Environ. Sci. Technol. 2007, 41, 6961-6968
Distribution of Polybrominated Diphenyl Ethers in Human Umbilical Cord Serum, Paternal Serum, Maternal Serum, Placentas, and Breast Milk from Madrid Population, Spain B. GO Ä MARA,† L. HERRERO,† J. J. RAMOS,† J . R . M A T E O , † M . A . F E R N AÄ N D E Z , † J . F . G A R C ´ı A , ‡ A N D M . J . G O N Z AÄ L E Z * , † Department of Instrumental Analysis and Environmental Chemistry, Institute of General Organic Chemistry, CSIC, Juan de la Cierva, 3, 28006-Madrid, Spain, and Environmental Health Service, Regional Health Ministry of Madrid Community, Julian Camarillo, 4B, 28037 Madrid, Spain
Median concentration of total PBDEs in maternal serum, paternal serum, umbilical cord serum, and breast milk samples were 12, 12, 17, and 6.1 ng/g lipid weight (lw) in Vallecas and 9.7, 12, 15, and 5.5 ng/g lw in Getafe. The median value found in placentas was 1.9 ng/g lw (in Vallecas). BDE 47 was the predominant congener in serum samples (maternal, paternal, and umbilical cord), while BDE 209 was predominant in placenta and breast milk samples. BDEs 196 and 197 were detected in most of the placenta and breast milk samples. The results show that PBDEs, like other POPs, can cross the placenta barrier, although the speed of the process seems to differ for each PBDE congeners. The total PBDE concentrations found in this study are consistent with research reported elsewhere. They are in the same range as those recently reported by other European and Asian studies and lower than those conducted in the U.S.A. No significant differences were found (p > 0.05) between regions, sexes, and ages, while statistically significant differences (p < 0.05) were found between maternal serum, umbilical cord serum, and breast milk samples. The presence of PBDEs in cord blood and placenta samples indicates that there is prenatal exposure of PBDEs, which could continue after birth via breast milk.
1. Introduction Polybrominated diphenyl ethers (PBDEs) are flame retardants, which are used as additives in electronic appliances, paints, textiles, and furniture to prevent the outbreak of fire (1). PBDEs migrate from the products in which they are used and enter the environment. In recent years, there has been a marked increase of their levels in human tissues all over the world. In Swedish human milk, PBDE concentrations increased exponentially from 1972 to 1997 (2), and similar trends have been observed in human samples from Japan, * Corresponding author phone: +34 91 5622900; fax: + 34 91 564485; e-mail:
[email protected]. † Institute of General Organic Chemistry. ‡ Regional Health Ministry of Madrid Community. 10.1021/es0714484 CCC: $37.00 Published on Web 09/12/2007
2007 American Chemical Society
Norway, Canada, and North America (3-6). The main reasons for this are as follows: a dramatic increase in their production and use starting in the 1980s (7), added to their persistence and their lipophilic character, which leads to their concentration in food chains and hence accumulation in the human body (8). In view of that, the European Community (EC) introduced a new regulation to reduce the use of penta- and octa-BDE technical mixtures (Directive 2003/11/EC) (9). Their structural similarity to PCBs, which are known to have neurotoxic and carcinogenic action, begs the question of potential biological hazards associated with PBDEs. The consumption of the different PBDE commercial mixtures available in European countries was estimated to be 150 metric tons of penta-, 400 metric tons of octa-, and 7000 metric tons of deca-BDE technical products (7), thus making the higher brominated congeners, especially BDE 209, a matter of special concern in European countries. However, most of the data available in the literature focus on tri- to hexa-BDE congeners, mainly BDEs 47, 99, 100, and 153, while little information is available on BDEs 183 and 209 (the major components of the commercial flame retardant mixtures octa-BDE and deca-BDE, respectively). In the same way, references to the concentrations of impurities in technical formulations and/or degradation products of BDE 209, such as BDEs 184, 191, 196, and 197 (10-11) are very scarce in the literature. Human samples like serum, breast milk, and placenta are nondestructive matrices adequate for monitoring human exposure to PBDEs indicating both parent and neonate body burdens. Although much is known about organochlorine contaminants in human tissues like serum and breast milk (12, 13), information regarding PBDE concentrations, especially those of high-brominated congeners in these kinds of samples, is still scarce in the literature. Regarding serum samples, only the most recent studies analyze the higher brominated congeners, including BDE 209 (3, 14-18). As for PBDE concentrations in human breast milk samples, most studies fail to include high brominated congeners (19-25), and only some report BDE 209 concentrations (26-30). The main objective of this study was to report, for the first time, the levels and accumulation profiles of tri- to decaBDEs in serum (maternal, paternal, and umbilical cord), placenta, and breast milk samples in the population living in two different areas of Madrid (Spain) and compare the results with previous studies in other countries. In addition, statistical comparisons were run between PBDE concentrations found in both areas and between the five studied matrices. The study further addressed the correlation of PBDE levels in maternal and paternal serum with age and sex, the ability of PBDE congeners to cross the placenta barrier, and prenatal and postnatal exposure to PBDEs.
2. Material and Methods 2.1. Study Design and Sample Collection. The study design and the sampling collection were conducted by the Public Health authorities of the community of Madrid and the Institute of Health Carlos III (Madrid, Spain). A total of 391 individual samples, including 113 maternal serum samples, 104 paternal serums, 92 umbilical cord serums, 30 placentas, and 52 breast milk samples were collected between October 2003 and May 2004, from volunteers living in two areas of the community of Madrid (Spain). Among all the samples collected, in 30 cases, triads consisted of maternal, paternal, and umbilical cord serum, then in 22 cases, triads consisted of maternal serum, placenta, and umbilical cord serum, and VOL. 41, NO. 20, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6961
then in 25 cases, triads consisted of maternal serum, umbilical cord serum, and breast milk samples. Only in seven cases, combinations of maternal serum, placenta, milk, and umbilical cord serum were obtained. Half of the samples were recruited from an urban district of Madrid City (Vallecas), while the other half was selected from its metropolitan belt (Getafe). Maternal and paternal blood samples from volunteers were obtained when the mothers were admitted to childbirth preparation classes (around 8 months of pregnancy) in the public health system, and umbilical cord blood was taken from the umbilical cord vein by syringe just after childbirth. Placentas were obtained after delivery, and breast milk samples were obtained 3 weeks after delivery. The goals and requirements of the study were explained to all participants. Donors were asked to participate and to sign a consent form. Pregnancies were full-term, and no medical problems were detected during pregnancy. All mothers were healthy and primiparas and older than 15 years. The mean age of parents was 30.5. None of the donors reported any work related potential for exposure to PBDEs. Once at the laboratory, serum samples were frozen at -20 °C, and breast milk and placenta samples were freeze-dried and stored at room temperature until analysis. The sample collection was approved by the local committee of medical ethics. 2.2. Reagents and Standards. All reagents used for the analysis of PBDEs were of trace analysis grade. n-Hexane, sulfuric acid (95-97%), acid formic, and silica gel were supplied by Merck Co. (Darmstadt, Germany), and granular anhydrous sodium sulfate was supplied by J.T. Baker (Deventer, The Netherlands). Acetone and toluene were supplied by SDS (Peypin, France). All PBDE standards were purchased from Wellington Laboratories (Ontario, Canada). A total of 15 PBDE congeners from tri- to deca-BDEs were analyzed. These were the most abundant in both, technical mixtures (penta-, octa-, and decabrominates) and environmental matrices, BDEs 17 (2,2′,4-tri-BDE), 28 (2,4,4′-tri-BDE), 47 (2,2′,4,4′-tetra-BDE), 66 (2,3′,4,4′-tetra-BDE), 85 (2,2′,3,4,4′penta-BDE), 99 (2,2′,4,4′,5-penta-BDE), 100 (2,2′4,4′,6-pentaBDE), 153 (2,2′,4,4′,5,5′-hexa-BDE), 154 (2,2′,4,4′,5,6′-hexaBDE),183(2,2′,3,4,4′,5′,6-hepta-BDE),and209(2,2′,3,3′,4,4′,5,5′,6,6′deca-BDE). Some other congeners, such as BDEs 184 (2,2′, 3,4,4′,6,6′-hepta-BDE), 191 (2,3,3′,4,4′,5′,6-hepta-BDE), 196 (2,2′,3,3′,4,4′,5′,6-octa-BDE), and 197 (2,2′,3,3′,4,4′,6,6′-octaBDE), which are impurities of PBDE formulations and formed during the reductive debromination of PBDE 209, were also analyzed. One 13C12-labeled standard, PBDE 139, was used as an injection standard for the quantification of samples. 2.3. Sample Treatment. Serum samples were extracted and purified using a semiautomated solid-phase extraction and cleanup method described elsewhere (31). Briefly, it consisted of a solid-phase extraction (SPE) of 0.5-1 mL of serum sample, previously treated with 1 mL of formic acid and 50 µL of acetonitrile, using Oasis HLB cartridges (Waters, Mildford, U.S.A). After loading of the sample into the cartridge and drying of the cartridge, the analytes eluted using 4 mL of toluene were directly passed through a multilayer column filled with 0.06 g of silica gel activated at 140 °C for 48 h and 0.24 g of silica gel activated and modified with sulfuric acid (44%, w/w) and anhydrous sodium sulfate. This eluent was concentrated under nitrogen up to an appropriate volume for its analysis by GC-MS. A variation of the method previously published for food sample analyses (32) was used for breast milk and placenta samples by reducing the solvent and adsorbent amounts. Briefly, this consisted of matrix solid-phase dispersion (MSPD) of 0.5 g of freeze-dried breast milk and 2.0 g of 6962
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 20, 2007
placenta. Further cleanup and lipid removal was performed by using acid and basic impregnated silica gel multilayer columns, and n-hexane was used as elution solvent. 2.4. Instrumental Determination of PBDEs Using GCECNI-MS. The 15 PBDEs selected, including from tri- to decasubstituted congeners, were determined using a 6890N gas chromatograph coupled with a 5975 quadrupole mass spectrometer (Agilent, Palo Alto, CA) working in the electron capture negative ionization mode (ECNI). Standards and samples were injected in hot splitless mode (300 °C, 1 µL; splitless time 2.0 min). A low bleed GC capillary column DB5MS (15 m, 0.2 mm i.d., 0.2 µm film thickness) purchased from J&W Scientific (U.S.A.) was used for separation. All GC working conditions and the quantification method used are detailed previously (33). 2.5. Quality Control Criteria. All PBDE analyses such as blanks, recoveries, and parallel analyses complied with analytical standards as recommended by the EU Commission in the directive for measuring dioxins in food (34) (European Commission, 2001). Special attention was paid to blanks and BDE 209, due to the high levels of this BDE in house dust. Satisfactory repeatability and intermediate precision were achieved when analyzing standard solutions; relative standard deviations (RSDs) were below 6 and 11%, respectively, for all congeners investigated, except BDE 209, whose RSD were 9% and 19%, respectively. Method detection limits were in the range 6-507 fg injected for standard solutions. Analyte recoveries were higher than 65% in serum and breast milk samples (31, 33). The accuracy of the whole method (sample preparation plus instrumental determination) was checked by participating in several international quality control studies for the analysis of PBDEs in different biological matrices, including serum and breast milk (35, 36). The results were consistent at all times with the consensus medians found in interlaboratory studies. 2.6. Statistical Analysis. In most cases the distribution of data was highly skewed. The variables did not follow a normal distribution (Shapiro-Wilk test), and nonparametric tests were used for statistical comparisons. The data set was analyzed by nonparametric Wilcoxon’s test to determine significant differences among the PBDE contents in human samples. Nonparametric correlation (Spearman) was used to compare PBDE levels between the different human samples. For statistical comparison, the nondetectable (ND) concentrations were taken to be half of the detection limits (LOD). Differences with p < 0.05 were considered statistically significant. The Statgraphic statistical package (version 4.0, STSC Inc., Rockville, MD) was used for the calculations.
3. Results and Discussion Concentrations (median and range) of 15 individual PBDE congeners as well as total PBDEs in the human samples in the two Madrid population areas studied are shown in Tables 1 and 2, expressed in ng/g lw (in the middle bound determination limit, assuming that nondetected values are equal to half of their corresponding limit of detection). Each sample was analyzed separately. 3.1. PBDE Levels in Different Human Samples from Both areas. Taking into account the results of both areas (Tables 1 and 2), BDEs 17, 28, 66, 154, 184, and 191 were not detected in hardly any sample, while BDEs 47, 99, and 100 were detected in most of them. BDEs 85, 153, 183, and 209 were detected in around 60% of the samples, and BDEs 196 and 197 were detected in around 30%, mainly in placentas and breast milk. The frequency of detection of tri- to heptaBDEs was similar in all the samples, while the frequency of detection of octa- to deca-BDEs was higher in umbilical cord serum, placenta, and breast milk than in paternal and maternal serum samples. The highest contribution of
TABLE 1. PBDE Median Concentrations and Range (Indicated in Parentheses) in Serum, Breast Milk, and Placenta Samples from Volunteers (Vallecas District) Living in the Community of Madrid (Spain)a maternal serum (n ) 61)
paternal serum (n ) 51)
