| Objectives:To analyze the epidemiological characteristics of subclinical hypothyroidism(SCH)in the elderly population in iodine suitable area;to establish the specific reference range of thyroid stimulating hormone(TSH)in the elderly population in different age ranges,to clarify the cut-off point of TSH for SCH in the elderly,and to explore the correlation between TSH in different TSH ranges and dyslipidemia.Methods:To establish TSH specific reference and the cut-off point of TSH for SCH in elderly by epidemiological investigation in two areas of Jiangsu Province;to evaluate the correlation of different TSH level and dyslipidemia.Results:There were significant differences in TSH,TPOAb and TgAb levels between men and women(P<0.01).TSH level(female vs male,2.87±4.00 mIU/L vs 2.63±4.92 mIU/L),TPOAb level(female vs male,28.59±73.05 IU/mL vs 20.65±57.03 IU/mL)and TgAb levels(female vs male,72.20±286.78 IU/mL vs 59.46±300.85 IU/mL)were significantly higher than those of men.In terms of blood lipid,there were significant differences in TC level,TG level and LDL-C level between men and women,and the level of blood lipid in women was higher than that in men(P<0.01).When using the laboratory-defined TSH reference range,the prevalence of SCH in the three age groups was 7.67%,9.98%and 11.88%,respectively.When the age-specific TSH reference range was used,the prevalence of SCH in 65-69,70-79 and≥80 years was 2.95%,2.96%and 3.45%,respectively.With the age-specific TSH reference range,the overall prevalence of SCH in 2460 participants decreased from 9.07%to 3.01%.After adjusting for age,sex,height,weight,BMI and HbAlc,the correlation between TSH level and serum lipids was analyzed by multiple linear regression.It was found that TSH level was positively correlated with TG,TC and LDL-C levels(R 0.01,0.03,0.01;P<0.01),but no correlation was observed with HDL-C level(P=0.06).The prevalence of dyslipidemia in different age groups and TSH intervals was further observed and compared.Among the 65-69 years group,only the prevalence of hypercholesteremia was statistically different at different TSH levels(P=0.04),and no statistical difference was observed in the prevalence of hypertriglyceridemia,hyperLDL-C at different TSH levels(P>0.05).However,in the 70-79 yearsgroup and≥80 years group,there were significant statistical differences in the prevalence of hypercholesteremia hypertriglyceridemia,and hyperLDL-C under different TSH levels(P<0.05);no significant difference was observed in the prevalence of hypoHDL-C at different TSH levels in three different age groups(P>0.05).Conclusions:The establishment of TSH reference range in different age ranges in elderly and the identification of the cut-off point for SCH in the elderly significantly reduce the prevalence of SCH in the elderly,thus reducing unnecessary diagnosis and treatment of SCH.The new TSH cut-off point could not reflect the effect of SCH on dyslipidemia in all ages,and more reliable biomarkers are needed to reflect the effect of SCH on dyslipidemia and CVD.Objectives:To identify the serum metabolomics information of SCH patients with CHD and without CHD in the elderly,analyze the differences of serum metabolomics information in different groups,and establish the clinical prediction model of CHD in elderly SCH patients based on metabolomics.Methods:A total of 32 patients with SCH in the elderly diagnosed as CHD were enrolled in the affiliated Suqian hospital of Xuzhou medical university,30 SCH patients without CHD in the elderly were enrolled as control group.Differential metabolites and metabolic pathways were identified by LC-MS.The risk prediction score of CHD patients with SCH in the elderly was established by multiple logistic regression model.Results:More than 30,000 metabolites were found in all serum samples.The serum metabolites of SCH patients with CHD were significantly changed.Differential metabolite peaks under electrospray ionization(ESI)were differentiated according to FC>1.2 and P<0.05.VIP values greater than 1.0 calculated by orthogonal partial least squares discriminant analysis(OPLS-DA)score were selected as the metabolic characteristics of significant changes,and 21 metabolites were identified and confirmed.Through dimensionality reduction analysis,three substances with the largest differential metabolites were selected as hypoxanthine,HIPPURATE and bilirubin.ROC evaluation showed that the AUC of hypoxanthine was 0.883(95%CI 0.796-0.971),the AUC of HIPPURATE was 0.776(95%CI 0.655-0.897),and the AUC of bilirubin was 0.634(95%CI 0.489-0.780).Multivariate logistic regression model was used to establish a clinical riks prediction score for CHD in elderly SCH patients based on hypoxanthine,HIPPURATE and bilirubin.The sensitivity and specificity were 75.0%and 96.7%,respectively,with an AUC of 0.897(95%CI 0.793-0.960).and a threshold of 0.7167.Correlation analysis between risk score and baseline lipid showed that risk score was positively correlated with LDL-C(R=0.435,P<0.001)and negatively correlated with HDL-C(R=-0.357,P<0.01).Conclusions:Serum metabolomics information of CHD patients with SCH in the elderly were identified by LC-MS of non-targeted metabonomics.On the basis of hypoxanthine,HIPPURATE and bilirubin,a clinical prediction model was established by multivariate logistic regression,and obtained the risk score,which significantly distinguished SCH patients with CHD and without CHD in the elderly.Risk prediction scores based on selected serum metabolite biomarkers are helpful for early prediction of CHD in elderly SCH patients. |
PDF Full Text Request