RESUMO
OBJECTIVES: Although obesity is a major risk factor for nonalcoholic fatty liver (NAFL), not all individuals with obesity develop the condition, suggesting that other factors such as diet may also contribute to NAFL development. We evaluated associations between fructose and total sugar intake and subsequent diagnosis of NAFL in adolescents with obesity and without obesity in a population-based cohort. METHODS: Adolescents participating in the Western Australian Pregnancy Cohort (Raine) Study completed 3-day food records and body mass index measurement at age 14 years. At age 17 years, participants underwent abdominal ultrasound to determine NAFL status. Multivariable logistic regression models were used to analyse associations between energy-adjusted fructose and total sugar intake and NAFL status. Food diaries and liver assessments were completed for 592 adolescents. RESULTS: The prevalence of NAFL at age 17 was 12.8% for the total group and 50% for adolescents with obesity. Fructose intake did not significantly differ between adolescents with or without NAFL in our cohort as a whole. Among adolescents with obesity, those without NAFL had significantly lower energy-adjusted fructose intake at age 14 years compared with those with NAFL (mean ± standard deviation [SD] 38.8 ± 19.8 g/day, vs 55.7 ± 14.4 g/day, P = 0.02). Energy-adjusted fructose intake was independently associated with NAFL in adolescents with obesity (OR [odds ratio] 1.09, 95% CI 1.01-1.19, P = 0.03) after the adjustment for confounding factors. Energy-adjusted total sugar intake showed less significance (OR 1.03, 95% CI 0.999-1.07, P = 0.06). No significant associations were observed in other body mass index categories. CONCLUSIONS: Lower fructose consumption in adolescents with obesity at 14 years is associated with a decreased risk of NAFL at 17 years. Fructose rather than overall sugar intake may be more physiologically relevant in this association.