<p>Brominated flame retardants (BFRs) are a diverse group of chemicals, which are used to slow down or inhibit the development of fires. BFRs are incorporated into a wide range of consumer products that are considered as potential fire hazards, such as TV-sets, household appliances, computers, and textiles. The production and use of BFRs is extensive and consists of mainly tetrabromobisphenol A (TBBPA), polybrominated diphenyl ethers (PBDEs) and hexabromocyclododecan (HBCD). BFRs in general, but in particular the PBDEs, have led to both scientific and public concern since they have been found to bioaccumulate in humans and wildlife. The general population is targeted by the PBDEs due to their applications and via the food web. Occupational exposure occurs not only during direct handling of BFRs, but also during handling, repair and dismantling of flame retarded goods.</p><p>This thesis is aimed to assess occupational exposure to BFRs. It is mainly focused PBDEs and especially the PBDEs with high bromine content, such as decabromodiphenyl ether (BDE-209). The work has been accomplished by analysis of BFRs in indoor air at industries handling BFRs or flame retarded goods, and by analysis of blood drawn from workers with potential exposure to BFRs. A referent group, abattoir workers with no occupational exposure to PBDEs, was also investigated. Data from these cross-sectional investigations and from serum sampling during vacation in PBDE-exposed workers have been used for calculation of apparent halflives of hepta- to decaBDE in serum.</p><p>The results clearly show that the workers were exposed to PBDEs when handling PBDE containing products or goods. The serum PBDE levels in computer technicians were found to correlate to the estimated cumulative work hours with computers. Exceptionally high concentrations of BDE-209, almost up to 300 pmol/g lipid weight (l.w.) were observed in serum from rubber workers manufacturing or handling rubber compound that was flame retarded with a technical mixture of decabromodiphenyl ether (DecaBDE). Elevated concentrations of PBDEs with eight or nine bromine substituents were also observed. In an electronics dismantling plant, where elevated levels of PBDEs previously had been observed, reduced serum levels of some, but not all PBDE congeners were observed after industrial hygiene improvements. Notably, it was observed that the BDE-209 concentrations in referents with no occupational exposure were similar to the concentrations of 2,2’,4,4’- tetrabromodiphenyl ether (BDE-47), often referred to as the most abundant PBDE congener in humans and wildlife. Additionally, PBDEs with high bromine content were found to have a fast rate of elimination or transformation in humans, based on serum analysis. BDE-209 had an apparent half-life in serum of only 15 days.</p><p>The possibility of quantifying BFRs, such as PBDEs, in human serum at low levels of detection has been achieved by reducing the contamination of the samples and procedural blanks. Major efforts have been done to develop routines and clean up methodology to enable an almost contamination-free environment at the laboratory. The use of a clean room has decreased PBDE levels in the blanks to acceptable limits. The modifications of the original analytical method have made it possible to quantify almost all PBDE congeners of interest in one GC/MS run.</p><p>Occupational and general background exposure of BFRs to humans will continue as long as these chemicals are a part of our daily life and present as environmental contaminants. The present scientific knowledge of the potential health risks of these BFRs still needs to be further developed. It should be stressed that health effects to PBDEs have not been assessed in this work. It is the author's wish that this thesis should add another piece of knowledge to the puzzle of BFRs and BFR exposure to humans and that these data will be used in future risk assessments of PBDEs in particular.</p>

Brominated flame retardants (BFRs) are a diverse group of chemicals, which are used to slow down or inhibit the development of fires. BFRs are incorporated into a wide range of consumer products that are considered as potential fire hazards, such as TV-sets, household appliances, computers, and textiles. The production and use of BFRs is extensive and consists of mainly tetrabromobisphenol A (TBBPA), polybrominated diphenyl ethers (PBDEs) and hexabromocyclododecan (HBCD). BFRs in general, but in particular the PBDEs, have led to both scientific and public concern since they have been found to bioaccumulate in humans and wildlife. The general population is targeted by the PBDEs due to their applications and via the food web. Occupational exposure occurs not only during direct handling of BFRs, but also during handling, repair and dismantling of flame retarded goods. This thesis is aimed to assess occupational exposure to BFRs. It is mainly focused PBDEs and especially the PBDEs with high bromine content, such as decabromodiphenyl ether (BDE-209). The work has been accomplished by analysis of BFRs in indoor air at industries handling BFRs or flame retarded goods, and by analysis of blood drawn from workers with potential exposure to BFRs. A referent group, abattoir workers with no occupational exposure to PBDEs, was also investigated. Data from these cross-sectional investigations and from serum sampling during vacation in PBDE-exposed workers have been used for calculation of apparent halflives of hepta- to decaBDE in serum. The results clearly show that the workers were exposed to PBDEs when handling PBDE containing products or goods. The serum PBDE levels in computer technicians were found to correlate to the estimated cumulative work hours with computers. Exceptionally high concentrations of BDE-209, almost up to 300 pmol/g lipid weight (l.w.) were observed in serum from rubber workers manufacturing or handling rubber compound that was flame retarded with a technical mixture of decabromodiphenyl ether (DecaBDE). Elevated concentrations of PBDEs with eight or nine bromine substituents were also observed. In an electronics dismantling plant, where elevated levels of PBDEs previously had been observed, reduced serum levels of some, but not all PBDE congeners were observed after industrial hygiene improvements. Notably, it was observed that the BDE-209 concentrations in referents with no occupational exposure were similar to the concentrations of 2,2’,4,4’- tetrabromodiphenyl ether (BDE-47), often referred to as the most abundant PBDE congener in humans and wildlife. Additionally, PBDEs with high bromine content were found to have a fast rate of elimination or transformation in humans, based on serum analysis. BDE-209 had an apparent half-life in serum of only 15 days. The possibility of quantifying BFRs, such as PBDEs, in human serum at low levels of detection has been achieved by reducing the contamination of the samples and procedural blanks. Major efforts have been done to develop routines and clean up methodology to enable an almost contamination-free environment at the laboratory. The use of a clean room has decreased PBDE levels in the blanks to acceptable limits. The modifications of the original analytical method have made it possible to quantify almost all PBDE congeners of interest in one GC/MS run. Occupational and general background exposure of BFRs to humans will continue as long as these chemicals are a part of our daily life and present as environmental contaminants. The present scientific knowledge of the potential health risks of these BFRs still needs to be further developed. It should be stressed that health effects to PBDEs have not been assessed in this work. It is the author's wish that this thesis should add another piece of knowledge to the puzzle of BFRs and BFR exposure to humans and that these data will be used in future risk assessments of PBDEs in particular.

Les déterminants de l’exposition aux substances per- et polyfluoroalkyliques (PFAS) chez les enfants sont mal connus. Cette étude visait à analyser les concentrations sériques de 9 PFAS chez 204 enfants participant à l’étude MIREC-Endo ; évaluer les associations entre ces concentrations et celles dans le sérum maternel (grossesse) et le lait maternel ; évaluer les déterminants des concentrations. Nous avons effectué des statistiques descriptives des concentrations et évalué leurs associations à l’aide de corrélations de Pearson et de tests de comparaisons de moyennes. Des analyses de régression ont été faites pour quantifier l’influence de l’allaitement sur les concentrations sériques. Les moyennes géométriques de PFOS, PFOA, PFHxS, PFNA et PFDA (détectés dans >67% des échantillons) étaient de 1,37 ; 1,21 ; 0,55 ; 0,42 et 0,13 µg/L, respectivement. Les concentrations sériques de certains PFAS étaient corrélées avec les concentrations sériques maternelles (r=0,315 [PFOS], 0,314 [PFOA], 0,328 [PFHxS]) et les concentrations dans le lait (r=0,273 [PFOA], 0,509 [PFHxS], 0,237 [PFNA]). Les concentrations sériques de certains PFAS chez les enfants étaient négativement associées avec l’âge maternel à l’accouchement (PFOS, PFHxS, PFDA), le tabagisme durant la grossesse (PFNA, PFDA), l’emploi de la mère (PFOS), alors qu’elles étaient positivement associées avec le niveau d’éducation maternel (PFNA) et le revenu familial (PFOA). Nous avons observé une augmentation de 2,8% (PFOS) et 1,85% (PFOA) dans les concentrations sériques par mois d’allaitement exclusif. En conclusion, les concentrations sériques de certains PFAS chez les enfants étaient associées aux expositions périnatales, à l’âge de la mère, au revenu familial et à la durée d’allaitement. / The determinants of exposure to per- and polyfluoroalkyl substances (PFAS) in children are unclear. The objectives of this study were to analyze serum concentrations of 9 PFAS in 204 children aged 7.5 to 11.8 years participating in the MIREC-Endo study, to assess associations between these concentrations and those in maternal serum (during pregnancy) and breast milk; and to evaluate the determinants of these concentrations. We performed descriptive statistics of the concentrations and assessed their associations using Pearson correlations and comparison tests. Regression analyses were performed to quantify the influence of breastfeeding on serum concentrations. The geometric means of PFOS, PFOA, PFHxS, PFNA, and PFDA (detected in >67% of samples) were 1.37, 1.21, 0.55, 0.42, and 0.13 µg/L, respectively. Serum concentrations of selected PFASs were correlated with maternal serum concentrations (r=0.315 [PFOS], 0.314 [PFOA], 0.328 [PFHxS]) and milk concentrations (r=0.273 [PFOA], 0.509 [PFHxS], 0.237 [PFNA]). Serum concentrations of children PFAS were negatively associated with maternal age at delivery (PFOS, PFHxS, PFDA), smoking during pregnancy (PFNA, PFDA), maternal employment (PFOS), whereas they were positively associated with maternal education level (PFNA) and family income (PFOA). We observed an increase of 2.8% (PFOS) and 1.85% (PFOA) in serum concentrations per month of exclusive breastfeeding. In conclusion, serum concentrations of selected PFAS in children were associated with perinatal exposures, maternal age, family income, and duration of breastfeeding.