Dioxin levels in cattle hide processed using different fuel sources for human consumption in Sokoto Central Abattoir, Sokoto State, Nigeria

Umar Zakariyau^{1, &}, Lawal Amadu¹, Junaidu Kabir², Shakir Bolugun³, Aboyowa Edukugho³, Lawali Bello Yahaya⁴

 

¹Nigerian Field Epidemiology and Laboratory Training Program, Abuja, Nigeria, ²Department of Veterinary Public Health and Preventive Medicine Ahmadu University Zaria, Kaduna, Nigeria, ³African Field Epidemiology Network, Abuja, Nigeria, ⁴Ministry of Animal Health and Fisheries Development Sokoto, Sokoto, Nigeria

 

 

^&Corresponding author
Umar Zakariyau, Nigerian Field Epidemiology and Laboratory Training Program.

 

 

 

 

In Nigeria, more than 90% of bovine hides produced in abattoirs are processed into a product called ponmo for human consumption. These hides are subjected to burning using different fuel sources to burn off the hairs from the hide. During this process, a lot of hazardous substances are used as fuel to sustain the flame, thus releasing dangerous substances to the hides, making them hazardous to humans, animals and the environment [1]. These hazardous substances include heavy metals, aromatic hydrocarbons, dioxins and furans[2].

 

Dioxins belong to a group of polychlorinated dibenzo-p-dioxins/furans (PCDD/Fs) which are halogenated aromatic hydrocarbons [3]. The basic structure of PCDD/Fs comprises two benzene rings joined by either a single (furan) or a double oxygen bridge (dioxin)[4]. They are formed as a result of incomplete combustion of hydrocarbons in the presence of chlorine such as metal processing, and domestic waste incineration [5], including municipal, hospital, and industrial waste. These are the most significant combustion sources of PCDDs and PCDFs [6]. They can also be formed from chemically unrelated compounds such as polyvinyl chloride (PVC) or other chlorocarbons and the burning of non-chlorinated organic matter such as polystyrene, cellulose, lignin, and coal in the presence of organic chlorine donors[7].Wood also contains chloride compounds that are oxidized under high heat, producing chlorine that readily reacts with organic compounds to form organochlorines, including dioxins[8].

 

In animals, dioxins are potent multisite carcinogens and were shown to disrupt multiple endocrine pathways [9]. Experiments examining the transgenerational effects of dioxins show that descendants of dioxin-exposed animals have an incidence of kidney disease in males, and ovarian abnormalities in females [10].

 

In humans, exposure to dioxins even in a smaller quantity may cause serious health problems such as liver problems, immunological, reproductive and hormonal systems, developmental problems and brain disorders in children [11]. Dioxins have also been found to cause testicular cancer decrease spermatogenesis, hormonal dysfunction, undescended testicles and endometriosis [12].

 

Environmental studies show that pollutants are released from burning plastic waste in a burn barrel which will be transported through the air either short or long distances, and are then deposited onto land or into bodies of water [13]. These pollutants such as mercury, polychlorinated biphenyls (PCBs), dioxins and furans persist for long periods in the environment and have a tendency to bio-accumulate which means they build up, at the top of the food web. Bioaccumulation of pollutants usually occurs indirectly through contaminated water and food rather than breathing the contaminated air directly [14].

 

Prevention and reduction of dioxin release into the environment are achieved through food safety management practices like Hazard Analysis Critical Control Point (HACCP), which targets chemical contaminants throughout the production process, beginning with raw materials [15].

 

There is an overall increase in cancer-related morbidity in Nigeria [16].

 

The impact of foods, animal products, and agriculture-related activities on cancer morbidity in Nigeria remains unexplored. Identifying critical intervention points within the food production system is essential for enhancing food safety [17]. The study aims to investigate the impact of fuel used in the processing of hides on the levels of dioxins in bovine hide meant for human consumption, in Sokoto Central abattoir.

 

 

Study area

 

The study was conducted at Sokoto Central Abattoir in Sokoto North local government area (LGA), Sokoto State, Nigeria (Figure 1). This is the largest abattoir in the state, and houses two large slaughterhouses and employs over 500 people, including butchers, livestock sellers, and veterinary and administrative staff. On average, 140 cattle, 70-80 sheep and goats, and 10 camels are slaughtered daily.

 

Study design

 

A quasi-experimental study design was used to determine the quantity of dioxin in cattle hide processed for human consumption.

 

Sample size determination

 

The sample size was determined using the formula by Thrusfield [18].

 

 

Where; N = Sample Size

 

Z = Degree of Confidence (1.96) p = Expected Prevalence (12%)[19] = 0.1

 

q = 1-p, d = Desired absolute precision (5%)=0.05

 

N =(1.96)2 x0.12x0.88/(0.05)2 =0.4029/0.0025 = 161.1

 

Sampling technique

 

A systematic random sampling technique was used to collect hide samples, On average a total of 140 cattle of both sexes are slaughtered daily. The sampling frame was 1960 cows and the first cow was selected at ramdom while the other cows were sampled at an interval of 12, over a period of 14 days and sample size per day was 12.

 

Hide sample collection

 

The 160 samples collected were allocated evenly across three distinct groups. The samples were evenly distributed across the three groups: Tyre, Plastic, and Firewood. Each group was further divided into two subgroups: Control and Test. The sample allocation was as follows: Tyre group: 54 samples (27 Control, 27 Test), Plastic group: 53 samples (26 Control, 27 Test) and Firewood group: 53 samples (26 Control, 27 Test). These were then processed using a variety of materials, specifically tyres, plastic or nylon bags, and firewood, to evaluate the differential effects of each. This methodical distribution was done to ensures a balanced comparison across the different processing materials

 

A 5 cm square hide sample was cut from cattle that was obtained using systematic random sampling. Each sample was split into two parts: The hide samples designated for the control group underwent a boiling water treatment, followed by the use of a knife to remove the hairs.and the other part was singed using either tyres, plastics, or firewood. After treatment, the samples were labeled and stored in sterile polythene bags seperately for further analysis.

 

Sample preparation and extraction

 

A sample of hide, weighing about 5 grams, was cut, measured, and then deposited into a 20 milliliter sampling vial, which was half-filled with 10 milliliters of Phosphate Buffered Saline (PBS) at a pH of 7.4. Subsequently, each sample batch underwent homogenization in an individual electric blender. The blender and cutting knife was promptly cleansed after processing each sample to prevent cross-contamination. The homogenate obtained was promptly placed into a clean, sterile container and then stored in a freezer set at -50°C. This low temperature was maintained to ensure that the rest of the sample could be homogenized effectively, and subjected to centrifugation at a speed of 3000 revolutions per minute (RPM) for a duration of 20 minutes. Following centrifugation, the clear supernatant fluid was extracted and stored in a freezer pending further analysis.

 

Detection procedure

 

We used the Enzyme-linked Immunosorbent Assay (ELISA) for Dioxin detection and it was conducted using a commercially available test kit. This kit included three essential components: a Microelisa strip plate consisting of 96 wells (12 wells per strip, 8 strips per block), and a set of dioxin standards with six concentrations ranging from 0 ppt to 0.8 ppt, provided in vials of 1 ml each. During the procedure, chromogen solutions A and B, along with the stop solution, were added to the blank well to calibrate the plate reader to zero. Subsequently, 50 μl of the standard dilution was dispensed into each well. This was followed by the addition of 50 μl of the sample dilution. In the final step we added 50 μl of the biotinylated antigen working solution to the mixture.

 

The sample well: 50ul of sample was added to the sample well and then later 50ul of working solution of biotinylated antigen was added to the wells. The ELISA Plate was then covered with a sealing membrane and gently shaken for 60 minutes at 37°C incubation.

 

During the first washing, we carefully removed the sealing membrane, the liquid from the plate was drained and dried up, each well was then filled with washing solution, kept aside for 1min and then drained, and this was repeated 5 times. We then added 50ul of avidin-HRP in standard well and sample well and covered with seal plate membrane, gently shook and mixed for 60 minutes at 37°C.

 

At the second washing the plate sealer was removed and the liquid was drained and dried,we then filled well with wash buffer and then allowed it to stand for 1min, the operations were repeated 5 times and dried by a clap.

 

Colour development: We added 50 μl of chromogen solution A to each well. Followed by adding of 50 μl of chromogen solution B to each well.

 

We then gently shook the plate and incubated for 15 minutes at 37°C, keeping it away from light for color development.

 

After that we went ahead to add 50 μl of stop solution to each well to stop the reaction, changing the color from blue to yellow.

 

The blank well was used as zero, and we then measured the absorbance (OD) of each well at 450 nm.

 

We calculated the linear regression equation of the standard curve using the standard concentrations and OD values.

 

We then determine the concentration of the samples based on their OD values.

 

Data analysis

 

Epi Info version 7.2.2.6, along with the Statistical Package for the Social Sciences (SPSS Version 20.0), was utilized to assess the variances pre- and post-processing. The level of concentration of dioxin was compared using the Paired T-test statistical method, one-way analysis of variance (ANOVA), and Post hoc Tukey's test.

 

The 'F' value in the ANOVA is the ratio of the mean square value between the groups to the mean square value within the groups. This value is used to determine the p-value, which tells us the probability that the observed results could have occurred by chance

 

The study applied a post hoc test (Tukey HSD) to analyze the mean difference in dioxin concentration before and after processing in three groups.

 

Ethical consideration

 

Animal welfare considerations were duly noted prior to the collection of hide samples. This is because the procurement of samples occurred posthumously, following the cows' slaughter and skinning processes. Furthermore, the research received authorization from the Ethical Approval Committee of the Ministry of Animal Health and Fisheries Development in Sokoto, Nigeria, under the approval number MAHFD/VET/190.

 

 

Table 1: we observe the following:

• Tyres: The dioxin concentration increased significantly from 0.679 PPT before processing to 26.305 PPT after processing.
• Plastics: The dioxin concentration increased from 0.263 PPT before processing to 2.366 PPT after processing.
• Firewood: The dioxin concentration increased from 0.501 PPT before processing to 8.829 PPT after processing.

 

The results of the bivariate analysis, conducted using the paired T-test, indicated that the groups utilizing tyres as a fuel source were the only ones that showed statistical significance, with a P-value of 0.04 (Table 2). In contrast, the remaining groups did not achieve statistical significance as their P-values exceeded the 0.05 threshold (values of 0.27 and 0.301, respectively).

 

The level of dioxin and confidence interval for hides processed using tyres was 11.9(3.9-20.05) respectively. The confidence interval for hides processed using plastics was 0.45(-1.2-0.37),also showing statistical significance. However, this interval includes zero, indicating a minimal or uncertain contribution to dioxin levels. The confidence interval for hides processed using firewood was 3.01(-8.8, 2.7), indicating no statistical significance. This means using firewood does not contribute to the dioxin levels in hides processed for human consumption.

 

The study find out that using tyres to singe hides significantly increases the amount of dioxin in the hides, as indicated by a statistically significant p-value of 0.04 (less than 0.05). In contrast, the p-values for hides processed using plastics (0.27) and firewood (0.301) were both greater than 0.05, indicating that these methods do not significantly contribute to dioxin levels in the hides (Table 3).

 

In this studies, an F value of 4.415 and a p-value of 0.015 suggest that there is a statistically significant difference between the group means at a conventional significance level. This means that it's very unlikely that the observed differences in means are due to random chance alone.

 

Analysis of variance (ANOVA) shows that the difference between and within the group was statistically significant, at a P=0.015 (Table 3).

 

The results, summarized in Table 4, showed:

• A statistically significant difference in dioxin concentration between hides processed with plastics and tires (p=0.017, p<0.05).
• No statistically significant difference between hides processed using firewood and tires, or between firewood and plastics (p>0.05).

 

 

The study determined the contribution of processing methods to the quantity of dioxin in bovine hide intended for human consumption using different fuel sources. All the three studied fuels resulted in a substantial increase in dioxin concentration, with tyres showing the highest increase, followed by firewood and then plastics. This suggests that burning tyres as a source of heat leads to the highest dioxin emissions among the three fuels listed.

 

In 2007, the World Health Organization (WHO) and the Food and Agricultural Organization (FAO) developed a toxic equivalent quantity (TEQ) range for dioxins, which spans from 0.0001-1.0 parts per trillion (ppt)[20]. The study found that the baseline concentration of dioxin in the control group (which was not singed or smoked) was, within the normal TEQ range set by WHO/FAO in 2007. This dioxin presence is likely due to livestock grazing on grasses contaminated by tyre and plastic smoke or pesticides used on farms [21]. Additionally, environmental sources such as soil and water that the cows came into contact with or drank from may have contributed to the dioxin levels detected in the control group [1]. The study found that the concentration of dioxin in hides processed with tyres, have exceeded the WHO/FAO TEQ limits set in 2007.

 

Bivariate analyses using a Paired T-test showed this concentration to be statistically significant, indicating a high dioxin level in tyre-processed hides [3, 5]. These results align with similar studies that found high concentrations of polycyclic aromatic hydrocarbons in roasted cowhide, though those studies did not show statistically significant differences. Additionally, this study concurs with research showing higher dioxin concentrations in poultry meat roasted with tires [22]. The high dioxin levels in the hides may be due to inadequate washing during processing before analysis. This aligns with another study's explanation that low PAH contamination in smoke might be because the compounds were trapped in the cowhide's pores after roasting [1]. The dioxin concentration in hides processed using plastics/nylon bags was, slightly higher than the value set up by WHO/FAO TEQ range of 0.0001-1.0 pp [23, 24]. This study agrees with research findings that burning plastics containing organochloride-based substances, such as polyvinyl chloride, releases high concentrations of dioxin [25]. Bivariate analysis using a paired T-test showed a p-value (p>0.05)., indicating no statistical significance. However, the quantity of dioxin detected aligns with Swedish research findings that large amounts of polycyclic aromatic hydrocarbons are produced annually by the rubber industry [26].

 

Limitations

 

One of the limitations of this study is the use of a sample size formula (N = Z2pq/d2) that is more suitable for survey/cross-sectional studies, rather than a formula that accounts for the expected effect size between intervention and control groups.

 

However, the impact of this limitation is mitigated by the use of appropriate statistical analysis and the clear reporting of the sample size and allocation.

 

 

This study indicated that cattle hide processed using tires has high dioxin content, compared to other methods and is potentially toxic to humans, Animals and the environment

 

Recommendation

 

There is need for Government regulatory agencies to enforce food safety laws that will protect the public from the practices of used of tires and plastics to singe hides.

 

There is need to educate meat processors and abattoir workers on the hazards to themselves, food safety and to environment.

 

There is need to educate the public on dangers of using tires to singe hide and consumptions of meat product processed using tires, and plastics materials.

 

 

The authors declare no competing interests.

 

 

Umar Zakariyau and Dr Lawal Amadu: contributed to design of the study, data collection, and analysis and writing of the manuscript; Junaidu Kabir offered his expertise in veterinary public health and supervised the Conduct of the study design, data collection, and analysis; Shakir Bolugun and Aboyowa Edukagho: offered A technical assistance and resources for the study; Lawali Bello Yahaya: granted permission for the study to be conducted and potentially provided access to the abattoir and animal samples.

 

 

We would like to acknowledge Ministry of animal health and fisheries development Sokoto for given us ethical approval and permission to use the abattoir as our study site. We also appreciate the African Field Epidemiology Network (AFENET) for funding our travel to present this work at the 8th AFENET Conference in Mombasa Kenya.

 

 

Table 1: Sociodemographic characteristics of study participants (n=148)

Table 2: Results of bivariate Analysis Using Paired T-test Statistical Methods

Table 3: Effect of mobile health intervention on perceptions of cervical cancer

Table 4: Result of Post hoc test using Turkey HSD

Figure 1: Map of Sokoto state showing location of the LGA where the abattoir is located

 

 

1. Oko OJ, Okoye COB. Quantification of Smoke Contributed PAHs in Roasted Cowhide (Ponmo) from Northern Nigeria. FUW Trends Sci Technol J [Internet]. 2017 April [cited 2024 Dec 18];2(1):55-9. Download 21Article 1.pdf.


2. Adam I, Teye M, Daniel O. Assessment of Heavy Metal Residues in Hides of Goats Singed with Tyres, and the Effect of Boiling on the Heavy Metal Concentrations in the Hides . J Vet Adv [Internet]. 2013 May 28 [cited 2024 Dec 18];3(5):165. https://doi.org/10.5455/jva.20130531104440 Download Assessment of Heavy Metal Residues in Hides of Goats Singed.pdf. Google Scholar


3. Reiner EJ, Clement RE, Okey AB, Marvin CH.Advances in analytical techniques for polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans and dioxin-like PCBs . Anal Bioanal Chem [Internet]. 2006 Jun 23 [cited 2024 Dec 18];386(4):791-806. https://doi.org/10.1007/s00216-006-0479-1 PubMed | Google Scholar


4. Lin DY, Lee YP, Li CP, Chi KH, Liang BW, Liu WY, Wang CC, Lin S, Chen TC, Yeh KJ, Hsu PC, Hsu YC, Chao HR, Tsou TC. Combination of a fast cleanup procedure and a DR-CALUX® bioassay for dioxin surveillance in Taiwanese soils . IJERPH [Internet]. 2014 May 6 [cited 2024 Dec 18];11(5):4886-90. https://doi.org/10.3390/ijerph110504886 PubMed | Google Scholar


5. Dada EO, Osilagun HO, Njoku KL.Physicochemical and Genotoxic Evaluations of Singed Cowhide Meat (Ponmo) Wastewater . Journal of Health and Pollution [Internet]. 2018 Dec 1 [cited 2024 Dec 18];8(20):181207. https://doi.org/10.5696/2156-9614-8.20.181207 PubMed | Google Scholar


6. Oluwole Oladele I.Effect of Chemical Treatments on the Physicochemical and Tensile Properties of Cow Hair Fibre for Low Load Bearing Composites Development. IJMSA [Internet]. 2015 May 15 [cited 2024 Dec 18];4(3):189. https://doi.org/10.11648/j.ijmsa.20150403.16 Google Scholar


7. Kim M, Kim DG, Choi SW, Guerrero P, Norambuena J, Chung GS. Formation of polychlorinated dibenzo-p-dioxins/dibenzofurans (PCDD/Fs) from a refinery process for zinc oxide used in feed additives: a source of dioxin contamination in Chilean pork . Chemosphere [Internet]. 2011 Jan 8[cited 2024 Dec 18];82(9):1225-9. https://doi.org/10.1016/j.chemosphere.2010.12.040 Google Scholar


8. Adamse P, Schoss S, Theelen RMC, Hoogenboom RLAP.Levels of dioxins and dioxin-like PCBs in food of animal origin in the Netherlands during the period 2001-2011. Food Additives & Contaminants: Part A [Internet]. 2016 Nov 17 [cited 2024 Dec 18];34(1):78-92. https://doi.org/10.1080/19440049.2016.1252065 Google Scholar


9. Warner M, Mocarelli P, Samuels S, Needham L, Brambilla P, Eskenazi B.Dioxin Exposure and Cancer Risk in the Seveso Women´s Health Study. Environ Health Perspect [Internet]. 2011 Aug 2[cited 2024 Dec 18];119(12):1700-5. http://doi.org/10.1289/ehp.1103720 PubMed | Google Scholar


10. Fürst P, Bernsmann T, Baumeister D, Sandy C. Determination of Dioxin-Like and Non-Dioxin-Like Polychlorinated Biphenyl Congeners in Foodstuffs and Animal Feed Using the Agilent 7000 Triple Quadrupole GC/MS System [Internet]. Agilent Technologies, Inc. (USA); 2011 Mar 2011 [cited 2024 Dec 18]; 11 p. Application Note 5990-6950.


11. Cheung YH, Neller A, Chu KH, Tam NFY, Wong CK, Wong YS, Wong MH. Assessment of sediment toxicity using different trophic organisms . Archives of Environmental Contamination and Toxicology [Internet]. 1997 Sep 1 [cited 2025 Jan 7];32(3):260-7.https://doi.org/10.1007/s002449900183 Purchase or subscription required to access full text. Google Scholar


12. Tanyildizi S, Çıftçı O. The Determination of Dioxin and Dioxin-like Compound Levels in Beef and Chicken Meat Samples Consumed in Turkey . Kafkas Univ Vet Fak Derg [Internet]. 2010 [cited 2025 Jan 7]; 16 (Suppl-A): S7-S11. Download pdf to view full text. https://doi.org/10.9775/kvfd.2009.792 Google Scholar


13. Institute of Medicine (US) Committee on the Implications of Dioxin in the Food Supply.Dioxins and Dioxin-like Compounds in the Food Supply: Strategies to Decrease Exposure [Internet]. Washington (DC): National Academies Press (US); 2003 [cited 2024 Dec 19]. 318 p. https://doi.org/10.17226/10763 Google Scholar


14. Pereira MD.Polychlorinated dibenzo-p-dioxins (PCDD), dibenzofurans (PCDF) and polychlorinated biphenyls (PCB): main sources, environmental behaviour and risk to man and biota . Química Nova [Internet]. 2004 Aug 9 [cited 2025 Jan 7]; 27 (6):934-43. Download 17-RV03179.p65.pdf. Google Scholar


15. Kim M.17-Chemical contamination of red meat . In: Schrenk D, editor. Chemical Contaminants and Residues in Food [Internet]. Sawston(UK):Woodhead Publishing; 2012 [Available online 2014 Mar 27; cited 2025 Jan 23]. p. 447-68. (Woodhead Publishing Series in Food Science, Technology and Nutrition) https://doi.org/10.1533/9780857095794.3.447 Subscription or purchase required to view full text. Google Scholar


16. Ogbonna ILP, Nwaocha KB, Appl A, Res S.Determination of levels of polycyclic aromatic hydrocarbons on singed cow hide ( Punmo ) and charcoal grilled meat ( Suya ) . Archives of Applied Science Research [Internet]. 2015 [cited 2024 Dec 19];7(4):1-6. Google Scholar


17. Abbas II, Sakwe QE, Iguisi EO.Dioxin emission and industrial solid waste in Kano metropolis, Nigeria. EJERM [Internet]. 2016 Dec [cited 2025 Jan 7];7(2):029-37. http://dx.doi.org/10.18685/EJERM(7)2_EJERM-16-017 Download 1480653564_Innocent et al.pdf. Google Scholar


18. World Health Organization. Evaluation of certain food additives and contaminants: Thirty-seventh report of the Joint FAO/WHO Expert Committee on Food Additives [Internet]. Geneva (Switzerland): World Health Organization; 1991[cited 2025 Jan 7]. 49 p. Download WHO_TRS_806.pd. Google Scholar


19. World Health Organization. Dioxins [Internet]. Geneva (Switzerland): World Health Organization; 2023 Nov 29[cited 2025 Jan 7].


20. United States Congress, Office of Technology Assessment. : Environmental Contaminants in Food[Internet]. Washington (DC): 1979 Dec. Chapter 2: Environmental Contamination of Food; [cited 2025 Jan 7]; p. 15-31. Library of Congress Catalog Card Number 79-600207.


21. Zhang M, Buekens A, Li X.Open burning as a source of dioxins . Critical Reviews in Environmental Science and Technology [Internet]. 2017 Jul 7 [cited 2025 Jan 7];47(8):543-620. https://doi.org/10.1080/10643389.2017.1320154 Google Scholar


22. Van Larebeke N. Food contamination with polychlorinated biphenyls and dioxins in Belgium. Effects on the body burden. Journal of Epidemiology & Community Health [Internet]. 2002 Nov 1 [cited 2024 Dec 19];56(11):828-30. https://doi.org/10.1136/jech.56.11.828 PubMed | Google Scholar


23. Word Health Organization. Assessment of the health risk of dioxins: re-evaluation of the Tolerable Daily Intake (TDI): Executive Summary [Internet]. Geneva (Switzerland): World Health Organization; 1998 May 29 [cited 2025 Jan 7]. 28 p Download exe-sum-final_may1998.pdf. PubMed | Google Scholar


24. Van Den Berg M, Birnbaum LS, Denison M, De Vito M, Farland W, Feeley M, Fiedler H, Hakansson H, Hanberg A, Haws L, Rose M, Safe S, Schrenk D, Tohyama C, Tritscher A, Tuomisto J, Tysklind M, Walker N, Peterson RE. The 2005 World Health Organization Reevaluation of Human and Mammalian Toxic Equivalency Factors for Dioxins and Dioxin-Like Compounds . Toxicological Sciences [Internet]. 2006 Jul 7 [cited 2024 Dec 19];93(2):223-41. https://doi.org/10.1093/toxsci/kfl055 PubMed | Google Scholar


25. World Health Organization. Food Safety [Internet]. Geneva (Switzerland): World Health Organization; 2024 Oct 4[cited 2025 Jan 7].


26. Boström CE, Gerde P, Hanberg A, Jernström B, Johansson C, Kyrklund T, Rannug A, Törnqvist M, Victorin K, Westerholm R.Cancer risk assessment, indicators, and guidelines for polycyclic aromatic hydrocarbons in the ambient air . Environ Health Perspect [Internet]. 2002 Jun 1[cited 2024 Dec 19];110(suppl 3):451-88. https://doi.org/10.1289/ehp.110-1241197 PubMed | Google Scholar