Dietary fatty acids and their longitudinal associations with fatty liver disease in adolescence

Fats are an important source of energy in the human diet. Fatty acids are the main components of fat, determining its biochemical properties. Dietary fatty acids are involved in many important processes in the human body, with established evidence showing that specific dietary fatty acid intake is closely related to metabolic activity. Fat accumulation is the central feature of fatty liver disease (FLD) and is the result of a number of factors of which dietary fatty acid intake has been shown to be strongly associated. An increasing prevalence of FLD in children and adolescents associated with overweight and obesity has been observed worldwide, which is usually unlikely to be the result of excessive alcohol consumption. Non-alcoholic fatty liver disease (NAFLD) is by far the most prevalent form of FLD among these populations affecting up to one-fifth of the global paediatric population based on different geographical and age groups. There are cost and ethical limitations to accurately quantifying NAFLD outcomes in the general population, making risk prevention and screening for diagnosis a challenge. Accurate and inexpensive non-invasive diagnostic scores such as the Fatty Liver Index (FLI) are required due to a lack of simple diagnostic tools. An FLI is potentially useful for monitoring changes in the extent of fatty liver in clinical practice. Based on the literature, a longitudinal relationship between dietary fatty acids and FLD in adolescents has been understudied; therefore, the aim of my thesis was to understand the longitudinal relationship between dietary fatty acids and adolescent FLD and the pathways through which these effects may occur. Three key objectives were addressed in the thesis as follows: 1) To discover the extent of FLD in an adolescent population as measured by a fatty liver disease index. 2) To determine dietary fatty acid intake in an Australian adolescent population. 3) To investigate the association between dietary fatty acid intake and longitudinal FLD risk. Data for this thesis were obtained from the Raine Study; a Western Australian pregnancy cohort followed up prospectively to 14, 17, 20 and 22 years of age. At birth, the study had 2868 enrolled infants who were followed at regular intervals throughout childhood, adolescence and young adulthood. At 14 years, dietary fatty acid data were collected using a semi-quantitative Food Frequency Questionnaire (FFQ), and fasting blood samples were collected and tested for erythrocyte fatty acids. At 17 years, a clinical diagnosis of FLD was assessed by ultrasound, and at 17, 20 and 22 years, anthropometry, and fatty liver biomarkers (gamma-glutamyl transferase, body mass index, waist circumference and triglycerides) were assessed to create the FLI. In addition, a wide range of potential covariate data was collected throughout the study. The first study was a validation analysis of adolescent FLD using the fatty liver index at the 17-16 year follow-up. We found that the prevalence of ultrasound-defined NAFLD in our population 17 was approximately 12% at the age of 17 years and showed that FLI as a continuous variable in 18 population-level screening was a good predictor of the risk of FLD for adolescents in this 19 population. The second study was an agreement analysis of dietary polyunsaturated fatty acids (PUFA) intake and erythrocyte fatty acids among the adolescents. For the major omega-3 (n-3) and omega-6 (n-6) fatty acids, there was high agreement between FFQ derived intake and erythrocyte levels, although some bias at high fatty acid intake levels was observed. The FFQ can be used as a tool to identify individuals with inadequate long-chain fatty acid levels, such as docosahexaenoic acid (DHA). The main dietary source of n-3 long-chain-PUFA (LC-PUFA) was 'fresh fish' (53% LC-PUFA) food group. The highest intake of n-3 PUFA was mainly from the 'butter, margarine' and 'savoury dishes, soup, stews ' food groups. The highest dietary n-6 PUFA intake was found to come from the 'butter, margarine' and 'cereals, breads' food groups. Finally, in the third study we identified fatty liver index (FLI) trajectories from adolescence to young adulthood and examined whether dietary fatty acids increased the risk of developing a specific fatty liver index (FLI) trajectory. Broadly, we found that a diet high in n-6 fatty acids or with a high n6: n3 ratio at 14 years was associated with an increased risk of FLD (as measured by the FLI trajectory) from adolescence to young adulthood. The prospective study design is a strength of the study, and the results are important given the lack of previous longitudinal data reported on the role of diet in the risk of FLD, particularly from adolescence to young adulthood. In this thesis, I contributed new knowledge through the following: 1) validation of a reproducible measurement epidemiological screening tool for fatty liver disease in the study population from a methodological perspective, addressing the difficulty of obtaining longitudinal quantitative data on fatty liver disease. This was followed by 2) the validation and analysis of dietary questionnaire data with biomarkers, which reinforced the understanding of the nutritional intake status of the study population. Finally (3), I innovatively constructed a natural history of the development of trajectories of FLD using the FLI from adolescence to young adulthood and considered the association of prospective dietary fatty acid intake with these trajectories. The findings of the three main studies of this thesis indicate that I have addressed our aim and objectives and that the development and progression of adolescent FLD into young adulthood is associated with validated dietary fatty acid exposure. Observational evidence that diets high in n-6 fatty acids or diets with a high n6: n3 ratio may be associated with a high risk of future FLD is provided. This study shows that disease risk profiles from adolescents in Western Australia follow a specific trajectory that is associated with the n6: n3 ratio in the diet. Understanding this relationship could help to reduce the incidence of FLD and improve progression in adolescents with FLD. Future nutritional epidemiological studies should focus on pathways associated with FLD by n-6 fatty acids, such as the mediation of the regulatory role of metabolic factors in biochemical approaches. In addition, if available, other studies could validate the relationship between fatty acids and fatty liver disease-related outcomes in larger or different adolescent populations.