Abstract
Thyroid disorders are the most common endocrine disorders affecting women commencing pregnancy. Thyroid hormone metabolism is strongly influenced by selenium status; however, the relationship between serum selenium concentrations and thyroid hormones in euthyroid pregnant women is unknown. This study investigated the relationship between maternal selenium and thyroid hormone status during pregnancy by utilizing data from a retrospective, cross-sectional study (Maternal Outcomes and Nutrition Tool or MONT study) with cohorts from two tertiary care hospitals in South East Queensland, Australia. Pregnant women (n = 206) were recruited at 26–30 weeks gestation and serum selenium concentrations were assessed using inductively coupled plasma mass spectrometry. Thyroid function parameters were measured in serum samples from women with the lowest serum selenium concentrations (51.2 ± 1.2 µg/L), women with mean concentrations representative of the entire cohort (78.8 ± 0.4 µg/L) and women with optimal serum selenium concentrations (106.9 ± 2.3 µg/L). Women with low serum selenium concentrations demonstrated reduced fT3 levels (P < 0.05) and increased TPOAb (P < 0.01). Serum selenium was positively correlated with fT3 (P < 0.05) and negatively correlated with TPOAb (P < 0.001). Serum fT4 and thyroid-stimulating hormone (TSH) were not different between all groups, though the fT4/TSH ratio was increased in the low selenium cohort (P < 0.05). Incidence of pregnancy disorders, most notably gestational diabetes mellitus, was increased within the low serum selenium cohort (P < 0.01). These results suggest selenium status in pregnant women of South East Queensland may not be adequate, with possible implications for atypical thyroid function and undesirable pregnancy outcomes.
Introduction
Thyroid dysfunction is the most common endocrine condition affecting women of reproductive age (Forehan 2012). It is reported to affect one in ten women of childbearing age and approximately 5% of all pregnant women with up to 18% of women presenting as thyroid antibody positive in their first trimester of pregnancy (Forehan 2012). Thyroid dysfunction has been shown to increase the risk of adverse maternal and fetal outcomes (Nazarpour et al. 2015). The most abundant thyroid hormone in circulation is tetraiodothyronine (T4), which is subsequently deiodinated to form the most bioactive thyroid hormone, triiodothyronine (T3). T3 is transported into the nucleus and regulates thyroid responsive gene transcription (Yen 2001); therefore, appropriate production of T4 as well as the conversion from T4 to T3 is essential to maintain maternal physiology throughout pregnancy.
It is estimated that, in populations with relative iodine-sufficiency, overt hypothyroidism affects 0.3–0.5% of pregnancies, with a further 3–5% of pregnancies affected by subclinical hypothyroidism (De Groot et al. 2012, Dulek et al. 2019). Adverse pregnancy outcomes such as hypertensive disorders of pregnancy (HDP), preterm birth (PTB), gestational diabetes mellitus (GDM), preeclampsia, placental abruption and recurrent miscarriage have been associated with hypothyroidism (Smallridge & Ladenson 2001). The two major causes of hypothyroidism are inadequate dietary iodine intake and the autoimmune condition known as Hashimoto’s autoimmune thyroiditis (Reid et al. 2013). An additional factor known to contribute to thyroid dysfunction is inappropriate concentrations of selenium. Selenium deficiency is common during pregnancy and is associated with many of the conditions commonly associated with thyroid dysfunction (Hofstee et al. 2018).
Selenium is required for the functioning of approximately 25 essential selenoproteins that are predominantly involved in regulating oxidative stress and thyroid hormone metabolism. Three selenium-dependent iodothyronine deiodinases (DIO 1, 2 and 3) control activation/deactivation of thyroid hormones and selenium deficiency can impair DIO activity and reduce thyroid hormone concentrations (Duntas & Benvenga 2015). The process of thyroid hormone synthesis produces reactive oxygen species (ROS). Selenium deficiency can reduce the activity of selenodependent antioxidant enzymes such as glutathione peroxidase, and thioredoxin reductase leading to ROS-induced thyrocyte damage and the subsequent production of thyroid antibodies, such as thyroid peroxidase antibodies (TPOAb) (Ventura et al. 2017).
During pregnancy and lactation there is increased demand and requirement for selenium. Currently, the recommended dietary intake (RDI) of selenium during pregnancy is 65 µg/day, only marginally higher than the 60 µg/day recommended for non-pregnant women (NHMRC 2017). While this small increase in the RDI during pregnancy is expected to be sufficient for most women, it is unlikely to maximize the activity of individual selenoproteins. For optimal plasma and cellular activity of key selenoproteins, such as selenoprotein P, an intake of ~105 µg/day of selenium is required (Hurst et al. 2010, Stoffaneller & Morse 2015).
There is cause for concern over the number of people who are predicted to be selenium deficient, with estimates from the early 2000s suggesting that 500 million to 1 billion people may be selenium deficient globally (Combs 2001, Hofstee et al. 2018, Shreenath & Dooley 2018). We have recently published findings from the Gold Coast, Australia, demonstrating serum selenium concentrations were, on average, 74.1 µg/L (McAlpine et al. 2019, 2020). This average plasma selenium concentration is similar to a study by Santos and colleagues from Spain which had maternal selenium serum concentrations of 69.1 µg/L (Santos et al. 2017), but higher than a study from Poland which reported 62.54 µg/L, a concentration lower than the current study (Lewandowska et al. 2019). Importantly, all of these studies demonstrate that blood selenium concentrations are well below the optimal range of 90–105 µg/L which is required to ensure adequate expression of Selenoprotein P (Hurst et al. 2010, Stoffaneller & Morse 2015). Data presented in a World Health Organisation report has demonstrated impaired thyroid activity occurs when plasma levels of selenium fall to approximately 71 µg/L, which is similar to the mean concentration detected in women from our cohort (World Health Organisation 2004).
The effects of different serum selenium concentrations on thyroid hormone status during pregnancy has not been well investigated. Recent research has advocated for the supplementation of selenium during pregnancy (Mantovani et al. 2019), however, it is not yet clear whether thyroid abnormalities during pregnancy are related to selenium deficiency, and if so, how (Hubalewska-Dydejczyk et al. 2019). Furthermore, many studies investigating the effect of selenium supplementation on thyroid function in pregnancy do not measure selenium concentrations prior to, during, or after supplementation, with primary outcomes measured through thyroid antibody levels (Ventura et al. 2017). This study aimed to delineate the relationship between maternal selenium concentrations during pregnancy and thyroid hormone status in a sample of pregnant women with no known thyroid disorders in South East Queensland, Australia.
Materials and methods
Study design and population
Pregnant women were recruited for the Maternal Outcomes and Nutrition Tool (MONT) study in South East Queensland from the Gold Coast University Hospital (GCUH) and Royal Brisbane and Women’s Hospital (RBWH). Women between 26- and 30-weeks’ gestation presenting for an Oral Glucose Tolerance test at GCUH or RBWH were recruited and informed consent was obtained prior to biological sampling,
At recruitment, fasting blood samples were collected by venepuncture into serum separator tubes (SST; BD Vacutainer SST II Advance; Cat. no. 367953). Samples were transported and processed at the School of Medical Science, Griffith University Gold Coast Campus. Once blood samples were collected, the SST was inverted 5–10 times, rested and allowed to clot for a further 30 min. SSTs were centrifuged at 4°C at 1500 g for 10 min. Serum was aliquoted into 1.5 mL Eppendorf tubes and stored at −80°C for subsequent analysis.
Inductively coupled plasma mass spectrometry
Elemental analysis of serum samples was performed using inductively coupled plasma mass spectrometry (ICP-MS) as recently described (McAlpine et al. 2019). This allowed us to measure serum concentrations of 29 different elements including selenium, iodide, manganese, copper, zinc, calcium, magnesium and iron (McKeating et al. 2020). An Agilent 7900 ICP-MS (Agilent Technologies) was used to measure samples with quality control standards between 10 and 100 µg/L analyzed every 12 samples. All samples were prepared in a 1:10 dilution using our standard protocol (De Blas Bravo et al. 2007, McAlpine et al. 2019). Scandium, Yttrium, Indium, and Terbium were added to all samples as an internal standard to account for instrument drift (final concentration of 10 µg/L). The limit of detection was set at three times the standard deviation of twenty blank replicates. The limit of quantification was set at ten times the standard deviation of the twenty blank replicates.
Inclusion criteria
A total of 206 blood samples were collected for the study and serum elemental analysis was performed using ICP-MS on all samples. Samples were sorted according to serum selenium concentrations and three groups were determined according to the concentration of serum selenium. The optimal selenium group (106.9 ± 2.3 µg/L, n = 21) consists of serum samples that had selenium concentrations detected over 90 µg/L, which is the threshold for maximal protein activity for most selenoproteins (Stoffaneller & Morse 2015). This was matched to a low selenium group (51.2 ± 1.2 µg/L, n = 21) consisting of the serum samples with the lowest concentration of selenium detected in the entire cohort. The mean selenium group (78.8 ± 0.4 µg/L, n = 21) consists of serum samples with the mean concentrations of selenium detected in the entire cohort.
Exclusion criteria
The final dataset comprised of 63 women. Serum concentrations of fT4, fT3, TSH, and TPOAb were determined in the samples collected from these women. One woman had TSH, fT3 and fT4 indicative of subclinical hypothyroidism. As it was important to investigate the parameters of interest within a group of individuals that would be clinically assessed as being euthyroid, her sample was excluded from subsequent analysis. Women with obstetric complications including hypertensive disorders of pregnancy, intrauterine growth restriction, low birth weight, small for gestational age, preterm birth, placenta previa and gestational diabetes were not excluded from the study. Clinical definitions for the aforementioned pregnancy complications are provided by the Australian Clinical Practice Guidelines: Pregnancy care (NHMRC 2018).
Thyroid function testing
Serum levels of fT4, fT3, TSH and TPOAb were determined by ELISA using commercial kits (Demeditec, Kiel, Germany). All the samples from the individual patients were analyzed in duplicate over two plates for each assay.
Free thyroxine
Quantitative in vitro levels for fT4 in serum were determined using a commercial ELISA kit (Demeditec Cat. no. DE3775) as per the manufacturer’s instructions. Reference values for the fT4 analysis were between 0.22–8.0 ng/dL. The intra-assay coefficient of variation was 3.8% and the inter-assay coefficient of variation was 3.1%.
Free triiodothyronine
An fT3 ELISA kit was utilized to determine quantitative in vitro serum fT3 levels (Demeditec Cat. no. DE3801) as per the manufacturer’s protocol. The range for the fT3 assay was between 0.38–20 pg/mL. The intra-assay coefficient of variation was 3.8% and the inter-assay coefficient of variation was 4.9%.
Thyroid-stimulating hormone
Quantitative in vitro levels for TSH in serum were determined using a commercial ELISA kit (Demeditec Cat. no. DE4171) as per the manufacturer’s protocol. Reference values for TSH were 0.06–15 mIU/L. The intra-assay coefficient of variation was 3.9% and the inter-assay coefficient of variation was 2.4%.
Thyroid peroxidase antibodies
Quantitative serum measurements of IgG class autoantibodies against TPO were performed with a commercial ELISA kit (Demeditec Cat. no. DE7580) as per the manufacturer’s protocol. The range for the TPOAb assay was between 0–3000 IU/mL, where; < 50 IU/L was considered negative, 50–75 IU/mL borderline and > 75 IU/L positive. The intra-assay coefficient of variation was 15.2% and the inter-assay coefficient of variation was 1.1%.
Statistical analysis
All data are presented as mean ± s.e.m. One-way ANOVA was used to compare the low, mean and optimal selenium pregnancy groups. Normality of distribution testing was conducted using Kolmogorov–Smirnov testing. If data were not normally distributed, Kruskal–Wallis nonparametric testing was used, with Dunn’s testing utilized for multiple comparisons. Correlations between serum selenium and thyroid status were performed using Pearson’s tests. When data were not normally distributed, Spearman rank correlation analysis was utilized. Analysis of pregnancy complication frequency between selenium groups was analyzed using chi-square analyses (χ2 (df), P value).
Ethical considerations
This research was approved by both the Griffith University Human Ethics Committees (HREC 2016/423) and Health Service Human Ethics (HREC 16/QGC/70). Written consent for the release of perinatal data from medical records was provided from all women included in the cohort.
Results
Clinical characteristics of groups
From the entire cohort of 206 women, blood samples were taken at 26–30 weeks. Details for fetal sex at birth were available for 172 women and a complete history of key pregnancy outcomes were available for 182 women (Table 1). The mean selenium concentration from all 206 women was 73.43 ± 1.15 µg/L (Table 1). Further analysis was performed on 63 matched samples (Table 2), with no difference between maternal age, BMI, gender ratio, birth weight, gestational length, parity, gravidity, ethnicity, alcohol intake, smoker status, or vitamin intake between the low, mean and optimal serum selenium cohorts. Of the total cohort of selected samples, 28 women had female offspring and 35 had male offspring. Furthermore, four pregnancies within the low serum selenium group were preterm (<37 weeks gestation), with a further three preterm pregnancies in the mean selenium group and one in the optimal serum selenium group.
Maternal outcome data for the MONT cohort.
| Average cohort data (n = 206) | ||
| Baby gender (M/F) | 88/84 | |
| Ave birth weight (g) | 3212 ± 51.60 | |
| Gestational length | 38 ± 0.18 | |
| Pregnancy complications | Negative | Positive |
| HDP | 166 | 16 |
| IUGR/LBW/SGA | 145 | 37 |
| PTB | 151 | 31 |
| PP | 166 | 16 |
| GDM | 160 | 22 |
| At least one pregnancy complication | 99 | 83 |
| Average maternal serum elemental concentrations (26–30 wks) | ||
| Selenium | 73.43 ± 1.15 µg/L | |
| Iodine | 93.05 ± 1.47 µg/L | |
| Manganese | 1.16 ± 0.02 µg/L | |
| Copper | 2.41 ± 0.04 mg/L | |
| Zinc | 0.08 ± 0.01 mg/L | |
| Calcium | 55.11 ± 0.30 mg/L | |
| Magnesium | 17.55 ± 0.13 mg/L | |
| Iron | 1.09 ± 3.45 mg/L | |
Data are mean ± s.e.m.
M, male; F, female; GDM, gestational diabetes mellitus; HDP, hypertensive disorder pregnancy; IUGR/LBW/SGA, intra-uterine growth restriction/low birth weight/small for gestational age; PTB, preterm birth; PP, placental previa.
Clinical characteristics of the study population (n = 63) from the MONT cohort.
| Serum selenium concentration | P value | |||
|---|---|---|---|---|
| Low (n = 21) | Mean (n = 21) | Optimal (n = 21) | ||
| Maternal age (years) | 32.00 ± 1.62 | 29.88 ± 1.25 | 30.50 ± 1.85 | 0.5296 |
| Baby sex | ||||
| M | 11, 52.4% | 11, 52.4% | 13, 61.9% | 0.7733 |
| F | 10, 47.6% | 10, 47.6% | 8, 38.1% | |
| BMI | 27.73 ± 2.02 | 27.59 ± 1.76 | 23.96 ± 1.49 | 0.1018 |
| Ethnicity | ||||
| Caucasian | 18, 85.7% | 16, 76.2% | 17, 81.0% | 0.7343 |
| Indigenous | 1, 4.8% | 2, 9.5% | 0, 0% | |
| Indian/Pakistani/Bangladeshi/Sri Lankan | 2, 9.5% | 0, 0% | 0, 0% | |
| Maori/Pacific islander | 0, 0% | 2, 9.5% | 0, 0% | |
| Papua New Guinean/Timorese | 0, 0% | 1, 4.8% | 0, 0% | |
| Chinese/Korean/Japanese | 0, 0% | 0, 0% | 3, 14.2% | |
| Central/South American | 0, 0% | 0, 0% | 1, 4.8% | |
| Alcohola | ||||
| Any | ||||
| No | 21, 100% | 16, 76.2% | 19, 90.5% | 0.3261 |
| Yes | 0, 0% | 5, 23.8% | 2, 9.5% | |
| Smokeb | ||||
| Any | ||||
| No | 20, 95.2% | 20, 95.2% | 21, 100% | 0.5967 |
| Yes | 1, 4.8% | 1, 4.8% | 0, 0% | |
| Birth weight (g) | 2956 ± 184.2 | 3316 ± 121.7 | 3245 ± 106.0 | 0.8138 |
| Parity | 0.71 ± 0.29 | 0.88 ± 0.32 | 0.93 ± 0.33 | 0.9213 |
| Gravidity | 2.50 ± 0.57 | 2.41 ± 0.38 | 2.71 ± 0.42 | 0.8662 |
| Gestation period (week) | 37.48 ± 0.70 | 38.43 ± 0.51 | 38.67 ± 0.29 | 0.1623 |
| TPOAb (IU/L) | 11.45 ± 7.27 | 2.711 ± 0.65 | 1.37 ± 0.54 | 0.0052 |
| Vitamins | ||||
| Any | ||||
| No | 6, 28.6% | 11, 52.4% | 6, 28.6% | 0.1805 |
| Yes | 15, 71.4% | 10, 47.6% | 15, 71.4% | |
| Multivitamin | 15 | 10 | 15 | |
| Selenium | 0 | 0 | 0 | |
Data are shown as mean ± s.e.m. Gender, ethnicity, alcohol intake, smoking status, and vitamin intake were analyzed by chi-square testing. Age, BMI, parity, gravidity, birth weight, and gestation length differences were analyzed by linear regression. TPOAb differences were analyzed by one-way ANOVA with post hoc comparisons performed with Tukey’s tests. Bold text indicates significance (P < 0.05).
aAny alcohol consumption during gestation. bAny smoking during pregnancy.
TPOAb, thyroid peroxidase antibody.
Serum micronutrient concentrations
Serum trace element concentrations were measured using ICP-MS. From these results, three groups from the cohort were selected (Fig. 1). Serum selenium concentrations were significantly decreased (P < 0.0001) in the low (n = 21) serum selenium group (51.2 ± 1.2 µg/L) compared to the mean serum selenium (78.8 ± 0.4 µg/L) group (n = 21). The optimal selenium (106.9 ± 2.3 µg/L) group (n = 21) had significantly higher (P < 0.0001) serum selenium concentrations compared to the mean serum selenium group.
Serum concentrations of selenium within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. ****P < 0.0001 vs the mean group.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Serum concentrations of selenium within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. ****P < 0.0001 vs the mean group.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Serum concentrations of selenium within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. ****P < 0.0001 vs the mean group.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
In addition to selenium, several other trace elements were analyzed in serum using ICP-MS (Table 3). Micronutrient concentrations between the mean serum selenium group and the low serum selenium group were not different. Within the optimal serum selenium group, there was a significant increase in iodine (P = 0.016), calcium (P = 0.004) and magnesium (P = 0.001) concentrations.
Serum concentrations of other trace elements in women with low, mean and optimal serum selenium concentrations at 26–30 weeks of gestation (n = 63).
| Element | Selenium Group | P | ||
|---|---|---|---|---|
| Low (n = 21) | Mean (n = 21) | Optimal (n = 21) | ||
| Iodine (µg/L) | 89.10 ± 4.68 | 89.71 ± 3.03 | 105.90 ± 5.53a | 0.016 |
| Manganese (µg/L) | 1.17 ± 0.06 | 1.04 ± 0.05 | 1.25 ± 0.14 | 0.28 |
| Copper (mg/L) | 2.32 ± 0.13 | 2.32 ± 0.15 | 2.41 ± 0.12 | 0.86 |
| Zinc (mg/L) | 0.76 ± 0.04 | 0.76 ± 0.02 | 0.82 ± 0.04 | 0.39 |
| Calcium (mg/L) | 52.24 ± 0.70 | 54.46 ± 0.56 | 59.16 ± 1.84a | 0.004 |
| Magnesium (mg/L) | 16.46 ± 0.39 | 17.19 ± 0.27 | 19.25 ± 0.76a | 0.001 |
| Iron (mg/L) | 1.01 ± 0.12 | 1.19 ± 0.20 | 1.26 ± 0.10 | 0.46 |
Data are mean ± s.e.m. Difference between multiple groups were tested by one-way ANOVA with bold text indicating significance (P < 0.05). Post hoc comparisons performed with Tukey’s tests.
aP < 0.05 compared to the mean selenium group.
fT3 and fT4 levels and fT3/fT4 ratio within the selenium groups
Thyroid function was determined by measuring serum concentrations of fT3, fT4 and TSH. As shown in Fig. 2A, fT3 was significantly different between selenium groups (P = 0.022). Specifically, fT3 was significantly reduced (P = 0.045) in the low serum selenium group (3.45 ± 0.13 pg/mL) compared to the mean group (3.89 ± 0.12 pg/mL) and optimal serum selenium group (3.91 ± 0.13 pg/mL). There was no significant difference in serum fT4 when comparing low (3.4 ± 0.12 pg/mL), mean (3.54 ± 0.11 pg/mL) and optimal (3.48 ± 0.18 pg/mL) serum selenium groups (Fig. 2B). Furthermore, the fT3/fT4 ratio did not differ between low (1.04 ± 0.07), mean (1.12 ± 0.05) and optimal (1.16 ± 0.07) serum selenium groups (Fig. 2C).
Serum levels of (A) fT3 and (B) fT4 within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. (C) The fT3/fT4 ratio. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. *P = 0.045. fT3, free triiodothyronine; fT4, free thyroxine.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Serum levels of (A) fT3 and (B) fT4 within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. (C) The fT3/fT4 ratio. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. *P = 0.045. fT3, free triiodothyronine; fT4, free thyroxine.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Serum levels of (A) fT3 and (B) fT4 within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. (C) The fT3/fT4 ratio. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. *P = 0.045. fT3, free triiodothyronine; fT4, free thyroxine.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
TSH levels and fT3/TSH, fT4/TSH, (fT3/fT4)/TSH ratios
Figure 3A shows that there were no statistical differences in serum TSH levels between the low serum selenium (0.88 ± 0.11 mIU/L), mean serum selenium (1.18 ± 0.14 mIU/L) or optimal serum selenium groups (1.10 ± 0.11 mIU/L). The fT3/TSH ratio (Fig. 3B) also did not differ significantly between women with low (5.20 ± 0.68), mean (4.02 ± 0.37) or optimal (4.36 ± 0.54) serum selenium. The fT4/TSH in low serum selenium women (5.33 ± 0.76) was significantly higher (P = 0.0258) than those within the mean (3.62 ± 0.32) and optimal (3.6 ± 0.33) serum selenium groups (Fig. 3C). In addition to the TSH ratios above, we also calculated the (fT3/fT4)/TSH ratio which is calculated by dividing the fT3/fT4 ratio by the TSH concentration. The (fT3/fT4)/TSH (Fig. 3D) ratio did not differ between the low (1.51 ± 0.20), mean (1.17 ± 0.12) or optimal (1.38 ± 0.22) serum selenium groups.
Serum levels of (A) TSH, (B) the fT3/TSH ratio, (C) the fT4/TSH ratio and (D) the (fT3/fT4)/TSH ratio within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. *P = 0.034. TSH, thyroid stimulating hormone; fT3, free triiodothyronine; fT4, free thyroxine.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Serum levels of (A) TSH, (B) the fT3/TSH ratio, (C) the fT4/TSH ratio and (D) the (fT3/fT4)/TSH ratio within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. *P = 0.034. TSH, thyroid stimulating hormone; fT3, free triiodothyronine; fT4, free thyroxine.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Serum levels of (A) TSH, (B) the fT3/TSH ratio, (C) the fT4/TSH ratio and (D) the (fT3/fT4)/TSH ratio within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Bars show mean ± s.e.m. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons test. *P = 0.034. TSH, thyroid stimulating hormone; fT3, free triiodothyronine; fT4, free thyroxine.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Correlations of clinical characteristics with thyroid status
Neither maternal age or gravidity was found to correlate with fT3, fT4, TSH, and TPOAb (Table 4). There was a statistically significant positive correlation between BMI and fT3 (R2 = 0.1144, P = 0.0306, Table 4). The addition of serum selenium to BMI for multivariate analysis also correlated to fT3 with statistical significance (R2 = 0.1933, P = 0.0168, Table 4).
Linear regression analysis of possible covariates with thyroid hormones.
| Variables | fT3 | fT4 | TSH | TPO-Ab | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| F | R2 | P value | F | R2 | P value | F | R2 | P value | F | R2 | P value | |
| Age | 0.2336 | 0.005 | 0.6314 | 0.0827 | 0.0019 | 0.775 | 0.5178 | 0.01247 | 0.4759 | 1.678 | 0.03932 | 0.2024 |
| Age + BMI | 2.464 | 0.1148 | 0.0985 | 0.9942 | 0.04497 | 0.3984 | 1.093 | 0.05439 | 0.3455 | 1.099 | 0.05467 | 0.3437 |
| Age + Selenium | 1.229 | 0.0578 | 0.3035 | 0.7953 | 0.03649 | 0.4581 | 0.2704 | 0.01334 | 0.7645 | 3.443 | 0.1469 | 0.0417 |
| Age + Gravidity | 0.3961 | 0.02042 | 0.6757 | 0.1907 | 0.00944 | 0.8271 | 0.3005 | 0.01557 | 0.7422 | 0.8191 | 0.04133 | 0.4484 |
| BMI | 5.036 | 0.1144 | 0.0306 | 1.931 | 0.04497 | 0.1722 | 2.217 | 0.05379 | 0.1445 | 1.208 | 0.03004 | 0.2785 |
| BMI + Selenium | 4.553 | 0.1933 | 0.01689 | 1.332 | 0.06244 | 0.2754 | 1.089 | 0.05422 | 0.3467 | 3.125 | 0.1412 | 0.05542 |
| BMI + Gravidity | 2.111 | 0.1049 | 0.1359 | 1.322 | 0.06506 | 0.2785 | 1.015 | 0.05339 | 0.3724 | 0.6384 | 0.03425 | 0.534 |
| Gravidity | 0.7037 | 0.0177 | 0.4067 | 0.2821 | 0.0068 | 0.5982 | 0.4343 | 0.011 | 0.5138 | 0.08437 | 0.0021 | 0.773 |
| Gravidity + Selenium | 0.7151 | 0.03627 | 0.4956 | 0.7779 | 0.03744 | 0.4662 | 0.3395 | 0.01755 | 0.7143 | 2.969 | 0.1351 | 0.06339 |
| Age + BMI + Gravidity | 1.38 | 0.1058 | 0.2649 | 0.8603 | 0.06521 | 0.4702 | 0.6602 | 0.05356 | 0.5821 | 0.7412 | 0.05974 | 0.5347 |
| Age + BMI + Gravidity + Selenium | 1.491 | 0.1492 | 0.2268 | 0.7153 | 0.07363 | 0.587 | 0.4884 | 0.05433 | 0.7442 | 1.639 | 0.1616 | 0.1872 |
Data analyzed through simple or multiple linear regression of variables in relation to hormones. Presented as the F-statistic, R2 value, and P value for each of the models with bold indicating significance (P < 0.05).
fT3, free triiodothyronine; fT4, free thyroxine; TSH, thyroid-stimulating hormone; TPOAb, thyroid peroxidase antibody.
Correlations of serum selenium with thyroid status
Correlation analysis showed that fT3 was positively correlated with serum selenium concentrations (P = 0.0201, Fig. 4A). There was no correlation between serum selenium concentration and fT4 (Fig. 4B). Correlation analysis also showed that TSH was positively correlated with serum selenium (Fig. 4C); however, this did not reach statistical significance (P = 0.077). Levels of TPOAb were significantly different between serum selenium groups (Table 2, P = 0.005) and further correlation analysis showed that TPOAb is negatively correlated with serum selenium concentration (Fig. 4D, P = 0.0002).
Correlation of serum selenium concentrations with (A) fT3, (B) fT4, (C) TSH and (D) TPOAb concentrations within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Correlations between the different variables were analyzed by correlation analysis using Pearson’s tests for fT3 and fT4 and Spearman’s test for TSH and TPOAb. P < 0.05 was considered statistically significant. fT3, free triiodothyronine; fT4, free thyroxine; TSH, thyroid-stimulating hormone; TPOAb, thyroid peroxidase antibody.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Correlation of serum selenium concentrations with (A) fT3, (B) fT4, (C) TSH and (D) TPOAb concentrations within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Correlations between the different variables were analyzed by correlation analysis using Pearson’s tests for fT3 and fT4 and Spearman’s test for TSH and TPOAb. P < 0.05 was considered statistically significant. fT3, free triiodothyronine; fT4, free thyroxine; TSH, thyroid-stimulating hormone; TPOAb, thyroid peroxidase antibody.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Correlation of serum selenium concentrations with (A) fT3, (B) fT4, (C) TSH and (D) TPOAb concentrations within the low, mean and optimal selenium groups of females at 26–30 weeks of gestation. Correlations between the different variables were analyzed by correlation analysis using Pearson’s tests for fT3 and fT4 and Spearman’s test for TSH and TPOAb. P < 0.05 was considered statistically significant. fT3, free triiodothyronine; fT4, free thyroxine; TSH, thyroid-stimulating hormone; TPOAb, thyroid peroxidase antibody.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Morbidity rates of pregnancy complications
Incidence of HDP, IUGR/LBW/SGA, PTB and PP were not significantly different between the three serum selenium groups (Table 5). Incidence of GDM was significantly different between the three selenium groups (χ2 = 7.370, df = 2, P = 0.025) with the Fisher’s exact test demonstrating that the incidence of GDM within the low selenium group was significantly increased compared to the mean serum selenium group (P = 0.044). In the low selenium group, 7/21 women (33.3%) experienced gestational diabetes mellitus, whereas the incidence of GDM in the mean and optimal Se group was 1/21 (4.76%) and 2/21 (9.52%), respectively. Furthermore, the total incidence of all complicated pregnancies (Fig. 5) was significantly different between the three selenium groups (χ2 = 12.85, df = 2, P = 0.002) with Fisher’s exact testing indicating the incidence of pregnancy complications within the low selenium group (61.9%) was significantly higher than the mean serum selenium group (42.86%, P = 0.007).
Percentage of normal and complicated pregnancies categorized according to serum selenium concentrations at gestational weeks 26–30. HDP, hypertensive disorders of pregnancy; IUGR/LBW/SGA, intra-uterine growth restriction/low birth weight/small for gestational age; PTB, preterm birth; PP, Placental previa; GDM, gestational diabetes mellitus.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Percentage of normal and complicated pregnancies categorized according to serum selenium concentrations at gestational weeks 26–30. HDP, hypertensive disorders of pregnancy; IUGR/LBW/SGA, intra-uterine growth restriction/low birth weight/small for gestational age; PTB, preterm birth; PP, Placental previa; GDM, gestational diabetes mellitus.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Percentage of normal and complicated pregnancies categorized according to serum selenium concentrations at gestational weeks 26–30. HDP, hypertensive disorders of pregnancy; IUGR/LBW/SGA, intra-uterine growth restriction/low birth weight/small for gestational age; PTB, preterm birth; PP, Placental previa; GDM, gestational diabetes mellitus.
Citation: Journal of Endocrinology 248, 1; 10.1530/JOE-20-0319
Presentation of complications analyzed by chi-square.
| Serum selenium concentration | χ2 (df) | P | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Low (51.2 ± 1.2 µg/L) | Mean (78.8 ± 0.4 µg/L) | Optimal (106.9 ± 2.3 µg/L) | |||||||
| Negative | Positive | Negative | Positive | Negative | Positive | ||||
| HDP | Count (n) | 19 | 2 | 18 | 3 | 19 | 2 | 0.3214 | 0.8515 |
| % within Group (n = 21) | 90.48% | 9.52% | 85.71% | 14.29% | 90.48% | 9.52% | |||
| % of Total (n = 63) | 30.16% | 3.17% | 28.57% | 4.76% | 30.16% | 3.17% | |||
| IUGR/LBW/SGA | Count (n) | 16 | 5 | 16 | 5 | 19 | 2 | 1.853 | 0.3959 |
| % within Group (n = 21) | 76.19% | 23.81% | 76.19% | 23.81% | 90.48% | 9.52% | |||
| % of Total (n = 63) | 25.40% | 7.94% | 25.40% | 7.94% | 30.16% | 3.17% | |||
| PTB | Count (n) | 17 | 4 | 17 | 4 | 20 | 1 | 2.333 | 0.3114 |
| % within Group (n = 21) | 80.95% | 19.05% | 80.95% | 19.05% | 95.24% | 4.76% | |||
| % of Total (n = 63) | 26.98% | 6.35% | 26.98% | 6.35% | 31.75% | 1.59% | |||
| PP | Count (n) | 20 | 1 | 20 | 1 | 19 | 2 | 0.5339 | 0.7657 |
| % within Group (n = 21) | 95.24% | 4.76% | 95.24% | 4.76% | 90.48% | 9.52% | |||
| % of Total (n = 63) | 31.75% | 1.59% | 31.75% | 1.59% | 30.16% | 3.17% | |||
| GDM | Count (n) | 14 | 7 | 20 | 1 | 19 | 2 | 7.370 | 0.0251 |
| % within Group (n = 21) | 66.67% | 33.33%a | 95.24% | 4.76% | 90.48% | 9.52% | |||
| % of Total (n = 63) | 22.22% | 11.11%a | 31.75% | 1.59% | 30.16% | 3.17% | |||
Chi-square analysis of morbidity incidence within the low, mean and optimal selenium groups of females at 26-30 weeks of gestation with bold text indicating significance (P < 0.05).
aSignificance from Fisher’s exact test (P < 0.05) compared to mean selenium group.
GDM, gestational diabetes mellitus; HDP, hypertensive disorders of pregnancy; IUGR/LBW/SGA, intra-uterine growth restriction/low birth weight/small for gestational age; PP, Placental previa; PTB, preterm birth.
Discussion
The relationship between selenium and thyroid hormone status has been extensively explored and published (Drutel et al. 2013, Ventura et al. 2017). Previous studies have demonstrated that aberrant selenium levels during pregnancy can have severe adverse effects on maternal and fetal outcomes through thyroid pathologies (Richard et al. 2017, Hofstee et al. 2018). In most studies that investigate the potential benefit of selenium supplementation for thyroid dysfunction, changes in selenium concentrations are not measured as a primary outcome (Negro et al. 2007). In this study, we have retrospectively investigated serum selenium status and thyroid hormone concentrations in pregnant women and correlated these findings with the incidence of pregnancy complications.
We observed that women with low serum selenium concentrations (51.2 ± 1.2 µg/L) had a decrease in serum levels of fT3 in comparison to the mean serum selenium group (78.8 ± 0.4 µg/L). This was concomitant with no changes in fT4 and TSH. Clinically, the assessment of fT3 is only made if levels of TSH and fT4 are abnormal (Smith et al. 2017). Consequently, the implications of independently decreased fT3 during pregnancy is unknown; however, reduced fT3 may be a consequence of reduced serum selenium concentration, limiting DIO function. DIO-1 and DIO-2 are selenoproteins that have critical roles in maintaining intracellular and circulating levels of T3, so reduced deiodinase activity would result in reduced conversion of T4 to T3, subsequently reducing fT3 levels (Ventura et al. 2017). Our findings suggest that selenium deficiency inhibits thyroid hormone metabolism during pregnancy, which may be implicated in hypothyroidism.
Although the reduction in fT3 concentrations were only modest (12% decreased), this hormone is normally maintained within a tight reference range. Maintaining a fT3 concentration within a very stable range is the primary target in subtle adjustments in TSH and T4 concentrations (Abdalla & Bianco 2014). Previous studies have demonstrated that even a moderate decline in fT3 of 10% may contribute to conditions such as diabetic nephropathy (Wu et al. 2015, Hofstee et al. 2020a). Other studies demonstrate that moderate decreases in fT3 are associated with insulin resistance in non-diabetic individuals (Wang et al. 2018). While the reference range used differs depending on gestational age, the population studied and the methodology employed, if a reference range of 3.1–6.35 pm/L for fT3 is employed, 23% (5/21) of samples from the low selenium cohort were below this threshold, while all samples from the mean and optimal groups were within this range (Sekhri et al. 2016). Clinically, T3 is not assessed when determining thyroid function in patients (Smith et al. 2017); however, it may be possible that women exhibit normal T4 concentrations but low T3 levels. Additionally, a positive correlation between serum selenium concentration and fT3 levels was also observed. This further supports that the bioavailability of fT3 in pregnancy is dependent on serum selenium concentration, likely due to the maintenance of DIO homeostatic function; however, previous studies indicate DIO function is maintained during prolonged selenium deficiency (Bates et al. 2000). Thus, it is important for future studies to confirm that selenium deficiency impacts thyroid function via reducing DIO1 and DIO2 function and to determine if this is unique to the pregnant state (Hofstee et al. 2019, 2020b). Alternatively, there may be additional mechanisms linking selenium deficiency to changes in serum thyroid levels. Reference intervals for thyroid hormones during pregnancy, specifically fT3, fT4 and TSH, are shown to be lower than non-pregnant women (Panesar et al. 2001, Moon et al. 2015). We reiterate the importance of developing gestational age-specific reference intervals when analysing thyroid function. Interestingly, a study analyzing pregnant women from China determined a set of gestation-related reference intervals for thyroid hormones with median serum levels for fT3 being 3.1–3.2 pmol/L with the lowest 2∙5th centile 2.3–2.5 pmol/L (Panesar et al. 2001). This indicates fT3 levels within this cohort remained within the reference range of healthy pregnant women, according to the aforementioned study.
The fact that thyroid hormone concentrations and pregnancy outcomes were similar in women from the mean group and the optimal group might suggest that in the absence of other major risk factors, selenium concentrations at mean values may be sufficient. However, previous studies have demonstrated adverse outcomes for thyroid function in non-pregnant individuals when levels reach approximately 71 µg/L, which is similar to the mean plasma concentration detected in women from our cohort (World Health Organisation 2004). Although adverse effects were not detected in these women, that would otherwise be considered mildly selenium deficient, it is likely that the use of a larger sample size may detect more subtle effects occurring in this group.
With regards to thyroid pathology, dietary selenium intake has been particularly associated with autoimmune disorders. In a study investigating the effects of selenium supplementation on 232 pregnant women with autoimmune thyroiditis, pregnant women who received 200 μg/day of selenomethionine were seen to have reductions in TPOAb levels and autoimmune thyroiditis progression, with improved thyroid echogenicity and a decreased incidence of thyroid dysfunction postpartum (Negro et al. 2007). We have observed that TPOAb levels were increased in pregnant women with low serum selenium concentrations. We also demonstrate that TPOAb levels were negatively correlated with serum selenium concentrations of women at 26–30 weeks of gestation. This supports research demonstrating that selenium deficiency may have implications with autoimmune thyroid disorders during pregnancy. Currently, selenium supplementation is documented to improve anti-thyroid antibody levels for pregnant women with high levels of TPOAb, significantly reducing the percentage of postpartum thyroiditis and definitive hypothyroidism (Drutel et al. 2013); however, this may not be relevant to iodine-deficient patients (Mao et al. 2016).
The human HPT axis maintains a physiologically inverse relationship between TSH and thyroid hormones. There was no change in fT4 or TSH levels between the three selenium groups; however, the fT4/TSH ratio was significantly increased within the low selenium group compared to both the other groups. This indicates the standardized estimate of T4 production per unit of TSH is increased within the low selenium cohort. This imbalance may be indicative of early stage capacity stress, where a compensatory increase in conversion efficiency is attempting to maintain a euthyroid state. This mechanism has been proposed as an early indicator of a progressively failing thyroid gland in conditions such as autoimmune thyroiditis (Hoermann et al. 2016). Although our samples are from 26 to 30 weeks of gestation, these results may suggest a minor form of hypothyroidism earlier in pregnancy in the low selenium group, though a compensatory mechanism has caused the thyroid to increase sensitivity to TSH in order to re-establish normal thyroxine levels. This compensatory shift in HPT function would permit re-establishment of the fT3/fT4 ratio; however, the reduction in fT3 suggests reduced selenium concentrations are impeding the capacity of fT3 production even with sufficient fT4, which may be a result of reduced DIO1 and 2 activity.
Clinical characteristics were observed to have no direct impact on selenium concentration; however, BMI was indicated to be positively correlated to fT3. When BMI and selenium status were used in conjunction, they correlated significantly with both fT3 and TPOAb. Previous studies have noted an association between T3/T4 ratio with BMI in obese Nigerian children and adolescents (Emokpae & Obazelu 2017). It was reported that as BMI increases there is an alteration in the T3/T4 ratio, predominantly due to an increase in T3 without changes in T4. It was not stated that the increased BMI was a result of altered thyroid hormones, or causative of (Emokpae & Obazelu 2017). Nevertheless, the primary aim of this investigation was to assess the effect of selenium on thyroid hormone status within these pregnant women; our data showed no difference in the BMI reported across the three study groups selected on selenium status.
Of interest, the incidence of GDM was increased in the low selenium cohort. The association between serum selenium concentrations and GDM has been shown to be complex. A recent meta-analysis conducted by Kong et al. attempted to elucidate the relationship between GDM and serum selenium. This study suggested that serum selenium concentrations are lower in women with GDM (Kong et al. 2016). Previous studies have also demonstrated the beneficial effects of selenium supplementation in pregnant women with GDM on biomarkers of oxidative stress and glucose metabolism (Asemi et al. 2015). Conversely, dietary selenium intake above recommended levels has also been implicated in the development of insulin resistance and type 2 diabetes mellitus (Stranges et al. 2007, Kohler et al. 2018). These studies have suggested that, in excess, selenium may act as an insulin mimetic, although mechanisms remain unclear (Stapleton 2000). These effects may also be dependent on the extent of selenium supplementation, with a study demonstrating that nearly three years of selenium supplementation (200 µg/day) had no effect on insulin sensitivity or β-cell function when compared to a placebo group (Jacobs et al. 2019). As biochemical alterations (reduced glutathione blood levels and prothrombin times) occur at intakes of selenium ranging from 700 to 850 µg/day, and clinically relevant signs of selenosis are only apparent between 1200 and 5000 µg/day, we highlight the need for further research to investigate the relationship between selenium intake, serum selenium concentration and metabolic health status. Maternal BMI is also significantly associated with incidence of GDM (Martin et al. 2015), thyroid dysfunction (Han et al. 2015) and selenium status (Carvalho et al. 2015). In future studies, maternal BMI should be considered when investigating these relationships as it may significantly modify or mediate the relationship between selenium, thyroid function and pregnancy complications.
Given these findings, the importance of sustaining selenium status to maintain thyroid health during pregnancy is imperative. One of the key limitations of this study was the lack of evidence indicating whether the increase in GDM in the low selenium group was causative or consequential. Additionally, whether atypical thyroid hormone status was implicated in the incidence of GDM also remains to be delineated. A further limitation is the lack of measurement of selenium-dependent iodothyronine deiodinases. Given the apparent limitations, this study does highlight pregnant women in Queensland may be experiencing selenium deficiency that correlates with atypical thyroid hormone levels. Future studies should further aim to delineate the relationship between selenium, selenoproteins and thyroid hormone function during pregnancy, which may also be further implicated in metabolic disorders such as GDM.
Conclusion
Thyroid disorders are a predominant endocrine disorder affecting women of reproductive age, with thyroid dysfunction affecting approximately 2–3% of all pregnant women (Forehan 2012). This study identified that euthyroid women from Queensland, Australia had low serum selenium concentrations at 26–30 weeks of gestation, which was associated with reduced fT3 and increased TPOAb, though not at physiologically detrimental levels. Importantly, this study also identified a positive correlation between serum selenium and fT3, as well as a negative correlation between serum selenium and TPOAb. We provide further insight into the effect of different serum selenium concentrations during pregnancy on thyroid hormone levels, and that selenium deficiency is implicated in aberrant thyroid status during pregnancy. Importantly, this may be associated with increased morbidity of pregnancy complications, most notably GDM.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
P H receives support from the Children’s Hospital Foundation, Queensland Health (Smarty Grants: RPCPHD0192017). The funding sources had no input in study design; in collection, analysis, and interpretation of data; in the writing of the report; and the decision to submit the report for publication. Study contents are the sole responsibility of the authors and do not necessarily represent the official views of CHF.
Data availability
The datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.
Author contribution statement
J S M C and A V P contributed equally as joint senior authors.
Acknowledgements
The authors would like to acknowledge the management and midwives of the Gold Coast University and Royal Brisbane and Women’s Hospitals.
References
Abdalla SM & Bianco AC 2014 Defending plasma T3 is a biological priority. Clinical Endocrinology 81 633–641. (https://doi.org/10.1111/cen.12538)
Asemi Z, Jamilian M, Mesdaghinia E & Esmaillzadeh A 2015 Effects of selenium supplementation on glucose homeostasis, inflammation, and oxidative stress in gestational diabetes: randomized, double-blind, placebo-controlled trial. Nutrition 31 1235–1242. (https://doi.org/10.1016/j.nut.2015.04.014)
Bates JM, Spate VL, Morris JS, St. Germain DL & Galton VA 2000 Effects of selenium deficiency on tissue selenium content, deiodinase activity, and thyroid hormone economy in the rat during development. Endocrinology 141 2490–2500. (https://doi.org/10.1210/endo.141.7.7571)
Carvalho RF, Rosa G, Huguenin GV, Luiz RR, Moreira AS & Oliveira GM 2015 The association of selenium status with thyroid hormones and anthropometric values in dyslipidemic patients. Nutricion Hospitalaria 31 1832–1838. (https://doi.org/10.3305/nh.2015.31.4.8363)
Combs GF 2001 Selenium in global food systems. British Journal of Nutrition 85 517–547. (https://doi.org/10.1079/bjn2000280)
De Blas Bravo I, Sanz Castro R, López Riquelme N, Tormo Díaz C & Apraiz Goyenaga D 2007 Optimization of the trace element determination by ICP-MS in human blood serum. Journal of Trace Elements in Medicine and Biology 21 (Supplement 1) 14–17. (https://doi.org/10.1016/j.jtemb.2007.09.017)
De Groot L, Abalovich M, Alexander EK, Amino N, Barbour L, Cobin RH, Eastman CJ, Lazarus JH, Luton D & Mandel SJ et al. 2012 Management of thyroid dysfunction during pregnancy and postpartum: an Endocrine Society clinical practice guideline. Journal of Clinical Endocrinology and Metabolism 97 2543–2565. (https://doi.org/10.1210/jc.2011-2803)
Drutel A, Archambeaud F & Caron P 2013 Selenium and the thyroid gland: more good news for clinicians. Clinical Endocrinology 78 155–164. (https://doi.org/10.1111/cen.12066)
Dulek H, Vural F, Aka N & Zengin S 2019 The prevalence of thyroid dysfunction and its relationship with perinatal outcomes in pregnant women in the third trimester. Northern Clinics of Istanbul 6 267–272. (https://doi.org/10.14744/nci.2018.51422)
Duntas LH & Benvenga S 2015 Selenium: an element for life. Endocrine 48 756–775. (https://doi.org/10.1007/s12020-014-0477-6)
Emokpae MA & Obazelu PA 2017 The association of triiodothyronine-to-thyroxine ratio with body mass index in obese Nigerian children and adolescents. Medical Sciences 5 36. (https://doi.org/10.3390/medsci5040036)
Forehan S 2012 Thyroid disease in the perinatal period. Australian Family Physician 41 578–581.
Han C, Li C, Mao J, Wang W, Xie X, Zhou W, Li C, Xu B, Bi L & Meng T et al. 2015 High body mass index is an indicator of maternal hypothyroidism, hypothyroxinemia, and thyroid-peroxidase antibody positivity during early pregnancy. BioMed Research International 2015 351831. (https://doi.org/10.1155/2015/351831)
Hoermann R, Midgley JE, Larisch R & Dietrich JW 2016 Relational stability of thyroid hormones in euthyroid subjects and patients with autoimmune thyroid disease. European Thyroid Journal 5 171–179. (https://doi.org/10.1159/000447967)
Hofstee P, McKeating DR, Perkins AV & Cuffe JS 2018 Placental adaptations to micronutrient dysregulation in the programming of chronic disease. Clinical and Experimental Pharmacology and Physiology 45 871–884. (https://doi.org/10.1111/1440-1681.12954)
Hofstee P, Bartho LA, McKeating DR, Radenkovic F, McEnroe G, Fisher JJ, Holland OJ, Vanderlelie JJ, Perkins AV & Cuffe JSM 2019 Maternal selenium deficiency during pregnancy in mice increases thyroid hormone concentrations, alters placental function and reduces fetal growth. Journal of Physiology 597 5597–5617. (https://doi.org/10.1113/JP278473)
Hofstee P, McKeating DR, Bartho LA, Anderson ST, Perkins AV & Cuffe JS 2020a Maternal selenium deficiency in mice alters offspring glucose metabolism and thyroid status in a sexually dimorphic manner. Nutrients 12 267. (https://doi.org/10.3390/nu12010267)
Hofstee P, Cuffe JSM & Perkins AV 2020b Analysis of selenoprotein expression in response to dietary selenium deficiency during pregnancy indicates tissue specific differential expression in mothers and sex specific changes in the fetus and offspring. International Journal of Molecular Sciences 21 2210. (https://doi.org/10.3390/ijms21062210)
Hubalewska-Dydejczyk A, Duntas L & Gilis-Januszewska A 2019 Pregnancy, thyroid, and the potential use of selenium. Hormones 19 47–53. (https://doi.org/10.1007/s42000-019-00144-2)
Hurst R, Armah CN, Dainty JR, Hart DJ, Teucher B, Goldson AJ, Broadley MR, Motley AK & Fairweather-Tait SJ 2010 Establishing optimal selenium status: results of a randomized, double-blind, placebo-controlled trial. American Journal of Clinical Nutrition 91 923–931. (https://doi.org/10.3945/ajcn.2009.28169)
Jacobs ET, Lance P, Mandarino LJ, Ellis NA, Chow HS, Foote J, Martinez JA, Hsu C-HP, Batai K & Saboda K et al. 2019 Selenium supplementation and insulin resistance in a randomized, clinical trial. BMJ Open Diabetes Research and Care 7 e000613. (https://doi.org/10.1136/bmjdrc-2018-000613)
Kohler LN, Foote J, Kelley CP, Florea A, Shelly C, Chow HS, Hsu P, Batai K, Ellis N & Saboda K et al. 2018 Selenium and type 2 diabetes: systematic review. Nutrients 10 1924. (https://doi.org/10.3390/nu10121924)
Kong FJ, Ma LL, Chen SP, Li G & Zhou JQ 2016 Serum selenium level and gestational diabetes mellitus: a systematic review and meta-analysis. Nutrition Journal 15 94. (https://doi.org/10.1186/s12937-016-0211-8)
Lewandowska M, Sajdak S & Lubiński J 2019 Serum selenium level in early healthy pregnancy as a risk marker of pregnancy induced hypertension. Nutrients 11 1028. (https://doi.org/10.3390/nu11051028)
Mantovani G, Isidori AM, Moretti C, Di Dato C, Greco E, Ciolli P, Bonomi M, Petrone L, Fumarola A & Campagna G et al. 2019 Selenium supplementation in the management of thyroid autoimmunity during pregnancy: results of the ‘Serena study’, a randomized, double-blind, placebo-controlled trial. Endocrine 66 542–550. (https://doi.org/10.1007/s12020-019-01958-1)
Mao J, Pop VJ, Bath SC, Vader HL, Redman CW & Rayman MP 2016 Effect of low-dose selenium on thyroid autoimmunity and thyroid function in UK pregnant women with mild-to-moderate iodine deficiency. European Journal of Nutrition 55 55–61. (https://doi.org/10.1007/s00394-014-0822-9)
Martin KE, Grivell RM, Yelland LN & Dodd JM 2015 The influence of maternal BMI and gestational diabetes on pregnancy outcome. Diabetes Research and Clinical Practice 108 508–513. (https://doi.org/10.1016/j.diabres.2014.12.015)
McAlpine JM, McKeating DR, Vincze L, Vanderlelie JJ & Perkins AV 2019 Essential mineral intake during pregnancy and its association With maternal health and birth outcomes in South East Queensland, Australia. Nutrition and Metabolic Insights 12 1178638819879444. (https://doi.org/10.1177/1178638819879444)
McAlpine JM, Vanderlelie JJ, Vincze LJ & Perkins AV 2020 Use of micronutrient supplements in pregnant women of South-East Queensland. Australian and New Zealand Journal of Obstetrics and Gynaecology 60 561–567. (https://doi.org/10.1111/ajo.13109)
McKeating DR, Fisher JJ, Zhang P, Bennett WW & Perkins AV 2020 Elemental metabolomics in human cord blood: method validation and trace element quantification. Journal of Trace Elements in Medicine and Biology 59 126419. (https://doi.org/10.1016/j.jtemb.2019.126419)
Moon HW, Chung HJ, Park CM, Hur M & Yun YM 2015 Establishment of trimester-specific reference intervals for thyroid hormones in Korean pregnant women. Annals of Laboratory Medicine 35 198–204. (https://doi.org/10.3343/alm.2015.35.2.198)
Nazarpour S, Tehrani FR, Simbar M & Azizi F 2015 Thyroid dysfunction and pregnancy outcomes. Iranian Journal of Reproductive Medicine 13 387–396.
Negro R, Greco G, Mangieri T, Pezzarossa A, Dazzi D & Hassan H 2007 The influence of selenium supplementation on postpartum thyroid status in pregnant women with thyroid peroxidase autoantibodies. Journal of Clinical Endocrinology and Metabolism 92 1263–1268. (https://doi.org/10.1210/jc.2006-1821)
NHMRC 2017 Nutrient Reference Values for Australia and New Zealand Including Recommended Dietary Intakes. Canberra, Australia: National Health and Medical Research Council Australian Government Department of Health and Ageing and New Zealand Ministry of Health.
NHMRC 2018 Department of Health Clinical Practice Guidelines: Pregnancy Care. Canberra, Australia: National Health and Medical Research Council.
Panesar NS, Li CY & Rogers MS 2001 Reference intervals for thyroid hormones in pregnant Chinese women. Annals of Clinical Biochemistry 38 329–332. (https://doi.org/10.1258/0004563011900830)
Reid SM, Middleton P, Cossich MC, Crowther CA & Bain E 2013 Interventions for clinical and subclinical hypothyroidism pre‐pregnancy and during pregnancy. Cochrane Database of Systematic Reviews 5 CD007752. (https://doi.org/10.1002/14651858.CD007752.pub3)
Richard K, Holland O, Landers K, Vanderlelie JJ, Hofstee P, Cuffe JSM & Perkins AV 2017 Review: Effects of maternal micronutrient supplementation on placental function. Placenta 54 38–44. (https://doi.org/10.1016/j.placenta.2016.12.022)
Santos C, García-Fuentes E, Callejón-Leblic B, García-Barrera T, Gómez-Ariza JL, Rayman MP & Velasco I 2017 Selenium, selenoproteins and selenometabolites in mothers and babies at the time of birth. British Journal of Nutrition 117 1304–1311. (https://doi.org/10.1017/S0007114517001155).
Sekhri T, Juhi JA, Wilfred R, Kanwar RS, Sethi J, Bhadra K, Nair S & Singh S 2016 Trimester specific reference intervals for thyroid function tests in normal Indian pregnant women. Indian Journal of Endocrinology and Metabolism 20 101–107. (https://doi.org/10.4103/2230-8210.172239)
Shreenath AP & Dooley J 2018 Selenium deficiency. StatPearls Publishing, 2020.
Smallridge RC & Ladenson PW 2001 Hypothyroidism in pregnancy: consequences to neonatal health. Journal of Clinical Endocrinology and Metabolism 86 2349–2353. (https://doi.org/10.1210/jcem.86.6.7577)
Smith A, Eccles-Smith J, D’Emden M & Lust K 2017 Thyroid disorders in pregnancy and postpartum. Australian Prescriber 40 214–219. (https://doi.org/10.18773/austprescr.2017.075)
Stapleton SR 2000 Selenium: an insulin mimetic. Cellular and Molecular Life Sciences 57 1874–1879. (https://doi.org/10.1007/PL00000669)
Stoffaneller R & Morse NL 2015 A review of dietary selenium intake and selenium status in Europe and the Middle East. Nutrients 7 1494–1537. (https://doi.org/10.3390/nu7031494)
Stranges S, Marshall JR, Natarajan R, Donahue RP, Trevisan M, Combs GF, Cappuccio FP, Ceriello A & Reid ME 2007 Effects of long-term selenium supplementation on the incidence of type 2 diabetes: a randomized trial. Annals of Internal Medicine 147 217–223. (https://doi.org/10.7326/0003-4819-147-4-200708210-00175)
Ventura M, Melo M & Carrilho F 2017 Selenium and thyroid disease: From pathophysiology to treatment. International Journal of Endocrinology 2017 1297658. (https://doi.org/10.1155/2017/1297658)
Wang CY, Yu TY, Shih SR, Huang KC & Chang TC 2018 Low total and free triiodothyronine levels are associated with insulin resistance in non-diabetic individuals. Scientific Reports 8 10685. (https://doi.org/10.1038/s41598-018-29087-1)
World Health Organization 2004 Joint FAO/WHO Expert Consultation on Human Vitamin and Mineral Requirements in Human Nutrition, pp. 289–302. World Health Organization.
Wu J, Li X, Tao Y, Wang Y & Peng Y 2015 Free triiodothyronine levels are associated with diabetic nephropathy in euthyroid patients with type 2 diabetes. International Journal of Endocrinology 2015 204893. (https://doi.org/10.1155/2015/204893)
Yen PM 2001 Physiological and molecular basis of thyroid hormone action. Physiological Reviews 81 1097–1142. (https://doi.org/10.1152/physrev.2001.81.3.1097)
