Impact of Exposure to Maternal PCBs and Dioxins on the Neonate’s Thyroid Hormone Status

The question of why human beings need thyroid hormone, as part of a complex regulatory system for their metabolism, is yet to be answered. Nevertheless, one can survive for years (“vegetate”) even in the absence of this hormone, and metabolism normalizes again after the thyroid hormone state has been restored. No such reversibility applies to brain development; indeed, lack of thyroid hormone during fetal life or during the first few years after birth and the subsequent irreversible brain damage make it clear that the developing brain is strongly dependent on thyroid hormone. 1 As early as the first trimester of pregnancy, fetal brain tissue expresses thyroid hormone receptors, 2 although at that time the fetal thyroid is still unable to produce significant amounts of hormone. 3 Thus, the fetus is totally dependent on maternal supply of thyroxine (T4) at least until midgestation. It is the task of the fetus and placenta together to regulate this supply throughout gestation, so that, initially, fetal needs are adequately fulfilled and, later on, the transfer of T4 is restricted in such a way that the fetal thyroid and its regulatory system in the hypothalamus and pituitary have the opportunity to develop as a self-supporting system. It is clear that, at term, the child is completely able to provide for its own T4 needs. Congenital hypothyroidism is a well-known cause of cerebral damage, resulting in impaired cognitive and motor development. The negative impact of overt maternal thyroid dysfunction (maternal thyroid disease, iodine deficiency, iodine excess) on fetal development has long been described. Recently, several articles focused on the possible harmful effects of apparently minor changes in maternal thyroid hormone levels during early pregnancy on the infant’s development. 4,5 In adults, such minor changes in the regulation of the thyroid hormone synthesis influence physical or mental well-being only slightly, reflected in terms like subclinical hyperthyroidism or hypothyroidism. However, even a low-normal maternal thyroid function appears to be associated with some loss of IQ points in the offspring. 6 The fact that low maternal free thyroxine (FT4) is related to impaired child-outcome does not necessarily mean that it is directly responsible. It might be hypothesized that low maternal FT4 itself is related to other aspects of (impaired) maternal-fetal transfer of thyroid hormone. Recently, delivery of maternal thyroid hormone to the fetal brain via iodothyronine sulfation and desulfation was shown to be a reliable alternative. 7,8 In fact, the supply of thyroid hormone to fetal tissues is a complex process, gradually changing throughout gestation, and even minor endogenous or exogenous disturbances might result in substantial brain dysfunction. Polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-para-dioxins and dibenzofurans (dioxins) are a large family of closely related chemicals, and share several toxic properties. PCBs were formerly used in a wide range of commercial products. Dioxins are unwanted byproducts formed during the process of synthesis and combustion of organochlorine chemicals. Because of their biochemical stability, these compounds persist in the environment and accumulate mainly in adipose tissue of animals and humans. A woman’s age appears to be related to the dioxin content of her breast milk, and there is a strong relation between dioxin concentration in milk fat and the consumption of animal, but not vegetable, fats, and proteins. 9 Although PCBs and dioxins can cause damage to many organs, one of the most promising areas of study appears to be their influence on the prenatal and postnatal thyroid hormone status, presumably because of the relation between brain development and thyroid function. Data about the influence of prenatal PCB/dioxin exposure on human brain development vary from no effect 10 or transient effects 11 to minor neuronal or psychomotor effects. 12–14 Longnecker and coworkers 15 report in this issue of Epidemiology on PCB exposure in relation to thyroid hormone levels in neonates born between 1978 and 1982 in North Carolina. The authors studied retrospectively a group of 160 children whose cord serum samples were stored, and whose prenatal PCB exposure was estimated on the basis of their mother’s breast milk and serum PCB levels. Infants were categorized into three groups according to PCB exposure. The measured concentrations of FT4 and thyroid stimulating hormone (TSH) in all three groups were well within the age-related normal values, although the median T4 and TSH levels, measured after some 20 years of storage at −20°C, were substantially lower than similar samples measured immediately, 16 presumably due to some degradation of the glycoproteins, thyroxine-binding globulin (TBG), and TSH. There was no relation between prenatal PCB exposure and cord serum FT4 levels and only a clinically unimportant difference (1 mU/L) in cord serum TSH levels between the group with the highest PCB exposure and the group with the lowest PCB exposure could be found. Moreover, there was definitely no inverse relation between the median values of FT4 and TSH of the three groups. These findings indicate that prenatal exposure to PCBs, at least in this population and in late gestation, does not lead to damage to the thyroid’s hormone production capacity and does not lead to measurable differences in binding to the circulating thyroid hormone binding proteins. The Netherlands is known as an area with a rather high exposure to PCBs and dioxins, and during the last decade two different studies were performed with a similar design as the study by Longnecker and coworkers. Pluim et al.17,18 studied 38 healthy term breast-fed infants, divided into two groups according to level of dioxin exposure in breast milk. Mean neonatal serum T4 concentrations and T4/TBG ratios (representing FT4 concentration) were higher in the high-exposure group. This group also had a slightly higher cord serum TSH concentration compared with the low-exposure group. Koopman-Esseboom et al.19 studied a similar group of 78 Dutch children. They found slightly lower serum FT4 levels and slightly higher TSH levels in the high-exposure group of neonates compared with the low-exposure group, within the age-related normal range. Considering both Dutch studies, there were no differences in the neonatal FT4 and TSH levels. Thus, proof is lacking that differences in mental or motor development correlated to PCB exposure are mediated by differences in the late-gestation fetal thyroid function. Taking the results of the U.S. and Dutch studies together, it appears that the measured effects on determinants of the thyroid hormone status might be due to chance. It cannot be excluded, however, that an excessive prenatal PCB and dioxin exposure might deteriorate the neonate’s T4 supply. Also, there is no information available on toxic effects to the fetus’ thyroid hormone supply in early pregnancy, nor on the thyroid hormone metabolism in fetal brain tissue. Therefore, it is still possible that the cerebral problems measured in some high-exposure children 11–14 are the result of negative influences on the thyroid hormone supply of the fetal brain. Since a substantial part of the T4 circulating in the fetus is of maternal origin, 20,21 and decreased or low-normal maternal FT4 levels during the first trimester seem to be associated with impaired child development, 22 it is important to investigate the maternal thyroid hormone status and the placental transfer of T4 in relation to PCB/dioxin exposure. Koopman-Esseboom et al.19 only measured the maternal thyroid hormone status during the last month of pregnancy; they found no meaningful correlation between breast milk PCB content and maternal serum FT4 and TSH concentrations. Data about earlier stages of pregnancy are lacking. An intriguing aspect of the influence of PCBs and dioxins on thyroid pathology is the generation of autoantibodies against thyroid tissue, 23 especially since children of healthy mothers who had high serum TPO-Ab concentrations during pregnancy end up with an IQ deficit of 10 points compared with children of mothers without TPO-Ab. 24 The presence of TPO-Ab in young women is not a rare condition, which suggests that a substantial proportion of offspring is at risk. Women with high levels of TPO-Ab had a sixfold increased risk of presenting with relatively low FT4 levels in early gestation, which is also a risk factor for impaired psychomotor development in the offspring. 4–6 It is still unclear why healthy children of healthy mothers who developed in a relatively hypothyroxinemic fetal environment are at risk for minor brain damage. 5 Iodine deficiency might play a role, 25 but other nutritional or environmental factors have to be taken into account. Only a large-scale, long-term prospective study will show convincingly that brain development and cerebral thyroid hormone metabolism are influenced by prenatal and postnatal PCB/dioxin exposure, and determine the pathogenetic mechanisms. The results of studies thus far are not that impressive that such a costly project should be a priority. As far as the effect of infant feeding on mental and psychomotor outcome, it is reassuring that even breast milk with a rather high PCB/dioxin content is as good as formula based on cow milk with a low PCB/dioxin content. 11