Novel halogenated flame retardants in Canadian human milk from the MIREC study (2008-2011).

Novel halogenated flame retardants (NHFRs) have been developed to replace those brominated flame retardants that have been restricted due to their persistence, bioaccumulation potential and toxicity, therefore, it is important to determine whether these replacement products are present at detectable concentrations in Canadians. NHFRs were measured in human milk samples (n = 541) collected from across Canada between 2008 and 2011, which is the first pan-Canadian dataset for these chemicals in human milk. Among the 15 measured NHFRs and eight methoxy-polybrominated diphenyl ethers (MeO-PBDEs), nine NHFRs and two MeO-PBDEs (6-MeO-PBDE 47 and 2-MeO-PBDE 68) were detected at a frequency of more than 9%. Despite benzene, 1,1'-(1,2-ethanediyl)bis [2,3,4,5,6-pentabromo-]/decabromodiphenylethane [DBDPE] being detected less frequently than the other observed NHFRs, its relative contribution to the sum of nine NHFRs was important when it was present. The maximum ΣNHFR concentration in Canadian human milk was 6930 pg g-1 lipid while the maximum ΣMeO-PBDEs was 1600 pg g-1 lipid. While most NHFR concentrations were significantly correlated with each other, no relationships between maternal age, parity or pre-pregnancy BMI were identified with ΣNHFR concentrations in the milk. In contrast, maternal age was significantly correlated with ΣMeO-PBDE concentrations (r = 0.237, p < 0.001). ΣNHFR concentrations were similarly not related to maternal education, although ΣMeO-PBDE concentrations were found to be higher in milk from women who had graduated from trade schools relative to the other education levels considered. NHFR detection frequency and concentrations observed in the Canadian human milk seem to align well with Europe.

Differences in mirex [dechlorane] and dechlorane plus [syn- and anti-] concentrations observed in Canadian human milk.

As part of the pan-Canadian Maternal-Infant Research on Environmental Chemicals (MIREC) study, human milk samples were collected between 2008 and 2011, and analyzed for mirex, an organochlorine insecticide and flame retardant, in addition to dechlorane plus (syn- and anti-DDC-CO), the flame retardant replacement for mirex. Mirex was analyzed separately, using a method for the analysis of existing organochlorine insecticides, while the presence of DDC-CO isomers was determined using a method developed for the detection of emerging flame retardants. Mirex was detected in all samples analyzed (n = 298), while syn- and anti-DDC-CO were present in 61.0% and 79.5% of the samples, respectively (n = 541). Mirex concentrations have declined in human milk since the 1990s. Since this is the first pan-Canadian dataset reporting DDC-CO concentrations in human milk, no temporal comparisons can be made. Maternal age was correlated with concentrations of both compounds although parity did not impact concentrations of either analyte. Given the presence of this relatively recently identified flame retardant (DDC-CO) in human milk from women across Canada, studies to identify dominant sources of this compound are critical. Despite low concentrations of environmental chemicals in human milk from Canadian women, Health Canada supports breastfeeding of infants because of the important health benefits to both the mothers and their infants.

Dioxins/furans and PCBs in Canadian human milk: 2008-2011.

Human milk was collected between 2008 and 2011 as part of the Maternal - Infant Research on Environmental Chemicals (MIREC) study that was initiated to establish Canadian national estimates of maternal and infant exposure to a broad suite of environmental contaminants (e.g., persistent organic pollutants [POPs], trace elements, phthalates, etc.). Among the 1017 human milk samples collected, 298 were analysed for polychlorinated dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs) and polychlorinated biphenyls (PCBs). World Health Organization (WHO) toxic equivalency concentrations (WHO TEQ2005) for PCDD/F+dioxin-like (DL) PCB ranged from 2.2pg TEQ2005 g-1 lipid to 27pg TEQ2005 g-1 lipid. The relative contribution of PCDDs to the overall WHO TEQ2005 (PCDD/F+DL PCB) has decreased from earlier investigations into POP levels in Canadian human milk. Significantly higher PCB concentrations were observed in milk from women born in Europe relative to those born in Canada (p<0.001), in contrast to results for the PCDD/Fs (p=0.496). Age was found to significantly impact milk ∑PCB concentrations (p=0.018), with elevated concentrations observed in milk from women >30years relative to those <30years of age. While this trend was also observed for the PCDD/Fs, this relationship was impacted by parity. WHO TEQ2005 concentrations were significantly higher in milk from primiparous women (p=0.019) and those >30years relative to those <30years of age (p<0.001). No significant differences were associated with education level or pre-pregnancy body mass index. PCB and PCDD/F concentrations have continued to decline in Canadian human milk since the last sampling of human milk was performed.

A method for the analysis of multiple novel halogenated flame retardants in cow's milk.

A method was developed for the extraction and analysis of cow's milk to measure 21 halogenated flame retardants (FRs), including individual isomers plus eight methoxy-polybrominated diphenyl ethers (MeO-PBDEs). Extraction was performed using homogenisation with acetone:hexane with size exclusion chromatography followed by adsorption chromatography clean-up. Analysis was undertaken using gas chromatography high-resolution mass spectrometry. The method was validated in fortified cow's milk with FRs and 2-methoxy-PBDE 68 at three levels (low [12.5-1250 pg g(-)(1)], mid [37.5-5000 pg g(-)(1)] and high [400-10 000 pg g(-)(1)]). Additional methoxy-PBDEs were tested at two fortification levels. Isotope dilution was used to correct for losses during sample preparation and average recoveries ranged from 58% (allyl 2,4,6-tribromophenyl ether [ATE]) to 121% (γ-tetrabromoethylcyclohexane (γ-TBECH)). Limits of detection ranged from 0.055 pg g(-)(1) (6-methoxy-PBDE 47) to 38.9 pg g(-)(1) (decabromodiphenylethane (DBDPE)). Matrix effects were overcome through the use of surrogate and performance standards. A single FR (1-bromomethyl-2,3,4,5,6-pentabromobenzene [PBBB]) and two methoxy-PBDEs were detected in commercially available cow's milk collected from local supermarkets in Ottawa, ON, Canada. Detection frequency was < 25% for these compounds and, where present, concentrations were low.

Brominated flame retardant concentrations in sera from the Canadian Health Measures Survey (CHMS) from 2007 to 2009.

Pooling of surplus serum from individual samples, collected between 2007 and 2009 during Cycle 1 of the Canadian Health Measures Survey (CHMS), was performed to develop a national baseline estimate of brominated flame retardants in Canadians. Serum samples were categorized by sex and distributed by five age groups ranging from 6 to 79years. Nearly 5000 (4583) serum samples were used to form 59 composite pools. Serum pools were created to ensure a high detection frequency of these analytes in serum because low volume samples had previously resulted in non-detectable concentrations. The analytes of interest in these serum pools included 23 polybrominated diphenyl ethers (PBDEs) and three hexabromocyclododecane (HBCD) isomers (α-, ß- and γ-HBCD). PBDEs were observed in all samples tested and total PBDE concentrations ranged from 27ngg(-1) lipid to 130ngg(-1) lipid (geometric mean [GM] 46ngg(-1) lipid). ∑PBDE concentrations were significantly elevated in samples representing the 6-11year old age group (GM 65ngg(-1) lipid) relative to ages above 40years, although no difference in concentration was observed between the sexes. PBDE concentrations in Canadian sera from the general population were higher than reported in Europe and Asia, but a little lower than observed in the US. PBDE 47 was the greatest contributor to ∑PBDE concentrations and the GM concentration for this congener was 22ngg(-1) lipid. The other dominant contributors to ∑PBDE concentrations were in descending order: 153 [GM 9.4ngg(-1) lipid]>99 [GM 4.6ngg(-1) lipid]≅100 [GM 4.1ngg(-1) lipid]>209 [GM 1.1ngg(-1) lipid] and 183 [GM 0.42ngg(-1) lipid]. ∑HBCD was detected in all samples analysed, although most samples were observed at concentrations <1ngg(-1) lipid, similar to global concentrations. α-HBCD was the dominant contributor to ∑HBCD concentrations in Canadians although ß- and γ-HBCD were detected in 23% and 35% of the samples, respectively. No differences in ∑HBCD concentration were associated with age or sex. This dataset represents the first national data describing HBCD isomers and some PBDEs (e.g., 183, 209) in Canadians.

The impact of production type and region on polychlorinated biphenyl (PCB), polychlorinated dibenzo-p-dioxin and dibenzofuran (PCDD/F) concentrations in Canadian chicken egg yolks.

Chicken eggs from five different production types (conventional, omega-3 enriched, free range, organic and free run) were collected, when available, from three regions (west, central and east) of Canada to determine persistent organic pollutant (POP) concentrations. Total polychlorinated biphenyl (PCB) concentrations (∑37 congeners) in yolks from the eggs ranged from 0.162 ng g(-1) lipid to 24.8 ng g(-1) lipid (median 1.25 ng g(-1) lipid) while the concentration of the sum of the 6 indicator PCBs ranged from 0.100 ng g(-1) lipid to 9.33 ng g(-1) lipid (median 0.495 ng g(-1) lipid). Total polychlorinated dibenzo-p-dioxin/dibenzofuran (PCDD/F) concentrations ranged from 2.37 pg g(-1) lipid to 382 pg g(-1) lipid (median 9.53 pg g(-1) lipid). The 2005 WHO toxic equivalency (TEQ) ranged from 0.089 pg TEQ(PCDD/F+dioxin-like[DL]-PCB) g(-1) lipid to 12.8 pg TEQ(PCDD/F+DL-PCB) g(-1) lipid (median 0.342 pg TEQ(PCDD/F+DL-PCB) g(-1) lipid). PCB and PCDD/F concentrations were significantly different (p<0.001) in egg yolks from different regions of collection. In contrast to observations in Europe, PCB and PCDD/F concentrations in Canadian egg yolks were not impacted solely by the production type (e.g., conventional, free range, organic, etc.) used to maintain the laying chickens. Additionally, only one Canadian free range yolk from western Canada (12.8 pg TEQ(PCDD/F+DL-PCB) g(-1) lipid) exceeded the European toxic equivalent concentration limits for eggs (5 pg TEQ(PCDD/F+DL-PCB) g(-1) lipid). This differs from observations in Europe where free range/home produced eggs frequently have higher POP concentrations than eggs from other production types. Median PCB dietary intake estimates based on consumption of eggs were less than 10 ng d(-1) while median PCDD/F intakes were less than 45 pg d(-1).

PCDD/F and PCB concentrations in sera from the Canadian Health Measures Survey (CHMS) from 2007 to 2009.

In order to establish a national baseline estimate of the concentrations of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) and polychlorinated biphenyls (PCBs) in Canadians, pooling of individual human sera was performed to ensure that a high frequency of detectable concentrations of analytes would be achieved. Nearly 5000 (4583) sera samples from Cycle 1 of the Canadian Health Measures Survey (CHMS) collected between 2007 and 2009 were used to form 59 composite pools of approximately 25 mL each. Pools were categorized by sex and age with participants ranging from 6 to 79 years. The pooled samples were analysed for 17 PCDD/Fs and 36 PCB congeners, and from these data, total toxic equivalent concentrations (TEQ(2005 PCDD/F+Dioxin-like [DL]-PCB)) were estimated. The average 2,3,7,8 tetrachlorodibenzodioxin (TCDD) concentration was <1 pg g⁻¹ on a lipid extractable basis. The average total TEQ(2005 PCDD/F+DL-PCB) was 11 pg TEQ g⁻¹ lipid and average ΣPCB concentrations were about 100 ng g⁻¹ lipid. Sex did not affect the concentrations, while PCB and PCDD/F concentrations were positively correlated with age (p<0.001). It appeared in some cases that the age group 6-11 years had higher concentrations of these persistent organic pollutants (POPs) than the concentrations observed in 12-19 year olds, however, the results were not statistically significant based on pair-wise comparisons. Concentration levels and patterns observed in this study of Canadians were similar to those reported in the US and European populations.

Effects of postharvest preparation on organophosphate insecticide residues in apples.

Apples were sampled directly from orchard trees at 96, 45, and 21 days postapplication with one of three organophosphate insecticides (azinphos methyl, phosalone, or phosmet, respectively). Individual apples were prepared for analysis following one of three postharvest preparations: no preparation, rinsed with deionized water for 10-15 s, or rinsed and peeled. Azinphos methyl, phosalone, and phosmet concentrations ranged from below the level of detection to 5.26 ng/g, 94.7 to 5720 ng/g, and 0.011 to 663 ng/g in the apples that received no postharvest preparation, respectively. Although rinsed apples had lower maximum concentrations than observed in apples with no preparation, levels were not significantly lower. Concentrations of all three OP insecticides in apples that were rinsed followed by peeling, however, were much lower (below detection limits to 0.733 ng/g, azinphos methyl; 0.322-219 ng/g, phosalone; and below detection limits to 44.0 ng/g, phosmet) than observed in apples that had been rinsed alone. Rinsing and peeling of apples resulted in a 74.5-97.9% reduction in OP residues, while rinsing alone lowered mean concentrations by 13.5-28.7% relative to apples that received no postharvest preparation.

Captan residue reduction in apples as a result of rinsing and peeling.

Apples, treated with captan for disease control in a commercial orchard in Quebec, Canada, were collected and sorted into post-harvest preparation types (no preparation; rinse; rinse and peel). Captan residues were greatest (25.5-5100ng/g) in apples with no post-harvest preparation and lowest (0.146-136ng/g) in apples that had been rinsed and peeled prior to extraction and analysis. Residues were significantly lower (p=0.003) in apples that had been rinsed prior to extraction than in apples with no post-harvest preparation. Similarly, apples subjected to rinsing and peeling had significantly lower captan residues than had apples that had been rinsed alone (p<0.0001). Although captan residues in rinsed apples were approximately 50% lower than those in apples that received no post-harvest preparation, the reduction associated with peeling of apples was much greater (98%). Estimated mean captan intakes resulting from consumption of raw apples were established and single day intakes, based on apples with no preparation, ranged from 2.58µg/kg in females >70 years to 9.48µg/kg for individuals aged three years (at this age no distinction is made between males and females). Mean intakes estimated using rinsed and peeled apples were approximately two orders of magnitude lower than intakes estimated using apples with no post-harvest preparation, demonstrating the effect of post-harvest preparation on captan intakes. Mean captan intake estimates from all post-harvest preparation types were well below the World Health Organization acceptable daily intake of 100µg/kg/day, based on raw apple consumption.

Variability in captan residues in apples from a Canadian orchard.

Apple trees in an orchard in Quebec, Canada were treated, following label directions, with the fungicide captan (1,2,3,6-tetrahydro-N-(trichloromethylthio)phthalimide) during the 2003 agricultural season. A total of 142 apples from three rows of trees were selected for determination of captan by GC/MS. Individual apples were found to contain captan levels ranging from 16.9 to 6350 ng g-1. Only two individual apple samples exceeded the Canadian maximum residue limit (5000 ng g-1) for captan in apples. Six composite samples, comprising half portions of eight individual apples, were analysed from each of the three experimental rows. Composite samples ranged in concentration from 166 to 2620 ng g-1. The greatest uncertainty associated with the measured concentrations was due to variability among apples rather than the measurement of residue levels.

Organophosphate levels in apple composites and individual apples from a treated Canadian orchard.

Azinphos-methyl, phosalone, and phosmet were applied individually to separate rows of trees within a commercial apple orchard in Quebec, Canada, during the 2003 agricultural season. Apples were collected for residue analysis immediately prior to the harvesting of the remaining apples for market distribution and were prepared for analysis as both individual apples and as composites of eight individuals. Analysis of the three applied compounds, as well as five organophosphate insecticides that were not applied, was performed using gas chromatography-mass spectrometry. Azinphos-methyl, phosalone, and phosmet, which were applied, were detected in all samples analyzed at concentrations ranging from 0.004 ng/g to 2260 ng/g. Methidathion was not observed in any sample. Chlorpyrifos, diazinon, dimethoate, and malathion concentrations ranged from below method detection limits to 0.71 ng/g, and the detection frequency for these compounds ranged from 20% to 100%. Residues measured in this study were all below the Canadian maximum residue limit for apples. Variability factors ranged from 2 to 19 for all compounds observed in this study. Composite samples may not accurately reflect the extremes of exposure from consumption of single servings of apples.