Relationship Between Thyroid Function and the Prevalence of Type 2 Diabetes Mellitus in Euthyroid Subjects
J Clin Endocrinol Metab, February
The Relationship Between Thyroid Function and the Prevalence of Type 2 Diabetes Mellitus in Euthyroid Subjects
Yeqing Gu 2
Huihui Li 2
Xue Bao 2
Qing Zhang 0
Li Liu 0
Ge Meng 2
Hongmei Wu 2
Huanmin Du 2
Hongbin Shi 0
Yang Xia 2
Qian Su 2
Liyun Fang 2
Fei Yu 2
Huijun Yang 2
Bin Yu 2
Shaomei Sun 0
Xing Wang 0
Ming Zhou 0
Qiyu Jia 0
Qi Guo 1
Hong Chang 2
Guolin Wang 0
Guowei Huang 2
Kun Song 0
Kaijun Niu 0 2
0 Health Management Centre, Tianjin Medical University General Hospital , 300070 Tianjin , China
1 Department of Rehabilitation and Sports Medicine Tianjin Medical University , 300070 Tianjin , China
2 Nutritional Epidemiology Institute and School of Public Health
Purpose: Thyroid hormones (THs) are primarily responsible for the regulation of energy balance and metabolism, suggesting that TH levels may contribute to the development of type 2 diabetes mellitus (T2DM). However, few studies have investigated the relationship between TH and T2DM in a general population. The aim of this study was to evaluate whether serum TH levels within the reference range are related to T2DM. Methods: A cross-sectional study (n = 15,296) was performed in Tianjin, China. Serum free triiodothyronine (FT3), free thyroxine (FT4), and thyroid-stimulating hormone (TSH) levels were measured by chemiluminescence immunoassay, and T2DM was defined according to the American Diabetes Association criteria. Multiple logistic regression models were used to assess the sexspecific relationships between FT3, FT4, FT3/FT4 ratios, and TSH quintiles and T2DM. Results: The prevalence of T2DM was 16.2% in males and 7.7% in females. In males, the multivariable-adjusted odds ratios (95% confidence interval) of T2DM for increasing quintiles of FT3, FT4, and FT3/FT4 ratios were 1.00, 0.75(0.63 to 0.89), 0.70(0.58 to 0.84), 0.63(0.52 to 0.76), 0.56 (0.46 to 0.68; P for trend , 0.0001); 1.00, 1.05(0.87 to 1.27), 1.16(0.96 to 1.40), 1.09(0.90 to 1.31), 1.29 (1.07 to 1.56; P for trend = 0.01); and 1.00, 0.69(0.58 to 0.83), 0.72(0.60 to 0.86), 0.59(0.48 to 0.71), and 0.55(0.46 to 0.66; P for trend , 0.0001), respectively. Similar results also were observed in females. In contrast, a strong negative correlation between TSH and T2DM was observed in males, but not in females. Conclusions: This study demonstrated that decreased FT3, FT3/FT4 ratios, and increased FT4 levels are independently related to a higher prevalence of T2DM in both males and females, and TSH is inversely related to T2DM in males only. (J Clin Endocrinol Metab 102: 434-442, 2017)
Tchronic endocrine disease, characterized by
hyperype 2 diabetes mellitus (T2DM) is the most common
glycemia resulting from impaired insulin secretion and/or
insulin resistance (
). The long-term complications of
T2DM can significantly increase the risks of
cardiovascular disease (CVD) and cancer, among many other
diseases, as well as significantly increase risk of mortality
). The global prevalence of diabetes mellitus is
rapidly increasing due to an ageing population,
urbanization and associated lifestyle changes. In 2013, an
estimated 382 million people worldwide had diabetes
mellitus, about 90 to 95% of whom had T2DM, and the
Abbreviations: BMI, body mass index; BP, blood pressure; CI, confidence interval; CVD,
cardiovascular disease; FT3, free triiodothyronine; FT4, free thyroxine; LDL, low-density
lipoprotein; OR, odds ratio; SD, standard deviation; T2DM, type 2 diabetes mellitus; T3,
triiodothyronine; T4, thyroxine;TC, total cholesterol; TG, triglyceride; TH, thyroid
hormone; TSH, thyroid-stimulating hormone.
number will increase to 592 million (8.8% of adults aged
20 to 79 years) by 2035 (1). In particular, the latest
national survey suggests that China has become the
global epicenter of the T2DM epidemic with more than
11.6% of the adult population (aged 18 years and over)
suffering from diabetes mellitus (
Clarifying the common pathophysiologic mechanisms
of T2DM is a crucial step toward providing early
prevention and treatment. Recently, increased interest has
focused on the relationship between thyroid function and
metabolic diseases, including T2DM (
). The thyroid
function is primarily responsible for the regulation of
energy balance and metabolism (
). Thyroid dysfunction
increases muscle and adipose tissue insulin resistance (
and decreases glucose transport in myocytes (
Meanwhile, thyroid hormone (TH) stimulates the basal
expressions of glucose transporters, which regulate the
intracellular glucose uptake on the surface of myocytes
). Moreover, recently studies have demonstrated that
free triiodothyronine (FT3) regulates insulin secretion (
). Because glucose metabolism and insulin secretion
are most closely related to the pathogenesis of T2DM,
it is hypothesized that TH is a useful predictive factor
for developing T2DM. On the other hand,
thyroidstimulating hormone (TSH) binds to receptors on
epithelial cells in the thyroid gland, stimulating synthesis and
secretion of TH by negative feedback inhibition (11).
However, to date, few studies have evaluated the
relationship between TH, TSH, and T2DM in the general
population with euthyroid status. The aim of this study
was to investigate whether serum TH concentrations
within the reference range as well as TSH levels are
related to the prevalence of T2DM among a large-scale
Materials and Methods
A large prospective dynamic cohort study, called The Tianjin
Chronic Low-Grade Systemic Inflammation and Health Cohort
Study, was carried out in a general adult population living in
Tianjin, China. The study was based on annual health
examinations conducted in Tianjin Medical University General
Hospital Health Management Center and focused on the
relationship between chronic low-grade systemic inflammation
and the health status. Participants who had received health
examinations (including medical examinations such as blood
tests, abdominal ultrasonography, etc.) and had completed
questionnaires regarding their smoking and drinking habits and
disease history over the course of January 2007 to December
2015 were recruited. Moreover, a detailed lifestyle
questionnaire covering family income, marital status, employment
status, educational level, physical activity, sleep habits, dietary
habits, overall computer/mobile device usage time, television
time, history of prior infections, and use of medicines, as well as
physical performance tests were administered to randomly
selected subjects from this population since May 2013.
This cross-sectional study used data from the Tianjin
Chronic Low-Grade Systemic Inflammation and Health Cohort
Study ranging from 2013 to 2015. The participant selection
process was described in Fig. 1. During the research period,
there were 18,682 participants who had received at least 1
health examination including blood glucose, TH, and TSH tests
agreed to participate and provided written informed consent for
their data to be analyzed. We excluded those with a history of
CVD (n = 1461) or cancer (n = 286). Moreover, participants
who having a level exceeding the standard reference range of
TH and/or TSH [FT3 ,3.5 pmol/L (n = 43) or .6.5 pmol/L
(n = 237), free thyroxine (FT4) ,11.5 pmol/L (n = 113)
or .22.7 pmol/L (n = 84), TSH ,0.55 mIU/L (n = 251)
or .4.78 mIU/L (n = 938)] were excluded. Moreover,
participants who had a history of thyroid disease, type 1 diabetes
mellitus, or who had used antithyroid drugs were not included
in the current study. Owing to these exclusions, the final
crosssectional study population comprised 15,269 participants
including 8970 males [mean age 6 standard deviation (SD): 48.1 6
10.6 years] and 6299 females (mean age 6 SD: 47.6 6 11.2 years).
Assessment of T2DM
Levels of fasting blood sugar were measured by glucose
oxidase method. Blood samples for analysis of HbA1c were
mixed with ethylenediaminetetraacetic acid (as an
anticoagulant) before testing. HbA1c separation and quantification were
performed using a high-performance liquid chromatography
analyzer (HLC-723 G8; Tosoh, Tokyo, Japan) with intra- and
interassay coefficients of variation of ,3%. To measure 2-hour
serum glucose, subjects were given a standard 75-g glucose
solution, and serum glucose was measured at 2 hours after
administration during the oral glucose tolerance test. In
undiagnosed participants, T2DM was defined as a fasting blood
sugar level $126 mg/dl (7.0 mmol/L), oral glucose tolerance
test $200 mg/dl (11.1 mmol/L), HbA1c $48 mmol/mol (6.5%),
or a history of T2DM based on the American Diabetes
Association 2013 criteria (
FT3, FT4, and TSH measurements
Serum FT3 and FT4 were measured by chemiluminescence
immunoassay using ADVIA Centaur FT3 analyzer and ADVIA
Centaur FT4 analyzer (Siemens Healthcare Diagnostics, New
York, NY) and expressed as pmol/L. The measuring range of
FT3 and FT4 were 0.3 to 30.8 pmol/L and 1.3 to 155 pmol/L,
respectively. Serum TSH was measured by chemiluminescence
immunoassay using ADVIA Centaur TSH3-Ultra analyzer
(Siemens Healthcare Diagnostics) and expressed as mIU/L. The
measuring range was 0.001 to 150 mIU/L. The reference ranges
of FT3, FT4, and TSH were 3.70;6.93 pmol/L, 11.61;21.41
pmol/L, and 0.55;4.87 mIU/L, respectively. We divided
participants into 5 categories (quintiles) according to the actual
concentrations of FT3, FT4, and TSH.
Assessment of other variables
Blood pressure (BP) was measured twice in the right arm
using an automatic device (Andon, Tianjin, China) after 5
minutes of rest in a seated position. The mean of these 2
measurements was taken as the BP value. Hypertension is
defined as having a BP higher than 140/90 mm Hg (systolic BP/
covariance for continuous variables and
multiple logistic regression analysis for
proportional variables after adjustment for
age. The prevalence of T2DM was used as
dependent variables, and quintiles of FT3,
FT4, and TSH concentrations were used as
independent variables. The multiple logistic
regression models were used to examine the
relationships between quintiles of FT3,
FT4, and TSH and the prevalence of T2DM
with adjustment for the covariates: age,
BMI, waist circumference, smoking status,
drinking status, hypertension,
hyperlipidemia, and family history of CVD,
hypertension, hyperlipidemia, and diabetes.
Because the previous studies suggested an
important role of thyroxine (T4) to
triiodothyronine (T3) conversion (
additionally assessed the relationship
between the quintiles of FT3/FT4 ratios and
the prevalence of T2DM. Odds ratios
(ORs) with their corresponding 95% CIs
were calculated. Furthermore, because
certain medicines (such as nonsteroidal
anti-inflammatory drug, estrogen,
antiepileptic drugs, etc.) may affect thyroid
), a sensitivity analysis was performed after
excluding the subjects who reported taking these drugs. All
P values for linear trends were calculated using the median value
of quintiles of FT3, FT4, FT3/FT4 ratios, and TSH. All tests
were 2-tailed and P , 0.05 was defined as statistically
The prevalence of T2DM is 16.2% (1449/8970) and
7.7% (488/6,99) in males and females, respectively.
Among participants with T2DM, 482 participants [368
(25.4%) males and 114 (23.4%) females] reported that
they take medications for diabetes. FT3 and FT4 levels
were both significantly higher in males than in females
[means (SD), in males: 5.5 (0.5) pmol/L, 16.8 (2.0) pmol/
L; in females: 5.0 (0.5) pmol/L, 15.7 (1.8) pmol/L, both
P , 0.0001], whereas the TSH level was significantly
lower in males than in females [means (SD), in males: 1.9
(0.8) mIU/L; in females: 2.3 (1.0) mIU/L, P , 0.0001].
Age-adjusted participant characteristics in relation to
T2DM were presented in Table 1. Compared with
participants without T2DM, those with T2DM tended to be
older and to have higher BMI, waist circumference, TC,
TG, systolic BP, diastolic BP, the proportion of family
history of hyperlipidemia and diabetes, and lower
highdensity lipoprotein (all P values , 0.01) in both males
and females. However, there are also some differences
between males and females. Compared with participants
without T2DM, males with T2DM tended to have lower
FT3 and FT3/FT4 ratios and higher FT4 levels (all
P values , 0.05); and females with T2DM tended to have
diastolic BP) or having a history of hypertension. As for lipids,
triglycerides (TGs) and total cholesterol (TC) were measured by
enzymatic methods. Low-density lipoprotein (LDL) was
measured by the polyvinyl sulfuric acid precipitation method, and
high-density lipoprotein was measured by the chemical
precipitation method using appropriate kits on a Cobas 8000
analyzer (Roche, Mannheim, Germany). Hyperlipidemia was
defined as TC $5.17 mmol/L, TG $1.7 mmol/L, LDL $3.37
mmol/L, or history of hyperlipidemia. Height and body weight
were measured using a standard protocol, and body mass index
(BMI) was calculated as weight/height2 (kg/m2). Waist
circumference was measured at the umbilical level with subjects
standing and breathing normally. Information on age, sex, and
smoking and drinking status was obtained from a questionnaire
survey. A detailed personal and family history of physical illness
and current medications was noted from “yes” or “no”
responses to relevant questions.
All statistical analyses were performed using the Statistical
Analysis System 9.3 edition for Windows (SAS Institute Inc.,
Cary, NC). Because previous studies have reported that
sexspecific difference was observed on the levels of TH, TSH, and
the incident of T2DM (
), we analyzed the relationships
between TH, TSH, and T2DM by sex. Distributions of
continuous variables were assessed for normality using the
Kolmogorov-Smirnov (n . 2,000) or Shapiro-Wilk (n # 2,000)
test. Because the distributions of all the continuous variables
were not normal in the current study, the natural logarithm was
applied to normalize the data before statistical analysis. The
continuous covariates after the log transformation approached
normal distribution. Descriptive data are presented as the
geometric mean [95% confidence interval (CI)] for adjusted
continuous variables and as percentages for categorical
variables. For baseline characteristics analysis, the differences
among T2DM categories were examined using analysis of
No. of subjects
Family history of
aAnalysis of covariance or multiple logistic regression analysis.
bGeometric mean (95% CI; all such values).
higher LDL and FT4 levels and lower FT3/FT4 ratios (all
P values , 0.01). No significant differences were observed
among the TSH levels, proportion of current smokers,
alcohol consumers and the proportion of those with a family
history of CVD and hypertension (all P values . 0.05) in
both males and females.
The crude and adjusted relationship between FT3, FT4,
TSH, and the prevalence of T2DM was indicated in Table 2.
In males, the adjusted ORs (95% CI) of T2DM were related
to the gradual increase of the FT3, FT4, FT3/FT4 ratios, and
TSH concentrations as compared with participants who had
the lowest concentrations and were as follows: FT3: 0.75
(0.63 to 0.89), 0.70 (0.58 to 0.84), 0.63 (0.52 to 0.76), and
0.56 (0.46 to 0.68; P for trend , 0.0001); FT4: 1.05 (0.87 to
1.27), 1.16 (0.96 to 1.40), 1.09 (0.90 to 1.31), and 1.29
(1.07 to 1.56; P for trend = 0.01); FT3/FT4 ratios: 0.69 (0.58
to 0.83), 0.72 (0.60 to 0.86), 0.59 (0.48 to 0.71), and 0.55
(0.46 to 0.66; P for trend , 0.0001); and TSH: 0.89 (0.74 to
1.08), 0.94 (0.78 to 1.13), 0.76 (0.63 to 0.92), and 0.80
(0.67 to 0.97; P for trend = 0.01). In females, the adjusted
ORs between FT3, FT4, FT3/FT4 ratios, TSH, and the
prevalence of T2DM were as follows: FT3: 1.00, 0.97 (0.72
to 1.30), 0.91 (0.67 to 1.22), 0.71 (0.52 to 0.96), 0.63 (0.46
to 0.86; P for trend , 0.01); FT4: 1.00, 0.81 (0.58 to 1.13),
1.15 (0.84 to 1.58), 1.12 (0.81 to 1.54), and 1.54 (1.14 to
2.08; P for trend , 0.0001); FT3/FT4 ratios: 1.00, 0.79
(0.59 to 1.05), 0.56 (0.41 to 0.76), 0.64 (0.47 to 0.85), and
0.57 (0.42 to 0.78; P for trend , 0.0001); and TSH: 1.00,
1.22 (0.91 to 1.66), 0.83 (0.60 to 1.15), 0.89 (0.65 to 1.22),
and 0.82 (0.60 to 1.12; P for trend = 0.04).
We further performed a sensitivity analysis after
excluding the subjects who reported taking certain
medicines (such as nonsteroidal anti-inflammatory drug,
estrogen, antiepileptic drugs, etc.; n = 220). However, the
relationship between TH, TSH and T2DM
change. The adjusted OR (95% CI) of T2DM for
increasing quintiles of TH and TSH for males were as
follows: FT3: 1.00 (reference), 0.75 (0.63 to 0.89), 0.70
(0.58 to 0.84), 0.63 (0.53 to 0.76), and 0.56 (0.46 to 0.69;
P for trend , 0.0001); FT4: 1.00 (reference), 1.05 (0.87
to 1.27), 1.14 (0.94 to 1.38), 1.09 (0.90 to 1.32), and 1.29
(1.07 to 1.57; P for trend = 0.01); FT3/FT4 ratios: 1.00
(reference), 0.70 (0.58 to 0.84), 0.73 (0.61 to 0.87), 0.59
(0.49 to 0.71), and 0.55 (0.46 to 0.67; P for trend
aMultiple logistic regression analysis.
bAdjusted ORs (95% CI; all such values).
cAdjusted for age, BMI, waist circumference, smoking status, drinking status, hypertension, hyperlipidemia, and family history of cardiovascular diseases,
hypertension, hyperlipidemia, and diabetes.
Number of subjects
Number of diabetes
Number of subjects
Number of diabetes
Number of subjects
Number of diabetes
Number of subjects
Number of diabetes
Number of subjects
Number of diabetes
Number of subjects
Number of diabetes
Number of subjects
Number of diabetes
Number of subjects
Number of diabetes
3.50 to 5.07
11.50 to 15.08
0.18 to 0.29
0.55 to 1.22
3.50 to 4.59
11.50 to 14.08
0.16 to 0.29
0.55 to 1.42
, 0.0001); and TSH: 1.00 (reference), 0.89 (0.74 to 1.08),
0.93 (0.77 to 1.12), 0.75 (0.62 to 0.91), and 0.80 (0.66 to
0.97; P for trend , 0.01). In females, the adjusted OR
(95% CI) of T2DM for increasing quintiles of TH and
TSH were as follows: FT3: 1.00 (reference), 0.95 (0.71 to
1.29), 0.87 (0.64 to 1.18), 0.70 (0.51 to 0.95), and 0.62
(0.45 to 0.86; P for trend , 0.001); FT4: 1.00 (reference),
0.88 (0.62 to 1.24), 1.21 (0.88 to 1.68), 1.20 (0.86 to
1.66), and 1.67 (1.23 to 2.28; P for trend , 0.0001); FT3/
FT4 ratios: 1.00 (reference), 0.77 (0.57 to 1.02), 0.52
(0.38 to 0.71), 0.61 (0.45 to 0.82), and 0.53 (0.39 to 0.72;
P for trend , 0.0001); and TSH: 1.00 (reference), 1.17
(0.86 to 1.59), 0.83 (0.60 to 1.15), 0.83 (0.60 to 1.14),
and 0.77 (0.56 to 1.06; P for trend = 0.02).
The current study has assessed the relationships between
TH, TSH, and T2DM in an adult population. The results
suggest that, after adjustment for confounding factors in
both males and females, a higher prevalence of T2DM
has a negative correlation to FT3 and a positive
correlation to FT4. Furthermore, a negative correlation was
observed between TSH and T2DM in males, but not in
females. To our knowledge, this study firstly
demonstrated that TH levels within reference range are
significantly related to the prevalence of T2DM.
We adjusted for multiple potentially confounding
factors in our analysis. This study suggests that numerous
factors (age, sex, BMI, drinking, smoking status, family
history of some diseases) are correlated with the
prevalence of T2DM. Because studies have shown that serum
TH and TSH levels are related to age (
) and BMI (
we first adjusted for these 2 variables. Adjustment for age
and BMI significantly affected the relationship between
serum FT4 levels and T2DM in males, leading us to
conclude that age and BMI are major confounding
factors. We subsequently adjusted for waist circumference,
smoking status, drinking status, hypertension,
hyperlipidemia, TC, TG [influential factors on TH and TSH
)], and genetic factors, such as family history
of CVD, hypertension, hyperlipidemia, and diabetes
[influential factors on T2DM (
)]. However, after these
adjustments, serum FT3 levels had a more obvious
correlation with T2DM in both males and females.
Several studies have explored the relationship between
TH, TSH, and insulin resistance or glucose levels. Four
cross-sectional studies reported that serum FT4 was
negatively related to insulin resistance, fasting insulin or
the homoeostasis model assessment index for insulin
resistance, whereas a positive relationship was found
between TSH and insulin resistance (
). Two other
cross-sectional and a small-scale case-control studies
found a positive correlation among higher T3 (30), T3 to
T4 ratio (
), FT3 and FT4 (
) levels, and high fasting
glucose levels and insulin resistance. In contrast, a
crosssectional study showed that FT3, FT3 to FT4 ratio, and
TSH are significantly and negatively associated with
HbA1c levels in patients with T2DM (
). Although the
reasons for these discrepancies remain unclear, the
differences in confounding factors, study setting, small
sample sizes for some studies, use of T3 instead of FT3, and
the difference of outcome indicators might partly explain
the cause for conflicting results. On the other hand, to date,
few studies have illustrated the relationship between TH,
TSH, and T2DM in the general population. This study
demonstrates that T2DM is negatively related to FT3 and
positively related to FT4 in both males and females.
Further studies are necessary to explore whether these
results can also be observed in other adult populations.
Our results showed a similar relationship between TH
and T2DM in both males and females, but sex difference
was observed in the relationship between TSH and
T2DM. Still, detailed molecular mechanisms remain
unclear, because sex hormones (such as estrogen, and
testosterone, etc.) can regulate the thyroid function (
and estrogen levels affect the development of T2DM (
The difference in sex hormones may partly explain the
sex-difference in the relationship between TSH and
T2DM. However, because levels of sex hormones such as
testosterone and estrogen were not measured in this
study, further research is needed to explore this issue. In
addition, because the sample size was smaller for females
and the prevalence of T2DM was markedly lower in
females (7.7%) than in males (16.2%), the precision and
statistical power of the analysis may be lower for females.
Further large-scale population studies are required to
confirm the above findings. On the other hand, in males,
the significance of the relationship between FT3 and
T2DM started from the second quintile, but from the
fourth quintile in females. Because FT3 concentrations on
the fourth quintile in females are equivalent to the second
quintile in males, it is speculated that the increased
prevalence of T2DM depends on the concentrations of
FT3 regardless of sex. Further studies are needed to
clarify this hypothesis.
Multiple putative mechanisms could explain the
relationship between thyroid function and T2DM. On the 1
hand, T3 regulates hepatic gluconeogenesis and
antagonize insulin action (
), even in the euthyroid state (
T3 modulates mRNA and protein expression of the
glucose transporter 4, adenosine monophosphate–activated
protein kinase, and acetyl coenzyme A carboxylase in
skeletal muscle (
). Furthermore, increases in plasma T3
levels impair the ability of insulin to suppress hepatic
glucose production and to increase glucose uptake in
). Interestingly, even subtle increases in the
levels of T3 or T4 within the physiological range have
been shown to induce insulin resistance (
). On the other
hand, inadequate insulin secretion is an early event in the
natural history of T2DM (
). There is convincing
evidence from independent research that T3 directly
increases islet b-cell mass via thyroid hormone receptor
a-dependent pathways (
). Moreover, insulin secretion
from b-cells is potentially controlled by the truncated
mitochondrial T3 receptor p43 (
). These changes have
been associated with a decrease in specific glucose
transporters, namely GLUT2 and Kir6.2, and may thus
be more broadly mediated by the control of intracellular
glucose availability, which may have implications for
other actions of T3 (
). FT3 is also a powerful inducer of
pancreatic acinar cell proliferation in rodents (
promotes expression of important proteins involved in
both glucose and lipid metabolism that may influence
insulin secretion (
). Finally, an interaction between
TSH and insulin sensitivity has been proposed by several
). Nonetheless, no direct effect of insulin or
insulin resistance on thyroid function in humans has yet
Interestingly, the current study has also found a strong
and inverse link between the FT3/FT4 ratios and T2DM.
It is known that TSH upregulates deiodinase expression
and activity (
). Higher peripheral deiodinase
activity increases conversion of FT4 to FT3 (
15, 16, 48
Thus, FT3/FT4 ratios can be considered an indicator of
peripheral deiodinase activity. Because the inhibition of
peripheral deiodinase activity lowers basal metabolic rate
), and basal metabolic rate is closely related to the
pathogenesis of T2DM, it is possible that the FT3/FT4
ratios were inversely related to T2DM because of
the regulation of basal metabolic rate by
deiodinasemediated TH signaling (
). In contrast, 3 small-scale
cross-sectional studies reported that the FT3/FT4 ratio
positively correlates with the homeostasis model
assessment for insulin resistance in obese adolescents with
nonalcoholic fatty liver disease (n = 200) (
), with both
waist circumference and BMI in obese women aged 18 to
68 years (n = 201) (
), and with nonalcoholic fatty liver
disease in patients with euthyroidism or hypothyroidism
(n = 115) (
). The differences in population and sample
size may be the cause for the observational discrepancies
between the aforementioned studies and the current
study. Furthermore, a recent study suggested that as TSH
levels increase, FT3/FT4 ratios increase until age 40, but
this differential increase does not occur in older age
). Based on this finding, we performed a
stratified analysis to ascertain whether age (,40 years
or $40 years) confounds the relationship between FT3/
FT4 ratios and T2DM. However, in the final multiple
logistic regression model, similar results were observed in
subjects from different age groups (P for interaction = 0.54
in males and 0.58 in females). Therefore, we speculated
that although age mediates the TSH regulation of FT3/FT4
ratios, FT3/FT4 ratios are significantly related to T2DM
independent of age. Further studies are needed to explore
the potential mechanisms underlying this association.
The current study has several limitations. Firstly, this is a
cross-sectional study, which is impossible to infer causality.
Further cohort studies and intervention trials should be
undertaken to establish a causal relationship between TH
and T2DM. However, the present large-scale
crosssectional study supports the important hypothesis that
TH levels even within the euthyroid range, may contribute
to the development of T2DM in the general population.
Secondly, because this population-based study was carried
out among apparently healthy adult population, and only
euthyroid subjects were included in the final analysis, only
1 serum TH test and 1 TSH test were measured in this
study. Therefore, further high-quality research is necessary
to confirm these results. Thirdly, although we adjusted for a
considerable number of potentially confounding factors,
we cannot exclude the possibility that T2DM is affected by
other lifestyle variables including supplementation of
iodine, which is intrinsically related to TH levels. Thus, a
well-designed randomized controlled trial is required to
verify these results. Finally, although the participants with
thyroid disease were excluded, serum thyroperoxidase
antibodies levels were not routinely checked in the general
population. Therefore, we cannot fully exclude the
possibility that the relationship between the categories of TH,
TSH, and T2DM is affected by these patients with positive
Decreased FT3 and FT3/FT4 ratios and increased FT4
levels were independently related to the prevalence of
T2DM among the adult population. A significantly
negative relationship between TSH and T2DM was
observed in males but not in females. Future studies
should be aimed at clarifying the cause-and-effect
relationship between TH and T2DM.
The authors gratefully acknowledge all the people that have
made this study.
Address all correspondence and requests for reprints to:
Kaijun Niu, Nutritional Epidemiology Institute and School of
Public Health, Tianjin Medical University, 22 Qixiangtai Road,
Heping District, Tianjin 300070, China. E-mail: nkj0809@
gmail.com or .
This study was supported by grants from the National
Natural Science Foundation of China (No. 81302422,
81673166, 81372118 and 81372467), the key technologies
R&D program of Tianjin (Key Project: No. 11ZCGYSY05700,
12ZCZDSY20400, 13ZCZDSY20200, and 15YFYZSY00020),
the National Science and Technology Support Program (No.
2012BAI02B02), 2012 and 2016 Chinese Nutrition Society
(CNS) Nutrition Research Foundation—DSM Research Fund
(No. 2014-071 and 2016-046), the Technologies development
program of Beichen District of Tianjin (No. bcws2013-21 and
bcws2014-05), the technologies project of Tianjin Binhai New
Area (No. 2013-02-04 and 2013-02-06), the Science
Foundation of Tianjin Medical University (No. 2010KY28 and
2013KYQ24), the Key Laboratory of Public Health Safety
(Fudan University), Ministry of Education (No. GW2014-5),
and the National Training Programs of Innovation and
Entrepreneurship for Undergraduates (No. 201510062013), China.
Disclosure Summary: The authors have nothing to disclose.
Gu et al
1. Guariguata L , Whiting DR , Hambleton I , Beagley J , Linnenkamp U , Shaw JE . Global estimates of diabetes prevalence for 2013 and projections for 2035 . Diabetes Res Clin Pract . 2014 ; 103 ( 2 ): 137 - 149 .
2. Leon BM , Maddox TM . Diabetes and cardiovascular disease: Epidemiology, biological mechanisms, treatment recommendations and future research . World J Diabetes . 2015 ; 6 ( 13 ): 1246 - 1258 .
3. Hua F , Yu JJ , Hu ZW . Diabetes and cancer, common threads and missing links . Cancer Lett . 2016 ; 374 ( 1 ): 54 - 61 .
4. Yang W , Lu J , Weng J , Jia W , Ji L , Xiao J , Shan Z , Liu J , Tian H , Ji Q , Zhu D , Ge J , Lin L , Chen L , Guo X , Zhao Z , Li Q , Zhou Z , Shan G , He J ; China National Diabetes and Metabolic Disorders Study Group. Prevalence of diabetes among men and women in China . N Engl J Med . 2010 ; 362 ( 12 ): 1090 - 1101 .
5. Iwen KA , Schr o¨der E , Brabant G. Thyroid hormones and the metabolic syndrome . Eur Thyroid J . 2013 ; 2 ( 2 ): 83 - 92 .
6. Bahi L , Garnier A , Fortin D , Serrurier B , Veksler V , Bigard AX , Ventura-Clapier R. Differential effects of thyroid hormones on energy metabolism of rat slow- and fast-twitch muscles . J Cell Physiol . 2005 ; 203 ( 3 ): 589 - 598 .
7. Dimitriadis G , Mitrou P , Lambadiari V , Boutati E , Maratou E , Panagiotakos DB , Koukkou E , Tzanela M , Thalassinos N , Raptis SA . Insulin action in adipose tissue and muscle in hypothyroidism . J Clin Endocrinol Metab . 2006 ; 91 ( 12 ): 4930 - 4937 .
8. Maratou E , Hadjidakis DJ , Kollias A , Tsegka K , Peppa M , Alevizaki M , Mitrou P , Lambadiari V , Boutati E , Nikzas D , Tountas N , Economopoulos T , Raptis SA , Dimitriadis G . Studies of insulin resistance in patients with clinical and subclinical hypothyroidism . Eur J Endocrinol . 2009 ; 160 ( 5 ): 785 - 790 .
9. Oda T , Taneichi H , Takahashi K , Togashi H , Hangai M , Nakagawa R , Ono M , Matsui M , Sasai T , Nagasawa K , Honma H , Kajiwara T , Takahashi Y , Takebe N , Ishigaki Y , Satoh J . Positive association of free triiodothyronine with pancreatic b-cell function in people with prediabetes . Diabet Med . 2015 ; 32 ( 2 ): 213 - 219 .
10. Kim TK , Lee JS , Jung HS , Ha TK , Kim SM , Han N , Lee EJ , Kim TN , Kwon MJ , Lee SH , Kim MK , Rhee BD , Park JH . Triiodothyronine induces proliferation of pancreatic b-cells through the MAPK/ERK pathway . Exp Clin Endocrinol Diabetes . 2014 ; 122 ( 4 ): 240 - 245 .
11. Sternberg RM , Thoemke KR , Korte JJ , Moen SM , Olson JM , Korte L , Tietge JE , Degitz SJ , Jr. Control of pituitary thyroid-stimulating hormone synthesis and secretion by thyroid hormones during Xenopus metamorphosis . Gen Comp Endocrinol . 2011 ; 173 ( 3 ): 428 - 437 .
12. American Diabetes Association. Standards of medical care in diabetes: 2013 . Diabetes Care . 2013 ; 36 ( Suppl 1 ): S11 - S66 .
13. Porcu E , Medici M , Pistis G , Volpato CB , Wilson SG , Cappola AR , Bos SD , Deelen J , den Heijer M , Freathy RM , Lahti J , Liu C , Lopez LM , Nolte IM , O'Connell JR , Tanaka T , Trompet S , Arnold A , Bandinelli S , Beekman M , B o¨hringer S , Brown SJ , Buckley BM , Camaschella C , de Craen AJ , Davies G , de Visser MC , Ford I , Forsen T , Frayling TM , Fugazzola L , Go¨ gele M, Hattersley AT , Hermus AR , Hofman A , Houwing-Duistermaat JJ , Jensen RA , Kajantie E , Kloppenburg M , Lim EM , Masciullo C , Mariotti S , Minelli C , Mitchell BD , Nagaraja R , Netea-Maier RT , Palotie A , Persani L , Piras MG , Psaty BM , R a¨ikko¨ nen K, Richards JB , Rivadeneira F , Sala C , Sabra MM , Sattar N , Shields BM , Soranzo N , Starr JM , Stott DJ , Sweep FC , Usala G , van der Klauw MM , van Heemst D , van Mullem A , Vermeulen SH , Visser WE , Walsh JP , Westendorp RG , Widen E , Zhai G , Cucca F , Deary IJ , Eriksson JG , Ferrucci L , Fox CS , Jukema JW , Kiemeney LA , Pramstaller PP , Schlessinger D , Shuldiner AR , Slagboom EP , Uitterlinden AG , Vaidya B , Visser TJ , Wolffenbuttel BH , Meulenbelt I , Rotter JI , Spector TD , Hicks AA , Toniolo D , Sanna S , Peeters RP , Naitza S. A meta-analysis of thyroid-related traits reveals novel loci and genderspecific differences in the regulation of thyroid function . PLoS Genet . 2013 ; 9 ( 2 ): e1003266 .
14. Krag MO , Hasselbalch L , Siersma V , Nielsen AB , Reventlow S , Malterud K , de Fine Olivarius N. The impact of gender on the long-term morbidity and mortality of patients with type 2 diabetes receiving structured personal care: a 13 year follow-up study . Diabetologia . 2016 ; 59 ( 2 ): 275 - 285 .
15. Bilgin H , Pirgon O ¨ . Thyroid function in obese children with nonalcoholic fatty liver disease . J Clin Res Pediatr Endocrinol . 2014 ; 6 ( 3 ): 152 - 157 .
16. G o¨ kmen FY , Ahbab S , Atao g˘lu HE, Tu¨ rker BC, Çetin F , Tu¨ rker F , Mamaç RY , Yenig u¨n M. FT3/FT4 ratio predicts non-alcoholic fatty liver disease independent of metabolic parameters in patients with euthyroidism and hypothyroidism . Clinics (Sao Paulo) . 2016 ; 71 ( 4 ): 221 - 225 .
17. De Pergola G , Ciampolillo A , Paolotti S , Trerotoli P , Giorgino R . Free triiodothyronine and thyroid stimulating hormone are directly associated with waist circumference, independently of insulin resistance, metabolic parameters and blood pressure in overweight and obese women . Clin Endocrinol (Oxf) . 2007 ; 67 ( 2 ): 265 - 269 .
18. Davies PH , Franklyn JA . The effects of drugs on tests of thyroid function . Eur J Clin Pharmacol . 1991 ; 40 ( 5 ): 439 - 451 .
19. Hamed SA . The effect of antiepileptic drugs on thyroid hormonal function: causes and implications . Expert Rev Clin Pharmacol . 2015 ; 8 ( 6 ): 741 - 750 .
20. Surks MI , Ortiz E , Daniels GH , Sawin CT , Col NF , Cobin RH , Franklyn JA , Hershman JM , Burman KD , Denke MA , Gorman C , Cooper RS , Weissman NJ . Subclinical thyroid disease: scientific review and guidelines for diagnosis and management . JAMA . 2004 ; 291 ( 2 ): 228 - 238 .
21. Knudsen N , Laurberg P , Rasmussen LB , Bu¨ low I, Perrild H , Ovesen L , Jørgensen T . Small differences in thyroid function may be important for body mass index and the occurrence of obesity in the population . J Clin Endocrinol Metab . 2005 ; 90 ( 7 ): 4019 - 4024 .
22. Altinova AE , To¨ r u¨ ner FB , Aktu¨ rk M, Bukan N , Cakir N , Ayvaz G , Arslan M. Adiponectin levels and cardiovascular risk factors in hypothyroidism and hyperthyroidism . Clin Endocrinol (Oxf) . 2006 ; 65 ( 4 ): 530 - 535 .
23. Asvold BO , Bjøro T , Nilsen TI , Vatten LJ . Tobacco smoking and thyroid function: a population-based study . Arch Intern Med . 2007 ; 167 ( 13 ): 1428 - 1432 .
24. Balhara YP , Deb KS . Impact of alcohol use on thyroid function . Indian J Endocrinol Metab . 2013 ; 17 ( 4 ): 580 - 587 .
25. Bergvall N , Cnattingius S. Familial (shared environmental and genetic) factors and the foetal origins of cardiovascular diseases and type 2 diabetes: a review of the literature . J Intern Med . 2008 ; 264 ( 3 ): 205 - 223 .
26. Park SB , Choi HC , Joo NS . The relation of thyroid function to components of the metabolic syndrome in Korean men and women . J Korean Med Sci . 2011 ; 26 ( 4 ): 540 - 545 .
27. Mehran L , Amouzegar A , Tohidi M , Moayedi M , Azizi F . Serum free thyroxine concentration is associated with metabolic syndrome in euthyroid subjects . Thyroid . 2014 ; 24 ( 11 ): 1566 - 1574 .
28. Gardun~ o-Garcia JdeJ , Alvirde-Garcia U , Lo´ pez-Carrasco G , Padilla Mendoza ME , Mehta R , Arellano-Campos O , Choza R , Sauque L , Garay-Sevilla ME , Malacara JM , Gomez-Perez FJ , Aguilar-Salinas CA . TSH and free thyroxine concentrations are associated with differing metabolic markers in euthyroid subjects . Eur J Endocrinol . 2010 ; 163 ( 2 ): 273 - 278 .
29. Roos A , Bakker SJ , Links TP , Gans RO , Wolffenbuttel BH . Thyroid function is associated with components of the metabolic syndrome in euthyroid subjects . J Clin Endocrinol Metab . 2007 ; 92 ( 2 ): 491 - 496 .
30. Luna-Vazquez F , Cruz-Lumbreras R , Rodrı´guez-Castel a´n J, Cervantes-Rodrı´guez M, Rodrı´guez- Antolı´n J , Arroyo-Helguera O , Castel a´n F, Martı´nez-Go´ mez M, Cuevas E. Association between the serum concentration of triiodothyronine with components of metabolic syndrome, cardiovascular risk, and diet in euthyroid post-menopausal women without and with metabolic syndrome . Springerplus . 2014 ; 3 : 266 .
31. Kim HJ , Bae JC , Park HK , Byun DW , Suh K , Yoo MH , Kim JH , Min YK , Kim SW , Chung JH . Triiodothyronine levels are independently associated with metabolic syndrome in euthyroid middle-aged subjects . Endocrinol Metab (Seoul) . 2016 ; 31 ( 2 ): 311 - 319 .
32. Lambadiari V , Mitrou P , Maratou E , Raptis AE , Tountas N , Raptis SA , Dimitriadis G . Thyroid hormones are positively associated with insulin resistance early in the development of type 2 diabetes . Endocrine . 2011 ; 39 ( 1 ): 28 - 32 .
33. Taneichi H , Sasai T , Ohara M , Honma H , Nagasawa K , Takahashi T , Ishii M , Fujiwara F , Yamashina M , Kajiwara T , Takabe N , Takahashi K , Satoh J . Higher serum free triiodothyronine levels within the normal range are associated with metabolic syndrome components in type 2 diabetic subjects with euthyroidism . Tohoku J Exp Med . 2011 ; 224 ( 3 ): 173 - 178 .
34. Boucai L , Hollowell JG , Surks MI . An approach for development of age-, gender-, and ethnicity-specific thyrotropin reference limits . Thyroid . 2011 ; 21 ( 1 ): 5 - 11 .
35. Mauvais-Jarvis F , Clegg DJ , Hevener AL . The role of estrogens in control of energy balance and glucose homeostasis . Endocr Rev . 2013 ; 34 ( 3 ): 309 - 338 .
36. Kim SR , Tull ES , Talbott EO , Vogt MT , Kuller LH . A hypothesis of synergism: the interrelationship of T3 and insulin to disturbances in metabolic homeostasis . Med Hypotheses . 2002 ; 59 ( 6 ): 660 - 666 .
37. Crunkhorn S , Patti ME . Links between thyroid hormone action, oxidative metabolism , and diabetes risk? Thyroid . 2008 ; 18 ( 2 ): 227 - 237 .
38. Dimitriadis G , Baker B , Marsh H , Mandarino L , Rizza R , Bergman R , Haymond M , Gerich J . Effect of thyroid hormone excess on action, secretion, and metabolism of insulin in humans . Am J Physiol . 1985 ; 248 ( 5 Pt 1 ): E593 - E601 .
39. Yavuz DG , Yu¨ ksel M, Deyneli O , Ozen Y , Aydin H , Akalin S . Association of serum paraoxonase activity with insulin sensitivity and oxidative stress in hyperthyroid and TSH-suppressed nodular goitre patients . Clin Endocrinol (Oxf) . 2004 ; 61 ( 4 ): 515 - 521 .
40. Lillioja S , Mott DM , Spraul M , Ferraro R , Foley JE , Ravussin E , Knowler WC , Bennett PH , Bogardus C . Insulin resistance and insulin secretory dysfunction as precursors of non-insulindependent diabetes mellitus . Prospective studies of Pima Indians. N Engl J Med . 1993 ; 329 ( 27 ): 1988 - 1992 .
41. Lin Y , Sun Z . Thyroid hormone potentiates insulin signaling and attenuates hyperglycemia and insulin resistance in a mouse model of type 2 diabetes . Br J Pharmacol . 2011 ; 162 ( 3 ): 597 - 610 .
42. Blanchet E , Bertrand C , Annicotte JS , Schlernitzauer A , Pessemesse L , Levin J , Fouret G , Feillet-Coudray C , Bonafos B , Fajas L , Cabello G , Wrutniak-Cabello C , Casas F. Mitochondrial T3 receptor p43 regulates insulin secretion and glucose homeostasis . FASEB J . 2012 ; 26 ( 1 ): 40 - 50 .
43. Ledda-Columbano GM , Perra A , Pibiri M , Molotzu F , Columbano A. Induction of pancreatic acinar cell proliferation by thyroid hormone . J Endocrinol . 2005 ; 185 ( 3 ): 393 - 399 .
44. Jansen MS , Cook GA , Song S , Park EA . Thyroid hormone regulates carnitine palmitoyltransferase Ialpha gene expression through elements in the promoter and first intron . J Biol Chem . 2000 ; 275 ( 45 ): 34989 - 34997 .
45. Bastemir M , Akin F , Emral R , Alkis E. Impact of insulin sensitivity in relationship with prolactin and thyroid stimulating hormone . Exp Clin Endocrinol Diabetes . 2007 ; 115 ( 4 ): 257 - 260 .
46. Imai Y , Toyoda N , Maeda A , Kadobayashi T , Fangzheng G , Nishikawa M , Iwasaka T. Type 2 iodothyronine deiodinase expression is upregulated by the protein kinase A-dependent pathway and is downregulated by the protein kinase C-dependent pathway in cultured human thyroid cells . Thyroid . 2001 ; 11 ( 10 ): 899 - 907 .
47. Borges M , Ingbar SH , Silva JE . Iodothyronine deiodinase activities in FRTL5 cells: predominance of type I 50-deiodinase . Endocrinology . 1990 ; 126 ( 6 ): 3059 - 3068 .
48. Ko ¨ hrle J. Thyroid hormone transporters in health and disease: advances in thyroid hormone deiodination . Best Pract Res Clin Endocrinol Metab . 2007 ; 21 ( 2 ): 173 - 191 .
49. Gereben B , Zavacki AM , Ribich S , Kim BW , Huang SA , Simonides WS , Zeo¨ ld A , Bianco AC . Cellular and molecular basis of deiodinase-regulated thyroid hormone signaling . Endocr Rev . 2008 ; 29 ( 7 ): 898 - 938 .
50. Strich D , Karavani G , Edri S , Gillis D. TSH enhancement of FT4 to FT3 conversion is age dependent . Eur J Endocrinol . 2016 ; 175 ( 1 ): 49 - 54 .