CORRELATION BETWEEN SUBCLINICAL HYPOTHYROIDISM AND DYSLIPIDEMIA IN WOMEN IN NORTHEAST CHINA

Abstract

Context

It is well known that thyroid hormones are important, being involved in affects the metabolism of carbohydrate, protein, lipids. The relationship between thyroid hormones and lipid metabolism is the focus of recent research.

Objective

To investigate the relationship between subclinical hypothyroidism and lipid metabolism in women.

Design

We conducted an epidemiological survey of thyroid diseases among women in Northeast China from September 2014 to December 2014.

Subjects and Methods

A total of 1397 women underwent physical examinations and laboratory tests for thyroid function and lipid metabolism.

Results

We found that the detection rate of subclinical hypothyroidism was 13.03%. Patients with subclinical hypothyroidism showed significantly higher levels of triglyceride (1.69±1.9 vs. 1.45±1.4) and the risk of hyper triglyceridemia in women with thyroid stimulating hormone (TSH) levels ≥10mIU/L was 4.96-fold higher compared with that in the normal population (P<0.01).

Conclusion

Disorders of lipid metabolism in women with subclinical hypothyroidism show a direct correlation with the level of TSH, and the risk of hyper triglyceridemia is significantly increased when the level of TSH ≥10mIU/L.

INTRODUCTION

Subclinical hypothyroidism (SH) is a type of endocrine metabolic disease with no obvious clinical manifestations but only elevated thyroid stimulating hormone levels. In adults, the incidence of SH is 3%–18%, and the detection rate of SH in China is 0.91%–6.05%. This disease is 3–5-fold more common among women than in men.

Studies have indicated a close association between dyslipidemia and SH in women. An increase in total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C), particularly triglyceride (TG) metabolism, was associated with SH (1-4). To observe the effects of SH on lipid metabolism in Asian, a cross-sectional study conducted by Rajendra et al. found a significant increase in TG and LDL-C levels in patients with SH compared with a control group (5, 6). Most studies are done on women with chronic diseases or small samples (7-10). There is no epidemiological investigation on large groups of women to analyze the correlation between thyroid hormone level and lipid metabolism. Women with SH and increased lipid levels have a higher risk of cardiovascular disease (11-14). In order to investigate if thyroid function and dyslipidemia among women with SH in Northeast China area are correlated, we analyzed epidemiological survey data with the aim of developing references for treatment of disorders of lipid metabolism.

MATERIALS AND METHODS

Participants

Participants were women from Northeast China who presented at the First Hospital of Jilin University from September 2014 to December 2014. Inclusion criteria were women aged >18 years who could understand clearly and were willing to participate in this study. Exclusion criteria were inability to communicate; patients with thyroid disease or have taken medications affecting thyroid function, such as glucocorticoids, amiodarone, etc., in the past 3 months, history of thyroid disease, neck radiotherapy, and neck surgery; patients with known dyslipidemia and using lipid-lowering drugs; and pregnancy and postpartum women within 1 year.

After screening 1464 women, 1430 were enrolled and categorized based on thyroid stimulating hormone (TSH) levels into those with euthyroidism (n =1250;TSH:0.27–4.2mIU/L) or SH (n =180;TSH>4.2mIU/L) groups.

Questionnaire Survey

The Holistic stratified sampling survey method with unified guiding terms was adopted. Questionnaires were distributed, participants’ responses were reviewed, and missing information was timely supplemented.

Physical Examination and Laboratory Data

Subjects were examined for body weight, height, waistline, blood pressure, collecting fasting venous blood in the morning and urine samples.

Laboratory measurement indexes included fasting blood glucose,75-g oral glucose tolerance tests (OGTTs), glycosylated hemoglobin (HbA1c), TG, TC, LDL-C, high-density lipoprotein cholesterol (HDL-C), TSH, free thyroxine, urinary iodine, anti-thyroid peroxidase antibody(TPO-Ab), and anti-thyroglobulin antibody (Tg-Ab).

Statistical Analyses

Epidemiological data were recorded by two authors using Epidata software, and checked and imported to Microsoft Excel by a third author. All statistical analysis was performed using SPSS version 22 software. Normal distribution data were represented by mean ± standard deviation and between two variables to which partial correlation analysis was applied and logistic regression analysis was applied to binary variables. Multivariable linear analysis was used for analysis of multiple factors. P<0.05 was considered statistically significant.

RESULTS

A total of 1430 questionnaires were distributed and 1397 valid questionnaires were retrieved. The effective response rate was 97.7%.The detection rate of SH among women in Northeast China was 13.03%. Comparison of serum lipids, blood glucose, and other related indicators between the two groups showed that age, body mass index (BMI), HbA1c, TG(1.69±1.9 vs. 1.45±1.4), and LDL-C (3.32±0.77 vs. 2.91±0.82) levels were significantly higher in the SH group than in the euthyroidism group (P<0.05) (Table 1).

Table 1.

General information of female subclinical hypothyroidism and normal population

Variable	Subclinical hypothyroidism population (n=161)	Normal population (n=1236)	P
Age (years)	45(28;60)	41(28;53)	0.005**
Weight (kg)	64(57;74)	65(56;74)	0.83
Height (cm)	162.2±8.8	164.5±9.3	0.15
SBP(mmHg)	132(120;147)	130(119;144)	0.16
DBP(mmHg)	81(73;88)	79(72;88)	0.36
HR(times/ min)	79.1±12.7	80.4±11.9	0.06
WL(cm)	83.7±13.3	83.4±18.2	0.73
BMI(kg/m2)	24.5±3.6	24.6±7.6	0.02*
FBG(mmol/L)	5.4±1.6	5.3±1.5	0.37
PBG(mmol/L)	6.2±2.4	6.2±3.0	0.54
HbA1c (%)	5.3(5.1;5.7)	5.2(4.9;5.6)	0.01*
UIC(ug/L)	304(250;386)	320(261;386)	0.25
TG(mmol/L)	1.69±1.9	1.45±1.4	0.000***
TC(mmol/L)	4.77±1.1	4.53±0.86	0.05
LDL(mmol/L)	3.32±0.77	2.91±0.82	0.018*
HDL(mmol/L)	1.35±0.20	1.37±0.45	0.15
TPO-Ab(IU/L)	52.6±105.5	21.7±53.8	0.002**
Tg-Ab(IU/L)	120.9±248.6	63.8±435.8	0.001**

Note: *P<0.05,**P<0.01,***P<0.001. BMI(body mass index), FBG (fasting blood glucose): 3.9~6.1mmol/L, PBG (postprandial blood glucose): 3.9~7.8mmol/L, HbA1c (hemoglobin A1c): 4.27~6.07%, UIC (urinary iodine concentration): 100~200ug/L, TG (triglyceride): 0.28~1.80mmol/ L, TC (total cholesterol): 2.6~6.0mmol/ L, LDL-C(low-density lipoprotein cholesterol): 2.06~3.10mmol/L, HDL-C (high-density lipoprotein cholesterol): 0.76~2.1mmol/L, TPO-Ab (Antithyroid peroxidas antibody): 0~34IU/L, Tg-Ab(Antithyroglobulin antibody): 0~110IU/L.

After controlling for age and BMI as confounding factors, partial correlation analysis of TSH and TG, TC, LDL-C, HDL-C in women showed a positive correlation between TSH and TG and LDL-C levels (r = 0.20, r = 0.09, P<0.01) (Table 2). To further clarify the effect of varying TSH levels on lipid metabolism, SH patients were stratified using TSH = 10mIU/L as cut-off, and the correlation between different TSH levels and TG and LDL-C was analyzed using logistic regression analysis, which showed a significant correlation between TG and TSH. Patients in the SH group with 4.2<TSH<10mIU/L and TSH≥10mIU/L showed 1.97-fold and 4.96-fold significantly higher risk of hyper triglyceridemia than subjects with normal TSH (OR = 1.97, P= 0.004; OR=4.96, P=0.002) (Table 3).

Table 2.

Partial Correlation Analysis of TSH and Lipids in Women after Controlling for Confounding Factors (Age, BMI)

	TSH
	r	P
Triglycerides	0.20	0.000***
Total cholesterol	0.08	0.05
Low-density lipoprotein cholesterol	0.09	0.002**
High-density lipoprotein cholesterol	-0.02	0.573

Note: **P<0.01,***P<0.001.

Table 3.

Compare the risk of hypertriglyceridemia in different TSH levels in the subclinical hypothyroidism

	Hypertriglyceridemia
	B	OR	P	95% CI
0.27<TSH≤4.2(mIU/L)		1
4.2<TSH<10(mIU/L)	0.68	1.97	0.004**	(1.48, 2.62)
TSH≥10(mIU/L)	1.84	4.96	0.002**	(1.83, 13.51)

Note: **P<0.01.

To determine whether thyroid antibodies affect the level of blood lipids, subjects were divided into four groups: TPO+/Tg-, Tg+/TPO-, TPO+/Tg+, and TPO-/Tg-. While there was no significant difference in the levels of blood lipids between the four groups, logistic regression analysis showed that the risk of LDL-C abnormality in TPO+/Tg- group was 1.69-foldsignificantly higher than that in TPO-/Tg- group (OR =1.69, 95% CI 1.07–2.71, P<0.05) (Table 4). Levels of TPO-Ab in the SH group was positively correlated with LDL-C (r = 0.15, P = 0.04) (Table 5).

Table 4.

The risk of hyperlipidemia in antibody positive group compared with antibody negative group

	TG		TC
	B	OR	P	(95% CI)	B	OR	P	(95% CI)
TPO+/Tg-	-0.30	0.74	0.25	(0.44,1.24)	0.40	1.49	0.10	(0.92,2.40)
Tg+/TPO-	-0.22	0.80	0.40	(0.47, 1.35)	0.24	1.27	0.35	(0.77,2.12)
TPO+/Tg+	-0.08	0.93	0.71	(0.63, 1.38)	0.10	1.10	0.64	(0.73,1.66)
	LDL		HDL
	B	OR	P	(95% CI)	B	OR	P	(95% CI)
TPO+/Tg-	0.53	1.69	0.03#	(1.07,2.71)	-0.28	0.76	0.37	(0.42,1.381)
Tg+/TPO-	0.12	1.13	0.65	(0.67,1.20)	-0.30	0.74	0.35	(0.40,1.382)
TPO+/Tg+	0.11	1.12	0.58	(0.74,1.69)	-0.26	0.77	0.28	(0.48,1.24)

Note: ^#P<0.05.

Table 5.

Correlation analysis of TPO antibody and LDL cholesterol

	LDL
	r	P
Normal population	0.03	0.20
Subclinical hypothyroidism population	0.15	0.04#

Note: ^#P<0.05.

DISCUSSION

We found that in a female population diagnosed with SH in Northeast China, age and TGs and LDL-C levels were significantly increased compared with those with euthyroidism. After adjusting for confounding factors of age and BMI, TSH in women showed a positive correlation with TGs and LDL-C, which was consistent with the results of Gardun-Garcia et al. (15).

We found that women with SH faced a high risk of hyper triglyceridemia and LDL-cholesterolemia compared to those with normal thyroid function. After stratifying patients according to their TSH levels, women with SH had a significantly higher risk of hyper triglyceridemia than women with normal thyroid function. A possible mechanism attributed to this phenomenon is the increase inexpression of interleukin (IL)-6 and tumor necrosis factor (TNF)-α in adipocytes when the TSH level is increased (16-18). Further, both IL-6 and TNF-α can promote lipodieresis, leading to insulin resistance and the occurrence of hyper triglyceridemia (19-21).

Anti-thyroid antibody is often used as an important index that reflects the prognosis of SH, and also aggravates vascular endothelial dysfunction leading to atherosclerosis (22, 23). From this it can be inferred that anti-thyroid antibodies may also affect blood lipid level, there by increasing the risk of cardiovascular disease, though this needs to be validated.

In this study, we showed that the TPO-Ab and Tg-Ab levels in patients with SH were significantly higher than that of the normal group, though the levels of blood lipids were similar between groups when TPO-Ab and Tg-Ab were positive. However, the higher risk of LDL-C abnormality in TPO-Ab+/Tg-Ab- group compared to TPO-Ab-/Tg-Ab- group indicated that the level of TPO-Ab in clinical hypothyroidism patients was significantly related to LDL-C, in concordance with results published by Teixeira et al. (24, 25).

It is plausible that an increase in TPO-Ab in patients with SH may enhance expression of apolipoproteins, ApoB/ApoA, which are protein components of lipoprotein. Therefore, female patients with SH with positive TPO-Ab may show an increased risk of cardiovascular disease.

In conclusion, our epidemiological study comprising a large cohort of women with SH residing in Northeast China showed that these patients had significantly elevated serum lipids and TSH levels, particularly TSH levels of 10 mIU/L, increased the risk for hyper triglyceridemia. Therefore, controlling TSH level during treatment for SH is of significant clinical importance to prevent hyperlipidemia and its associated complications.

Conflict of interest

The authors declare that they have no conflict of interest.