COT statement on the health hazards of polychlorinated biphenyls (April 1997)

The COT was asked to review the toxicity of the polychlorinated biphenyls (PCBs). It was also asked to comment on the concentrations of these, with polychlorinated dibenzo- p-dioxins and dibenzofurans (PCDDs and PCDFs), that had been found in the UK diet and in human breast milk in the course of surveillance exercises carried out by the Ministry of Agriculture, Fisheries and Food.

The COT welcomed the decline in the level of these contaminants present in the diet and in human milk but noted that the potential intake of PCBs, PCDDs and PCDFs by breast fed infants or by toddlers consuming bottled fish oils was undesirably high. However, this intake of the contaminants is not dangerous and the COT reaffirmed support for breast feeding because of its well-established benefits to infants.

The Committee's Conclusions are listed below.

Conclusions
i. PCBs produce a wide spectrum of adverse effects in experimental animals, including reproductive toxicity, immunotoxicity and carcinogenicity.

ii. We accept the advice of the COM that PCBs do not have significant mutagenic activity and that any carcinogenesis in animal studies is likely to be due to a 'non-genotoxic' mechanism.

iii. We accept the advice of the COM and COC that it would be prudent to assume that all PCB congeners are potential human carcinogens.

iv. We note the results of preliminary work which indicates that current human body burdens of PCBs may be affecting thyroid hormone levels. We consider that further work in a larger population is required to investigate this further.

v. We conclude that the available epidemiological studies of individuals exposed to high background dietary levels of PCBs are not of sufficient quality to allow a quantitative estimate of risk to be made of reproductive dysfunction following dietary exposure.

vi. Most animal studies have been conducted using commercial mixtures of PCBs. The individual congeners and the relative proportion of these congeners in these mixtures are different from those found in food. There are only limited data on individual congeners. For these reasons, we have not been able to set a Tolerable Daily Intake for total PCBs.

vii. We are of the opinion that, despite the limitations in the derivation of the TEFs (see paragraphs 5 and 21 of the main text), the use of TEFs to assess the health risks of certain coplanar ('dioxin-like') PCB congeners offers a pragmatic approach to the evaluation of these compounds and we recommend that TEFs be tentatively accepted and used in a limited manner for this purpose. We recommend that the TEFs proposed by the 1994 WHO-ECEH/IPCS task force are used at present and that a TDI of 10 pg TEQ/kg bw/day can be employed to assess the health risks of combinations of PCDDs, PCDFs and dioxin-like PCB congeners. Considerable further work is necessary to improve the scientific basis for validating TEFs.

viii. Further work is needed to develop an approach to assessing the health risks of the non-coplanar PCB congeners.

ix. We welcome the new data on UK dietary intakes of both total and individual congeners of PCBs provided by MAFF. We note that the data on intakes by toddlers are provisional and suggest that further work is carried out to give more reliable estimates of intake by this age group. We consider that there is unlikely to be a health risk from current intakes of PCBs from food.

x. We welcome the new data on levels of PCBs in human milk. Although intakes of PCBs by breast-feeding babies are higher than is desirable, we reaffirm our previous advice that breast-feeding should continue to be encouraged on the basis of convincing evidence of the benefits of human milk to the overall health and development of the infant.

xi. We note the data on levels of PCBs, PCDDs and PCDFs in fish oil dietary supplements and medicinal products. The intake of such levels is undesirable, since it potentially leads to the TDI for PCBs, PCDDs and PCDFs being exceeded by toddlers and schoolchildren for a sustained period and thus reduces the safety margin between intake and the toxicity observed in animal studies. However, we consider that this intake is unlikely to pose a risk to health.

xii. We recommend that it would be desirable to carry out studies to improve the understanding of PCB, PCDD and PCDF accumulation and disposition in humans in the first year of life, eg by physiologically-based toxicokinetic modelling.

xiii. We consider that it would be prudent to seek action to ensure that consumer exposure to these contaminants is kept below the TDI. However, we recognise that PCBs are likely to persist as contaminants of the environment for many years to come. Therefore, we recommend that levels in food and in human milk should continue to be monitored at regular intervals to confirm that the downward trend continues. If it does not, we recommend that a further review is instigated to determine how human exposure can be reduced.