The study is aimed

• To quantify the change of adipose tissues, triglyceride in liver and pancreas and cholesterol after lifestyle intervention or bariatric surgery.
• To test the hypothesis that Brown fat is an independent biomarker for the development of Non Alcoholic Fat Liver Disease (NAFLD)
• To study the association among Brown fat, NAFLD and obesity.

Obesity is associated with different chronic disorders such as metabolic syndrome, non-alcoholic fatty liver disease (NAFLD) and diabetes mellitus type 2 (DM2). It can be classified into mild, moderate and severe using different cut-off values of body mass index (BMI). Many interventions and treatments against obesity have been proposed. Patients with mild to moderate obesity are suggested to undergo lifestyle monitoring or intervention programs while patients with severe or morbid obesity may be suggested to undergo bariatric surgery. Brown adipose tissue (BAT) is a specific type of adipose tissue with unique ability to produce heat by non-shivering thermogenesis. It increases energy expenditure and lower blood glucose levels and therefore it is suggested to be a potential therapy against obesity, NAFLD, DM2 etc.

Research studies demonstrated that obesity is strongly associated with NAFLD. From our previous observation, most obese individuals suffer from NAFLD. Nevertheless, a small portion of them are protected from NAFLD. We try to understand why and how this small portion of severe obese patients can keep themselves protected from NAFLD while some lean individual suffers from it. We would also like to test whether the amount of BAT is a biomarker for NAFLD. Besides, we would like to validate the accuracy of liver inflammation and fibrosis (LIF) scores with liver biopsy results.

The change of adipose tissues (i.e. brown, subcutaneous, visceral), triglyceride in liver and pancreas and cholesterol (i.e. HDL and LDL) in bloodstream would be measured in patients with NAFLD who undergo either bariatric surgery or lifestyle intervention. 40 subjects will be recruited. Half of them for the lifestyle intervention group while half for the bariatric group. Baseline MRI and blood test will be performed before the start of their treatments and two follow-up scans as well as blood tests in six-month and twelve-month intervals.

This study allows objective measurement to evaluate efficacy of different treatments against obesity and the response of NAFLD after weight loss. Furthermore, this study provides quantitative measurement of BAT and tests whether BAT is associated with NAFLD or other metabolites disorders such as DM2 and hyperlipidemia.

b) Research plan and methodology Subjects Eighty volunteers will be recruited for this project. 40 of them with mild-moderate obesity and the other 40 subjects with morbid obesity.

Morbid obesity subjects will be further divided into bariatric group (those who gave consent for surgery i.e. 20 subjects and 20 for lifestyle intervention group.

The mild- moderate obesity group will also be divided into two subgroups (20 in each group) through a randomization process in a 1:1 ratio to participate in the dietitian led life style modification program or receive usual care. The computer will be used to generate a list of random numbers for randomization of the two groups and this will be done in blocks of six by Nursing officers working in the clinic. The treatment assignments will be concealed in numbered sealed envelopes which will only be opened sequentially upon patient enrollment. Blinding methods to treatment assignments will be employed to both clinicians and radiographers during analysis and patient assessments. Inclusion criteria involve individuals between the age of 18-65, BMI ≥ 25-30 kg/m2 for mild-moderate obesity, BMI ≥ 30 kg/m2 with metabolic syndrome or BMI ≥ 35 kg/m2 for morbid obesity, who have been diagnosed with NAFLD. Exclusion criteria include other kind of hepatic diseases or under medications that are known to affect liver fat accumulation, weight≥250 kg, waist circumference≥150 cm. The maximum weight and waist circumference are set due to the tolerance of the scanner and its bore size. Written informed consents will be obtained from all volunteers.

Protocol The protocol in this part of the study will be as described by Wong et al. At the commencement of the study, all the 80 patients will initially be assessed by the dietitian and relevant individual recommendations will be provided. This will include complete behavioral change assessments, covering important areas such as the patient's current eating habits, lifestyles patterns etc.

Surgical group (bariatric surgery) The surgical group will be identified from the surgical team and will be enrolled by the surgical team accordingly.

Lifestyle Modification Group Patients randomized to the intervention group will participate in a dietitian-led lifestyle modification programme for 12 months. The programme will be held at 2 urban centres that are open to the public for the management of obesity and related disorders. The programme will be based on a strategy of increasing energy expenditure and reducing caloric intake using lifestyle behavioral change to achieve long-lasting impact. The patients will attend dietary consultation sessions weekly in the first 4 months, and monthly in the following 8 months. At the first session (about 1 h), the dietitian will carry out a complete behavioral assessment, covering important areas such as the patient's current eating and lifestyle patterns, specific eating-related behaviors, knowledge of risks associated with current eating patterns, concerns and feelings about specific lifestyle changes. The dietitian will also discuss the expected duration and specific dietary and lifestyle advices to achieve a desirable weight status with the patients.

In the follow-up sessions (about 20 minutes), the dietitian will review the patient's dietary practice and provide recommendations. Each patient will be given an individualized menu plan. The dietary component and portion sizes of the menu plan will be based on the recommendations of the American Dietetic Association . A varied balanced diet with an emphasis on fruit and vegetables, and moderate-carbohydrate, low-fat, low-glycemic index (GI) and low calorific products in appropriate portions will be encouraged. This diet will result in a relative increase in energy consumption from proteins, which also will promote satiety. Each patient will be provided with two booklets, one for food portion size exchange and tips for eating out, and another listing the low-GI food options and meal plans (GI <55). Furthermore, techniques for coping at-risk situations such as parties and festival celebrations will be taught. Recipes will also be provided to the patients to encourage healthy cooking. Adherence to dietary intervention will be assessed by calculating the percentage attendance to the intervention sessions and evaluating the dietary intakes and meal patterns using a weekly food record.

Besides, patients will be encouraged to see an exercise instructor at least once during the lifestyle modification programme. During the first exercise consultation (about 30 min), the exercise instructor will review the patient's medical history and exercise habits, and design a suitable exercise regime for the patient. The patients will first be instructed to do moderate intensity aerobic exercise for 30 min, 3 to 5 days a week and encouraged to increase daily physical activities.

During subsequent appointments, the exercise instructor will evaluate the patient's exercise progress on aerobic exercise and stretching during follow-ups. When the patients will be able to develop a routine exercise habit, they will be instructed to perform resistance training to increase their muscle endurance and strength for better aerobic performance and liver fat reduction (30). The intensity of exercise will be gradually increased to 30 minutes every day. The target will be a reduction of body mass index (BMI) towards 23 kg/m2.

Control group Patients in the control group will receive routine care at the medical clinic of the Prince of Wales Hospital, Hong Kong. At baseline, a clinician will explain the laboratory test results and the natural history of NAFLD to the patients. The patients will be encouraged to reduce carbohydrate and fat intake, and to exercise for at least 3 times per week, 30 minutes per session.

Follow up assessments in all groups The patients will attend the clinic at months 4 and 12 for metabolic assessment, and receive further advice from a clinician as appropriate. Before each visit, patients will be asked to fast overnight for at least 8 hours, then blood samples will be taken for liver biochemistry, glucose, insulin and lipids. Insulin resistance will be estimated using the homeostasis model assessment (HOMA-IR) calculated as fasting plasma glucose (mmol/L) x insulin (mIU/L) / 22.5 and quantitative insulin-sensitivity check index (QUICKI), calculated as 1/ [log fasting insulin (μU/mL) +log fasting glucose (mg/dL)]). Alanine aminotransferase (IU/L), fasting glucose (mmol/L), hemoglobin A1c (%), total cholesterol LDL, HDL(mmol/L) and triglycerides (mmol/L) will also be measured. Additionally, Weight, height will be measured during their clinical visit before each of their MRI scan and BMI will be calculated by weight (kg) / height^2 (m). Waist circumference will be measured at a level midway between the lower rib margin and iliac crest with the tape all around the body in the horizontal position. Physical activities will be recorded as the total duration of active exercise (min) per week.

Image Acquisition Subjects from all the groups will undergo MRI at baseline before their intervention and two follow-up scans at 4-months and 12-months intervals. All scanning will be performed using a Philips Achieva 3.0 Tesla MRI Scanner (Philips Medical System, Best, The Netherlands). Chemical-shift water-fat images will be acquired by the 3D spoiled multi-echo modified DIXON sequence with a 16-channel SENSE-XL-Torso array also from Philips Healthcare. Imaging parameters as follow: Repetition Time(TR) = 5.7-5.9 (ms), Echo time (TE)/echo spacing = 1.2-1.4 (ms) / 1.0-1.2 (ms), number of echoes = 6, flip angle = 3°, SENSE acceleration = 2, reconstructed slice thickness/number of slices = 3.0 mm / 50 to yield co-registered water, fat, fat-fraction and T2* image series. Non-breath-hold acquisition will be performed from the base of the skull to the base of the thoracic cavity for BAT measurement and breath-hold acquisition will be performed from the dome of the diaphragm to the pubic symphysis for abdominal SAT and VAT measurements.

T1 liver images will be acquired using 3D T1-fast-field echo (FFE) sequence, 2-echoes: TE1=1.8 msec, TE2=4.0 msec, TR=5.2 msec, flip angle=15, SENSE parallel imaging with acceleration factor 2.0 in phase-encoding direction.

In vivo quantification of liver fat will be performed using single voxel STEAM spectroscopy (TE=15 ms, TR=5000 ms, Number of signal avaregaes (NSA)=24, spectral width=2000 Hz, no water suppression). A 30×30×30-mm^3 (27 mL) voxel will be located on the right liver lobe to avoid major vessels according to the institutional protocol for hepatic fat measurement. Short TE and long TR will be selected to minimize T2 and T1 effects. Non-breath hold scan will be carried out and the acquisition time will be approximately 2 min.

Quantitative Measurement of Brown and Abdominal White Adipose Tissues Volume Brown adipose tissue (BAT) will be extracted from MR images using an in-house algorithm that has been validated and published in a recognized domain journal. This method utilizes the unique features of fat-fraction and T2* values in BAT to segment the tissue. Volume of BAT will be measured in mm^3 and be converted to mL.

Abdominal white adipose tissue will be separated into subcutaneous and visceral adipose tissues (SAT and VAT) using another in-house algorithm which has also been validated and published in a domain journal. Volume will be measured in mililitre (ml) and region of interest will be covered from the dome of diaphragm to the pubic symphysis.

Spectrum Analysis for Intrahepatic Triglycerides (IHTG). Magnetic resonance spectroscopy (MRS) data will be exported for offline time-domain analysis using the AMARES algorithm with Gaussian line shapes fitting in jMRUI software package. IHTG will be measured for relative fat signal integrals in terms of a percentage of the total signal amplitude. IHTG content is calculated as [signal Intensity of fat/(Signal Intensity of fat+Signal Intensity of water)] x 100. The definition of NAFLD will be based on liver fat content higher than the threshold of 5.5%.

Liver Inflammation and Fibrosis Liver inflammation and fibrosis (LIF) scores based on iron-corrected T1 (cT1) and T2* values will be calculated using LiverMultiScan™ (LMS, Perspectum Diagnostics, Oxford, UK). LIF scores range from 0-4 which 0 indicates the least severe and 4 as the most severe as show in figure 3. To validate the LIF scores, liver biopsy will be performed during the bariatric surgeries for assessment of degree of inflammation and fibrosis within the liver in association with NAFLD.

Statistical Analysis Within groups comparison will be measured using repeated measures ANOVA for results acquired from 3 different time points (baseline, Month 4 and Month 12). Variables with significant differences will be carried forward for pairwise comparison with bonferroni correction. Between groups comparison (life-style intervention and bariatric groups) will be measured in relative changes between baseline to Month 4 and Month 4 to Month12. Statistical analyses will be conducted using Statistical Package for the Social Sciences (SPSS) 25 (IBM, New York , USA). Significant level is selected at p < 0.05 (2-tailed).

Inclusion Criteria:

• BMI ≥ 25-30 kg/m2
• BMI ≥ 30 kg/m2 with Metabolic syndrome
• BMI ≥ 35 kg/m2 for morbid obesity

Exclusion Criteria:

• Other kind of hepatic diseases
• Under medications known to affect liver fat
• Waist circumference ≥ 150 cm
• Weight ≥ 250 kg
• MRI contraindications