Scientific Opinion on Polybrominated Biphenyls (PBBs) in Food

EFSA was asked by the European Commission to deliver a scientific opinion on polybrominated biphenyls (PBBs) in food. PBBs are additive flame retardants which were applied in synthetic fibres and polymers. PBBs are present in the environment at low concentrations and likewise in biota and in food and feed. Data from the analysis of 16 PBB congeners in 794 food samples were provided to EFSA by 6 Member States, covering the period from 2003 to 2009. Toxicity studies were carried out with technical PBB mixtures of which the exact congener composition is not known. Main targets were the liver, thyroid hormone homeostasis and the reproductive, nervous and immune systems. PBBs are not directly genotoxic. The Panel on Contaminants in the Food Chain (CONTAM Panel) selected the hepatic carcinogenic effects as the critical effect, with a no-observed-effect level (NOEL) of 0.15 mg/kg body weight (b.w.). Since this NOEL was obtained in a study with a technical PBB mixture, the congener profile of which differs from that currently found in food, the CONTAM Panel concluded that it was inappropriate to use this NOEL to derive a health based guidance value. The intake of PBBs by high and frequent consumers of fatty fish, the subgroup with the highest dietary exposure, was approximately 6 orders of magnitude less than this NOEL. Exposure for high consuming breast-fed infants is 5 orders of magnitude less than this NOEL. Therefore the CONTAM Panel concluded that the risk to the European population from exposure to PBBs through the diet is of no concern. Since PBBs are no longer produced or used in Europe and taking into account low and declining environmental concentrations, the CONTAM Panel concluded that PBBs are a low priority for further research or monitoring efforts.

© European Food Safety Authority, 2010

Following a request from the European Commission, the Panel on Contaminants in the Food Chain (CONTAM Panel) was asked to deliver a scientific opinion on polybrominated biphenyls (PBBs) in food.

PBBs are additive flame retardants which were specially applied in synthetic fibres and polymers. As they are not chemically bound to the polymers, they can leach into the environment. PBBs were produced until the mid 1980s, except DecaBB which was produced up till around 2000.

PBBs are a class of brominated hydrocarbons with a basic structure consisting of two phenyl rings to which bromine atoms are attached. There are 209 possible compounds, referred to as PBB congeners, which differ in the number and position of the bromine atoms in the two phenyl rings. Like polychlorinated biphenyls (PCBs), the benzene ring can rotate around the central bond that connects both phenyl rings adopting a planar and a non-planar configuration depending on the degree of substitution in the ortho positions. PBBs where the hydrogen atoms in the ortho positions are substituted by bromine atoms are called ortho PBBs, and those where the hydrogen atoms in the ortho positions are not substituted by bromine are called non-ortho PBBs. This difference in molecular structure is relevant for the interaction with different receptors determining the toxicological properties of PBBs.

PBBs are lipophilic compounds with a low vapour pressure and low water solubility which decreases with increasing degree of bromination. They are generally chemically stable, persistent in the environment and bioaccumulative. It has been reported that higher brominated biphenyls can undergo photolysis and reductive debromination, thereby producing lower brominated congeners.

In 1983 it was regulated in the European Union (EU) that PBBs may not be used in textile articles, such as garments, undergarments and linen, intended to come into contact with the skin, and in 2006 regulation was placed in force ensuring that electrical and electronic equipment put on the market does not contain PBBs. However, a maximum concentration value of 0.1 % of PBBs by weight in homogeneous materials shall be tolerated.

PBBs are present in the environment at low concentrations and likewise in biota and in food and feed.

Following a CONTAM Panel advice, a monitoring exercise was carried out from 2006 and results obtained from the analysis of 16 PBB congeners on 794 food samples were provided to EFSA by 6 Member States, covering the period from 2003 to 2009.

The CONTAM Panel reviewed the available data on composition of technical mixtures, occurrence in food and toxicology of the various PBB congeners. Based on the reported data on occurrence in food, the Panel put special emphasis on the PBB congeners BB-3, -15, -29, -49, -52, -77, -80, -101, -103, 126, -153, -169, -180, -194, -206 and -209.

The food category “Fish and other seafood (including amphibians, reptiles, snails and insects)” dominated the total samples, followed by “Meat and meat products (including edible offal)” and “Animal and vegetable fats and oils” and “Milk and dairy products”. The data were characterised by a high proportion of non detects for the various congeners (overall more than 80 %), with some food categories, i.e. “Animal and vegetable fats and oils”, “Milk and dairy products”, close to 100 % non detects. The lowest proportion of non detects was reported for the food category of “Fish and other seafood (including amphibians, reptiles, snails and insects)”. For 7 congeners with a proportion of non detects below 90 % (BB-49, -52, -77, -80, -101, -153 and -209) analysed in the specific food category of “Fish meat”, an increasing fat content corresponded with increasing PBB levels (except for BB 209).

Due to the high proportion of non detects for certain food categories and certain congeners, the estimation of upper and lower bound was driven by the reported limits of detection (LODs) or limits of quantification (LOQs). Therefore, the CONTAM Panel decide to focus only on a restricted list of food categories including “Fish and other seafood”, “Meat and meat products”, “Animal and vegetable fats and oils”, “Milk and dairy products” and “Food for infants and small children”. Additionally, in order to allow for a more reliable application of the upper and lower bound approach and to prevent an unrealistic exposure estimate, the exposure estimation was performed only on those congeners in the respective food categories where the proportion of non detects was lower than 80 %.

Limited studies have been published on the toxicokinetics of PBBs, demonstrating that gastrointestinal absorption of PBBs may occur to a significant extent (90 % of the dose for BB-153 in rodents) and that highly brominated congeners may accumulate in lipid-rich tissues. There is evidence for debromination and hydroxylation of PBBs. Limited data indicate that the apparent half life of BB 153 in rats varies between 9 and 69 weeks, depending on the tissue concerned. Epidemiological data indicate that the median serum half life of PBBs in humans varies between about 10 and 30 years.

The toxicological studies on PBBs date back decades ago, reflecting the phase-out of the manufacture and use of PBBs. Oral toxicity studies were carried out with technical PBB mixtures of which the exact composition of congeners is not known. Main targets were the liver, the reproductive system, thyroid hormone homeostasis and the nervous and immune systems.

PBBs have low acute oral toxicity, with lethal dose (LD₅₀) values > 1,000 mg/kg body weight (b.w.) after single exposure. After repeated exposure (60 days), lethality was in the range of 65 150 mg/kg b.w.

PBBs (i.e. FireMaster preparations) caused liver enlargement, hepatocellular hypertrophy, fatty degeneration and enzyme induction in experimental animals. Evidence from animal studies indicate that exposure to PBBs influences the thyroid hormone homeostasis. The observed effects include decreases in serum levels of thyroid hormones (thyroxine (T4) and triiodothyronine (T3)), elevated thyroid stimulating hormone (TSH) levels, thyroid enlargement and morphological changes in follicular cells. Based on the available data there is evidence that PBBs affect neurobehavioral development and the immune system. These effects occur at slightly higher levels than those on liver and thyroid hormones. Exposure to PBBs during early pregnancy can lead to resorption of foetuses and foetal malformations.

In vitro and in vivo genotoxicity studies indicate that PBBs are not directly genotoxic.

PBBs are carcinogenic in the liver of rodents, by a non-genotoxic mode of action, which is assumed to have a threshold in the dose-response curve, with a no-observed-effect level (NOEL) of 0.15 mg/kg b.w. There is evidence that ortho substituted congeners may cause cancer through interaction with nuclear receptors, such as the constitutive androstane receptor (CAR), whereas the non-ortho congeners appear to cause tumours as a consequence of arylhydrocarbon receptor (AhR) activation and cytotoxicity, presumably via stimulation of regenerative proliferation.

The technical PBB mixtures used in the different toxicity tests comprise both ortho and non-ortho substituted congeners. The non-ortho congeners have been shown to activate the AhR receptor and a number of the toxic effects observed are consistent with dioxin-like activity. There is some evidence that ortho-substituted PBB congeners can activate other receptors such as CAR and pregnane X receptor (PXR). Activation of these receptors can lead to increased catabolism of thyroid hormones. The effects on the liver including hepatocarcinogenesis and on the thyroid hormone homeostasis may be a consequence of such receptor activation.

Epidemiological studies indicate that there are some associations between exposure to PBBs and changes in health, such as neurodevelopmental effects, site-specific cancer and effects on fertility and offspring. However, these findings were limited and inconsistent, and confounding by other compounds and/or lifestyle factors and limitations in the study design hamper interpretation of the epidemiological results.

In considering the available toxicological information, the CONTAM Panel selected the hepatic carcinogenic effects of PBBs as the critical effect for the derivation of a reference point for gauging the potential health risks of dietary exposure to PBBs. The NOEL for this end-point is 0.15 mg/kg b.w. The CONTAM Panel noted however, that this NOEL represents a worst case situation as it was obtained in a study with a technical PBB mixture, the congener composition of which is not representative of the congener profiles currently found in food. Therefore, the CONTAM Panel concluded that it was inappropriate to use this NOEL to derive a health based guidance value for PBBs.

The CONTAM Panel identified a specific group of the population comprising high and frequent fish consumers consuming fatty fish meat (>8 % fat) as those with the highest exposure to PBBs in the diet of all of the subgroups considered, other than breast-fed infants. The upper bound estimate of exposure to the sum of the 5 PBB congeners (BB-49, -52, -77, -101 and -153) for which the percentage of non detects was less than 80 % is 0.15 ng/kg b.w. per day. Compared to the NOEL for hepatocarcinogenesis of 0.15 mg/kg b.w. per day in rats, the CONTAM Panel noted that exposure in this specific high consumer group was approximately 6 orders of magnitude less than this NOEL. Dietary exposure of all other groups of the population even at the upper bound for the high consumers, was appreciably lower than that of the frequent and high fish consumers group.

The mean exposure for infants with high human milk consumption was in the region of 0.9 to 1.4 ng/kg b.w. per day. This is 5 orders of magnitude less than the toxicological reference point.

Due to the potential toxicological concerns related to the non-ortho PBBs, additional exposure estimates were performed for the three congeners BB-77, -126 and -169. A calculation for high consumers (95^th percentile) based on median upper bound concentrations resulted in exposures of around 0.3 pg/kg b.w. per day for the sum of the three congeners. Assuming similar toxicity equivalency factors (TEF) as for non-ortho PCBs, the exposure to non-ortho PBBs was estimated to be in the region of 0.01 pg toxicity equivalents (TEQ)/kg b.w. per day. Compared to background exposure of the European population to dioxins and dioxin-like compounds the CONTAM Panel considered this highly overestimated exposure to non-ortho PBBs as negligible.

The CONTAM Panel concluded that the risk to the European population from exposure to PBBs through the diet in Europe, even considering the difference in half-lives between rats and humans, is of no concern.

Since PBBs are no longer produced or used in Europe and taking into account low and declining environmental concentrations, the CONTAM Panel concluded that PBBs are a low priority for further research or monitoring efforts.

Polybrominated biphenyls, PBBs, risk assessment, food, toxicity, exposure, occurrence, brominated flame retardants