LAUSR

Introduction The purpose of this study was to investigate associations between intake of ultra-processed food (UPF) and liver fat, pancreas fat and visceral adipose tissue (VAT) but also subcutaneous adipose tissue (SAT), VAT/SAT ratio and total fat mass. Materials and Methods Cross-sectional analysis of n = 286 50-year old men and women. Energy percentage (%E) from UPF was calculated from a semi-quantitative food frequency questionnaire. Food items were categorized according to the NOVA-classification system and fat depots were assessed using magnetic resonance imaging (MRI) and bioelectrical impedance analysis (BIA). Associations were analyzed using linear regression, adjusted for sex, education, physical activity, smoking, dietary factors and BMI. Results Mean intake of UPF was 37.8 ± 10.2 %E and the three largest contributors to this were crisp- and wholegrain breads and spreads, indicating overall healthy food choices. Consumption of UPF was associated with higher intake of energy, carbohydrates and fiber and lower intake of protein and polyunsaturated fat but no differences were observed for total fat, saturated fat (SFA), monounsaturated fat, sugar or alcohol between tertiles of UPF. Intake of UPF was positively associated with liver- and pancreas fat, VAT, VAT/SAT and inversely associated with total fat mass in crude models. The association for VAT remained after full adjustment (β = 0.01 (95% CI: 0.002, 0.02), P = 0.02) and was driven by women. Conclusion Energy intake from UPF is not associated with ectopic fat, SAT or total fat after adjustment for multiple confounders in this population having overall healthy food habits. However, a positive association between UPF and VAT was observed which was driven by women.