میزان دی بنزو پی دیوکسین پلی کلرینه، dibenzofurans (PCDD/F) و بی فنیل (PCB) در خون کارگران غیررسمی در آگبوگبلوشی کشور غنا و کنترل آنها Levels of polychlorinated dibenzo-p-dioxins, dibenzofurans (PCDD/Fs) and biphenyls (PCBs) in blood of informal e-waste recycling workers from Agbogbloshie, Ghana, and controls

Abstract The formation and environmental release of highly toxic organohalogen compounds associated with informal recycling of waste electric and electronic equipment (e-waste) is a growing problem at e-waste dumps/recycling sites (EWRSs) in many developing countries worldwide. We chose a cross-sectional study design to measure the internal exposure to polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) as well as polychlorinated biphenyls (PCBs) of individuals working on one of the largest EWRSs of Africa, located at Agbogbloshie, Accra, Ghana, and in controls from a suburb of Accra without direct exposure to EWRS activities. In whole blood samples of 21 age matched male exposed individuals (mean age: 24.7 years, SD 6.0) and 21 male controls (mean age: 24.4 years, SD 5.7) 17 PCDD/F congeners were determined. Moreover three indicator PCB congeners (#138, #153 and #180) were measured in blood of 39 exposed (mean age: 27.5 years, SD 11.7) and 19 non-exposed (mean age: 26.8 years, SD 9.7) patients. Besides a health examination, biometric and demographic data, residential and occupational history, occupational exposures and working conditions were recorded using a standardized questionnaire. In the exposed group, median PCDD/F-concentrations were 6.18 pg/g lipid base WHO2005-TEq (range: 2.1–۴۲٫۷) and significantly higher compared to the control group with 4.60 pg/g lipid base WHO2005-TEq (range: 1.6–۱۱٫۶). Concentrations were differentfor 2,3,7,8-TetraCDD, three HexaCDD and all 10 PCDF congeners, indicating a combustion pattern. Using a multivariate regression analysis exposure to EWRS activities was the most important determinant for PCDD/F exposure. Median PCB levels for the indicator congeners #138, #153 and #180 were 0.011, 0.019 and 0.008 μg/l whole blood (ranges: 0.002–۰٫۱۸, ۰٫۰۰۳–۰٫۱۶, ۰٫۰۰۲–۰٫۰۷۸) in the exposed group and, surprisingly, significantly higher in the controls (0.037, 0.062 and 0.022; ranges: 0.005–۰٫۴۶, ۰٫۰۱۰–۰٫۴۶, ۰٫۰۰۴–۰٫۲۱). In a multivariate regression approach e-waste related activities had no positive influence on internal PCB exposure, but rather the time living in Accra. The internal PCB exposure is in particular notable for a country where PCBs have historically never been produced or used. The impact of EWRS activities on organohalogen compound exposure of individuals working at and living in the surroundings of the Agbogbloshie EWRS, and the surprisingly high PCB exposure of people living in Accra not involved in e-waste activities require further investigation. ۱٫ Introduction Formation of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) by thermal processes was first identified and quantified in emissions from municipal waste incinerators in 1977 (Olie et al., 1977). Intense research followed and it rapidly became obvious that PCDD/Fs and other halogenated (aromatic) compounds are formed in all combustion processes involving organic carbon, oxygen and halogens, especially in the presence of metals like copper working as catalysts (Fiedler, 1998). In most industrialized countries, significant emission sources, including cable incinerator plants, were identified and emissions were reduced both by technical and legislative measures at the end of the last century. As a result, a continuous reduction of the internal exposure to PCDD/Fs was observed in Germany and many other industrialized countries (Wittsiepe et al., 2000; Ulaszewska et al., 2011; Consonni et al., 2012). With the increasing production of home appliances and consumer electronics devices, such as refrigerators, washing machines, air conditioners, lighting equipment, TVs/monitors, computers, printers, photocopiers, fax machines, cell/smart phones, game consoles and batteries, in combination with the downward trend in prices and their ever shorter life span, a growing uncontrolled stream of those products at their end-of-life-time developed. The wastes from electric and electronic equipment (WEEE or ewaste) contain relatively high amounts of valuable materials which are highly integrated into each other and therefore are difficult to separate. E-waste has been exported (sometimes illegally) as “second-handgoods” by developed countries and informal e-waste dumps/recycling sites (EWRSs) were set up since the middle of the 1990s in developing countries in Africa and Asia, e.g. Ghana (Accra), Nigeria (Lagos), China (Guiyu in Shantou, Guangdong Province), India (Delhi) and Pakistan (Karachi) (Kuper and Hojsik, 2008; Robinson, 2009; Chi et al., 2011; Schluep et al., 2012). Informal recycling processes used can be different at EWRSs, but often include steps of manual dismantling, acid leaching, heating and wanton burning (Brigden et al., 2005, 2008; Jian et al., 2014). The e-waste material itself can contain halogenated persistent organic pollutants (POPs), such as polychlorinated biphenyls (PCBs) used as dielectric or flame-retardant plasticizers and brominated flame retardants (BFRs) like polybrominated diphenyl ethers (PBDEs), biphenyls (PBBs) and others. Moreover, the open burning of e-waste, that contain halogens, e.g. BFR containing or polyvinyl chloride (PVC) coated copper cables, provide optimal conditions for de novo formation of halogenated aromatic compounds, such as PCDD/Fs, its brominated equivalents PBDD/Fs, mixed brominated/chlorinated homologies (PXDD/Fs) and other dioxin-related compounds (DRCs) (Weber and Kuch, 2003; Hedman et al., 2005; Gullett et al., 2007; Duan et al., 2011; Hibbert and Ogunseitan, 2014). Due to uncontrolled combustion and thermal processing of e-waste, these POPs have been found in the air, bottom ash, dust, soil, water and sediment samples from EWRSs worldwide, partly in tremendous high concentrations (Li et al., 2007; Wong et al., 2007; Brigden et al., 2008; Liu et al., 2008; Ma et al., 2009a, 2009b; Wen et al., 2009, 2011; Ni et al., 2010; Tue et al., 2010b; Zhang et al., 2012b; Chan and Wong, 2013; Hu et al., 2013; Labunska et al., 2013; Ren et al., 2013; Wang et al., 2013; Hosoda et al., 2014; Xiao et al., 2014). The share of annual mass of e-waste derived PCDD/F in China alone is estimated to be in the range of several kilograms (Ni et al., 2010). At EWRSs direct involvement in open burning without protective gear, and environmental contamination with PCDD/Fs and DRCs may result in accumulation in humans, but only very few data on internal human exposure exist (Yang et al., 2013; Tue et al., 2014). Typical pathways are direct soil/dust ingestion, inhalation of fumes and consumption of contaminated local food, especially fish. Data from Chinese, Indian and Vietnamese EWRSs including human health risk assessments have been summarized recently (Sepulveda et al., 2010; Tsydenova and Bengtsson, 2011; Zhang et al., 2012c; Chan and Wong, 2013; Tue et al., 2013; Song and Li, 2014). At the Agbogbloshie EWRS near Accra, Ghana, the technical processes used seem to be more outmoded compared to Asian sites. Workers, often children and adolescents, working 10–۱۲ h per day without any form of protective gear are exposed to frequent burns, cuts, and continuous inhalation of highly contaminated fumes (Brigden et al., 2008; Akormedi et al., 2013; Sthiannopkao and Wong, 2013; Jian et al., 2014). To our knowledge this is the first study on internal exposure to PCDD/Fs and PCB at an African EWRS and the first study on blood levels of PCDD/F of directly exposed EWRS workers worldwide. In this paper, we report levels of PCDD/Fs and PCB in blood samples of workers from the Agbogbloshie EWRS in Accra, Ghana, and of controls from the surrounding area without exposure to e-waste recycling. ۲٫ Materials and methods ۲٫۱٫ Study design We conducted a cross-sectional study at the Agbogbloshie EWRS, one of the largest EWRS in Africa, located in Accra, the capital of Ghana. About 40,000 people live in this area under the most deplorable environmental conditions and largely represent a migrant population from northern parts of Ghana. For comparison, a second group of individuals was recruited in Kwabenya North, a suburb of Accra, which is located about 25 km north of the city center of Accra. Compared to Agbogbloshie, the environment and ambient air in Kwabenya North are relatively pristine. The population in Kwabenya North consisted predominantly of migrants who are transitioning from rural areas to an urban fringe area, mostly unskilled and of low socio-economic status, similar to the migrant group working on the e-waste dumpsite. Details of the recruitment, study procedures and sample collection have been described in detail before (Feldt et al., 2014). A total of 75 exposed and 40 unexposed individuals were recruited in October 2011. Due to technical complexity and high costs, PCDD/F analysis was conducted in those patients with the highest risk of exposure and in agematched controls. Among patients with a sufficient blood sample volume of at least 30 ml, participants who were directly involved in the burning of e-waste materials, who worked at the site for at least one year, and who also lived at the EWRS, were selected for PCDD/F analysis. For 21 participants of the exposed group who met these criteria, 21 participants from the control group were chosen by best-matching age (see Table 1a). After this selection, analysis of PCB was conducted in all patients with sufficient sample volume being available for testing (39 exposed subjects and 19 controls) (see Table 1b). ۲٫۲٫ Laboratory analysis The method for determination of PCDD/Fs in blood samples was described in detail before (Wittsiepe et al., 2007). In brief, the method includes: (a) extraction of blood fat with organic solvents, (b) multiplecolumn chromatography cleanup using modified silica gels and activated charcoal, and (c) instrumental determination with capillary gas chromatography and high-resolution mass spectrometry. Toxicity equivalent concentrations were calculated according to the WHO 2005 model (Van den Berg et al., 2006). All concentrations were adjusted on a lipid base. Determination of PCB was performed using a recently developed method featuring on-line solid-phase extraction and large volume injection gas chromatographic high resolution mass spectrometry (Wittsiepe et al., 2014). Concentrations were calculated on a whole blood volume base. Because most values of the lower chlorinated indicator PCBs (#28, #52, #101) were below LOD or in the range of blank samples, these data were not included here. ۲٫۳٫ Statistical analysis For statistical calculations values below the limit of detection (LOD) were set to 1/2 LOD. Statistical analysis was performed using STATISTICA software system (version 10 & 12.5, Statsoft, Inc., Tulsa, www.statsoft. com). The Mann–Whitney U test was used to compare differences of between POP concentrations in the exposed and control group. In addition, we used the t-test for the same purpose with log-transformed POP concentrations. To investigate associations between e-waste related exposure and internal POP concentrations, and to control for confounding factors, multivariate regression models were used. The covariates years working/living at site, years living in Accra in general, age, and exposed/non-exposed to ewaste recycling processes were included. The variables describing the time working/living at the EWRS and in Accra were together and optionally used in the model. We used the log10-transformed concentration values as dependent variables to ensure normal distribution of residuals. The variables years working/living at EWRS, years living in Accra in general and age were log2-transformed, so that their associations with POP-levels were estimated for a doubling in the natural scale. Multivariate regression analysis was computed by the SAS software (version 9.3, SAS Institute, Cary, NC, USA) procedure GENMOD, which fits a generalized linear model to the data by maximum likelihood estimation. The results of these regression analyses are expressed as adjusted regression coefficients (β) together with their 95% CIs and Wald p-values. ۲٫۴٫ Ethical considerations The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Ethical clearance was obtained from the institutional review board of the Noguchi Memorial Institute for Medical Research, University of Ghana, Accra.