Screening Diabetes Patients for NAFLD With Controlled Attenuation Parameter and Liver Stiffness Measurements

1. Patient recruitment: eligible patients in the department of endocrinology and department of gastroenterology are encouraged to participate in this study. 2. Non-diabetic patients admitted to the department of gastroenterology for other gastroenterological diseases. If available, these patients' data will be collected as control groups. Each diabetic patient is matched with a non-diabetic patient by factors of age, gender and BMI. 3. Hepatic steatosis and fibrosis are diagnosed via controlled attenuation parameter (CAP) and liver stiffness measurement by FibroScan ®, respectively. According to the meta-analysis of individual patient data published recently, the CAP in dB/m will be adjusted by deducting 10dB/m from the CAP value for NAFLD/NASH patients, 10dB/m for diabetes patients and deducting/adding 4.4dB/m for each unit of BMI above/below 25kg/m2 over the range of 20-30kg/m2. 4. Regular laboratory tests are performed, including complete blood counts, liver function, renal function, coagulation function, glycosylated hemoglobin, lipid metabolism, urine albumin/creatinine ratio, 24-hour urine albumin. 5. Anthropometric data consists of height, weight, body mass index (the weight in kilograms divided by the square of the height in meters) and waist circumference. 6. The diabetic duration, diabetic complications, the grading of hypertension, and the history of anti-diabetic drugs are gathered from medical records. 7. The data are collected via EpiData software. They are input into computer twice by a single researcher to avoid typing errors. 8. Site monitoring and auditing: As a project for postgraduate thesis, this study is monitored and audited by the school of postgraduate. The research records will be checked in December, 2016, as a mid-term inspection 9. Statistical analysis plan: data are summarized and presented using appropriate descriptive statistics. The normality of continuous variables is assessed by skewness statistic and graphically by normal probability plot. Patients characteristics between those with and without diabetes, and in diabetes group, between those with and without elevated CAP are compared using independent t test, Chi-square, or Fisher's exact tests as appropriate.