| Part I Relationship between liver fat content and serum osteocalcin levels in middle-aged and elderly population:The Shanghai Changfeng studyObjective:The aim of this study was to explore the relationship between non-alcoholic fatty liver disease (NAFLD) and serum osteocalcin levels in middle-aged and elderly Chinese population. Furthermore, we assessed the independent risk factors for NAFLD and evaluated the contribution osteocalcin made to liver fat content in each gender. Methods:This was a cross-sectional community-based study.4576 subjects at least 45 years old were enrolled from Shanghai Changfeng study since May 2010 to June 2012. A standard interview, anthropometrics and laboratory parameters were performed for each participant. Liver fat content were assessed by the standardized ultrasound quantitative method. Body mass indexes, liver fat content, HOMA-1R, HOMA-beta were calculated.1268 subjects with incomplete data, failed to finish the ultrasound liver examination, drinking history, positive for hepatitis B surface antigen and/or hepatitis C antibody, abnormal liver function (ALT and/or AST> 1.5 times normal) or renal dysfunction (eGFR<60 mL/min per 1.73 m2) were excluded. Finally,3308 participants (male 1146, female 2162) were taken into analyses. Correlation analyses between NAFLD and osteocalcin were performed and a total of 2970 subjects (male 999, female 1971) without diabetes history were selected for subgroup analyses,Results:1.3308 participants (1146 males and 2162 females) were included in the analyses. The median of age, serum osteocalcin levels and liver fat content were 60.4(54.6-68.7) years,20.1(15.8-25.1)ng/ml and 5.4(2.5-11.5)%, respectively. Men had significantly higher age, weight, BMI, waist circumference, hip circumference, waist-hip ratio, systolic blood pressure, diastolic blood pressure, FBG, PBG, ALT, GGT, UA, BUN and Cr levels (P<0.05), and lower levels of FINS, HOMA-IR, HOMA-β,TC, LDL-c, HDL-c, OCN, LFC levels than women (P<0.05).2. Subjects were divided into two groups according to NAFLD diagnosed by liver fat content above 9.15%. NAFLD patients had significantly higher levels of weight, BMI, waist circumference, hip circumference, waist-hip ratio, systolic blood pressure, diastolic blood pressure, ALT, AST, GGT, UA, FBG, PBG, FINS, HOMA-IR, HOMA-β, TC and TG(P<0.001), and lower levels of age and serum osteocalcin levels (P<0.01) compared with those without NAFLD. No statistically significant differences in LDL-c levels were found between the two groups (P> 0.05).3. Spearman partial correlation analysis revealed that, after adjustment of gender, age and BMI, liver fat content was positively associated with body weight, waist circumference, waist-hip ratio, DBP, ALT, AST, GGT, UA, FBG, PBG, FINS, HOMA-IR, HOMA-β, TC and TG (P<0.05), and negative correlated with HDL-c, Cr and OCN (P<0.05); no association was found between liver fat content and LDL-c.4. Multivariate logistic regression analysis showed that for women, the independent risk factors for NAFLD include weight, UA, eGFR, TG and FINS (P <0.05), while osteocalcin (P=-0.017, OR (95% CI) turned to be a weak protective factors. For men, the risk factors included body weight, TG and waist circumference(P<0.05), while age was a protective factor against NAFLD(P <0.05). But osteocalcin showed no independent association with NAFLD in men. In addition, independent association between NAFLD and osteocalcin in non-diabetes subgroup were in consistent with these results above.5. Multiple linear regression analysis indicated for women, weight, TG, FINS, UA, eGFR, FBG independently and positively related to LFC (P<0.01), whereas HDL-c had a negative independent association with LFC; for men, TG, UA, and waist circumference were independent risk factors for LFC (P<0.05). osteocalcin levels were not an independent factors for LFC in both gender. In addition, no independent correlation was found between OCN and LFC in non-diabetes subgroup.6. Non-diabetes subgroup were stratified to five groups by the cut-off points of liver fat content 5%,10%,15%,20%. Weight, BMI, waist circumference, hip circumference, FINS, HOMA-IR, HOMA-p, TC, TG, LDL-c and ALT increased significantly from the second layer (LFC=5.1-10%) (P<0.05), while waist-hip ratio, blood pressure, blood glucose, AST, GGT and UA level significantly increased since the third layer (LFC=10.1-15.0%) (P<0.05). HDL-c levels decreased significantly from the second layer (LFC= 5.1-10%), and declined in line with the increments of liver fat content (P<0.05). Serum osteocalcin levels significantly decreased when the LFC exceeded 10% compared with those of the first two layers (P<0.05), and the significant differences still existed after adjusting for age (P<0.05).7. In non-diabetes subgroup, multiple linear regression analysis for osteocalcin indicated for women, age, SBP and HDL-c had independent positive correlation with osteocalcin (P<0.05), while LFC (P=-0.058, P=0.020), BMI, PBG and HOMA-P had independent negative correlation (P<0.05); for men, the osteocalcin levels were independently and negatively associated with age and PBG (P<0.05).Conclusions:In the middle-aged and elderly Chinese population from Shanghai Changfeng study, subjects with NAFLD had lower serum osteocalcin levels than those without. Serum osteocalcin levels were an independent protective factor for NAFLD in women rather than men, but it was not independent associated with liver fat content in both gender.PART II Assocciation between nonalcoholic fatty liver disease and elevated liver enzyme levels in middle-aged and elderly population:The Shanghai Changfeng studyObjective:The purpose of this study was to explore the relationships between the elevation of liver enzyme levels and the presence of NAFLD in middle-aged and elderly population of Shanghai Changfeng study, and further determine the cut-off value of liver enzyme used for NAFLD diagnosis. Methods;A total of 4576 community residents at least 45 years old were recruited from May 2010 to June 2012. All of them were from Shanghai Changfeng community. A standard interview, anthropometrics, laboratory parameters were performed for each participant. Liver fat content were assessed by the standardized hepatic ultrasonographic examination.1268 subjects with incomplete data, failed to finish the ultrasound liver examination, drinking history, positive for hepatitis B surface antigen and/or hepatitis C antibody, abnormal liver function (ALT and/or AST> 1.5 times normal) or renal dysfunction (eGFR<60 mL/min per 1.73 m2) were excluded. Finally,3308 participants(male 1146, female 2162) were taken into analyses. Fatty liver was diagnosed according to the liver fat content by ultrasound method exceeds 9.15%. The receiver operating characteristic (ROC) curve for NAFLD diagnosis were applied to find the optimum cut-off value of liver enzymes.Results:1.3308 participants (2162 females and 1146 males) were taken into analyses. The median of age, BMI and liver fat content were 61.0 (55-69.6) years,23.9 (21.9-26.2)kg/m2 and 5.5 (2.5-11.7)%, respectively. In men, the median levels of ALT,AST and GGT were 17(13-24)U/L,20(17-24)U/L and 25 (19-36)U/L respectively. While in women, the median levels of ALT, AST and GGT were 15 (12-21)U/L,20 (17-23)U/L and 20 (16-29)U/L, respectively.2. Spearman partial correlation analysis showed that ALT, AST and GGT were significantly associated with obesity related indexes, blood pressure, blood glucose, insulin resistance and lipid profiles(P<0.05).3. Cut-off value of ALT、AST、GGT and ALT/AST reveal by receiver operating characteristic curve for NAFLD diagnosis were 17.5U/L、23.5U/L、 28.5U/L and 0.90 in males and 16.5U/L、21.5U/L、8.5U/L and 0.85 in females.4. Logistic regression analyses for NAFLD in relation to the ALT and ALT/AST quintiles indicated that, the ORs for NAFLD increased progressively with increasing liver enzyme levels(P<0.001). After adjustment for age, BMI, waist, systolic blood pressure, fasting blood glucose, fasting insulin, triglycerides, and HDL-c, the ORs for the presence of NAFLD significantly increased in the third quintile of ALT in both gender(P<0.05), and in the third quintile of ALT/AST in men while fourth in women(P<0.05).Conclusions:In the middle-aged and elderly Chinese population, liver enzyme levels closely associated with various metabolic indexes. Slightly elevation of serum liver enzyme levels within the normal range revealed an increase risk for NAFLD. As a result, we suggested the optimum cut-off value of ALT for NAFLD diagnosis should be 17.5U/L for men and 16.5U/L for women, or of ALT/AST ratio 0.90 for men and 0.85 for women.PART Ⅲ Associations between non-alcoholic fatty liver disease and sexual hormones in middle-aged and elderly population:The Shanghai Changfeng studyObjective:The purpose of our study was to investigate the relationship between non-alcoholic fatty liver disease (NAFLD) and sexual hormones including estradiol and testosterone.Methods:During May 2010 to June 2012, a total of 4576 participants above 45 years old were enrolled from Shanghai Changfeng community. The history information, demographic information, past history, anthropometric parameters, biochemical parameters and liver ultrasound were collected for all participants. Body mass indexes, liver fat content, HOMA-IR, HOMA-beta were calculated.1268 subjects with incomplete data, failed to finish the ultrasound liver examination, drinking history, positive for hepatitis B surface antigen and/or hepatitis C antibody, abnormal liver function (ALT and/or AST> 1.5 times normal) or renal dysfunction (eGFR<60 mL/min per 1.73 m2) were excluded. Finally,3308 participants (male 1146, female 2162) were taken into analyses. Of the 3308 subjects, data from 3 were not available for estradiol (E2), and E2 levels of 119 cases (males 23, females 1096 including 1037 were postmenopausal) were below the detection limit (< 18.35 pmol/L),and the remaining 2186 participants (male 1121,female 1065) had detailed serum estradiol levels data. In addition, data from 4 were not available for testosterone (T), and T levels of 132 subjects were below the detection limit (<0.087 nmol/L), while 24 were above the detection limit (>57.05 nmol/L), as a result, the remaining 3148 participants (male 1125, female 2023) had detailed serum testosterone level data. The present study evaluated the associations between sexual hormones and NAFLD and LFC in each gender.Results:1.249 women were premenopausal with the median of age 50.8(49.3-53)yrs, BMI 23.2(21.5-25.5)kg/m2, and LFC 5.5 (2.7-11)%; 199 of them had E2 levels above 18.35pmol/L with the median of E2 262.7(77.7-543.0) pmol/L; 242 of them had T levels within the test range with the median of T 0.64(0.45-0.91) nmol/L.1891 women were postmenopausal with the median of age 61.6(56.1-70.0)yrs, BMI 23.8(21.7-26.1)kg/m2, and LFC 5.8(2.7-12.1)%; 854 of them had E2 levels above 18.35pmol/L with the median of E2 35.8(26.1-35.8) pmol/L; 1761 of them had T levels within the test range with the median of T 0.58(0.37-0.85) nmol/L. Premenopausal women had significantly lower levels of age, waist circumference, waist-hip ratio, SBP, FBG, PBG, FINS, HOMA-IR, TC, TG, LDL-c, liver enzymes, BUN, Cr and UA (P<0.05) compared with postmenopausal women.2. Postmenopausal women were divided into five layers:1037 postmenopausal women had an E2 levels... |
PDF Full Text Request