Associations between thyrotropin level and glycated hemoglobin A1c in non-diabetic population

刘人心,张巧,彭年春,徐淑静,张淼,王睿,胡颖,时立新
DOI: https://doi.org/10.3760/cma.j.cn115791-20200312-00143
2020-01-01
Abstract:Objective:To investigate the relationship between serum thyrotropin (TSH) and glycated hemoglobin A 1c (HbA 1c) in non-diabetic patients. Methods:From May to October 2011, 10 140 permanent residents aged over 40 years old were sampled by cluster sampling method in yunyan district of guiyang city, and 5 819 residents were eventually included. All the subjects were investigated by detailed epidemiological questionnaire, and the basic data were collected to measure body mass index (BMI), waist circumference, blood pressure, blood glucose, blood lipid, insulin, glycosylated hemoglobin and thyrotropin. Non parameter test t test, and logistic regression analysis were used. Results:According to TSH level, the population was divided into the normal TSH group and the elevated TSH group. Compared with the normal TSH group, the HbA 1c value of the elevated TSH group was lower ( P<0.05),while the BMI, fasting insulin (FINS), homesostasis model assessment of insulin resistance (HOMA-IR), triglyceride (TG), low-density lipoprotein-cholesterol (LDL-C) and total cholesterol were higher , with statistically significant differences ( P<0.05).After adjusting for age, systolic pressure, diastolic pressure, BMI, waist circumference, FPG, OGTT 2 h blood glucose, FINS, HOMA-IR, TG, HbA 1c,LDL-C , logistic regression analysis results showed that compared with the normal TSH group, the elevated TSH group was the factor affecting HbA 1c. TSH was negatively correlated with HbA 1c ( OR=0.845, P=0.024, 95% CI=0.730 to 0.978).After further gender stratification, the negative correlation between TSH and HbA 1c was found only in males ( OR=0.672, P=0.034, 95% CI=0.466 to 0.970), but not in females. Conclusion:In non-diabetic population, increased TSH may reduce the level of HbA 1c. Serum TSH in male non-diabetic population is negatively correlated with HbA 1c, while serum TSH in female non-diabetic population is not significantly correlated with HbA 1c.
What problem does this paper attempt to address?