Fibrillation and Nonalcoholic Fatty Liver Disease

Abstract

Background: The clinical impact of nonalcoholic fatty liver disease (NAFLD) in patients with atrial fibrillation (AF) is still controversial.

Aim: To evaluate the 1-year risk of all-cause death, thromboembolic events, and bleeding in patients with AF-NAFLD.

Methods: Retrospective study with a health research network (TriNetX). Patients with AF on oral anticoagulation (OAC) were categorized according to the presence of NAFLD into 2 groups. The primary outcomes were the 1-year risks of (1) a composite cardiovascular outcome (all-cause death, myocardial infarction, stroke, cardiac arrest, and pulmonary embolism) and (2) a composite hemorrhagic outcome (intracranial hemorrhage and gastrointestinal bleeding). Cox regression analysis before and after propensity score matching was used to estimate hazard ratio (HR) and 95% 95% CI,. Sensitivity analyses investigated the risk associated with cirrhosis, thrombocytopenia, and type of OAC (warfarin vs non-vitamin K antagonist oral anticoagulants (NOACs).

Results: We identified 22 636 patients with AF-NAFLD (69 ± 12 years, 46.7% females) and 391 014 patients with AF and without liver disease (72 ± 12 years, 42.7% females). NAFLD was associated with a higher risk of composite cardiovascular (HR, 1.54; 95% CI, 1.47-1.61) and hemorrhagic (HR, 1.56; 95% CI, 1.42-1.72) outcomes. This was consistent also for all the single outcomes. Cirrhotic and thrombocytopenic patients with AF-NAFLD showed the highest risks. Compared to patients with AF-NAFLD on NOACs, those on warfarin were associated with a higher risk of cardiovascular and hemorrhagic outcomes.

Conclusion: In patients with AF, NAFLD is associated with a higher 1-year risk of adverse events, with the risk of adverse events progressively increasing from noncirrhotic to cirrhotic and from nonthrombocytopenic to thrombocytopenic patients. NOACs were associated with a better effectiveness and safety profile compared to warfarin.

Keywords: all-cause death; atrial fibrillation; bleeding; cardiovascular events; non-alcoholic fatty liver disease.

Figure 1.

1-year Aalen–Johansen cumulative incidence curves for all-cause death and cardiovascular and hemorrhagic outcomes.

Figure 2.

Survival curves for the primary outcomes in patients with atrial fibrillation and non-alcoholic fatty liver disease and controls patients without liver disease.