Svetlana Popovic

Levels of Bisphenol A in Canned Soft Drink Products in Canadian Markets

The method developed previously for the determination of bisphenol A (BPA) in liquid infant formula was adapted and validated for determination of BPA in soft drink products. This method was based on solid phase extraction and derivatization with acetic anhydride followed by gas chromatography-mass spectrometry in selected-ion monitoring mode. The average method detection limit was 0.045 microg/L for a 10 mL sample. The average extraction recoveries were 101 and 99.9% obtained with seven different soft drink products spiked with BPA at 0.5 and 2.5 microg/L, respectively. Good repeatability of the method was observed with replicate analyses of seven different soft drinks; relative standard deviations ranged from 1.3 to 6.6%. This method was used to analyze samples of 72 canned soft drink products for BPA. Except for three products from which BPA-d16 could not be recovered at all due to interference of product compositions (e.g., quinine hydrochloride in tonic water), BPA was detected in samples of all the other products at levels ranging from 0.032 to 4.5 microg/L. About 75% of the products had BPA levels of <0.5 microg/L, and 85% of the products had BPA levels of <1 microg/L. Exposure to BPA through consumption of canned soft drink products is low; dietary intake of BPA was estimated at 0.027 microg/kg of body weight/day on the basis of the consumption of one canned soft drink with the highest BPA level (4.5 microg/L) for an adult with a 60 kg body weight, well below the provisional tolerable daily intake of 25 microg/kg of body weight/day established by Health Canada.

Concentrations of bisphenol A in the composite food samples from the 2008 Canadian total diet study in Quebec City and dietary intake estimates

A total of 154 food composite samples from the 2008 total diet study in Quebec City were analysed for bisphenol A (BPA), and BPA was detected in less than half (36%, or 55 samples) of the samples tested. High concentrations of BPA were found mostly in the composite samples containing canned foods, with the highest BPA level being observed in canned fish (106 ng g−1), followed by canned corn (83.7 ng g−1), canned soups (22.2–44.4 ng g−1), canned baked beans (23.5 ng g−1), canned peas (16.8 ng g−1), canned evaporated milk (15.3 ng g−1), and canned luncheon meats (10.5 ng g−1). BPA levels in baby food composite samples were low, with 2.75 ng g−1 in canned liquid infant formula, and 0.84–2.46 ng g−1 in jarred baby foods. BPA was also detected in some foods that are not canned or in jars, such as yeast (8.52 ng g−1), baking powder (0.64 ng g−1), some cheeses (0.68–2.24 ng g−1), breads and some cereals (0.40–1.73 ng g−1), and fast foods (1.1–10.9 ng g−1). Dietary intakes of BPA were low for all age–sex groups, with 0.17–0.33 µg kg−1 body weight day−1 for infants, 0.082–0.23 µg kg−1 body weight day−1 for children aged from 1 to 19 years, and 0.052–0.081 µg kg−1 body weight day−1 for adults, well below the established regulatory limits. BPA intakes from 19 of the 55 samples account for more than 95% of the total dietary intakes, and most of the 19 samples were either canned or in jars. Intakes of BPA from non-canned foods are low.

Bisphenol A in Canned Food Products from Canadian Markets

A method based on solid phase extraction followed by derivatization and gas chromatography-mass spectrometry analysis was validated for the determination of bisphenol A (BPA) in canned food products. This method was used to analyze 78 canned food products for BPA. Concentrations of BPA in canned food products differed considerably among food types, but all were below the specific migration limit of 0.6 mg/kg set by the European Commission Directive for BPA in food or food simulants. Canned tuna products had the highest BPA concentrations in general, with mean and maximum values of 137 and 534 ng/g, respectively. BPA concentrations in the condensed soup products were considerably higher than those in the ready-to-serve soup products, with mean and maximum values of 105 and 189 ng/g, respectively, for the condensed soups and 15 and 34 ng/g, respectively, for the ready-to-serve soups. BPA concentrations in canned vegetable products were relatively low; about 60% of the products had BPA concentrations of less than 10 ng/g. Canned tomato paste products had lower BPA concentrations than did canned pure tomato products. The mean and maximum BPA concentrations were 1.1 and 2.1 ng/g, respectively, for tomato paste products and 9.3 and 23 ng/g, respectively, for the pure tomato products.

Sources of low concentrations of bisphenol A in canned beverage products.

Although migration from can coatings is likely the source of bisphenol A (BPA) for the canned soft drink products with relatively high BPA concentrations, questions have been raised concerning the exact sources of BPA for those canned soft drink products with low BPA concentrations. Information is also needed for BPA concentrations in canned beer products to conduct proper exposure assessment for BPA under the Government of Canadas Chemicals Management Plan. In this work, 22 soft drink samples and 16 beer samples in both cans and plastic and/or glass bottles were analyzed for BPA. BPA was not detected in any of the soft drink samples in either plastic or glass bottles except for one product with a BPA concentration (0.018 microg/liter) close to the limit of quantification (0.015 microg/liter). BPA was detected in all of the corresponding soft drink products in cans, indicating that migration from can coatings is the likely source for BPA in canned products. Because considerable interference with ions m/z 213 and m/z 228 from sample matrices was observed for all beer samples, BPA concentrations in beer samples were measured using the ion m/z 270 instead. BPA was detected in only one of the seven beer products in glass bottles (0.054 microg/liter) but was detected in all corresponding beer samples in cans at low concentrations ranging from 0.081 to 0.54 microg/liter, indicating that migration from can coatings is likely the source of BPA in canned beer products.

Bisphenol A in Baby Food Products in Glass Jars with Metal Lids from Canadian Markets

A method based on solid phase extraction and derivatization with acetic anhydride followed by gas chromatography-mass spectrometry was validated for the determination of bisphenol A (BPA) in baby foods. The average method detection limit (MDL) was 0.18 ng/g for a 5 g sample. Method repeatability was demonstrated with the replicate analyses of various different types of baby foods; relative standard deviations (RSD) ranged from 1.2 to 16.1% with an average of 8.7%. Extraction recoveries ranged from 93.5 to 102.5% for different types of baby foods spiked at levels of 1-8 ng/g. This method was used to analyze 122 baby food products of 7 brands in glass jars with metal lids for BPA. The presence of BPA could not be confirmed and quantified for 23 of the 122 products due to interference from sample matrices. For the other 99 products, 15% had BPA levels of less than the average MDL, about 70% had BPA levels of less than 1 ng/g, and the average BPA levels in all 99 products was 1.1 ng/g. The average BPA level in the baby food products from brand E (3.9 ng/g) is higher than the average BPA levels in the products from the other brands (0.54-1.1 ng/g). The highest level of BPA, 7.2 ng/g, was found in two products from brand E as well. The average BPA level in the fruit products from all brands (0.60 ng/g) is lower than those in the mixed-dish products (1.1 ng/g) and the vegetable products (1.2 ng/g).

Determination of free and total bisphenol A in human milk samples from Canadian women using a sensitive and selective GC-MS method

A sensitive and selective GC-MS method was developed and used to analyse human milk samples for both free and total bisphenol A (BPA). Total BPA was detected in 72 of the 278 human milk samples (25.9%) at concentrations from < 0.036 to 2.5 ng g–1 with a geometric mean (GM) of 0.13 ng g–1 and median of 0.11 ng g–1, while free BPA was detected in fewer samples, 46 of the 278 samples (16.5%) at concentrations ranging from < 0.036 to 2.3 ng g–1 with a GM of 0.11 ng g–1 and median of 0.10 ng g–1. Ratios of [free BPA]/[total BPA] for the positive samples ranged from 7.9% to 100% with a GM of 57.2% and median of 70.3%. Concentrations of free and total BPA in most samples were low with 0.39 and 0.65 ng g–1 at the 95th percentile for free and total BPA, respectively, and they are also lower than those reported in other countries. Based on the low frequency of detection of free BPA in human milk samples, in general, dietary exposure to BPA for Canadian breast-fed infants is expected to be somewhat lower compared with exposure among formula-fed infants.

Migration of bisphenol A from can coatings to liquid infant formula during storage at room temperature.

Information on migration of bisphenol A (BPA) from can coatings to foods during storage at room temperature is very limited, and the conclusions from the available studies are not always consistent. To investigate the effect of storage time on BPA migration from can coatings to liquid infant formula at room temperature, samples of 21 canned liquid infant formula products from different cans but the same lot as those analyzed for BPA previously were analyzed for BPA again after storage at room temperature for 10 months. Additional migration of BPA from can coatings to liquid formula during the 10-month storage period at room temperature was observed for 9 of the 21 products, with increases in BPA levels ranging from 29.8 to 110%. Significant differences between the 2007 and 2008 results (P = 0.026) were observed for only one brand of product, which had the lowest BPA levels in the 2007 survey. The BPA levels in the milk-based formula products analyzed in 2008 (mean, 6.8 ng/g) were significantly higher (P = 0.00023) than those in the milk-based formula products analyzed in 2007 (mean, 5.0 ng/g), whereas the differences in BPA levels between the soya-based formula products analyzed in 2008 (mean, 5.3 ng/g) and those analyzed in 2007 (mean, 5.8 ng/g) were not significant (P = 0.097). No obvious correlation between the product expiration date and the level of BPA migration from can coatings was observed.

Background bisphenol A in experimental materials and its implication to low-dose in vitro study.

In vitro low-dose studies are important to understand the mechanisms of bisphenol A (BPA) action. BPA doses used in current in vitro studies varied considerably, and doses as low as 10(-15)M have been reported. The actual doses of BPA used in the in vitro low-dose studies were rarely checked analytically, and the background BPA levels in experimental materials, which will determine the lowest BPA dose to be used, should be investigated or considered. In this study, the background BPA levels in various materials typically used in in vitro low-dose studies for BPA were investigated. Background BPA levels from the use of disposable pipettes and pipette tips were low (<0.20 ng mL(-1) or 0.88 nM). BPA was also detected in several commercial buffer solutions at levels close to the method limit of quantification (LOQ) (0.02 ng mL(-1); 0.088 nM). However, BPA was detected in all cell culture media obtained from various sources at levels ranging from 0.080 to 4.26 ng mL(-1) (or 0.35 to 19 nM) with an average of 0.83 ng mL(-1) (3.5 nM). We suggest that culture media used for low-dose BPA studies should be analysed for background BPA levels prior to use, and the medium with the lowest BPA levels should be used.

Bisphenol A and Three Other Bisphenol Analogues in Canned Fish Products from the Canadian Market 2014.

A sensitive and selective gas chromatography-mass spectrometry method was developed and validated for simultaneous analysis of bisphenol A (BPA) and three other bisphenols, bisphenol B (BPB), bisphenol E (BPE), and bisphenol F (BPF). This method was used to analyze samples of 52 canned fish products to follow up a previous study conducted 5 years ago to investigate any changes in BPA levels since then and levels of other bisphenols due to possible changes in can coating formulations. BPB and BPE were not detected in any of the 52 canned fish products, and BPF was detected in only four products at low levels from 1.8 to 5.7 ng/g, indicating that BPA is likely still the dominant bisphenol used in current can coating formulations. BPA was detected in all 52 canned fish products, but at much lower levels compared with a previous study; levels ranged from 0.96 to 265 ng/g (average, 28 ng/g). The few products with high BPA levels (>100 ng/g) are exclusively from a new brand that has become available on the market only recently. Further analysis of canned fish products is planned in the future to capture any changes in BPA levels in these products and to update the exposure assessment of BPA due to consumption of canned fish products.

Levels and temporal trend of bisphenol A in composite food samples from Canadian Total Diet Study 2008–2012

Food composite samples from the Canadian Total Diet Study which was conducted each year from 2008 to 2012 rotating between different cities were analysed for bisphenol A (BPA). The overall levels of BPA in the composite food samples from each of the five years from 2008 to 2012 were similar in general with averages (range) of 7.7 ng/g (0.20–106 ng/g), 7.8 ng/g (0.26–110 ng/g), 6.9 ng/g (0.20–84 ng/g), 7.7 ng/g (0.20–105 ng/g) and 9.0 ng/g (0.15–90 ng/g) for 2008, 2009, 2010, 2011 and 2012, respectively. Levels of BPA in most of the non-canned food composite samples were low and no particular trends were observed. In contrast, the trend of BPA levels in canned food composite samples over the five years (2008–2012) varies. BPA levels in most of the canned food composite samples from 2008 to 2012 were consistent in general (e.g. canned luncheon meat: 10–18 ng/g, canned baked beans: 18–25 ng/g). While BPA levels over the five years were found to decrease for some canned food composite samples (e.g., canned fish: 109 ng/g in 2009 vs. 51 ng/g in 2012), they were also found to increase for some other canned food composite samples (e.g. canned meat soups: 90–104 ng/g in 2011–2012 vs. 29 ng/g in 2008). Thus, recent changes in can coating for food packaging to BPA-free alternatives may have not been fully reflected in all canned food products over the period from 2008 to 2012. Continued monitoring is necessary to more fully assess the potential impact on dietary exposure by the use of BPA alternatives in food contact materials.