PDF icon PDF herunterladen
Horm Metab Res 2025; 57(02): 88-95
DOI: 10.1055/a-2508-7964
Original Article: Endocrine Care

Free Triiodothyronine Concentrations and Gestational Diabetes Mellitus: Unveiling the Correlation and Implications

Hongying Zha‡
1   Department of Endocrinology and Metabolism, The First Affiliated Hospital With Nanjing Medical University, Nanjing, China (Ringgold ID: RIN74734)
2   Department of Endocrinology, The Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi People’s Hospital, Wuxi Medical Center, Nanjing Medical University, Wuxi, China
,
Shasha Li‡
1   Department of Endocrinology and Metabolism, The First Affiliated Hospital With Nanjing Medical University, Nanjing, China (Ringgold ID: RIN74734)
,
Lu Sun
3   Department of Endocrinology and Metabolism, The Affiliated People’s Hospital of Jiangsu University, Zhenjiang, China (Ringgold ID: RIN196541)
,
Lin Yu
2   Department of Endocrinology, The Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi People’s Hospital, Wuxi Medical Center, Nanjing Medical University, Wuxi, China
,
Qingxin Yuan 
4   Department of Endocrinology and Metabolism, Jiangsu Province Hospital and Nanjing Medical University First Affiliated Hospital, Nanjing, China (Ringgold ID: RIN74734)
› Institutsangaben
› Weitere Informationen
Auch verfügbar aufeRef
 
 als PDF herunterladen Lizenzen und Reprints

Abstract

Thyroid disease and gestational diabetes mellitus (GDM) are frequent complications during pregnancy. We observed the relationship between thyroid indicators and blood glucose to analyze whether thyroid function is associated with the development of GDM. We enrolled a total of 575 pregnant women diagnosed with GDM and 573 pregnant women without GDM. The correlation between thyroid indicators and blood glucose levels was established through correlation analysis. In addition, stratified analysis and restricted cubic spline curves were employed to describe the association between thyroid indicators and the incidence of GDM. We found no significant difference in urine iodine levels between the GDM and non-GDM groups throughout the second trimester. The levels of free triiodothyronine (FT3) and both fasting blood glucose and post-load blood glucose showed a robust positive connection. Thyroid-stimulating hormone (TSH) and free thyroxine (FT4), on the other hand, showed a weakly positive connection with these glucose values. A nonlinear correlation between FT3 and the risk of GDM was also found (pNonlinear=0.0007, p<0.0001). Particularly, those in the top quartile of FT3 had a 6.99-fold greater risk than those in the lowest. Notably, FT3 levels below 4.04 pmol/l were linked to a decreased chance of developing GDM, but levels over 4.04 pmol/l were linked to a greater risk. Our study successfully established the correlation between thyroid indicators and the risk of GDM. Notably, we discovered a non-linear association between FT3 levels and GDM. The study suggests that ensuring optimal thyroid function during pregnancy may decrease the likelihood of developing GDM.


 

Keywords

gestational diabetes mellitus - thyroid function - free triiodothyronine

Introduction

The global rise in obesity and diabetes has resulted in a higher incidence of gestational diabetes mellitus (GDM). The primary underlying mechanisms of GDM involve insulin resistance and dysfunction of b cells [1]. The prevalence of GDM varies significantly across different regions, ranging from 1% to 30%. In China, the reported incidence of GDM is approximately 14.8% [2].

Thyroid hormone plays a critical role in ensuring the well-being of both the mother and the developing fetus during pregnancy. Monitoring thyroid hormone levels is a standard part of prenatal testing. Thyroid dysfunction is relatively common among pregnant women [3]. Notably, both thyroid dysfunction and GDM are common complications experienced by pregnant women. Research has demonstrated a significant impact of thyroid hormones on glucose metabolism and homeostasis, highlighting the crucial role of thyroid dysfunction in the development of GDM [4] [5] [6] [7]. However, some studies have reported no significant relationship between thyroid hormone and GDM [8].

In recent years, the significance of iodine during pregnancy has garnered increased attention. Goiter and hypothyroidism have been reported to be more common in regions with iodine deficiency. It is crucial to note that an excessive dietary iodine consumption may also have an effect on thyroid function. Iodine influences insulin homeostasis both directly – by modulating thyroid hormone levels, and indirectly – by altering oxidative stress, which in turn affects glucose metabolism and insulin sensitivity. The evidence, however, is still equivocal regarding the link between GDM and iodine status. While some studies have associated iodine deficiency with a higher risk of GDM [9], others have found no conclusive evidence of this association [10].

Our study aimed to investigate the correlation between thyroid indicators during pregnancy and blood glucose levels, as well as the risk of GDM. Additionally, we explored the impact of iodine nutrition status during pregnancy on these indicators of thyroid function. The findings of our study will have potential implications in guiding clinical practice and decision-making.


 

Subjects and Methods

Participants

We conducted a retrospective study at the First Affiliated Hospital of Nanjing Medical University from 2017 to 2023, focusing on GDM. The study comprised 575 pregnant women diagnosed with GDM and 573 pregnant women with normal glucose levels. Exclusion criteria for the study population were as follows: 1). Pregnant women with a history of diabetes mellitus or abnormal glucose tolerance prior to pregnancy. 2). Pregnant women with pre-existing hyperthyroidism or hypothyroidism detected before pregnancy and receiving medication. 3). Pregnant women who underwent thyroid surgery for conditions such as thyroid cancer, thyroid nodules, or hyperthyroidism. 4). Pregnant women with a history of tumors. 5). Pregnant women with systemic immune diseases such as systemic lupus erythematosus. 6). Pregnant women with severe liver and kidney impairment. All participants received comprehensive care from regular prenatal visits to delivery. At approximately 20–28 weeks of gestation, they underwent an oral glucose tolerance test (OGTT), and their thyroid function and urine iodine levels were assessed. If GDM was diagnosed, the pregnant women were referred to dietary modifications and regular monitoring. Insulin treatment was initiated if their blood glucose levels did not reach the target range.


 

Laboratory measurements

Thyroid-stimulating hormone (TSH), free triiodothyronine (FT3), free thyroxine (FT4), and thyroid peroxidase antibody (TPOAb) levels were measured using chemiluminescence immunoassay (Roche Cobas 6000, ZB/GEM1815-2010, Mannheim, Switzerland). Urine iodine levels were determined using arsenic-cerium-catalyzed spectrophotometry.


 

Diagnostic criteria

According to the International Association of Diabetes and Pregnancy Study Group’s (IADPSG) criteria, GDM is diagnosed if the fasting blood glucose level exceeds 5.1 mmol/l, or the 1-hour postprandial blood glucose level exceeds 10.0 mmol/l, or the 2-hour postprandial blood glucose level exceeds 8.5 mmol/l following the ingestion of 75 g of glucose in the fasting state. As per the Guidelines for the Diagnosis and Management of Thyroid Diseases in Pregnancy and Postpartum issued by the American Thyroid Association (ATA) in 2017 and the Guidelines for Diagnosis and Treatment of Thyroid Diseases in Pregnancy and Postpartum in China, the reference intervals for thyroid function during pregnancy are as follows: TSH levels ranging from 0.45 to 4.32 mU/l, FT4 levels ranging from 9.77 to 18.89 pmol/l during the second trimester, and TSH levels ranging from 0.30 to 4.98 mU/l, FT4 levels ranging from 9.04 to 15.22 pmol/l during the third trimester. A TPOAb titer exceeding 34 IU/ml is considered positive for the presence of thyroid peroxidase antibodies.


 

Definition of some pregnancy-related disorders

Preterm birth was defined as the delivery before 37 weeks of gestation, excluding cases of iatrogenic preterm birth caused by conditions such as preeclampsia, placenta previa, fetal growth restriction, and other factors. Low birth weight was defined as the birth weight of a live infant less than 2500 g. Fetal macrosomia was defined as a birth weight greater than 4000 g, regardless of gestational age. Preeclampsia was defined as the development of hypertension and proteinuria after 20 weeks of gestation in women who were previously normotensive. For birth weight classification, a birth weight greater than the ninetieth percentile was considered large for gestational age (LGA), while a birth weight below the ninetieth percentile was classified as small for gestational age (SGA). Neonatal intensive care admissions encompassed adverse outcomes such as hyperbilirubinemia, preterm birth, respiratory distress, and neonatal hypoglycemia.


 

Statistical analysis

Continuous variables are presented as medians (interquartile) according to the normality. Chi-square tests were used to compare the differences between the categorical variables of these two groups. Pearson correlation analysis was used to describe the relationship between thyroid function and blood glucose, and a further correlation graph was made whether GDM was negative or positive for TPOAb. The three indicators (TSH, FT3 and FT4) were divided into four groups according to the quartile. The relation between thyroid function and GDM was investigated by Logitic model analysis, and BMI, family history and age were adjusted. Cubic splines were also used to further characterize the relationship between thyroid function and GDM. All analyses were performed in R Studio version 4.1.3 and a two-sided p-value<0.05 was considered to indicate statistical significance.


 

 

Results

The study included 1148 pregnant women, comprising 575 with GDM and 573 without ([Table 1]). Significant differences were observed between the two groups in terms of age, family history of type 2 diabetes mellitus (T2DM), body mass index (BMI), prevalence of polycystic ovarian syndrome (PCOS), systolic and diastolic blood pressure, and thyroid function (p<0.05). There were no statistically significant differences between the two groups in the prevalence of thyroid disorders or the proportion of the positive for TPOAb. We also analyzed the urine iodine levels between the two groups to assess the potential influence of iodine status on thyroid function, but we were unable to find a statistically significant difference.

Table 1 Characteristics of subjects.

Level

Overall

Non-GDM

GDM

p

n

1148

573

575

Age (median [IQR])

31.00 [28.00, 34.00]

30.00 [28.00, 33.00]

32.00 [29.00, 35.00]

<0.001

Parity (%)

0

791 (69.0)

409 (71.4)

382 (66.6)

0.059

1

336 (29.3)

159 (27.7)

177 (30.8)

2

18 (1.6)

5 (0.9)

13 (2.3)

3

2 (0.2)

0 (0.0)

2 (0.3)

Family (%)

no

1050 (91.5)

552 (96.3)

498 (86.6)

<0.001

yes

98 (8.5)

21 (3.7)

77 (13.4)

PCOS (%)

no

1128 (98.3)

569 (99.3)

559 (97.2)

0.013

yes

20 (1.7)

4 (0.7)

16 (2.8)

SBP (median [IQR])

119.00 [112.00, 126.00]

118.00 [112.00, 125.00]

120.00 [113.00, 127.00]

0.024

DBP (median [IQR])

75.00 [69.00, 80.00]

75.00 [69.00, 80.00]

76.00 [70.00, 80.00]

0.004

Height (median [IQR])

162.00 [158.00, 165.00]

162.00 [159.00, 165.00]

161.00 [158.00, 165.00]

0.005

BMI (median [IQR])

27.05 [25.14, 29.58]

26.85 [24.98, 29.04]

27.29 [25.41, 30.07]

0.002

FBG (median [IQR])

4.59 [4.29, 4.98]

4.43 [4.20, 4.64]

4.90 [4.49, 5.33]

<0.001

PBG1h (median [IQR])

8.65 [7.31, 10.11]

7.61 [6.63, 8.41]

10.07 [9.04, 10.73]

<0.001

PBG2h (median [IQR])

7.54 [6.40, 8.83]

6.54 [5.86, 7.31]

8.81 [7.97, 9.54]

<0.001

TSH (median [IQR])

2.12 [1.42, 3.10]

2.30 [1.51, 3.26]

2.01 [1.34, 2.97]

0.005

FT3 (median [IQR])

4.04 [3.63, 4.45]

3.87 [3.51, 4.23]

4.25 [3.82, 4.70]

<0.001

FT4 (median [IQR])

12.53 [11.26, 14.08]

12.71 [11.50, 14.37]

12.19 [11.02, 13.87]

<0.001

TPOAB (%)

–

914 (89.9)

467 (89.0)

447 (90.9)

0.368

+

103 (10.1)

58 (11.0)

45 (9.1)

UIC (%)

1

458 (46.0)

238 (44.5)

220 (47.8)

0.145

2

260 (26.1)

151 (28.2)

109 (23.7)

3

274 (27.5)

143 (26.7)

131 (28.5)

4

3 (0.3)

3 (0.6)

0 (0.0)

Thyroid function (%)

euthyroid

1007 (91.2)

504 (90.8)

503 (91.6)

0.491

Isolated hypothyroxinemia

22 (2.0)

10 (1.8)

12 (2.2)

Overt hyperthyroidism

16 (1.4)

11 (2.0)

5 (0.9)

Subclinical hyperthyroidism

13 (1.2)

5 (0.9)

8 (1.5)

Subclinical hypothyroidism

46 (4.2)

25 (4.5)

21 (3.8)

Several maternal pregnancy outcomes differed significantly between the GDM and non-GDM groups ([Table 2]). GDM group had a higher rate of threatened preterm birth (p=0.011), preeclampsia (p=0.001), and cesarean section (p<0.001). Infants born to mothers with GDM were more likely to be classified as large for gestational age (LAG) (p=0.014) and admitted to the neonatal critical care unit (p<0.001).

Table 2 Pregnancy outcomes and pregnancy-related complications of GDM and non-GDM groups.

Non-GDM

GDM

OR

p.ratio

p.overall

n=573

n=575

Threatened preterm birth

0.001

No

531 (92.7%)

499 (86.8%)

Ref.

Ref.

Yes

42 (7.33%)

76 (13.2%)

1.92 [1.30; 2.88]

0.001

Preterm

0.074

No

544 (94.9%)

530 (92.2%)

Ref.

Ref.

Yes

29 (5.06%)

45 (7.83%)

1.59 [0.99; 2.60]

0.058

Gestational hypertension

0.079

No

562 (98.1%)

553 (96.2%)

Ref.

Ref.

Yes

11 (1.92%)

22 (3.83%)

2.02 [0.98; 4.39]

0.055

Preeclampsia

0.011

No

559 (97.6%)

543 (94.4%)

Ref.

Ref.

Yes

14 (2.44%)

32 (5.57%)

2.34 [1.25; 4.58]

0.007

Placental abruption

0.725

No

569 (99.3%)

572 (99.5%)

Ref.

Ref.

Yes

4 (0.70%)

3 (0.52%)

0.76 [0.14; 3.63]

0.723

Cesarean section

<0.001

No

390 (68.1%)

308 (53.6%)

Ref.

Ref.

Yes

183 (31.9%)

267 (46.4%)

1.85 [1.45; 2.35]

<0.001

Anemia of pregnancy

0.255

No

549 (95.8%)

559 (97.2%)

Ref.

Ref.

Yes

24 (4.19%)

16 (2.78%)

0.66 [0.34; 1.25]

0.199

Postpartum hemorrhage

0.536

No

528 (92.1%)

523 (91.0%)

Ref.

Ref.

Yes

45 (7.85%)

52 (9.04%)

1.17 [0.77; 1.78]

0.471

LGA

0.014

No

444 (77.5%)

408 (71.0%)

Ref.

Ref.

Yes

129 (22.5%)

167 (29.0%)

1.41 [1.08; 1.84]

0.012

SGA

0.598

No

564 (98.4%)

569 (99.0%)

Ref.

Ref.

Yes

9 (1.57%)

6 (1.04%)

0.67 [0.22; 1.89]

0.448

Low birth weight infant

0.218

No

556 (97.0%)

549 (95.5%)

Ref.

Ref.

Yes

17 (2.97%)

26 (4.52%)

1.54 [0.83; 2.94]

0.170

Macrosomia

0.793

No

512 (89.4%)

510 (88.7%)

Ref.

Ref.

Yes

61 (10.6%)

65 (11.3%)

1.07 [0.74; 1.55]

0.722

Gender

0.213

No

259 (45.2%)

282 (49.0%)

Ref.

Ref.

Yes

314 (54.8%)

293 (51.0%)

0.86 [0.68; 1.08]

0.193

Umbilical cord around neck

0.101

No

408 (71.2%)

435 (75.7%)

Ref.

Ref.

Yes

165 (28.8%)

140 (24.3%)

0.80 [0.61; 1.04]

0.089

Admission to newborn intensive care

<0.001

No

535 (93.4%)

491 (85.4%)

Ref.

Ref.

Yes

38 (6.63%)

84 (14.6%)

2.40 [1.62; 3.63]

<0.001

Both [Fig. 1] and [Fig. 2] show the relationship between thyroid function and blood glucose. A negative correlation was observed between TSH and FT4 levels and blood glucose, implying that higher TSH and FT4 levels were associated with lower blood glucose levels. In contrast, we discovered a positive correlation between FT3 levels and blood glucose, indicating that higher FT3 levels were associated with higher blood glucose levels. We observed a stronger association between blood glucose and thyroid function in the TPOAb-negative group (n=1003). The correlation coefficient between FT3 levels and the presence of GDM was 0.111 when the GDM group (n=1148) was considered.

Zoom
Fig. 1 Scatter plots of blood glucose and thyroid function grouped by TPOAb negative and positive. “0” means negative for TPOAb and “1” means positive for TPOAb.
Zoom
Fig. 2 Scatter plots of blood glucose and thyroid function grouped according to GDM or not. “0” means non-GDM and “1” means GDM.

As demonstrated in [Fig. 3], higher concentrations of FT4 and TSH were associated with a lower risk of GDM. In contrast, higher FT3 concentrations were positively associated with an increased risk of GDM.

Zoom
Fig. 3 Associations between quartile of thyroid indicators and the risk of GDM.

After adjusting for age, BMI, SBP, DBP, and family history, the incidence of GDM was found to be 6.99 times higher in individuals with the highest FT3 concentration compared to those with the lowest [Q4 vs. Q1, odds ratio (OR)=6.99, 95% confidence interval (CI) (4.59–10.64)]. In addition, the incidence of GDM increased between the third and first quartiles [Q3 vs. Q1, OR=2.47, 95% CI (1.65–3.67)]. To describe the relationship between thyroid indicators and GDM, we used a cubic spline curve ([Fig. 4]). The cubic spline curve revealed a significant non-linear association, with an OR of 1 for FT3 at a concentration of 4.04 pmol/l (pNonlinear=0.0007, p<0.0001). In contrast, no statistically significant differences were observed between the FT4 (pNonlinear=0.4726, p=0.1387) and TSH (pNonlinear=0.3142, p=0.0038) curves, indicating a probably linear correlation between these variables and the incidence of GDM.

Zoom
Fig. 4 RCS between quartile of thyroid indicators and the risk of GDM.

 

Discussion

In our study, we observed differences in thyroid indicators between the GDM group and the non-GDM group. Specifically, we observed that FT4 and TSH levels were negatively correlated with GDM risk, whereas FT3 levels were positively correlated with GDM risk. In addition, we investigated the relationship between FT3 levels and the incidence of GDM using a cubic spline analysis. Our findings indicated that the incidence of GDM may increase nonlinearly as FT3 levels increase.

By promoting glucose and fatty acid oxidation, thyroid hormones play a vital role in the development of diabetes. In addition, thyroid hormones affect glucose metabolism by accelerating hepatic glycogen breakdown, facilitating intestinal glucose absorption, and enhancing catecholamine sensitivity. These mechanisms demonstrate the potential effect of thyroid hormones on glucose homeostasis and the pathogenesis of glucose disorders [11]. As a bioactive hormone, FT3 is more essential for stimulating endogenous glucose production and insulin secretion. This exactly corresponds with the results of our correlation analysis. FT3 was positively correlated with the incidence of GDM and some previous studies have reported similar findings [4] [12] [13] [14]. In fact, we found that FT3 had a stronger correlation with fasting blood glucose and postprandial blood glucose than TSH and FT4 and further observed a non-linear increase in the risk of GDM with rising FT3 levels. Two previous prospective studies [15] [16] have reported no significant difference in FT3 levels between women with GDM and those without. However, it is important to note that the sample size of GDM cases in both studies was relatively small, which may have impacted the results. Thus, increased FT3 levels are strongly linked to an increased risk of developing GDM. Compared to the lowest quartile, the incidence of GDM was found to be 6.99 times higher in the highest quartile of FT3 concentration. In addition, our cubic spline analysis revealed a progressive rise in GDM risk when FT3 concentrations exceeded 4.04 pmol/l. These discoveries suggest that variations in FT3 levels have a greater effect on blood glucose during pregnancy. Nonetheless, further research is required to elucidate the pathophysiological mechanisms underlying this association.

We observed decreased levels of FT4 in patients with GDM. Consistent with previous studies in both pregnant and non-pregnant populations [17] [18] [19], our findings imply a negative association between FT4 concentrations and BMI. The correlation between FT4 and GDM weakened after adjusting for covariates such as maternal BMI, age, and family history. Only the highest quartile of FT4 concentration demonstrated statistical significance when compared to the lowest quartile, indicating that FT4 may play a role in mediating insulin resistance via its effect on body weight [20]. In turn, this may contribute to the abnormal glucose levels associated with GDM.

We conducted a stratification analysis based on TSH quartiles, despite the fact that the fitted curve did not attain statistical significance. When TSH levels were below 0.59 mIU/l or above 2.11 mIU/l, there was a protective trend against GDM, according to our findings. However, the incidence of GDM increased when TSH concentrations decreased between 0.59 and 2.11 mIU/l. A study [21] conducted in Tianjin, China, reported that TSH concentrations below 3.2 mIU/l were associated with an increased incidence of GDM, whereas those above 3.2 mIU/l exhibited no significant correlation. Similarly, a study [22] conducted in Shanghai, China, found that the risk of GDM was lower when TSH levels were above 1.96 mIU/l. Nevertheless, it is worth noting that several studies have not found a significant association between TSH and GDM [4] [23] [24]. These discrepancies in findings demonstrate the need for additional research to clarify the relationship between TSH and GDM. TPOAb, an antibody that inhibits thyroid peroxidase activity and causes destruction of thyroid tissue, can also lead to antibody-dependent cell-mediated cytotoxic effects [25]. In our study, we did not observe a statistically significant difference in the rate of TPOAb positivity between pregnant women with GDM (9.1%) and those without GDM (11%) in our study population. Interestingly, previous studies have indicated that the combination of subclinical hypothyroidism and positive TPOAb antibodies may increase the prevalence of GDM [16] [26] [27] [28]. In our future research, we can further explore this aspect to gain more insights into the relationship between TPOAb, thyroid function, and the development of GDM.

In our previous study, we investigated the impact of iodine levels during pregnancy on thyroid function and pregnancy outcomes in normally pregnant women [29]. According to WHO standards, 46% of the study population showed an iodine deficiency. We did not observe a statistically significant difference in urine iodine concentrations between the GDM and non-GDM groups in the present study. Despite the fact that iodine deficiency may affect thyroid function and pregnancy outcomes, its association with GDM needs further study. There was no statistically significant difference between the GDM and non-GDM groups with regard to the prevalence of thyroid disease and TPOAb status. It indicates that the impact of these factors on pregnancy outcomes may have been minimized. Nonetheless, it is necessary to recognize the limitations of our investigation. If obtain TgAb data, we will have a fuller assessment of thyroid autoimmune status. Moreover, our study was retrospective, and the data were obtained from the hospital’s electronic medical record system. To attain a definitive conclusion, additional prospective studies must be designed and carried out.


 

Conclusions

Our study successfully established a correlation between indicators of thyroid function and the risk of GDM. Specifically, we discovered a non-linear relationship between FT3 levels and GDM, with the highest quartile exhibiting a 6.99-fold increase in GDM incidence compared to the lowest quartile. The study suggests that ensuring optimal thyroid function during pregnancy may decrease the likelihood of GDM.


 

 

Conflict of Interest

The authors declare that they have no conflict of interest.


Correspondence

Dr. Qingxin Yuan
Department of Endocrinology and Metabolism, Jiangsu Province Hospital and Nanjing Medical University First Affiliated Hospital
210029 Nanjing
China   

Publikationsverlauf

Eingereicht: 15. September 2023

Angenommen nach Revision: 07. Dezember 2024

Artikel online veröffentlicht:
10. Februar 2025

© 2025. Thieme. All rights reserved.

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany


  Lizenzen und Reprints
Zoom
Fig. 1 Scatter plots of blood glucose and thyroid function grouped by TPOAb negative and positive. “0” means negative for TPOAb and “1” means positive for TPOAb.
Zoom
Fig. 2 Scatter plots of blood glucose and thyroid function grouped according to GDM or not. “0” means non-GDM and “1” means GDM.
Zoom
Fig. 3 Associations between quartile of thyroid indicators and the risk of GDM.
Zoom
Fig. 4 RCS between quartile of thyroid indicators and the risk of GDM.