The safety of bisphenol A

Canned food

Bisphenol A can be found in many canned food products (Credit: andrivete – Flickr)

Let’s face it, we are all exposed to bisphenol A (BPA), if not daily, at least several times a week. It as an unwelcome contamination of our food, with most of it coming from migration of BPA from the epoxy lining covering the inside of food cans, except for babies. Babies get their BPA either indirectly through breast milk or directly from liquid formula served out of baby bottles made of polycarbonate plastic. We are also “foolish” enough to effectively absorb it from the gut to allow the BPA to circulate in our bodies. But here is the good bit. Within minutes we start to attack the circulating BPA to neutralise its potentially dangerous effects and then we fairly quickly excrete the modified BPA through the urine. However, a small amount of free BPA might linger for a bit longer. The big question is what damage will that small amount cause, if any? And here there is big disagreement between scientists. So let’s look at how they see it by following their traditional way of assessing food safety risks. Bear with me, it is only four steps.

What is all the fuss about (hazard analysis)?

As mentioned upfront, it is very difficult to totally avoid consuming some BPA. It can actually be found in the urine of well over 90% of people examined worldwide. This is no surprise since BPA is a high volume industrial chemical used since the 1960s as a major ingredient in making clear polycarbonate plastic. This plastic is used to make a variety of common products including baby and water bottles, sports equipment, medical and dental devices, dental fillings and sealants, CDs and DVDs, household electronics, and eyeglass lenses. More importantly, BPA is also found in epoxy resins, which act as a protective lining on the inside of metal-based food and beverage cans and in the lid of glass bottles. There are many other applications of BPA outside of the food area. It is used as a colour developer in carbonless copy paper and thermal point-of-sale receipt paper. Close to 4 million tonnes of BPA are used by manufacturers yearly. In other words it is all around us and most of us carry some daily traces of BPA in our blood.

But does it matter? Well, yes it might. It is well-known that BPA acts as an endocrine disruptor, that is it is influencing hormone responses in our bodies. As a matter of fact, because of its oestrogen-like effects, it was briefly used as a growth promoter in cattle and poultry in the 1930s and even as a hormone-replacement therapy for women. However, the oestrogen effect was quite weak so it was quickly replaced by more powerful substances. The biggest worry is for the developing foetus. There are concerns that exposure in the womb to even small amounts of the chemical at this very sensitive stage of life could disturb normal development.

SO How much do we eat (exposure assessment)?

So now we have to find out if the levels of BPA we consume are high enough to cause damage. We do that by looking at contamination levels in different foods and checking how much of the respective food we eat. Sounds simple enough, but it isn’t and here is why. Since food is generally contaminated by BPA through migration from food packaging and food consumption information rarely indicates the exact source of the food, it makes accurate calculations of BPA exposure difficult.

Of course most foods would not be contaminated by BPA, but a quick review of “suspect” foods for infants up to adults indicates levels from 1 to 70 μg/kg. BPA levels in breast milk for the very young can vary from an average of about 2 μg/kg up to a maximum of just below 8 μg/kg. Liquid formula can contain slightly higher levels from an average of about 4 μg/kg to a maximum of around 11 μg/kg of BPA calculated as consumed. BPA in toddler food in glass jars can vary from an average of about 1 μg/kg to a high of around 7 μg/kg. For adult foods, average concentrations can range from 10 to 70 μg/kg in solid canned food and from 1 to 23 μg/L in liquid canned food.

As a next step we have to make qualified guesses about the amount of the respective food consumed to be able to estimate BPA exposure. The following theoretical calculations assume consumption of BPA-containing food on that day.

High exposure in children (Credit: paparutzi – Flickr

It is easiest for exclusively breastfed babies (0–6 months) where average daily exposure to BPA can be estimated to be 0.3 μg/kg bodyweight, up to a high of 1.3 μg/kg bodyweight. Bottle fed babies in the same age range go higher if the liquid formula is fed using polycarbonate bottles. Average exposure for such babies can be estimated to be 2.4 μg/kg bodyweight per day, up to a high of 4.5 μg/kg bodyweight per day. This is the reason why polycarbonate baby bottles are phased out in several countries. Once solid foods are introduced, exposure to BPA decreases. For children older than 3 years, average daily exposure would probably not exceed 0.7 μg/kg bodyweight with a possible high of up to 1.9 μg/kg bodyweight.

ŠFor adults, consumption of canned food might give an average daily exposure not exceeding 1.4 μg/kg bodyweight, with a possible high exposure of 4.2 μg/kg bodyweight when you really splash out. Handling of point-of-sale receipts and BPA circulating in the air might contribute to exposure, but the common view is that food is by far the major contributor of overall exposure to BPA for most population groups.

The theoretical calculations can be compared with biomonitoring data from analysing the total BPA excreted through urine on any day irrespective of the consumption of BPA-contaminated food or not. As was to be expected, children data compared the best with the average for infants with fairly consistent day-to-day food consumption patterns at 0.07–1.61 μg/kg bodyweight per day and children 3-5 years of age at 0.12-0.78 μg/kg bodyweight per day. Adult exposure estimates on the other hand were in the range of 0.05-0.27 μg/kg bodyweight per day, reflecting the irregular consumption of BPA-contaminated food.

But what about the unborn baby? Here we have to rely on reported normal blood levels of free BPA in pregnant women of 0.3-4.0 ng/ml as typical serum levels. Since BPA passes freely over the placenta, the same levels could be expected in the unborn baby.

The potential public health issue (HAZARD CHARACTERISATION)?

Public health concerns have been raised by findings that BPA exposure can influence brain development and modify normal behaviour. Exposure to BPA in the immediate period before and after birth has been shown to modify sex differentiation in the brain of rodents. In mice it was associated with increased anxiety, aggression, cognitive impairments, and decreased novelty seeking. In the offspring of BPA-exposed monkeys, males displayed less social activity.

Animal experiments point to the damage that can be caused by BPA (Credit: Brian_Kellett – Flickr)

But that’s in animals, right? And we all know that it is the dose that makes the poison. In most cases the animals were given quite high doses of BPA. So do we even come near the levels of BPA needed to cause damage? Many scientists don’t believe that is the case, but recent finding of suspected low-dose effects might change this around. And associations in human epidemiological studies between BPA exposure before birth and later behavioural changes would need a closer review.

In recent reports covering humans, BPA exposure has been associated with hyperactivity and aggression in 2-year-old female children when exposed before birth and with anxious, depressive, and hyperactive behaviours related to impaired behavioural regulation at 3 years of age. In contrast, childhood BPA exposure was not associated with changes in behaviour or executive function at 3 years of age.

Together, these reports and many others demonstrate that BPA exposure before birth affects several types of behaviour in a number of species and at blood levels that might be relevant for humans. What makes it even worse is that it can have immediate and long-lasting, transgenerational effects in brain development and social behaviour. That is the changes can move along to following generations by modifying gene response in something called epigenetic changes.

So what about the controversy surrounding the high BPA doses given to experimental animals compared to relatively low amounts consumed by humans? In most experiments animals have been given amounts of BPA higher than 50 μg/kg bodyweight and often at mg level while human daily consumption would be 5 μg/kg bodyweight at the most. Curiously enough the resulting blood levels in animals and humans have been very similar. Animals given 50 μg/kg bodyweight showed serum levels of free BPA of 2.0-4.6 ng/ml, well within the range of normally exposed humans. There is now speculation that other sources could contribute to human exposure or that metabolism after repeated exposure as is typical in humans could be different.

To be clear there is still remaining uncertainty about the exact doses of BPA needed for the described effects to occur. Additional large prospective cohort studies are needed to confirm and validate findings from animal studies. However, even in the absence of epidemiological studies, concern over adverse effects of BPA is warranted given the unique vulnerability of the developing foetus and child.

What does it all mean for us (risk characterisation)?

Based on previous scientific reports a Tolerable Daily Intake (TDI) of 50 μg/kg body weight per day has been set for the substance – that is the amount that can be ingested daily over a lifetime without appreciable risk. This is well above the calculated intake levels of BPA for most people.

However, although initially considered to be a weak environmental oestrogen, more recent studies have demonstrated that BPA may be similar in potency to the natural oestradiol hormone in stimulating some cellular responses. Emerging evidence also suggests that BPA may influence multiple endocrine-related pathways. Some scientists suggest that studies in rodents have identified adverse effects of BPA at exposure levels at or below the current acceptable daily intake level for this compound. However, when scrutinising the administered doses given to animals they seem to be much higher then the estimated intake in the exposure chapter above. Still they show that the circulating blood levels of free BPA in the age group with highest dietary exposures in humans (i.e. infants 0–6 months of age) is the same as measured in the animals. This is puzzling.

Pregnant women

The child during pregnancy is the most sensitive to BPA (Credit: molly_darling – Flickr)

Much more investigation is needed to understand the potential adverse health effects of BPA exposure in humans and to understand the multiple pathways through which it may act. Although many questions remain to be answered, it is becoming increasingly apparent that exposure to BPA is ubiquitous and that the effects of this endocrine disruptor are complex and wide-ranging.

At this stage it seems prudent to focus on the effects on the most vulnerable population group, the developing foetus and child. The most recent research seems to point to real life effects at this vulnerable life stage at current exposure levels. The only way to reduce this risk would be to limit BPA exposure in pregnant and lactating females. However, since this would be impractical, a general phasing out of the use of BPA in products in contact with food seems to be the only safe solution. New research underway and the current review of the scientific literature by food safety authorities will hopefully be able to resolve the issue once and for all.

Official views on bisphenol A

SummaryBisphenol A has been used in the production of food contact materials since the 1960s. Some recent findings point to possible low dose effects at levels of current human exposure. As a precaution some official authorities have introduced limited bans on bisphenol A to protect babies and young children, while others take a wait and see approach pointing to the need for more research.

bisphenol A

Structure of bisphenol A (credit: Wikipedia)

The use of the industrial chemical bisphenol A (BPA) to make food contact plastics and resins, although questioned for some time as a potential risk to human health by some scientists, is still approved by official authorities in most countries. Polycarbonate baby bottles is a recent exception. Question is, with some new scientific findings, will official risk assessment authorities change their general view and will risk managers take further action?

The European Union situation

BPA is permitted for use in food contact materials in the European Union (EU), most recently as listed under Regulation 10/2011/EU. The European Food Safety Authority (EFSA) completed a full risk assessment of BPA in 2006 and set a Tolerable Daily Intake (TDI) of 0.05 mg/kg body weight per day for the substance – that is the amount that can be ingested daily over a lifetime without appreciable risk. Since the initial opinion, EFSA has been asked three times to review the initial assessment and as a result published updated scientific advice in 2008, 2010 and 2011. EFSA also evaluated intakes of BPA through food and drink, for adults, infants and children and found that they were all well below the TDI.

Due to the controversial debate on BPA and despite EFSA’s view on the safety of BPA, the EU member states Denmark and France banned BPA in baby bottles in 2010. The ban was imposed for reasons of precautionary consumer protection. In order to create a consistent legal framework within the EU and following a similar move in Canada, the European Commission adopted Directive 2011/8/EU in January 2011 prohibiting the use of BPA for the manufacture of polycarbonate infant feeding bottles and their placing on the market within the EU.

Contrary to the prevailing view of EFSA, on 27 September 2011 the French L’Agence nationale de sécurité sanitaire de l’alimentation, de l’environnement et du travail (ANSES) published two reports on bisphenol A reaching a different conclusion. One report related to the health effects associated with exposure to BPA while the other concerned its uses. In the report on health effects – “Effets sanitaires du bisphénol A” – ANSES concluded that health effects have been proven in animals and suspected in humans, even at low levels of exposure below current regulatory thresholds. On the basis of these findings it recommended that exposure to bisphenol A should be limited in the most susceptible populations of infants, young children, pregnant and breastfeeding women. A previous Danish report produced by the DTU Fødevareinstituttet on behalf of the Danish government reached similar conclusions. It noted that neurodevelopment effects were seen in young male rats even at low doses of BPA raising some concern of possible human effects.

The CEF Panel did not consider that the information in the ANSES report would immediately change the views the Panel expressed in 2010, but wanted more time to review in depth the new studies. Consequently, in February 2012, the CEF Panel decided to undertake a full re-evaluation of the human risks associated with exposure to BPA through both dietary and non-dietary sources. The intention is to review all the available data and scientific studies on exposure published since 2006. The Panel will further evaluate uncertainties about the possible relevance to human health of some BPA-related effects observed in rodents at low dose levels. New findings from ongoing studies on low dose effects as well as on dietary and non-dietary exposure to BPA will be considered as they become available during 2012, with the intention to publish the new opinion by May 2013.

The Canadian situation

Health Canada’s Food Directorate has conducted periodic reviews of BPA as new information has become available relating to its toxicity and potential exposure from food packaging applications. The purpose of these reviews was to determine whether dietary exposure to BPA could pose a health risk to consumers. Based on the overall weight of evidence, the view is still that current dietary exposure to BPA through food packaging uses is not expected to pose a health risk to the general population, including newborns and infants.

However, due to the uncertainty raised in some animal studies relating to the potential effects of low levels of BPA, the Government of Canada banned the use of BPA in baby bottles in March 2010 to enhance the protection of infants and young children. It further added BPA to the list of substances deemed potentially harmful to health or the environment in October the same year to make it easier to take further action should this be deemed necessary.

The Bureau of Chemical Safety, in the Food Directorate of Health Canada, is investing further resources in exploring the potential impact of BPA on the health of the Canadian population. It will improve estimates of dietary exposure to BPA from all Canadian food sources. It is also a partner in assessing information gaps on the occurrence and significance of BPA in human milk from Canadian mothers in a project funded by Health Canada, the Canadian Institutes of Health Research and the Ontario Ministry of the Environment.

The situation in the USA

BPA was first approved by the Food and Drug Administration in the early 1960s. Due to some concerns raised, the FDA released a draft report in August 2008 finding that BPA remained safe in food contact materials. However, a few months later a subcommittee of FDA’s science board raised questions about whether the review had adequately considered the most recent scientific information available.

On January 15, 2010 and again on March 30, 2012, the FDA issued interim updates on BPA. They claimed that studies employing standardised toxicity tests had thus far supported the safety of current low levels of human exposure to BPA. However, on the basis of results from recent studies using novel approaches to test for subtle effects, FDA, supported by the view of the National Toxicology Program at the National Institutes of Health, expressed some concern about the potential effects of BPA on the brain, behaviour, and prostate gland in foetuses, infants, and young children. This concern was expressed because it was noted that developmental changes in some animal studies occurred at BPA exposure levels similar to those experienced by humans. A collaborative effort is underway to carry out further in-depth studies to clarify the outstanding uncertainties about the risks of BPA.

While awaiting the new results, FDA expressed support for actions to stop the production of BPA-containing baby bottles and infant feeding cups for the U.S. market and the recommendations issued by the Department of Health and Human Services for infant feeding and food preparation to reduce exposure to BPA.

The Australian situation

The Australian Government has limited its action on BPA to an introduction of a voluntary phase out of BPA use in polycarbonate baby bottles. Food Standards Australia New Zealand (FSANZ) is firm in its view that there are no health issues associated with BPA at the levels people are exposed to. They state that extremely large amounts of food and beverages would have to be consumed to even reach still acceptable intake level.

There is no suggestions of further activities as they claim that countries that banned the use of BPA in some products did so without any support of risk assessment conclusions on the safety of BPA. To further support this view the FSANZ website lists shortcomings in 17 scientific studies purporting to show the toxicity of BPA.

The international view

According to the World Health Organization website the scientific controversy around the safety of BPA is driven by the disparate results reported in the many studies exploring toxicity and endocrine activity of BPA in animals. This has led to different risk management decisions taken by national authorities. The issue has also received much attention in the media and among the general public. In light of possible adverse human health effects at low doses of BPA, especially on reproduction, the nervous system and on behavioural development, and considering the relatively higher exposure of very young children compared with adults, a meeting of ad hoc experts was jointly organised by FAO and WHO in late 2010 to assess the safety of BPA.

It was difficult for the meeting to reach definitive conclusions on the safety of BPA because of a lack of data from experimental animal studies suitable for risk assessment. Controversy persisted over the biological significance of many of the more sensitive end-points and it was uncertain if all relevant end-points had been covered by the studies used for past risk assessments.

It was clear that for many traditional end-points, human exposure was much too low to be of any health concern. However, for some emerging new end-points (sex-specific neurodevelopment, anxiety, pre-neoplastic changes in mammary glands and prostate in rats, and impaired sperm parameters) a few studies showed associations at levels close to the estimated human exposure, which was clearly of concern if their toxicological significance were to be confirmed. Further research would be needed before these results could provide a basis for a realistic estimate of the human health risk.

The maligned bisphenol A

Summary – Bisphenol A is an industrial chemical that has been present in many hard plastic bottles and metal-based food and beverage cans since the 1960s. Recent studies using novel approaches to test for subtle effects raised some concern about its potential influence on the brain, behaviour and prostate gland development in foetuses, infants and young children. It is currently under extensive review.

Bisphenol A (BPA) is an industrial chemical used since the 1960’s as an ingredient in making a clear plastic known as polycarbonate. Polycarbonate is used to make a variety of common products including baby and water bottles, sports equipment, medical and dental devices, dental fillings and sealants, CDs and DVDs, household electronics, and eyeglass lenses. BPA is also found in epoxy resins, which act as a protective lining on the inside of metal-based food and beverage cans. It can leach into food, and a study of over 2,000 people found that more than 90 percent of them had BPA in their urine. Traces have also been found in breast milk, the blood of pregnant women and umbilical cord blood.

BPA is permitted for use in food contact materials in the European Union (EU) and in most other countries.  In the past, the European’s Scientific Committee on Food, the European Chemicals Bureau, the European Food Safety Authority, and the US Food and Drug Administration all concluded that current levels of BPA present no risk to the general population. However, recently it was found to exert detectable hormone-like properties, raising concerns about its presence in consumer products and foods contained in such products. Starting in 2008, several governments questioned its safety, prompting some retailers to voluntarily withdraw polycarbonate products. Canada was the first country to ban the use of BPA in baby bottles from 2010 followed by the European Commission in January 2011. In 2012, also the United States banned the use of BPA in baby bottles including in infant feeding cups.

Bisphenol A toxicity

There is a great diversity of opinion about the health effects of BPA. Overall, empirical evidence in support of negative health effects of BPA varies significantly across studies.  Standardised toxicity tests used globally for regulatory decision-making long supported the safety of current low levels of human exposure to BPA. However, results of recent studies using novel approaches and different endpoints found detrimental effects in laboratory animals at very low doses similar to estimated human exposure.

Several new studies evaluated developmental or behavioural effects that are not typically assessed in standardised tests. For example, perinatal exposure to BPA in rodents modified sex differences in the brain. In mice, prenatal exposure to BPA was associated with increased anxiety, aggression and cognitive impairment. In the offspring of BPA-exposed monkeys, males displayed less social behaviours and were more exploratory. In humans, BPA exposure during gestation has been associated with hyperactivity and aggression in young children and with anxiety and depression in older children. Together, these reports and many others demonstrate that BPA exposure during gestation affects several types of behaviours in a number of species.

It is clear that bisphenol A is a weak endocrine disruptor, which can mimic oestrogen and may lead to negative health effects. Early developmental stages seem to be the period of greatest sensitivity to its effects. Regulatory bodies have determined safety levels for humans, but those safety levels are being questioned or are under review as a result of the new findings. Experts in the field of endocrine disruptors have stated that the entire population may suffer adverse health effects from current BPA levels. In 2009, the Endocrine Society released a statement citing the adverse effects of endocrine-disrupting chemicals, and the controversy surrounding BPA.

In 2010, the WHO organised an expert meeting to review toxicological and health aspects of BPA supported by Health Canada, the European Food Safety Authority, the U.S. National Institute of Environmental Health Sciences and the US Food and Drug Administration. Although the meeting concluded that doses much higher than estimated human exposure were necessary for most toxicological effects, it did agree that for some emerging effects, like sex-specific neurodevelopment, anxiety, preneoplastic changes in mammary glands and prostate in rats, and impaired sperm parameters, a few studies showed associations at levels close to estimated human exposure. Because of the considerable uncertainty of the validity and relevance of these observations it was recommended that further research should be undertaken to reduce the uncertainty.

Re-evaluation of human risks

EFSA completed a full risk assessment of BPA in 2006 and set a limit of 0.05 mg/kg body weight per day that could be ingested daily over a lifetime without appreciable risk (Tolerable Daily Intake – TDI). It evaluated intakes of BPA through food and drink, for adults, infants and children and found that they were all well below the TDI.

EFSA has updated its scientific advice on BPA several times since 2006, most recently updating its risk assessment in 2011, reaching similar conclusions.

In February 2012, following further consideration of new scientific studies, EFSA decided to undertake a full re-evaluation of the human risks associated with exposure to BPA through the diet, also taking into consideration the contribution of non-dietary sources to the overall exposure to BPA. The new opinion will review all the available data and scientific studies on dietary exposure published since EFSA’s 2006 Opinion. The Panel will further evaluate uncertainties about the possible relevance to human health of some BPA-related effects observed in rodents at low dose levels.

FDA is also continuing to consider the low dose toxicity studies of BPA as well as other recent peer-reviewed studies related to BPA. At this stage, FDA has stated its current perspective on BPA, its support for further studies, its establishment of a public docket for its assessment of BPA use in food contact applications. FDA has issued interim public health recommendations, including its view of the appropriate regulatory framework for BPA use in food contact applications. It is expressing the wish to pursue the issue in collaboration with international partners.