Synergism between exposure to mercury and use of iodine supplements on thyroid hormones in pregnant women

Environmental Research 138 (2015) 298–305

Contents lists available at ScienceDirect

Environmental Research journal homepage: www.elsevier.com/locate/envres

Synergism between exposure to mercury and use of iodine supplements on thyroid hormones in pregnant women Sabrina Llop a,b,n, Maria-Jose Lopez-Espinosa a,b, Mario Murcia a,b, Mar Alvarez-Pedrerol b,c,d, Jesús Vioque b,e, Xabier Aguinagalde f, Jordi Julvez b,c,d, Juan J. Aurrekoetxea g,h,i, Mercedes Espada j, Loreto Santa-Marina b,g,h, Marisa Rebagliato a,b,k, Ferran Ballester a,b a FISABIO–Universitat de València–Universitat Jaume I Joint Research Unit of Epidemiology and Environmental Health, Av. Catalunya 21, 46020 Valencia, Spain b Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Madrid, Spain c Centre for Research in Environmental Epidemiology (CREAL), Doctor Aiguader 88, 08003 Barcelona, Spain d Municipal Institute of Medical Research (IMIM-Hospital del Mar), Doctor Aiguader 88, 08003 Barcelona, Spain e Universidad Miguel Hernandez, Avenida de Alicante KM 87, 03550 Sant Joan d´Alacant, Spain f Laboratorio de Salud Pública de Alava, Santiago 11, 01002 Vitoria Gasteiz, Spain g Departamento de Sanidad Gobierno Vasco, Subdirección de Salud Pública de Gipuzkoa, Avenida de Navarra 4, 20013 San Sebastián, Spain h Biodonostia, Instituto de Investigación Biomédica, Doctor Begiristain, s/n, 20014 San Sebastián, Spain i Departamento de Medicina Preventiva y Salud Pública, University of the Basque Country (UPV/EHU), Apartado 1599, 20014 San Sebastian, Spain j Laboratorio de Salud Pública de Bizkaia, Departamento de Salud del Gobierno Vasco, Parque Tecnológico de Bizkaia, C/Ibaizabal, Edificio 502, 1ª Pt, 48160 Derio, Spain k Universitat Jaume I, Av. de Vicent Sos Baynat, s/n, 12071 Castelló de la Plana, Spain

art ic l e i nf o

a b s t r a c t

Article history: Received 21 November 2014 Received in revised form 16 February 2015 Accepted 23 February 2015 Available online 5 March 2015

Objective: To evaluate the association between mercury exposure and thyroid-stimulating hormone (TSH), total triiodothyronine (TT3) and free thyroxine (FT4) levels during pregnancy as well as to explore if there is any synergic action between mercury and intake of iodine from different sources. Methods: The study population was 1407 pregnant women participating in the Spanish INMA birth cohort study. Total mercury concentrations were analyzed in cord blood. Thyroid hormones (THs) were measured in serum samples collected at 13.2 71.5 weeks of gestation. The association between mercury and TH levels was evaluated with multivariate linear regression models. Effect modification caused by iodine intake from supplements and diet was also evaluated. Results: The geometric means of TSH, TT3, FT4 and mercury were 1.1 μU/L, 2.4 nmol/L, 10.5 pmol/L and 7.7 μg/L, respectively. Mercury levels were marginally significantly associated with TT3 (β:
0.05; 95%CI:
0.10, 0.01), but were neither associated with TSH nor FT4. The inverse association between mercury and TT3 levels was stronger among the iodine supplement consumers (
0.08; 95%CI:
0.15,
0.02, interaction p-value¼ 0.07). The association with FT4 followed the same pattern, albeit not significant. Conclusion: Prenatal mercury exposure was inversely associated with TT3 levels among women who took iodine supplements during pregnancy. These results could be of public health concern, although further research is needed. & 2015 Elsevier Inc. All rights reserved.

Keywords: Prenatal exposure Endocrine disruption Mercury Thyroid Iodine supplementation

1. Introduction

Abbreviations: CI, confidence interval; D1, deiodinase type I; D2, deiodinase type II; D3, deiodinase type 3; FT4, free thyroxine; GM, geometric mean; sd, standard deviation; TH, thyroid hormones; TSH, thyroid-stimulating hormone; TT3, total triiodothyronine n Corresponding author at: Foundation for the Promotion of Health and Biomedical Research in the Valencian Region, FISABIO, Av. Catalunya 21, 46020, Valencia Spain. E-mail address: [email protected] (S. Llop). http://dx.doi.org/10.1016/j.envres.2015.02.026 0013-9351/& 2015 Elsevier Inc. All rights reserved.

Thyroid hormones (THs) are essential for cellular metabolism, growth, and development, especially for normal brain maturation (Porterfield and Hendrich, 1993). In fact, deficiencies in maternal THs may lead to neuropsychological disorders, especially during the first half of pregnancy, when the fetus is totally dependent on THs of maternal origin (Morreale de et al., 2004). Some epidemiological studies have observed an association between subclinical alterations of TH status during pregnancy and delays in child neuropsychological development later in life (Haddow et al.,

S. Llop et al. / Environmental Research 138 (2015) 298–305

1999; Julvez et al., 2013; Li et al., 2010; Pop et al., 2003, 1999). A large number of environmental pollutants are known to interfere with TH equilibrium (Langer, 2008; Pearce and Braverman, 2009). Conversely, published data on mercury and thyroid function is limited and mostly focused on adults (Abdelouahab et al., 2008; Chen et al., 2013; Yorita Christensen, 2013) and only two studied pregnant populations. In a Canadian study (n ¼147) of pregnant women and their offspring (Takser et al., 2005), an inverse association between cord free thyroxine (FT4) and inorganic mercury was reported, however, such an association was not found in a larger study from the Faroe Islands (Steuerwald et al., 2000). An adequate iodine intake during pregnancy is essential for the synthesis of maternal THs (Glinoer, 2007). It has been suggested that both deficient and excessive iodine intake can affect the occurrence of subclinical thyroid diseases (Laurberg et al., 2010). In fact, a previous study in pregnant women from the INMA study showed that iodine supplement intake of 200 daily grams or more in the first half of pregnancy was associated to maternal thyroid dysfunction, specifically increasing risk of levels of thyroid-stimulating hormone (TSH) above 3 mU/mL (adjusted odds ratio ¼2.5; 95% confidence interval¼1.2–5.4) (Rebagliato et al., 2010). Cord blood mercury concentrations in INMA cohort study were relatively high (the geometric mean of total mercury was 8.2 μg/L) (Ramon et al., 2011). However, the effect of the exposure to this metal on TH has yet to be explored. The aim of this study is to evaluate the association between exposure to mercury and maternal TH levels (TSH, FT4, and total triiodothyronine [TT3]) in the INMA birth cohort study. We also propose to explore if there is any synergic action between mercury and the use of different sources of iodine intake during pregnancy.

2. Materials and methods 2.1. The study population The study subjects were pregnant women participating in the INMA birth cohort. The INMA—INfancia y Medio Ambiente (Childhood and the Environment)—project is a multicenter cohort study which aims to investigate the effect of environmental exposures and diet during pregnancy on fetal and child development (http://www.proyectoinma.org/). Details of protocol and study population in INMA project were reported previously (Guxens et al., 2012). Briefly, for this study, women were recruited at the beginning of their pregnancy (November 2003–February 2008) in three regions of Spain: Valencia (N ¼ 855), Sabadell (N ¼657), and Gipuzkoa (N ¼ 638). A total sample of 2021 women (92.0% in Valencia, 94.7% in Sabadell and 95.8% in Gipuzkoa) were followed up until delivery. Among them, those women who reported having been diagnosed with thyroid pathology (n ¼ 87), regardless of whether or not they continued in treatment, were excluded as was a woman from Gipuzkoa with an outlier value in TT3 (15.2 nmol/L). Only women with data available on cord mercury concentrations, maternal THs, as well as questionnaire information were included in this study. We were unable to collect cord blood samples from all participants in the study. This is the main reason for the differences between included (n ¼1407) and excluded (n ¼614) participants. Characteristics of the areas at study have been described previously (Guxens et al., 2012). Briefly, the Valencia area at study has been sub-divided in four different zones as function of the population density and main uses of land: the urban area is the part of the city of Valencia included in the project, the metropolitan are composed by towns near the city within the ring road, the semi-

299

urban are composed by towns where agriculture and a certain amount of industrial activity are combined with residential areas, and the rural area with villages with a low population density. The Sabadell area is exclusively urban composed by the medium size city of Sabadell (Catalonia). The area of Gipuzkoa (Basque Country) included 25 municipalities and is divided in three narrow valleys that have a high grade of unevenness. Metallurgy is the principal industrial activity in the area. Women participating in the study signed an informed consent form and the Ethics Committees of the centres involved in the study approved the research protocol. 2.2. Mercury analysis Whole cord blood samples were collected by using venipuncture of cord vessels before the placenta was delivered. Samples were processed, separated into aliquots of 1 mL, and then frozen to
80 °C until analysis. One aliquot was used to analyze total mercury by thermal decomposition, amalgamation, and atomic absorption spectrometry by using a single purpose AMA254 advanced mercury analyzer (LECO Corporation, St. Joseph, Michigan). The limit of determination was 2 mg/L. More details of analytical procedure can be found in Ramon et al. (2011). We have categorized the variable mercury into quartiles (1st: o 4.8, 2nd: 4.8–7.8, 3rd: 7.9–13.0, 4th: 413 mg/L). 2.3. Thyroid hormone analysis We measured TSH, TT3, FT4 as biomarkers of thyroid function in maternal serum samples taken at 13.1 71.5 (mean 7standard deviation [sd]) weeks of gestation by means of a solid-phase, timeresolved sandwich fluoroimmunoassay (AutoDELFIA, Perkin Elmer Life and Analytical Sciences, WallacOy, Turku, Finland) using a lanthanide metal europium label. Measurements were performed at the Public Health Laboratory of Bilbao (Spain). Between-assay coefficients of variance were 3.0%, 3.1%, and 2.6% for TSH, 7.2%, 5.5%, and 5.2% for TT3, and 6.1%, 4.1%, and 4.0% for FT4 at low, medium, and high concentrations, respectively. 2.4. Other variables At the end of the first trimester of pregnancy (12.9 7 1.7 weeks of gestation), women filled in a questionnaire on socio-demographic, environmental, dietary and lifestyle characteristics, as well as iodine supplement information. The maternal covariates used in this study were maternal age (years), country of birth (Spain, South America, other), level of education (up to primary, secondary, university studies), working situation (unemployed, employed), smoking at the beginning of pregnancy (no, yes), parity (0, 1, 4 1 children), season of samples collection (winter, spring, summer, autumn), caffeine intake (0, 40–1, 41 mg/day), alcohol intake (0, 4 0–2 servings/week), and social class, defined according to the most privileged occupation during pregnancy of the mother or the father using a widely used Spanish adaptation of the international ISCO88 coding system. Class I included managerial jobs, senior technical staff, and commercial managers; class II included skilled non-manual workers; and class III included manual workers. Information on brand names, dose and timing of consumption of specific potassium iodide supplements, or vitamin/mineral preparations containing iodine was obtained with a reference time window from 3 months before conception until the date of the interview. The iodine content per daily dose was obtained from the composition referred to in the reference manual or in the product label information. We defined supplement consumers as those women who were taking supplements at the time of TH

300

S. Llop et al. / Environmental Research 138 (2015) 298–305

testing (categorized as no vs. yes, as well as r100 mg/day vs. 4100 mg/day, and r 200 mg/day vs. 4200 mg/day). The daily iodine intake from diet was obtained from a validated food frequency questionnaire (FFQ) of 100 food items. Pregnant women were asked how often, on average, they had consumed each type of item since their last menstrual period until the time of the interview at the 1st trimester of pregnancy (Vioque et al., 2013). Iodine intake from foods and salt was estimated and adjusted for total energy by the residual method (Willett et al., 1985). The variable was dichotomized according to tertiles (1st tertiler150.1 mg/day; 150.1 mg/day o2nd tertiler 224.9 mg/day; 3rd tertile 4224.9 mg/day). Maternal PCB (congeners 118, 153, 138 and 180) concentrations were measured in maternal blood samples taken at the 12th week of gestation by gas chromatography with electron capture detection (Llop et al., 2010). The sum of the four PCBs was included in the analysis. 2.5. Urinary iodine analysis Iodine concentrations were measured in a spot urine sample taken at the end of the first trimester of pregnancy by a paired-ion reversed-phase high-performance liquid chromatography with electrochemical detection and a silver working electrode. Urine samples were stored at
20 °C, until they were delivered to the reference laboratory (Normative Public Health Laboratory of Bilbao, Basque Country) (Rebagliato et al., 2010). WHO established as a normal range of urinary iodine for pregnant women between 150 and 249 mg/L (World Health Organization (WHO), 2007). This variable was categorized according to the WHO recommendations (o150, 150–249, Z250 mg/L). 2.6. Statistical analysis Differences in socio-demographic, environmental and dietary characteristics between the included and excluded participants were first evaluated using the χ2 test. A descriptive analysis by geometric means of mercury and THs by region was performed. Differences according to mercury and TH levels by region were contrasted by ANOVA test. For further analysis TH levels were standardized by weeks of gestation at the time of blood sampling. Mercury and TH levels were log 2 transformed due to their skewed distribution. Spearman correlations adjusted by region between THs and total mercury, iodine intake from supplements and diet, and urinary iodine were obtained. General additive models using cubic smoothing splines with 2– 4 degrees of freedom were performed in order to study the shape of the relationship between the TH levels and the exposure to mercury. We compared linear and non-linear models with the aid of graphical examination and the LR test (p o0.05). Spline regression models did not significantly improve the fit over the linear models; hence the linear model was used in all cases. Multivariate linear regression models were performed in order to explore the relationship between prenatal exposure to mercury and TH levels. Beta coefficients and 95% confidence intervals (CIs) were obtained. A 2-step procedure was used for multivariate model building. In the first step, a core model was built for TSH, TT3 and FT4 levels using all the marginally significant covariates in the bivariate analysis (p-value o0.1). Following a backward elimination procedure, all the covariates associated with the TH levels at a level of p o0.1 in the likelihood ratio test were retained in the models. We included mercury levels in these final adjusted models (as a continuous variable and categorized as 4th quartile vs. 1st quartile). Factors associated with individual mercury levels (Ramon et al., 2011) were adjusted for as potential confounders if they

changed the magnitude of the main effects by more than 10%. Sensitivity analysis including the sum of four PCBs (118, 138, 153, and 180) in the multivariate models was also performed. Effect modification by iodine intake from supplements and from diet was assessed by including cross-product (Hg  different iodine sources) terms in the models. Subsequently, the study population was stratified and the association between mercury and TH levels was evaluated among groups. The change (%) in TH levels related to mercury exposure was graphically represented. The analyses were carried out using the Stata version 11 statistical package (StataCorp LP, College Station, Texas) and R.

3. Results Socio-demographic characteristics of the study population are shown in Table 1. Most of the women (91%) were born in Spain, 40.3% had finished secondary school, 86.8% were employed during pregnancy, around 41.6% belonged to the lowest social class, 31.3% smoked at the beginning of the pregnancy, 55.4% were primiparous and the mean age was 30.5 (sd: 4.2) years. We found statistically significant differences between the study population and excluded women according to educational level, alcohol consumption and age. Excluded women were less educated, younger and consumed more alcohol than the study population (Table 1). Mercury and TH levels according to the maternal covariates are shown in Supplemental material Table 1. Levels of total mercury in cord blood and THs in maternal serum are shown in Table 2. The geometric means (95%CIs) of mercury, TSH, TT3 and FT4 levels were 7.7 mg/L (7.3, 7.9), 1.14 mIU/L (1.09, 1.19), 2.43 nmol/L (2.40, 2.45) and 10.51 pmol/L (10.43, 10.58), respectively. There were no statistically significant differences in TSH and TT3 levels between regions, but FT4 levels were higher in women from Valencia than those from Gipuzkoa and Sabadell. Women from Gipuzkoa consumed more iodine in their diet than women from Sabadell and from Valencia. There was a higher percentage of women from Gipuzkoa who consumed iodine supplements during the 1st trimester of pregnancy than in Valencia and Sabadell. Urinary iodine was also higher in Gipuzkoa than in Valencia and in Sabadell. Spearman correlation coefficients between THs, mercury concentrations, iodine intake from diet and supplements, and urinary iodine are shown in Table 3. The correlation between mercury and TT3 was negative and statistically significant. The correlation between urinary iodine and TT3 was also negative but marginally significant. Results from linear regression models between mercury and THs are shown in Table 4. A doubling in mercury levels associated with TT3 levels in the unadjusted model (β:
0.07; 95%CI:
0.12,
0.02) and marginally significantly associated in the multivariate model (β:
0.05; 95%CI:
0.10, 0.01). When we studied the impact of high vs. low exposure to mercury (4th quartile [ 4 13 mg/L] vs. 1st quartile [4.8 mg/L]) on THs we observed a significant association with TSH (β: 0.16; 95%CI: 0.00, 0.31) and TT3 (β:
0.20; 95%CI:
0.35,
0.05) in the unadjusted models. In the adjusted models the association between mercury and TT3 diluted and became marginally significant (β:
0.14; 95%CI:
0.31, 0.03). These associations did not change after including maternal PCBs in the model (data not shown). We found a marginal significant interaction between mercury concentrations and the intake of iodine supplements (p ¼0.07) and the intake of 4200 mg/day of iodine supplements (p ¼0.07) on TT3 concentrations. When we stratified the study population in order to evaluate the association between mercury and TH levels according to their iodine intake from supplements and diet, and the urinary iodine (Fig. 1), we did not find any significant change on

S. Llop et al. / Environmental Research 138 (2015) 298–305

Table 1 Socio-demographic characteristics of the study population, INMA Study, Spain, 2003–2008. Study population (n¼1407)

Excluded populationa (n¼ 614)

N

%

N

%

1278 89 37

91.0 6.3 2.6

554 37 22

90.4 6.0 3.6

0.494

346 567 491

24.6 40.3 34.9

177 250 185

28.9 40.8 30.2

0.053

Working situation Unemployed 186 Employed 1221

13.2 86.8

95 519

15.5 84.5

0.178

Social classc I II III

452 370 585

32.1 26.3 41.6

178 152 284

29.0 24.8 46.3

0.141

Smoking habit No Yes

952 433

68.7 31.3

383 212

64.4 35.6

0.059

Parity (children) 0 1 41

779 526 100

55.4 37.4 7.1

333 234 47

54.2 38.1 7.7

0.846

Season of sampling Winter 326 Spring 331 Summer 374 Autumn 360

23.4 23.8 26.9 25.9

125 156 160 154

21.0 26.2 26.9 25.9

0.719

Daily caffeine intake (mg/day) 0 950 40–100 317 4100–200 127

68.1 22.7 9.1

408 133 67

67.1 21.9 11.0

0.406

Alcohol consumption (servings/week) 0 1098 78.8 40–2 296 21.2

454 154

74.7 25.3

0.044

(4.0)

0.028

Country of birth Spain South America Other

Level of education Up to primary Secondary University

p-Valueb

301

mercury exposure and FT4 levels were also found among women with iodine supplement intake, although it was not statistically significant (doubling exposure to mercury was associated with a decrease in FT4 concentrations of
3.4% (95%CI:
7.58, 1.04) (Fig. 1C).

4. Discussion

Maternal age (years)d Mean (sd) 30.5

(4.2)

30.0

a

Women followed until delivery but who did not take part in this study. p-Value from χ2 test. c Class I: managerial jobs, senior technical staff, and commercial managers; class II: skilled non-manual workers; and class III: manual workers. d p-Value from Mann–Whitney test. b

TSH coefficients (Fig. 1A). However, among women who consumed iodine supplements during pregnancy we found an inverse and statistically significant association between mercury and TT3 levels. Doubling exposure to mercury was associated with a decrease in TT3 concentrations of
5.5% (95%CI:
10.0,
0.9) (Fig. 1B). Among women who consumed 4200 mg/day of iodine supplements the inverse association was stronger (
6.7% (95%CI:
12.3,
0.8). No significant association was found when the study population was stratified according to their iodine intake from diet and the urinary iodine. A negative relation between

In this cohort study of pregnant women from a European Mediterranean country we have found that mercury exposure was inversely but not statistically significantly associated with maternal TT3 and FT4 levels. This association was stronger among women who consumed iodine from supplements and it was statistically significant in the case of TT3. The inverse association between mercury and TT3 levels was stronger among the highest iodine supplement consumers (4200 mg/day). Literature on the association between prenatal exposure to mercury and maternal TH levels is sparse. In line with our results, Takser et al. (2005) observed a negative effect of maternal inorganic mercury on TT3 levels in pregnant women from Quebec (Canada), although no association between methylmercury and levels of none of the studied THs emerged. A study in Slovakia reported that TT3 and FT3 levels in 6-month-old children were inversely related to total mercury in cord blood and mothers with dental amalgam fillings had statistically significantly lower TT4 and FT4 levels (Ursinyova et al., 2012). Another study conducted in the Faroe Islands (Steuerwald et al., 2000) did not report any significant association between prenatal mercury and TH levels measured during pregnancy and at birth. The cord blood total mercury concentrations observed in this population is relatively high in comparison with other populations. The GM was 7.7 (7.3, 7.9) mg/L and around 20% and 60% of newborns had concentrations above the World Health Organization (WHO) and US-EPA recommendations (United Nations Environment Programme, 2007; US Environmental Protection Agency, 2007), respectively. In the present study, concentrations were around twice those reported in a US population (Lederman et al., 2008); higher than in a study in a Korean population (Lee et al., 2010); much higher (ten times) than concentrations found in some European countries such as Poland and Sweden (Bjornberg et al., 2003; Jedrychowski et al., 2007), but lower than levels reported in heavily exposed populations in Faeroe and Greenland (Bjerregaard and Hansen, 2000; Grandjean et al., 1997; Muckle et al., 2001; Steuerwald et al., 2000). In previous studies we described a strong association between these cord blood mercury concentrations and fish intake by pregnant women, especially the intake of large oily fish (Ramon et al., 2011). WHO recommends a iodine intake for pregnant and lactating women of 250 mg/day (Andersson et al., 2007) and suggests that the total daily intake should preferably not exceed 500 mg/day, as higher intake might be associated with impaired thyroid function. However, iodine supplementations is still under debate due to the fact that supplementary or excessive iodine intake during pregnancy has been related to altered TH levels in both mothers (Moleti et al., 2011; Orito et al., 2009; Rebagliato et al., 2010) and neonates (Nohr and Laurberg, 2000). We have observed an interaction between mercury exposure and the intake of iodine from supplements but not from diet. Previous results in our study population suggested that a rapid iodine load through supplements during early pregnancy might result in an iodine excess that could negatively affect the maternal thyroid function (Rebagliato et al., 2010). Around 7% of our study population who consumed iodide supplements during pregnancy exceeded the WHO guidelines of 500 mg/day when also considering the iodine from diet and salt. These results also support the relative importance of an adequate

302

S. Llop et al. / Environmental Research 138 (2015) 298–305

Table 2 Levels of cord blood total mercury and maternal serum thyroid hormones, INMA Study, Spain, 2003–2008.

Total mercury (mg/L) TSH (mIU/L) TT3 (nmol/L) FT4 (pmol/L) Iodine intake from diet (mg/day) Iodine supplement intake, n (%) Urinary iodine (mg/L)

All GM (95%CI)

Gipuzkoa (n¼508) GM (95%CI)

Sabadell (n¼ 429) GM (95%CI)

Valencia (n¼470) GM (95%CI)

p-Value

7.7 1.14 2.43 10.51 180.3 775 129.7

7.5 1.15 2.41 10.09 196.3 484 168.0

6.3 1.13 2.40 10.65 153.2 33 90.6

9.6 1.14 2.46 10.84 192.8 258 130.5

o 0.001 0.649 0.860 o 0.001 o 0.001 o 0.001a o 0.001

(7.3, 7.9) (1.09, 1.19) (2.40, 2.45) (10.43, 10.58) (176.5, 184.3) (55.1%) (124.6, 134.9)

(7.0, 7.9) (1.06, 1.24) (2.38, 2.44) (9.98, 10.19) (189.4, 203.4) (95.3%) (158.5, 178.1)

(5.9, 6.8) (1.05, 1.22) (2.36, 2.44) (10.52, 10.79) (146.3, 158.4) (7.7%) (84.1, 97.7)

(8.9, 10.4) (1.06, 1.22) (2.42, 2.50) (10.71, 10.96) (186.3, 199.6) (54.9%) (122.4, 139.1)

CI: confidence interval; FT4: free thyroxine; GM: geometric mean; TSH: thyroid-stimulating hormone; TT3: total triiodothyronine. p Value from ANOVA test (differences by region). Levels of mercury, thyroid hormones and urinary iodine have been log 2 transformed. a

p-Value from χ2 test.

Table 3 Spearman correlationsa between THs and total mercury concentrations, iodine intake from diet and supplements, and urinary iodine, INMA Study, Spain, 2003– 2008.

TSH TT3 FT4

Total mercury

Iodine intake from diet

Iodine supplement intake

Urinary iodine

0.04
0.09nn
0.01


0.01
0.01 0.02

0.05
0.02
0.03


0.02
0.04n
0.01

FT4: free thyroxine; TSH: thyroid-stimulating hormone; TT3: total triiodothyronine. a

Adjusted by cohort. po 0.100. nn p o0.050. n

baseline iodine nutritional status, rather than sudden increases of iodine intake during pregnancy. Experimental studies have shown that excess of iodine could inactivate the deiodinase enzymes and this effect can be stronger in a selenium deficiency situation (Xu et al., 2006; Yang et al., 2006). In the thyroid system, the deiodinases make up 3 isoforms of selenoenzymes that convert THs from one iodinated form to another. Deiodinase type I (D1) found in liver, kidney, muscle tissue and thyroid gland produces most of the circulating T3. Similar to D1 also D2, found in brain, testis and thyroid gland, is the major activating enzyme and catalyzes the T3 production from T4 by deiodination at the phenolic ring. Finally the deiodinase type 3 (D3) found in fetal tissue and placenta removes iodide from the tyrosyl ring and therefore is an obligatory inactivating enzyme of THs and their metabolites (Kohrle, 1999). Deiodinases involved in TH metabolism might also be a potential target of mercury (Soldin et al., 2008). It has been

demonstrated that perinatal exposure to low doses of methylmercury resulted in a decrease in brain D3 activity in neonatal mice (Mori et al., 2006). Barregard et al. (1994) observed that changes in TH levels in the plasma of workers exposed to mercury vapor may be due to an inhibitory effect of mercury on D1 or D2 that converts T4 to T3 in both the thyroid and peripheral tissue. Results from Ellingsen et al. (2000) also suggested that mercuryvapour exposure could alter the function of D1, which also deiodinates reverse T3. Exposure to mercury may affect deiodinases production, a selenium dependent enzyme, by sequestering the selenium in the brain and endocrine system. In our study, the strongest inverse association between mercury and TT3 levels was observed among women who consumed 4200 mg/day of iodine from supplements, which may be indicating a synergic effect of both mercury and elevated iodine on deiodinases activity. To the best of our knowledge, this is the first epidemiological study reporting an interaction between prenatal exposure to mercury and iodine supplementation during pregnancy. These results could be of concern taking into account the medium-high exposure to mercury observed in our study population (Ramon et al., 2011) as well as the health policy of several countries, including Spain, of encouraging pregnant and lactating women to follow iodine supplementation programs, regardless of their iodine status (Becker et al., 2006). These programs could be safe and effective in iodinedeficient areas, but previous studies in our cohort showed that iodine supplementation Z200 mg/day during pregnancy was associated with maternal thyroid dysfunction (Rebagliato et al., 2010) and Z150 mg/day was related to impaired mental and psychomotor development during the second year of age (Rebagliato et al., 2013). This study has some limitations. Firstly, the exposure to mercury was evaluated in cord blood samples while THs were

Table 4 Results from unadjusted and adjusted linear regression models between cord blood total mercury and maternal serum thyroid hormone levels, INMA Study, Spain, 2003– 2008. Doubling in total mercurya

Total mercury(high vs. low)b

Unadjusted

TSH TT3 FT4

β

95%CI

0.04
0.07 0.00


0.01
0.12
0.05

Adjusted

0.09
0.02 0.04

Unadjusted

p-Value

β

95%CI

0.100 o 0.001 0.860

0.02
0.05
0.01


0.03
0.10
0.06

0.07 0.01 0.03

p-Value

β

95%CI

0.500 0.070 0.520

0.16
0.20
0.01

0.00
0.35
0.15

Adjusted

0.31
0.05 0.14

p-Value

β

95%CI

0.047 0.010 0.920

0.09
0.14
0.08


0.07
0.31
0.23

CI: confidence interval; FT4: free thyroxine; TSH: thyroid-stimulating hormone; TT3: total triiodothyronine. TSH model adjusted for region, parity, smoking habit and iodine intake from supplements. TT3 model adjusted for maternal age, region, level of education, social class, season of blood sampling and smoking habit. FT4 model adjusted for maternal age, country of birth, region and caffeine intake. a b

Log 2 transformed. 4th quartile ( 413 mg/L) vs. 1st quartile ( o 4.8 mg/L).

p-Value 0.26 0.03 0.07

0.284 0.108 0.314

S. Llop et al. / Environmental Research 138 (2015) 298–305

303

analysed in serum samples from the first trimester of pregnancy. It is interesting to take into account, however, that mercury has a medium-term life and exposure is related with fish consumption so individual levels of exposure are not expected to change dramatically during pregnancy. Some studies analysed mercury at different times throughout pregnancy and at birth and found that cord blood mercury levels were strongly correlated (r around 0.7) with mercury analyzed in the first trimester (Morrissette et al., 2004; Vahter et al., 2000). Secondly, we lack information about selenium levels during pregnancy, a key element for discussing the deoidinases hypothesis. Additionally, we have no data on TT4 levels to reinforce the findings of TT3 and the possible change of TH levels through pregnancy and the postpartum period. Likewise, although our findings are of interest, it is yet to be determined if these associations persist and have clinically important long-term impacts on physical health. Further work on this population in relation to the association between iodine intake and maternal thyroid function indicators on perinatal outcomes and long-term child development will allow some of these concerns to be addressed. This is the largest study to date on the association between mercury and TH status as well as the first reporting an interaction between prenatal exposure to mercury and iodine supplementation during pregnancy. Furthermore, and as part of the ongoing cohort study, in the future we will be able to evaluate the association between iodine intake and indicators of maternal thyroid function on perinatal outcomes and long-term child development. Finally, other strengths from this study are the low rate of missing participants between recruitment and delivery, and detailed information on many potential confounders from early pregnancy. In conclusion, the results in our birth cohort study showed that prenatal exposure to mercury associated with lower TT3 levels among women who took iodine supplements during the 1st trimester of pregnancy, especially among those who consumed 4200 mg/day of iodine from supplements. The association between mercury and FT4 levels followed a similar, albeit not significant, pattern. Experimental studies suggested that both mercury and iodine excess have similar action mechanisms of TH disruption, i.e. the inactivation of deiodinases enzymes which are crucial in T4 and T3 levels. These results could be of public health concern due to the high levels of mercury reached by populations with medium-high fish consumption during pregnancy and also the extended recommendation of systematic iodine supplementation during pregnancy even in iodine-sufficient populations. Further research is needed, as well as surveillance and public debate on this practice. Fig. 1. Association between total mercury and TSH (A), TT3 (B) and FT4 (C) levels in all pregnant women and stratified according to their iodine intake from supplements and diet, INMA Study, Spain, 2003–2008. TSH models adjusted for region, parity, smoking, and iodine intake from supplements (only in the model with all women). Study population: all women (n¼ 1380), women who did not take iodine supplements (n¼ 620), women who took iodine supplements (n¼760), women who took 4100 mg/day of iodine supplements (n¼ 608), women who took 4200 mg/day of iodine supplements (n¼ 509), women who took iodine from diet 43rd tertile (224.9 mg/day) (n¼ 466). TT3 models adjusted for maternal age, region, parity, level of education, social class, season of sampling, and smoking. Study population: all women (n¼ 1362), women who did not take iodine supplements (n¼ 617), women who took iodine supplements (n¼745), women who took 4100 mg/day of iodine supplements (n¼ 593), women who took 4200 mg/day of iodine supplements (n¼ 494), women who took iodine from diet 43rd tertile (224.9 mg/day) (n¼ 455). FT4 models adjusted for region, maternal age, country of birth and caffeine intake. Study population: all women (n¼ 1390), women who did not take iodine supplements (n¼622), women who took iodine supplements (n¼ 768), women who took 4100 mg/day of iodine supplements (n¼ 615), women who took 4200 mg/day of iodine supplements (n¼ 516), women who took iodine from diet 43rd tertile (224.9 mg/day) (n¼ 476).

Disclosure statement The authors have nothing to disclose.

Funding This study was funded by grants from Instituto de Salud Carlos III (Red INMA G03/176 and CB06/02/0041), UE (FP7-ENV-2011 cod 282957 and HEALTH.2010.2.4.5-1), FIS FEDER (03/1615, 04/1509, 04/1112, 04/1931, 04/1436, 05/1052, 06/0867, 06/1213, 07/0314, 08/115105/1079, 09/02647, 09/00090, 11/01007, 11/02591, 11/ 00178, 13/1944, 14/00891, and 14/01687), Conselleria de Sanitat Generalitat Valenciana, Generalitat de Catalunya (CIRIT 1999SGR 00241), Department of Health of the Basque Government (2005111093 and 2009111069), the Provincial Government of Gipuzkoa (DFG06/004 and DFG08/001).

304

S. Llop et al. / Environmental Research 138 (2015) 298–305

Acknowledgements The authors would particularly like to thank all the participants for their generous collaboration. A full roster of the INMA Project Investigators can be found at http://www.proyectoinma.org/pre sentacion-inma/listado-investigadores/en_listado-investigadores. html.

Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.envres.2015.02. 026.

References Abdelouahab, N., Mergler, D., Takser, L., Vanier, C., St-Jean, M., Baldwin, M., Spear, P. A., Chan, H.M., 2008. Gender differences in the effects of organochlorines, mercury, and lead on thyroid hormone levels in lakeside communities of Quebec (Canada). Environ Res. 107, 380–392. Andersson, M., de, B.B., Delange, F., Zupan, J., 2007. Prevention and control of iodine deficiency in pregnant and lactating women and in children less than 2-yearsold: conclusions and recommendations of the technical consultation. Public Health Nutr. 10, 1606–1611. Barregard, L., Lindstedt, G., Schutz, A., Sallsten, G., 1994. Endocrine function in mercury exposed chloralkali workers. Occup. Environ Med 51, 536–540. Becker, D.V., Braverman, L.E., Delange, F., Dunn, J.T., Franklyn, J.A., Hollowell, J.G., Lamm, S.H., Mitchell, M.L., Pearce, E., Robbins, J., Rovet, J.F., 2006. Iodine supplementation for pregnancy and lactation—United States and Canada: recommendations of the American Thyroid Association. Thyroid 16, 949–951. Bjerregaard, P., Hansen, J.C., 2000. Organochlorines and heavy metals in pregnant women from the Disko Bay area in Greenland. Sci. Total Environ. 245, 195–202. Bjornberg, K.A., Vahter, M., Petersson-Grawe, K., Glynn, A., Cnattingius, S., Darnerud, P.O., Atuma, S., Aune, M., Becker, W., Berglund, M., 2003. Methyl mercury and inorganic mercury in Swedish pregnant women and in cord blood: influence of fish consumption. Environ. Health Perspect. 111, 637–641. Chen, A., Kim, S.S., Chung, E., Dietrich, K.N., 2013. Thyroid hormones in relation to lead, mercury, and cadmium exposure in the National Health and Nutrition Examination Survey, 2007–2008. Environ. Health Perspect. 121, 181–186. Ellingsen, D.G., Efskind, J., Haug, E., Thomassen, Y., Martinsen, I., Gaarder, P.I., 2000. Effects of low mercury vapour exposure on the thyroid function in chloralkali workers. J Appl. Toxicol. 20, 483–489. Glinoer, D., 2007. The importance of iodine nutrition during pregnancy. Public Health Nutr. 10, 1542–1546. Grandjean, P., Weihe, P., White, R.F., Debes, F., Araki, S., Yokoyama, K., Murata, K., Sorensen, N., Dahl, R., Jorgensen, P.J., 1997. Cognitive deficit in 7-year-old children with prenatal exposure to methylmercury. Neurotoxicol. Teratol. 19, 417–428. Guxens, M., Ballester, F., Espada, M., Fernandez, M.F., Grimalt, J.O., Ibarluzea, J., Olea, N., Rebagliato, M., Tardon, A., Torrent, M., Vioque, J., Vrijheid, M., Sunyer, J., 2012. Cohort profile: the INMA—INfancia y Medio Ambiente—(Environment and Childhood) Project. Int. J. Epidemiol. 41, 930–940. Haddow, J.E., Palomaki, G.E., Allan, W.C., Williams, J.R., Knight, G.J., Gagnon, J., O’Heir, C.E., Mitchell, M.L., Hermos, R.J., Waisbren, S.E., Faix, J.D., Klein, R.Z., 1999. Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N. Engl. J. Med. 341, 549–555. Jedrychowski, W., Perera, F., Jankowski, J., Rauh, V., Flak, E., Caldwell, K.L., Jones, R.L., Pac, A., Lisowska-Miszczyk, I., 2007. Fish consumption in pregnancy, cord blood mercury level and cognitive and psychomotor development of infants followed over the first three years of life: Krakow epidemiologic study. Environ. Int. 33, 1057–1062. Julvez, J., varez-Pedrerol, M., Rebagliato, M., Murcia, M., Forns, J., Garcia-Esteban, R., Lertxundi, N., Espada, M., Tardon, A., Riano, G.I., Sunyer, J., 2013. Thyroxine levels during pregnancy in healthy women and early child neurodevelopment. Epidemiology 24, 150–157. Kohrle, J., 1999. Local activation and inactivation of thyroid hormones: the deiodinase family. Mol. Cell Endocrinol. 151, 103–119. Langer, P., 2008. Persistent organochlorinated pollutants (PCB, DDE, HCB, dioxins, furans) and the thyroid—review 2008. Endocr. Regul. 42, 79–104. Laurberg, P., Cerqueira, C., Ovesen, L., Rasmussen, L.B., Perrild, H., Andersen, S., Pedersen, I.B., Carle, A., 2010. Iodine intake as a determinant of thyroid disorders in populations. Best Pract. Res. Clin. Endocrinol. Metab. 24, 13–27. Lederman, S.A., Jones, R.L., Caldwell, K.L., Rauh, V., Sheets, S.E., Tang, D., Viswanathan, S., Becker, M., Stein, J.L., Wang, R.Y., Perera, F.P., 2008. Relation between cord blood mercury levels and early child development in a World Trade Center cohort. Environ. Health Perspect. 116, 1085–1091. Lee, B.E., Hong, Y.C., Park, H., Ha, M., Koo, B.S., Chang, N., Roh, Y.M., Kim, B.N., Kim, Y. J., Kim, B.M., Jo, S.J., Ha, E.H., 2010. Interaction between GSTM1/GSTT1

polymorphism and blood mercury on birth weight. Environ. Health Perspect. 118, 437–442. Li, Y., Shan, Z., Teng, W., Yu, X., Li, Y., Fan, C., Teng, X., Guo, R., Wang, H., Li, J., Chen, Y., Wang, W., Chawinga, M., Zhang, L., Yang, L., Zhao, Y., Hua, T., 2010. Abnormalities of maternal thyroid function during pregnancy affect neuropsychological development of their children at 25–30 months. Clin. Endocrinol. 72, 825–829. Llop, S., Ballester, F., Vizcaino, E., Murcia, M., Lopez-Espinosa, M.J., Rebagliato, M., Vioque, J., Marco, A., Grimalt, J.O., 2010. Concentrations and determinants of organochlorine levels among pregnant women in Eastern Spain. Sci. Total Environ. 408, 5758–5767. Moleti, M., Di, B.B., Giorgianni, G., Mancuso, A., De, V.A., Alibrandi, A., Trimarchi, F., Vermiglio, F., 2011. Maternal thyroid function in different conditions of iodine nutrition in pregnant women exposed to mild-moderate iodine deficiency: an observational study. Clin. Endocrinol. 74, 762–768. Mori, K., Yoshida, K., Hoshikawa, S., Ito, S., Yoshida, M., Satoh, M., Watanabe, C., 2006. Effects of perinatal exposure to low doses of cadmium or methylmercury on thyroid hormone metabolism in metallothionein-deficient mouse neonates. Toxicology 228, 77–84. Morreale de, E.G., Obregon, M.J., Escobar del, R.F., 2004. Role of thyroid hormone during early brain development. Eur. J. Endocrinol. 151 (Suppl. 3), U25–U37. Morrissette, J., Takser, L., St-Amour, G., Smargiassi, A., Lafond, J., Mergler, D., 2004. Temporal variation of blood and hair mercury levels in pregnancy in relation to fish consumption history in a population living along the St. Lawrence River. Environ. Res. 95, 363–374. Muckle, G., Ayotte, P., Dewailly, E.E., Jacobson, S.W., Jacobson, J.L., 2001. Prenatal exposure of the northern Quebec Inuit infants to environmental contaminants. Environ. Health Perspect. 109, 1291–1299. Nohr, S.B., Laurberg, P., 2000. Opposite variations in maternal and neonatal thyroid function induced by iodine supplementation during pregnancy. J. Clin. Endocrinol. Metab. 85, 623–627. Orito, Y., Oku, H., Kubota, S., Amino, N., Shimogaki, K., Hata, M., Manki, K., Tanaka, Y., Sugino, S., Ueta, M., Kawakita, K., Nunotani, T., Tatsumi, N., Ichihara, K., Miyauchi, A., Miyake, M., 2009. Thyroid function in early pregnancy in Japanese healthy women: relation to urinary iodine excretion, emesis, and fetal and child development. J. Clin. Endocrinol. Metab. 94, 1683–1688. Pearce, E.N., Braverman, L.E., 2009. Environmental pollutants and the thyroid. Best Pract. Res. Clin. Endocrinol. Metab. 23, 801–813. Pop, V.J., Brouwers, E.P., Vader, H.L., Vulsma, T., van Baar, A.L., de Vijlder, J.J., 2003. Maternal hypothyroxinaemia during early pregnancy and subsequent child development: a 3-year follow-up study. Clin. Endocrinol. 59, 282–288. Pop, V.J., Kuijpens, J.L., van Baar, A.L., Verkerk, G., van Son, M.M., de Vijlder, J.J., Vulsma, T., Wiersinga, W.M., Drexhage, H.A., Vader, H.L., 1999. Low maternal free thyroxine concentrations during early pregnancy are associated with impaired psychomotor development in infancy. Clin. Endocrinol. 50, 149–155. Porterfield, S.P., Hendrich, C.E., 1993. The role of thyroid hormones in prenatal and neonatal neurological development—current perspectives. Endocr. Rev. 14, 94–106. Ramon, R., Murcia, M., Aguinagalde, X., Amurrio, A., Llop, S., Ibarluzea, J., Lertxundi, A., varez-Pedrerol, M., Casas, M., Vioque, J., Sunyer, J., Tardon, A., Martinez-Arguelles, B., Ballester, F., 2011. Prenatal mercury exposure in a multicenter cohort study in Spain. Environ. Int. 37, 597–604. Rebagliato, M., Murcia, M., Espada, M., Alvarez-Pedrerol, M., Bolumar, F., Vioque, J., Basterrechea, M., Blarduni, E., Ramon, R., Guxens, M., Foradada, C.M., Ballester, F., Ibarluzea, J., Sunyer, J., 2010. Iodine intake and maternal thyroid function during pregnancy. Epidemiology 21, 62–69. Rebagliato, M., Murcia, M., varez-Pedrerol, M., Espada, M., Fernandez-Somoano, A., Lertxundi, N., Navarrete-Munoz, E.M., Forns, J., Aranbarri, A., Llop, S., Julvez, J., Tardon, A., Ballester, F., 2013. Iodine supplementation during pregnancy and infant neuropsychological development: INMA Mother and Child Cohort Study. Am. J. Epidemiol. 177, 944–953. Soldin, O.P., O’Mara, D.M., Aschner, M., 2008. Thyroid hormones and methylmercury toxicity. Biol. Trace Elem. Res. 126, 1–12. Steuerwald, U., Weihe, P., Jorgensen, P.J., Bjerve, K., Brock, J., Heinzow, B., BudtzJorgensen, E., Grandjean, P., 2000. Maternal seafood diet, methylmercury exposure, and neonatal neurologic function. J. Pediatr. 136, 599–605. Takser, L., Mergler, D., Baldwin, M., de, G.S., Smargiassi, A., Lafond, J., 2005. Thyroid hormones in pregnancy in relation to environmental exposure to organochlorine compounds and mercury. Environ. Health Perspect. 113, 1039–1045. United Nations Environment Programme, 2007. Mercury Programme. Available: 〈http://www.chem.unep.ch/mercury/Report/JECFA-PTWI.htm〉 (accessed 13.04.13). Ursinyova, M., Uhnakova, I., Serbin, R., Masanova, V., Husekova, Z., Wsolova, L., 2012. The relation between human exposure to mercury and thyroid hormone status. Biol. Trace Elem. Res. 148, 281–291. US Environmental Protection Agency, 2007. Mercury. Human Exposure. Available: 〈http://www.epa.gov/mercury/exposure.htm〉 (accessed 13.04.07). Vahter, M., Akesson, A., Lind, B., Bjors, U., Schutz, A., Berglund, M., 2000. Longitudinal study of methylmercury and inorganic mercury in blood and urine of pregnant and lactating women, as well as in umbilical cord blood. Environ. Res. 84, 186–194. Vioque, J., Navarrete-Munoz, E.M., Gimenez-Monzo, D., Garcia-de-la-Hera, M., Granado, F., Young, I.S., Ramon, R., Ballester, F., Murcia, M., Rebagliato, M., Iniguez, C., 2013. Reproducibility and validity of a food frequency questionnaire among pregnant women in a Mediterranean area. Nutr. J. 12, 26.

S. Llop et al. / Environmental Research 138 (2015) 298–305

Willett, W.C., Sampson, L., Stampfer, M.J., Rosner, B., Bain, C., Witschi, J., Hennekens, C.H., Speizer, F.E., 1985. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am. J. Epidemiol. 122, 51–65. World Health Organization (WHO), 2007. Assessment of Iodine Deficiency Disorders and Monitoring Their Elimination. A Guide for Programme Managers, 3rd edn. International Council for the Control of Iodine Deficiency Disorders, Geneva, Switzerland. Xu, J., Yang, X.F., Guo, H.L., Hou, X.H., Liu, L.G., Sun, X.F., 2006. Selenium supplement alleviated the toxic effects of excessive iodine in mice. Biol. Trace Elem. Res. 111, 229–238.

305

Yang, X.F., Hou, X.H., Xu, J., Guo, H.L., Yinq, C.J., Chen, X.Y., Sun, X.F., 2006. Effect of selenium supplementation on activity and mRNA expression of type 1 deiodinase in mice with excessive iodine intake. Biomed. Environ. Sci. 19, 302–308. Yorita Christensen, K.L., 2013. Metals in blood and urine, and thyroid function among adults in the United States 2007–2008. Int. J. Hyg. Environ. Health 216, 624–632.