The effect of gas versus charcoal open flames on the induction of polycyclic aromatic hydrocarbons in cooked meat: a systematic review and meta-analysis

Abstract

Purpose

Open flames of gas and charcoal can induce polycyclic aromatic hydrocarbons (PAHs) in cooked meat. The current study aimed to compare the effect of gas and charcoal open flames on the induction of PAHs in cooked meat using a meta-analysis approach.

Methods

A systematic review of the literature was conducted electronically based on the PRISMA guidelines. Experimental studies comparing the PAHs content of cooked meat over open flames of gas and charcoal were searched using the appropriate keywords until June 2018.

Results

Of 1137 papers retrieved, 7 with a total sample size of 474 meat samples were used in the meta-analysis. The mean difference (MD) between the gas and charcoal cooking methods in the induction of each PAH was 2.053 μg/Kg. (95%CI: 1.022–3.085 μg/Kg; P < 0.001). The subgroup analysis of 17 trials indicated the difference between the two cooking methods increases when red meat rather than white meat is cooked (MD in red meat: 3.499 μg/Kg; 95%CI: 2.030–4.967; P < 0.0001 vs. MD in white meat: 3.319 μg/Kg; 95% CI: 1.689–4.950; P < 0.0001). Interestingly, studies that analyzed meat samples for fewer PAHs (cut-off ≤7) found a much wider difference between gas and charcoal-cooked meat (MD: 5.106 μg/Kg; (95% CI: 2.162–8.049; P < 0.001 in studies with ≤7 PAHs vs. MD: 1.447 μg/Kg; 95% CI: 0.628–2.266; P < 0.001 in studies with >7 PAHs).

Conclusions

It is necessary to avoid open flames of charcoal as the heat source or change the geometry of charcoal-fired cookstoves to prevent fat dripping on the fire and thus, excessive PAHs induction.

Electronic supplementary material

The online version of this article (10.1007/s40201-020-00457-0) contains supplementary material, which is available to authorized users.

Introduction

Polycyclic aromatic hydrocarbons (PAHs) are a group of organic compounds composed of at least two aromatic rings. As persistent pollutants, PAHs are known by the International Agency for Research on Cancer (IARC) for their probable carcinogenicity and mutagenicity to humans [1]. The International Union of Pure and Applied Chemistry (IUPAC) has identified more than 100 PAHs [2], 16 of which have been listed by JECFA (the Joint FAO/WHO Expert Committee on Food Additives) as genotoxic and carcinogenic compounds [3].

Although they are naturally occurring, the most important way of exposure to PAHs for non-smokers and non-occupationally-exposed individuals is through the ingestion [4]. The dietary exposure is very common when PAHs-containing food is consumed. Many studies have shown cooking through smoking [5, 6], grilling [7, 8], barbecuing [9, 10], roasting [11, 12], and frying [13] can lead to the formation of PAHs in meat and meat products.

The level of PAHs may vary from zero to several hundred micrograms per kilogram of the edible portion of cooked meat [14]. The factors affecting the concentration of PAHs in cooked meat include the type of meat, fat content of meat, cooking method (frying, grilling, roasting, boiling, smoking, etc.), temperature and duration of cooking, the type of heat supply (electricity, gas, wood, and charcoal), distance to the heat source, and the type of contact with the heat source (direct or indirect) [4, 15, 16].

The exact mechanism of PAH induction during meat processing is still unknown. However, free radical reactions, intramolecular addition, and polymerization of small molecules have been mentioned in the literature [12, 17]. Undoubtedly, the high temperature in food processing has a contribution to the PAH formation. The pyrolysis of the organic content of meat including fat, proteins, and carbohydrates may occur especially at high temperatures (above 200 °C) [4]. In addition, the direct contact of lipids dripping on the heat source is another major contributor, resulting in the generation of volatile PAHs rising to adhere to the surface of the meat placed over the heat source [12]. The incomplete combustion of fuel used for heat supply, especially in the case of charcoal, is also responsible for PAH occurrence in food [8].

The consumption of grilled meat is increasingly popular worldwide at home, outdoor, and in restaurants because of its unique taste [18]. Charcoal is traditionally used for heat supply in grill ovens since it is an easy-available and low-cost fuel [19, 20]. In addition, charcoal grilling is usually associated with the improved taste of meat and specific aroma. However, charcoal grilling is an intense thermal process that can intensify PAH formation. In recent decades, due to the increased availability, natural gas (methane, propane, and butane) has been also used as fuel in meat cooking [21].

Concerning the popularity of both gas and charcoal heating in meat cooking, it seems necessary to compare them in the view of food safety. A literature review showed that there are some studies aiming to compare the PAH content of gas and charcoal-cooked meat by utilizing various methodologies, with reporting different results. To shed light on the issue, the current study aimed to conduct a systematic review and meta-analysis of experimental studies comparing the PAH content of cooked meat prepared on open flames of gas and charcoal.

Methods

Study protocol

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [22] was used for reporting the current study (see Supplemental File appendix 1). The study protocol was registered in the international prospective register of systematic reviews database (PROSPERO) on June 2018 with registration number CRD42018098525.

Search strategy

A systematic search was done to find studies comparing the effects of gas and charcoal cooking on the levels of PAHs in cooked meat. To this aim, we searched Web of Knowledge (ISI), MEDLINE, ScienceDirect, and Scopus using the following MeSH and non-MeSH terms in title, abstract, or keywords until June 2018, with no limitation in time: (Pork OR Meat OR Fish OR Beef OR Chicken OR Duck OR Barbecue OR Kebab OR Kokoretsi OR Salmon OR Lamb OR Steak OR Veal OR Rib OR Sausage OR Burger) AND (Gas OR Flame) AND (Charcoal OR Char) AND (“Polycyclic Aromatic Hydrocarbon” OR PAH OR “Polynuclear Aromatic Hydrocarbon” OR “Polyaromatic Hydrocarbon” OR “Aromatic Polycyclic Hydrocarbon” OR “Polycyclic Aromatic Compound”).

Inclusion and exclusion criteria

Original studies with an experimental design in English dealing with the quantification of PAHs in gas or charcoal-cooked meat on open flames were included in this study. Studies not comparing gas and charcoal as the heat source in the induction of PAHs, not measuring PAHs content in the same meat sample, not analyzing ready-to-eat meat, taking just one sample size, and with non-extractable data were excluded from the study. The retrieved papers from electronic searching were independently reviewed by two authors (FBA and HN) for the inclusion and exclusion criteria and disagreements were solved by consulting with a third author (MD). The reference lists of the retrieved studies were also reviewed manually for possibly related papers that were not found in the electronic search.

Data extraction

After selecting the relevant papers, two independent researchers (MD and MGh) read the full texts carefully and extracted the following data: author, publication year, country of origin, name of the journal, the type of meat, the number of PAHs determined in the samples, sample size, and the mean and SD of PAHs in the cooked meat. The extracted data were compared between examiners and any disagreement was discussed to reach consensus and ensure the high accuracy of data. If a study did not reflect enough information, we would contact the authors by sending e-mails two times, two-week apart. In some cases, we contacted the authors by phone.

Taking a glance at the extracted data, we found out that the studies were heterogeneous in terms of the number of PAHs measured in meat samples. Therefore, there was a wide variation between studies in terms of the sum of PAHs, making it impossible to obtain a reliable pooled effect based on the sum of PAHs. Thereby, we divided the sum of PAHs measured in cooked meat samples by the number of measured PAHs to calculate mean PAH, as follows:

$$\mathit{\text{Mean}}\mathit{PAH}=\frac{\sum _{i=1}^n{\mathit{PAH}}_s}{n}$$ 1

where n is the number of PAHs analyzed in each meat sample. The standard error (SE) was calculated to indicate the between-sample variations.

Quality assessment

A modified version of CONSORT [23] adapted for experimental studies was used by two investigators (MGh and MD) to examine the quality of the selected studies. The tool included 50 items to assess the quality of studies in the title, abstract, introduction, materials and methods, results, discussion, references, and general principles.

Statistical analysis

The mean difference (MD) of PAHs content between gas and charcoal-cooked meat and its corresponding SE were used as the effect size for meta-analysis. A pooled estimate of MD with a 95% confidence interval (CI) was obtained using a random-effects model. The Q-statistic was used to find the heterogeneity of the studies, and it was quantified by the I² statistic. A Q-statistic with p < 0.10 or I² statistic >50% was the signs of significant heterogeneity between studies. The between-study variance was assessed by the tau-squared (Tau²) statistic [24–26].

Subgroup analyses based on the type of meat (red or white) and the number of PAHs (≤7 or > 7) determined in meat samples were done to find the possible sources of heterogeneity. Sensitivity analysis was done by removing studies from the meta-analysis one by one. The publication bias was investigated using Begg’s funnel plot and the asymmetry tests (Egger’s and Begg’s test). All statistical analyses were conducted using comprehensive meta-analysis software (CMA; version 2.2.064). P-values of ˂0.05 were considered statistically significant.

Results

Flow and characteristics of included studies

Figure 1 depicts the PRISMA 2009 flow diagram of study selection. Briefly, our search retrieved 1137 papers (38 in Pubmed, 141 in ScienceDirect, 802 in Scopus, 130 in ISI Web of Science, and 26 in the reference list of relevant studies) of which, 1097 remained after the removal of duplicates. As depicted in Fig. 1, 963 articles were excluded after titles/abstracts screening. Therefore, 134 articles remained to be carefully checked in the full-text. Of the remaining articles, 120 were excluded because of not comparing gas and charcoal-cooked meat in the induced PAHs. Moreover, seven other studies were removed because of the following reasons: not comparing the same meat (n = 2), non-extractable data (n = 2), taking just one sample size (n = 2), and not analyzing ready-to-eat meat (n = 1). In total, seven experimental studies comprising 17 trials with a total sample size of 474 meat samples were eligible to enter the systematic review and meta-analysis study [4, 12, 14, 15, 17, 27, 28].

Fig. 1.

PRISMA Flow Diagram of the selected studies

Study characteristics

The characteristics of the studies, which met the eligibility criteria of meta-analysis, are demonstrated in Table 1. The articles were published between 1997 and 2016; they were from Kuwait [4], Republic of Korea [12], Malaysia [14], Iran [15], China [17], Turkey [27], and the UK [28]. Among all included studies, the largest study had a sample size of 160 meat samples conducted by Gorji et al. [15] and the smallest one recruited 18 samples carried out by Chen, Lin [17]. Different types of meat had been analyzed in the studies. We divided them into two general groups named red meat (including lamb, beef, pork, sausage, and burger) and white meat (including chicken, salmon, and duck meat). Farhadian et al. [14], Alomirah et al. [4], Rose et al. [28], Gorji et al. [15] used both red and white meat in their studies while Terzi et al. [27], Chung et al. [12] used only red meat and Chen, Lin [17] used only white meat. The PAHs content of meat samples had been measured by either GC-MS [4, 15, 28] or HPLC [12, 14, 17, 27]. A different number of PAHs had been measured in various studies in a wide range of 1 PAH [27] to 28 PAHs [28]. The quality score of the seven studies was in the range of 56% to 86%, indicating that all studies were qualified based on the modified CONSORT scale (Table 1). The completed tool for studies included in the meta-analysis can be found in Supplemental file appendix 2.

Table 1.

The characteristics of the studies included in the meta-analysis

Row	Authors	Journal	Country of origin	Name of meat	Type of meat	No. PAHs	Gas cooking			Charcoal cooking			Quality score (%)
							Sample size	Mean PAH (μg/kg)	SD	Sample size	Mean PAH (μg/kg)	SD
1	Chen and Lin (1997)	Journal of Agricultural and Food Chemistry	China	Duck meat	White meat	16	6	6.91	1.25	12	14.88	4.33	56
2	Terzi, Ģelik et al. (2008)	Irish Journal of Agricultural and Food Research	Turkey	Beef	Red meat	1	20	5.7	0.85	20	24.2	0.48	72
3	Farhadian, Jinap et al. (2010)	Food Control	Malaysia	Chicken	White meat	3	36	3.99	0.85	36	10.96	1.89	68
				Beef	Red meat	3	18	3.78	0.47	18	34.44	4.25
4	Chung, Yettella et al. (2011)	Food Chemistry	Republic of Korea	Beef	Red meat	7	10	0.004	0.006	10	0.001	0.002	76
				Pork	Red meat	7	20	0.004	0.004	20	0.008	0.006
5	Alomirah, Al-Zenki et al. (2011)	Food Control	Kuwait	Lamb	Red meat	16	6	1.69	0.45	4	4.77	1.38	76
				Chicken	White meat	16	6	1.48	0.42	4	10.75	0.95
6	Rose, Holland et al. (2015)	Food and Chemical Toxicology	The UK	Sausage	Red meat	28	6	0.85	1.01	6	3.29	1.05	86
				Burger	Red meat	28	6	15.51	5.44	6	23.27	8.29
				Beef	Red meat	28	6	0.69	0.23	6	1.97	0.67
				Lamb	Red meat	28	2	0.57	0.41	2	4.54	1.22
				Pork	Red meat	28	2	0.79	0.91	2	2.78	0.37
				Chicken	White meat	28	6	0.29	0.22	6	1.89	0.99
				Salmon	White meat	28	6	1.05	0.7	6	3.48	0.81
7	Gorji, Ahmadkhaniha et al. (2016)	Food Control	Iran	Chicken	White meat	15	40	0.67	0.15	40	0.93	0.22	76
				Beef	Red meat	15	40	0.71	0.19	40	0.89	0.21

Effects of gas or charcoal open flames on the levels of PAHs in cooked meat

To combine studies (n = 7) and estimate the pooled effect, we first selected studies as the unit of analysis. The results showed significant heterogeneity between the studies comparing the effects of gas and charcoal on the induction of PAHs in cooked meat (Q-statistic P < 0.001, I² = 94.57%). Therefore, a random-effects model was used for data analysis. The forest plot summarizing the meta-analysis of comparing the effects of gas and charcoal cooking on the levels of PAHs in cooked meat samples is illustrated in Fig. 2. As can be seen, the highest difference between gas and charcoal cooking in PAHs induction in meat was reported by Terzi et al. [27] as 26.802 μg/Kg and the lowest one was found by Chung et al. [12] as 0.310 μg/Kg, both in favor of charcoal cooking. The meta-analysis of data derived from seven studies showed that charcoal cooking induced significantly higher concentrations of PAHs in the cooked meat than gas cooking did (MD: 2.053 μg/Kg; 95% CI: 1.022 to 3.085; P < 0.001). This effect size was robust in the leave-one-out sensitivity analysis; that is, the removal of each study did not change the significance of the pooled effect (Fig. 3).

Fig. 2.

The Forest plot of the effects of gas or charcoal on the levels of PAHs in cooked meat

Fig. 3.

Forest plot displaying the leave-one-out sensitivity analysis

Subgroup analysis

The type of meat is an important factor that might play a significant role in the effects of gas and charcoal heating on PAHs production in meat during open-flame cooking. For subgrouping based on the type of meat, we first used trials (n = 17) as the unit of analysis. The results of subgroup analysis based on the type of meat are shown in Fig. 4. As can be seen, 11 out of 17 trials dealt with the analysis of red meat and 6 trials investigated white meat. When the studies were categorized based on the type of meat, it was observed that red meat generated a wider difference between gas and charcoal cooking (MD in red meat: 3.499 μg/Kg; 95% CI: 2.030, 4.967; P < 0.0001 vs. MD in white meat: 3.319 μg/Kg; 95% CI: 1.689, 4.950; P < 0.0001).

Fig. 4.

The forest plot of the effects of gas or charcoal on the levels of PAHs in cooked meat according to the type of meat

Another factor that had a significant role in the effects of gas and charcoal heating on PAHs production in meat during open-flame cooking was the number of PAHs measured in each study. The results of subgroup analysis based on the number of PAHs (≤7 or > 7) are shown in Fig. 5. When the studies were categorized in accordance with the number of PAHs, the MD of studies that measured ≤7 PAHs in each meat sample was 5.106 μg/Kg (95% CI: 2.162, 8.049; P < 0.001) and the MD of studies that measured >7 PAHs was 1.447 μg/Kg (95% CI: 0.628, 2.266; P < 0.001).

Fig. 5.

The forest plot of the effects of gas or charcoal on the levels of PAHs in cooked meat according to the number of PAHs

Publication Bias

The visual inspection of the funnel plot did not suggest a significant potential publication bias in the meta-analysis of the difference between gas and charcoal in the induction of PAHs in cooked meat (Fig. 6). This observation was confirmed by the Egger’s linear regression (intercept = 5.66; S.E. = 2.43; 95% CI: − 0.60, 11.92; t = 2.32; df = 5; two-tailed P = 0.068) and Begg’s rank correlation (Kendall’s Tau with continuity correction = 0.33; z = 1.05; two-tailed P value = 0.29) tests. Duval and Tweedie ‘trim-and-fill’ correction resulted in the imputation of three potentially missing studies and an adjusted effect size of 0.96 μg/Kg (95% CI: − 0.16 to 2.08). The result of the ‘fail-safe N’ test showed that 301 studies would be required to bring the effect size to a non-significant value.

Fig. 6.

The funnel plot of the effects of gas or charcoal on the levels of PAHs in cooked meat

Discussion

The results of the meta-analysis showed that charcoal-cooked meat contains significantly more PAHs than gas-cooked meat. The mean difference between the cooking methods in the induction of each PAH was 2.053 μg/Kg. The natural gas such as butane, propane, and methane are easily available, particularly in urban areas. Gas grills have inherent advantages over charcoal grills that make them convenient to handle. The advantages include the ease of lighting, faster heating up, and no residues as ash to manage. Charcoal is a traditional heat source produced from incomplete combustion of various carbonaceous compounds including wood, peat, lignite, nutshells, bones, vegetables, etc. [29]. Charcoal grills produce a dry heat of higher temperature, compared to gas grills, that sears the meat ending in moisture [30]. Charcoal-grilled meat also cooks faster with a better taste.

One of the possible reasons for the higher PAHs concentration in charcoal-cooked meat over open flames can be due to the incomplete combustion of charcoal compared to gas [15]. The incomplete combustion might be attributed to the uneven distribution of heat in the burning parts of the charcoal mass. Another reason may be the higher temperature due to uncontrollable charcoal burning. In the studies included in the present meta-analysis, just one study [15] reported the temperature of cooking, which was the same (200–230 °C) for both methods of meat cooking. However, surveying the literature shows that charcoal usually burns with higher temperatures [30] as much as 400–450 °C in the fireplace of the grill [31], 280–300 °C near to the charcoal bed [32], and 230 to 300 °C next to the grid [33]. Studies show that temperatures above 200 °C initiate the pyrolysis of meat ingredients, especially fat, and accelerate it in the range of 500–900 °C [4], leading to the formation of heat-induced PAHs. That is why it is suggested that lower temperatures and longer duration be used for meat grilling to reduce PAHs formation [34].

The other explanation for the higher generation of PAHs in meat cooked by open flames of charcoal rather than natural gas, which may be the most significant one, is the use of a bed fully covered of burning charcoal in the oven. This configuration allows all fat to drip from the meat and pyrolyze when falling on the hot charcoal bed and generate smoke [35]. The rising smoke specifically carries low-molecular-weight PAHs (with one or two rings) and deposits them on the surface of the meat [4]. On the other hand, gas-burning ovens usually utilize several pipes that direct the gas to the oven chamber through nuzzles. The free space between the pipes allows fat droplets to fall onto a tray on the bottom of the chamber, thus preventing fat pyrolysis and PAHs formation due to avoiding the direct contact of fat with the heating source. However, it is noteworthy to consider that charcoal not always but often is placed on the bottom of the oven and the skewers or grilling grids are placed over the flame; this is while nuzzles in gas ovens can also be located above or in the side of the meat skewer. The latter configuration does not allow the fat to drop on the naked flame. In the examined papers in this study, the heat source location relative to the meat skewer was not the same in 3 out of 7 studies. This might affect the true difference between gas and charcoal-cooked meat in terms of PAHs content, which needs further investigation.

The results of the subgroup analysis of 17 trials dealing with red or white meat cooking showed that the type of meat had a significant contribution to the increased difference between gas and charcoal cooking in PAH induction. In other words, the mean difference between gas and charcoal cooking in terms of PAHs induction increased when red meat rather than white meat was subjected to an open flame. The MD for red meat was 3.499 μg/Kg while it was 3.319 μg/Kg for white meat. It seems this difference between red and white meat is due to the fat content of the meat as it is higher in red meat than in white meat [36]. Selected studies in the meta-analysis rarely reported fat percentage in their meat samples, making it impossible to subgrouping based on the fat content. However, Gorji et al. [15] measured the fat content of beef and chicken meat as 7.3% and 3.2%, respectively.

Another interesting finding of subgroup analysis was the effect of PAH numbers of interest detected in the meat samples on the outcome. Based on the analysis, studies investigating fewer PAHs (cutoff point of 7) reported higher differences between charcoal and gas cooking of meat in terms of PAH induction (MD: 5.106 vs. 1.447 μg/Kg). Therefore, one may conclude that changing the heat source from gas to charcoal leads to exponentially increased levels of some specific PAHs in cooked meat. However, this finding may simply be due to the publication bias if we assume that studies tend to measure and report just those PAHs whose concentrations were high enough to be quantified in samples, making more significant results. Therefore, for future studies, it is suggested that a set of specific PAHs be studied to increase the homogeneity, including benzo[a]pyrene, low-molecular-weight (LMW) PAHs (acenaphthylene, acenaphthene, fluorene, phenanthrene, and anthracene), high-molecular-weight (HMW) PAHs (benz[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, indeno[1,2,3-cd]pyrene, dibenz[ah]anthracene, benzo-[g,h,i]perylene, fluoranthene, and pyrene), PAH4 (benz[a]anthracene, chrysene, benzo[b]fluoranthene, and benzo[a]pyrene), and PAH8 (benz[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, indeno[1,2,3-cd]pyrene, dibenz[ah]anthracene, and benzo-[g,h,i]perylene) [28]. Benzo[a]pyrene (BaP) is usually measured in a mixture of PAHs as the established indicator [37, 38]. It was measured by all seven studies reviewed in the current meta-analysis. Therefore, it can be the best candidate for further studies.

PAHs are carcinogenic and genotoxic compounds. The literature survey showed preheating with steam and microwave and warping with aluminum foil and banana leaf are some methods of preventing PAHs production during meat cooking on open flames [39]. The effect of marinating with antioxidants such as beer as antiradical agents [40] is also notable. Cooking methods such as electric grilling [28], steaming [17], boiling [41], and even frying [28] have been mentioned in various studies to be safer than gas and charcoal open-flame grilling for meat cooking. However, the obtained results are largely dependent on the methodology. Therefore, there is a need for a comprehensive review of the literature to search for the safest methods and conditions of meat cooking in terms of PAHs induction.

The current study faced limitations. First, there was limited homogeneity between studies in terms of preparation and cooking conditions that made it impossible to conduct subgrouping based on parameters including the fat percentage of the meat, heat source location, skewer direction, distance from the heat source, cooking time, and cooking temperature. In addition, our study was not designed to estimate the concentration of PAHs in gas and charcoal-cooked meat and therefore, it was not possible to compare the results with the recommended PAHs values in ready-to-eat meat products. Moreover, cooking methods such as grilling, barbequing, broiling, roasting, etc. have been used alternatively in many studies that made it difficult to elucidate the exact conditions of cooking. On the other hand, the most significant feature of the study was the quantification of the difference between gas and charcoal as the most common fuel sources in PAHs induction when they were used for open-flame cooking of meat, which is of great importance in food safety assessments.

Conclusion

The study was conducted to determine the effect of gas and charcoal cooking on the PAHs induction in meat products. It was revealed that charcoal induces significantly more PAHs in open-flame cooked meat. The overall difference was 2.053 μg per kilogram of the edible portion of the meat. It is crucial to notice that this value reflects the mean difference per each PAH measured in the meat. Since JECFA has listed 16 PAHs as genotoxic and carcinogenic compounds produced in meat during cooking, one can conclude that the sum difference between charcoal and gas-cooked meat in terms of toxic PAH content is 2.053 × 16 = 32.85 μg/Kg, which is a considerably high difference. This finding emphasizes the necessity of avoiding open flames of charcoal for cooking meat or, at least, the need for changing the geometry of charcoal-fueled cookstoves to prevent fat dripping on the fire, which is possibly the main route of PAHs induction in the cooked meat. In addition, red meat displays the higher differences between charcoal and gas in terms of PAH induction, possibly due to the higher fat content of red meat than white meat. Greater differences are observed when fewer PAH numbers of interest are detected in cooked meat. This suggests the need for a consensus on a specific set of PAHs to be focused on in future studies.

Statement of authorship

Hossein Najafi Saleh conceived the study; Hossein Najafi Saleh and Fateme Barjasteh-Askari conducted database searching; Mohammad Ghorbani and Mojtaba Davoudi participated in qualitative and quantitative study selection; Mohammad Ghorbani conducted data analysis; Mojtaba Davoudi drafted the manuscript; Fateme Barjasteh-Askari polished the manuscript; Simin Nasseri critically revised the manuscript for important intellectual content. All the authors read and approved the final manuscript.

Compliance with ethical standards

Conflict of interests

None declared.