Surveillance of dioxins and polychlorinated biphenyls (PCBs) in food commercialized in Barcelona, Spain

Transcription

1 Food Additives and Contaminants: Part B Vol. 2, No. 1, June 2009, Surveillance of dioxins and polychlorinated biphenyls (PCBs) in food commercialized in Barcelona, Spain M. Fontcuberta ab *, J.F. Arque s ab, J.R. Villalbı ab, M. Martı nez a, E. Serrahima a, F. Centrich ab,m.ábalos c, E. Abad c, J. Duran a and C. Casas ab a Age`ncia de Salut Pu blica de Barcelona, Plaça Lesseps 1, E-08023, Barcelona (Catalonia), Spain; b CIBERESP, Spain; c Laboratory of Dioxins, Mass Spectrometry Laboratory, Department of Environmental Chemistry, 1D ÆA-CSIC, Jordi Girona 18, 08034, Barcelona, Spain (Received 25 August 2008; final version received 26 March 2009) This study explored the potential use of seven congeners of polychlorinated biphenyls (PCBs-7) as indicator compounds for the presence of dioxins and PCBs in food samples, as part of the routine surveillance programme of a public health agency. Samples of 24 foodstuffs with high fat content were collected (ten fresh fish, six dairy products, five meat and three eggs). Duplicate analyses were performed. A research laboratory tested samples for seven polychlorinated dibenzo-p-dioxin (PCDDs), ten dibenzofurans (PCDFs) and twelve dioxin-like PCBs, with limits of detection in the range of ng kg 1 (ppt). The public health services official control laboratory tested samples for PCBs-7, with a limit of quantification of 5 mgkg 1 (ppb). The research laboratory detected the presence of dioxins and dioxin-like PCBs in all samples; fish samples had the highest levels ( pg WHO-TEQ g 1 ). The public health service official control laboratory detected PCBs-7 only in five samples, which were all fish. Comparing the results in the two laboratories there seems to be an association between the detection of PCB-7 and the presence of higher levels of PCDD/Fs and dioxin-like PCBs. The use of PCB-7 as an indicator compound may be a practical surveillance strategy for those foodstuffs with higher concentrations of dioxin-like congeners. Keywords: chromatography gas chromatography-mass spectrometry (GC/MS); dioxins toxic equivalent quantities (TEQs); polychlorinated biphenyls (PCBs); animal products meat; cheese; eggs; fish Introduction Dioxins (polychlorinated dibenzo-p-dioxins and dibenzofurans, PCDD/Fs) and polychlorinated biphenyls (PCBs) are lipophilic organic compounds of concern to public health. Among the 210 existing PCDD/F congeners, 17 are of toxicological concern. Among the 209 PCB congeners, twelve are considered to be dioxin-like PCBs because a chlorine substitution and their planar geometry make them toxicologically similar to dioxins. Although acute effects of PCDD/F and PCBs exposure have been described, the greater public and scientific concern relates to its chronic effects on the immune and endocrine systems, and their neurotoxic and carcinogenic effects. The Global Stockholm Convention on persistent organic pollutants (United Nations Environment Programme 2001) and European Regulation 850/2004 (European Parliament and European Council 2004) shape the current European regulatory frames to reduce human exposure through their progressive elimination and emission control. PCDD/Fs and to some extent PCBs can also be formed as unwanted by-products of many anthropogenic processes (steel and iron plants, waste management, hazardous waste incinerations, etc.). Besides, even if current legislation bans PCB production and industrial use, human exposure may derive from equipment produced before In Spain, the existence of around 50,000 tons of equipment containing at least 50 mg kg 1 of PCBs was estimated for 2004 (Spanish Ministry of Environment 2006). Once released into the atmosphere, these pollutants can easily enter the food chain due to their lipophilic properties and bioaccumulate in fatty compartments of tissues and organs. For the general population, dietary intake is the main source of PCDD/Fs and PCBs, estimated to be responsible for the 95% of daily exposure (King 1999). International organizations have defined tolerable intakes for these compounds: the World Health Organization (WHO) established a tolerable daily intake (TDI) of 1 4 pg I-TEFs (Toxic Equivalent) for each kg of body weight per day. Later, the Scientific Commission on Food (European Commission 2001a) settled a tolerable weekly intake (TWI) of 14 pg WHO-TEQs kg 1 body weight, apparently exceeded by a large proportion of the *Corresponding author. mfontcub@aspb.cat ISSN print/issn online ß 2009 Taylor & Francis DOI: /

2 European population (European Commission 2001b). The European Union has established maximum residue levels for PCDD/Fs and the twelve dioxin-like PCBs in several foodstuffs in 2006 (European Commission 2006), while previous regulations from 2001 set maximum levels only for dioxins. During the Belgian dioxin alert in chicken and pork (1999), and due to the complexity of the PCB mixture involved, seven congeners known as PCB-7 (PCB-28, PCB-52, PCB-101, PCB-118, PCB-138, PCB-153 and PCB-180) were chosen as markers (Covaci et al. 2002). These were predominantly present in the commercial mixtures of PCB congeners produced by industry and in environmental samples, and some of them (PCB- 138, PCB-153 and PCB-180) were also the predominant congeners found in human tissues (National Center for Environmental Health 2003). Health authorities across Europe have been using different combination of non-toxic PCBs as markers, depending on the commercial PCBs mixture predominantly used in each country, due to their relatively simple analytical procedure compared with dioxin analysis (European Commission 2001b). The main goal of this study was to explore the adequacy of a monitoring programme using PCB-7 as indicators for the surveillance of dioxins and PCBs in marketed food. The public health official control laboratory results for PCB-7 are compared with the specific PCDD/Fs and DL-PCBs congeners found in a research laboratory. This study was undertaken as part of a review of the current strategy for the surveillance of persistent organic pollutants in the programme IQSA (an acronym in Catalan for Research on the Health Quality of Food), developed since 1984 by the public health services of the city of Barcelona (Catalonia, Spain) (Vicente et al. 2004). Materials and methods Design The seven congeners of PCBs chosen to monitor the mixture of PCDD/Fs and PCBs released in the Belgium incident in 1999 were included in the Barcelona IQSA surveillance programme in 2006 and analysed in the public health official control laboratory. Some duplicate samples were also sent to the CSIC research laboratory, where the 17 toxic PCDD/ Fs and the twelve dioxin-like PCBs were analysed. The results of these duplicate samples are compared, one-by-one and by food groups. Food samples Samples were collected as part of the IQSA programme. This food surveillance programme has the aim of monitoring the presence and amount of Food Additives and Contaminants: Part B 67 additives and contaminants (chemical or microbiological), in food sold to consumers in the city (through retail trade in municipal markets, supermarkets and grocery stores). Foodstuffs chosen for this comparative study were commonly consumed items with a relatively high fat content, susceptible of containing lipophilic pollutants. Altogether, 24 samples pertaining to four food groups were included: fresh fish (ten), dairy products (six), meat products (five) and eggs (three). Duplicate samples were sent to two different laboratories: in fish, different individuals from the same batch were collected for small fish species (red mullet, mackerel and sardine) and two fillets from the same individual for large fish (salmon and tuna); for packaged foods, samples were taken from the same lot. Results are reported per g of fresh weight (f.w.) for fish samples, and per g of total fat for other samples. PCDD/F and DL-PCBs analyses in the CSIC research laboratory The analytical method validated in the laboratory was partially based on EPA methods 1613 and Briefly, a portion of the freeze-dried sample, in the case of fish, meat, eggs and cheese, was spiked with known amounts of mixtures of 13 C 12 -PCDD/Fs and 13 C 12 - dioxin-like-pcbs (EPA-1613LCS and WP-LCS, Wellington Laboratories, Inc., Guelph, Canada) and then extracted with cyclohexane:toluene (1:1) by Soxhlet. Next, the extract was rotary concentrated and kept in the oven overnight (105 C) to eliminate the solvents before gravimetrical fat determination. Afterwards, the fat residue was redissolved in n-hexane (for butter samples a portion was directly dissolved in n-hexane). Organic components, fat and other interfering substances were removed by treating the n-hexane extract with sulphuric acid. The extract was then rotary concentrated again and filtered before the clean-up process. An automated Power-Prep TM system (Fluid Management Systems, Inc., Waltham, MA, USA) based on the use of multilayer silica, basic alumina and carbon adsorbents was used to carry out the clean-up step (Abad et al. 2000). Finally, instrumental analyses were performed by high-resolution gas chromatography coupled to high-resolution mass spectrometry (HRGC/HRMS) on a 6890N Network GC System Agilent gas chromatograph (Agilent Technologies, Inc., Palo Alto, CA, USA) coupled to an Autospec Ultima NT high-resolution mass spectrometer (EBE geometry) (Micromass, Manchester, UK), using a positive electron ionization (EIþ) source operating in SIM mode. The chromatographic column used was a DB-5ms from J&W Scientific (Agilent Technologies) (Abad et al. 1997). Quantification was carried out by the isotopic dilution method. Relative response factors were obtained for each individual

3 68 M. Fontcuberta et al. analyte by the analysis of six different calibration solutions for PCDDs/PCDFs and seven in the case of DL-PCBs. The results are expressed TEQ values using WHO-TEF. TEQs values were calculated in upper bound assuming the limit of quantification (LOQ) for those non-quantifiable or below the LOQ. The criteria for ensuring the quality of the data included the application of the quality assurance and quality control (QA/QC) measures. The limit of detection (LOD) and limit of quantification (LOQ) were determined following a statistical approach based on the uses of control blank covering the whole methodology samples as described by the International Union of Pure and Applied Chemistry (IUPAC). Analytes were mostly below the LOQ or not detected and only traces of highly chlorinated compounds were quantified. The figures varied from 0.11 to 2.82 (OCDD) pg for PCDD and PCDFs, whereas values ranged between 0.82 and (PCB-118) pg for DL-PCBs. The validity of the data was also contrasted by analysing certified reference materials (CRM) for PCDD/Fs such as the BCR-607 (milk powder). Therefore, uncertainty values obtained during the validation process were around 17% (units expressed as WHO-TEQPCDD/ F þ DL-PCB). To complement analysis of CRMs the continuous participation in intercalibration exercises is current practice in the laboratory as a tool to ensure the quality of generated data. Table 1a. Method characteristics for the analysis of PCBs-7: Barcelona, Analyte Matrix Limit of quantification (LOQ) (mg g 1 ) PCBs-7 analysis (Public Health Official Control Laboratory) The method used for PCB-7 determination in food was developed in the Barcelona Public Health Agency laboratory and fully validated through parameters such as precision, recovery, and linearity. It was accredited by the National Accreditation Body (Entidad Nacional de Acreditacio n (ENAC)) following the ISO Before extraction with dichloromethane, sample portions of 10 g were homogenized. The method was based on a primary extraction using the accelerated solvent extraction (ASE) with dichloromethane (DCM). Gel permeation chromatography (GPC) with DCM was used as the clean-up method. PCBs analysis was performed with an Agilent 6890N gas chromatograph with mass-spectrometry 5975 (MS- SIM mode). The oven temperature was programmed starting at 70 C and held for 2 min, followed by increases of 25 C min 1 to 150 C, 3 C min 1 to 200 C and 8 C min 1 to 280 C. Injection port temperature was 300 C, as was detection temperature. Helium was the carrier gas (1.8 ml min 1 ). The LOQ was mg kg 1 for fish, meat, cheese and eggs and mg kg 1 for butter or fat (Table 1). Calibration curves were linear in the range mg kg 1 with correlation coefficients and a relative standard deviation (RSD) of response factors 525% for all analytes. Precision and accuracy of the method were Recovery range (%) RSD (%) (n ¼ 18) U(2*% RSD) Accreditation PCB_28 Meat/fish Yes PCB_52 Meat/fish Yes PCB_101 Meat/fish Yes PCB_118 Meat/fish Yes PCB_153 Meat/fish Yes PCB_138 Meat/fish Yes PCB_180 Meat/fish Yes PCB_28 Cheese Yes PCB_52 Cheese Yes PCB_101 Cheese Yes PCB_118 Cheese Yes PCB_153 Cheese Yes PCB_138 Cheese Yes PCB_180 Cheese Yes PCB_28 Eggs Yes PCB_52 Eggs Yes PCB_101 Eggs Yes PCB_118 Eggs Yes PCB_153 Eggs Yes PCB_138 Eggs Yes PCB_180 Eggs Yes PCB_28 Fat Yes PCB_52 Fat Yes PCB_101 Fat Yes PCB_118 Fat Yes PCB_153 Fat Yes PCB_138 Fat Yes PCB_180 Fat Yes

4 Food Additives and Contaminants: Part B 69 tested with spiked and latter extraction of the samples with standard for each matrix (fish, eggs, meat, cheese, fat). Reproducibility was tested with duplicate samples on 3 days for three concentration levels, and it is expressed as percentage RSD (total n ¼ 18). Recovery values were between 70% and 110% and the RSD recovery acceptation criteria was 20%. Statistics Standard descriptive quantitative methods (frequencies, means and standard deviations, and maximum levels) were calculated for each group of compounds (PCDD/Fs, dioxin-like PCBs and PCBs-7) and for each individual sample and food group. A concentration equal to zero was assumed for samples with dioxins and PCBs levels below the LQ. Table 1b. Method characteristics for PCDD/F and DL-PCBs analysis: Barcelona, Analyte Reference value (pg g 1 ) I (pg g 1 ) Matrix Non-parametric H Kruskal Wallis tests were performed to estimate the significance of differences among groups. Results Dioxins and dioxin-like PCBs concentrations The PCDD/Fs and DL-PCBs concentration in pg WHO-TEQ g 1 is given in Table 2 for each sample. Congener groups with highest concentrations (pg g 1 ) were mono-ortho PCBs, especially PCB-118 values ranged from 23 to 8176 pg g 1 f.w. in fish samples and from 134 to 643 pg g 1 fat in other samples. Non-ortho PCBs had much lower values, and their highest concentration levels were found again in fish samples. PCB-77 was the most important congener (values from 0.99 to pg g 1 f.w.), except in butter where Limit of detection (LOD) (pg g 1 ) Limit of quantifica tion (LOQ) (pg g 1 ) Recovery range (%) (WHO- TEQ) RSD% (n ¼ 10) Measurement uncertainty I(k ¼ 2) PCDD/Fs þ DL-PCBs (TEQ) Fish Yes PCDD/Fs þ DL-PCBs (TEQ) Pork Yes PCDD/Fs þ DL-PCBs (TEQ) Fish oil Yes PCDD/Fs þ DL-PCBs (TEQ) Milk powder Yes Table 2. PCDD/F and PCBs dioxin-like concentration (pg g 1 WHO-TEQs) in fish samples and non-fish samples examined: Barcelona, Samples PCDD/Fs Non-ortho PCBs Mono-ortho PCBs PCDD/Fs + PCBs Concentration (pg g 1 WHO-TEQ f.w.) Red mullet A Red mullet B Mackerel A Mackerel B Tuna A Tuna B Salmon A Salmon B Sardine A Sardine B Concentration (pg g 1 WHO-TEQ fat) Cheese A Cheese B Cheese C Butter A Butter B Butter C Eggs A Eggs B Eggs C Chicken A Chicken B Chicken C Chicken D Chicken E Accredit ation

5 70 M. Fontcuberta et al. PCB-126 had higher concentrations. OCDD was the predominant PCDD/F congener in most of dairy samples (with values ranging from 0.22 to 0.94 pg g 1 fat), eggs (from 1.69 to 9.76 pg g 1 fat) and meat products (from 1.58 to pg g 1 fat). In fish samples, the predominant congener was 2,3,7,8-TCDF (TCDF, tetrachlorodibenzofuran; values ranged from 0.03 to 2.32 pg g 1 ). Considering the total sum in WHO-TEQs, highest values were found in fish samples, with a mean value of 3.32 pg WHO-TEQ g 1 (Figure 1) and levels ranging from 0.03 to pg WHO-TEQ g 1 (Table 2). Other food groups had much lower values: chicken meat ¼ 0.73 pg WHO-TEQ g 1, dairy products ¼ 0.71 pg WHO-TEQ g 1, and eggs ¼ 0.38 pg WHO-TEQ g 1. This difference between fish and non-fish products was due to variation in PCB levels, as PCDD/Fs were of similar magnitude. PCBs-7 detection in the official control laboratory PCBs-7 were detected in five out of ten fish samples, with levels ranging from 6 to 38 ng g 1 of fresh weight (Table 3), which were much higher levels than for the toxic congeners as expected. All samples with measurable levels (two red mullets, one mackerel and one sardine) were from the Mediterranean, except for one tuna caught in the Atlantic (but near the Mediterranean). No PCBs-7 were detected in the samples of dairy products, eggs nor meat. Figure 1. Mean PCDD/F and dioxin-like PCBs levels (pg/g WHO TEQs) and percentage of contribution to the total TEQ by food group in food samples examined in the research laboratory of the CSIC, Barcelona, Table 3. Detection and concentration (ng g 1 or ppb) of PCBs-7 in fish samples tested, by nature and origin of the fish. Official control laboratory of the city public health services, Barcelona, Samples Origin Concentration (ng g 1 ) PCB-138 PCB-153 PCB-180 Red mullet A Mediterranean (Valencia, Spain) Red mullet B Mediterranean (Catalonia, Spain) Mackerel A Atlantic (Cantabria, Spain) n.d. n.d. n.d. Mackerel B Mediterranean (Catalonia, Spain) 6 8 n.d. Tuna A Atlantic (Canary islands, Spain) n.d. n.d. n.d. Tuna B Atlantic (Andalucia, Spain) Salmon A Atlantic (Norway) n.d. n.d. n.d. Salmon B Pacific (Alaska, USA) n.d. n.d. n.d. Sardine A Mediterranean (Catalonia, Spain) Sardine B Atlantic (Galicia, Spain) n.d. n.d. n.d. Note: n.d., Not detected.

6 Food Additives and Contaminants: Part B 71 Table 4. Mean concentration of PCDD/F and dioxin-like PCBs (pg g 1 WHO TEQs) in food samples examined, by PCB-7 detection in the public health official control laboratory. Barcelona, Mean (pg g 1 WHO-TEQ) Samples with PCB-7 detected (n ¼ 5) Samples with PCB-7 not detected (n ¼ 19) p-value PCBs mono-ortho PCBs non-ortho PCDD/F Sum of PCDD/F and PCBs dioxin-like Congeners 138 and 153 were detected in all positive samples, and PCB-180 was detected in four of the five positive samples. The other PCBs markers were not detected. Association between PCBs detection and PCDD/F and dioxin-like PCBs levels Sorting out samples according to the detection of PCB-7 as markers (Table 4), fish samples with detectable levels of PCB-7 had a mean concentration of 5.8 pg WHO-TEQ g 1, while the 19 samples with no detectable levels of PCB-7 had a mean of 0.7 pg WHO- TEQ g 1 (p ¼ 0.001). As can be seen in Table 4, these differences in WHO-TEQs concentrations are mainly due to the differences in PCBs concentrations (and mostly to non-ortho PCBs). Discussion The presence of the 17 toxic congeners of dioxins and twelve dioxin-like PCBs was detected in all food samples studied in a research laboratory, although the highest levels were in fish samples. Other studies in the same context have consistently shown that PCDD/ F and PCBs contamination is highest in fish (Llobet et al. 2003). The European Union recently established maximum permissible levels for fish of 4 and 8 ng WHO TEQ kg 1 for PCDD/F and for PCDD/F plus dioxinlike PCBs. All samples analysed showed concentrations under the European Union maximum permissible levels, except one tuna sample with ng WHO- TEQ kg 1, which exceeded the PCDD/F plus dioxinlike compounds limit, mainly due to the high level of non-ortho PCBs (7.23 ng WHO-TEQ kg 1 ). Similarly to other studies (Bordajandi et al. 2006) this pattern is followed in all fish samples: the most important contribution to total WHO-TEQs is due to non-ortho PCBs, that represent around 60% of total WHO-TEQs. Regarding the individual contribution of each congener, the non-ortho PCB-126, due to its specific toxic pattern with the highest TEF among PCBs, accounted for the 97% of the total non-ortho PCBs which is 58% of the total TEQ content. The mean PCDD/F contribution in fish samples is 18% of the total TEQ content, being the two penta congeners (2,3,4,7,8-PeCDF and the 1,2,3,7,8-PeCDD) the main contributors to this percentage, followed by the 2,3,7,8-TCDF. These results are in disagreement with some data that show the tetra congener 2,3,7,8- TCDF as one of the most important congeners in fish samples (Focant et al. 2002). Concerning other food groups, overall congener distribution is quite constant over matrix types. The non-ortho PCBs concentration is much lower and PCDD/F contribution to total WHO-TEQ became more important: in eggs and dairy products PCDD/F contribution is similar to PCBs but in chicken samples is clearly higher (75% of the total TEQ content). These results are in good agreement with contributions reported in aquatic and terrestrial foodstuffs (Focant et al. 2002) showing a much higher level of PCBs in aquatic samples than in terrestrial ones. The most important PCDD/F congeners are again the two penta congeners. Except for the tuna sample mentioned above, extracted from the Atlantic just at the gates of the Mediterranean, fish samples with higher levels of contamination were of Mediterranean origin. Some data on chlorinated compounds concentration in mammals suggested a higher contamination in the Mediterranean sea due to its semi-enclosed nature and the high population density around (Borrell and Aguilar 2007), although a slight decrease of PCBs levels in marine biota was also reported. In fact, some authors suggest also a decline in food PCDD/F levels in recent years (Fernandez et al. 2004; Gomara et al. 2005) and a less evident decrease or even a slightly increase in PCBs levels (Bocio et al. 2007). Comparing the present results with these studies, levels of both PCDD/F and PCBs are slightly higher: fish samples analysed by Bocio et al. had a PCDD/F mean level of 0.19 pg WHO-TEQ g 1 and PCBs mean level (as the sum of PCB-77, PCB-126, PCB-169 and PCB-105) of 1.16 pg WHO-TEQ g 1 ; while the present results were 0.54 pg WHO-TEQ g 1 for PCDD/F and 2.12 pg WHO-TEQ g 1 for the sum of the same PCB congeners. Gomara et al. (2005) also reported a lower

7 72 M. Fontcuberta et al. PCDD/F mean level (0.39 pg WHO-TEQ g 1 ) and a sum of PCBs dioxin-like (2.74 pg WHO-TEQ g 1 ) just under our result (2.78 pg WHO-TEQ g 1 ). None of the farmed salmon samples showed high contamination, despite reports of high contamination in European farmed salmon, suggesting that the salmon farming industry may have reviewed feeding practices (Easton et al. 2002; Shaw et al. 2006). On the other hand, PCB-7 used as markers were detected by the official control laboratory in five out of ten fish samples, while no meat, dairy nor egg samples had detectable levels. The levels found (ranging from 14 to 91 ng g 1 f.w.) are of the same order as those observed in other studies (Bayarri et al. 2001). The three congeners detected by the official control laboratory (PCB-138, PCB-153 and PCB-180) are also reported in other studies as the three predominant congeners among the PCB-7 studied in fish samples (Ferrante et al. 2007). They are also commonly found in human tissue and serum (National Center for Environmental Health 2003). Di-ortho PCBs such as the PCB-7 (except congener 118) do not display dioxinlike toxicity, have a different toxicological profile but are more persistent in the environment and in organisms: some authors argue that their danger is sub estimated (Hansen 1998). There is no maximum residue limit (MRL) for these PCBs in fish or seafood, although some authors have advocated for its adoption (Ferrante et al. 2007). In fact, the current reference value of 200 ng g 1 of fat weight was defined only for terrestrial animals during the Belgian incident, which involved chicken feed. Covaci et al. (2002) reported an ability of chicken to metabolize to some extent congeners of tri- to penta-chlorinated PCBs (except PCB-28), but not these highly chlorinated PCBs, which may explain their frequent accumulation in chicken samples. There have been no reported incidents of massive contamination in Spain, and thus the presence of PCBs and dioxin in marketed food reflects the background pollution and how it impacts on the food chain. However, when sorting out samples according to the detection of PCB-7 as markers, there was a huge difference statically significant in the concentrations of PCDD/Fs and dioxin-like PCBs between both groups: fish samples with detectable levels of PCB-7 had concentrations of toxic congeners between 3.28 and pg WHO-TEQ g 1, while the other fish samples and other food groups had concentrations under 2 pg WHO-TEQ g 1. These results suggest the utility of analysing the PCB-7 as a routine analytical methodology in a monitoring programme managed by a public administration, with food safety and control responsibilities, limited budget and time, to detect samples that presumably present levels of toxic PCDD/F and PCBs congeners above 3 pg WHO-TEQ g 1. Conclusions This study confirms that the use of PCB-7 as a marker compound in a surveillance programme is a viable and practical strategy to detect samples with the highest levels of dioxin and PCB contamination. Seventeen toxic congeners of dioxins and twelve dioxin-like PCBs were studied in 24 food samples acquired from food retailers in Barcelona, Spain. Some presence of these PCDD/Fs and dioxin-like PCBs was detected in all samples, although highest levels were in fish samples. PCB-7 used as markers were detected in five fish samples. These samples had concentrations of PCDD/Fs and dioxin-like PCBs much higher than other samples. Using the PCB-7 indicators as a surveillance system to identify foodstuffs with higher dioxin and PCB contamination, all samples with a dioxin and dioxin-like PCBs concentration above 3 pg WHO-TEQ g 1 would be detected. As current regulations for fish define as acceptable levels up to 8 pg WHO-TEQ g 1, this may be a practical surveillance system for a public health administration which may not have access to all the needed research instruments for the continuous and detailed monitoring in food of all these environmental chemicals. Acknowledgements This project received partial funding from the Instituto de Salud Carlos III (Fondo de Investigacio n Sanitaria, PI and CIBERESP). References Abad E, Caixach J, Rivera J Application of DB-5ms gas chromatography column for the complete assignment of 2,3,7,8-substituted polychlorodibenzop-dioxins and polychlorodibenzofurans in samples from municipal waste incinerator emissions. J Chromatogr A. 786: Abad E, Saulo J, Caixach J, Rivera J Evaluation of a new automated cleanup system for the analysis of polychlorinated dibenzo-p-dioxins and dibenzofurans in environmental samples. J Chromatogr A. 893: Bayarri S, Baldassarri LT, Lacovella N, Ferrara F, Domenico A PCDDs, PCDFs, PCBs and DDE in edible marine species from the Adriatic Sea. Chemosphere. 43: Bocio A, Domingo JL, Falco G, Llobet JM Concentrations of PCDD/PCDFs and PCBs in fish and seafood from the Catalan (Spain) market: estimated human intake. Environ Int. 33:

8 Bordajandi LR, Martin I, Abad E, Rivera J, Gonzalez MJ Organochlorine compounds (PCBs, PCDDs and PCDFs) in seafish and seafood from the Spanish Atlantic Southwest Coast. Chemosphere. 64: Borrell A, Aguilar A. (Forthcoming) Organochlorine concentrations declined during in western Mediterranean bottlenose dolphins, a coastal top predator. Chemosphere. 66(2): Covaci A, Ryan JJ, Schepens P Patterns of PCBs and PCDD/F in chicken and pork fat following a Belgian food contamination incident. Chemosphere. 47: Easton MDL, Luszniak D, Der Geest E Preliminary examination of contaminant loadings in farmed salmon, wild salmon and commercial salmon feed. Chemosphere. 46: European Commission. 2001a. Opinion of the Scientific Committee on Food on the risk assessment of dioxins and dioxin-like PCBs in food. Brussels: European Commission Health & Consumer Protection Directorate- General. p. 29. European Commission. 2001b. Reports on tasks for scientific cooperation: Assessment of dietary intake of dioxins and related PCBs by the population of EU Member States. Brussels: European Commission Health & Consumer Protection Directorate-General. p European Commission Commission Regulation 1881/ 2006 of 19 December setting maximum levels for certain contaminants in foodstuffs. Off J Eur Union L 364, European Parliament and European Council European Regulation 850/2004 of the European Parliament and of the Council of 29 April 2004 on persistent organic pollutants. Off J Eur Union L 158, Fernandez MA, Gomara B, Bordajandi LR, Herrero L, Abad E, Abalos M, Rivera J, Gonzalez MJ Dietary intakes of polychlorinated dibenzo-p-dioxins, dibenzofurans and dioxin-like polychlorinated biphenyls in Spain. Food Addit Contam. 21: Ferrante MC, Cirillo T, Naso B, Clausi MT, Lucisano A, Cocchieri RA Polychlorinated biphenyls and Food Additives and Contaminants: Part B 73 organochlorine pesticides in seafood from the Gulf of Naples (Italy). J. Food Prot. 70: Focant JF, Eppe G, Pirard C, Massart AC, Andre JE, De Pauw E Levels and congener distributions of PCDDs, PCDFs and non ortho PCBs in Belgian foodstuffs assessment of dietary intake. Chemosphere. 48: Gomara B, Bordajandi LR, Fernandez MA, Herrero L, Abad E, Abalos M, Rivera J, Gonzalez MJ Levels and trends of polychlorinated dibenzo-p-dioxins/ furans (PCDD/Fs) and dioxin-like polychlorinated biphenyls (PCBs) in Spanish commercial fish and shellfish products, J Agric Food Chem. 53: Hansen LG Stepping backward to improve assessment of PCB congener toxicities. Environ Hlth Perspect. 106: King K Compilation of EU dioxin exposure and health data. Task 4: Human exposure. Compilation of EU dioxin exposure and health data. Brussels: European Commission. Llobet JM, Bocio A, Domingo JL, Teixido A, Casas C, Muller L Levels of polychlorinated biphenyls in foods from Catalonia, Spain: estimated dietary intake. J Food Prot. 66: National Center for Environmental Health Second national report on human exposure to environmental chemicals. Atlanta, GA: National Center for Environmental Health. Shaw SD, Brenner D, Berger ML, Carpenter DO, Hong CS, Kannan K PCBs, PCDD/Fs, and organochlorine pesticides in farmed Atlantic salmon from Maine, eastern Canada, and Norway, and wild salmon from Alaska. Environ Sci Technol. 40: Spanish Ministry of Environment National Implementation Plan for the Stockholm Convention. Madrid: Spanish Ministry of Environment. United Nations Environment Programme Stockholm convention on persistent organic compounds. New York, NY: United Nations. Vicente A, Arque s JF, Villalbı JR, Centrich F, Serrahima E, Llebaria X Pesticides in the diet: adding pieces to the puzzle. Gac Sanit. 18: