Long-term environmental exposure to perchlorate through drinking water and thyroid function during pregnancy and the neonatal period.

Tellez, R.T., Chacon, P.M., Abarca, C.R., Blount, B.C., Van Landingham, C.B., Crump, K.S. and Gibbs, J.P. (2005) Thyroid, 15 (9); 963-75.

Recently concern has been raised about the possibility of environmental perchlorate in drinking water inducing a relative iodine deficiency during pregnancy, especially if women have a low iodine intake. Such a deficiency may result in adverse neurodevelopmental effects in the foetus. A longitudinal epidemiologic study was conducted to assess the possible impact of perchlorate in drinking water at concentrations of 100-120 µg/L on maternal thyroid function during pregnancy, neonatal thyroid function and developmental status at birth, and breast milk iodide and perchlorate levels during lactation. It is hypothesised that long-term exposure to perchlorate in drinking water will cause a situation similar to iodine deficiency and therefore cause increases in thyrotropin (TSH) and thyroglobulin (Tg) levels and decreased levels of free thyroxine (FT 4 ) in either the mother during early stages of gestation or the neonate at birth, or cause growth retardation in the foetus

The study was conducted between November 2002 and April 2004 in three coastal cities in northern Chile where saltpeter, which contains perchlorate, has been mined for over 200 years. The three cities studied were Taltal with 114 µg/L, Chanaral with 6 µg/L, and Antofagasta with 0.5 µg/L perchlorate in their public drinking water supplies. Women were enrolled in the study when first presenting for prenatal care. Women were excluded if they were beyond 24 weeks gestation, had lived in their respective cities less than 6 months, or were taking thyroid medications or medication containing iodine during the 3 months before starting the study.

During the first prenatal study visit subjects answered a questionnaire on socioeconomic and cultural information and provided a sample of home tap water. During the post-partum study visit, a breast milk sample was collected when possible. During both visits, maternal serum and urine samples were taken and analysed. Levels of TSH, free FT 4 and triiodothyronine (T 3) were measured using chemiluminescence. Tg was analysed by radioimmune separation and antibodies to thyroid peroxidase (TPO) were analysed using hemagluttination. Urine iodine determinations were made using oxidation with ammonium persulfate. A physical examination of maternal thyroid was conducted during both study visits. Tap water samples were analysed for perchlorate. There were also 63 serum samples and 297 urine samples analysed for perchlorate. All breast milk samples were analysed for perchlorate and for iodine. For all hospital births, neonatal weight, length and head circumference were measured and gestational age at birth was estimated.

To determine if nitrate, selenium, iodine, lithium and arsenic were present in mineral deposits in the study region, some of the home tap water samples from each of the three cities were analysed for these substances. These substances have been linked to thyroid dysfunction and might confound the results. Postpartum serum samples from 40 nonsmokers were analysed for thiocyanate to ensure the local diet was not high in cyanogens.

No increases in Tg or TSH and no decreases in FT 4 were found among women during early pregnancy, late pregnancy or neonates at birth in relation to perchlorate in drinking water. The birth weight, length and head circumference of neonates were similar among the three cities and were similar to the distribution reported in the National Health and Nutrition Examination Survey in the U.S (NHANES III) in the early 1990s. Therefore perchlorate levels in drinking water found here did not cause changes in neonatal thyroid function or foetal growth retardation. The levels of arsenic, lithium, nitrate and iodine were not sufficiently high to cause concern about confounding.

The mean urinary iodine level of the entire cohort was 269 µg/L which is intermediate between that of pregnant women in the U.S. NHANES I in the early 1970s and NHANES III, and also consistent with current World Health Organization (WHO) recommendations. It has been hypothesised that perchlorate from the environment might cause a decrease in breast milk iodine concentrations however this was not the case with no decrease in iodine levels found in cities with detectable perchlorate. Maternal urinary perchlorate excretion in Antofagasta and Chanaral were higher than expected if drinking water was the only source of perchlorate, indicating that a significant additional dietary source of perchlorate is present.