Skip to content

Environmental Health Perspectives

Facebook Page EHP Twitter Feed Open Access icon  

Research Volume 125 | 2017

Environ Health Perspect; DOI:10.1289/EHP508

Fetal Thyroid Function, Birth Weight, and in Utero Exposure to Fine Particle Air Pollution: A Birth Cohort Study

Bram G. Janssen,1* Nelly D. Saenen,1* Harry A. Roels,1,2 Narjes Madhloum,1 Wilfried Gyselaers,3,4 Wouter Lefebvre,5 Joris Penders,3,6 Charlotte Vanpoucke,7 Karen Vrijens,1 and Tim S. Nawrot1,8

Author Affiliations open
1Centre for Environmental Sciences, Hasselt University, Hasselt, Belgium; 2Louvain Centre for Toxicology and Applied Pharmacology (LTAP), Université catholique de Louvain, Brussels, Belgium; 3Biomedical Research Institute, Hasselt University, Hasselt, Belgium; 4Department of Obstetrics, East-Limburg Hospital, Genk, Belgium; 5Flemish Institute for Technological Research (VITO), Mol, Belgium; 6Department of Clinical Biology, East-Limburg Hospital, Genk, Belgium; 7Belgian Interregional Environment Agency, Brussels, Belgium; 8Department of Public Health and Primary Care, Occupational and Environmental Medicine, Leuven University, Leuven, Belgium

PDF icon PDF Version (1.1 MB)

  • Background: Thyroid hormones are critical for fetal development and growth. Whether prenatal exposure to fine particle air pollution (≤ 2.5 μm; PM2.5) affects fetal thyroid function and what the impact is on birth weight in normal healthy pregnancies have not been studied yet.

    Objectives: We studied the impact of third-trimester PM2.5 exposure on fetal and maternal thyroid hormones and their mediating role on birth weight.

    Methods: We measured the levels of free thyroid hormones (FT3, FT4) and thyroid-stimulating hormone (TSH) in cord blood (n = 499) and maternal blood (n = 431) collected after delivery from mother–child pairs enrolled between February 2010 and June 2014 in the ENVIRONAGE birth cohort with catchment area in the province of Limburg, Belgium.

    Results: An interquartile range (IQR) increment (8.2 μg/m3) in third-trimester PM2.5 exposure was inversely associated with cord blood TSH levels (–11.6%; 95% CI: –21.8, –0.1) and the FT4/FT3 ratio (–62.7%; 95% CI: –91.6, –33.8). A 10th–90th percentile decrease in cord blood FT4 levels was associated with a 56 g decrease in mean birth weight (95% CI: –90, –23). Assuming causality, we estimated that cord blood FT4 mediated 21% (–19 g; 95% CI: –37, –1) of the estimated effect of an IQR increment in third-trimester PM2.5 exposure on birth weight. Third-trimester PM2.5 exposure was inversely but not significantly associated with maternal blood FT4 levels collected 1 day after delivery (–4.0%, 95% CI: –8.0, 0.2 for an IQR increment in third-trimester PM2.5).

    Conclusions: In our study population of normal healthy pregnancies, third-trimester exposure to PM2.5 air pollution was associated with differences in fetal thyroid hormone levels that may contribute to reduced birth weight. Additional research is needed to confirm our findings in other populations and to evaluate potential consequences later in life.

  • Citation: Janssen BG, Saenen ND, Roels HA, Madhloum N, Gyselaers W, Lefebvre W, Penders J, Vanpoucke C, Vrijens K, Nawrot TS. 2017. Fetal thyroid function, birth weight, and in utero exposure to fine particle air pollution: a birth cohort study. Environ Health Perspect 125:699–705;

    *These authors contributed equally to this work.

    Address correspondence to T.S. Nawrot, Centre for Environmental Sciences, Hasselt University, Agoralaan gebouw D, 3590 Diepenbeek, Belgium. Telephone: 32-11-268382. E-mail:

    The authors thank the participating women and neonates, as well as the staff of the maternity ward, midwives, and the staff of the clinical laboratory of East-Limburg Hospital in Genk.

    The ENVIRONAGE birth cohort is supported by the European Research Council (ERC-2012-StG.310898), by the Flemish Scientific Fund (FWO, G073315N/G082317N) and the Bijzonder Onderzoeksfonds (BOF) of Hasselt University. K.V. is a postdoctoral fellow of the FWO.

    None of the funding agencies had a role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.

    The authors declare they have no actual or potential competing financial interests.

    Received: 25 January 2016
    Revised: 8 August 2016
    Accepted: 16 August 2016
    Published: 13 September 2016

    Note to readers with disabilities: EHP strives to ensure that all journal content is accessible to all readers. However, some figures and Supplemental Material published in EHP articles may not conform to 508 standards due to the complexity of the information being presented. If you need assistance accessing journal content, please contact Our staff will work with you to assess and meet your accessibility needs within 3 working days.

  • PDF icon Supplemental Material PDF (440 KB)

    Note to readers with disabilities: EHP has provided a 508-conformant table of contents summarizing the Supplemental Material for this article (see below) so readers with disabilities may determine whether they wish to access the full, nonconformant Supplemental Material. If you need assistance accessing journal content, please contact Our staff will work with you to assess and meet your accessibility needs within 3 working days.

    PDF icon Supplemental Table of Contents PDF (124 KB)


During prenatal life, thyroid hormones are critical for fetal growth and development, especially neurodevelopment (Burrow et al. 1994; Morreale de Escobar et al. 2004). Unbalanced thyroid function influences pregnancy outcomes and adversely affects the fetus. In particular, both maternal hypo- and hyperthyroidism are associated with increased risk of low birth weight (Blazer et al. 2003; Millar et al. 1994), whereas other studies also suggest an important role of fetal thyroid function in regulating fetal growth (Medici et al. 2013; Shields et al. 2011).

Thyroxine (T4), the major form of thyroid hormone, and triiodothyronine (T3), the active form, are controlled by thyroid-stimulating hormone (TSH) and released by the thyroid gland. Bound to plasma proteins, these thyroid hormones are transported throughout the body and diffuse from maternal blood across the placenta to reach the fetus (Calvo et al. 2002). However, it is the unbound, free fractions of these hormones (FT4 and FT3) that are taken up by different cell types to regulate their functioning (Hennemann et al. 2001). From the second trimester of gestation onward, the fetal thyroid gland becomes functional, and the fetus is able to produce its own supply of thyroid hormones in addition to the maternal supply (Morreale de Escobar et al. 2004).

Findings from previous studies suggest that airborne persistent organic pollutants (Abdelouahab et al. 2013; Baccarelli et al. 2008; Maervoet et al. 2007), cadmium (Iijima et al. 2007), and exposure to active and passive cigarette smoke (Soldin et al. 2009) may affect thyroid hormone regulation and function in neonates and adults; however, epidemiological studies on the impact of exposure to particulate matter (PM) air pollution on thyroid hormones are lacking. In large areas of the world, PM air pollution is an omnipresent environmental risk factor of public health concern, especially the fine particles with an aerodynamic diameter ≤ 2.5 μm (PM2.5). Exposure to ambient PM2.5 pollution during pregnancy has been found to be significantly associated with increased risk of low birth weight at term in mother–child cohorts of 12 European countries (Pedersen et al. 2013) and preterm birth (20–36 weeks of gestation) in a very large cohort of singleton pregnancies from three states of the United States (Rappazzo et al. 2014).

Despite the well-established link between PM2.5 air pollution and adverse gestational outcome (Pedersen et al. 2013), the role of fetal thyroid function in this association has never been investigated. Therefore, we hypothesized that airborne PM2.5 exposure during gestation affects fetal thyroid hormone function in normal healthy pregnancies and contributes to reduced birth weight. We tested this hypothesis in the framework of a mother–child cohort by studying the impact of third-trimester PM2.5 exposure on fetal and maternal thyroid hormone function, as reflected by the levels of FT3, FT4, and TSH in cord blood and maternal blood, and their mediating role on birth weight.


Study Population

From February 2010 through June 2014, we recruited 640 mother–child pairs after delivery at the East-Limburg Hospital in Genk, Belgium. They were enrolled in the on-going ENVIRONAGE birth cohort study (ENVIRonmental influence ON early AGEing) following procedures previously approved by the Ethical Committee of Hasselt University and the East-Limburg Hospital (Janssen et al. 2012). The study was conducted according to the principles outlined in the Declaration of Helsinki for investigation of human subjects. The participation rate of eligible mothers in the birth cohort (mothers able to fill out a Dutch language questionnaire and those without planned cesarean section) was 61%, and enrollment was spread equally over all seasons of the year. Midwives recorded the reason for nonparticipation. The main reasons (in descending importance) were failure to ask for participation, communication problems, or complications during labor. Participating mothers provided written informed consent when they arrived at the hospital for delivery. They completed study questionnaires in the postdelivery ward to provide detailed information on maternal age, prepregnancy body mass index (BMI), maternal education, occupation, self-reported smoking status, alcohol consumption, place of residence, use of medication, parity, and newborn’s ethnicity. Former smokers were defined as those who had quit smoking before pregnancy. Smokers continued smoking during pregnancy. Based on the native country of the newborn’s grandparents, we classified his/her ethnicity as European-Caucasian when two or more grandparents were European, or non-European when at least three grandparents were of non-European origin. Maternal education was coded as “low” (no diploma or primary school), “middle” (high school), or “high” (college or university degree). After birth, we collected perinatal parameters from the medical files such as birth date, gestational age, newborn’s sex, birth weight and length, length of labor, Apgar score, pH of arterial cord blood, and ultrasonographic data.

The main analysis of our investigation was conducted in a subcohort of the 640 singleton pregnancies in the ENVIRONAGE birth cohort, after excluding 16 mothers with hyper- or hypothyroidism, 79 mothers from whom we had no complete set of cord blood thyroid hormone values, 28 cesarean sections, and 18 preterm births (< 37 weeks), leaving 499 mother–child pairs for the main analysis (Figure 1). Additionally, maternal blood could not be collected from 68 mothers, resulting in a study population of 431 for the maternal thyroid hormone analysis (mother group) (Figure 1). Our study population was generally similar to all births in Flanders [data obtained from the Study Centre for Perinatal Epidemiology (SPE)] as to maternal age, education, parity, sex, ethnicity, and birth weight (see Table S1) (Cox et al. 2013).

Figure 1. Flow chart of study participation and eligibility.

Figure 1. Flow chart depicting the selection procedure of study participants from the ENVIRONAGE birth cohort, Limburg, Belgium.

View larger image (TIF File)

Ambient Exposure Assessment

For each mother’s residential address, we interpolated the regional background PM2.5 (μg/m3) using a spatial temporal interpolation method (kriging method) (Janssen et al. 2008) that uses pollution data collected by the official fixed-site monitoring network and land-cover data obtained from satellite images (CORINE land-cover data set) in combination with a dispersion model (Lefebvre et al. 2011). This model chain provides daily PM2.5 values using data from the Belgian telemetric air quality network and point and line sources, which are then interpolated in a high-resolution receptor grid. In the Flemish region of Belgium, the interpolation tool explained > 80% of the temporal and spatial variability (Maiheu et al. 2013). We defined the third trimester of pregnancy as from week 27 to delivery and calculated the mean PM2.5 values for this trimester. The date of conception was estimated on the basis of the first day of the mother’s last menstrual period, combined with the first ultrasound exam. Complete information for the residential address during pregnancy was obtained by questionnaire and checked with hospital records. For those who moved residence during pregnancy (n = 54; 10.8%), we calculated the third-trimester exposure window allowing for the changes in address during this period.

Mean daily temperatures and relative humidity for the study region were provided by the Royal Meteorological Institute (Brussels, Belgium). We calculated the third-trimester apparent temperature by using the following formula (Kalkstein and Valimont 1986; Steadman 1979): –2.653 + (0.994 × Ta) + (0.0153 × Td2), where Ta is air temperature and Td is dew point temperature (degrees Celsius).

Blood Collection and Thyroid Hormone Measurements

Umbilical cord and maternal blood samples (8 mL each) were collected in plastic BD Vacutainer® Lithium Heparin Tubes (BD, Franklin Lakes, NJ, USA) immediately after delivery and 1 day after delivery, respectively. The samples were centrifuged (3,200 rpm for 15 min) to retrieve plasma which was instantly frozen at –80°C. The plasma levels of FT4 (pmol/L), FT3 (pmol/L), and TSH (mU/L) were measured with an electro-chemiluminescence immunoassay using the Modular E170 automatic analyzer (Roche, Basel, Switzerland) at the clinical laboratory of East-Limburg Hospital.

Statistical Analysis

For database management and statistical analysis, we used the SAS software program (version 9.2; SAS Institute Inc., Cary, NC, USA). Thyroid hormone levels were log10-transformed to improve the normality of the distributions and described by geometric mean and 10th–90th percentile. The ratio FT4/FT3 was calculated using untransformed values and had a normal distribution. Pearson correlation coefficients were calculated among the different thyroid hormone levels in blood (FT3, FT4, and TSH) and between thyroid hormone levels and birth weight. We performed multiple linear regressions to assess the associations between newborn or maternal thyroid hormones and PM2.5 exposure during third trimester of gestation, and between newborn or maternal thyroid hormones and birth weight. Exposures to PM2.5 were fitted as linear variables in the models, and effect estimates on thyroid hormones were calculated for an interquartile range (IQR) increment in PM2.5. The effect estimates of cord blood FT4 on birth weight were calculated for a 10th–90th percentile decrease in FT4, which corresponds to an 11% difference in FT4. All cord blood models were adjusted for sex, gestational age (weeks), season of delivery [winter (21 December–20 March)/spring (21 March–20 June)/summer (21 June–20 September)/autumn (21 September–20 December)], Apgar score (< 9/9/10), maternal age (years), prepregnancy BMI (kg/m2), smoking status (never-smoker/cessation before pregnancy/smoker), parity (1/2/≥ 3), ethnicity (European-Caucasian, yes or no), maternal education (low/middle/high), and third-trimester apparent temperature (°C), and all models for maternal blood were adjusted for the same covariates except newborn’s sex and Apgar score. In an additional analysis, we adjusted the cord blood models for maternal thyroid hormones. The Shapiro–Wilk statistic and Q-Q plots of the residuals were used to test the assumptions of model linearity.

We used mediation analysis to investigate potential associations that may underlie the relation between the exposure variable (PM2.5) and the continuous outcome variable (birth weight, g) by examining how they relate to a third variable, the mediator (cord blood FT4 levels) (Valeri and Vanderweele 2013). We accomplished this by decomposing the total effect into a direct effect (DE; exposure effect on outcome at a fixed level of the mediator) and an indirect effect (IE; exposure effect on outcome that operates through the mediator). Mediation analysis is based on several assumptions: All associations are causal, with no uncontrolled confounding of associations between the exposure and mediator, the exposure and the outcome, or the mediator and the outcome; no measured mediator–outcome confounder is affected by exposure; and no interaction occurs between the exposure and mediator (Valeri and Vanderweele 2013).

Sensitivity Analysis

Tobacco smoke exposure, a form of personalized airborne PM exposure, has been shown to influence maternal and fetal thyroid function through changes in thyroid hormone levels (Männistö et al. 2012; McDonald et al. 2008; Shields et al. 2009; Soldin et al. 2009). In a sensitivity analysis, we performed linear regression analysis to examine the associations between newborn or maternal thyroid hormones and smoking, adjusting for the same co-variables as mentioned above except smoking. Additionally, we repeated the analysis between cord blood thyroid hormones and PM2.5 exposure while excluding smokers.

Thyroid hormones may also show seasonal variations linked to changes in temperature (Reed 1995). To account for possible seasonal differences between subjects, we calculated for each subject an exposure window covering a 1-year period: 365 days calculated backward from the date of delivery.

Furthermore, it is known that cord blood thyroid hormone levels are influenced by different external factors. We explored whether covariates such as cord plasma estradiol (Lv et al. 2014), passive smoking (Soldin et al. 2009), alcohol consumption (Herbstman et al. 2008), pH of arterial cord blood (Chan et al. 2001), or length of labor (Parate et al. 2010), known for their interference with thyroid hormones, may alter the association between cord blood thyroid hormones and third-trimester PM2.5 exposure.


Demographics of Participants

Table 1 shows demographic characteristics and perinatal traits of the mother–child group (n = 499). Mean (10th–90th percentile) maternal age was 29.1 years (23–35) and mean prepregnancy BMI was 23.9 (19.6–29.8) kg/m2. Most women never smoked (n = 316), 113 stopped smoking before pregnancy, and 70 mothers reported to continue with smoking during pregnancy (on average 8.6 cigarettes/day). More than 80% of the mothers reported no consumption of alcoholic beverages during pregnancy. The newborns, among them 254 girls (50.8%), had a mean gestational age of 39.4 weeks (38–41) and comprised 275 primiparous and 170 secundiparous newborns. About 90% of the newborns were Europeans of Caucasian ethnicity. The mean birth weight of the newborns was 3,446 (2,915–3,990) g. Five minutes after delivery, > 90% of the newborns had an Apgar score ≥ 9.

Table 1. Select View Table (HTML Version) for a 508-conformant version

Table 1. Characteristics and thyroid hormone levels of the mother–child pairs (n = 499).

View Table (HTML Version)
View larger image (TIF File)

Thyroid Hormone Levels in Cord Blood and Maternal Blood

The geometric means of thyroid hormone levels in cord blood (n = 499) were 10.3 mU/L for TSH, 2.5 pmol/L for FT3, and 15.7 pmol/L for FT4, whereas in maternal blood (n = 431) it was 2.1 mU/L, 4.2 pmol/L, and 12.5 pmol/L respectively (Table 1). A positive correlation was observed between FT3 and FT4 (cord blood: r = 0.30; p < 0.0001; maternal blood: r = 0.27; p < 0.0001) and between FT3 and TSH (cord blood: r = 0.11; p = 0.01; maternal blood: r = 0.19; p < 0.0001). Maternal FT4 levels were positively correlated with cord blood FT4 levels (r = 0.21; p < 0.0001), whereas an inverse correlation was observed with cord blood FT3 levels (r = –0.11; p = 0.01). Compared with maternal values, the measured cord blood FT3 levels were approximately 2-fold lower and the TSH levels much higher. The thyroid hormone concentrations in cord blood were similar to values published by others (Abdelouahab et al. 2013).

Ambient Exposure Levels

Average (25th–75th percentile) PM2.5 exposure and apparent temperature for the third gestational trimester were respectively 16.0 μg/m3 (11.6–19.8) and 8.7°C (3.2–14.7). Mean levels of both parameters were similar throughout the trimesters of pregnancy (data not shown).

Thyroid Hormones and PM2.5 Exposure During Gestation

In cord blood (n = 499), TSH levels and FT4/FT3 ratios correlated inversely with PM2.5 exposure during the third trimester of pregnancy (Figure 2). After adjustment for sex, gestational age, season of delivery, Apgar score, maternal age, prepregnancy BMI, smoking status, parity, ethnicity, maternal education, and apparent temperature, an IQR increment (8.2 μg/m3) in PM2.5 exposure during the third trimester was associated with a lowering of 11.6% [95% confidence interval (CI): –21.8, –0.1; p < 0.05] in cord blood TSH levels (Figure 3A) and a lowering of 62.7% (95% CI: –91.6, –33.8; p < 0.0001) in cord blood FT4/FT3 ratio (Figure 3A). Considering the FT4 and FT3 levels in cord blood separately (Figure 3B), we observed opposite associations for an IQR increment of PM2.5 exposure on these two hormones during the third trimester (FT4, –3.7%; 95% CI: –6.4, –0.9; p = 0.009, and FT3, +6.4%; 95% CI: 1.8, 11.1; p = 0.006). Additional adjustment for maternal thyroid hormones in the cord blood models did not alter our findings for cord blood (data not shown).

Figure 2. Two scatterplots with regression lines for cord blood TSH and cord blood FT4 to FT3 ratio, respectively (y-axes) according to third trimester PM2.5 exposure (x-axes).

Figure 2. Unadjusted correlation between third-trimester PM2.5 exposure (μg/m3) and cord blood TSH (mU/L, log10) levels (left) and the FT4/FT3 ratio (right).

View larger image (TIF File)

Figure 3. A) Two plots of estimated percentage differences (with 95% CI) in TSH and FT4 to FT3 ratios, respectively (y-axes) in cord blood and maternal blood, respectively (x-axes). B) Two plots of estimated percentage differences (with 95% CI) in FT4 and FT3, respectively (y-axes) in cord blood and maternal blood, respectively (x-axes).

Figure 3. Difference in cord and maternal blood thyroid hormones in association with third-trimester PM2.5. The estimated relative difference in percentage (95% CI) is calculated for an IQR increment (8.2 μg/m3) in third-trimester PM2.5 exposure. Panel A displays the difference in TSH (left) and the difference in FT4/FT3 ratio (right). Panel B displays the difference in FT4 (left) and FT3 (right). The cord blood models were adjusted for sex, gestational age, season of delivery, Apgar score, maternal age, prepregnancy BMI, smoking status, parity, ethnicity, maternal education, and third-trimester apparent temperature, whereas for maternal blood sex and Apgar score were excluded.

*p < 0.05. **p < 0.01. ***p < 0.001.

View larger image (TIF File)

In maternal blood (n = 431), TSH and FT4 levels correlated inversely with third-trimester PM2.5 exposure (r = –0.10; p = 0.04 and r = –0.13; p = 0.005 respectively). After adjustment for gestational age, season of delivery, maternal age, prepregnancy BMI, smoking status, parity, ethnicity, maternal education, and apparent temperature, only maternal FT4 levels were inversely but not significantly associated with an IQR increment in third-trimester PM2.5 exposure (–4.0%; 95% CI: –8.0, 0.2; p = 0.06) (Figure 3B). Neither TSH nor the FT4/FT3 ratio in maternal blood showed a significant difference with an IQR increment in third-trimester PM2.5 exposure.

Thyroid Hormones and Birth Weight

After adjustment for gestational age and sex, neither FT3 nor TSH levels in maternal or cord blood were associated with birth weight (p ≥ 0.47). However, a 10th–90th percentile decrease (11%) in cord blood FT4 (log10 values) was associated with a lowering in birth weight of 71 g (95% CI: –103, –38; p < 0.0001). After additional adjustment for maternal age, prepregnancy BMI, smoking status, parity, season of delivery, Apgar score, ethnicity, maternal education, and apparent temperature, the association for the cord blood model remained significant (–56 g; 95% CI: –90, –23; p = 0.001). In contrast, a 10th–90th percentile decrease (15%) in maternal FT4 was positively associated with birth weight (45 g; 95% CI: 6, 83; p = 0.02), though the association was weaker and no longer significant after full adjustment (31 g; 95% CI: –7, 69; p = 0.11).

We performed mediation analysis to estimate the proportion of the PM2.5 exposure effect on birth weight as mediated by cord blood FT4. Although we did not observe a significant association between third-trimester PM2.5 exposure and birth weight (p = 0.70), there is consensus among statisticians that the relationship between exposure (e.g., PM2.5) and outcome (e.g., birth weight) does not need to be statistically significant for a variable (e.g., FT4) to be a mediator (Valeri and Vanderweele 2013). Assuming causality, adjusted estimates of the proportion of mediation suggest that cord blood FT4 levels explained 21% (indirect effect: –19 g; 95% CI: –37, –1; p = 0.03) of the association between the third-trimester IQR PM2.5 exposure and birth weight (Figure 4). Because maternal thyroid hormones did not meet the assumptions for mediation (no association between maternal FT4 and birth weight), we did not perform a mediation analysis.

Figure 4. Diagram showing the estimated proportion of the association between PM2.5 and birth weight that was mediated by cord blood FT4, assuming that assumptions of the mediation analysis were valid.

Figure 4. Estimated proportion of the association between PM2.5 exposure and birth weight (g) mediated by cord blood FT4 levels. The figure displays cord blood FT4 levels as mediator, the estimate of the indirect effect (IE), the estimate of the direct effect (DE), and the proportion of mediation (IE/DE + IE). The estimated effect is calculated for an IQR increment (8.2 μg/m3) in PM2.5 exposure during the third trimester of pregnancy. The mediation model was adjusted for sex, gestational age, season of delivery, Apgar score, maternal age, prepregnancy BMI, smoking status, parity, ethnicity, maternal education, and third-trimester apparent temperature.

View larger image (TIF File)

Sensitivity Analysis

After adjustment for newborn’s sex, gestational age, season of delivery, Apgar score, maternal age, prepregnancy BMI, parity, ethnicity, and apparent temperature, cord blood TSH levels were lower in mothers who continued smoking during pregnancy (–18.7%; 95% CI: –29.1, –6.7; p = 0.003) and also in those who stopped smoking before pregnancy (–10.3%; 95% CI: –19.6, 0.1; p = 0.05) in comparison with never-smoking mothers (see Table S2). When excluding never-smokers, we observed, as expected, an inverse association between smoking years and TSH levels in cord blood (r = –0.21; p = 0.004; n = 173). We did not find differences in cord blood FT4 levels between smokers and former smokers compared with nonsmokers. Newborns from women who stopped smoking before pregnancy had slightly lower FT3 cord blood levels (–3.7%; 95% CI: –7.5, 0.1; p = 0.06), but, surprisingly, those who continued smoking had higher levels of cord blood FT3 (3.7%; 95% CI: –1.3, 9.0; p = 0.15) compared with nonsmokers. We did not find an association between maternal thyroid hormones and smoking status during pregnancy (p ≥ 0.31). The associations between cord blood thyroid hormones and PM2.5 exposures did not alter when women who smoked during pregnancy were excluded (data not shown).

The analysis to account for seasonal differences invariably showed for an IQR increment (3.7 μg/m3) of PM2.5 during a 1-year period inverse associations with cord blood TSH levels (–8.8%; 95% CI: –14.8, –2.4; p = 0.008), the FT4/FT3 ratio (–32.7%; 95% CI: –48.8, –16.6; p < 0.0001), and the FT4 levels (–1.7%; 95% CI: –3.2, –0.1; p = 0.03), and a positive association with FT3 levels (3.5%; 95% CI: 1.0, 6.0; p = 0.006), corroborating the associations found for the third trimester of pregnancy (Figure 3).

Additional adjustments for cord blood plasma estradiol (n = 498), passive indoor tobacco smoke exposure (n = 486), alcohol consumption (n = 485), or pH of arterial cord blood (indicator of hypoxemia) (n = 431) did not alter the associations between PM2.5 exposure and FT3 and FT4 thyroid hormones shown in the main analysis (see Table S3). Length of labor may influence thyroid hormone levels possibly due to the high energy demand during labor (Parate et al. 2010). Adjusting the main models for length of labor (n = 427) also did not alter the reported associations except that the association between cord blood TSH levels and third-trimester PM2.5 exposure was no longer significant (see Figure S1).


To the best of our knowledge, our study is the first to show associations between airborne PM2.5 exposure and cord blood thyroid hormones. A key finding is that PM2.5 exposure during the third trimester of gestation is inversely associated with TSH levels and the FT4/FT3 ratio in cord blood, but not with thyroid hormones in maternal blood. The FT4/FT3 ratio in cord blood is a useful indicator of how effectively the body is able to convert T4 into T3 (Yoshimura Noh et al. 2005). In addition, results of the mediation analysis suggested that cord blood FT4 is a partial mediator of the association between third-trimester pregnancy PM2.5 exposure and birth weight, assuming the underlying causal assumptions of mediation analysis are valid. Our findings highlight the potential influence of early-life environmental exposure to PM2.5 on fetal thyroid function and fetal growth. In addition, our results remained robust in multiple sensitivity analyses comprising maternal tobacco smoking, passive indoor smoking, seasonal variations, alcohol consumption, fetal hypoxemia, maternal estrogen levels, and length of labor.

During pregnancy, thyroid hormones regulate metabolism, stimulate differentiation and growth of the fetus, and influence neurocognitive development (Burrow et al. 1994; Morreale de Escobar et al. 2004). Despite the fact that the fetus starts secreting small amounts of thyroid hormone from mid-gestation onward (Thorpe-Beeston et al. 1991), the mother already supplies thyroid hormones to the fetal circulation from the first trimester without compromising her own supply (Vulsma et al. 1989). The rise of maternal thyroid hormones in the first trimester of pregnancy is considered critical to ensure normal (neurological) development (Morreale de Escobar et al. 2004). Maternal T4 and T3 diffuse across the placenta to reach concentrations in the fetus that are in the same range as those in adult tissues (Calvo et al. 2002). Fetal T3 is generated locally from T4 by type 2 deiodinase, has a high affinity for nuclear binding sites in the placenta, and stimulates the production of factors that control trophoblast growth and development (Maruo et al. 1991). This suggests that thyroid hormones play an important role in normal placentation and development of the fetus. Shields et al. (2011) showed in women with normal healthy pregnancies that placental weight was positively associated with cord blood FT4 levels and inferred that thyroid hormones may influence fetal growth indirectly by affecting placental growth. These authors found that lower FT4 levels in cord blood were associated with reduced birth weight, and their results are corroborated by our study and two studies from the Netherlands (Korevaar et al. 2016; Medici et al. 2013). Moreover, our study estimated that during the third trimester of pregnancy the estimated effect of PM2.5 exposure on birth weight was for 21% (on average –19 g) mediated by cord blood FT4 levels. As in all observational studies, these estimates should be interpreted with caution because the underlying assumptions of causality between each pair of factors in the mediation analysis cannot be verified. Nevertheless, this finding suggests that the third trimester of pregnancy, when the fetus significantly increases in size, is an important window of susceptibility to PM2.5 exposure. Shields et al. (2011), Medici et al. (2013), and León et al. (2015), as well as our study, report an inverse association between maternal FT4 and birth weight, which is opposite to cord blood FT4. In a study of pregnant women without history of thyroid dysfunction, it has been shown that lower concentrations of FT4 in maternal blood were related with increased placental growth (Bassols et al. 2011). These observations together suggest a functional discrepancy for FT4 between maternal and fetal blood, especially with regard to fetal growth. In our study, we observed an inverse but no significant association (p = 0.06) between maternal FT4 and third-trimester PM2.5 exposure in accordance with our findings in cord blood.

Contrary to maternal T4 and T3, perfusion experiments with TSH on human term placentas have shown that TSH crosses placental tissue and fetal membranes only sparingly (Bajoria and Fisk 1998). Hence, our finding of an inverse association between cord blood TSH levels and PM2.5 exposure during pregnancy suggests a potential effect of PM2.5 on fetal thyroid function. Experimental studies showed that PM exposure in healthy rats modulates the hypothalamic–pituitary–thyroid axis and leads to increases in markers of glucocorticoid activity (Thomson et al. 2013), which are known to suppress TSH release (Wilber and Utiger 1969). In the context of anti-inflammatory actions of glucocorticoids, previous findings from our birth cohort suggest that ambient PM2.5 exposure may induce a systemic oxidative stress response (Janssen et al. 2012) and increase placental protein-bound 3-nitrotyrosine (Saenen et al. 2016).

The FT4/FT3 ratio in cord blood, a useful indicator of how effectively the body is able to convert T4 into T3 (Bassols et al. 2011), was inversely associated with PM2.5 during pregnancy. This finding could be explained by the fact that placental type 2 deiodinase activity increases when the availability of T4 decreases, thus representing a potential homeostatic mechanism for maintaining T3 production when T4 concentrations are reduced (Glinoer 2004). In a population of 4,837 euthyroid Finnish mothers, Männistö et al. (2012) observed that mothers who smoked before, or continued smoking during the first trimester of pregnancy, had reduced blood levels of FT4 and increased levels of FT3 compared with nonsmokers. Constituents of tobacco smoke may stimulate the conversion of T4 to T3 in tissues by boosting type 2 deiodinase activity, as shown in cultured rat brain glial cells (Gondou et al. 1999). Low levels of TSH and FT4 are suggestive of central hypothyroidism (defect of thyroid hormone production due to insufficient stimulation by TSH of an otherwise normal thyroid gland) (Persani 2012). Recently, it has been shown that intrauterine exposure to insufficient maternal thyroid hormone levels, characterized by low levels of FT4 coexisting with reference TSH levels, was associated with higher scores for attention deficit/hyperactivity disorder (ADHD) symptoms in 127 children at 8 years of age of a population-based birth cohort in the Netherlands (Modesto et al. 2015). Additional research is needed to confirm our findings in other populations and to evaluate potential consequences later in life.

Our study has some limitations. First, thyroid hormones are responsive to environmental temperature (Reed 1995) and show a seasonal pattern, with lower values in the cold period than in the warm period of the year. Nevertheless, our results were robust for both seasonal differences between subjects as well as adjustment for third-trimester apparent temperature. Second, iodine is required for the synthesis of thyroid hormones, but we did not have information on iodine levels in maternal or cord blood. However, we excluded a priori clinically confirmed cases of hypo- and hyperthyroidism. Last, we acknowledge the fact that we cannot fully exclude residual or unmeasured confounding by other factors such as noise, polychlorinated biphenyls, heavy metals, or pesticides that could be associated with both ambient air pollution and thyroid function. Overall, the characteristics of the ENVIRONAGE birth cohort were generally similar compared with all births in Flanders, except that we excluded cesarean sections and preterm births, so our findings might be generalizable to the gestational segment of the population at large (see Table S1). We used a standardized fine-scale exposure model for the estimation of residential fine particle air pollution levels of the pregnant mothers (on average 16.0 μg/m3) which are comparable with other European and U.S. cohorts with mean PM2.5 exposure values amounting to 18.5 (Pedersen et al. 2013) and 14.5 μg/m3 (Rappazzo et al. 2014), respectively.


Our epidemiological finding of differences in fetal thyroid function in association with PM2.5 exposure is in line with the known effects of cigarette smoking on thyroid function during pregnancy (Männistö et al. 2012). Although confirmation in other study populations is needed, our findings suggest that cord blood FT4 may play a mediating role between PM2.5 exposure and birth weight during late pregnancy. The potential mechanisms and possible later-in-life adverse consequences are far from elucidated. Our findings are of critical public health importance because of the ubiquity of fine PM air pollution and the possibility of long-term health consequences of early-life alterations in thyroid function. Therefore, to promote a healthier living environment for children, our findings support a down-revision of the current European Union air pollution limit for PM2.5 of 25 μg/m3 (annual average threshold) in the direction of the World Health Organization–recommended limit of 10 μg/m3 (annual average) (World Health Organization 2006).


Abdelouahab N, Langlois MF, Lavoie L, Corbin F, Pasquier JC, Takser L. 2013. Maternal and cord-blood thyroid hormone levels and exposure to polybrominated diphenyl ethers and polychlorinated biphenyls during early pregnancy. Am J Epidemiol 178(5):701–713.

Baccarelli A, Giacomini SM, Corbetta C, Landi MT, Bonzini M, Consonni D, et al. 2008. Neonatal thyroid function in Seveso 25 years after maternal exposure to dioxin. PLoS Med 5(7):e161, doi: 10.1371/journal.pmed.0050161.

Bajoria R, Fisk NM. 1998. Permeability of human placenta and fetal membranes to thyrotropin-stimulating hormone in vitro. Pediatr Res 43(5):621–628.

Bassols J, Prats-Puig A, Soriano-Rodríguez P, Garcia-González MM, Reid J, Martínez-Pascual M, et al. 2011. Lower free thyroxin associates with a less favorable metabolic phenotype in healthy pregnant women. J Clin Endocrinol Metab 96(12):3717–3723.

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(2):232–241.

Burrow GN, Fisher DA, Larsen PR. 1994. Maternal and fetal thyroid function. N Engl J Med 331(16):1072–1078.

Calvo RM, Jauniaux E, Gulbis B, Asunción M, Gervy C, Contempré B, et al. 2002. Fetal tissues are exposed to biologically relevant free thyroxine concentrations during early phases of development. J Clin Endocrinol Metab 87(4):1768–1777.

Chan LYS, Leung TN, Lau TK. 2001. Influences of perinatal factors on cord blood thyroid-stimulating hormone level. Acta Obstet Gynecol Scand 80(11):1014–1018.

Cox B, Martens E, Nemery B, Vangronsveld J, Nawrot TS. 2013. Impact of a stepwise introduction of smoke-free legislation on the rate of preterm births: analysis of routinely collected birth data. BMJ 346:f441, doi: 10.1136/bmj.f441.

Glinoer D. 2004. The regulation of thyroid function during normal pregnancy: importance of the iodine nutrition status. Best Pract Res Clin Endocrinol Metab 18(2):133–152.

Gondou A, Toyoda N, Nishikawa M, Yonemoto T, Sakaguchi N, Tokoro T, et al. 1999. Effect of nicotine on type 2 deiodinase activity in cultured rat glial cells. Endocr J 46(1):107–112.

Hennemann G, Docter R, Friesema ECH, de Jong M, Krenning EP, Visser TJ. 2001. Plasma membrane transport of thyroid hormones and its role in thyroid hormone metabolism and bioavailability. Endocr Rev 22(4):451–476.

Herbstman J, Apelberg BJ, Witter FR, Panny S, Goldman LR. 2008. Maternal, infant, and delivery factors associated with neonatal thyroid hormone status. Thyroid 18(1):67–76.

Iijima K, Otake T, Yoshinaga J, Ikegami M, Suzuki E, Naruse H, et al. 2007. Cadmium, lead, and selenium in cord blood and thyroid hormone status of newborns. Biol Trace Elem Res 119(1):10–18.

Janssen S, Dumont G, Fierens F, Mensink C. 2008. Spatial interpolation of air pollution measurements using CORINE land cover data. Atmos Environ 42(20):4884–4903.

Janssen BG, Munters E, Pieters N, Smeets K, Cox B, Cuypers A, et al. 2012. Placental mitochondrial DNA content and particulate air pollution during in utero life. Environ Health Perspect 120:1346–1352, doi: 10.1289/ehp.1104458.

Kalkstein LS, Valimont KM. 1986. An evaluation of summer discomfort in the United States using a relative climatological index. Bull Am Meteorol Soc 7:842–848.

Korevaar TI, Chaker L, Jaddoe VW, Visser TJ, Medici M, Peeters RP. 2016. Maternal and birth characteristics are determinants of offspring thyroid function. J Clin Endocrinol Metab 101(1):206–213.

Lefebvre W, Vercauteren J, Schrooten L, Janssen S, Degraeuwe B, Maenhaut W, et al. 2011. Validation of the MIMOSA-AURORA-IFDM model chain for policy support: modeling concentrations of elemental carbon in Flanders. Atmos Environ 45(37):6705–6713.

León G, Murcia M, Rebagliato M, Álvarez-Pedrerol M, Castilla AM, Basterrechea M, et al. 2015. Maternal thyroid dysfunction during gestation, preterm delivery, and birthweight. The Infancia y Medio Ambiente cohort, Spain. Paediatr Perinat Epidemiol 29(2):113–122.

Lv PP, Meng Y, Lv M, Feng C, Liu Y, Li JY, et al. 2014. Altered thyroid hormone profile in offspring after exposure to high estradiol environment during the first trimester of pregnancy: a cross-sectional study. BMC Med 12:240, doi: 10.1186/s12916-014-0240-0.

Maervoet J, Vermeir G, Covaci A, Van Larebeke N, Koppen G, Schoeters G, et al. 2007. Association of thyroid hormone concentrations with levels of organochlorine compounds in cord blood of neonates. Environ Health Perspect 115:1780–1786, doi: 10.1289/ehp.10486.

Maiheu B, Veldeman B, Viaene P, De Ridder K, Lauwaet D, Smeets N, et al. 2013. Identifying the Best Available Large-Scale Concentration Maps for Air Quality in Belgium. Study Commissioned by the Flemish Environment (MIRA) [in Dutch].​0_onderzoeksrapporten/2013/Eindrapport_C​oncentratiekaarten_29_01_2013_TW.pdf [accessed 16 December 2015].

Männistö T, Hartikainen AL, Vääräsmäki M, Bloigu A, Surcel HM, Pouta A, et al. 2012. Smoking and early pregnancy thyroid hormone and anti-thyroid antibody levels in euthyroid mothers of the Northern Finland Birth Cohort 1986. Thyroid 22(9):944–950.

Maruo T, Matsuo H, Mochizuki M. 1991. Thyroid hormone as a biological amplifier of differentiated trophoblast function in early pregnancy. Acta Endocrinol (Copenh) 125(1):58–66.

McDonald SD, Walker MC, Ohlsson A, Murphy KE, Beyene J, Perkins SL. 2008. The effect of tobacco exposure on maternal and fetal thyroid function. Eur J Obstet Gynecol Reprod Biol 140(1):38–42.

Medici M, Timmermans S, Visser W, de Muinck Keizer-Schrama SM, Jaddoe VW, Hofman A, et al. 2013. Maternal thyroid hormone parameters during early pregnancy and birth weight: the Generation R Study. J Clin Endocrinol Metab 98(1):59–66.

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(6):946–949.

Modesto T, Tiemeier H, Peeters RP, Jaddoe VW, Hofman A, Verhulst FC, et al. 2015. Maternal mild thyroid hormone insufficiency in early pregnancy and attention-deficit/hyperactivity disorder symptoms in children. JAMA Pediatr 169(9):838–845.

Morreale de Escobar G, Obregon MJ, Escobar del Rey F. 2004. Role of thyroid hormone during early brain development. Eur J Endocrinol 151(suppl 3):U25–U37.

Parate VR, Rode M, Pande S, Ansari T, Kamble P. 2010. Thyroid function in mothers during the process of normal delivery. Int J Endocrinol Metab 1:39–45.

Pedersen M, Giorgis-Allemand L, Bernard C, Aguilera I, Andersen AMN, Ballester F, et al. 2013. Ambient air pollution and low birthweight: a European cohort study (ESCAPE). Lancet Res Med 1(9):695–704.

Persani L. 2012. Central hypothyroidism: pathogenic, diagnostic, and therapeutic challenges. J Clin Endocrinol Metab 97(9):3068–3078.

Rappazzo KM, Daniels JL, Messer LC, Poole C, Lobdell DT. 2014. Exposure to fine particulate matter during pregnancy and risk of preterm birth among women in New Jersey, Ohio, and Pennsylvania, 2000–2005. Environ Health Perspect 122:992–997, doi: 10.1289/ehp.1307456.

Reed HL. 1995. Circannual changes in thyroid hormone physiology: the role of cold environmental temperatures. Arctic Med Res 54(suppl 2):9–15.

Saenen ND, Vrijens K, Janssen BG, Madhloum N, Peusens M, Gyselaers W, et al. 2016. Placental nitosative stress and exposure to ambient air pollution during gestation: a population study. Am J Epidemiol 184(6):442–449.

Shields B, Hill A, Bilous M, Knight B, Hattersley AT, Bilous RW, et al. 2009. Cigarette smoking during pregnancy is associated with alterations in maternal and fetal thyroid function. J Clin Endocrinol Metab 94(2):570–574.

Shields BM, Knight BA, Hill A, Hattersley AT, Vaidya B. 2011. Fetal thyroid hormone level at birth is associated with fetal growth. J Clin Endocrinol Metab 96(6):E934–E938.

Soldin OP, Goughenour BE, Gilbert SZ, Landy HJ, Soldin SJ. 2009. Thyroid hormone levels associated with active and passive cigarette smoking. Thyroid 19(8):817–823.

Steadman RG. 1979. The assessment of sultriness. Part I: A temperature-humidity index based on human physiology and clothing science. J Appl Meteor 18:861–873.

Thomson EM, Vladisavljevic D, Mohottalage S, Kumarathasan P, Vincent R. 2013. Mapping acute systemic effects of inhaled particulate matter and ozone: multiorgan gene expression and glucocorticoid activity. Toxicol Sci 135(1):169–181.

Thorpe-Beeston JG, Nicolaides KH, Felton CV, Butler J, McGregor AM. 1991. Maturation of the secretion of thyroid hormone and thyroid-stimulating hormone in the fetus. N Engl J Med 324(8):532–536.

Valeri L, Vanderweele TJ. 2013. Mediation analysis allowing for exposure–mediator interactions and causal interpretation: theoretical assumptions and implementation with SAS and SPSS macros. Psychol Methods 18(2):137–150.

Vulsma T, Gons MH, de Vijlder JJM. 1989. Maternal-fetal transfer of thyroxine in congenital hypothyroidism due to a total organification defect or thyroid agenesis. N Engl J Med 321(1):13–16.

Wilber JF, Utiger RD. 1969. The effect of glucocorticoids on thyrotropin secretion. J Clin Invest 48(11):2096–2103.

World Health Organization. 2006. WHO Air Quality Guidelines for Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide.​f_file/0005/78638/E90038.pdf?ua=1 [accessed 26 November 2015].

Yoshimura Noh J, Momotani N, Fukada S, Ito K, Miyauchi A, Amino N. 2005. Ratio of serum free triiodothyronine to free thyroxine in Graves’ hyperthyroidism and thyrotoxicosis caused by painless thyroiditis. Endocr J 52(5):537–542.

WP-Backgrounds Lite by InoPlugs Web Design and Juwelier Schönmann 1010 Wien