| PART I Prevalence of Non-alcoholic Fatty Liver Diseases in Middle-aged and Elderly Population of Shanghai Changfeng CommunityObjectiveThis study was designed to estimate the prevalence of non-alcoholic fatty liver disease (NAFLD) and its distribution in different genders and age groups among the middle-aged and elderly population of Shanghai Changfeng Community, by utilizing the quantitative liver ultrasound to diagnose NAFLD.MethodsA cross-sectional survey was performed in 4576 citizens recruited from the Shanghai Changfeng Community from May 2010 to June 2012. The history information, demographic information, past history, anthropometric parameters, biochemical parameters and liver ultrasound were collected in all subjects. Alcohol abuse history, other previous liver diseases and serum biomarkers of viral hepatitis were specially and carefully evaluated. The liver fat content (LFC) over 9.15% indicated by the quantitative liver ultrasound was diagnosed as a fatty liver.Results1. There were 4178 subjects with complete clinical and laboratory data enrolled in the study. The mean age of the population was 62.7±9.6.2. Applying the quantitative liver ultrasound, the median liver fat content was 5.48%. The median liver fat content of male and female was 5.15% and 5.73% respectively.3. The prevalence of fatty liver was 33.3%. The general prevalence of NAFLD in this population was 26.7%.4. In male subjects, the NAFLD prevalence decreased as the age increased, and the peak prevalence was 27.6% in the age group of 45-50 years old. In female, the NAFLD prevalence increased with the increasing age, and reached the peak of 38.5% in the age group of 65-70, and then decreased as the age grew.ConclusionAccording to the quantitative liver ultrasound results and NAFLD diagnostic criteria, fatty liver is found in over one third of the middle-aged and elderly residents in Shanghai, and 26.7% was NAFLD. The median liver fat content was 5.48%. The prevalence of NALFD was fairly high in the Chinese population, badly in need of early diagnosis and treatment. Quantitative liver ultrasound was an accurate and reliable method to diagnose NAFLD.PART II Establishment of A Non-invasive Scoring Equation to Predict NAFLDObjectiveIt is of great importance of early and accurate diagnosis of NAFLD. This study was designed to establish a non-invasive scoring equation and evaluate its diagnostic power in predicting NAFLD, in order to provide an alternative to diagnose NAFLD and estimate NAFLD prevalence, by applying the population data and NAFLD condition of Shanghai Changfeng community.MethodsAfter excluding subjects with alcohol abuse history,other previous liver diseases and history of viral hepatitis,this study recruited 3430 subjects of Changfeng community from May 2010 to June 2012. The clinical history, past history, demographic information,anthropometric parameters, routine biochemical parameters and liver ultrasound were collected in all subjects, and those whose liver fat content(LFC) was over 9.15% indicated by the quantitative liver ultrasound were diagnosed as NAFLD patients. Subjects were randomly divided into estimation group(n=1729) and validation group(n=1701). We applied univariate and multivariate logistic regression analysis to assess the association of the parameters mentioned above with LFC and NAFLD, and to establish the NAFLD scoring equation, with quantitative liver ultrasound as the diagnosing criteria. We then evaluated the diagnostic power of the newly built equation, by utilizing the area under receiver operating curve(AUROC) and its comparison with the Finnish NAFLD fat score.Results1.Univariate and multivariate logistic regression analysis showed that the presence of type2 diabetes(T2DM), increasing BMI, fasting serum insulin(fs-insulin), fasting serum triglyceride(fs-TG), and decreasing AST/ALT ratio, were independently associated with NAFLD.2.The non-invasive NAFLD scoring equation we established was the following; score:--5.545+0.010*fs-insulin+0,256*T2DM(yes=2,no=0)+0.199*BMI+0.417*fs-TG-0732 xAS T/ALT。3.The AUROC of this equation in the estimation group was 0.78±0.01, and optimal cut-off was determined as-0.92. Values higher than-0.92 predicted NAFLD with sensitivity of 77%and specificity of 65.6%. In the validation group, the AUROC of the equation was0.77土0.01, and values higher than-0.92 predicted NAFLD with sensitivity of 79.2% and specificity of 61.1%. The AUROC of Finnish NAFLD fat score was 0.75士0.01,and values higher than-0.64 predicted NAFLD with sensitivity of 50.6% and specificity of 83.6%. Our equation was slightly superior to the Finnish NAFLD fat score(P=0.012).4.By applying this equation, the general prevalence of NAFLD in the community population was 31.0%.JConclusionThe NAFLD scoring equation established in this study was superior to Finnish NAFLD fat score. By applying this equation, the prevalence of NALFD was slightly higher than the prevalence achieved by quantitative liver ultrasound. This NAFLD scoring equation proved to be a simple, non-invasive, high-efficacy method to diagnose NAFLD. To apply it to clinical practice or community population study can help to predict the presence of NAFLD in individuals and population.PART III Association of NAFLD with Type 2 Diabetes Mellitus among Middle-aged and Elderly population of Shanghai Changfeng CommunityObjectiveNAFLD is closely associated with type 2 diabetes mellitus(T2DM). This study was designed to explore the association of liver fat content(LFC) with abnormal glucose metabolism, and NAFLD conditions in different pre-diabetes and diabetes phenotypes in the Shanghai Changfeng Community, utilizing the quantitative liver ultrasound.MethodsAfter excluding subjects with alcohol abuse history, history of viral hepatitis,other previous liver diseases, type 1 or other types of diabetes, this study recruited 3285 subjects of Changfeng community from May 2010 to May 2012. The clinical history, past history,demographic information, anthropometric parameters, routine biochemical parameters and liver ultrasound were collected in all subjects. Past diabetes history, alcohol abuse history,other previous liver diseases and history of viral hepatitis were specially surveyed. Subjects with no prior known diabetes history went through the oral glucose tolerance test, and their fasting blood glucose(FBG), 2h postprandial glucose(2hPG), fasting serum insulin(fs-insulin) were recorded. Liver fat content(LFC) over 9.15% indicated by the quantitative liver ultrasound was diagnosed as a fatty liver. Data and parameters were compared in the LFC quartiles.Results1.The general prevalence of abnormal glucose metabolism was 43.6%, in which simple impaired fasting glucose(IFG),simple impaired glucose tolerance(IGT), IFG+IGT,newly-T2 DM and prior T2 DM was 2.0%, 17.34%, 1.9%, 12.0% and 10,3% respectively.2.After adjusting for age and gender, from IFG, IGT, IFG+IGT, prior-T2 DM to newly-T2 DM, the mean liver fat content(LFC) was 6.3%, 8.5%, 8.8%, 8.5% and 10.8%respectively; the prevalence of NAFLD rose(IFG: 37.9%, IGT:42.8%, IFG+IGT: 42.9%,prior-DM: 40.2%, newly-DM 52.8%).3.The cohort was stratified according to quartiles ofLFC. After adjusted for age and gender,the average concentrations of FBG, 2hPG and fs-insulin rose across the quartiles, and increased significantly in the fourth quartile(LFC from 11.64%, P<0.001).4.Logistic regression analysis showed that 1% LFC increment independently predicted prediabetes and diabetes(OR 1.028,1.015-1.04], PO.OOl; 1.040,1.026-1.055, P<0.001,respectively) after adjustment for all potential confounders. Participants with LFC higher than 10% had higher odds ratios of impaired glucose regulation as compared with those with LFC below 10% in fully adjusted logistic models.ConclusionEvery 1% LFC increment independently predicted 2.8% and 4.0% respectively greater risk of pre-diabetes and diabetes. When LFC was over 10%, the risk of impaired glucose regulation increased significantily. |
PDF Full Text Request