Authors:
Longnecker, MP., Gladen, BC., Patterson, DG., Rogan, WJ.

Title:
Polychlorinated Biphenyl (PCB) Exposure in Relation to Thyroid Hormone Levels in Neonates.

Journal:
Epidemiology; 11: 249-254

Summary:
Polychlorinated Biphenyls (PCBs) are a group of synthetic compounds that are a blend of up to 209 different individual chlorinated congeners. Their inflammability and insulating properties made them useful in electrical, heat transfer and hydraulic equipment. Experimental and epidemiologic studies have shown that PCBs can alter thyroid hormones in pregnant women and their offspring. They are structurally similar to thyroid hormones and interfere with thyroid metabolism through competitive binding to thyroid transport proteins, damage to the thyroid gland and induction of thyroid-metabolizing enzymes therefore warranting further investigation. Although they were banned in most countries over two decades ago for their adverse health effects and persistence, they remain detectable in the majority of humans.

The authors of this study drew a subset of subjects from an established cohort study that looked at early life organochlorine exposure in relation to neurodevelopment and other outcomes. The cohort consisted of 880 North Carolina mothers exposed to background levels of PCBs that delivered a total of 930 children between 1978 and 1982. Several specimens were collected for the purpose of the original study including: samples of the mother's serum at two times; breast milk samples at multiple intervals; umbilical cord blood; and placenta specimen. The samples were collected with glassware documented to be PCB and phthalate free. PCB levels were measured within 2 years of collection and cord serum was stored at -20 degrees Celsius. Subjects were included into the current study if cord serum specimens were still available after initial analysis (N=161) and if any measures of PCB exposure were available (N=160). PCB measures were then done by gas chromatography with electron capture. Thyroid function was assessed through cord serum thyroxine (T4), free T4, triiodothyronine (T3) and thyroid simulating hormone (TSH) measurements. Additionally, in order to facilitate comparison of the North Carolina exposures to those found in other studies, milk specimens that were collected soon after birth from 10 mothers were analyzed for PCB content. The specimens were selected on the basis so that PCB levels in the milk spanned the spectrum of the cohort from the 5th to 95th percentile.

The results showed that thyroid hormone levels in umbilical cord serum were not strongly correlated with in utero exposure to background levels of PCBs. Controlling for potential confounders did not alter the association between PCB level and cord hormone levels. The milk sample analysis showed that overall exposure levels to PCBs were comparable to a former Dutch study.

A strength of this study was the inclusion of measures of thyroid function as a potential mechanistic explanation for the potential health effects of PCBs. Thyroid hormones are essential for normal brain development in utero and in the early postnatal period. Unfortunately Longnecker et al. found that over 90% of the subjects had total T3 levels that were below the expected range, suggesting decomposition. Measurement error may have limited the findings of this study. At the time of hormone measurement the specimens were nearly 20 years old. Thus, although an interesting question, results of this study do not permit a conclusion to be reached concerning the effects of PCB exposure and thyroid function.