Metabolic syndrome is a medical disorder characterized by obesity, hyperglycemia, dyslipidemia, and hypertension. Thyroid hormone has been shown to affect many metabolic processes. This study was undertaken to explore the relationship between serum thyrotropin and components of metabolic syndrome in Chinese adolescents. Waist circumference (76.4 ± 10.7 versus 70.0 ± 10.6 cm,
The prevalence of metabolic syndrome and obesity has continued to increase worldwide in recent years. According to the guidelines of the US National Cholesterol Education Program (NCEP), the age-standardized prevalence of metabolic syndrome is 9.8% in men and 17.8% in women in China [
Thyroid hormones have been shown to affect many metabolic processes, and overt hypothyroidism is clearly related to body weight and greater adiposity [
A total of 936 students (aged 11–16 years; 46.6% female) from junior and senior high schools in Liaoyang, a medium-sized city in northeastern China, were initially recruited for the study after informed consent was obtained from their parents or legal guardians. The study and research protocols were approved by the Medical Ethics Committee of Shengjing Hospital of China Medical University.
Participants with a history of thyroid disease under thyroxine or antithyroid drugs treatment, those taking medications affecting thyroid function, those with overt hyperthyroidism or overt hypothyroidism, and those with a history of diabetes or who were taking hypoglycemic agents were excluded from the study. After 17 participants were excluded, a total of 919 participants with complete data were enrolled in the study.
A questionnaire was given to each participant to obtain general information as well as information on history of thyroid disease, diabetes, hypertension, and dyslipidemia with the associated treatments. Anthropometric data were collected by qualified physicians. Height and weight were measured to calculate body mass index (BMI): BMI = weight (kg)/height² (m²). With the subjects standing with their feet 25–30 cm apart, waist circumference was measured by placing the measuring tape horizontally around the patient’s abdomen from the horizontal waistline of the two middle points of the anterosuperior iliac spine through the inferior margin of 12th ribs with the tape held tightly to the skin, but without pressing. The measurement precision was ±0.1 cm. The thyroid gland was examined by qualified clinical endocrinologists according to the standard procedure combined with inspection and palpation. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured twice after a 10-minute rest with the patient in the sitting position using a mercury-gravity sphygmomanometer. There was a 2-minute interval between the two measurements, and the mean value of the two measurements was used. Blood samples were obtained after an overnight fast lasting at least 10 hours; a portion of the samples were sent to the laboratory in the Liaoyang Diabetes Hospital within 30 minutes of collection for measurement of fasting plasma glucose (FPG), total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), high density lipoprotein cholesterol (HDL-C), and triglycerides (TG). The serum samples for fasting insulin (FINS), TSH, and thyroid peroxidase antibodies (TPOAb) were stored at −80°C until analysis. The Homoeostasis Model of Insulin Resistance (HOMA-IR) score was calculated using the following formula: HOMA-IR = (FPG [mmol/L] × insulin [
Serum TSH and TPOAb were tested using an electrochemiluminescence immunoassay (Roche Cobas 6000, Roche Ltd, Switzerland). FPG was measured using the glucose oxidase method. TC, LDL-C, HDL-C, and TG were measured by routine enzymatic methods (Olympus 400, Olympus Optical Company, Tokyo, Japan). Insulin was determined by radioimmunoassay (China Institute of Atomic Energy, Beijing, China).
Subclinical thyroid diseases: the TSH reference range (1.01–5.73 mIU/L) was set by the National Academy of Clinical Biochemistry (NACB) guidelines [
Data processing and statistical analysis were performed using SPSS 17.0 software. Data are expressed as mean ± standard deviation (SD) or median (interquartile range). Statistical comparisons were performed using independent-samples
Waist circumference (76.4 ± 10.7 versus 70.0 ± 10.6 cm,
Comparison of metabolic syndrome and its components in groups with different serum TSH levels.
TSH (mIU/L) |
| |||
---|---|---|---|---|
<1.01 ( |
1.01–5.73 ( |
>5.73 ( | ||
Age (yr) | 14 ± 2 | 14 ± 1 | 14 ± 2 | 0.689 |
Waist circumference (cm) | 71.3 ± 10.9 | 70.0 ± 10.6 | 76.4 ± 10.7* | 0.021 |
BMI (kg/m2) | 21.68 ± 4.58 | 21.51 ± 4.16 | 23.90 ± 4.20* | 0.040 |
FPG (mmol/L) | 4.70 (4.50–5.10) | 4.80 (4.40–5.00) | 4.60 (4.40–5.10) | 0.803 |
TC (mmol/L) | 4.19 ± 0.85 | 4.49 ± 0.80 | 4.62 ± 0.90 | 0.160 |
HDL-C (mmol/L) | 1.01 ± 0.25 | 1.06 ± 0.26 | 1.04 ± 0.21 | 0.666 |
LDL-C (mmol/L) | 3.30 ± 0.13 | 3.37 ± 0.17 | 3.40 ± 0.27 | 0.136 |
TG (mmol/L) | 0.63 (0.52–1.16)* | 0.95 (0.68–1.30) | 0.82 (0.66–1.37) | 0.006 |
SBP (mmHg) | 117 ± 14 | 114 ± 14 | 113 ± 12 | 0.285 |
DBP (mmHg) | 69 ± 12 | 70 ± 11 | 69 ± 11 | 0.508 |
FINS ( |
19.00 (15.20–24.50) | 18.00 (13.00–24.00) | 21.00 (18.75–26.50) | 0.803 |
HOMA-IR | 3.88 (3.09–5.47) | 3.76 (2.72–5.21) | 4.20 (3.87–5.48) | 0.939 |
Data are presented as mean ± SD or median (interquartile range); compared with euthyroid group,
Comparison of the prevalence of metabolic syndrome and its components in groups with different serum TSH levels.
TSH (mIU/L) |
| |||
---|---|---|---|---|
<1.01 ( |
1.01–5.73 ( |
>5.73 ( | ||
Overweight/obesity | ||||
Prevalence (%) | 30.4 | 22.0 | 50.0 | 0.002 |
OR | 1.470 | Ref. | 3.444* | |
Hyperglycemia | ||||
Prevalence (%) | 4.2 | 6.4 | 10.3 | 0.621 |
OR | 0.615 | Ref. | 2.636 | |
Hypertension | ||||
Prevalence (%) | 16.7 | 17.6 | 10.7 | 0.633 |
OR | 0.853 | Ref. | 0.362 | |
Low HDL-C | ||||
Prevalence (%) | 54.2 | 46.3 | 53.6 | 0.574 |
OR | 1.346 | Ref. | 1.052 | |
High TG | ||||
Prevalence (%) | 8.3 | 9.8 | 13.8 | 0.754 |
OR | 0.831 | Ref. | 1.246 | |
Metabolic syndrome | ||||
Prevalence (%) | 8.7 | 7.7 | 14.3 | 0.449 |
OR | 1.117 | Ref. | 1.128 |
Ref. is control group; compared with euthyroid group,
Based on the IDF criteria for diagnosis, the prevalence of metabolic syndrome in adolescents was 8.0% (72.2% were male). After adjusting for age and gender, the TSH level in the metabolic syndrome group was significantly higher than that in nonmetabolic syndrome group (2.65 (2.28–3.80) versus 2.53 (1.92–3.45) mIU/L,
Comparison of serum TSH levels in different components of metabolic syndrome between the case and control groups.
Overweight/obesity | Hyperglycemia | Hypertension | Low HDL-C | High TG | Metabolic syndrome | |
---|---|---|---|---|---|---|
TSH (mIU/L) | ||||||
Case group | 2.57 (2.01–3.70) | 2.62 (2.01–3.45) | 2.61 (1.97–3.55) | 2.58 (2.00–3.60) | 2.76 (2.13–3.77) | 2.65 (2.28–3.80) |
Control group | 2.55 (1.91–3.41) | 2.55 (1.92–3.46) | 2.50 (1.92–3.45) | 2.50 (1.85–3.34) | 2.50 (1.90–3.42) | 2.53 (1.92–3.45) |
|
||||||
|
0.026* | 0.481 | 0.314 | 0.151 | 0.037* | 0.032* |
Data are presented as median (interquartile range);
The levels of TG in the high-normal TSH group were significantly higher than those in the low-normal TSH group (1.02 (0.71–1.42) versus 0.88 (0.63–1.25) mmol/L,
Comparison of the components of metabolic syndrome in groups with normal TSH levels.
TSH (mIU/L) |
| |||
---|---|---|---|---|
Low-normal TSH group |
Mid-range TSH group |
High-normal TSH group | ||
Age (yr) | 14 ± 1 | 14 ± 1 | 14 ± 1 | 0.338 |
Waist circumference (cm) | 69.3 ± 10.4 | 70.5 ± 10.4 | 70.3 ± 10.9 | 0.287 |
BMI (kg/m2) | 21.25 ± 3.88 | 21.60 ± 4.02 | 21.67 ± 4.55 | 0.381 |
FPG (mmol/L) | 4.70 (4.40–5.01) | 4.80 (4.40–5.00) | 4.80 (4.40–5.00) | 0.849 |
TC (mmol/L) | 4.41 ± 0.73 | 4.50 ± 0.82 | 4.55 ± 0.83 | 0.177 |
HDL-C (mmol/L) | 1.08 ± 0.27 | 1.06 ± 0.26 | 1.05 ± 0.26 | 0.377 |
LDL-C (mmol/L) | 3.35 ± 0.15 | 3.37 ± 0.17 | 3.38 ± 0.19 | 0.075 |
TG (mmol/L) | 0.88 (0.63–1.25) | 0.95 (0.68–1.27) | 1.02 (0.71–1.42)* | 0.022 |
SBP (mmHg) | 114 ± 13 | 113 ± 15 | 116 ± 14 | 0.051 |
DBP (mmHg) | 70 ± 12 | 69 ± 12 | 71 ± 10 | 0.079 |
FINS ( |
18.00 (13.43–24.00) | 18.00 (13.00–25.00) | 18.00 (13.00–26.00) | 0.849 |
HOMA-IR | 3.75 (2.70–4.92) | 3.76 (2.73–5.22) | 3.77 (2.69–5.43) | 0.986 |
Data are presented as mean ± SD or median (interquartile range); compared with low-normal TSH group,
Comparison of the prevalence of metabolic syndrome and its components in groups with normal TSH levels.
TSH (mIU/L) |
| |||
---|---|---|---|---|
1.01–2.14 ( |
2.15–3.07 ( |
3.08–5.72 ( | ||
Overweight/obesity | ||||
Prevalence (%) | 19.4 | 24.6 | 21.9 | 0.315 |
OR | 0.752 | Ref. | 0.827 | |
Hyperglycemia | ||||
Prevalence (%) | 5.9 | 6.6 | 6.6 | 0.923 |
OR | 0.988 | Ref. | 0.990 | |
Hypertension | ||||
Prevalence (%) | 15.7 | 17.4 | 19.8 | 0.439 |
OR | 0.931 | Ref. | 1.166 | |
Low HDL-C | ||||
Prevalence (%) | 40.8 | 48.9 | 49.3 | 0.070 |
OR | 0.681* | Ref. | 1.020 | |
High TG | ||||
Prevalence (%) | 7.3 | 10.5 | 11.7 | 0.179 |
OR | 0.728 | Ref. | 1.004 | |
MS | ||||
Prevalence (%) | 4.9 | 9.2 | 9.0 | 0.099 |
OR | 0.618 | Ref. | 0.722 |
Ref. is control group; compared with mid-range TSH group,
The relationship between TSH and the components of metabolic syndrome by multiple linear regression model.
Model |
|
| |
---|---|---|---|
Waist circumference | |||
T | 1 | 1.512 | 0.019* |
N | 1 | 1.206 | 0.152 |
BMI | |||
T | 1 | 0.562 | 0.024* |
N | 1 | 0.383 | 0.240 |
TC | |||
T | 2 | 0.160 | 0.003* |
N | 2 | 0.173 | 0.013* |
LDL-C | |||
T | 2 | 0.032 | 0.008* |
N | 2 | 0.031 | 0.043* |
HDL-C | |||
T | 2 | −0.002 | 0.923 |
N | 2 | −0.020 | 0.379 |
TG | |||
T | 2 | 0.095 | 0.001* |
N | 2 | 0.132 | 0.021* |
FPG | |||
T | 2 | 0.002 | 0.766 |
N | 2 | 0.005 | 0.613 |
SBP | |||
T | 2 | −0.139 | 0.875 |
N | 2 | 1.216 | 0.296 |
DBP | |||
T | 2 | 0.675 | 0.377 |
N | 2 | 1.692 | 0.093 |
FINS | |||
T | 2 | −0.002 | 0.766 |
N | 2 | −0.005 | 0.613 |
HOMA-IR | |||
T | 3 | 0.010 | 0.745 |
N | 3 | 0.017 | 0.679 |
Our study showed that among 919 participants, 866 (94.2%) had normal thyroid function, 29 (3.2%) had subclinical hypothyroidism, and 24 (2.6%) had subclinical hyperthyroidism. We found that the TSH level in the metabolic syndrome group was significantly higher than that in the non-metabolic syndrome group (Table
Among the 919 adolescents, those with a waist circumference at the 90th percentile or above were regarded as obese; the total percentage meeting this criterion was 23.0%, which further confirmed the high prevalence of obesity in adolescents. We speculate that leptin may play an important role in the development of physiological insulin resistance, which is a major component of metabolic syndrome. Soliman et al. [
In our current study, a correlation also was found between elevated TSH and dyslipidemia. With every 1 mIU/L increase in TSH, TG was increased by 0.095 mmol/L; this positive linear correlation also existed between normal range TSH and TG. Among patients with hypertriglyceridemia, TSH levels were higher than in participants with normal TG levels. The above findings illustrate the positive relationship between TSH and TG. Regardless of whether TSH was within the normal range, there were positive linear correlations between TSH and TC as well as TSH and LDL-C, which indicated that both TC and LDL-C could be increased with an increase in TSH. For HDL-C, the risk of low HDL-C in the low-normal TSH group was less than that in the mid-range TSH group (OR = 0.681, 95% CI: 0.480–0.967), which indicated that a decrease in TSH within the normal range could reduce the risk of low levels of HDL-C.
Thyroid hormones influence various metabolic pathways, and both the composition and transport of lipoproteins are impaired in thyroid diseases. Hypothyroidism is a major cause of secondary dyslipidemia, which is characterized by hypercholesterolemia and a marked increase in LDL-C, which results from decreased fractional clearance by the reduced number of LDL receptors in the liver. Althaus et al. [
We also found the TSH level in the metabolic syndrome group was significantly higher than that in non-metabolic syndrome group, in agreement with the results of our previous study, which showed that serum TSH is higher among adult individuals with metabolic syndrome than among those without metabolic syndrome [
We also found a significant positive correlation between normal range TSH and DBP, with an increase in DBP of 1.991 mmHg for every 1 mIU/L increase in TSH. However, after adjusting for age, gender, HOMA-IR scores, and BMI, the correlation disappeared. Moreover, this relationship was not seen for SBP. Meanwhile, no difference was found for SBP, DBP, or the risk of hypertension among groups with different levels of TSH. The relationship between TSH and blood pressure is not clear yet. Åsvold et al. [
Our study did not show a correlation between TSH and FPG, possibly because adolescents with overt hyperthyroidism or hypothyroidism were excluded. All the participants had normal FT3 and FT4, and the effects of thyroxine were not strong enough to influence FPG. Additionally, we used a cross-sectional design in our study, which did not address the question of whether adolescents with subclinical hypothyroidism should receive thyroid hormone replacement therapy. In summary, we found that an increase in serum TSH was positively correlated with components of metabolic syndrome and might be a risk factor for metabolic syndrome in adolescents. Further investigations are essential to further confirm the relationship between TSH and components of metabolic syndrome in adolescents as well as the underlying mechanism(s).
The authors declare no conflict of interests.
The authors are indebted to the students from junior and senior high schools in Liaoyang City who participated in this study. The authors would like to thank all the nurses and physicians at Liaoyang Diabetes Hospital for assisting in the acquisition of data.