Author(s): Daniela Liccardo 1 , Anna Alisi 2 3 , Gilda Porta 4 , Valerio Nobili [*] 5 6
dietary pattern; food frequency; nonalcoholic fatty liver disease; obesity
Methods & results
This study includes adolescents participating in the Western Australian Pregnancy Cohort (Raine) Study. At age 14 years, 1613 adolescents from this cohort had completed, with their primary caregivers, a food frequency questionnaire (FFQ) that permitted to distinguish between Western and health dietary patterns, mainly characterized by a difference in fat and sugar intakes. At the 14-year follow-up, time spent watching television and hours of physical activity per week were assessed using a self-reported written questionnaire. Liver ultrasonography was performed at the 17-year follow-up in 1170 adolescents of those with a reasonable FFQ. Information on alcohol intake and medications over the past 12 months was obtained from the FFQ and lifestyle questionnaire, and secondary causes of NAFLD through a careful collection of medical history were excluded.
Auxological parameters, including body weight, height and waist circumference (WC) were measured at 14 and 17 years by experienced operators, and central obesity was defined as WC of 80 cm in females and 94 in males, accordingly to the International Diabetes Federation, criteria  . At 14 years, BMI was also evaluated to categorize adolescents into underweight, normal weight, overweight and obese categories using International Obesity Task Force criteria at 14 years [2,3] .
The χ2 tests were used to compare the characteristics of study participants at 14 years (sex, family income, BMI status, dietary misreporting, physical activity per week and sedentary behavior), according to NAFLD diagnosis at 17 years. At age 17 years, 995 adolescents with a complete FFQ presented ultrasound assessment of NAFLD, and 151 of them had the disease with 54.1% of obesity/overweight observed at 14 years.
Biochemical parameters, including serum levels of insulin, glucose, triglycerides, high-density lipoprotein cholesterol (HDL-C), alanine transaminases, [gamma]-glutamyltransferase and high-sensitivity C-reactive protein, and homeostasis model assessment for insulin resistance were measured at 17 years. Statistical analysis by independent sample t-test showed that subjects with NAFLD had increased triglyceride and reduced HDL-C levels, elevated values of alanine transaminase, [gamma]-glutamyltransferase and high-sensitivity C-reactive protein and higher adiposity (BMI, WC and waist-to-hip ratio) and homeostasis model assessment for insulin resistance than no NAFLD adolescents.
Each subject received a z-score for each dietary pattern, indicating how closely their reported dietary intake corresponded with the two patterns. The Goldberg method was applied to estimate dietary misreporting and to classify respondents as likely under reporters, plausible reporters or over reporters.
The associations between the dietary pattern scores at 14 years and the presence of NAFLD at 17 years were evaluated by using logistic regression analysis and presented for three different models. The first model included odds ratio (OR) adjusted for dietary patterns, sex and dietary misreporting, whereas, a second model was additionally adjusted for family income, frequency and intensity of physical activity and sedentary behavior at 14 years. The healthy dietary pattern showed no association with NAFLD both in model 1 and model 2. On the contrary,...