Abstract
BACKGROUND: This study aimed to investigate how body composition, particularly adiposity distribution, mediates the association between circulating fatty acids (FAs) and the incidence of severe non-alcoholic fatty liver disease (NAFLD). METHODS: In the UK Biobank cohort, plasma FAs were measured using nuclear magnetic resonance spectroscopy in 107,593 participants at baseline. Cox regression models were utilized to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between FAs and the incidence of severe NAFLD. Additionally, mediation analyses were conducted to quantify the mediating effects of different body composition indices in these associations. Cross-sectional association of FAs with metabolic-associated fatty liver disease (MAFLD) and metabolic dysfunction-associated steatotic liver disease (MASLD) were also explored by multivariate logistic regression. RESULTS: During an average follow-up period of 13.1 years, 1,480 severe NAFLD cases were identified. In multivariable analyses, the HRs per SD and CIs were 0.76 (95% CI: 0.68-0.86) for plasma polyunsaturated fatty acids (PUFA), 1.28 (95% CI: 1.12-1.46) for monounsaturated fatty acids (MUFA), and 1.20 (95% CI: 1.10-1.32) for saturated fatty acids (SFA) (all P < 0.001). Results for MAFLD and MASLD were comparable. SFA and MUFA were positively correlated, while PUFA was inversely correlated to all body composition indices. Mediation analysis revealed that central obesity indices, such as waist circumference and body roundness index, mediated 19-36% of the associations between FAs and severe NAFLD (all P < 0.001). Although other body composition indices also exhibited significant mediating effects, their contributions were relatively smaller. Specifically, leg fat mass index mediated around 13-26% and arm fat mass index mediated around 9-21% of the association between FAs and severe NAFLD. CONCLUSION: This study highlights that body compositions, particularly central obesity, may partially mediate the relationship between plasma FAs and the incidence of severe NAFLD, suggesting that dietary interventions, when integrated with broader strategies to reduce central adiposity, could be critical for preventing NAFLD.