Relationship Between Androgen Level And Body Fat Distribution In Postmenopause Patients With Type 2 Diabetes

Objective: Aim to investigate women with type 2 diabetes and non-diabetic postmenopausal women,to explore the relationship between androgen level and body fat distribution,and the influence degree of estrogen and androgen on visceral fat area.What's more,to explore a more convenient,quick and easy method for measuring visceral fat content.Methods: Total 88 postmenopausal women subjects with type 2diabetes(Case group)were recruited in the present study who were hospitalized in Qinhuangdao First Hospital during 2015.10-2016.10.And 88 healthy postmenopausal women(control group)were recruited at the same time.All of these subjects volunteered to participate in this study and signed informed consent.The name,age,menstrual history,history of type 2 diabetes and other basic information of all subjects were recorded.Some human parameters were measured,such as height,weight,waist circumference,hip circumfe-rence.All subjects with empty stomach were extracted the elbow venous blood5 ml in the morning,measured fasting blood glucose(FBG),testosterone(TT),estradiol(E2),sex hormone binding globulin(SHBG),follicle-stimulating hormone(FSH),luteinizing hormone(LH),triglyceride(TG),total cholesterol(TC),low density lipoprotein-cholesterol(LDL-c),high density lipoproteincholesterol(HDL-c).Calculated free androgen index(FAI=TT×100/SHBG).All of subjects(fasting)were measured liver fat content by the same diplomate of color doppler ultrasound.Measured the visceral fat area(VFA),body fat percentage(BFP)and total body fat(TBF)by bioelectrical impedance analysis.All analyses were performed using the SPSS 13.0 statistical software.Test the normality and homogeneity of variance before analysis the data.Values were expressed as mean with standard deviation.When not normallydistributed,the data were expressed as binary with interquartile ranges.A simple scatter plot and Pearson correlation coefficient were used to measure the strength of the association between two variables.Multivariate linear regression analysis was used in multivariate analysis,P<0.05 difference was statistically significant.Results:1 Case group:The history of diabetes(10.0±7.95),glycosylated hemoglobin(9.03±2.07)%,both of them were not statistically significant with LFC,BFP,TBF and VFA,P>0.05.2 General comparison of the two groups:The waist circumference(92.31±7.72)cm,hip circumference(87.72±3.91)cm,waist to hip ratio(1.05±0.06),waist height ratio(0.59±0.05),BMI(26.52±3.12)kg/m2 of the case group were higher than those of the control group.Control group waist circumference(89.49± 9.62)cm,hip circumference(85.13±4.13)cm,waist to hip ratio(1.03±0.09),waist height ratio(0.56±0.06),BMI(24.78±4.25)kg/m2.The difference between the two groups was statistically significant(P<0.05).3 In case group,serum SHBG(1.06±0.22)ng/mL,TT(0.71±0.28)nmol/L,Free androgen index(FAI),(75.32±48.18),FSH(43.47±17.01)mIU/mL,LH(19.99± 8.75)mIU/mL,TG(1.97±1.27)mmol/L,TC(5.77±1.38)mmol/L,LDL(3.50± 1.15)mmol/L,E2(less than 10pg/mL were 45 cases,more than10pg/mL were 43 cases),Liver fat content(LFC)and VFA were(44.35±18.88)% and(64.79±21.10)cm2.In control group,serum SHBG and TT were(0.88± 0.08)ng/mL and(0.58±0.36)nmol/L by respectively.FAI was(61.11±32.27).E2(less than 10pg/mL and more than 10pg/mL were half and half).LFC and VFA were(36.38±14.88)%,(58.69±16.38)cm2 by respectively.The levels of TT,FAI,LFC,TC,LDL and visceral fat in case group were higher than control group,and the FSH,LH,and TG level were lower than those of control group,the difference was statistically significant,P<0.05.4 Comparison of each index in group:(1)It was found that TT,FAI,TC,TG and LH level were positivecorrelation with LFC by single factor correlation analysis,P<0.05;E2 level was negative correlation with LFC;Both of the groups were mainly increased with visceral fat content compared with total body fat,and the whole fat were central distribution.(2)The case group,taking LFC as the dependent variable(Y),and taking TT(X1),SHBG(X2),FAI(X3),E2(X4),WHtR(X5),TG(X6),TC(X7),LH(X8)as independent variables for multiple linear regression analysis:regression equation:^Y=120.824-67.879X2+0.169X3-8.117X4+3.458X7+0.664X8(P<0.05).Complex correlation coefficient R=0.720,it explained that the correlation between the actual LFC value and the estimated LFC value of the regression equation was 0.720,Determination coefficient R2=0.518,The results showed that the LFC of the multiple linear regression equation could be explained by SHBG,FAI,E2,TC,LH,accounting for 57.4%.The standardized regression coefficient of the independent variable(the larger the value,the greater the effect of the independent variable on Y)is 12.491(X2),0.058(X3),2.978(X4),1.195(X7),0.174(X8).Taking BFP as the dependent variable(Y),and taking TT(X1),SHBG(X2),FAI(X3),E2(X4),WHtR(X5),TC(X6)as independent variables for multiple linear regression analysis:regression equation:^Y =25.928+29.49X7-2.685X4-6.444X2(P<0.05).Complex correlation coefficient R=0.522.Determination coefficient R2=0.272.The results showed that the BFP of the multiple linear regression equation can be explained by SHBG,E2 and WHt R only accounting for 27.2%.(3)For control group,taking LFC as the dependent variable(Y),and takingFAI(X1),SHBG(X2),E2(X3),TG(X4),TC(X5),WHtR(X6),FSH(X7),LDL(X8),TT(X9),LH(X10)as independent variables for multiple linear regression analysis: regression equation:^Y=-3.335-38.664X2+ 0.625X3+ 2.276X5+114.421X6+0.301X10(P<0.05).Complex correlation coefficient R=0.663,it explain-ed that the correlation between the actual LFC value and the estimated LFC value of the regression equation was 0.684,Determination coefficient R2=0.468,The results showed that the LFC of the multiple linear regression equation could be explained by SHBG,E2,TC,LH,WHt R,accou-nting for 46.8%.The standardized regression coefficient of the independent variable is 38.664(X2),0.206(X3),0.994(X5),22.454(X6),0.075(X10).Conclusions:1 Postmenopausal women fat tends to central distribution.The distribution of type 2 diabetes are more obvious than that of non-diabetic women,and liver fat content and visceral fat area increase respectively by 22% and 10%compared with non-diabetic women`s.2 The androgen level is positively correlated with central fat distribution of postmenopausal women.Estrogen level is negatively correlated with visceral fat content.The correlation between the sex hormone binding globulin and the central distribution of fat in postmenopausal women is stronger than that of estrogen,and the correlation between free androgen index and liver fat content is lower than that of estrogen.3 In subjects with type 2 diabetes,testosterone,sex hormone binding globulin,free androgen index and liver fat content and visceral fat area of the correlation was higher than non-diabetic group.4 The visceral fat content could be reflected effectively by liver fat content and visceral fat area,Liver fat content showed good correlation of visceral fat area,but the detection method of visceral fat area(bioelectrical impedance analysis)was more convenient,fast and good compliance.