Fetal thyroid hormone level as measured in the cord blood is associated with birth weight, supporting the role of thyroid hormone in fetal growth.
Abstract
Context:
Thyroid function is known to play an important role in fetal neurological development, but its role in regulating fetal growth is not well established. Overt maternal and fetal thyroid disorders are associated with reduced birth weight. We hypothesized that, even in the absence of overt thyroid dysfunction, maternal and fetal thyroid function influence fetal growth.
Aim:
In normal, healthy pregnancies, we aimed to assess whether fetal thyroid hormone at birth (as measured in cord blood) is associated with fetal growth. We also aimed to study whether fetal thyroid hormone at birth is associated with maternal thyroid hormone in the third trimester.
Methods:
In 616 healthy mother-child pairs, TSH, free T4 (FT4), and free T3 (FT3) were measured in mothers at 28 wk gestation and in umbilical cord blood at birth. Birth weight, length, head circumference, and tricep and bicep skinfold thicknesses were measured on the babies.
Results:
Cord FT4 was associated with birth weight (r = 0.25; P < 0.001), length (r = 0.17; P < 0.001), and sum of skinfolds (r = 0.19; P < 0.001). There were no associations between birth measurements and either cord TSH or cord FT3. Maternal FT4 and cord FT4 were correlated (r = 0.14; P = 0.0004), and there were weaker negative associations between maternal TSH and cord FT4 (r = −0.08; P = 0.04) and FT3 (r = −0.10; P = 0.01).
Conclusion:
Associations between cord FT4 and birth size suggest that fetal thyroid function may be important in regulating fetal growth, both of skeletal size and fat. The correlation between third-trimester maternal FT4 and cord FT4 supports the belief that maternal T4 crosses the placenta even in late gestation.
Thyroid hormone plays an important role in neurodevelopment of the fetus (1), and both overt and subclinical maternal hypothyroidism during pregnancy are associated with impaired neurological development in infancy and early childhood (2, 3). However, the role of thyroid hormone in regulating other aspects of fetal development is not well established.
Several lines of evidence suggest that thyroid hormone could be important in regulating fetal growth. Low birth weight is a well-recognized complication of both uncontrolled Graves' disease and hypothyroidism during pregnancy (4, 5). In cases of thyroid hormone resistance, unaffected infants born to affected mothers with high thyroid hormone levels tend to be born smaller (6). Fetal thyrotoxicosis, due to activating mutations in the TSHR gene, has been found to be associated with reduced birth weight (7). These observations suggest that not only maternal but also fetal thyroid hormone levels are important. It is thought that low birth weight associated with overt maternal thyroid dysfunction is secondary to suboptimal transplacental transfer of maternal thyroid hormone (8).
Although overt maternal and fetal thyroid dysfunctions are associated with reduced birth weight, the impact of maternal and fetal thyroid function within the normal range on fetal growth is less clear. We hypothesized that, even in the absence of overt thyroid dysfunction, maternal and fetal thyroid function influence fetal growth. We therefore aimed to assess in normal healthy pregnancies whether fetal thyroid hormone levels at birth (as measured in cord blood) and maternal thyroid hormone levels in the third trimester are associated with fetal growth. In addition, we examined whether fetal cord blood thyroid hormone levels are correlated with maternal third-trimester thyroid hormone levels.
Subjects and Methods
Study cohort
Detailed anthropometric measurements and serum samples were taken on 974 women without known thyroid disease who were recruited into the Exeter Family Study of Childhood Health (EFSOCH) at 28 wk gestation (9). Routine blood test results (including glucose) were taken at this time, and women reported whether they smoked or not during pregnancy. Socioeconomic status (SES) was assessed by Townsend score based on postal code, with negative scores representing more affluence and positive scores indicating more deprivation (9). All subjects gave informed consent, and ethical approval was obtained from the North and East Devon Local Research Ethics Committee. TSH, free T4 (FT4), free T3 (FT3), and thyroid peroxidase antibodies (TPOAb) were measured in the stored serum samples. Based on these results, 10 women with overt hypothyroidism (TSH > 4.5 mIU/liter and FT4 < 11 pmol/liter) and one with overt hyperthyroidism (TSH < 0.01 mIU/liter, and FT4 > 24 pmol/liter) were excluded from analysis.
Of the offspring of these 963 women, 42 were born premature (<37 wk gestation), data were missing on 10, two had congenital abnormalities (one had cerebral palsy, one had chromosome abnormalities), and four children were twins. They were excluded from analysis. Detailed anthropometric measurements were taken on all offspring at birth, including weight, length, head circumference, and tricep and bicep skinfold thicknesses. Gestational age was determined by the last menstrual period when periods were regular and thought reliable, or by ultrasound scan in all other cases or if it differed from the last menstrual period estimate by more than 10 d. Analysis of maternal results and birth measures was based on 905 mother-child pairs.
Umbilical cord blood was taken at the time of delivery and stored at −80 C. TSH, FT4, and FT3 were measured in the stored plasma samples. Results were available on 616 babies, and therefore analysis involving cord results is based on this smaller cohort.
Analysis of thyroid function and thyroid antibodies
Serum (mothers) and plasma (cord) TSH, FT4, and FT3 were analyzed using the electrochemiluminescent immunoassay, run on the Modular E170 Analyzer (Roche, Burgess Hill, UK). Intraassay coefficients of variation were: TSH, <5.3%; FT4, <5.3%; and FT3, <5.1%. The manufacturer's population reference ranges were: TSH, 0.35–4.5 mIU/liter; FT4, 11–24 pmol/liter; and FT3, 3.9–6.8 pmol/liter. TPOAb were analyzed using the competitive immunoassay (Roche), and a titer above 34 IU/ml was considered positive.
Statistical analysis
Thyroid function measures showed a skewed distribution; therefore, median and interquartile range (IQR) are presented for these data. All maternal results and cord FT3 and TSH required natural log (ln) transformation to achieve normal distribution for subsequent analyses. Pearson correlation coefficients were used to assess associations between maternal and cord thyroid function tests. Partial correlations were used to determine associations between cord thyroid function results and measures of birth size, adjusting for sex and gestation. ANOVA with contrasts was used to assess differences in birth weight, adjusted for sex and gestation, between the quartiles of cord FT4. To provide some indication of effect size, β coefficients from regression analysis were used to determine the increase in birth weight (or FT4) per unit increase in thyroid function results. In the case of log-transformed variables, the effect size was calculated in terms of unit increase in the dependent variable per 10% increase in the independent variable using the formula β*ln(1.1).
Results
Characteristics of mothers and babies (Supplemental Table 1, published on The Endocrine Society's Journals Online web site at http://jcem.endojournals.org)
Mothers had a mean (sd) age of 30 (5.2) yr and a median (IQR) 28-wk body mass index (BMI) of 27 (24.7, 30.2) kg/m2. A total of 412 (45.5%) were primiparous, and 119 (13%) smoked. The babies had a mean (sd) gestation of 40.1 (1.2) wk and a mean (sd) birth weight of 3500 (474) g. A total of 466 (51.5%) were male, and 59 (6.5%) women were TPOAb positive.
Associations between cord thyroid function tests and birth weight
Table 1 shows correlations between cord thyroid function measures and birth size and gestation. Cord FT4 was associated with birth weight, even when adjusted for sex and gestation (r = 0.25; P < 0.001), but there was no association between either cord FT3 or TSH and birth weight. The association between cord FT4 and birth weight remained when adjusting for other potential confounders of maternal smoking, BMI, parity (primiparous or multiparous), and glucose (r = 0.22; P < 0.001), and when looking at only the babies born to the 843 women who were TPOAb negative (r = 0.19; P < 0.001). Similar associations were seen with cord FT4 and other measures of birth size, including measures of skeletal size (birth length and head circumference), fat (sum of skinfolds), and placental weight (Table 1).
Table 1.
Correlations (r) between birth measures (adjusted for sex and gestational age) or gestational age (37–42 wk) and cord thyroid function tests (n = 616) and maternal FT4 (n = 905)
| Cord TSHa | Cord FT4 | Cord FT3a | Maternal FT4a | |
|---|---|---|---|---|
| Birth weight | 0.04 | 0.25*** | 0.02 | −0.18*** |
| Birth length | −0.02 | 0.17*** | 0.03 | −0.11* |
| Birth head circumference | 0.05 | 0.11 | 0.02 | −0.14*** |
| Birth sum of skinfolds | −0.02 | 0.19*** | 0.05 | −0.18*** |
| Gestational age | −0.03 | −0.10 | 0.009 | 0.01 |
| Placental weight | 0.004 | 0.20*** | 0.05 | −0.16*** |
Bonferroni adjusted P values are presented: ***, <0.001; **, <0.01; *, <0.05.
Log-transformed variables.
Regression analysis showed that a 1-pmol/liter increase in cord FT4 was associated with a 64-g increase in birth weight, adjusted for sex and gestational age [β (95% confidence interval or CI) = 64 (45, 84)].
These significant associations were emphasized further when examining quartiles of cord FT4. Babies in the highest quartile of cord FT4 values had higher birth weights, adjusted for sex and gestational age, than babies in the lowest quartile of cord FT4 (3682 vs. 3381 g, respectively; P < 0.0001).
Within this group of term babies, a weak negative association between gestational age and cord FT4 was seen (r = −0.10; P = 0.01).
Associations between maternal thyroid function tests and birth weight
In contrast to the associations seen with cord thyroid function tests, log maternal FT4 was negatively associated with birth weight (r = −0.18, when adjusting for sex and gestation; P < 0.001). Regression analysis showed that a 10% increase in maternal FT4 was associated with a 59-g decrease in birth weight [β*ln(1.1) = −59 (95% CI, −80, −37)]. This was further emphasized when looking at quartiles of maternal FT4 concentrations, where birth weight decreased with increasing FT4 (mean birth weights, 3600, 3511, 3481, and 3411 g; P < 0.001, for increasing quartiles of FT4).
The association weakened but remained significant when adjusting for other potential confounders of maternal smoking, BMI, parity, SES, TPOAb, and glucose (r = −0.11; P = 0.001), and when looking at only the 843 women who were TPOAb negative (r = −0.09; P = 0.002). Maternal FT4 was also negatively associated with placental weight (r = −0.16; P < 0.001) but was not associated with gestational age (r = 0.02; P = 0.6).
There was no significant difference in birth weight between babies born to TPOAb- positive mothers and TPOAb-negative mothers (3495 vs. 3575 g, respectively; P = 0.17).
Associations between maternal and cord thyroid function tests
Table 2 shows the median (IQR) for maternal and cord TSH, FT3, and FT4 values and correlations between all the maternal and cord thyroid function tests. Maternal and cord FT4 results were correlated (r = 0.14; P = 0.001), with a 10% increase in maternal FT4 leading to a 0.18-pmol/liter increase in cord FT4 [β*ln(1.1) = 0.18 (95% CI, 0.08, 0.29)]. There were also weaker negative associations between maternal TSH and both cord FT4 (r = −0.09; P = 0.02) and cord FT3 (r = −0.10; P = 0.02).
Table 2.
Correlations between maternal and cord thyroid function tests (n = 616), adjusted for sex and gestational age
| Maternal TSHa, 1.86 (1.38, 2.50) |
Maternal FT4a, 12.03 (11.19, 13.01)a |
Maternal FT3a, 4.15 (3.88, 4.47)a |
||||
|---|---|---|---|---|---|---|
| r | P | r | P | r | P | |
| Cord TSH, 8.11 (5.88, 11.29)a | 0.004 | 0.92 | −0.78 | 0.81 | −0.02 | 0.64 |
| Cord FT4, 14.33 (13.29, 15.48)a | −0.08 | 0.039 | 0.14 | 0.0004 | 0.04 | 0.31 |
| Cord FT3, 4.15 (3.88, 4.47)a | −0.10 | 0.014 | −0.02 | 0.56 | 0.04 | 0.36 |
Median (IQR) is presented for each test. Data represent term babies only. Units for TSH, FT4, and FT3 are mIU/liter, pmol/liter, and pmol/liter, respectively.
Log-transformed variables used for correlations.
Discussion
We have demonstrated, for the first time in normal healthy pregnancy, significant positive associations between FT4 in the cord blood at birth and birth weight. Every 1-pmol/liter increase in cord FT4 was associated with a 64-g increase in birth weight. The associations persist when looking at birth length, head circumference and sum of skinfold thicknesses, suggesting that fetal thyroid hormone may influence generalized growth, including skeletal size.
The association between thyroid hormone and measures of fetal growth is consistent with previous studies showing an association between overt maternal hypothyroidism and low birth weight (5, 10). Thorpe-Beeston et al. (11) found significantly lower FT4 and higher TSH in fetal blood samples from small-for-gestational-age fetuses compared with those from appropriate-for-gestational-age fetuses. In normal pregnancy, the role of thyroid hormone in fetal growth is indirectly supported by previous work showing that babies born to women with inadequate dietary iodine intake in the third trimester had lower birth weights than those born to women with adequate dietary iodine intake (12). However, maternal thyroid function in the third trimester or fetal thyroid function was not assessed in this study.
Our findings are in contrast to a previous study that found no association between neonatal FT4 levels from heel prick bloodspots 2 d after delivery and birth weight (13). However, it is possible that thyroid hormone levels measured 2 d after birth are less representative of fetal thyroid status during gestation than those measured in cord blood at birth.
Associations cannot establish cause and effect, so the correlations seen in our data could suggest either that thyroid function may influence fetal growth or the correlations may simply be a consequence of a bigger baby producing more FT4. However, the suggestion that thyroid hormones may have a direct role in regulating fetal growth is supported by animal studies showing that a lack of thyroid hormone results in abnormal skeletal development (14, 15) and in vitro studies showing expression of thyroid hormone receptors in human skeletal tissue (16), although little research has looked at the impact of thyroid hormone on fetal adiposity. It is thought that thyroid hormone binding to specific nuclear receptors results in expression of target genes to regulate skeletal development (16). In addition to this, it is well known that growth of placenta contributes to fetal growth (17), and because placental weight was positively associated with cord FT4 level in our study, it is plausible that thyroid hormone may influence fetal growth indirectly by affecting placental growth. Taken together, these observations provide strong evidence to support the role of thyroid hormone in fetal growth.
We have also presented data showing an association between maternal FT4 at 28 wk of pregnancy and fetal FT4 at birth as measured in the cord blood. This is in contrast to Oken et al. (13) who found no association between maternal FT4 levels in the first trimester and neonatal FT4 levels 2 d after birth. These contrasting findings may result from the idea that the human fetal thyroid gland starts producing thyroid hormones only from the second trimester (after 12 to 14 wk gestation) and the fetus depends upon the placental transfer of maternal thyroid hormone for its requirement before that period. It is also possible that the neonatal TSH surge, which occurs shortly after birth, could have influenced the neonatal FT4 levels in the study by Oken et al. (13). In contrast, neonatal thyroid hormone levels in cord blood taken at the time of birth are less likely to be affected by the TSH surge. The correlation between 28-wk maternal FT4 and cord FT4 in our study is consistent with the notion of maternal transfer of T4 across the placenta even during late gestation, established from studies of congenital hypothyroidism (18). However, we found no association between maternal FT3 and cord FT3. This may be partly explained by the marked placental deiodinase activity resulting in a homeostatic mechanism for maintaining T3 production in the placenta (19). The lack of association between maternal TSH and cord TSH in our study is consistent with the suggestion that TSH is poorly transferred across the placenta (20).
Given the positive associations between maternal and cord FT4 and between cord FT4 and birth weight, the negative association of maternal FT4 and birth weight was unexpected and difficult to explain. There are several possible explanations for this intriguing observation. First, we checked maternal thyroid function only once at 28 wk gestation, and weight measurements were taken at birth. Maternal thyroid function at 28 wk may not represent the thyroid status all through gestation. Second, it is possible that the negative association is caused by confounding factors affecting maternal FT4 and birth weight in different directions independently. Analysis after adjustment for a number of potential confounders including maternal smoking, BMI, parity, glucose, SES, and TPOAb weakened the association, but it remained significant (r = −0.11; P = 0.001). Finally, it is possible that, during late gestation, placental type 3 deiodinase activity leads to increased deiodination in those with high maternal FT4 to limit the impact of excess maternal thyroid hormone levels during late gestation on neonatal outcomes, including fetal growth. However, this hypothesis does not explain the positive association between maternal FT4 and cord blood FT4. Therefore, the negative association between maternal FT4 and fetal birth weight should be considered as tentative and needs confirming in further studies.
In conclusion, in this cohort of healthy, singleton, Caucasian pregnancies, we have demonstrated an association between fetal FT4 as measured in the cord blood and birth weight, highlighting the role of thyroid hormone in fetal growth and development.
Supplementary Material
Acknowledgments
This study was supported by The Wellcome Trust, the Endocrine Research Fund, and The Research and Development Directorate (Royal Devon and Exeter National Health Service Foundation Trust) and the National Institute for Health Research (NIHR). The views expressed in this publication are those of the authors and not necessarily those of the National Health Service, the NIHR, or the Department of Health, United Kingdom.
Disclosure Summary: The authors have nothing to disclose.
Footnotes
- BMI
- Body mass index
- CI
- confidence interval
- FT3
- free T3
- FT4
- free T4
- ln
- natural log
- SES
- socioeconomic status
- TPOAb
- thyroid peroxidase antibodies.
References
- 1. Obregon MJ, Calvo RM, Del Rey FE, de Escobar GM. 2007. Ontogenesis of thyroid function and interactions with maternal function. Endocr Dev 10:86–98 [DOI] [PubMed] [Google Scholar]
- 2. Haddow JE, Palomaki GE, Allan WC, Williams JR, Knight GJ, Gagnon J, O'Heir CE, Mitchell ML, Hermos RJ, Waisbren SE, Faix JD, Klein RZ. 1999. Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N Engl J Med 341:549–555 [DOI] [PubMed] [Google Scholar]
- 3. Pop VJ, Kuijpens JL, van Baar AL, Verkerk G, van Son MM, de Vijlder JJ, Vulsma T, Wiersinga WM, Drexhage HA, Vader HL. 1999. Low maternal free thyroxine concentrations during early pregnancy are associated with impaired psychomotor development in infancy. Clin Endocrinol (Oxf) 50:149–155 [DOI] [PubMed] [Google Scholar]
- 4. Millar LK, Wing DA, Leung AS, Koonings PP, Montoro MN, Mestman JH. 1994. Low birth weight and preeclampsia in pregnancies complicated by hyperthyroidism. Obstet Gynecol 84:946–949 [PubMed] [Google Scholar]
- 5. Blazer S, Moreh-Waterman Y, Miller-Lotan R, Tamir A, Hochberg Z. 2003. Maternal hypothyroidism may affect fetal growth and neonatal thyroid function. Obstet Gynecol 102:232–241 [DOI] [PubMed] [Google Scholar]
- 6. Anselmo J, Cao D, Karrison T, Weiss RE, Refetoff S. 2004. Fetal loss associated with excess thyroid hormone exposure. JAMA 292:691–695 [DOI] [PubMed] [Google Scholar]
- 7. Vaidya B, Campbell V, Tripp JH, Spyer G, Hattersley AT, Ellard S. 2004. Premature birth and low birth weight associated with nonautoimmune hyperthyroidism due to an activating thyrotropin receptor gene mutation. Clin Endocrinol (Oxf) 60:711–718 [DOI] [PubMed] [Google Scholar]
- 8. Chan SY, Vasilopoulou E, Kilby MD. 2009. The role of the placenta in thyroid hormone delivery to the fetus. Nat Clin Pract Endocrinol Metab 5:45–54 [DOI] [PubMed] [Google Scholar]
- 9. Knight B, Shields BM, Hattersley AT. 2006. The Exeter Family Study of Childhood Health (EFSOCH): study protocol and methodology. Paediatr Perinat Epidemiol 20:172–179 [DOI] [PubMed] [Google Scholar]
- 10. Idris I, Srinivasan R, Simm A, Page RC. 2005. Maternal hypothyroidism in early and late gestation: effects on neonatal and obstetric outcome. Clin Endocrinol (Oxf) 63:560–565 [DOI] [PubMed] [Google Scholar]
- 11. Thorpe-Beeston JG, Nicolaides KH, Snijders RJ, Felton CV, McGregor AM. 1991. Thyroid function in small for gestational age fetuses. Obstet Gynecol 77:701–706 [PubMed] [Google Scholar]
- 12. Alvarez-Pedrerol M, Guxens M, Mendez M, Canet Y, Martorell R, Espada M, Plana E, Rebagliato M, Sunyer J. 2009. Iodine levels and thyroid hormones in healthy pregnant women and birth weight of their offspring. Eur J Endocrinol 160:423–429 [DOI] [PubMed] [Google Scholar]
- 13. Oken E, Braverman LE, Platek D, Mitchell ML, Lee SL, Pearce EN. 2009. Neonatal thyroxine, maternal thyroid function, and child cognition. J Clin Endocrinol Metab 94:497–503 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Capelo LP, Beber EH, Huang SA, Zorn TM, Bianco AC, Gouveia CH. 2008. Deiodinase-mediated thyroid hormone inactivation minimizes thyroid hormone signaling in the early development of fetal skeleton. Bone 43:921–930 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Bassett JH, Williams AJ, Murphy E, Boyde A, Howell PG, Swinhoe R, Archanco M, Flamant F, Samarut J, Costagliola S, Vassart G, Weiss RE, Refetoff S, Williams GR. 2008. A lack of thyroid hormones rather than excess thyrotropin causes abnormal skeletal development in hypothyroidism. Mol Endocrinol 22:501–512 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Abu EO, Bord S, Horner A, Chatterjee VK, Compston JE. 1997. The expression of thyroid hormone receptors in human bone. Bone 21:137–142 [DOI] [PubMed] [Google Scholar]
- 17. Murphy VE, Smith R, Giles WB, Clifton VL. 2006. Endocrine regulation of human fetal growth: the role of the mother, placenta, and fetus. Endocr Rev 27:141–169 [DOI] [PubMed] [Google Scholar]
- 18. Vulsma T, Gons MH, de Vijlder JJ. 1989. Maternal-fetal transfer of thyroxine in congenital hypothyroidism due to a total organification defect or thyroid agenesis. N Engl J Med 321:13–16 [DOI] [PubMed] [Google Scholar]
- 19. Glinoer D. 1997. The regulation of thyroid function in pregnancy: pathways of endocrine adaptation from physiology to pathology. Endocr Rev 18:404–433 [DOI] [PubMed] [Google Scholar]
- 20. Bajoria R, Fisk NM. 1998. Permeability of human placenta and fetal membranes to thyrotropin-stimulating hormone in vitro. Pediatr Res 43:621–628 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.