Environ. Sci. Technol. 2008, 42, 3453–3458
Serum Levels of Polybrominated Diphenyl Ethers (PBDEs) in Foam Recyclers and Carpet Installers Working in the United States H E A T H E R M . S T A P L E T O N , * ,† ANDREAS SJÖDIN,‡ RICHARD S. JONES,‡ SARA NIEHÜSER,‡ YALIN ZHANG,‡ AND DONALD G. PATTERSON, JR.‡ Nicholas School of the Environment & Earth Sciences, Duke University, LSRC Box 90328, Durham, North Carolina 27708, and Center for Disease Control and Prevention, National Center for Environmental Health, Division of Laboratory Sciences, Organic Analytical Toxicology Branch, 4770 Buford Highway, Mail Stop F-17, Atlanta, Georgia 30341-3717
Received November 16, 2007. Revised manuscript received January 22, 2008. Accepted February 1, 2008.
Increased exposure to the flame retardants known as polybrominated diphenyl ethers (PBDEs) may be expected to occur during the recycling of polyurethane foam containing these chemicals. To date, no studies in the United States have investigated occupational exposure to these flame retardants during recycling processes. The objective of the present study was to determine if individuals working in foam recycling facilities, and/or carpet installers who may install carpet padding manufactured from recycled foam, possess significantly higher PBDE serum levels relative to that of the general U.S. population. As a control group, serum was collected from four spouses and one clerical worker. In addition, levels in workers were also compared to the recently published national health and nutrition examination survey (NHANES) data set on PBDEs in the general U.S. population. Serum samples were collected in duplicate and analyzed by two different laboratories as quality control. Total PBDE levels were found to be significantly higher (p < 0.05) in the individuals recycling foam and installing carpet (n ) 15) relative to the control group (n ) 5). Median ΣPBDE levels in the foam recyclers, carpet layers, and control group were 160, 178, and 19 ng/g lipid, respectively. In contrast, concentrations of a polybrominated biphenyl (BB153) and a polychlorinated biphenyl (CB-153) were equivalent among all groups tested. The PBDE congeners BDE-47, 99, 100, and 153 contributed 90% of the ΣPBDE concentration in serum and no differences in congener patterns were apparent among the different groups. Relative to concentrations measured in the NHANES, foam recyclers and carpet layers have body burdens that are an order of magnitude higher. These datasuggestindividualsrecyclingfoam-containingproducts,and/ or using products manufactured from recycled foam (i.e., carpet padding), have higher body burdens of PBDEs, and thus may be at higher risk from adverse health effects associated with brominated flame retardant exposure. * Corresponding author e-mail:
[email protected]; phone: (919) 613-8717; fax: (919) 684-8741. † Duke University. ‡ Center for Disease Control and Prevention. 10.1021/es7028813 CCC: $40.75
Published on Web 03/18/2008
2008 American Chemical Society
Introduction Polybrominated diphenyl ethers (PBDEs) are a group of flame retardant chemicals added to a variety of resins and polymers to decrease their flammability. Typical polymers and resins containing these chemicals include high-impact polystyrene (HIPS), acrylonitrile-butadiene styrene (ABS), and polyurethane foam. These treated polymers are then used to manufacture consumer products and include electronic items such as TVs, DVD players, and cellular phones, and most foam-containing furniture. Three different PBDE commercial formulations were manufactured and marketed for different polymer applications: PentaBDE, OctaBDE, and DecaBDE. These three mixtures differ in their relative contributions of isomers with varying degrees of bromination. Currently, PentaBDE and OctaBDE are banned from use in the European Union, and they have been voluntarily phased out in the United States. In the U.S., DecaBDE was recently (May 2007) banned from use in mattresses, furniture, TVs, and electronic enclosures in the states of Washington and Maine. This ban will go into effect starting in 2008 and a few application exemptions (e.g., in vehicles and airplanes) apply. Prior to these bans, the world market demand for PBDEs was estimated at approximately 70,000 t, of which 33,100 t was used in the United States alone (1). Brominated flame retardants (BFRs) are incorporated into products through additive or reactive methods (1). Reactive flame retardants are covalently bound to the materials they are protecting; in contrast, additive BFRs are mixed in with the resin or polymer, typically during the extrusion process. PBDEs are a type of additive flame retardants and are thus more likely to leach out over the lifetime of the products. Presently, PBDEs are found in almost every environment studied and are quite ubiquitous (2, 3). Because of their physicochemical properties, they are now considered a new persistent and bioaccumulative contaminant (4, 5). Scientific studies have found that PBDEs are present in human serum, blood, milk in females, and adipose tissues of almost all individuals (6–10). The primary routes of human exposure to PBDEs are currently unclear but are believed to be either from consumption of contaminated food or from inadvertent ingestion of contaminated indoor dust (11–13). Analysis of food items collected from supermarkets, in addition to recreational and commercial seafood, have found PBDEs to be present at low (ppb) levels in almost all types of foods (14–16)with higher concentrations typically observed in fish. One study investigating PBDEs in urban anglers found PBDEs levels in anglers to be higher, but not significantly higher, than non fish-eating consumers (17). In contrast, indoor environments contain elevated levels of PBDEs in air and dust (18, 19)suggesting exposure in homes and offices may also be a primary pathway. Furthermore, a recent study found a positive and significant relationship between PBDEs in breast milk and PBDE concentrations in indoor dust and from consumption of meat and dairy products (20). The recycling of products containing flame retardant chemicals may lead to increased exposure to employees working in these facilities. In fact, a recent study measured the indoor air concentrations of PBDEs near an electronics recycling facility in California and found levels that were approximately 1000× higher than air levels measured in indoor environments (21). To date the only occupational studies investigating exposures to PBDEs that have been conducted have been in electronic recycling facilities in Europe and China (22–25). These studies have demonstrated that workers in these facilities are receiving exposure to the VOL. 42, NO. 9, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3453
higher brominated PBDE congeners that are associated with the OctaBDE and DecaBDE commercial mixtures (1, 26, 27). In contrast, the congeners most frequently detected in U.S. human tissues are the congeners present in the PentaBDE formulation, which was primarily applied to polyurethane foam (28). Therefore, increased exposure to these more bioaccumulative lower brominated PBDE congeners may occur in employees working in close contact with foam and in foam recycling facilities. The present study was undertaken to determine if employees working in facilities recycling polyurethane foam contain higher body burdens of PBDEs relative to the general United States population. Because the recycled foam is often used to produce carpet padding underlay, carpet installers may also be receiving greater exposure to PBDEs. Our objectives were to determine if the serum PBDE concentrations were greater in these workers relative to the general United States population, and to determine if there were any significant differences in the PBDE congener patterns.
Materials and Methods Study Population. Institutional review board approval was granted for this study prior to collection of samples (Protocol 7647-05-10R0ER). Participants for this study were solicited through fliers and personal communications with employers at two factories that currently recycle polyurethane foam. Serum samples were collected from individuals working in the Baltimore, MD factory during March 2006 and were collected from individuals working in the Santa Ana, CA factory during October 2006. Clerical staff employed at the factories, and spouses of employees, were invited to participate as a control group. Professional carpet installers associated with the factory in Santa Ana, CA were also invited to participate. Informed consent was received and no individuals were compensated for their participation in this study. Serum Collection. Serum samples were collected by two medical facilities: Concentra Medical Laboratory in Columbia, MD and Tustin Irving Medical Facility in Santa Ana, CA. Clean collection tubes (serum separator tubes) and bovine serum (Gibco, Auckland, New Zealand) were shipped to the medical facilities for use in blood collection and for providing material for field blanks, respectively. Participants were asked to fill out and sign a consent form and a short questionnaire providing information regarding age and occupation history. Duplicate blood samples (approximately 8–10 mL each) were collected from each individual and centrifuged to isolate serum. Duplicate samples were collected for analysis by two different laboratories: Duke University and the Centers for Disease Control and Prevention (CDC) (Atlanta, GA) for quality assurance purposes. The serum was transferred to precleaned amber vials and frozen at the medical center for storage. The questionnaires, consent forms, and samples were shipped on dry ice, and after arrival serum samples were stored at -20 °C until analysis. Chemicals. Solvents used for the analysis of PBDEs by the Duke University laboratory were HPLC-grade purchased from VWR Scientific (Atlanta, GA). Two fluorinated BDE standards, 4′-2,3′,4,6-tetrabromodipheyl ether (FBDE-69) and 4′-fluoro-2,3,3′,4,5,6-hexabromodiphenyl ether (FBDE-160) (Chiron Inc., Trondheim, Norway) and 13C labeled BDE-209 (Wellington Laboratories, Guelph, Ontario) were used as internal and recovery standards. PBDE calibration solutions were prepared from neat standards purchased from Accustandard Inc. (New Haven, CT). Sample Extraction. Serum samples analyzed at Duke University were extracted according to a previously published method (29). Serum samples were transferred to 50-mL centrifuge tubes and spiked with the two internal standards, FBDE-69 and 13C BDE 209. The serum was allowed to 3454
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 9, 2008
equilibrate with the standards overnight before the extraction continued. The next day 1 mL of 6 M HCl and 5 mL of 2-propanol were added to the serum and it was vortexed for 30 s. Ten mL of a 1:1 mixture of hexane/methyl-tert-butyl ether (MTBE) was then added and vortexed for an additional 30 s. The mixture was allowed to sit for one hour and then centrifuged at 2500 RMP for 20 min. The organic layer was transferred to a clean test tube and the aqueous layer was re-extracted twice with 10 mL hexane/MTBE aliquots. The combined organic layers were reduced in volume to 5.0 mL by use of an automated rapid evaporation system (TurboVap, Zymark). Lipid was removed by adding 0.5 mL of concentrated sulfuric acid. The extracts were centrifuged at 2500 RPM for 15 min and the organic layer was transferred to a clean test tube. The aqueous fraction was re-extracted with 5 mL of hexane in a similar manner. The combined hexane extracts were again reduced in volume to 0.5 mL. They were then cleaned by use of solid-phase alumina chromatography. The extracts were eluted through 4.0 g of 6% deactivated alumina using 30 mL of petroleum ether. The eluent was reduced in volume a second time to 0.5 mL, transferred to a GC autosampler vial, and spiked with the recovery standard FBDE-160. Samples analyzed by the CDC laboratory were extracted using a semiautomated high-throughput extraction method previously published (30). Lipid content was determined by the CDC laboratory by measuring total triglycerides and cholesterol and using a previously determined regression equation to calculate total lipid (31). Sample Analysis. Extracts were analyzed for PBDEs using a gas chromatograph (GC, Agilent 5890) coupled to a mass spectrometer (Agilent 5975) operated in electron capture negative ionization (GC/ECNI-MS) mode. A 0.25 mm (i.d.) × 15 m fused silica capillary column coated with a 5% phenyl methylpolysiloxane (0.25 µm film thickness) was used for the separation of PBDE congeners. Programmable temperature vaporizing (PTV) injection was employed in the GC. The injection port temperature was held at 50 °C for 0.3 min followed by a temperature ramp of 700 °C/min to 275 °C. The oven temperature program was held at 80 °C for 2 min followed by a temperature ramp of 12 °C /min to 140 °C, followed by a temperature ramp of 5 °C /min to a final temperature of 280 °C, which was held for an additional 20 min. The transfer line temperature was maintained at 280 °C and the ion source was held at 200 °C. A suite of 27 individual BDE congeners was measured in all samples. Trithrough octa-BDE congeners were quantified by monitoring bromide ions (m/z 79 and 81). All three nona-BDE congeners and BDE-209 were quantified by monitoring molecular ion fragments (m/z 486.6 and 409.6) while BDE-209L was monitored through m/z 494.6 and 415.6. The method of analysis used at the CDC laboratory has been published previously (30). Quality Control. Bovine serum was used as a field blank by the medical centers using the same methods as for human blood. At least four bovine serum blanks were analyzed by each laboratory to monitor potential collection contamination. Minor laboratory contamination with BDE-47 and BDE99 was observed at the Duke University Laboratory in two of the three batches analyzed. Field blanks in these two groups contained 0.67 ( 0.1 and 0.58 ( 0.2 ng of BDE-47 and BDE99, respectively. Therefore all serum samples were blankcorrected for these two congeners. The mass subtracted was equivalent to the average BDE level in the blank. Recovery of the internal standard, FBDE-69 averaged 65 ( 5%; however, recovery of the 13C BDE-209 compound averaged about 35%. Therefore, BDE-209 was not quantified in these samples. PBDE measurements of field blanks by the CDC laboratory were low (median
