Associations between use of aspirin and magnetic resonance imaging‐derived liver fat and fibroinflammation

BACKGROUND: The effects of aspirin on hepatic steatosis and fibroinflammation are unclear. The study aimed to examine the association between aspirin use and liver magnetic resonance imaging (MRI)-derived liver fat and corrected T1 (cT1). METHODS: We used UK Biobank imaging cohort data. Aspirin use was self-reported at baseline and imaging assessment, and the main exposures were aspirin use at imaging assessment and longitudinal aspirin use patterns (never users, initiators, discontinuers, vs. persistent users). Outcomes were MRI-derived liver fat (%) and cT1 (ms). Multivariable adjustment analyses and inverse probability of treatment weighting (IPTW) analyses were performed, accounting for demographic, lifestyle and clinical factors. RESULTS: We included 36 413 participants (mean age 64.6 years, 51.4% female). Aspirin use at imaging assessment was associated with lower liver fat (-0.35; 95% confidence interval [CI]: -0.51, -0.20) and slightly higher cT1 (5.13; 95% CI: 3.23, 7.03). Analyses on longitudinal aspirin use patterns showed that compared to never users, initiators and persistent users showed lower liver fat (-0.48; -0.69, -0.28) and (-0.24; -0.45, -0.02) and higher cT1 (2.94; 0.38, 5.49) and (8.31; 5.65, 10.97). IPTW analyses showed consistent results. CONCLUSION: In this large population-based cohort, aspirin use was linked to reduced liver fat, but a small, clinically insignificant (i.e.,