Associations between perfluoroalkyl substances and thyroid hormones after high exposure through drinking water

https://doi.org/10.1016/j.envres.2020.110647Get rights and content

Highlights

  • Fire-fighting foam contamination led to high PFOS and PFHxS in drinking water.

  • Thyroid hormones were compared in exposed and reference groups.

  • Associations between thyroid hormones and serum PFAS were studied.

  • No associations were found in adults and seniors.

  • Inconsistent associations were found in preteens and teenagers.

Abstract

Background

The reported associations for several per- and polyfluoroalkyl substance (PFAS) with thyroid hormones are inconsistent in epidemiological studies. The purpose of the current study was to investigate the possible association of thyroid hormones in relation to serum levels of perfluorohexane sulfonate, perfluorooctane sulfonate and perfluorooctanoic acid, in a Swedish general population, highly exposed through contaminated drinking water, and if the associations with PFAS remained in a comparison to a reference group based only on residency in areas with contrasting PFAS levels.

Method

3297 participants from Ronneby, a municipality with drinking water highly contaminated by PFAS (exposed group), and a reference group (N = 226) from a nearby municipality with non-contaminated drinking water supply were included. Regression analysis was used to investigate the associations between PFAS exposure, assessed as exposure groups (Ronneby and reference groups) and measured serum PFAS levels, and thyroid hormone levels, with adjustments for age, sex and BMI.

Result

No cross-sectional associations were found between PFAS and thyroid hormones in adults and seniors except for a positive association between PFAS and fT4 in males over 50. Higher thyroid hormone levels were found in the preteen children from Ronneby compared to the reference group. In contrast, within Ronneby, there was weak evidence of associations between increased PFAS levels and decreased fT3 in preteen boys, and decreased TSH in teenage males. No such pattern was found in preteen and teenage girls.

Conclusion

The present study found no consistent evidence to support association of PFAS with thyroid hormones.

Introduction

Per- and polyfluoroalkyl substances (PFAS) are ubiquitous in the environment and human population, due to their extensive use, persistence and ability to bioaccumulation (Conder et al., 2008). Human background exposure comes mainly from diet and the indoor environment (Eriksson et al., 2013; Gebbink et al., 2015; Sjogren et al., 2016; Vestergren et al., 2012). In some local populations, PFAS-contaminated drinking water originating from fire-fighting training areas and PFAS using or production facilities, has been the principal significant source of PFAS exposure (Domingo et al., 2019; Ericson et al., 2008; Frisbee et al., 2009; Glynn et al., 2020; Gyllenhammar et al., 2015; Ingelido et al., 2018.).

On December 16, 2013, it was unexpectedly discovered that one out of two municipal waterworks in Ronneby, southern Sweden, had alarmingly high levels of PFAS. The contamination source was a nearby military airfield, where Aqueous Film-Forming Foam (AFFF) was used, probably since the mid-1980s. About one-third of the households in Ronneby had been supplied drinking water with PFAS for decades. Biomonitoring in more than 3200 individuals of the altogether 28,000 inhabitants in the municipality started in June 2014 (approximately 6 months after the provision of clean water), and comprised the Ronneby Biomarker Cohort. An early report from this cohort showed very high serum levels of perfluorohexane sulfonate (PFHxS) of a median 152 ng/ml, perfluorooctane sulfonate (PFOS) of median 176 ng/ml and to a lesser extent, perfluorooctanoic acid (PFOA) of median 10 ng/ml (Li et al., 2018). Although clean water was provided promptly following the discovery of the contamination, the potential health concern remains due to the long serum half-lives of these PFAS, which have been estimated to be mean 5.3, 3.4 and 2.7 years for PFHxS, PFOS and PFOA, respectively, as calculated for small subset of the population in Ronneby (Li et al., 2018).

Potential effects of PFAS on the thyroid hormone system have gained attention in research. Thyroid hormones play an important role in human metabolism, growth and maturation. Thyroid homeostasis is controlled by the hypothalamus-pituitary-thyroid axis feedback mechanism, which depends on the interaction between thyroid stimulating hormone (TSH), thyroxin (T4) and triiodothyronine (T3) (Zoeller et al., 2007). A number of epidemiological studies, mostly with a cross-sectional design, have investigated thyroid hormones in association with PFAS exposure in the general population (Audet-Delage et al., 2013; Bloom et al., 2010; Dallaire et al., 2009; Jain 2013; Ji et al., 2012; Lewis et al., 2015; Li et al., 2017; Lin et al., 2013; Shrestha et al., 2015; Wen et al., 2013) and high exposed mainly to PFOA groups (in high exposure communities, Knox et al., 2011; Lopez-Espinosa et al., 2012; and in a high occupationally exposed group, Olsen and Zobel 2007). Overall, there is inconsistent evidence of associations between any particular PFAS and thyroid hormones (EFSA 2018). Notably, a registry-based cohort study from Ronneby of approximately 63,000 persons of which 16,000 had resided in the contaminated water district, did not find an increased risk for hypothyroidism or hyperthyroidism, neither in men nor women (Andersson et al., 2019). These inconsistent findings limit current understandings of the potential effects of PFAS on thyroid homeostasis. Furthermore, cross-sectional studies cannot establish whether there is a causal effect of PFAS on thyroid function.

The present study aimed to advance knowledge about thyroid function in relation to PFAS exposure by investigating associations between PFHxS, PFOS and PFOA and thyroid hormones [free T3 (fT3), free T4 (fT4) and TSH] in the Ronneby Biomarker Cohort, and a reference population from a nearby municipality without PFAS-contaminated water supply. With the current cross-sectional study, we examined if the associations were presented in ecological comparison, i.e. by the comparison based only on residency in areas with contrasting exposure to PFAS, and if associations were presented for serum PFAS, within the Ronneby Biomarker Cohort. We also hypothesized that younger participants with early-life exposure may show stronger associations between PFAS and thyroid hormones.

Section snippets

Study population

The Ronneby Biomarker Cohort consists of 3297 participants (aged 1–94 years) from Ronneby, who attended extensive biomonitoring by open invitations and free of cost from June 2014 to December 2015. In addition to recruitment in the high exposure area, considerable efforts were made to recruit persons thought to have low PFAS exposure, i.e. who did not live in the area with highly contaminated water supply area in Ronneby, in order to ensure a broad range of serum PFAS levels. In addition, a

Results

The distributions of the serum PFAS and thyroid hormone levels across all age and sex strata are shown for the Ronneby Biomarker Cohort (Table 1) and the reference group (Table 2). The median levels of PFAS (i.e. PFOS, PFHxS and PFOA) in the exposed group were 4 to 300-fold higher than in the reference group, with PFHxS having the most dramatic contrast. The serum PFAS levels in the reference group were comparable to the general population in Sweden. The median age and BMI in the Ronneby

Discussion

The present study investigated a relatively large population aged from 1 to 94 years, with a wide range of PFAS exposure, dominated by PFOS and PFHxS. We observed that PFAS serum levels were positively associated with fT4 in senior males. In the preteen children, we found some evidence for associations between PFAS and thyroid hormones. However, the findings from the area-based comparison between exposure groups (e.g. higher thyroid hormones in the children from Ronneby compared to the

Conclusions

In conclusion, we did not observe consistent evidence suggesting effects of PFAS on thyroid hormones, especially not in adults. The possible inverse association between serum PFAS levels and fT3 in preteen boys was contradicted by the finding of higher thyroid hormones levels in Ronneby (exposed group) compared to the non-exposed reference group, therefore, we do not consider that there is evidence to support a causal interpretation of either of these associations. While we do not find

Credit author statement

Ying Li: Methodology, Software, Formal analysis, Writing-original draft, Writing-review & editing. Yiyi Xu: Methodology, Writing-original draft, Writing-review & editing. Kristin Scott: Project administration, Writing-review & editing. Christel Nielsen: Writing - review & editing. Daniela Pineda: Resources, Laboratory work, Writing - review & editing. Christian H. Lindh: Resources, Laboratory work, Writing - review & editing. Eva M Andersson: Writing-review & editing. Daniel S. Olsson: Writing

Funding

This work was supported by FORMAS, Sweden (dnr 216-2014-1709) and FORTE, Sweden (dnr 2015-732).

Ethical approval

The study was approved by the Regional Ethical Review Board in Lund, Sweden. All participants have given written consent for the current study.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors are grateful to Ronneby residents who took part in the study and the field and laboratory staffs for their assistance with sample collection and analysis.

References (43)

  • K. Ji et al.

    Serum concentrations of major perfluorinated compounds among the general population in Korea: dietary sources and potential impact on thyroid hormones

    Environ. Int.

    (2012)
  • C.-Y. Lin et al.

    The associations between serum perfluorinated chemicals and thyroid function in adolescents and young adults

    J. Hazard Mater.

    (2013)
  • C.H. Lindh et al.

    Blood serum concentrations of perfluorinated compounds in men from Greenlandic Inuit and European populations

    Chemosphere

    (2012)
  • S. Shrestha et al.

    Perfluoroalkyl substances and thyroid function in older adults

    Environ. Int.

    (2015)
  • P. Sjogren et al.

    Circulating levels of perfluoroalkyl substances are associated with dietary patterns–a cross sectional study in elderly Swedish men and women

    (2016)
  • R. Vestergren et al.

    Dietary exposure to perfluoroalkyl acids for the Swedish population in 1999, 2005 and 2010

    (2012)
  • C. Virili et al.

    “With a little help from my friends”-the role of microbiota in thyroid hormone metabolism and enterohepatic recycling

    (2017)
  • Y. Audet-Delage et al.

    Persistent organic pollutants and transthyretin-bound thyroxin in plasma of Inuit women of childbearing age

    Environ. Sci. Technol.

    (2013)
  • J.M. Conder et al.

    Are PFCAS bioaccumulative? A critical review and comparison with regulatory criteria and persistent lipophilic compounds

    Environ. Sci. Technol.

    (2008)
  • R. Dallaire et al.

    Thyroid function and plasma concentrations of polyhalogenated compounds in inuit adults

    Environ. Health Perspect.

    (2009)
  • C.M. Dayan et al.

    Chronic autoimmune thyroiditis

    N. Engl. J. Med.

    (1996)
  • Cited by (16)

    • Association between serum levels of TSH and free T4 and per- and polyfluoroalkyl compounds concentrations in pregnant women

      2023, Journal of Environmental Sciences (China)
      Citation Excerpt :

      Maternal thyroid dysfunction has been associated with cognitive deficits (Lyall et al., 2017), preterm delivery and neurodevelopmental deficits (Escobar et al., 2004). However, the results of epidemiological studies on the association between PFASs exposure and thyroid hormone levels are often inconsistent (Wang et al., 2013, 2014; Preston et al., 2018; Andersson et al., 2019; Aimuzi et al., 2020; Silva et al., 2020; Li et al., 2021). Some studies have reported that maternal perfluorohexanesulfonic acid (PFHxS) and PFOS concentration were positively associated with thyroid stimulating hormone (TSH) levels (Wang et al., 2013, 2014).

    • Target analysis and suspect screening of per- and polyfluoroalkyl substances in paired samples of maternal serum, umbilical cord serum, and placenta near fluorochemical plants in Fuxin, China

      2022, Chemosphere
      Citation Excerpt :

      Serum H-PFCA (C11, C12), 6:2 Cl-PFESA, and Cl-PFCA (C9, C10) in newborns were positively associated with FT3 levels in cord sera (p = 0.034, p = 0.031, p = 0.013, p = 0.018, p = 0.021), while serum levels of FT4 would also be associated with increased 5:2 Cl-PFESA and UPFCA (C8) (p = 0.030, p = 0.033). Some studies earlier reported that PFOS, PFOA, and PFNA could elevate FT3 in cord sera and PFOS and PFDoA might have positive associations with serum FT4 in newborns (Aimuzi et al., 2019; Guo et al., 2021; Liang et al., 2020), although strong associations of legacy PFASs with TSH, FT3, or FT4 in cord sera were not supported by other recent studies (Lebeaux et al., 2020; Li et al., 2021). Hence the thyroid hormone levels in newborns could be associated with serum Cl-PFESAs, serving as the alternatives of PFOS, and serum H–PFCAs, Cl–PFCAs, and UPFCAs, being the substitutes for long-chain PFCAs.

    • Reproductive and developmental toxicity of perfluorooctane sulfonate (PFOS) in the white-footed mouse (Peromyscus leucopus)

      2022, Reproductive Toxicology
      Citation Excerpt :

      Therefore, there is no clear conclusion that thyroid structure is affected at the histological level. Importantly, the scientific community recognizes that PFAS exposure is linked to thyroid modulation—although studies have found positive correlations [14,48,7], negative correlations [24,25,41,57], or no clear association between blood PFAS concentrations and thyroid hormones [8,9,23,31,42,54,55]. Further, thyroid endpoints were unaffected by PFAS in three species of amphibians [20].

    View all citing articles on Scopus
    View full text