Polybrominated diphenyl ethers in paired samples of maternal and umbilical cord blood plasma and associations with house dust in a Danish cohort

Page 1

International Journal of Hygiene and Environmental Health 213 (2010) 233–242

Contents lists available at ScienceDirect

International Journal of Hygiene andEnvironmental Health

journa l homepage: www.e lsev ier .de / i jheh

Polybrominated diphenyl ethers in paired samples of maternal and umbilicalcord blood plasma and associations with house dust in a Danish cohort

Marie Frederiksena,b,1, Cathrine Thomsenc, May Frøshaugc, Katrin Vorkampb,Marianne Thomsend, Georg Becherc,e, Lisbeth E. Knudsena,∗

a Department of Environment and Health, Institute of Public Health, University of Copenhagen, Oester Farimagsgade 5, 1014 Copenhagen K, Denmarkb Department of Environmental Chemistry and Microbiology, National Environmental Research Institute (NERI), Aarhus University, Frederiksborgvej 399, 4000 Roskilde, Denmarkc Department of Analytical Chemistry, Division of Environmental Medicine, Norwegian Institute of Public Health, P.O. Box 4404 Nydalen, 0403 Oslo, Norwayd Department of Policy Analysis, National Environmental Research Institute (NERI), Aarhus University, Frederiksborgvej 399, 4000 Roskilde, Denmarke Department of Chemistry, University of Oslo, P.O. Box 1033 Blindern, 0315 Oslo, Norway

a r t i c l e i n f o

Article history:Received 25 January 2010Received in revised form 21 April 2010Accepted 28 April 2010

Keywords:PBDEBrominated flame retardantHuman exposurePlacentaUmbilical cord bloodDust

a b s t r a c t

Brominated flame retardants (BFRs), in particular the polybrominated diphenyl ethers (PBDEs), havebeen used in consumer products for many years to increase fire resistance. Recently, developmentalneurotoxicity at very low levels has increased the concern about these compounds. The major objectivesof this study were to investigate the maternal and fetal exposure to PBDEs on the basis of maternal andumbilical cord plasma samples and to study the extent of placental transfer for different PBDE congeners.The findings were also compared with previously observed PBDE levels and patterns determined inplacental tissue from the same individuals, and the relationship with the external exposure from housedust from the participants’ homes was explored. Samples of maternal and umbilical cord plasma from acohort of 51 pregnant women from the Copenhagen area were collected. Paired maternal and umbilicalcord plasma were analysed for BDE-28, 37, 47, 85, 99, 100, 119, 138, 153, 154, 183, 209 and the brominatedbiphenyl BB-153 using automated SPE extraction and GC-HRMS for the tri- to hepta-BDEs and GC-LRMS(ECNI) for BDE-209. PBDEs were detected in all maternal and umbilical cord plasma samples. The sum oftri- to hexa-BDEs (�PBDE) in maternal plasma varied between 640 and 51,946 pg/g lipid weight (lw) witha median level of 1765 pg/g lw. In the umbilical cord samples �PBDE varied between 213 and 54,346 pg/glw with a median of 958 pg/g lw. The levels observed in fetal and maternal plasma were highly correlated,but the placental transport of PBDE congeners was found to decrease with increasing diphenyl etherbromination. Maternal concentrations were significantly correlated (p < 0.05) for most congeners withthe previously determined concentrations in placental tissue from the same individuals. Furthermore,positive correlations (p < 0.05) were found for BDE-28, 47, 100, 209 and �PBDE in maternal plasma and

house dust as well as for �PBDE in umbilical cord plasma and house dust. The positive correlations for

and uuma

I

ri2

rD

RD

1d

PBDEs for both maternalis a significant source of h

ntroduction

Brominated flame retardants (BFRs) are used to increase the fireesistance of a variety of materials and consumer products includ-ng plastics, upholstery, textiles, and electronic equipment (BSEF,010). Although flame retardants contribute to saving lives as they

∗ Corresponding author at: University of Copenhagen, Occupational and Envi-onmental Health, Oster Farimagsgade 5, room 5.2.12, DK 1014 K Copenhagen,enmark. Tel.: +45 35327653; fax: +45 35327686.

E-mail addresses: [email protected], [email protected] (L.E. Knudsen).1 Current address: Department of Construction and Health, Danish Buildingesearch Institute, Aalborg University, Dr. Neergaards Vej 15, 2970 Hoersholm,enmark.

438-4639/$ – see front matter © 2010 Elsevier GmbH. All rights reserved.oi:10.1016/j.ijheh.2010.04.008

mbilical cord plasma with house dust showed that domestic house dustn exposure to PBDEs in Denmark including in utero exposure.

© 2010 Elsevier GmbH. All rights reserved.

reduce the likelihood that products will catch fire and also slowthe spread of fire, many are persistent organic pollutants (POPs),being persistent, toxic, bioaccumulative and subject to long-rangetransport (de Wit et al., 2010). Concern about these properties hasresulted in a voluntary phase-out or ban of several groups of BFRs.Polybrominated biphenyls (PBBs) were banned in the US in themid-1970s after they had been accidentally added to livestock feedand thereby contaminated a great variety of food items (Fries andKimbrough, 1985). Production in Europe was discontinued in 2000

(Alaee et al., 2003). Two commercial mixtures of polybrominateddiphenyl ethers (PBDE), Penta- and Octa-BDE, have been bannedor phased out in Europe and North America. Furthermore, bothPenta- and Octa-BDE have been added to Annex A of the StockholmConvention, a global treaty that aims to protect human health and

Page 2

2 giene

tcetearaial2HtPedl2

cadprestishmaehaeha2ot

doFtcwo

M

S

(s5Cawpun

34 M. Frederiksen et al. / International Journal of Hy

he environment from POPs (The Stockholm Convention, 2010). Inontrast, the use of Deca-BDE is continued in many applications forxample in textiles, plastics and polymers for use in the transporta-ion and construction sectors, though it is restricted from use inlectronic equipment within the EU (BSEF, 2010). While the Penta-nd some of the Octa-BDE congeners biomagnify, and typicallyeach the highest levels in lipid rich tissues of marine predatorsnd mammals, high levels of Deca-BDE have primarily been foundn abiotic compartments such as sediments (de Wit, 2002; Law etl., 2006), while levels in biota, e.g. aquatic food webs, usually wereower than those of the less brominated congeners (Ross et al.,009). Consequently, humans may be exposed to PBDEs via the diet.owever, in addition to this traditional route of exposure for POPs,

he indoor environment may contribute significantly to the totalBDE exposure. High PBDE levels have been found in the indoornvironment for example in house dust where PBDEs have beenetected in very high concentrations, with European dust being

argely dominated by BDE-209 (Allen et al., 2008; Harrad et al.,008).

The concern for human PBDE exposure arises from toxicologi-al studies, which have shown that PBDEs are endocrine disruptorscting through several pathways (Legler, 2008) and exhibitedevelopmental neurotoxicity in lambs at low doses (<1 �g/kg) afterrenatal exposure (Abdelouahab et al., 2009). The prenatal stage isegarded as a very vulnerable phase for exposure to toxicants whereven very low levels of for example endocrine disruptors may causeevere irreversible damage (Welsh et al., 2008). Disturbance of thehyroid hormones during development can lead do lifelong deficitsn cognitive performance and behaviour (Darras, 2008). Recently,everal epidemiological studies on health effects of PBDE exposureave been published, linking PBDE exposure and changes in hor-one levels. Most studies have focused on thyroid hormones, but

lso effects on reproductive hormones have been observed (Meekert al., 2009; Turyk et al., 2009). Increased PBDE levels in breast milkave also been linked to the prevalence of cryptorchidism (Main etl., 2007), which is an indicator of hormonal imbalance (Schnackt al., 2008). Furthermore, in a longitudinal study, children withigher PBDE levels in cord blood scored lower in tests of mentalnd physical development at 1–4 and 6 years (Herbstman et al.,010). Most noteworthy is that these studies have been performedn the general population rather than focusing on groups subjectedo high exposure.

In the present study, concentrations of PBDEs and BB-153 wereetermined in pairs of maternal and umbilical cord blood. The mainbjective was to investigate maternal and fetal exposure to PBDEs.urthermore, different congeners were compared with regard toheir fetal/maternal concentrations and their associations with pla-ental tissue concentrations were established. Finally, the relationith house dust concentrations for the same cohort was studied, in

rder to elucidate its significance as a source of exposure to PBDEs.

aterials and methods

tudy subjects and sampling

The sampling procedure has previously been described in detailFrederiksen et al., 2009a; Pedersen et al., 2009). In brief, pairedamples of maternal and umbilical cord blood were collected from1 healthy pregnant women scheduled for caesarean section atopenhagen University Hospital (Rigshospitalet) between March

nd December 2007. Peripheral blood was collected from theomen when admitted for the caesarean section, and a blood sam-le was drawn from the umbilical cord immediately after birthsing heparinised tubes. From eight of the umbilical cords, it wasot possible to draw a sufficient volume of blood for PBDE analy-

and Environmental Health 213 (2010) 233–242

sis. The collected blood was processed (650 G, 10 min) and plasmawas stored at −20 ◦C in pre-cleaned glass containers until analysis.Information on general life style, dietary habits, etc. was collectedby self-administered questionnaires. PBDE concentrations werealso available for placental tissue, breast milk as well as indoor airand house dust from the participants’ homes (e.g. Frederiksen et al.,2009a; Vorkamp et al., in press). The PBDE levels in the domestichouse dust were determined in samples from vacuum cleaner bagswhich had been collected at two occasions three months apart, oneday prior to and approximately three months after blood sampling.Details on dust sampling and results have been described and dis-cussed by Vorkamp et al. (in press). For comparison with the plasmasamples the mean of the two sampling rounds was used, with thefollowing exceptions: From 11 participants only one dust samplewas available. For participants who moved between the samplings,the results from the first sampling were used (n = 3). All partici-pants had signed informed consent and the study was approved bythe Regional Ethics Committee of the Capital Region of Denmark(H-KF-01-327603) as well as the Danish Data Protection Agency.

Plasma analyses

ChemicalsAll PBDE standards solutions (both native and labelled) as well

as BB-153 were obtained from Wellington Laboratories (Guelph,Ontario, Canada), CIL (Andover, MA, USA) or AccuStandard (NewHaven, CT, USA). All solvents used were of Pestiscan quality andobtained from Lab-Scan (Gliwice, Poland). Sulphuric acid, formicacid, silica and sodium sulphate were from Merck (Darmstadt, Ger-many). Prior to use the silica was pre-cleaned with methanol anddichloromethane and activated at 130 ◦C, sodium sulphate washeated overnight at 600 ◦C.

GlasswareAll glassware was washed in Extran MA01 (Merck), rinsed with

distilled water and subsequently heated at 450 ◦C for 4 h (volumet-ric equipment was not heated).

Sample preparationThe method for extraction of PBDEs from plasma was similar

to the previously described method for human serum (Thomsen etal., 2007) but with minor modifications. Paired samples of maternaland umbilical cord plasma were prepared on the same day. Briefly,the plasma was thawed at room temperature, sonicated and cen-trifuged (2000 rpm, 3 min) to remove precipitates formed duringfreezing. Between 2 and 5 g of plasma were accurately weighed,spiked with 30 �l internal standards solution (75 pg of 13C-labelledBDE-28, 47, 99, 100, 153, 154, 183 and 750 pg of 13C-labelled BDE-209), sonicated for 5 min and left to equilibrate for at least one hourunder dark conditions. Subsequently, the proteins were denatu-rated using 5 ml formic acid:2-propanol (4:1, v/v), whirli-mixed,sonicated (5 min) and kept at room temperature for one hour beforeovernight storage at 5 ◦C. Prior to solid phase extraction (SPE), thesamples were diluted with 5 ml 10% 2-propanol in distilled waterand sonicated (5 min). Automated SPE was performed using anASPEC XL4 (Gilson Inc., Middleton, MI, USA) and Oasis HLB (WatersCorp., Milford, MA, USA) (540 mg/3 ml) columns as described forserum samples (Thomsen et al., 2007). The extracts were evapo-rated to a few microliters on a TurboVap LV (Zymark, Hopkinton,MA, USA) and redissolved in 3.5 ml n-heptane:dichloromethane(3:1, v/v). For clean-up empty SPE-cartridges were packed from

bottom to top with 0.7 g sulphuric acid impregnated silica (1:3,w/v), 0.2 g silica, and 0.5 g sodium sulphate. The columns werewashed, conditioned and eluted as described for serum samples.Prior to analysis, the extracts were evaporated to approximately30 �l using a TurboVap LV and a gentle stream of nitrogen.

Page 3

giene

ec(

I

aAp2gcpittBahptm11rssMv

p1Bbcnsof

V

siociaiwt

Q

cwmUAt5o1

to

The internal PBDE exposure data in human populations usually

M. Frederiksen et al. / International Journal of Hy

Cholesterols, phospholipids, and triglycerides were determinednzymatically at Copenhagen University Hospital and the total lipidontent was calculated by the method described by Grimvall et al.1997).

nstrumentation, analysis and quantificationThe tri- to hepta-BDEs were determined by GC-HRMS using

n HP 6890 GC (Agilent, Avondale, PA, USA) coupled to anutoSpec Premier P677 (Waters, Milford, MA, USA). Two �l of thelasma extract was injected in splitless mode (injector temperature90 ◦C). Helium (99.9999%, Aga, Norway) was used as the carrieras, and the flow was kept constant at 1.0 ml/min. The 25 m DB-5MSolumn and GC temperature program used have been describedreviously (Thomsen et al., 2007). The MS was operated in electron

onisation mode with electron energy of 37 eV and an ion sourceemperature of 265 ◦C. The monitored ions were 405.8026 m/z forri-BDEs, 483.7131 m/z for tetra-BDEs, 563.6215 m/z for penta-DEs, 483.7131 m/z for hexa-BDEs, 563.6215 m/z for hepta-BDEsnd 467.7182 m/z for BB-153. In addition, qualifier ions of 2 amuigher than the target ion were monitored for confirmation pur-oses. Deviations of ±20% of the isotope ratio in samples comparedo standards were accepted. Data were collected in selected ion

onitoring mode at a resolution of 5000. BDE-28, 47, 99, 100, 153,54 and 183 were quantified via isotope dilution, while BDE-37, 85,19 and 138 were quantified using 13C-BDE-28, 99, 100 and 153,espectively. BB-153 was quantified using 13C-BDE-154 as internaltandard. The GC-HRMS detection method was validated in thistudy (see below). BDE-209 was determined by GC–low resolutionS (HP6890/5973, Agilent) as has been previously described and

alidated by Thomsen et al. (2007).For a 5 g plasma sample, the LOQ was approximately 0.06 pg/g

lasma for BDE-28, 0.1 pg/g plasma for BDE-47, 153, 154 and BB-53, 0.6 pg/g plasma for BDE-99 and 100, and 6 pg/g plasma forDE-209. The higher relative LOQ of the penta-BDEs was causedy increased noise in the penta-window compared to the otherongeners. Compounds below the LOQ or compounds that wereot detected have been set to zero for the calculations of means andums giving a lower bound measure of exposure. The applicationf ½LOQ was tested but did not yield more informative results ofor example FM-ratios.

alidation of the GC-HRMS methodThe GC-HRMS detection method was validated by analysing

tandard solutions in n-nonane at nine concentration levels cover-ng 0.05–40 pg/�l (n = 6, 6, 1, 1, 6, 1, 1, 1, 6 injections) and samplesf breast milk (n = 3) and cod liver oil (n = 3) from interlaboratoryomparison studies (ILS) (Haug and Becher, 2005, 2006). Linear-ty of the calibration curve, limits of detection (LODs), repeatabilitynd accuracy were assessed. A DB-5MS column (60 m × 0.25 mm.d. × 0.25 �m film thickness, Agilent) was used in the validation

ith the following temperature program: 90 ◦C for 1 min, 20 ◦C/mino 200 ◦C, 5 ◦C/min to 300 ◦C and 10 ◦C/min to 325 ◦C (3 min).

uality control and blank correctionFor approximately every eighth plasma sample one in-house

ontrol sample, one procedural blank and one horse serum blankere analysed. The in-house control sample was American humanale EDTA plasma (Lampire Biological Laboratories, Pipersville, PA,SA). As the levels of PBDEs are generally 10 times higher in Northmerica compared to Europe (Frederiksen et al., 2009b) only 2 g of

he control sample were used. The procedural blank consisted of

ml purified water, which was treated as a real sample through-ut the pre-treatment and extraction procedures. The recovery of3C-BDE-209 in the procedural blanks were often quite low dueo the extreme hydrophobicity of BDE-209 possibly leading to lossf internal standard by adsorption to surfaces in the early steps

and Environmental Health 213 (2010) 233–242 235

of the procedure. Therefore horse serum (H-1270, Sigma–Aldrich)was used to determine blank levels for BDE-209, assuming thatthe matrix of the horse serum would increase the recovery of13C-BDE-209 and thereby give a more accurate picture of theactual blank levels. Since the horse serum contained trace levelsof BDE-47, both procedural blanks and horse serum blanks werenecessary. Three serum samples from the AMAP ring test for persis-tent organic pollutants in human serum (2008-3) were analysed asexternal control samples together with the plasma samples of thisstudy.

The levels observed in the blank samples were used to cor-rect the real samples for procedural contamination. Approximatelyhalf way through the extraction of samples a contamination ofthe denaturation agent became evident. The contamination con-sisted primarily of BDE-209, which was approximately 10 timesthe normal blank level, but trace levels of BDE-47, 99, 100, 153,and 154 were also observed. Therefore, during data processingthe samples were divided into two groups for blank correctioncorresponding to the two contamination situations. For BDE-47,153 and 154 (plus 99 and 100 in the first group), the mean back-ground levels calculated from the two groups of blanks weresubtracted from the sample concentrations according to the respec-tive groups of analyses. Concerning BDE-209, the horse serumdid not increase the recovery of 13C-BDE-209 significantly, thusprocedural blanks as well as horse serum blanks and fetal sam-ples below the average procedural blank were used to estimatethe blank level of BDE-209. For the first group, the BDE-209results could not be used as the contamination of blank sam-ples by far exceeded the levels in the samples. In the secondgroup BDE-209 blank levels were established by the above pro-cedure and subtracted. To avoid false positivies, only samplesabove the highest blank were quantified, though one should beaware of the risk of a biased dataset if many samples are belowLOQ.

Assessment of placental transport

In the assessment of the placental transport, the terminol-ogy and approaches from human ex vivo placenta perfusions asdescribed by Mose et al. (2007) and Mathiesen et al. (2010) havebeen adopted. This includes fetal–maternal ratios (FM-ratio), whichare calculated as ratios between the concentration in the fetal andmaternal compartments, in this case umbilical cord plasma andmaternal plasma, and describe the extent of the placental transferat equilibrium (Frederiksen et al., in press). Similar use of cord-maternal ratios has been described by Meijer et al. (2008) for invivo placental transport of PBDEs. Because of the great differencesin lipid content between maternal and umbilical cord plasma andto facilitate comparison with both ex vivo and in vivo studies,FM-ratios based on both wet weight and lipid weight were cal-culated. FM-ratios were calculated for the individual congeners inthose pairs which had levels above LOQ in both the maternal andfetal samples. Due to the low detection frequency in umbilical cordplasma, FM-ratios could not be calculated for BDE-99, 100, 154, and209.

Statistical methods

exhibit a skewed distribution; therefore the nonparametric Spear-man rank correlation was used for all correlation analyses. For allstatistical analyses, only samples above LOQ were included. Thecorrelations were calculated using GraphPad Prism 5.0 (GraphPadSoftware Inc., La Jolla, CA, USA).

Page 4

236 M. Frederiksen et al. / International Journal of Hygiene and Environmental Health 213 (2010) 233–242

Table 1Results from the validation of the GC-HRMS method. The concentrations are in pg/�l for standard solutions and in pg/g wet weight for the samples.

LODa r2b 6 injections of standard solution Breast milk from ILS Cod liver oil from ILS

0.05 pg/�l 40 pg/�l 3 samples ILS consensus 3 samples ILS consensus

Mean RSD Mean RSD Mean RSD Consensus RSD %c Mean RSD Consensus RSD %c

BDE-28 0.01 0.999 0.053 9.7 38 2.6 2.6 0.61 3.3 34 −20 576 0.53 580 24 −0.66BDE-37 0.01 0.997 0.062 16 38 5.0 130 2.6BDE-47 0.02 0.998 0.058 38 38 3.8 33 3.9 39 20 −15 10619 1.0 10416 16 2.0BDE-100 0.02 0.999 0.053 28 37 3.6 7.3 6.0 10 23 −27 1597 1.2 1570 27 1.7BDE-119 0.03 0.998 0.055 25 37 3.9BDE-99 0.03 0.999 0.065 16 35 3.1 8.4 3.3 8.6 28 −2.7 301 6.5 301 37 0.02BDE-85 0.04 0.998 0.058 27 40 4.3BDE-154 0.02 0.999 0.055 19 38 3.2 1.0 1.9 1.1 28 −11 820 0.73 725 30 13BB-153 0.03 0.998 0.053 37 36 3.6 3.6 6.2 29 2.3BDE-153 0.02 0.998 0.053 19 39 3.5 15 2.3 16 18 −6.7 45.0 4.8 42 28 7.0BDE-138 0.03 0.998 0.052 15 41 4.4BDE-183 0.04 0.996 0.050 28 40 5.8 ND 1.1 41 ND 6.6 42

a LOD in pg/�l extrapolated from the calibration curve.

NI

R

V

GfL02BlTstfra

tftAwsA

P

wstepsfwtBftom

b r2 is the correlation coefficient from the calibration curve.c Deviation from the consensus value.D: Not detected.

LS: Interlaboratory comparison studies.

esults

alidation of the GC-HRMS method and quality control

A summary of the validation experiments is given in Table 1. TheC-HRMS method was proven to be linear through 0.05–40 pg/�l

or all PBDEs, with correlation coefficients (r2) 0.996–0.999. TheODs extrapolated from the calibration curve ranged from 0.01 to.04 pg/�l. A repeatability of standards in the range 9.7–38% and.6–5.8% RSD was demonstrated at 0.05 and 40 pg/�l, respectively.DE-28, 47, 99, 100, 153, and 154 were measured in the inter-

aboratory comparison samples of breast milk and cod liver oil.he quantified concentrations were within ±1 SD of the consen-us value corresponding to a deviation of 2.7–27% and 0.02–13% forhe breast milk and cod liver oil, respectively. This illustrates satis-actory accuracy in the analysis of biological samples. The RSDs ofesults for the three replicate samples were in the range of 0.6–6.0%nd 0.5–6.5% for breast milk and cod liver oil, respectively.

The in-house reference material human EDTA plasma was usedo monitor the stability of the method. RSD below 10% was observedor all congeners except for BDE-99 and 154, which were affected byhe contamination mentioned above. The PBDE levels in the threeMAP samples analysed along with the current plasma samplesere quantified within ±20% of the assigned values, which has been

et as that level which is indicative for excellent performance byMAP (2008).

BDE levels and profile

A total of 51 maternal and 43 umbilical cord plasma samplesere analysed for PBDEs, and of these 17 maternal and 12 fetal

amples were successfully analysed for BDE-209. The recovery ofhe internal standards in three umbilical cord samples was too lownable quantification. Although, PBDEs were detected in all sam-les, congeners BDE-37, 85, 119, and 183 were not detected in anyample, while BDE-138 was detected in only one sample and there-ore is not reported. Concentrations of the detected congeners asell as �PBDE, which here and in the following text comprises up

o six tri- to hexa-BDEs, are given in Table 2. The detection limits of

DE-99 and 100 were higher than those for the other PBDEs, and

or a large proportion of the samples, BDE-99 and 100 were belowheir LOQs. The distribution of �PBDE was highly skewed with 5%f the samples having �PBDE levels more than 10 and 19 times theedian level of maternal and umbilical cord plasma, respectively

(range maternal plasma 640–51,946 pg/g lw, range umbilical cordplasma 213–54,346 pg/g lw). As expected, the lipid content of thetwo types of plasma was very different, with a mean lipid con-tent of 8.91 g/l in maternal plasma and 2.88 g/l in umbilical cordplasma. Generally, concentrations on a wet weight basis were muchlower in umbilical cord plasma than in maternal plasma, whilstthe lipid-normalised levels were more similar. Therefore PBDE lev-els are given on both lipid and wet weight basis in Table 2. In themajority of both maternal and fetal samples, the dominating con-gener among the tri- to hexa-BDEs was BDE-153 (84% and 70% ofsamples, respectively). The remaining samples were dominated byBDE-47. In those maternal samples where the BDE-209 analysis wassuccessful, the concentrations of BDE-209 were similar to that of�PBDE and on average contributed 50% to the total PBDE burden(range: 19–86%). In the umbilical cord plasma samples, BDE-209was detected in all samples, but the concentrations were belowor close to those found in the blanks in spite of low blank levelsof approximately 60 pg/sample. The brominated biphenyl BB-153was detected in the majority of both maternal and umbilical cordsamples at concentrations higher than BDE-154. In one sample,high levels of all congeners, in particular BDE-153, were detectedin both maternal and umbilical cord plasma. The congener patternalso deviated somewhat from the general picture, as the level ofBDE-100 was higher than that of BDE-99 and similar to BDE-47.The same pattern was observed previously in the placental tissueof the same individual (Frederiksen et al., 2009a).

Statistical analysis did not reveal any correlation between PBDElevels and maternal age, number of children or previous breast-feeding history.

Placental transfer

A high degree of inter-correlation of PBDEs in maternal plasmaand umbilical cord plasma was observed (data not shown). For indi-vidual congeners highly significant correlations (Spearman rank)were observed between the concentrations of BDE-28, 47, 153 andBB-153 in maternal and umbilical cord plasma (Fig. 1). For theremaining congeners, the detection frequency was too low to allowstatistical treatment. The extent of the placental transfer was esti-

mated for the individual congeners based on FM-ratios (Table 3).The higher the FM-ratio, the higher the relative amount trans-ported across the placenta. Differences between congeners wereobserved in terms of a decreasing transport with increasing degreeof bromination (Table 3), independent of lipid adjustment. An FM-

Page 5

M. Frederiksen et al. / International Journal of Hygiene and Environmental Health 213 (2010) 233–242 237

Table 2Median, mean and range of PBDE levels in maternal and umbilical cord plasma on lipid weight basis (pg/g lw); wet weight basis results [pg/g ww] are given in square brackets.Total lipid content based on cholesterol, triglycerides and phospholipids.

Maternal plasma (n = 51) Umbilical cord plasma (n = 40)

Frequency > LOQ Median/mean (range) Frequency > LOQ Median/mean (range)

BDE-28 88% 32.9/57.1 (<6.09–398)[0.320/0.496 (<0.059–3.47)]

73% 50.7/55.7 (<17.5–271)[0.137/0.165 (<0.052–0.911)]

BDE-47 80% 381/859 (<11.0–7883)[3.49/7.68 (<0.109–73.1)]

45% <67.9/694 (<21.4–5262)[<0.194/2.10 (<0.083–17.7)]

BDE-99 37% <105/552 (<53.4–18554)[<0.935/5.34 (<0.557–183)]

28% <290/694 (<128–7022)[<0.804/2.03 (<0.499–21.9)]

BDE-100 27% <104/290 (<53.4–6218)[<0.935/2.66 (<0.538–54.4)]

5% <271/102 (<128–2733)[<0.781/0.340 (<0.499–9.18)]

BDE-153 98% 1126/1916 (<13–35962)[9.80/16.8 (<0.134–315)]

100% 507/808 (202–9472)[1.46/2.42 (0.592–31.8)]

BDE-154 45% <18.2/68.5 (<8.91–2010)[<0.158/0.660 (0.093–19.8)]

20% <46.2/56.2 (<21.4 651)[<0.131/0.163 (<0.083–1.74)]

BB-153 90% 181/209 (<14.6–848)[1.59/1.90 (<0.116–8.27)]

50% 68.6/75.3 (<29.2–351)[0.205/0.220 (<0.083–0.902)]

BDE-209a 94% 1709/1805 (<664–3849)[14.2/15.3 (<5.84–31.7)]

0% <2413b

[<6.76]

�tri–hexaBDE 100% 1765/3743 (640–51946)[16.7/33.7 (5.27–454)]

100% 958/3733 (213–54346)[3.06/11.5 (0.592–174)]

Total lipids (g/l) 8.75/8.91 (5.68–12.35) 2.92/2.88 (1.97–3.95)

Hereof Cholesterol (g/l) 4.53/4.48 (2.51–6.52) 1.32/1.34 (0.84–2.11)Triglycerides (g/l) 1.43/1.59 (0.82–4.29) 0.23/0.23 (0.14–0.38)Phospholipids (g/l) 2.88/2.84 (1.81–3.88) 1.29/1.29 (0.86–1.71)

a The number of samples for BDE-209 determinations were n = 17 and n = 12 for maternal and cord plasma, respectively.b Median LOQ depending on samples amount and lipid content.

Fig. 1. Scatter plot of BDE-28, 47, 153 and BB-153 in maternal and fetal (umbilical cord) plasma pairs (pg/g lw). The p and r values of the Spearman rank correlation are given.Two samples have been marked in grey; their FM-ratios deviated noticeably for most congeners from the other samples. *In the plot of BDE-153 one datapoint was outsideaxis limits due to extremely high levels of BDE-153 in both maternal and fetal plasma (35,961 ng/g lw and 9472 ng/g lw, respectively).

Page 6

238 M. Frederiksen et al. / International Journal of Hygiene and Environmental Health 213 (2010) 233–242

Table 3Fetal–maternal ratios (FM-ratios) of some PBDE congeners and BB-153 as indicatorsfor the extent of placental transfer, median and (25th; 75th percentile).

Congener npairs FM-ratio (ww) FM-ratio (lw) log Kowa,b

BDE-28 28 0.45 (0.30; 0.58) 1.3 (1.0; 2.1) 5.9BDE-47 18 0.30 (0.26; 0.47) 1.0 (0.80; 1.25) 6.8BB-153 20 0.18 (0.14; 0.20) 0.57 (0.47; 0.62) 6.4

rFcb

C

ctadsaewIrpaaf(

C

earvteanhhpispBdl(

D

P

Daeu

corr

elat

ion

coef

fici

ents

ofth

eco

rrel

atio

ns

ofPB

DEs

inp

lace

nta

ltis

sue

and

pla

sma

(pg/

glw

;an

dal

sop

g/g

ww

for

BD

E-20

9).S

tati

stic

ally

sign

ifica

nt

(p<

0.05

)co

rrel

atio

ns

are

inbo

ld.�

pen

tain

clu

des

BD

E-17

tou

des

BD

E-15

3to

183,

thu

sB

DE-

153

and

154

are

incl

ud

edin

both

pen

taan

doc

taas

fou

nd

inco

mm

erci

alm

ixtu

res

byLa

Gu

ard

iaet

al.(

2006

).

Plac

enta

ltis

sue

(n)

BD

E-28

(11)

BD

E-49

(20)

BD

E-47

(50)

BD

E-10

0(3

6)B

DE-

99(4

9)B

DE-

154a

(44)

BD

E-15

3(5

0)B

DE-

209

(lw

)(4

4)B

DE-

209

(ww

)(4

4)�

pen

ta(5

0)�

octa

(50)

a(4

5)−0

.07

0.21

0.61

0.49

0.39

0.26

0.29

−0.1

0−0

.08

0.53

0.28

(41)

0.18

−0.0

60.

740.

640.

570.

300.

150.

070.

120.

640.

15(1

4)−0

.70

−0.9

90.

680.

470.

550.

600.

550.

240.

030.

720.

65(1

9)0.

40−0

.18

0.58

0.15

0.53

0.15

0.00

0.44

0.43

0.57

0.17

(23)

0.37

−0.2

40.

590.

190.

470.

220.

30−0

.01

0.08

0.61

0.42

(48)

0.42

−0.1

10.

050.

04−0

.01

0.66

0.11

−0.1

4−0

.07

0.18

0.25

(50)

−0.2

2−0

.16

0.31

0.30

0.17

0.13

0.77

−0.0

8−0

.06

0.38

0.58

(16)

−0.2

30.

32−0

.23

0.51

−0.6

10.

220.

28−0

.39

−0.4

6)

(16)

−0.2

00.

18−0

.22

0.61

−0.5

40.

200.

27−0

.41

−0.3

6(5

0)0.

020.

530.

460.

320.

160.

50−0

.09

−0.0

60.

500.

37

plas

ma (29)

0.39

0.64

0.24

0.54

0.12

−0.0

20.

000.

510.

17(1

8)−0

.11

−0.0

60.

570.

490.

340.

300.

22−0

.09

−0.0

90.

29−0

.05

(11)

0.26

−0.0

90.

410.

500.

360.

100.

04−0

.27

−0.3

0−0

.28

0.09

(8)

−0.2

00.

43−0

.32

−0.1

8−0

.38

−0.1

70.

080.

11−0

.25

−0.1

90.

29(2

0)−0

.50

0.80

−0.5

0−0

.21

−0.6

90.

100.

26−0

.31

−0.4

8−0

.18

0.23

(40)

0.14

−0.0

2−0

.31

−0.4

4−0

.35

0.49

0.12

−0.3

9−0

.39

0.19

0.36

(40)

−0.0

50.

050.

170.

04−0

.11

0.06

0.55

−0.2

6−0

.19

0.43

0.21

elu

ted

wit

hB

B-1

53in

anal

ysis

ofp

lace

nta

sam

ple

s.

BDE-153 39 0.15 (0.12: 0.18) 0.44 (0.38; 0.54) 7.9

a Braekevelt et al. (2003).b UNEP (2006).

atio >1 was calculated for BDE-28. However, a high variation of allM-ratios was observed, and this was most pronounced for thoseongeners with a low detection frequency and levels close to LOQs,ut the actual ratio seemed independent of concentration.

orrelation with placental tissue

The concentration in maternal plasma was found to be highlyorrelated with concentrations previously observed in the placen-al tissue (Frederiksen et al., 2009a), with the exception of BDE-28nd 209 (Table 4). Significant correlations were also found betweenifferent penta-BDE congeners in placenta and plasma, which pos-ibly illustrates inter-correlation of the congeners or other factorsffecting the congener composition in specific compartments. Forxample, BDE-47 in placenta was significantly correlated (p < 0.05)ith BDE-28, 47, 99, 100, 153, and 154 in maternal plasma (Table 4).

n contrast, BDE-209 in plasma was found to be inversely cor-elated with BDE-153 in placenta. PBDE levels in umbilical cordlasma and placental tissue were also found to be correlated,nd significant correlations (p < 0.05) were observed for BDE-153nd �PBDE. A significant negative correlation was also observedor BDE-209 in placenta with BDE-153 in umbilical cord plasmaTable 4).

orrelation with house dust

In Table S1 of the Supplementary material, the mean PBDE lev-ls in house dust used for the comparison with the plasma levelsre presented. The individual concentrations in the two samplingounds were significantly correlated, though differences in absolutealues were observed. Further details on the PBDE dust concen-rations have been reported and discussed elsewhere (Vorkampt al., in press). Fig. 2 shows the scatter plot of BDE-28, 47, 100nd �PBDE in dust and maternal plasma. There were also sig-ificant correlations between the concentrations of BDE-209 inouse dust and maternal plasma (Table S2). In contrast, BDE-153 inouse dust was not significantly correlated to BDE-153 in maternallasma, although it was significantly correlated to BDE-28 and 47

n maternal plasma (Table S2). Correlation was observed betweeneveral penta-BDE congeners in plasma (not shown) and betweenlasma and house dust (Table S2), as was the case for several penta-DE congeners in house dust alone (Vorkamp et al., in press). Theust concentrations of BDE-28 and �PBDE could be directly corre-

ated with the respective concentrations in umbilical cord plasmap = 0.0003/r = 0.57 and p = 0.0004/r = 0.53, respectively).

iscussion

BDE levels and profile

The concentrations of PBDEs found in maternal plasma fromenmark are in the same range as previously reported for serumnd plasma of adults from Europe (Fromme et al., 2009; Guveniust al., 2003; Meijer et al., 2008; Thomsen et al., 2007). The levels inmbilical cord plasma are also consistent with previously reported Ta

ble

4Sp

earm

anra

nk

154;

�oc

tain

cl

Mat

erna

lpla

smB

DE-

28B

DE-

47B

DE-

100

BD

E-99

BD

E-15

4B

B-1

53B

DE-

153

BD

E-20

9(l

w)

BD

E-20

9(w

w�

pen

ta

Um

bilic

alco

rdB

DE-

28B

DE-

47B

DE-

99B

DE-

154

BB

-153

BD

E-15

3�

pen

ta

aB

DE-

154

co

Page 7

M. Frederiksen et al. / International Journal of Hygiene and Environmental Health 213 (2010) 233–242 239

F , and Pc

lai(te

1hwtegt4FsIihisa2mitotiB

ig. 2. Scatter plot of PBDEs in house dust and maternal plasma for BDE-28, 47, 100orrelations are given. *In dust samples BDE-154 co-eluted with BB-153.

evels of PBDEs in umbilical cord serum from Europe (Antignac etl., 2009; Guvenius et al., 2003; Meijer et al., 2008), but approx-mately one order of magnitude lower than in the United StatesHerbstman et al., 2007; Mazdai et al., 2003). This is in line withhe general picture of the geographical differences of human PBDExposure.

In the present, as well as other recent European studies, BDE-53 was found to be the most abundant congener in the majority ofuman blood samples (Roosens et al., 2009; Thomas et al., 2006),hile in US and earlier European studies BDE-47 has been observed

o be the dominating congener (Guvenius et al., 2003; Herbstmant al., 2007). The reasons for these shifts and differences in the con-ener patterns could be due to a longer half-life of BDE-153 relativeo BDE-47 (Geyer et al., 2004) resulting in a decreased intake of BDE-7 from the food chain and increased excretion relative to BDE-153.urthermore there have been differences between regions in expo-ure and usage patterns of the technical mixtures (de Wit, 2002).nterestingly, while the levels of �PBDE and BDE-47 have decreasedn Swedish breast milk since the late 1990s, the levels of BDE-153ave been found to increase (Fangstrom et al., 2008). Differences

n congener patterns can also be observed between tissues of theame subjects; e.g. in placental tissue from the same cohort themount of BDE-47 and 153 was almost equal (Frederiksen et al.,009a), while BDE-153 had a higher contribution than BDE-47 inost plasma samples. Huwe et al. (2008) found that after oral dos-

ng of rats the hepta- to deca-BDE were not distributed equally in

he body lipids resulting in different congener patterns dependingn the matrix, which has implications for comparison of biomoni-oring data across matrices. This is in agreement with other studiesndicating the most pronounced biomagnfication for tetra- to hexa-DEs, while the higher brominated PBDEs (hepta- to deca-BDEs)

entaBDE (BDE-28, 47, 99, 100, 153, 154*). The p and r values of the Spearman rank

do not bioaccumulate to the same extent (Yogui and Sericano,2009).

Placental transfer of PBDEs

In accordance with other studies, we also observed strong cor-relations between maternal and fetal levels for all congeners forwhich there was a reasonable number of detects (Antignac etal., 2009; Kawashiro et al., 2008; Meijer et al., 2008). This indi-cates direct transfer of PentaBDE congeners and BB-153 across theplacenta. In contrast, BDE-209 was not detected in quantifiableamount in the umbilical cord plasma even given the low blank lev-els, which could be caused by a very restricted transport across theplacenta (Frederiksen et al., in press). However, other studies havedetected BDE-209 in cord blood with LOQs similar to those of thisstudy indicating that to some extent transport of BDE-209 doestake place (Antignac et al., 2009; Gomara et al., 2007). Antignacet al. (2009) reported a median level of fetal exposure to BDE-209that was higher than the LOQ of the present study, thus BDE-209 isprobably present but below the LOQ.

The difference in PBDE levels in maternal and umbilical cordplasma became more pronounced with increasing bromination asillustrated by the decreasing FM-ratios (Table 3). The median FM-ratios are associated with variation which is partially due to highermeasurement uncertainty at levels close to the limits of detection.Therefore the absolute values of the FM-ratios should be inter-

preted with some caution. In Fig. 1 two individuals are markedwith grey as for most congeners their FM-ratios deviated notice-ably from the other individuals, and they had marked influence onthe observed variation. Nevertheless, a clear trend of decreasingFM-ratio with increasing degree of bromination could be observed.

Page 8

2 giene

rrptaLmto

t2tvalwpcmombdcTsaiho

aFrieBtBcfortme

C

ssappdwlltpstw

40 M. Frederiksen et al. / International Journal of Hy

Meijer et al. (2008) describe a similar decrease in cord-maternalatio with increasing bromination. Guvenius et al. (2003) alsoeport a decrease of higher brominated congeners in umbilical cordlasma, though without quantifying the ratio. Decreasing placen-al transfer of PBDEs with increasing degree of bromination haslso been observed in marine mammals (Kajiwara et al., 2008;aw et al., 2002) as well as in a human ex vivo placenta perfusionodel (Frederiksen et al., in press). For PCBs decreasing placental

ransport with increasing molecular weight has previously beenbserved by Covaci et al. (2002).

Log Kow is an indicator for lipid solubility and increases withhe degree of bromination (Table 3) (Braekevelt et al., 2003; UNEP,006). In addition, the molecular weight and size also increase withhe degree of bromination. Since all congeners have high log Kow

alues they are all highly lipid soluble and assumed to be 100%ssociated with lipids. The observed FM-ratios were highest for theess brominated congeners BDE-28 and BDE-47 and close to one

hen lipid adjusted, thus indicating that lipid dependent trans-ort across the placental barrier occurs for the lower brominatedompounds. An FM-ratio (lw) above one for BDE-28 would nor-ally indicate accumulation in the fetus; however, the uncertainty

f the absolute number is large and enhanced by the lipid adjust-ent, probably overestimating the relevant ratio. For the higher

rominated BDE-153 and BB-153 congeners the FM-ratios (lw)ecrease, which means that the differences in lipid content aloneannot explain the differences observed in absolute concentration.he restricted transport could possibly be caused by an increasingteric hindrance to the transport with the degree of brominationnd thereby also size (Guvenius et al., 2003). Other factors such asnvolvement of transport proteins controlling the transport of theigh molecular weight congeners (Huwe et al., 2008) could also bef importance.

For comparison with ex vivo placenta perfusion studies usingrtificial medium without lipids but containing proteins, etc. theM-ratios based on wet weight have been calculated. The kineticsegarding the transport of BDE-47, 99 and 209 have been studiedn an ex vivo placenta perfusion system and have been reportedlsewhere (Frederiksen et al., in press). Here, it was found thatDE-47 was transported faster than BDE-99 and at equilibrium,he FM-ratio (ww) was higher for BDE-47 (FM47 = 0.47) than forDE-99 (FM99 = 0.25). Although ex vivo placenta perfusions is ofourse a model of the in vivo conditions, the observed FM-ratioor BDE-47 was similar to that of the present study. No transportf BDE-209 was observed in the perfusion model supporting theesults obtained here and in agreement with the generally observedransport barrier properties of membranes for compounds with a

olecular weight above 600 Da, i.e. the hexa- to decaBDEs (Hewittt al., 2007; Myren et al., 2007).

orrelation with placental tissue

The PBDE concentrations on a lipid weight basis seem to belightly higher in maternal plasma than in placental tissue from theame participants (median �PBDE = 1220 pg/g lw) (Frederiksen etl., 2009a), which again is slightly higher than in the umbilical cordlasma. A correlation of PBDE concentrations between maternallasma and placenta was expected as the placenta is embed-ed in the maternal uterus and the maternal blood exchangesith the placental tissue. However, a previous study found that

evels in placental tissue were not correlated with the plasmaevels (Gomara et al., 2007). The shifting of congener composi-

ion and the inverse correlation between BDE-209 in maternallasma and BDE-153 in placenta illustrates the differences in per-istence and metabolic pathways between congeners, and howhey are distributed in the body. In general, the BDE-209 levelas significantly higher than the BDE-153 level in placental tissue

and Environmental Health 213 (2010) 233–242

(Frederiksen et al., 2009a), which may be explained by increasedretention in placental tissue as a result of the limited placentaltransport.

Correlation with house dust

The relative importance of dust and diet as sources for humanexposure is still unclear, and seems to vary considerably betweencountries (Frederiksen et al., 2009b). In North America, dust seemsto be a very significant if not the main source for human expo-sure (Lorber, 2008), and significant correlations between breastmilk concentrations and house dust have been found (Wu et al.,2007). However, a recent large scale study of diet and PBDE expo-sure in the US population revealed an association between intakeof poultry and red meat and PBDE levels in serum (Fraser et al.,2009). Due to much lower levels in Europe and Asia, the relativeimportance of the exposure pathways is still being discussed. Instudies using duplicate diets it has been estimated that the dietcontributes up to 97% and 96% of the PBDE exposure in the Ger-man and Belgian population, respectively, and that the contributionfrom dust is minor (Fromme et al., 2009; Roosens et al., 2009).There are several studies describing the linkage between externaland internal exposure of high exposure groups such as workers orsports fishermen (Sjodin et al., 1999; Thomsen et al., 2008). How-ever, for the general European population only a few investigationshave been able to show a significant correlation between exter-nal and internal exposure. Roosens et al. (2009) studied both thediet and dust contribution for 19 students in Belgium. However,no correlation was found for either exposure pathway. Karlssonet al. (2007) found a significant correlation between house dustand serum concentrations in five subjects from Sweden, but thiscorrelation was strongly driven by a single subject. In the presentstudy we found significant correlations between house dust andmaternal plasma concentrations for BDE-28, 47, 100 and �PBDE aswell as for �PBDE in dust and umbilical cord plasma. Furthermore,correlations between different penta-BDE congeners in dust andmaternal plasma were also observed, possibly as a result of stronginter-correlation within one matrix. However, metabolic processesas well as differences in bioavailability and compound uptake canalso play a role, but were outside the scope of this study. Our resultsindicate that domestic house dust indeed can be a significant sourcefor PBDE exposure of the general population as well as in utero, alsoin Europe.

Conclusions

Concentrations of PBDE in maternal and umbilical cord plasmawere highly correlated and similar to previous results for Europeanmaternal and umbilical cord blood samples. The FM-ratios showeda decreasing transport of the PBDE congeners across the placentawith increasing degree of bromination. While the transport of tri-and tetra-BDEs seems to be primarily lipid-dependent, other fac-tors than lipid solubility apparently affect the transport of the largerPBDE congeners. These might include restricted membrane perme-ability and therefore a higher degree of accumulation within theplacental tissue, compared to the lower brominated PBDEs. ThePBDE concentrations in maternal plasma were highly correlatedwith previously observed concentrations in placental tissue andincluded both direct correlations and correlations between differ-ent congeners. Finally, the levels of several PBDE congeners in both

the maternal and umbilical cord plasma were found to be positivelycorrelated with the levels of PBDEs measured in the house dust ofthe study participants, indicating that house dust is a significantsource of human PBDE exposure, including in utero exposure, inDenmark.

Page 9

giene

A

vKcdUmD0CFt

A

t

R

A

A

A

A

A

B

B

C

D

d

d

F

F

F

F

F

F

F

G

G

M. Frederiksen et al. / International Journal of Hy

cknowledgements

The authors gratefully acknowledge the contribution from theolunteer donors in this study, Marie Pedersen for recruitment, andillian Smith for proofreading the manuscript. Furthermore, theontribution from Professor Lars Bo Nielsen and Charlotte Wan-el from the Department of Clinical Biochemistry, Copenhagenniversity Hospital (Rigshospitalet) for the enzymatic lipid deter-inations is greatly appreciated. The study was supported by theanish Agency for Science, Technology and Innovation (271-06-355) and the Danish Ministry of the Interior and Health, Researchentre for Environmental Health’s Fund (0-302-02-18/4). Marierederiksen’s stay at NIPH was funded by the PhD study board athe Faculty of Health Sciences, University of Copenhagen.

ppendix A. Supplementary data

Supplementary data associated with this article can be found, inhe online version, at doi:10.1016/j.ijheh.2010.04.008.

eferences

bdelouahab, N., Suvorov, A., Pasquier, J.C., Langlois, M.F., Praud, J.P., Takser, L., 2009.Thyroid disruption by low-dose BDE-47 in prenatally exposed lambs. Neonatol-ogy 96, 120–124.

laee, M., Arias, P., Sjödin, A., Bergman, Å., 2003. An overview of commercially usedbrominated flame retardants, their applications, their use patterns in differentcountries/regions and possible modes of release. Environ. Int. 29, 683–689.

llen, J.G., McClean, M.D., Stapleton, H.M., Webster, T.F., 2008. Critical factors inassessing exposure to PBDEs via house dust. Environ. Int. 34, 1085–1091.

MAP, 2008. AMAP Ring test for Persistent Organic Pollutants in Human Serum.Report 2008-3.

ntignac, J.P., Cariou, R., Zalko, D., Berrebi, A., Cravedi, J.P., Maume, D., Marchand, P.,Monteau, F., Riu, A., Andre, F., Le Bizec, B., 2009. Exposure assessment of Frenchwomen and their newborn to brominated flame retardants: determination oftri- to deca-polybromodiphenylethers (PBDE) in maternal adipose tissue, serum,breast milk and cord serum. Environ. Pollut. 157, 164–173.

raekevelt, E., Tittlemier, S.A., Tomy, G.T., 2003. Direct measurement ofoctanol–water partition coefficients of some environmentally relevant bromi-nated diphenyl ether congeners. Chemosphere 51, 563–567.

SEF, 2010. Bromine Science an Environmental Forum, available at www.bsef.org.(January, 2010).

ovaci, A., Jorens, P., Jacquemyn, Y., Schepens, P., 2002. Distribution of PCBs andorganochlorine pesticides in umbilical cord and maternal serum. Sci. Tot. Envi-ron. 298, 45–53.

arras, V., 2008. Endocrine disrupting polyhalogenated organic pollutants interferewith thyroid hormone signalling in the developing brain. Cerebellum 7, 26–37.

e Wit, C.A., 2002. An overview of brominated flame retardants in the environment.Chemosphere 46, 583–624.

e Wit, C.A., Herzke, D., Vorkamp, K., 2010. Brominated flame retardantsin the Arctic environment—trends and new candidates. Sci. Tot. Environ.,doi:10.1016/j.scitotenv.2009.08.037.

angstrom, B., Athanassiadis, I., Odsjo, T., Norén, K., Bergman, A., 2008. Tempo-ral trends of polybrominated diphenyl ethers and hexabromocyclododecane inmilk from Stockholm mothers, 1980–2004. Mol. Nutr. Food Res. 52, 187–193.

raser, A.J., Webster, T.F., McClean, M.D., 2009. Diet contributes significantly to thebody burden of PBDEs in the general U.S. population. Environ. Health Perspect.117, 1520–1525.

rederiksen, M., Thomsen, M., Vorkamp, K., Knudsen, L.E., 2009a. Patterns and con-centration levels of polybrominated diphenyl ethers (PBDEs) in placental tissueof women in Denmark. Chemosphere 76, 1464–1469.

rederiksen, M., Vorkamp, K., Thomsen, M., Knudsen, L.E., 2009b. Human internaland external exposure to PBDEs—a review of levels and sources. Int. J. Hyg.Environ. Health 212, 109–134.

rederiksen, M., Vorkamp, K., Mathiesen, L., Mose, T., Knudsen, L.E., in press. Placentaltransfer of the polybrominated diphenyl ethers BDE-47, BDE-99, and BDE-209in a human placenta perfusion system: An Experimental study. Environ. Health.

ries, G.F., Kimbrough, R.D., 1985. The PBB episode in Michigan: an overall appraisal.Crit. Rev. Toxicol. 16, 105–156.

romme, H., Körner, W., Shahin, N., Wanner, A., Albrecht, M., Boehmer, S., Parlar, H.,Mayer, R., Liebl, B., Bolte, G., 2009. Human exposure to polybrominated diphenylethers (PBDE), as evidenced by data from a duplicate diet study, indoor air, housedust, and biomonitoring in Germany. Environ. Int. 35, 1125–1135.

eyer, H.J., Schramm, K.W., Darnerud, P.O., Aune, M., Feicht, E.A., Fried, K.W., Henkel-mann, B., Lenoir, D., Schmid, P., McDonald, T.A., 2004. Terminal eliminationhalf-lives of the brominated flame retardants TBBPA, HBCD, and lower bromi-nated PBDEs in humans. Organohal. Comp. 66, 3867–3872.

omara, B., Herrero, L., Ramos, J.J., Mateo, J.R., Fernandez, M.A., Garcia, J.F., Gonzalez,M.J., 2007. Distribution of polybrominated diphenyl ethers in human umbilical

and Environmental Health 213 (2010) 233–242 241

cord serum, paternal serum, maternal serum, placentas, and breast milk fromMadrid population, Spain. Environ. Sci. Technol. 41, 6961–6968.

Grimvall, E., Rylander, L., Nilsson-Ehle, P., Nilsson, U., Strömberg, U., Hagmar, L., Öst-man, C., 1997. Monitoring of polychlorinated biphenyls in human blood plasma:methodological developments and influence of age, lactation, and fish consump-tion. Arch. Environ. Contam. Toxicol. 32, 329–336.

Guvenius, D.M., Aronsson, A., Ekman-Ordeberg, G., Bergman, A., Norén, K., 2003.Human prenatal and postnatal exposure to polybrominated diphenyl ethers,polychlorinated biphenyls, polychlorobiphenylols, and pentachlorophenol. Env-iron. Health Perspect. 111, 1235–1241.

Harrad, S., Ibarra, C., Diamond, M., Melymuk, L., Robson, M., Douwes, J., Roosens, L.,Dirtu, A.C., Covaci, A., 2008. Polybrominated diphenyl ethers in domestic indoordust from Canada, New Zealand, United Kingdom and United States. Environ.Int. 34, 232–238.

Haug, L.S., Becher, G., 2005. Interlaboratory comparison on dioxins in food. Report2005:7. Norwegian Institute of Public Health, Oslo, Norway.

Haug, L.S., Becher, G., 2006. Interlaboratory comparison on dioxins in food. Report2006:7. Norwegian Institute of Public Health, Oslo, Norway.

Herbstman, J., Sjodin, A., Patterson, D.G., Apelberg, B.J., Witter, F.R., Halden,R.U., Heidler, J., Needham, L.L., Goldman, L.R., 2007. Determinants of pre-natal exposure to polychlorinated biphenyls (PCBs) and polybrominateddiphenyl ethers (PBDEs) in an urban population. Environ. Health Perspect. 115,1794–1800.

Herbstman, J., Sjodin, A., Kurzon, M., Lederman, S.A., Jones, R.S., Rauh, V., Needham,L., Tang, D., Niedzwiecki, M., Wang, R.Y., Perera, F., 2010. Prenatal exposure toPBDEs and neurodevelopment. Environ. Health Perspect 118, 712–719.

Hewitt, M., Madden, J.C., Rowe, P.H., Cronin, M.T.D., 2007. Structure-based modellingin reproductive toxicology: (Q)SARs for the placental barrier. SAR and QSAREnviron. Res. 18, 57–76.

Huwe, J.K., Hakk, H., Birnbaum, L.S., 2008. Tissue distribution of polybrominateddiphenyl ethers in male rats and implications for biomonitoring. Environ. Sci.Technol. 42, 7018–7024.

Kajiwara, N., Kamikawa, S., Amano, M., Hayano, A., Yamada, T.K., Miyazaki, N.,Tanabe, S., 2008. Polybrominated diphenyl ethers (PBDEs) and organochlo-rines in melon-headed whales, Peponocephala electra, mass stranded alongthe Japanese coasts: maternal transfer and temporal trend. Environ. Pollut. 156,106–114.

Karlsson, M., Julander, A., van Bavel, B., Hardell, L., 2007. Levels of brominated flameretardants in blood in relation to levels in household air and dust. Environ. Int.33, 62–69.

Kawashiro, Y., Fukata, H., Omori-Inoue, M., Kubonoya, K., Jotaki, T., Takigam, H.,Sakai, S.I., Mori, C., 2008. Perinatal exposure to brominated flame retardantsand polychlorinated biphenyls in Japan. Endocr. J. 55, 1071–1084.

La Guardia, M.J., Hale, R.C., Harvey, E., 2006. Detailed polybrominated diphenyl ether(PBDE) congener composition of the widely used penta-, octa-, and deca-PBDEtechnical flame-retardant mixtures. Environ. Sci. Technol. 40, 6247–6254.

Law, R.J., Allchin, C.R., Bennett, M.E., Morris, S., Rogan, E., 2002. Polybrominateddiphenyl ethers in two species of marine top predators from England and Wales.Chemosphere 46, 673–681.

Law, R.J., Allchin, C.R., de Boer, J., Covaci, A., Herzke, D., Lepom, P., Morris, S., Tron-czynski, J., de Wit, C.A., 2006. Levels and trends of brominated flame retardantsin the European environment. Chemosphere 64, 187–208.

Legler, J., 2008. New insights into the endocrine disrupting effects of brominatedflame retardants. Chemosphere 73, 216–222.

Lorber, M., 2008. Exposure of Americans to polybrominated diphenyl ethers. J. Exp.Sci. Environ. Epidemiol. 18, 2–19.

Main, K.M., Kiviranta, H., Virtanen, H.E., Sundquist, E., Tuomisto, J.T., Toumisto, J.,Vartiainen, T., Skakkebæk, N.E., Toppari, J., 2007. Flame retardants in placentaand breast milk and cryptorchidism in newborn boys. Environ. Health Perspect.115, 1519–1526.

Mathiesen, L., Mose, T., Mørck, T.J., Nielsen, J.K.S.N.L.K., Maroun, L.L., Dziegel, M.H.,Larsen, L.G., Knudsen, L.E., 2010. Quality assessment of a placental perfusionprotocol. Reprod. Toxicol., doi:10.1016/j.reprotox.2010.01.006.

Mazdai, A., Dodder, N.G., Abernathy, M.P., Hites, R.A., Bigsby, R.M., 2003. Polybromi-nated diphenyl ethers in maternal and fetal blood samples. Environ. HealthPerspect. 111, 1249–1252.

Meeker, J.D., Johnson, P.I., Camann, D., Hauser, R., 2009. Polybrominated diphenylether (PBDE) concentrations in house dust are related to hormone levels in men.Sci. Tot. Environ. 407, 3425–3429.

Meijer, L., Weiss, J., van Velzen, M., Brouwer, A., Bergman, ke Sauer, P.J.J., 2008. Serumconcentrations of neutral and phenolic organohalogens in pregnant women andsome of their infants in the Netherlands. Environ. Sci. Technol. 42, 3428–3433.

Mose, T., Mortensen, G.K., Hedegaard, M., Knudsen, L.E., 2007. Phthalate monoestersin perfusate from a dual placenta perfusion system, the placenta tissue andumbilical cord blood. Reprod. Toxicol. 23, 83–91.

Myren, M., Mose, T., Mathiesen, L., Knudsen, L.E., 2007. The human placenta—analternative for studying foetal exposure. Toxicol. In Vitro 21, 1332–1340.

Pedersen, M., Wichmann, J., Autrup, H., Dang, D.A., Decordier, I., Hvidberg, M., Bossi,R., Jakobsen, J., Loft, S., Knudsen, L.E., 2009. Increased micronuclei and bulky DNAadducts in cord blood after maternal exposures to traffic-related air pollution.

Environ. Res. 109, 1012–1020.

Roosens, L., Abdallah, M.A.-E., Harrad, S., Neels, H., Covaci, A., 2009. Factors influ-encing concentrations of polybrominated diphenyl ethers (PBDEs) in studentsfrom Antwerp, Belgium. Environ. Sci. Technol. 43, 3535–3541.

Ross, P.S., Couillard, C.M., Ikonomou, M.G., Johannessen, S.C., Lebeuf, M., Macdonald,R.W., Tomy, G.T., 2009. Large and growing environmental reservoirs of Deca-

Page 10

2 giene

S

S

TT

T

T

Wu, N., Herrmann, T., Paepke, O., Tickner, J., Hale, R., Harvey, E., LaGuardia, M.,

42 M. Frederiksen et al. / International Journal of Hy

BDE present an emerging health risk for fish and marine mammals. Mar. Pollut.Bull. 58, 7–10.

chnack, T.H., Zdravkovic, S., Myrup, C., Westergaard, T., Wohlfahrt, J., Melbye, M.,2008. Familial aggregation of cryptorchidism—a nationwide cohort study. Am.J. Epidemiol. 167, 1453–1457.

jodin, A., Hagmar, L., Klasson-Wehler, E., Kronholm-Diab, K., Jakobsson, E., Bergman,A., 1999. Flame retardant exposure: polybrominated diphenyl ethers in bloodfrom Swedish workers. Environ. Health Perspect. 107, 643–648.

he Stockholm Convention, 2010. available at: http://chm.pops.int/. (January 2010).homas, G.O., Wilkinson, M., Hodson, S., Jones, K.C., 2006. Organohalogen chem-

icals in human blood from the United Kingdom. Environ. Pollut. 141,30–41.

homsen, C., Liane, V.H., Becher, G., 2007. Automated solid-phase extraction forthe determination of polybrominated diphenyl ethers and polychlorinated

biphenyls in serum—application on archived Norwegian samples from 1977 to2003. J. Chromatogr. B 846, 252–263.

homsen, C., Liane, V.H., Frøshaug, M., Kvalem, H., Haugen, M., Meltzer, H.M., Alexan-der, J., Becher, G., 2008. Consumption of fish from a contaminated lake stronglyaffects the concentrations of polybrominated diphenyl ethers and hexabromo-cyclododecane in serum. Mol. Nutr. Food Res. 52, 228–237.

and Environmental Health 213 (2010) 233–242

Turyk, M.E., Persky, V.W., Imm, P., Knobeloch, L., Chatterton, R., Anderson, H.A.,2009. Hormone disruption by PBDEs in adult male sport fish consumers. Environ.Health Perspect. 116, 1635–1641.

UNEP, 2006. Hexabromobiphenyl: Risk Profile. United Nations Environment Pro-gramme (2006), UNEP/POPS/POPRC. 2/17, Persistent Organic Pollutants ReviewCommittee, Second Meeting, Geneva, 6–10 November 2006.

Vorkamp, K., Frederiksen, M., Thomsen, M., Pedersen, M., Knudsen, L.E., in press.Polybrominated diphenyl ethers (PBDEs) in the indoor environment and asso-ciations with prenatal exposure. Environ. Int.

Welsh, M., Saunders, P.T.K., Fisken, M., Scott, H.M., Hutchison, G.R., Smith, L.B.,Sharpe, R.M., 2008. Identification in rats of a programming window for repro-ductive tract masculinization, disruption of which leads to hypospadias andcryptorchidism. J. Clin. Invest. 118, 1479–1490.

McClean, M.D., Webster, T.F., 2007. Human exposure to PBDEs: associationsof PBDE body burdens with food consumption and house dust concentrations.Environ. Sci. Technol. 41, 1584–1589.

Yogui, G.T., Sericano, J.L., 2009. Polybrominated diphenyl ether flame retardants inthe U.S. marine environment: a review. Environ. Int. 35, 655–666.