Brominated Flame Retardants in Archived Serum Samples from

other brominated phenols are known to be formed by marine biota (29−33). .... temperature was 250 °C. Helium (99.998%, Aga) was used as a carri...
0 downloads 0 Views 69KB Size
Download PDF
Environ. Sci. Technol. 2002, 36, 1414-1418

Brominated Flame Retardants in Archived Serum Samples from Norway: A Study on Temporal Trends and the Role of Age C A T H R I N E T H O M S E N , * ,† E L S A L U N D A N E S , ‡ A N D G E O R G B E C H E R †,‡ National Institute of Public Health, P.O. Box 4404 Nydalen, N-0403 Oslo, Norway, and Department of Chemistry, University of Oslo, P.O. Box 1033 Blindern, N-0315 Oslo, Norway

The temporal trends and influence of age and gender on levels of selected brominated flame retardants (BFRs) in human serum have been assessed by analyzing archived samples from Norway. Serum from 40 to 50 year old men collected at six time periods during 1977 to 1999 and from eight groups of differing age and gender sampled in 1998 were pooled into six and eight samples, respectively. The BFRs were isolated using solid-phase extraction (SPE) and the serum lipids decomposed by treatment with concentrated sulfuric acid directly on the polystyrene-divinylbenzene SPE column, prior to elution of the BFRs. Following diazomethane derivatization, the samples were analyzed by gas chromatography-electron capture mass spectrometry. Eight BFRs were quantified in the serum samples: 2,4,4′tribromodiphenyl ether (BDE-28), 2,2′,4,4′-tetrabromodiphenyl ether (BDE-47), 2,2′,4,4′,5-pentabromodiphenyl ether (BDE99), 2,2′,4,4′,6-pentabromodiphenyl ether (BDE-100), 2,2′,4,4′,5,5′-hexabromodiphenyl ether (BDE-153), 2,2′,4,4′,5,6′hexabromodiphenyl ether (BDE-154), 2,4,6-tribromophenol (TriBP), and tetrabromobisphenol A (TBBP-A). The serum concentrations of all the BFRs, increased during the entire period with the exception of TriBP, and the sum of the six polybrominated diphenyl ethers increased from 0.44 ng/g lipids in 1977 to 3.3 ng/g lipids in 1999. The BFR concentrations in the serum from the different age groups were relatively similar, except for the age group 0-4 years, which had 1.6-3.5 times higher serum concentrations. Women older than 25 years had lower serum concentrations of BFRs compared to the corresponding group of men. No trend related to age or gender, nor time during the period 1977 to 1999 was observed for TriBP. The present study indicates an ongoing increase in human exposure to BFRs, and the current body burden appears to be independent of age, except for infants (0-4 years old), who seem to experience elevated exposure.

Introduction Flame retardants constitute a diverse group of compounds that are added to materials in order to suppress, reduce, or * Corresponding author phone: +47 22 04 23 41; fax: +47 22 04 26 86; e-mail: [email protected]. † National Institute of Public Health. ‡ University of Oslo. 1414

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 7, 2002

delay fire. The most widely used brominated flame retardants (BFRs) are tetrabromobisphenol A (TBBP-A) and polybrominated diphenyl ethers (PBDEs), while others such as tribromophenol (TriBP) and pentabromophenol (PeBP) are also used (1). The BFRs are used in polymers either reactively or as additives, in a wide range of products, such as printed circuit boards, television and computer housings and other electronic household equipment, automotive parts, and construction materials (1-3). The use of BFRs has expanded considerably during recent decades, and the annual global demand was estimated to 150 000 tons in 1992 (1) and 200 000 tons in 1999 (4). The BFRs might be released into the environment during production, use, and especially from disposal of the flame retarded products. The environmental distribution of PBDEs (5-8) and TBBP-A (7) has recently been reviewed. So far, knowledge on the toxicity of BFRs is limited to a few congeners. Concerning the studied PBDE congeners, hepatotoxicity, embryotoxicity, and thyroid effects seem to be characteristic end points in animal toxicity, and behavioral effects have been demonstrated (8). Also, TBBP-A, TriBP, PeBP, and the hydroxylated metabolites of PBDE have been reported to bind to transthyretin in vitro (9-11). A recent study indicated that several PBDE congeners, but especially hydroxylated PBDEs and brominated bisphenol A analogues, are agonists of the two estrogen receptors, in vitro (12). Thus, these BFRs have potential for posing a threat to human health and the environment. Due to their high lipophilicity and resistance to degradation numerous BFRs are present in biota, and several of the PBDEs have shown potential for biomagnification in the food chain (13-15). Humans are mainly exposed through the diet, especially from intake of fatty food of animal origin (8). In addition, volatilization of BFRs from hot electrical equipment to the surrounding air might also be of importance, and occupational exposure has been demonstrated in certain occasions (16, 17). PBDEs have been found in human blood (16-20), adipose tissue (15, 21-24), and breast milk (2528), and TBBP-A has been determined in blood samples (16, 20). Some hydroxylated PBDEs and several other brominated phenols are known to be formed by marine biota (29-33). However, neither TBBP-A nor PBDEs have been reported to occur naturally in the environment, and their presence in biological samples is of anthropogenic origin. Retrospective time trend studies are a valuable tool for judging the development of a pollution situation. In Sweden, the levels of PBDEs have been found to increase in sediments (34), guillemot eggs (14), and pike (35) since the 1960s. The PBDE concentrations in human milk have been shown to increase exponentially, at a rate corresponding to a doubling of the concentrations every 5 years in the period between 1972 and 1997 (26). Trends of increasing PBDE concentrations have also been observed in human plasma samples from Germany (19), in lake trout (36), beluga (37), and ringed seal (38) from Canada and in Norwegian sediments (39). On the other hand, recent investigations have reported an apparent declining trend of PBDE concentrations in guillemot eggs (7) and human milk (40). Further well designed investigations seem to be necessary to assess the time-related trends of BFR levels in the environment and the body burden of BFRs in the general population. From 1975, serum samples from different county hospitals in Norway have been thoroughly registered and stored in a specimen bank at the National Institute of Public Health, creating a valuable archive of historical samples. The aim of the present study was to determine the concentrations and 10.1021/es0102282 CCC: $22.00

 2002 American Chemical Society Published on Web 03/01/2002

TABLE 1. Serum Concentrations of the Brominated Flame Retardants in ng/g Lipids (pmol/g Lipids in Parentheses)a serum pool

fat %

n

BDE-28

BDE-47

BDE-100

BDE-99

BDE-154

BDE-153

TBBP-A

TriBP

1977 1981 1986 1990 1995 1999 0-4 years 4-14 years 15-24 years, f 15-24 years, m 25-59 years, f 25-59 years, m >60 years, f >60 years, m

0.62 0.67 0.74 0.68 0.69 0.65 0.50 0.53 0.51 0.62 0.75 0.57 0.64 0.65

34 17 24 20 19 29 14 10 10 13 12 11 13 10

nd 0.096 (0.23) nd 0.066 (0.16) 0.14 (0.34) 0.24 (0.59) 0.26 (0.64) 0.20 (0.49) 0.088 (0.22) 0.094 (0.23) 0.093 (0.23) 0.093 (0.23) 0.36 (0.90) 0.096 (0.24)

0.25 (0.52) 0.32 (0.67) 0.41 (0.86) 0.89 (1.8) 1.4 (2.9) 1.5 (3.1) 6.2 (13) 2.0 (4.2) 2.5 (5.2) 2.3 (4.7) 1.3 (2.8) 2.3 (4.7) 1.2 (2.5) 3.4 (7.1)

nd 0.079 (0.14)b 0.12 (0.22) 0.13 (0.24) 0.32 (0.56) 0.35 (0.62) 1.7 (3.0) 0.66 (1.2) 0.49 (0.87) 0.61 (1.1) 0.28 (0.50) 0.52 (0.93) 0.37 (0.66) 0.45 (0.80)

0.087 (0.15) 0.13 (0.23) 0.13 (0.22) 0.24 (0.42 0.33 (0.58) 0.31 (0.55) 1.6 (2.8) 0.37 (0.66) 0.71 (1.3) 0.68 (1.2) 0.32 (0.56) 0.40 (0.71) 0.25 (0.44 0.36 (0.64)

nd 0.22 (0.35) 0.26 (0.40) 0.23 (0.36) 0.50 (0.77) 0.35 (0.55) 0.45 (0.70) 0.39 (0.60) 0.27 (0.42) 0.19 (0.30) 0.23 (0.36) 0.24 (0.37) 0.26 (0.40) 0.38 (0.59)

0.10 (0.16) 0.18 (0.28) 0.14 (0.21) 0.27 (0.42) 0.52 (0.81) 0.59 (0.91) 1.5 (2.3) 0.86 (1.3) 0.56 (0.87 0.66 (1.0) 0.34 (0.53) 0.71 (1.2) 0.36 (0.56) 0.59 (0.92)

nd nd 0.44 (0.82) 0.42 (0.78) 0.59 (1.1) 0.65 (1.2) 0.71 (1.3) 0.40 (0.73) 0.40 (0.75) 0.56 (1.0) 0.34 (0.63) 0.65 (1.2) 0.42 (0.78) 0.31 (0.57)

0.43 (1.3) 0.077 (0.23) 1.3 (4.0) 0.21 (0.63) 0.30 (0.91) 0.30 (0.91) 0.95 (2.9) 14 (41) 1.5 (4.6) 1.8 (5.6) 26 (79) 2.6 (7.9) 0.45 (1.4) 0.20 (0.60)

a The fat content of the serum samples and the number of individuals represented in each pool are given in column 2 and 3. Abbreviations: f, females; m, males; nd, not detected. b The compound was identified at a level below the LOQ.

the temporal trends of selected BFRs in serum samples from this unique collection, sampled during the period 1977 to 1999. To avoid a possible variation of body burden with age and sex, this study was restricted to male subjects within a relatively close age range. It is well documented that the body burden of other persistent organic pollutants (POPs) such as polychlorinated dibenzo-p-dioxines (PCDDs), polychlorinated dibenzofurans (PCDFs), and polychlorinated biphenyls (PCBs) increases with age and for women decreases with the number of breast fed children (41,42). A corresponding knowledge concerning the BFRs is presently not available. We therefore also assessed the dependence of BFR levels in serum on age and gender for samples collected in 1998.

Materials and Methods Chemicals. TBBP-A and chlorotribromobisphenol A (CtriBBPA) were a gift from the Wallenberg Laboratory (University of Stockholm, Sweden). 2,4,4′-Tribromodiphenyl ether (BDE28), 2,2′,4,4′-tetrabromodiphenyl ether (BDE-47), 2,2′,4,4′,5pentabromodiphenyl ether (BDE-99), 2,2′,4,4′,6-pentabromodiphenyl ether (BDE-100), 2,2′,4,4′,5,5′-hexabromodiphenyl ether (BDE-153), 2,2′,4,4′,5,6′-hexabromodiphenyl ether (BDE-154), 2,2′,3,4,4′,5′,6-heptabromodiphenyl ether (BDE-183), and 13C-3,3′,4,4′-tetrabromodiphenyl ether (BDE77) were obtained from by Wellington Laboratories (Guelph, Ontario, Canada). TriBP, tetrabromo-o-cresol (TBCr), and N-methyl-N-nitroso-p-toluenesulfonamide (Diazald) were purchased from Aldrich (Milwaukee, WI), PeBP from Acros (Geel, Belgium), 2,2′,4,4′,5,5′-hexachlorobiphenyl (CB-153), and 3,3′,4,4′-tetrabromobiphenyl (BB-77) from AccuStandard Inc. (New Haven, CT) and 1,3,5-tribromobenzene (TriBB) from Fluka (Buchs, Switzerland). All solvents were pesticide grade from Labscan (Dublin, Ireland), except ethyl acetate, which was from SDS (Peypin, France). Formic acid, sulfuric acid, and sodium acetate-trihydrate were of analytical grade (Merck, Darmstadt, Germany). Water was purified using an Elga Option 4 Water Purifier device (Elga, Bucks, U.K.). Glassware. All glassware was washed in 2.5% RBS 25 foaming cleaner (Chemical Products, Brussels, Belgium), rinsed with distilled water, and subsequently heated at 450 °C for 4 h (volumetric equipment was not heated). Serum Samples. The study was conducted on serum samples archived in a serum bank at the National Institute of Public Health in Norway. The serum had been sampled from patients at five different county hospitals, regardless of disease and the reason for hospitalization, during the period of 1975 to present date, and stored at -20 °C. In the investigation for a temporal trend, the study was restricted to men aged 40-50 years, to limit variation of body burden

with gender and age. About 0.5-1 mL serum from each individual was pooled and stored at -20 °C. Serum sampled in 1998 was chosen for studying the current concentration of BFRs in different age groups. In the serum bank, the samples had originally been divided into eight age groups, and samples from between 10 and 14 individuals of each age group were pooled. The lipid content of the pooled serum samples was determined at The National Hospital of Norway (Oslo, Norway) according to a method described by Grimvall et al. (43). Details concerning fat content and number of individuals represented in each pool are summarized in Table 1. Sample Preparation. The frozen serum samples were thawed in a refrigerator (4 °C) and brought to room temperature before 5.0 g of each sample was added the following amounts of internal standards 8.4 pg of TBCr, 23.3 pg of BDE-77, 46.5 pg of BB-77, and 26.3 pg of CtriBBP-A were dissolved in 30 µL of ethyl acetate. The samples were kept overnight in a refrigerator and subsequently extracted according to a previously described method (44). In brief, 5.0 g of serum was diluted by formic acid, 2-propanol, and water and applied to a prewashed and conditioned solidphase extraction (SPE) column of cross-linked polystyrenedivinylbenzene material (Isolute ENV+, (200 mg) from International Sorbent Technology, Mid Glamorgan, U.K.). The lipids were decomposed by treatment with concentrated sulfuric acid directly on the SPE column. The sorbent was completely dried using nitrogen (99.999%, Aga, Oslo, Norway) prior to elution of the BFRs by 6 mL of dichloromethanemethanol (7 + 3, v/v). The extracts were concentrated under a gentle stream of nitrogen at 50 °C to about 70 µL, 15 µL of reference standard (TriBB) was added, and the mixture was concentrated further to about 50 µL. TriBP and TBBP-A were methylated by addition of 30 µL of diazomethane as previously described (45). The derivatization is quantitative, and the methyl ethers show a better chromatographic performance than the phenolic compounds. The presence of native methylated derivatives of the phenols, i.e., 1,3,5tribromo-2-methoxybenzene and tetrabromobisphenol A dimethyl ether, was controlled by analysis prior to derivatization. All solutions were stored in amber glass vials at -18 °C. GC-MS Instrumentation. The chromatographic separation was performed on an HP (Avondale, PA) 6890 gas chromatograph equipped with an HP 7683 automatic liquid sampler. The column, a CP-Sil 5 CB fused silica capillary column (30 m × 0.25 mm i.d., 0.25 µm film thickness, Chrompack, Middelburg, The Netherlands) was connected to the injector via a deactivated retention gap of 1.5 m × 0.32 mm i.d. fused silica (J & W Scientific, Folsom, CA). Samples VOL. 36, NO. 7, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1415

of 1 µL were injected in pulsed splitless mode with a pulse pressure of 1.72 bar for 1.5 min. The injector temperature was 250 °C. Helium (99.998%, Aga) was used as a carrier gas, and the flow was held constant at 1.2 mL/min. The column temperature was initially 90 °C for 1 min and then raised by 20 °C/min to 250 °C, 10 °C/min to 300 °C, and 30 °C/min to 325 °C, which was held for 3 min. The mass spectrometer, an HP 5973 MSD with chemical ionization option, was operated in the electron capture mode with methane (99.99%, Aga) as buffer gas. The temperatures were 106 °C, 250 °C, and 300 °C for the quadrupole, the ion source, and the interface, respectively, and an electron energy of 235 eV was applied. Identification and Quantification. The analytes were identified by comparing their retention time with the respective retention time of the calibration standards. The brominated compounds were monitored at m/z 79;81, and their identity was tentatively confirmed by controlling the isotope abundance ratio. The PCB CB-153, a well-studied biomarker of exposure to persistent organochlorine compounds, was detected at m/z 360. The sample extracts were also analyzed using a second temperature program with a slower temperature ramp, to resolve possible overlapping peaks. Due to a previously observed matrix effect (44), the GCMS calibration solutions were prepared by adding internal standard and standard solutions to concentrated extracts of 5.0 g of fetal bovine serum from Sigma (St. Louis, MO), followed by further concentration and derivatization. The final volume of the calibration solutions was comparable to the volume of the sample extracts. As an exception, TriBP was quantified using calibration standards in ethyl acetate, due to a relatively high native content of TriBP in the fetal bovine serum. In addition, the matrix effect was less pronounced for this compound. The calibration concentration range for TriBP was 0.008-0.122 pg/µL ethyl acetate (n ) 4) and 1.8-27 pg/g serum for all other compounds (n ) 4). The BFRs were quantified using area ratios and an internal standard calibration. BB-77 was used as an internal standard for quantification of BDE-47, BDE-99, BDE-100, BDE153 and BDE-154, and BDE-77 for BDE-28. CtriBBP-A and TBCr were used as an internal standard for TBBP-A and TriBP, respectively. The uncertainty of the analysis was previously found to be about 20% (44). The calibration curves were corrected for the native content of BFRs in the fetal bovine serum used for the calibration solutions. Laboratory air has previously been shown to contain low levels of BFRs (46) ,and procedural blanks were included in each SPE-series. BDE-47 and TriBP were occasionally detected, but the level never exceeded 10% of the amount of BDE-47 and TriBP in the analyzed serum samples, and the contribution from the procedural blanks were thus considered insignificant in this study. The serum samples had been stored in two different types of plastic tubes. Tubes of these brands were extracted using isooctane or ethyl acetate in search for possible contaminants. However, none of the investigated BFRs were detected in the tube extracts. The signal-to-noise ratio (S/N) of the analytes in the calibration solution containing 1.8 pg BFR/g serum (except for TriBP) was used for an estimation, by extrapolation, of the quantification limits (LOQ, S/N ) 3). The estimations concerning TriBP was based on the native content in the fetal bovine serum. The LOQs were in the range of 0.4-1.6 pg/g serum both for TriBP, TBBP-A, and the PBDEs. In cases where compounds were identified at a level below the quantification limit, the calculated value is still reported in Table 1, according to the guidelines given by the Analytical Methods Committee (47). 1416

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 7, 2002

FIGURE 1. The concentration of TBBP-A and the total concentration of the six PBDE congeners in ng/g lipids in pooled serum samples from 40 to 50 year old Norwegian men, sampled in 1977, 1981, 1986, 1990, 1995, and 1999. An exponential trend line is shown for the sum of the PBDEs.

Results The method for determination of BFRs in serum samples applied in this study has recently been validated (44) and found satisfactory. The quality of the quantification in the present study was controlled by assessing the mean recovery of the internal standards, which were 56, 69, and 25% for BDE-77, BB-77, and CtriBBP-A, respectively. These recoveries were in close agreement with the results from the validation, except for CtriBBP-A (44). The phenolic compounds are strongly bound to the SPE material. In the validation experiments, fetal bovine serum was used, and the lower recovery of CtriBBP-A observed in the present study could be due to a reduced release from the sorbent during elution, after application of human serum. The concentrations of BFRs in the different serum samples, and additional information concerning the serum pools are given in Table 1. To permit adequate comparisons between compounds of highly variable molar masses, the concentrations are also given on a molar basis in parentheses in Table 1. The heptabrominated diphenyl ether BDE-183 and PeBP were not observed above the detection limit in any of the samples. The congeners BDE-47, BDE-99, and BDE153 were found in all samples, while BDE-28, BDE-100, BDE-154, and TBBP-A were found in the most recent pools (Table 1). In all time periods BDE-47 was the dominant congener. Concerning the pools from men at age 40-50 years, the serum concentration of all the BFRs, except TriBP, increased during the entire period, with the greatest changes occurring between 1986 and 1995 (Table 1). The sum of the six PBDEs increased from 0.44 ng/g lipids in 1977 to 3.3 ng/g lipids in 1999 (Figure 1). To get a more mathematical depiction on the development of body burden of PBDEs during these decades, the six points were subjected to trend analysis. The best curve fit was obtained with an exponential trend line, which resulted in an R-value of 0.966 (Figure 1). With the exception of TriBP, the BFR concentrations in the serum from the group aged 0-4 years were 1.6-3.5 times higher than those of the other age groups, which on the other hand were relatively similar (Figure 2). Also, for the group 25 years and older, women seemed to have lower serum concentrations of BFRs compared to the corresponding group of men. The serum levels of the polychlorinated biphenyl CB-153 decreased in the samples from 1977 to 1999 and were considerably higher in the older subjects compared to the younger ones for the 1998 samples (results not shown). The

FIGURE 2. The concentration of TBBP-A and the individual PBDE congeners in ng/g lipids in pooled serum from different age groups of Norwegian male (m) and/or female (f) subjects sampled in 1998. relative amount of CB-153 in the serum pools was established using TriBB as reference standard. The sample extracts were also analyzed prior to derivatization in search for native methylated derivatives of the phenols. 1,3,5-Tribromo-2-methoxybenzene was detected in all the samples; however, its concentration was less than 10% of that of TriBP and was therefore neglected. Tetrabromobisphenol A dimethyl ether was only observed in the serum pool of the 0-4 year olds at a level close to the detection limit.

Discussion In general populations, the body burdens of POPs such as PCDDs, PCDFs, or PCBs, are known to increase during life (41, 42). On the other hand, decreasing levels have been observed both in human milk (26) and blood (41) during the recent decades, due to restrictions in use and/or measures made to reduce the emission of such compounds to the environment. The relative levels of CB-153 in the present serum pools were consistent with these time and age trends, which implies that the pools can be regarded reliable and representative for the general population. However, a similar age trend was not observed for any of the BFRs, where the serum concentration seemed to be relatively constant for the different age groups above 4 years (Figure 2). This is in accordance with studies on BFR concentrations in human adipose tissue (24), breast milk (28), and plasma (48) where no age related trends were observed either. The absence of an age dependency of BFR body burdens may be explained by the fact that BFRs, in contrast to PCBs and other POPs, are relatively new contaminants in the environment. Thereby, also the time period for human exposure is relatively short which is demonstrated by the rather recent increase in PBDE blood levels (see Figure 1). Different age groups will thus have experienced a similar lifetime exposure. BDE-47 was by far the most abundant congener in all the serum samples from 1998 and represented between 44 and 65% of the total concentration of PBDEs on weight basis (mean 52%). This is in accordance with the congener patterns reported in human whole blood (19), plasma (17), and milk (26). When excluding the age group 0-4 years, the current concentration of BDE-47 in serum from the general population in Norway was in the range of 1.2-3.4 ng/g lipids. This is comparable to previously reported BDE-47 concentrations in human blood samples which were 1.6 ng/g lipids (median, n ) 19, 1997, Sweden), 0.63 ng/g lipids (median, n ) 12, 1988, U.S.A.) (18), and 3.0 ng/g lipids (median, n ) 20, 1999, Germany) (19). In the present study, the serum concentration of the PBDEs, except for BDE-28, were higher in the male pool, compared to female pool, for people 25 years and older. Lower PBDE levels in women compared to men have

previously been reported in adipose tissue (23) and whole blood (19). So far, these findings have not been explained; however, for the youngest group of women in this study, the reduction in body burden of the persistent BFRs might be related to breast-feeding. The most remarkable observation concerning the samples from 1998 was the unambiguously elevated concentration of BFRs in the serum pool from the youngest individuals (except for TriBP). As previously stated, food is regarded as the major source for human exposure to BFRs. Based on Swedish data on PBDE levels in market basket samples from 1999, the total PBDE intake was estimated to 51 ng/day (8). The presence of PBDEs in breast milk has been shown to increase exponentially since 1972, and the intake for infants corresponding to consumption of 0.7 L of milk daily was calculated to 110 ng/day based on PBDE concentrations in milk from the late 1990s (8). Accordingly, a recent probabilistic analysis of chronic daily intake (CDI) of pentaBDE, based on eight exposure pathways, showed a clear decrease in CDI during life (breast milk was included in the age group of 0-2 years old) (49). Both estimates suggest an approximately twice as high intake of BFRs in infants compared to adults. This is strongly supported by the findings in the present study, where the total PBDE concentration in the serum from the youngest group on average was 2.8 times higher than that of the other age groups. Even though breast milk is considered to be the main source for infants’ exposure to BFRs in Norway, also prenatal exposure might contribute to the elevated BFR levels in infants. PBDEs have been found in placenta (27), and transplacental transfer might expose the fetus to the BFRs. The total PBDE concentration in pooled serum samples from Norwegian men of age 40-50 years increased from 0.44 ng/g lipids in 1977 to 3.3 ng/g lipids in 1999 (Figure 1). Accordingly, an increase from 0.07 ng/g lipids in 1972 to 4.0 ng/g lipids in 1997 has been shown in Swedish human milk (26). Furtermore, the total PBDE concentrations increased from 3.9 ng/g lipids in 1985 to 5.6 ng/g lipids 1999 in German whole blood (19). It is reasonable to believe that the increasing BFR levels observed in these samples reflect the enhanced use of products containing BFRs. On the contrary, the concentrations of other POPs, such as p,p′-DDT, PCBs, and dioxins, in human samples have been shown to decrease during the past decades (26, 50). The phenolic flame retardant TBBP-A is industrially the most important individual BFR used; however, reports on TBBP-A in human samples and in the environment in general are scarce. Concentrations in plasma in the low ng/g lipids have been reported (16, 20), which is in accordance with this study where serum levels of TBBP-A ranged from 0.34 to 0.71 ng/g lipids. TBBP-A was not identified in the serum pools from 1977 and 1981, but a slight increase in serum concentration was observed from 1986 to 1999 (Figure 1). As for the other BFRs, the highest serum concentration of TBBP-A was found in the serum pool from the group aged 0-4 years. Concerning the serum concentrations of TriBP, no trend related to age or increase during the period 1977 to 1999 was observed. Compared to the other BFRs, the concentration range of TriBP in the different pools was rather wide, ranging from 0.077 to 26 ng/g lipids. TriBP has previously been quantified in individual plasma samples in the range from 0.17 to 81 ng/g lipids (16). Only limited knowledge concerning the sources of human exposure to TriBP exists. Besides the apparent possibility of leakage from flame retarded materials, TriBP has been reported in exhaust from cars using leaded fuel (51) and is also formed naturally in the marine environment (29). Our findings indicate an exponential increase in human exposure to BFRs during the past decades in the general population in Norway, which is consistent with investigations VOL. 36, NO. 7, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1417

in other European countries (19, 26). Even though the rate of increase in these studies are slightly different and the present BFR levels still in human tissues 10-100 times lower than that of PCBs, human body burden of BFRs should be carefully attended in the forthcoming years. The present study also suggests that the current body burden is independent of age, except for infants 0-4 years old, who seem to experience elevated exposures. This shows the importance of further monitoring of BFRs in human samples such as breast milk and blood and investigations concerning possible health effects as a consequence of BFR exposure.

Acknowledgments We are thankful to the Department of Virology at the National Institute of Public Health for providing the serum samples, Wellington Laboratories for providing the PBDE standards and the Research Council of Norway for financial support.

Literature Cited (1) Environmental Health Criteria 192. Flame Retardants: A General Introduction; World Health Organization: Geneva, Switzerland, 1997. (2) Environmental Health Criteria 162. Brominated Diphenyl Ethers; World Health Organization: Geneva, Switzerland, 1994. (3) Environmental Health Criteria 172. Tetrabromobisphenol A and derivatives; World Health Organization: Geneva, Switzerland, 1995. (4) BSEF. An introduction to Brominated Flame Retardants; Bromine Science and Environmental Forum: Brussels, Belgium. http:// www.bsef.com. (accessed Oct 2000). (5) Hooper, K.; McDonald, T. A. Environ. Health Perspect. 2000, 108, 387-392. (6) de Boer, J.; de Boer, K.; Boon, J. P. In The Handbook of Environmental Chemistry; Paasivirta, J., Ed.; Springer-Verlag: Berlin, 2000; Vol. 3, Part K, pp 62-95. (7) de Wit, C. Brominated Flame Retardants; Report 5065; Swedish Environmental Protection Agency: Stockholm, Sweden, 2000. (8) Darnerud, P. O.; Eriksen, G. S.; Jo´hannesson, T.; Larsen, P. B.; Viluksela, M. Environ. Health Perspect. 2001, 109, 49-68. (9) Ghosh, M.; Meerts, I. A. T. M.; Cook, A.; Bergman, Å.; Brouwer, A.; Johnson, L. N. Acta Crystallogr. 2000, 56, 1085-1095. (10) Meerts, I. A. T. M.; Marsh, G.; Van Leeuwen-Bol, I.; Luijks, E. A. C.; Jakobsson, E.; Bergman, Å.; Brouwer, Organohalogen Compd. 1998, 37, 309-312. (11) Meerts, I. A. T. M.; Van Zanden, J. J.; Luijks, E. A. C.; Van LeeuwenBol, I.; Marsh, G.; Jakobsson, E.; Bergman, Å.; Brouwer, A. Toxicol. Sci. 2000, 56, 95-104. (12) Meerts, I. A. T. M.; Letcher, R. J.; Hoving, S.; Marsh, G.; Lemmen, J. G.; Van der Burg, B.; Brouwer, A. Environ. Health Perspect. 2001, 109, 399-407. (13) Burreau, S.; Broman, D.; Zebuhr, Y. Organohalogen Compd. 1999, 40, 363-366. (14) Sellstro¨m, U.; Jansson, B.; Kierkegaard, A.; de Wit, C. Chemosphere 1993, 26, 1703-1718. (15) Haglund, P.; Zook, D. R.; Buser, H. R.; Hu, J. Environ. Sci. Technol. 1997, 31, 3281-3287. (16) Thomsen, C.; Lundanes, E.; Becher, G. J. Environ. Monit. 2001, 3, 366-370. (17) Sjo¨din, A.; Hagmar, L.; Klasson-Wehler, E.; Kronholm-Diab, K.; Jakobsson, E.; Bergman, Å. Environ. Health Perspect. 1999, 107, 643-648. (18) Patterson, D. G., Jr.; Sjo¨din, A.; Bergman, Å. Environ. Sci. Technol. 2001, 35, 3830-3833. (19) Schro¨ter-Kermani, C.; Helm, D.; Herrmann, T.; Pa¨pke, O. Organohalogen Compd. 2000, 47, 49-52. (20) Sjo¨din, A. Ph.D. Dissertation, Department of Environmental Chemistry, Stockholm University, 2000. (21) Stanley, J. S.; Cramer, P. H.; Thornburg, K. H.; Remmers, J. C.; Breen, J. J.; Schwemberger, J. Chemosphere 1991, 23, 11851195.

1418

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 7, 2002

(22) Lindstro¨m, G.; Hardell, L.; van Bavel, B.; Wingfors, H.; Sundelin, E.; Liljegren, G.; Lindholm, P. Organohalogen Compd. 1998, 35, 431-434. (23) Meneses, M.; Wingfors, H.; Schumacher, M.; Domingo, J. L.; Lindstro¨m, G.; van Bavel, B. Chemosphere 1999, 39, 2271-2278. (24) Strandman, T.; Koistinen, J.; Kivaranta, H.; Vuorinen, P. J.; Tuomisto, J.; Tuomisto, J.; Vartiainen, T. Organohalogen Compd. 1999, 40, 355-358. (25) Ryan, J. J.; Patry, B. Organohalogen Compd. 2000, 47, 57-60. (26) Nore´n, K.; Meironyte, D. Chemosphere 2000, 40, 1111-1123. (27) Strandman, T.; Koistinen, J.; Vartiainen, T. Organohalogen Compd. 2000, 47, 61-64. (28) Darnerud, P. O.; Atuma, S.; Aune, M.; Cnattingius, S.; Wernroth, M. L.; Wicklund-Glynn, A. Organohalogen Compd. 1998, 35, 411-414. (29) Gribble, G. W. Chem. Soc. Rev. 1999, 28, 335-346. (30) Carte´, B.; Faulkner, D. J. Tetrahedron 1981, 37, 2335-2339. (31) Handayani, D.; Edrada, R. A.; Proksch, P.; Wray, V.; Witte, L.; Van Soest, R. W. M.; Kunzmann, A.; Soedarsono, J. Nat. Prod. 1997, 60, 1313-1316. (32) Fu, X.; Schmitz, F. J.; Govindan, M.; Abbas, S. A.; Hanson, K. M.; Horton, P. A.; Crews, P.; Laney, M.; Schatzman, R. C. J. Nat. Prod. 1995, 58, 1384-1391. (33) Fu, X.; Schmitz, F. J. J. Nat. Prod. 1996, 59, 1102-1103. (34) Nylund, K.; Asplund, L.; Jansson, B.; Jonsson, P.; Litze´n, K. Sellstro¨m, U. Chemosphere 1992, 24, 1721-1730. (35) Kierkegaard, A.; Sellstro¨m, U.; Bignert, A.; Olsson, M.; Asplund, L.; Jansson, B.; de Wit, C. Organohalogen Compd. 1999, 40, 367370. (36) Alaee, M.; Luross, J.; Sergeant, D. B.; Muir, D. C. G.; Whittle, D. M.; Solomon, K. Organohalogen Compd. 1999, 40, 347-350. (37) Stern, G. A.; Ikonomou, M. G. Organohalogen Compd. 2000, 47, 81-84. (38) Ikonomou, M. G.; Fischer, M.; He, T.; Addison, R. F.; Smith, T. Organohalogen Compd. 2000, 47, 77-81. (39) Zegers, B. N.; Lewis; W. E.; Boon, J. P. Organohalogen Compd. 2000, 47, 229-232. (40) Guvenius, D. M.; Nore´n, K. Proceedings from The Second International Workshop on Brominated Flame Retardants; Stockholm, Sweden, 2001; pp 303-305. (41) Pa¨pke, O. Environ. Health Perspect. 1998, 106, 723-731. (42) Rylander, L.; Dyremark, E.; Stro¨mberg, U.; O ¨ stman, C.; Hagmar, L. Sci. Total Environ. 1997, 207, 55-61. (43) Grimvall, E.; Rylander, L.; Nilsson-Ehle, P.; Nilsson, U.; Stro ¨ mberg, U.; Hagmar, L.; O ¨ stman, C. Arch. Environ. Contam. Toxicol. 1997, 32, 329-336. (44) Thomsen, C.; Lundanes, E.; Becher, G. J. Sep. Sci. 2001, 24, 282290. (45) Thomsen, C.; Jana´k, K.; Lundanes, E.; Becher, G. J. Chromatogr. B 2001, 750, 1-11. (46) Thomsen, C.; Leknes, H.; Lundanes, E.; Becher, G. J. Chromatogr. A 2001, 923, 299-304. (47) Analytical Methods Committee. Analyst 2001, 126, 256-259. (48) Sjo¨din, A.; Hagmar, L. Klasson-Wehler, E.; Bjo¨rk, J.; Bergman, Å. Environ. Health Perspect. 2000, 108, 1035-1041. (49) Wenning, R. J. Proceedings from The Second International Workshop on Brominated Flame Retardants; Stockholm, Sweden, 2001; pp 397-400. (50) Ward, E. M.; Schulte, P.; Grajewski, B.; Andersen, A.; Patterson, D. G., Jr.; Turner, W.; Jellum, E.; Deddens, J. A.; Friedland, J.; Roeleveld, N.; Waters, M.; Butler, M. A.; DePietro, E.; Needham, L. L. Cancer Epidemiol. Biomark. Prev. 2000, 9, 1357-1367. (51) Buser, H. R. Anal. Chem. 1986, 58, 2919-2923.

Received for review September 6, 2001. Revised manuscript received January 7, 2002. Accepted January 28, 2002. ES0102282