Scholarly article on topic 'Maternal and infant exposure to environmental phenols as measured in multiple biological matrices'

Maternal and infant exposure to environmental phenols as measured in multiple biological matrices Academic research paper on "Health sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Science of The Total Environment
OECD Field of science
Keywords
{"Bisphenol A" / Triclosan / Urine / Infant / Pregnancy / Meconium}

Abstract of research paper on Health sciences, author of scientific article — Tye E. Arbuckle, Lorelle Weiss, Mandy Fisher, Russ Hauser, Pierre Dumas, et al.

Abstract Background Results of recent national surveys have shown the high prevalence of exposure to bisphenol A (BPA) and triclosan (TCS) among the general population; however biomonitoring data for pregnant women and infants are limited. Methods Women (n=80) were recruited from early prenatal clinics and asked to collect urine samples multiple times during pregnancy and once 2–3months post-partum. Samples of infant urine and meconium as well as breast milk and infant formula were also collected. Biospecimens were analyzed by GC–MS/MS for BPA, TCS and triclocarban (TCC). Results Triclosan was detected in over 80% of the maternal urines (geometric mean (GM): 21.61μg/L), 60% of the infant urines (GM: 2.8μg/L), 46% of the breast milk and 80% of the meconium samples. Triclocarban was rarely detected in any of the biospecimens. Median total BPA concentrations were 1.21 and 0.24μg/L in maternal and infant urines, respectively. Free BPA was detected in only 11% of infant urine samples. The meconium of female infants had significantly higher concentrations of total BPA and TCS than those of males, while no differences were observed in infant urine concentrations by sex. Conclusions We found widespread exposure among pregnant women and infants to environmental phenols, with large inter-individual variability in exposure to triclosan. These data will contribute to the risk assessment of these chemicals, especially in susceptible sub-populations.

Academic research paper on topic "Maternal and infant exposure to environmental phenols as measured in multiple biological matrices"

ELSEVIER

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Maternal and infant exposure to environmental phenols as measured in multiple biological matrices

Tye E. Arbuckle a'*, Lorelle Weiss b, Mandy Fisher a, Russ Hauserc,d, Pierre Dumas e, René Bérubé e, Angelica Neisa a, Alain LeBlanc e, Carly Lang a, Pierre Ayotte e,f, Mark Walker g, Mark Feeley h, Diane Koniecki ', George Tawagij

a Population Studies Division, Healthy Environments and Consumer Safety Branch, Health Canada, Ottawa ON, Canada b Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa ON, Canada c Department of Environmental Health, Harvard School of Public Health, Boston, MA, United States d Department of Epidemiology, Harvard School of Public Health, Boston, MA United States

e Centre de Toxicologie du Québec (CTQ), Institut National de Santé Publique du Québec (INSPQ), Québec, QC, Canada f Axe Santé des Populations et Pratiques Optimales en Santé, Centre de recherche du CHU Québec, Québec, QC, Canada g Department of Obstetrics and Gynecology, University of Ottawa, Ottawa ON, Canada h Bureau of Chemical Safety, Health Products and Food Branch, Health Canada, Ottawa, ON, Canada

i Consumer Product Safety Directorate, Healthy Environments and Consumer Safety Branch, Health Canada, Ottawa, ON, Canada j Department of Obstetrics and Perinatal Medicine, Ottawa Hospital, Ottawa, ON, Canada

HIGHLIGHTS

• BPA and triclosan were measured in maternal and infant biospecimens

• Triclosan was detected in over 80% of maternal urines and meconium

• BPA was detected in over 90% of maternal urines and 40% of infant urines

ARTICLE INFO

ABSTRACT

Article history:

Received 27 May 2014

Received in revised form 29 October 2014

Accepted 29 October 2014

Available online 4 December 2014

Editor: Adrian Covaci

Keywords:

Bisphenol A

Triclosan

Infant

Pregnancy

Meconium

Background: Results of recent national surveys have shown the high prevalence of exposure to bisphenol A (BPA) and triclosan (TCS) among the general population; however biomonitoring data for pregnant women and infants are limited.

Methods: Women (n = 80) were recruited from early prenatal clinics and asked to collect urine samples multiple times during pregnancy and once 2-3 months post-partum. Samples of infant urine and meconium as well as breast milk and infant formula were also collected. Biospecimens were analyzed by GC-MS/MS for BPA, TCS and triclocarban (TCC).

Results: Triclosan was detected in over 80% of the maternal urines (geometric mean (GM): 21.61 ^ig/L), 60% of the infant urines (GM: 2.8 ^ig/L), 46% of the breast milk and 80% of the meconium samples. Triclocarban was rarely detected in any of the biospecimens. Median total BPA concentrations were 1.21 and 0.24 ^ig/L in maternal and infant urines, respectively. Free BPA was detected in only 11% of infant urine samples. The meconium of female infants had significantly higher concentrations of total BPA and TCS than those of males, while no differences were observed in infant urine concentrations by sex.

Conclusions: We found widespread exposure among pregnant women and infants to environmental phenols, with large inter-individual variability in exposure to triclosan. These data will contribute to the risk assessment of these chemicals, especially in susceptible sub-populations.

Crown Copyright © 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/3.0/).

* Corresponding author at: Population Studies Division, Healthy Environments and Consumer Safety Branch, Health Canada, 50 Colombine Dr., AL 0801A, Ottawa, ON K1A 0K9, Canada. Tel.: +1 613 941 1287.

E-mail address: Tye.Arbuckle@hc-sc.gc.ca (T.E. Arbuckle).

1. Introduction

Environmental phenols such as bisphenol A (BPA), triclosan (TCS) and triclocarban (TCC) are non-persistent ubiquitous chemicals that are primarily excreted in urine. BPA and TCS are commonly detected in urine samples from national surveys in Canada and the United

http: //dx.doi.org/10.1016/j.scitotenv.2014.10.107

0048-9697/Crown Copyright © 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

States, whereas TCC exposure is less prevalent (Health Canada, 2013; CDC, 2013; Ye et al., 2011a).

Triclosan is used to preserve materials such as textiles, leather, paper, plastic and rubber and as an anti-bacterial and anti-fungal agent in a number of cosmetics and personal care consumer products including non-prescription drugs and natural health products (Health Canada and Environment Canada, 2012). The US Department of Health and Human Services lists a number of household products in their database which contain TCS or TCC (http://householdproducts.nlm. nih.gov/index.htm, accessed February 19, 2014). TCS can be found in antibacterial hand soaps, dishwashing liquid, toothpaste, shaving gel, antiperspirant, deodorant, face and body wash, hand lotion, lipcolor, and antibacterial dog shampoo. Self-reported use of personal care products in adolescents has been reported to be a significant determinant of urinary TCS concentrations (Den Hond et al., 2013).

TCC is also used in a number of consumer products as an antibacterial agent in the US; however, published data to date indicates that the number of products containing TCC is far fewer than those containing TCS. Unlike in the US, no drug products are marketed in Canada with triclocarban as an active medicinal ingredient nor is it an ingredient in any licensed natural health product (http://webprod3.hc-sc.gc.ca/ lnhpd-bdpsnh/start-debuter.do?lang=eng; http://webprod5.hc-sc.gc. ca/dpd-bdpp/index-eng.jsp, accessed August 20, 2013).

BPA is an industrial chemical that has been used in the manufacture of polycarbonate plastics and in hardeners, paperboard packaging, adhesives, custom color powder coatings and as a curing agent for resurfacing concrete, and is also found in epoxy resins used to line metal food and beverage cans (Environment Canada and Health Canada, 2008). Approximately two-thirds of the intake of BPA in adults is estimated to come from dietary sources, with the balance from other routes (Christensen et al., 2013).

Only two epidemiologic studies have explored potential health effects of prenatal exposure to TCS and reported no significant impact on birth weight, length or head circumference (Wolff et al., 2008; Philippat et al., 2012); however one of these studies did report that male infants were significantly shorter when prenatally exposed to higher levels of TCS (Wolff et al., 2008).

Epidemiologic studies on the effects of prenatal exposure to BPA on pregnancy outcomes are limited, with often no significant effects observed or in the case of total BPA and fetal growth, conflicting results (Lee et al., 2014; Chen et al., 2013; Tang et al., 2013; Snijder et al., 2013; Robledo et al., 2013; Philippat et al., 2012; Cantonwine et al., 2010; Wolff et al., 2008). Similarly some studies have reported no observed effects of prenatal BPA exposure on child behavior (Miodovnik et al., 2011; Yolton et al., 2011) while others found significant effects (Harley et al., 2013a; Perera et al., 2012; Braun et al., 2011a) with support from the toxicological literature for sexually dimorphic pheno-types (Kundakovic et al., 2013; McCaffery et al., 2013; Wolstenholme etal., 2011).

Given the potential for endocrine disruption associated with BPA (Meeker, 2012), triclosan (evidence in rodents with uncertain implications for humans) (Allmyr et al., 2009; Cullinan et al., 2012; Koeppe et al., 2013) and triclocarban (Witorsch and Thomas, 2010), and the paucity of data on early life exposure to these chemicals, a prospective pregnancy study was conducted. As the free phenol is presumed to be more biologically active, we measured both total (free plus conjugated) and free BPA in infant urine, meconium and breast milk.

2. Materials and methods

2.1. Study population

Women in early pregnancy (<20 weeks gestation) were approached at prenatal clinics in Ottawa (Ontario, Canada) between December 2009 and December 2010 and invited to participate in the Plastics and

Personal-care Products use in Pregnancy (P4) Study. A poster and pamphlets were also placed in the obstetrical and ultrasound clinics of The Ottawa Hospital and physician offices. Eligibility criteria included ability to consent and to communicate in English or French, age 18 years or older and planning on delivering within the City of Ottawa. Women with major medical conditions such as renal disease, epilepsy, heart disease and cancer or with known fetal abnormalities or major malformations were excluded from the study. In addition, women who were already participating in 2 or more research studies were disqualified. The study was approved by human studies ethics committees at Health Canada and the Ottawa Hospital and all participants signed an informed consent form.

2.2. Data and biospecimen collection

2.2.1. Maternal urine collection

Women who consented to participate were asked to collect every urine void in separate containers over a 24-hour period during early pregnancy (between 6 and 19 weeks), as well as a spot urine void during the 2nd (24-28 weeks) and 3rd (32-36 weeks) trimesters and 2-3 months post-partum (Table 1). During early pregnancy, women were asked to collect and record the dates and times of all urine voids over a 24-hour period on a week-day (T1a) and/or a week-end day (T1b). Women were provided with pre-screened urine cups (polypropylene) and a cooler bag with ice packs. To avoid degradation of the target chemicals, the urine was kept cool (4 °C) during the collection period, mixed well and aliquoted within 36 h of collection and then stored frozen at — 80 °C. A research assistant visited the participants' homes to retrieve the urine samples that were collected over the 24-hour period. The spot samples were collected during regularly scheduled clinic visits or at home and women were asked to provide information on the time of the void and the time since last void.

2.2.2. infant urine collection

Infant urine was collected twice in the early post-natal period using pre-screened newborn urine-bags (U-bags) (Hollister Inc. Libertyville, IL; and Mabis Healthcare, Waukegan, IL) (Table 1). The first infant urine sample was collected either at the hospital, if possible, or at home up to one month after birth (T4).

Prior to collecting infant urine, the genital area was cleansed using only warm water and a washcloth, allowed to air dry and then a U-bag was attached. When at least 5 mL of urine was collected, the sample was transferred from the U-bag into sterile 30 mL Nalgene® containers. The date and time the bag was removed were noted and the urine refrigerated. Within 24 h of collection, the urine was aliquoted and frozen at — 80 °C. The maximum length of time the U-bag was left attached to the infant was 4 h. Infant urine was also collected 23 months post-partum (T5) during a home visit.

2.2.3. Meconium collection

Pre-screened Mère Hélène® bioliners (Mère Hélène, Quebec Canada) were inserted into the diapers. After the meconium was passed, a wooden spatula was used to transfer the meconium to a 50 mL Sarstedt tube and then refrigerated. The sample was collected on one or more occasions within the first two days after delivery until approximately 10 g was collected. A note was to be made if the diaper was wet with urine and of any lotions, powders, wipes or creams that had been applied to the baby's bottom. The samples were pooled and frozen within 72 h of collection at — 20 °C.

2.2.4. Breast milk and/or infant formula

Two to three months post-partum, a sample of breast milk (minimum 20-30 mL) was collected in a 150 mL glass jar. The women were provided with a Medela® (Medela International, Zug, Switzerland) manual breast pump prior to delivery. Alternatively, women could choose to hand express directly into the glass container. Immediately

Data and biospecimen collection by visit for the P4 Study.

Pregnancy visits Post delivery visits

Early pregnancy (6-<20 weeks) 2nd trimester (24-28 weeks) 3rd trimester (32-36 weeks) <1 month 2-3 months

T1a (weekday) T1b (weekend) T2 T3 T4 T5

Questionnaire Maternal urine Meconium Infant urine Breast milk Infant formula • • (Serial over 24 h) • (Serial over 24 h) • • • • • • (n = 45) • • • (n = 55) • •

after using the breast pump, the milk was transferred directly from the pump bottle to the glass jar. Mothers were asked not to use any creams or cleansers on her breast, prior to pumping. Women who were not breastfeeding or who were supplementing with formula were asked to document the brand of formula and provide a minimum of 2030 mL in the glass jar provided. The breast milk and infant formula were stored in a cooler bag with freezer packs or refrigerated until delivery to hospital. At the lab, the samples were aliquoted into 30 mL Nalgene® containers and frozen at — 20 °C.

22.5. Questionnaires and exposure diary

Upon recruitment, women were asked to complete a short questionnaire and given an exposure diary to complete. The questionnaire collected demographic, socio-economic, obstetrical history, smoking history and current pregnancy information, as well as information on their occupation. Short questionnaires were also completed during pregnancy to update information on the pregnancy, as well as postna-tally to collect data on infant feeding and care practices. Additional data on the pregnancy and the birth were abstracted from medical charts.

2.3. Screening of collection materials and field blankks

All collection materials provided to participants and used for storage were pre-screened for potential contamination. Diaper bioliners were screened with acetonitrile extraction after being soaked overnight. Detectable levels of free BPA were observed (0.007 ^g/g liner). No TCS or TCC was detected in the bioliners and no BPA, TCS or TCC was measured in any of the urine collection or storage materials. While no BPA or TCC was detected in the breast pump, 0.62 ^g/L of TCS was detected (LOD = 0.58 ^g/L).

For each collection period and participant, a field blank was included to assess potential risks of contamination at the collection premises. These would not only consider risks from the sampling collection materials (tubes, transfer vials, etc.) but the collection sites' environment as well. Steril.O reagent grade deionized distilled water was used as the sampling media. Field blanks were handled and analyzed at the same time and similarly to the biospecimens. To create field blanks for the infant urines, 5 ml of the deionized water was added to sample U-bags, jostled to simulate infant movement while wearing the bag and left to sit at room temperature for at least 30 min and then transferred into sterile 30 mL Nalgene® containers.

2.4. Laboratory chemical analysis

To account for varying urine dilutions, specific gravity was measured using an Atago UG-a digital urine refractometer (Atago U.S.A. Inc., Bellevue, WA) on urine samples that had undergone a freeze-thaw cycle.

Biological samples and field blanks were shipped in dry ice to the laboratory in batches where they were kept at — 20 °C until they were placed at 4 °C the day before analysis to be thawed.

For all methods described below, the samples were first thawed, mixed thoroughly and then subsamples were enriched with their corresponding isotopically-labeled carbon 13 analogs (BPA-13C12, TCS-13C12 and TCC-13C13) before treatment.

2.4.1. Measurement of total BPA and TCS, TCC and free BPA in urine

For urine measurements, the analytical methodology and technical information have previously been described (Provencher et al., 2014); for TCC, further details are provided in the Supplemental material. In brief, when the total forms ofBPA, TCS and TCC were measured, samples were, prior to treatment, submitted to enzymatic hydrolysis with (3-glucuronidase in acetate buffer at pH 5 for 16 h at 37 °C. 100 ^L of urine sample were derivatized at 70 °C with pentafluorobenzyl bromide (PFB-Br) and potassium carbonate for 2 h. Total TCC was separately analyzed using sulfuric acid hydrolysis at 80 °C and PFB-Br derivatization catalyzed with cesium carbonate for 1 h at 80 °C. BPA, TCS and TCC pentafluorobenzyl (PFB) derivatives were extracted with a mixture of dichloromethane:hexane (20:80, v:v) concentrated and analyzed by GC-MS/MS.

2.4.2. Measurement of free and total BPA, TCS and TCC in meconium, breast milk and infant formula

The samples of meconium (100 mg), breast milk (100 |jL) or infant formula (100 |jL) were mixed with acetate buffer. For the meconium, the mix of samples was performed by homogenization to assure their homogeneity. When the total forms of BPA, TCS and TCC were measured, the samples were incubated with ( -glucuronidase for 3 h at 37 °C. After protein precipitation with acetonitrile and addition of NaCl, phenols were liquid-liquid extracted using chlorobutane. The organic phase was evaporated and compounds derivatized at 80 °C in the presence of PFB-Br and cesium carbonate for 1 h. After cooling, water and a solution of dichloromethane:hexane (8:92, v:v) was added to the samples for extraction of the PFB derivatives of BPA, TCS and TCC. Extracts were dried under vacuum and reconstituted with a solution of dichloromethane:hexane (20:80, v:v) and analyzed by GC-MS/MS.

2.4.3. Gas chromatographic analysis of BPA, TCS and TCC

All extracts were analyzed using a GC-MS/MS (Agilent 6890 gas chromatograph coupled with a Waters Quattro Micro GC tandem mass spectrometer) equipped with a HP-5MS 30 m, 0.25 mm internal diameter, 0.25 |am film thickness analytical capillary column (Agilent Technologies, Mississauga, ON, Canada). The measurement of ions generated was performed in multiple reaction monitoring (MRM) mode with negative ion chemical ionization (NICI) using methane as the reagent gas.

2.4.4. Method performance

The methods were fully validated for each matrix using ISO 17025 guidelines. Method limits of detection based on repeatability were evaluated as three times the standard deviation of ten replicate analyses of the spiked matrix at levels around two times the LOQ. Limits of

detection, intra-day and inter-day precision are presented in Supplemental Material Table S1.

As described in Provencher et al. (2014), the ID-GC-MS-MS method for total BPA and TCS in urine produces results in excellent agreement with those obtained by ID-LC-MS/MS. The authors only noted a small over-estimation of BPA concentrations by the LC-MS-MS method. The GC-MS/MS method accuracy for total BPA in urine has also been extensively evaluated with success over the last several years by participating in the German external quality assessment scheme.

2.4.5. Laboratory contamination

Procedural reagent blanks were systematically prepared in duplicate in each analytical sequence of 20 samples. TCS and TCC blank levels were always undetected for all types of samples treated. However, BPA blanks were systematically positive with BPA originating mainly from cumulative slight contamination of reagents that remained after extensive elimination of all other potential procedural external contamination. These positive BPA values of reagent blanks were close to the LOD and were highly reproducible (inter- and intra-batches) and were subtracted from the back-calculated results of unknown samples.

2.5. Statistical analysis

Raw data for results less than the limits of detection were available and used in calculating percentiles and geometric means (GM) and 95% confidence intervals of phenols in the various matrices. Since normality assumptions were not satisfied, BPA and TCS concentrations were log transformed and we imputed 0.0001 for values of 0. Both unadjusted and specific gravity adjusted results were calculated, using the following formula for the specific gravity adjustments:

Pc = Pi[(SGm-1)/(SGi-1)]

where Pc is the SG-adjusted metabolite concentration (ng per mL), Pi is the observed metabolite concentration, and SGi is the specific gravity of the urine sample and SGm is the median SG for the maternal or infant cohort.

To compare with other studies, both fresh weight (TCS concentration/1.03 {density of milk}) and lipid-adjusted (fresh weight TCS/% lipids/100) TCS concentrations in breast milk were calculated.

Spearman correlations were calculated to examine associations between maternal and infant matrix concentrations. For calculating correlations of T1a and T1b urinary concentrations with other variables, bootstrapping was used to calculate the median of all correlations and a permutation test was done to determine the statistical significance of the correlation (p-values < 0.05 were deemed statistically significant).

The covariates selected a priori as potential predictors of maternal urinary BPA and TCS concentrations included: maternal age, born in Canada or elsewhere, body mass index (BMI), parity, season of collection, total volume, smoking status, time since last void and time of day of urine collection, household income, education, and occupation. Bivariate modeling was performed to examine associations between each covariate and the log transformed maternal urinary concentrations of BPA and TCS across all visit samples. Specific gravity was included in each model as a covariate. Given that the data involved repeated measurements from the same individual, a linear mixed effects model with random subject effect was used to account for potential correlations of measurements within an individual. SG-adjusted geometric means of maternal urinary chemical concentrations (BPA, TCS) were calculated for all maternal urines by categories of the covariates. Similarly, predictors of infant urinary and breast milk concentrations were evaluated using linear modeling.

Statistical analysis was performed using SAS (Statistical Analysis System) Enterprise Guide 4.2.

3. Results

3.1. Study population

Given the burden on participants, it was difficult to recruit women representative of the study area, with only 11% of those eligible agreeing to participate. The average age of the 80 participants was 32.4 years, with 89% having a college or university degree, 46% having their first pregnancy and about 28% being overweight or obese (See Supplemental Table S2). Smoking prevalence in the study population was very low. Specifically, two participants reported smoking occasionally at T1, one reported smoking at T2 and none at T3 (data not shown). As a result women were categorized into ever (including the 2 occasional smokers) (68%) versus never (32%) smokers.

3.2. Field blanlks

Field blanks for total and free BPA, triclosan and triclocarban in maternal and infant urines and breast milk were generally at or below the limit of detection. Four of the 67 maternal urine field blanks (6%) had total BPA results exceeding the limit of detection, with a maximum reading of 0.40 |ag/L (LOD was 0.2 ^g/L). Three percent (2 of 63) of the breast milk field blanks exceeded the limit of detection of 0.3 |ag/L total BPA with the maximum value of 0.32 |ag/L. As no consistent results were observed for the field blanks and those that were positive were close to the LOD, no adjustments were made.

33. Descriptive statistics

The SG-adjusted median maternal and infant urinary TCS concentrations were 23.3 and 6.1 ^g/L, respectively. Triclosan was detected in 84% of maternal urines, 61% of infant urines, 22% of infant formulas, 46% of breast milk samples, and 81% of the meconium samples; however the geometric means for the infant formula and breast milk were less than the limit of detection (Table 2). Triclocarban was rarely detected in any of the biospecimens with none measured in breast milk or infant formula and less than 4% in urine or meconium (data not shown).

Median SG-adjusted maternal and infant urinary concentrations of total BPA were 1.26 and 0.21 |ag/L, respectively; however, the infant urine median was near the LOD. Most (92%) of the maternal urines, 40% of the infant urines, 54% of the meconium, 30% of the infant formula and only 5% of the breast milk samples had detectable concentrations of total BPA. Only 11% of the infant urines but 48% of the meconium samples had detectable concentrations of free BPA; however, the latter figure is questionable as the liners may have been a source of free BPA. All women had at least one urine sample with detectable concentrations of total BPA, 60% of the infants had at least one urine sample with detectable concentrations of total BPA and 13% with detectable concentrations of free BPA (data not shown).

The geometric mean of the ratios of free to total BPA in infant urine was similar for both the T4 and T5 urine samples (0.4 and 0.5, respectively) with the 50th percentile of 0.12 and 0.17, respectively; however, the geometric mean ratio of free to total BPA in meconium was 0.8 (data not shown).

3.4. Predictors of matrix concentrations

3.4.1. TCS

Maternal urinary triclosan concentrations did not differ significantly whether the sample was collected during or post-pregnancy (Supplemental Fig. S1). Infant urinary triclosan did not differ significantly by sex; however, females had significantly higher geometric mean meconium concentrations (3.25 ng/g) than males (0.97 ng/g). Meconium concentrations of triclosan were significantly correlated with both maternal (especially during the 2nd and 3rd trimesters; r = 0.8) and

Geometric mean (GM) and selected percentiles of triclosan and bisphenol A concentrations in all maternal urines, infant urines, infant formula, breast milk, and meconium for a Canadian population: data from P4 Study 2009-2011.

Matrix GM (95% CI) 10 th 50th 90th 95th Max No. % < LOD

Triclosan

Maternal urine (|og/L) 21.6 (18.17-25.71) <LOD 25.3 (21.2-29.6) 523.2 (471.3-591.7) 833.4(740.7-918.1) 3229.3 1247 16.4

SG-adjusted 22.9 (19.2-27.2) <LOD 23.3 (20.7-26.8) 526.4 (466.2-576.2) 774.9 (673.6-880.8) 2452.4 1247 13.2

Infant urine (|g/L) 2.8 (1.6-4.9) <LOD 3.9 (3.1-5.1) 21.5 (14.9-61.0) 52.0 (22.7-100.0) 99.9 100 39.0

SG-adjusted 2.5 (1.5-4.4) <LOD 6.1 (4.8-8.5) 35.3 (16.9-53.4) 53.4 (35.2-229.8) 229.8 98 24.0

Infant formula (|g/L) < LOD <LOD < LOD 1.0 (0.9-3.8) 1.1 (1.0-3.8) 3.8 23 78.3

Breast milk (|g/L) 0.05 (0.01-0.2) <LOD < LOD 3.6 (1.7-16.0) 8.8 (3.3-75.4) 75.4 56 53.6

Fresh weight (ng/g) 0.05 <LOD 0.45 3.52 8.51 73.18 56 56.3%

Lipid (ng/g lipids) 2.5 <LOD 19.2 95.2 156.0 2287.0 52 56.3%

Meconium (ng/g) 2.24 (0.97-5.2) <LOD 3.2 (1.3-12.2) 36.5 (24.6-77.0) 68.8 (30.9-77.0) 77.0 52 19.2

Total BPA

Maternal urine ( | g/L) 1.1 (1.0-1.2) 0.3 1.2 4.0 6.4 297.8 1238 8

SG-adjusted 1.2 (1.1-1.3) 0.5 1.3 3.7 5.6 191.4 1238 4

Infant urine ( | g/L) < LOD <LOD 0.2 2.3 3.4 9.4 100 60

SG-adjusted < LOD <LOD 0.2 1.4 3.4 12.3 100 61

Infant formula ( | g/L) < LOD <LOD <LOD 1.2 1.5 9.0 23 70

Breast milk (| g/L) < LOD <LOD < LOD <LOD 0.4 1.9 56 95

Meconium (ng/g) 0.52 (0.39-0.69) <LOD 0.60 1.65 2.65 3.93 54 46

Free BPA

Infant urine ( | g/L) < LOD <LOD < LOD 0.2 0.6 9.2 100 89

SG-adjusted < LOD <LOD < LOD 0.2 0.4 12.0 100 89

Breast milk (| g/L) <LOD <LOD < LOD <LOD <LOD 1.6 56 96

Infant formula ( | g/L) <LOD <LOD < LOD 1.0 1.6 9.0 23 74

LOD: limit of detection.

LODs in urine: free BPA: 0.1; total BPA: 0.2; triclosan: 3.0 |g/L LODs in breast milk/formula: free BPA: 0.30; total BPA: 0.30; triclosan: 0.58 |g/L. LODs in meconium: total BPA: 0.48; triclosan: 0.49 ng/g.

infant urinary concentration shortly after birth (r = 0.5). Significant correlations were also observed between triclosan concentrations in breast milk and maternal urine collected at visits T2 (r = 0.5), T3 (r = 0.4) and T5 (r = 0.6) and infant urine collected at the same time as the breast milk (r = 0.4). However, the geometric mean TCS in infant urine at T5 did not vary by whether or not the infant was exclusively breastfed, nor for breast milk, by how the sample was collected (Table 3).

Significantly lower maternal urinary concentrations of TCS were found in samples collected between 4 p.m. and midnight, less than 90 min from the previous void, and on a weekday (Table 4). TCS concentrations were highest in samples collected in the autumn and in primiparous women.

3.4.2. BPA

No significant differences in maternal urinary concentrations of BPA were observed by visit (Supplemental Fig. S2). Infant urinary concentrations of total and free BPA, adjusted for specific gravity did not differ

significantly by visit or by sex (data not shown); however, females had significantly higher geometric mean concentrations (p < 0.05) of both total (0.75 ng/g) and free BPA (0.45 ng/g) in their meconium than males (0.31 and 0.14 ng/g, respectively). No significant correlations were observed between maternal or infant urine concentrations of total BPA and meconium or breast milk (data not shown). A significant correlation was observed between free BPA in infant urine and breast milk (r = 0.36), albeit based on small numbers and low detection frequency.

Infants who were fed exclusively with formula had significantly higher total BPA in their urine, compared to those who were exclusively breast fed or by a combination of both formula and breast feeding (Table 3).

Some factors describing when the urine was collected were significant predictors of elevated maternal total BPA urinary concentration, specifically time of day (between 4 p.m. and midnight) and on weekends (Table 4). Urine collected within 90 min of the previous void had significantly lower concentrations of total BPA. While the sample size

Table 3

Specific gravity-adjusted infant urinary and breast milk concentrations of phenols by infant feeding and breast milk collection practices.

Infant urine (T5)

Total BPA

Free BPA

Total TCS

p-Valuea

GM (95% CI)

p-Valuea

GM (95% CI)

p-Valuea

GM (95% CI)

Infant feeding Exclusively breast fed Exclusively formula Combination of the two

Breast milkb Collected by: Hand

Mechanical pump Electrical pump

33 6 14

16 20 17

0.09 (0.03, 0.25) 1.04(0.20, 5.36) 0.35 (0.18, 0.69)

0.01 (0.00, 0.05)

0.01 (0.00,4.07) 0.02 (0.00, 0.16)

2.70 (1.13, 6.44) 6.21 (1.49,25.85) 2.18 (0.23,21.00)

0.02 (0.00, 0.45) 0.05 (0.00, 0.45) 0.23 (0.03,1.94)

a p-Value for overall group effect using ANOVA

b BPA and free BPA concentrations not shown because >95% below the detection limit.

Specific gravity (SG) adjusted geometric means (GM) and bivariate predictors of maternal urinary BPA and TCS with specific gravity included as a covariate in the fixed effects regression model (n = up to 1269 urines from up to 80 women).

Characteristic

N (# samples)

BPA(|jg/L)

p-Valuea

SG-adjusted GM

TCS (|g/L)

p-Valuea

SG-adjusted GM

Time of day

00:00-07:59

08:00-11:59

12:00-15:59

16:00-23:59

282 302 235 412

< 0.0001 < 0.0001 0.01

Referent

1.18 1.05 1.31 1.63

0.02 0.01 0.04

Referent

29.47 28.55 28.11 18.37

Time since last void (min.)

90-149

150-209

284 331 214 315

0.02 0.58 0.47

Referent

1.16 1.27 1.35 1.45

0.05 0.71 0.61

Referent

15.60 30.70 30.74 29.53

Total volume (mL)

362 296 455

0.43 0.33 0.72

Referent

1.30 1.18 1.37 1.35

0.99 0.72 0.28

Referent

31.91 28.11 21.66 23.09

Collection Day

Weekday

Weekend

686 574

Referent 0.02

1.20 1.42

Referent 0.008

21.28 29.17

Season of collection

Spring

Summer

Autumn

Winter

299 297 387 277

Referent

1.29 1.21 1.41 1.24

0.06 0.05 0.001

Referent

21.08 23.86 33.22 19.42

Pre-pregnancy BMI Underweight or normal Overweight or obese

Referent 0.25

1.29 1.52

Referent 0.35

18.75 30.11

Parity 0 1

546 569 145

0.35 0.57

Referent

1.55 1.11 1.19

0.04 0.29

Referent

41.11 21.19 6.35

Collection period

6-19 weeks (weekday)

6-19 weeks (weekend day)

24-28 weeks

32-36 weeks

2-3 months post-partum

Born in Canada

512 544

237 1023

0.44 0.09 0.15 0.86

Referent

Referent 0.02

1.42 1.04 1.13

0.95 1.39

0.81 0.21 0.94 0.64

Referent

Referent 0.89

21.68 28.81 25.11 20.45 20.63

18.01 26.37

Maternal age (years) <30 30-35 > 35

218 588 436

Referent

1.32 1.32 1.27

Referent

18.53 26.77 21.93

Household income <$100,000 >$100,000

468 732

Referent

1.21 1.36

Referent

14.63 33.63

Education

Less than college diploma College diploma University degree MSc or PhD

114 198 590 358

0.01 0.63

Referent 0.03

1.83 1.20 1.11 1.55

0.89 0.27

Referent 0.64

15.78 39.73 22.02 25.87

Smoking status

825 402

Referent 0.37

1.22 1.47

Referent

29.65 13.33

Marital status Married

Single or with partner

Employment status Health care worker Office or government worker Unemployed

Other jobs (teachers, sales, etc.)

1029 231

254 472 198 336

Referent

0.07 0.02

Referent 0.006

1.22 1.68

1.27 1.36 0.92 1.50

Referent

0.10 0.11

Referent 0.10

29.58 66.92

37.45 30.96 5.37 31.58

a p-Values calculated on log-transformed SG-adjusted concentrations using a linear mixed effects model with random subject effect to account for potential correlations of measurements within an individual.

was small, no significant differences in BPA urinary concentrations were observed by maternal age, household income, parity or pre-pregnancy BMI categories. Unemployed women had significantly lower exposure to BPA while women born in Canada had significantly higher urinary BPA.

4. Discussion

4.1. TCS and TCC

The geometric mean and median urinary triclosan concentrations of all maternal samples (n = 1247) were 21.6 and 25.3 |ag/L, respectively, which are higher than those reported in studies of pregnant women in California (California, 2013), New York (Wolff et al., 2008; Philippat et al., 2013), Spain (Casas et al., 2011), Norway (Bertelsen et al., 2014), Denmark (Tefre de Renzy-Martin et al., 2014; Frederiksen et al., 2014), US NHANES (Woodruff et al., 2011), US pilot sites for the National Children's Study (NCS) (Mortensen et al., 2014) and Canada (Arbuckle et al., 2014a), but comparable to those measured in France (Philippat et al., 2012) and Puerto Rico (Meeker et al., 2013) (Supplemental Fig. S3). It is unknown whether differences in laboratory methods may account for the disparities observed; however, multiple urine samples per woman were collected in our P4 Study while most of the other studies, including the other Canadian study collected only one urine sample, thereby reducing the impact of intra-individual variability over time.

A recent Chinese study reported mean total triclosan in urine of infants < 6 months of age of 4.04 ng/mL (LOD 0.13) (Liu et al., 2014). A study of infants in neonatal intensive care units has measured triclosan in infant urine and reported that 81% of the samples were below the LOD (2.3 ^g/L) with a maximum value of 16.7 |ag/L (Calafat et al., 2009). Thirty-nine percent of our infant urines were < LOD (3 ^g/L) and the maximum result was 99.9 |ag/L.

Triclosan has been measured in breast milk in several countries. On a fresh weight basis, higher maximum values were reported in our study (median: 0.45 ng/g; range: 0-73.18 ng/g) than in Australia (median: 0.26; range: <0.019-19 ng/g) (Toms et al., 2011); however a Swedish study reported the highest maximum value (median in "exposed" 0.54; range: <0.018-0.95 ng/g) (Allmyr et al., 2006). Our maximum lipid-adjusted results (2287 ng/g lipids) were higher than those reported in the U.S. (2100 ^g/kg) (Dayan, 2007) and China (308.6 ng/g lipids) (Wang et al., 2011).

Similar to the Flemish Health and Environment Study (Den Hond et al., 2013), we observed very large inter-individual variability in urinary concentrations of TCS (from none detected to 3229 ^g/L). In the Flemish Study, higher urinary TCS was associated with increased use of personal care products but not with age, BMI, educational level, smoking or urbanization (Den Hond et al., 2013). Higher maternal urinary TCS concentrations have been reported in women with more education and income (Bertelsen et al., 2014; Arbuckle et al., 2014a). We observed no consistent significant associations between maternal exposure to TCS and any socio-demographic variables; however, somewhat higher TCS concentrations in women with more education and income were noted.

Triclocarban was rarely detected in any of the biospecimens tested in our study (4% of maternal urines), which may reflect limited use of this substance in Canada as well as a less sensitive method for measuring TCC (LOD in urine: 1.1 ^g/L; whereas the CDC method had a LOD of 0.1 ng/mL and detected TCC in 35% of urines (Zhou et al., 2012)). An analysis of 100 urine samples in Greece for TCC (LOQ was 0.5 ng/mL) found only 4% with detectable concentrations (Asimakopoulos et al., 2014). In contrast, among 200 pregnant women in Denmark, 54% had detectable concentrations ofTCC in their urine, with a LOD of 0.01 ^g/L (Tefre de Renzy-Martin et al., 2014); however TCC was detected in only 18% of urines from a larger group of the Danish cohort (Frederiksen et al., 2014) and TCC was detected in 25%

of first morning voids from mothers in another study (Frederiksen et al., 2013).

42. BPA

In contrast to triclosan, there have been a number of studies published which have measured urinary concentrations of total BPA during pregnancy (Supplemental Fig. S4). Median concentrations were at least 2.0 ^g/L in studies conducted in France (Philippat et al., 2012), Cincinnati (Braun et al., 2011b), Spain (Casas et al., 2013), Australia (Callan et al., 2013) and Puerto Rico (Meeker et al., 2013) and less than 1.0 ^g/L in New York Mennonites (Martina et al., 2012), Mexico (Lewis et al., 2013) and Canada (Arbuckle et al, 2014b). The median BPA in our study (1.21 |ag/L) was comparable to that of cohorts in California (Harley et al, 2013b), New York (Wolff et al, 2008), Korea (Lee et al., 2014), NCS pilot sites (Mortensen et al., 2014) and Denmark (Tefre de Renzy-Martin et al, 2013; Frederiksen et al., 2014). Differences in dietary exposures (through types and amounts of contaminated foods consumed), sampling protocols and to a lesser extent, other sources of BPA exposure such as paper and paper products, including thermal receipts (Liao and Kannan, 2011) may explain variances in exposure between populations.

We noted that infants who were exclusively fed with formula had significantly higher total BPA in their urine, compared to those who were exclusively breastfed or by a combination of both formula and breast feeding. As BPA was not detected in 70% of the formula and 95% of the breast milk, this suggests that the bottles may be the main source of BPA for the infants. However, information was not available on the frequency of feeding with bottles containing expressed breast milk.

A few studies have measured infant exposure to BPA. Studies of hospitalized infants (NICU) reported urinary concentrations of free and total BPA at least one order of magnitude higher (Calafat et al., 2009; Duty et al., 2013) than in studies of healthy infants, who generally had median free BPA below the limit of detection and total BPA less than 2 |ag/L (Liu et al., 2014; Mendonca et al., 2014; Nachman et al., 2013), comparable to what we observed.

Several studies have measured free and total BPA in breast milk but at higher concentrations than those measured in our study where approximately 95% of the samples had no detectable BPA. For example, the median total BPA in three US studies ranged from around 0.8 ^g/L (Ye et al., 2008; Mendonca et al., 2014) to 1.3 ^g/L (Duty et al., 2013), and free BPA ranged from less than the LOD to around 0.7 ^g/L (Ye et al., 2008; Zimmers et al., 2014). While the most recent American study (Zimmers et al., 2014) had a lower LOD (0.22 ng/mL), than that of our study (0.3 ng/mL), this likely does not explain the higher detection of free BPA (60%) versus 4% in our study. Another American study (Mendonca et al., 2014) had the same LOD for free BPA as our study but reported 20% of the milk samples had detectable free BPA.

A New York City cohort study has reported that geometric means for maternal prenatal BPA concentrations in urine were significantly lower than paired children's postnatal concentrations at ages 3 to 7 years and suggested that BPA in pregnant women may be temporarily diverted from the excretory process due to transfer across the placenta (Hoepner et al., 2013). However there is no evidence from our study or that of Braun and colleagues (Braun et al., 2012) that urinary BPA concentrations are overall lower in pregnancy than in the non-pregnant state. Braun et al., 2011b did report median BPA decreasing across pregnancy, while our study reported the highest concentrations at 32-36 weeks.

While we observed that autumn collections were a significant predictor of BPA exposure, one study has reported BPA geometric mean concentrations in urine samples collected in the summer months were consistently higher than concentrations in non-summer months (Hoepner et al., 2013). Urinary BPA concentrations have been positively

associated with water consumption from re-usable polycarbonate bottles in summer months among females (Makris et al., 2013).

4.3. Meconium

No previous study has measured phenols in meconium. As meconi-um begins to form in the fetus as early as the 12th week of gestation, chemicals can accumulate in meconium through to delivery (Ostrea et al., 2006). Meconium is therefore a cumulative repository of many of the xenobiotics that the fetus is exposed to throughout pregnancy, compared to an acute phase matrix such as blood or urine. Hence, meconium may be a better matrix to measure prenatal exposure to short-lived chemicals such as phenols. While meconium can be messy to handle and complex to analyze, large amounts can be collected. Unlike cord blood, there is no competition with tissue banks for this material. We have no explanation for the observation of significantly higher concentrations of phenols in meconium from females than males. The meconium results must be interpreted with caution, however, especially given that the phenols are non-persistent chemicals frequently detected in urine. Cross-contamination of the meconium sample with urine may have occurred which would reflect exposure during gestation as well as during delivery (Calafat and Needham, 2009). The sex differences in concentrations in meconium were not reflected in infant urines. This suggests that contamination of the meconium with infant urine was not an issue in this study or anatomical differences resulted in female meconium being more contaminated with urine than male meconium. However, our examination of phthal-ate metabolites in meconium did not show any differences by sex (unpublished data). The length of time that the meconium is in contact with the diaper may also affect chemical concentrations as the fluid in the meconium may be retained by the diaper, resulting in artificially lower concentrations. Although staff were told to record whether there was evidence of urine in the diaper, in most cases, no information was provided. In addition, in a few cases, the provided diaper liner was not used. Vaseline® and Pampers® wipes were commonly used in the hospitals; however, there is no evidence that these products contained triclosan or BPA (US Department of Health and Human Services, http://hpd.nlm.nih.gov/index.htm, accessed December 24, 2013). Additionally, the meconium concentration could be affected if all the meconium was not collected and well homogenized, as meconium excreted first can vary in concentration to later excretions which may also contain postnatal stool (Zelner et al., 2012).

4.4. Potential contamination ofsamples

Unintended random or systematic contamination of the samples in our study cannot be ruled out. Samples of the collection materials were pre-screened and field blanks were included to try to minimize blatant sources of contamination. Our analysis showed only trace contamination of 6% of the maternal and 3% of the breast milk field blanks with total BPA, which suggests that potential contamination from the processes was minimal. BPA contamination of milk samples has been reported from the leaching of BPA from a specific lot of microcentrifuge tubes, even though the tubes were presented as made of polypropylene, which is not known to contain BPA (Ye et al., 2013). Contamination of the laboratory can also come from indoor dust which has been found to contain BPA (Loganathan and Kannan, 2011) and triclosan (Fan et al., 2010). The absorbent underpad used to cover the workspace in some laboratory benches may also be a source of BPA contamination (Ye et al., 2013). Random contamination with triclosan has also been observed in labs where the hand soap in the restroom dispensers contained triclosan and when one of the analysts used triclosan containing toothpaste (Ye et al., 2013). As Ye and colleagues noted, random contamination can be difficult to identify and trace even by experienced researchers adhering to a comprehensive quality control protocol.

5. Conclusions

Our results showed that the main predictors of maternal exposure to BPA or TCS, adjusted for specific gravity were time of day, weekend and autumn collections. Time since last void was only a predictor of maternal TCS exposure. However, these results are tempered by the fact that the sample size is small and the women are primarily from a higher socio-demographic group than the general population. After finding considerable within day variability for BPA, Ye et al. (2011b) suggested that the time of day of urine collection and the time since last void be collected when measuring this chemical. Both the timing of the sample relative to food consumption and previous bladder voiding has a direct impact on the estimated exposure concentrations in spot urines (Ye et al., 2011b). We found significantly higher concentrations of BPA, in the evening between 4:00 pm and midnight versus during the day or overnight; whereas TCS concentrations were significantly higher in the mornings and early afternoons. Therefore we recommend that time of urine collection and time since last void be routinely collected in biomonitoring studies.

The correlations observed between maternal urinary triclosan with meconium, breast milk and infant urine concentrations, suggest that maternal exposures may be the main source of exposure for the young infant. Our small sample size limited the study's ability to identify maternal characteristics (e.g., household income, age, education, occupation) that were predictive of maternal urinary phenol concentrations.

In conclusion, our study showed that the majority of pregnant women in this highly educated population were exposed to BPA and TCS and that their fetuses and infants are also exposed but to a lesser extent. Data from extensive biomonitoring studies in susceptible subpopulations such as that in the P4 Study, will add to the quality of the risk assessment.

Acknowledgments

The authors acknowledge the work of Ruth White, Pauline Shields and Adam Probert on recruitment and data and biospecimen collection and/or data management. We are grateful for the advice of Antonia Calafat on specimen collection and comments on an earlier draft of the manuscript. Special thanks to the women who took the time and effort to participate in this demanding study. This work was funded by Health Canada's Chemicals Management Plan.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.10.107.

References

Allmyr M, Adolfsson-Erici M, McLachlan MS, Sandborgh-Englund G. Triclosan in plasma and milk from Swedish nursing mothers and their exposure via personal care products. Sci Total Environ 2006;372(1):87-93. [Dec 15 Epub 2006 Sep 26]. Allmyr M, Panagiotidis G, Sparve E, Diczfalusy U, Sandborgh-Englund G. Human exposure to triclosan via toothpaste does not change CYP3A4 activity or plasma concentrations of thyroid hormones. Basic Clin Pharmacol Toxicol 2009 Novv;105(5):339-44. http://dx.doi.org/10.1111/j.1742-7843.2009.00455.x. [Nov Epub 2009 Jul 20]. Arbuckle TE, Marro L, Davis K, Fisher M, Ayotte P, Bélanger P, et al. Exposure to free and conjugated forms of bisphenol A and triclosan among pregnant women in the MIREC Cohort. Environ Health Perspect 2014a. http://dx.doi.org/10.1289/ehp. 1408187. [online Nov 21, 2014]. Arbuckle TE, Davis K, Marro L, Fisher M, Legrand M, LeBlanc A, et al. Phthalate and bisphenol A exposure among pregnant women in Canada — results from the MIREC study. Environ Int 2014b;68C:55-65. http://dx.doi.org/10.1016/j.envint.2014.02.010. [Apr 3 [Epub ahead of print]]. Asimakopoulos AG, Thomaidis Ns, Kannan K. Widespread occurrence of bisphenol A diglycidyl ethers, p-hydroxybenzoic acid esters (parabens), benzophenone type-UV filters, triclosan, and triclocarban in human urine from Athens, Greece. Sci Total Environ 2014;470-471:1243-9.

Bertelsen RJ, Engel SM. Jusko TA, Calafat AM, Hoppin JA. London SJ. Reliability of triclosan measures in repeated urine samples from Norwegian pregnant women. J Expos Sci Environ Epidemiol 2014;24(5):517-21. http://dx.doi.org/10.1038/jes.2013.95. [Jan 29 [Epub ahead of print]].

Braun JM, Kalkbrenner AE, Calafat AM, Yolton K, Ye X, Dietrich KN, et al. Impact of early-life bisphenol A exposure on behavior and executive function in children. Pediatrics 2011a;128(5):873-82. http://dx.doi.org/10.1542/peds.2011-1335. [Nov [Epub 2011 Oct 24]].

Braun JM, Kalkbrenner AE, Calafat AM, Bernert JT, Ye X, Silva MJ, et al. Variability and predictors of urinary bisphenol A concentrations during pregnancy. Environ Health Perspect 2011b;119(1):131—7. [Jan].

Braun JM, Smith KW, Williams PL, Calafat AM, Berry K, Ehrlich S, et al. Variability of urinary phthalate metabolite and bisphenol A concentrations before and during pregnancy. Environ Health Perspect 2012;120(5):739-45. [May Epub 2012 Jan 19].

Calafat AM, Needham LL. What additional factors beyond state-of-the-art analytical methods are needed for optimal generation and interpretation of biomonitoring data? Environ Health Perspect 2009;117(10):1481-5. http://dx.doi.org/10.1289/ ehp.0901108. [Oct Epub 2009 Jun 24].

Calafat AM, Weuve J, Ye X, Jia LT, Hu H, Ringer S, et al. Exposure to bisphenol A and other phenols in neonatal intensive care unit premature infants. Environ Health Perspect 2009;117(4):639-44. [Apr Epub 2008 Dec 10].

California. MIEEP project. http://biomonitoring.ca.gov/projects/maternal-and-infant-environmental-exposure-project-mieep. [accessed July 22, 2013].

Callan AC, Hinwood AL, Heffernan A, Eaglesham G, Mueller J, Odland JO. Urinary bisphenol A concentrations in pregnant women. Int J Hyg Environ Health 2013;216(6):641-4.

Cantonwine D, Meeker JD, Hu H, Sánchez BN, Lamadrid-Figueroa H, Mercado-García A, et al. Bisphenol A exposure in Mexico City and risk of prematurity: a pilot nested case control study. Environ Health 2010;9:62. http://dx.doi.org/10.1186/1476-069X-9-62. [Oct 18].

Casas L, Fernández MF, Llop S, Guxens M, Ballester F, Olea N, et al. INMA Project. Urinary concentrations ofphthalates and phenols in a population of Spanish pregnant women and children. Environ Int 2011;37(5):858-66. [Jul Epub 2011 Mar 25].

Casas M, Valvi D, Luque N, Ballesteros-Gomez A, Carsin AE, Fernandez MF, et al. Dietary and sociodemographic determinants of bisphenol A urine concentrations in pregnant women and children. Environ Int 2013;26(56C):10-8. http://dx.doi.org/10.1016/j. envint.2013.02.014. [Mar [Epub ahead of print]].

CDC. Fourth national report on human exposure to environmental chemicals. Updated tables March 2013. Department of Health and Human Services Centers for Disease Control and Prevention. http://www.cdc.gov/exposurereport/pdf/FourthReport_ UpdatedTables_Mar2013.pdf.

Chen X, Chen M, Xu B, Tang R, Han X, Qin Y, et al. Parental phenols exposure and spontaneous abortion in Chinese population residing in the middle and lower reaches of the Yangtze River. Chemosphere 2013;93(2):217-22. http://dx.doi.org/ 10.1016/j.chemosphere.2013.04.067. [Sep Epub 2013 May 25].

Christensen KL, Lorber M, Ye X, Calafat AM. Reconstruction of bisphenol A intake using a simple pharmacokinetic model. J Expo Sci Environ Epidemiol 2013. http://dx.doi.org/ 10.1038/jes.2013.81. [Nov 20 [Epub ahead of print]].

Cullinan MP, Palmer JE, Carle AD, West MJ, Seymour GJ. Long term use of triclosan toothpaste and thyroid function. Sci Total Environ 2012;416:75-9. http://dx.doi.org/10. 1016/j.scitotenv.2011.11.063. [Feb 1 Epub 2011 Dec 22].

Dayan AD. Risk assessment of triclosan [Irgasan] in human breast milk. Food Chem Toxicol 2007;45(1):125-9. [Jan Epub 2006 Aug 30].

Den Hond E, Paulussen M, Geens T, Bruckers L, Baeyens W, David F, et al. Biomarkers of human exposure to personal care products: results from the Flemish Environment and Health Study (FLEHS 2007-2011). Sci Total Environ 2013;20(463-464C):102-10. http://dx.doi.org/10.1016/j.scitotenv.2013.05.087. [Jun [Epub ahead of print]].

Duty SM, Mendonca K, Hauser R, Calafat AM, Ye X, Meeker JD, et al. Potential sources of bisphenol A in the neonatal intensive care unit. Pediatrics 2013;131(3):483-9. http://dx.doi.org/10.1542/peds.2012-1380. [Mar Epub 2013 Feb 18].

Environment Canada and Health Canada. Screening assessment for the Challenge Phenol, 4,4' -(1-methylethylidene)bis-(bisphenol A), chemical abstracts service registry number 80-05-7, October 2008. http://www.ec.gc.ca/ese-ees/default.asp?lang= En&n=3C756383-1#a5. [accessed August 20, 2013].

Fan X, Kubwabo C, Rasmussen P, Jones-Otazo H. Simultaneous quantitation of parabens, triclosan, and methyl triclosan in indoor house dust using solid phase extraction and gas chromatography-mass spectrometry. J Environ Monit 2010;12(10):1891-7. http://dx.doi.org/10.1039/c0em00189a. [Oct 6 Epub 2010 Sep 6].

Frederiksen H, Nielsen JKS, Morck TA, Hansen PW, Jensen JF, Nielsen O, et al. Urinary excretion of phthalate metabolites, phenols and parabens in rural and urban Danish mother-child pairs. Int J Hyg Environ Health 2013;216(6):772-83. [March 13 http://dx.doi.org/10.1016/j.ijheh2013.02.006].

Frederiksen H, Jensen TK, Jorgensen N, Boye Kyhl H, Husby S, Skakkebaek NE, et al. Human urinary excretion of non-persistent environmental chemicals: an overview of Danish data collected 2006-2012. Reproduction 2014;147(4):555-65. [Jan 6 [Epub ahead of print]].

Harley KG, Gunier RB, Kogut K, Johnson C, Bradman A, Calafat AM, et al. Prenatal and early childhood bisphenol A concentrations and behavior in school-aged children. Environ Res 2013a;126:43-50. http://dx.doi.org/10.1016/j.envres.2013.06.004. [Oct Epub 2013 Jul 17].

Harley KG, Aguilar Schall R, Chevrier J, Tyler K, Aguirre H, Bradman A, et al. Prenatal and postnatal bisphenol A exposure and body mass index in childhood in the CHAMACOS cohort. Environ Health Perspect 2013b;121(4):514-20. http://dx.doi.org/10.1289/ ehp.1205548. [Apr 520e1-6. Epub 2013 Feb 14].

Health Canada and Environment Canada. Preliminary assessment — triclosan. March

2012 http://www.ec.gc.ca/ese-ees/6EF68BEC-5620-4435-8729-9B91 C57A9FD2/Tri-closan_EN.pdf.

Health Canada. Second report on human biomonitoring of environmental chemicals in Canada. Results of the Canadian Health Measures Survey Cycle 2 (2009-2011), 2. 978-1-100-22140-3; 2013. [April 2013. HC Pub.: 130019, Cat.: H128-1/10-601-1E-PDF www.healthcanada.gc.ca/biomonitoring].

Hoepner LA, Whyatt RM, Just AC, Calafat AM, Perera FP, Rundle AG. Urinary concentrations of bisphenol A in an urban minority birth cohort in New York City, prenatal through age 7 years. Environ Res 2013;122:38-44. http://dx.doi.org/10.1016/j. envres.2012.12.003. [Apr Epub 2013 Jan 8].

Koeppe ES, Ferguson KK, Colacino JA, Meeker JD. Relationship between urinary triclosan and paraben concentrations and serum thyroid measures in NHANES 2007-2008. Sci Total Environ 2013;445-446:299-305. http://dx.doi.org/10.1016/j.scitotenv. 2012.12.052. [Feb 15 Epub 2013 Jan 20].

Kundakovic M, Gudsnuk K, Franks B, Madrid J, Miller RL, Perera FP, et al. Sex-specific epigenetic disruption and behavioral changes following low-dose in utero bisphenol A exposure. Proc Natl Acad Sci U S A 2013;110(24):9956-61. http://dx.doi.org/10. 1073/pnas.1214056110. [Jun 11 Epub 2013 May 28].

Lee BE, Park H, Hong YC, Ha M, Kim Y, Chang N, et al. Prenatal bisphenol A and birth outcomes: MOCEH (Mothers and Children's Environmental Health) study. Int J Hyg Environ Health 2014;217(2-3):328-34. http://dx.doi.org/10.1016/j.ijheh.2013.07. 005. [Jul 11 [Epub ahead of print]].

Lewis RC, Meeker JD, Peterson KE, Lee JM, Pace GG, Cantoral A, et al. Predictors of urinary bisphenol A and phthalate metabolite concentrations in Mexican children. Chemosphere 2013;93(10):2390-8. http://dx.doi.org/10.1016/j.chemosphere.2013. 08.038. [Sep 13. [Epub ahead of print]].

Liao C, Kannan K. Widespread occurrence of bisphenol A in paper and paper products: implications for human exposure. Environ Sci Technol 2011;45(21):9372-9. http:// dx.doi.org/10.1021/es202507f. [Nov 1 Epub 2011 Oct 5].

Liu L, Xia T, Zhang X, Barr DB, Alamdar A, Zhang J, et al. Biomonitoring of infant exposure to phenolic endocrine disruptors using urine expressed from disposable gel diapers. Anal Bioanal Chem 2014;406(20):5049-54. [Jun 13].

Loganathan SN, Kannan K. Occurrence of bisphenol A in indoor dust from two locations in the eastern United States and implications for human exposures. Arch Environ Contam Toxicol 2011;61(1):68-73. http://dx.doi.org/10.1007/s00244-010-9634-y. [Jul Epub 2011 Jan 8].

Makris KC, Andra SS, Jia A, Herrick L, Christophi CA, Snyder SA, et al. Association between water consumption from polycarbonate containers and bisphenol A intake during harsh environmental conditions in summer. Environ Sci Technol 2013;47(7): 3333-43. http://dx.doi.org/10.1021/es304038k [Apr 2 Epub 2013 Mar 15].

Martina CA, Weiss B, Swan SH. Lifestyle behaviors associated with exposures to endocrine disruptors. Neurotoxicology 2012;33(6):1427-33. http://dx.doi.org/10.1016/j.neuro. 2012.05.016. [Dec Epub 2012 Jun 26].

McCaffrey KA, Jones B, Mabrey N, Weiss B, Swan SH, Patisaul HB. Sex specific impact of perinatal bisphenol A (BPA) exposure over a range of orally administered doses on rat hypothalamic sexual differentiation. Neurotoxicology 2013;36:55-62. http://dx. doi.org/10.1016/j.neuro.2013.03.001. [May Epub 2013 Mar 13].

Meeker JD. Exposure to environmental endocrine disruptors and child development. Arch Pediatr Adolesc Med 2012;166(10):952-8. [Oct].

Meeker JD, Cantonwine DE, Rivera-González LO, Ferguson KK, Mukherjee B, Calafat AM, et al. Distribution, variability, and predictors of urinary concentrations of phenols and parabens among pregnant women in Puerto Rico. Environ Sci Technol 2013; 47(7):3439-47. [Mar 19 [Epub ahead of print]].

Mendonca K, Hauser R, Calafat AM, Arbuckle TE, Duty SM. Bisphenol A concentrations in maternal breast milk and infant urine. Int Arch Occup Environ Health 2014;87(1): 13-20. http://dx.doi.org/10.1007/s00420-012-0834-9. [Jan Epub 2012 Dec 5].

Miodovnik A, Engel SM, Zhu C, Ye X, Soorya LV, Silva MJ, et al. Endocrine disruptors and childhood social impairment. Neurotoxicology 2011;32(2):261-7. http://dx.doi.org/ 10.1016/j.neuro.2010.12.009. [Mar Epub 2010 Dec 21].

Mortensen ME, Calafat AM, Ye X, Wong Y-L, Wright DJ, Pirkle JL, et al. Urinary concentrations of environmental phenols in pregnant women in a pilot study of the National Children's Study. Environ Res 2014;129:32-8.

Nachman RM, Fox SD, Golden WC, Sibinga E, Veenstra TD, Groopman JD, et al. Urinary free bisphenol A and bisphenol A-glucuronide concentrations in newborns. J Pediatr 2013;162(4):870-2. http://dx.doi.org/10.1016/jjpeds.2012.11.083. [Apr Epub

2013 Jan 11].

Ostrea Jr Em, Bielawski DM, Posecion Jr NC. Meconium analysis to detect fetal exposure to neurotoxicants. Arch Dis Child 2006;91 (8):628-9. http://dx.doi.org/10.1136/adc. 2006.097956. [August].

Perera F, Vishnevetsky J, Herbstman JB, Calafat AM, Xiong W, Rauh V, et al. Prenatal bisphenol a exposure and child behavior in an inner-city cohort. Environ Health Perspect 2012;120(8):1190-4. http://dx.doi.org/10.1289/ehp.1104492. [Aug Epub 2012 Apr 27].

Philippat C, Mortamais M, Chevrier C, Petit C, Calafat AM, Ye X, et al. Exposure to phthalates and phenols during pregnancy and offspring size at birth. Environ Health Perspect 2012;120(3):464-70. http://dx.doi.org/10.1289/ehp.1103634. [Mar Epub 2011 Sep 7. Erratum in: Environ Health Perspect. 2012 Mar;120(3):470].

Philippat C, Wolff MS, Calafat AM, Ye X, Bausell R, Meadows M, et al. Prenatal exposure to environmental phenols: concentrations in amniotic fluid and variability in urinary concentrations during pregnancy. Environ Health Perspect 2013. [Aug 13 [Epub ahead of print]].

Provencher G, Bérubé R, Dumas P, Bienvenu JF, Gaudreau G, Bélanger P, et al. Determination of bisphenol A, triclosan and their metabolites in human urine using isotope-dilution liquid chromatography-tandem mass spectrometry. J Chromatogr A 2014; 1348:97-104. [Jun].

Robledo C, Peck JD, Stoner JA, Carabin H, Cowan L, Koch HM, et al. Is bisphenol-A exposure during pregnancy associated with blood glucose levels or diagnosis of gestational diabetes? J Toxicol Environ Health A 2013;76(14):865-73. http://dx.doi.org/10. 1080/15287394.2013.824395.

Snijder CA, Heederik D, Pierik FH, Hofman A, Jaddoe VW, Koch HM, et al. Fetal growth and prenatal exposure to bisphenol A: the generation R study. Environ Health Perspect 2013;121(3):393-8. http://dx.doi.org/10.1289/ehp.1205296. [Mar Epub 2012 Dec21].

Tang R, Chen MJ, Ding GD, Chen XJ, Han XM, Zhou K, et al. Associations of prenatal exposure to phenols with birth outcomes. Environ Pollut 2013;178:115-20. http://dx.doi.org/10.1016/j.envpol.2013.03.023. [Jul Epub 2013 Apr 3].

Tefre de Renzy-Martin K, Frederiksen H, Christensen J, Boye Kyhl H, Andersson AM, Husby S, et al. Current exposure of 200 pregnant Danish women to phthalates, parabens and phenols. Reproduction 2014;147(4):443-53. [Nov 26 [Epub ahead of print]].

Toms LM, Allmyr M, Mueller JF, Adolfsson-Erici M, McLachlan M, Murby J, et al. Triclosan in individual human milk samples from Australia. Chemosphere 2011;85(11):1682-6. http://dx.doi.org/10.1016/jxhemosphere.2011.08.009. [Dec Epub 2011 Oct 13].

Wang H, Zhang J, Gao F, Yang Y, Duan H, Wu Y, et al. Simultaneous analysis of synthetic musks and triclosan in human breast milk by gas chromatography tandem mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci 2011;879(21):1861-9. http://dx.doi.org/10.1016/jjchromb.2011.04.036. [Jul 1 Epub 2011 May 7].

Witorsch RJ, Thomas JA. Personal care products and endocrine disruption: a critical review of the literature. Crit Rev Toxicol 2010;40(Suppl. 3):1-30. http://dx.doi.org/10. 3109/10408444.2010.515563. [Nov].

Wolff MS, Engel SM, Berkowitz GS, Ye X, Silva MJ, Zhu C, et al. Prenatal phenol and phthal-ate exposures and birth outcomes. Environ Health Perspect 2008;116(8):1092-7. http://dx.doi.org/10.1289/ehp.11007. [Aug].

Wolstenholme JT, Taylor JA, Shetty SR, Edwards M, Connelly JJ, Rissman EF. Gestational exposure to low dose bisphenol A alters social behavior in juvenile mice. PLoS One 2011;6(9):e25448. http://dx.doi.org/10.1371/journal.pone.0025448. [Epub 2011 Sep 28].

Woodruff TJ, Zota AR, Schwartz JM. Environmental chemicals in pregnant women in the United States: NHANES 2003-2004. Environ Health Perspect 2011;119(6):878-85. http://dx.doi.org/10.1289/ehp.1002727. [Jun Epub 2011 Jan 10].

Ye X, Bishop AM, Needham LL, Calafat AM. Automated on-line column-switching HPLC-MS/MS method with peak focusing for measuring parabens, triclosan, and other environmental phenols in human milk. Anal Chim Acta 2008;622(1-2):150-6. http:// dx.doi.org/10.1016/j.aca.2008.05.068. [Aug 1 Epub 2008 Jun 3].

Ye X, Zhou X, Furr J, Ahn KC, Hammock Bd, Gray EL, et al. Biomarkers of exposure to triclocarban in urine and serum. Toxicology 2011a;286(1-3):69-74. http://dx.doi. org/10.1016/j.tox.2011.05.008. [Aug 15 Epub 2011 May 23].

Ye X, Wong LY, Bishop AM, Calafat AM. Variability of urinary concentrations of bisphenol A in spot samples, first morning voids, and 24-hour collections. Environ Health Perspect 2011b Jul; 119(7):983-8. http://dx.doi.org/10.1289/ehp.1002701. Epub 2011 Mar 15.

Ye X, Zhou X, Hennings R, Kramer J, Calafat AM. Potential external contamination with bisphenol A and other ubiquitous organic environmental chemicals during biomonitoring analysis: an elusive laboratory challenge. Environ Health Perspect 2013; 121(3):283-6. http://dx.doi.org/10.1289/ehp.1206093. [Mar Epub 2013 Jan 15].

Yolton K, Xu Y, Strauss D, Altaye M, Calafat AM, Khoury J. Prenatal exposure to bisphenol A and phthalates and infant neurobehavior. Neurotoxicol Teratol 2011;33(5): 558-66. http://dx.doi.org/10.1016/j.ntt.2011.08.003. [Sep-Oc Epub 2011 Aug 10].

Zelner I, Hutson JR, Kapur BM, Feig DS, Koren G. False-positive meconium test results for fatty acid ethyl esters secondary to delayed sample collection. Alcohol Clin Exp Res 2012;36(9):1497-506. http://dx.doi.org/10.1111 /j.1530-0277.2012.01763.x. [Sep Epub 2012 Mar 20].

Zhou X, Ye X, Calafat AM. Automated on-line column-switching HPLC-MS/MS method for the quantification of triclocarban and its oxidative metabolites in human urine and serum. J Chromatogr B Anal Technol Biomed Life Sci 2012;881-882:27-33. http:// dx.doi.org/10.1016/j.jchromb.2011.11.024. [Jan 15 Epub 2011 Nov 26].

Zimmers SM, Browne EP, O'Keefe PW, Anderton DL, Kramer L, Reckhow DA, et al. Determination of free Bisphenol A (BPA) concentrations in breast milk of U.S. women using a sensitive LC/MS/MS method. Chemosphere 2014;104:237-43. http://dx.doi.org/10. 1016/j.chemosphere.2013.12.085. [Feb 4 pii: S0045-6535(14)00022-8].