Abstract
Non-alcoholic fatty liver disease (NAFLD) is an increasingly prevalent condition, and is often referred to as the hepatic manifestation of the metabolic syndrome. It affects up to 30% of adults and up to 60-70% of those with type 2 diabetes. Diet appears to be a critical factor in the pathogenesis, although what initiates the disease and causes progression is not well understood. In the absence of pharmacological therapy, diet and lifestyle modification aiming for 5-10% weight loss is recommended as the first line of treatment; however there is no particular recommended weight loss approach.
My PhD project consisted of two separate studies. The first, a crossover study, investigated possible dietary triggers of hepatic fatty acid accumulation, that is, dietary energy and macronutrients, in 12 healthy weight premenopausal women (Trigger Study); the effect of two 4-week hypercaloric (excess energy intake and weight gaining) diets and two 4-week isocaloric (similar calorific intake and weight maintaining) diets on liver fat, and markers of NAFLD were compared. The second 6-month study examined the feasibility of two weight-loss approaches in 20 overweight participants with confirmed NAFLD; a dietitian-guided partial meal replacement for 8 weeks followed by regular nurse follow-up, and a dietitian-guided ‘slow and steady’ weight loss approach based on evidence-based diabetes prevention and treatment dietary advice.
TRIGGER Study: What triggers non-alcoholic fatty liver disease – fat, carbohydrate or excess energy?
This crossover study compared the effect of two hypercaloric (excess energy intake personalised to produce a 3-5% bodyweight gain) diets with 35-40%TE from fat (15-25%TE from protein, 35-50%TE from carbohydrate) or 55-60%TE from refined carbohydrate (15-25%Te from protein, 15-30%TE from fat), with two isocaloric diets with the same macronutrient distributions as above, on liver fat and markers of NAFLD. Of the 16 healthy normal weight pre-menopausal New Zealand females aged 20-54 years randomised to one of the two diets, 12 women completed the full study. Participants were randomised to either the hypercaloric or isocaloric diet. The order in which they completed the two different 4-week diets containing different proportions of fat (35-40%) and refined carbohydrates (55-60%) was randomised. There was an 8-week washout period between the experimental diets. At the start and the end of each experimental diet period, liver fat was measured by Proton magnetic resonance spectroscopy (1H-MRS). Blood tests, including liver enzymes, and clinical measures (height, weight, waist circumference) were also taken at these times. Dietary intake was measured by multiple 3-day weighed food records (3d-WFR) and analysed using Kaiculator, a New Zealand based dietary assessment programme. Medians, ranges, interquartile ranges and means were calculated. Baseline dietary patterns included vegetarian, vegan, meat-eating, and high fat diets and median energy intake was 7844 kilojoules (kJ) (Interquartile range (IQR) 6474-8575kJ). The proportion of energy from each of the macronutrients varied between woment: protein from 11-21% with a median of 15%, carbohydrate varied from 24-63% with a median of 50%, and fat varied from 18-50% with a median of 36%. The 5 participants randomised to the hypercaloric diets gained 0.8-4.8% bodyweight, while bodyweight was maintained among those randomised to the isocaloric diets. Median liver fat content increased from 38.2% (IQR -3.2 – 54.1) (absolute median measures: 1.2% to 1.6%) with the hypercaloric diets, and a minimal change (median of 1.3% (IQR -20.3 – 15.0)) (absolute median measures: 1.4% to 1.4%) was observed with the isocaloric diets. The difference between the two groups was close to statistical significance (p=0.07). Both hypercaloric diets increased liver fat content; the high carbohydrate diet by 53.9% and the high fat diet by 35.3%, and there was no statistical difference between the two hypercaloric diets (p=0.14). The isocaloric diets had little effect on liver fat content. Alanine aminotransferase (ALT) levels increased during 80% of the hypercaloric diets, but only in 43% of the isocaloric diets. Participants were followed up after each experimental diet to ensure their weight returned to their pre-study usual weight, and their liver fat was also monitored to ensure it returned to the pre-study level, however this was only after the first experimental diet period and not the second due to cost.
INTERVENTION Study: Non-alcoholic fatty liver disease: The feasibility of two different weight loss dietary approaches
This 6-month feasibility study compared an 8-week low-calorie partial meal replacement diet followed by regular nurse ‘weigh-ins’ for 22 weeks (Optifast diet), and an intensive evidence-based nutritional approach for the treatment and prevention of diabetes (Lifestyle Over and Above Drugs in Diabetes (LOADD) diet). Twenty people aged 20-65 years with confirmed NAFLD were randomised to the Optifast diet (two meal replacements and one light meal daily for 4 weeks, then one meal replacement and two light meals daily for 4 weeks), or the LOADD diet (dietitian-guided weight loss using evidence-based nutrition guidelines for diabetes prevention and treatment). The frequency of intervention visits was the same for both diets. A questionnaire to update medical details, clinical measures, blood samples, a 3 day-Weighed Food Record (3d-WFR) and liver fat measured by H-MRS were completed at baseline, 8 weeks and 6 months. Median body mass index (BMI) at baseline was 35 kg/m² for both groups. Baseline dietary energy intake for the study population was 7209kJ, with 18% TE from protein, 35% TE from fat and 46% TE from carbohydrate. The participant retention for this study was high with only two participants withdrawing, and this was due to unrelated health issues. Both dietary approaches were acceptable to most participants with the exception of one participant finding the Optifast products unacceptable. Median weight loss was higher for the Optifast diet (4.1kg) than the LOADD diet (2.9kg) at 8 weeks (p = 0.047), but the same at 6 months (3.8kg) (p= 0.594). Weight gain at 6-months was observed for one Optifast diet and two LOADD diet participants. Liver fat decreased from 9.5% to 6.6% overall, with a 37.7% reduction in the Optifast diet and a 22.3% reduction in the LOADD diet. Both diets led to weight loss and liver fat reduction at 6 months.
Conclusion: The Trigger study showed that eating in excess of energy requirements increased bodyweight and liver fat content suggesting a prominent role of excess energy in the accumulation of liver fat, the first step in the pathogenesis of NAFLD. This also supports current recommendations to lose weight for the treatment of NAFLD. The INTERVENTION feasibility study demonstrated two potentially effective and acceptable weight loss strategies for NAFLD, both of which showed moderate weight losses over 6 months. These two strategies showed different weight loss patterns over 6 months but the overall weight loss at 6 months was similar. A larger and longer-term follow up study is warranted to assess the effectiveness and sustainability of these strategies.