Chemical hazards in smoked meat and fish

Abstract This review aims to give an insight into the main hazards currently found in smoked meat and fish products. Literature research was carried out on international databases such as Access to Global Online Research in Agriculture (AGORA) database, Science direct, and Google scholar to collect and select 92 relevant publications included in this review. The smoking process was described and five hazards mostly found in smoked fish and meat were presented. The heat‐induced compounds such as polycyclic aromatic hydrocarbons, heterocyclic amines, and nitrosamines were found in smoked fish and meat. Other hazards such as biogenic amines and heavy metals were also present in smoked fish and meat. The levels of these hazards reported from the literature exceeded the maximal limits of European Union. A brief description of risk assessment methodology applicable to such toxic compounds and risk assessment examples was also presented in this review. As most of the hazards reported in this review are toxic and even carcinogenic to humans, actions should be addressed to reduce their presence in food to protect consumer health and to prevent public health issue.

| 6905 IKO AFÉ et Al. amine, amides, secondary amino acids, quaternary ammonium salts, etc.) and a nitrosating agent (nitrite, nitrates, and nitrogen oxides) through several reactions (Filho et al., 2003; INERIS (Institut National de l'Environnement Industriel et des RISques), 2014; Reinik, 2007). Nitrogen oxides are formed either from the addition of nitrate and/or nitrite to foods or from the heating process of food such as smoking, during which nitrogen molecular can be oxidized or present in the smoke (INERIS (Institut National de l'Environnement Industriel et des RISques), 2014; Jakszyn et al., 2005). Al Bulushi et al. (2009) reported NPYR and NPIP in vitro formation at high temperature (160°C, 2 h). Microorganisms (Aspergillus sp.; Pseudomonas sp.; P. stutzeri; E. coli) can be involved in N-nitrosamine formation by reducing nitrates to nitrites, by degrading proteins to amines and amino acids, or by producing enzymes working at a suitable pH (2)(3)(4) for nitrosation (Al Bulushi et al., 2009;Ayanaba & Alexander, 1973;Drabik-Markiewicz et al., 2009;Jägerstad & Skog, 2005;Jägerstad et al., 1998;Mills & Alexander, 1976;Rostkowska et al., 1998;Yurchenko & Molder, 2007). Nitrite and nitrate are frequently used in meat preservation and lead to nitrosamines formation due to reaction with amino compounds either in the stomach or within the food product (Filho et al., 2003;Pan et al., 2011;Sebranek & Bacus, 2007;Swann, 1977). It is the case of meat products such as sausages, ham, and salami where the addition of nitrite and nitrate was used to inhibit the formation of spoilage bacteria (Drabik-Markiewicz et al., 2009;Filho et al., 2003;Hustad et al., 1973). The nitrosamines are found in smoked meat, grilled meat, canned meat, and pickled meat at different levels (Table 1), but not in raw meat where there is not enough nitrite and amines for its production (Yurchenko & Molder, 2007). Studies carried out on nitrosamine determination in fish products mostly focused on NDMA determination because of its precursor dimethylamine (DMA) which is widely formed in marine fish (Al Bulushi et al., 2009). NDMA is classified in Group 2A (probably carcinogenic to humans) by the International Agency for Research on Cancer (IARC (International Agency for Research on Cancer), 2010), whereas N-nitrosonornicotine (NNN) and 4-(N-nitrosomethylamino)-1-(3-pyridyl)-1-butanone (NNK) are classified in Group 1 (carcinogenic to humans). Belitz et al. (2009) reported NDMA in cured meat processed with pickling with levels ranging between 0.5 and 15 μg/kg (Table 1). Herrmann et al. (2014) also reported NDMA in smoked pork fillet (1.3 µg/kg) and smoked ham (2.1 µg/kg).
Different analytical methods were used to analyze nitrosamines.
The method of gas chromatography and mass spectrometry detection with ion monitoring using different columns has been used by several authors to identify and quantify nitrosamines (Filho et al., 2003;Herrmann et al., 2014;Swann et al., 1977;Yurchenko & Molder, 2007).
High levels of PhIP (till 480 ng/g) were reported from the literature (Table 2). Even though no maximal limit of heterocyclic amines was reported in the literature, their presence in food is a health concern and adequate food preparation procedures should be implemented having the ALARA (ALARA = as low as reasonably achievable) principle in mind. Lu et al. (2018) , 2008). Heterocyclic amines can also be extracted by solidphase extraction and analyzed by reverse phase HPLC or LC/MS (Oz & Yuze, 2016;Santos et al., 2004;Sinha et al., 1998;Viegas, Novo, Pinto, et al., 2012). Ingenbleek et al., 2019; Yusuf et al., 2015). Several studies reported that fat dropping in the flame during grilling processing contributes to PAHs formation (Chen et al., 2013;Viegas, Novo, Pinto, et al., 2012).

| Polycyclic aromatic hydrocarbons
Additionally, studies showed that PAHs formation depends on the type of raw material, smoking methods, fuel and kiln type, smoke composition and degree of exposure to smoke, and combustion temperature (Chen et al., 2013;Codex Alimentarius, 2009;Kpoclou et al., 2014;Stołyhwo & Sikorski, 2005). Traditional smoking or grilling is responsible for the production of high amounts of PAH in meat Consumers are exposed to PAHs according to three possible ways: by inhalation, contact with the skin, and consumption of con-

F I G U R E 4
Chemical structure of the PAH4 for which a maximum limit in food has been set in EU (PubChem, 2020). BaP at a concentration of 0, 5, 25, or 100 mg/kg of diet for 2 years, papillomas and carcinomas were observed in the forestomach, oesophagus, and tongue (Culp et al., 1998). Several authors also associate colorectal cancer with meat consumption and some of them established colorectal cancer (Gunter et al., 2007;Ronco et al., 2011;Sinha et al., 2005). Sinha et al. (2005) reported an increased risk of colorectal adenomas resulting from high BaP intake from both meat consumption and other food sources.
Benzo ( The European commission set maximum levels of 2 and 12 μg/kg for benzo(a)pyrene (BaP) and the PAH4, respectively, in smoked meat and smoked fish products (EC (European Commission), 2006). Table 3 shows some examples of PAH4 levels reported from the literature (between 2015 and 2020), far above the EU limit of 12 µg/kg (25 times (Iko Afé et al., 2020) or 52 times (Rozentale et al., 2018) this limit).
Determination of PAH in food can be performed after using an accelerated solvent extractor (ASE) for the extraction, and HPLC coupled with fluorescence and photo diode array detectors (FLD/ PDA) or gas chromatography coupled with mass spectrometry (GC/ MS) for quantification (Brasseur et al., 2007;Kendirci et al., 2014;Saito et al., 2014;.
During these past decades, several studies dealt with PAHs in processed food, especially in smoked meat and fish. Some of these studies reported in Tables 3 and 4 were from different continents: Asia (15.8%), Africa (42.1%), and Europe (42.1%). In Africa, Nigeria is the country in which more studies were carried out on PAHs.
The PAHs data reported in Table 4 showed that most studies are recent (published between 2017 and 2021), showing that there is a new interest for scientists to update data on the presence of PAHs in smoked or grilled fish and meat. However, for countries such as Benin and Egypt, very few relevant data were available on PAHs contamination in fish and meat products before 2016 (Table 4). Most of the reported concentrations were above the EU maximal limit for BaP, and the highest BaP level (288 µg/kg) was about one hundred and forty-four times above this limit, showing that consumers could be highly exposed to PAHs through the consumption of this kind of food.

TA B L E 3
Examples of levels of PAH4 above the maximal EU limit of 12 µg/kg in various smoked or grilled meat and fish Heavy metals such as cadmium (Cd), mercury (Hg), and lead (Pb) are toxic even at low concentrations (Amos-Tautua et al., 2013;Daniel et al., 2013;Ersoy et al., 2006;Şireli et al., 2006 Table 5. Perello et al. (2008) reported the increase of Pb, As, and Hg contents in fish and meat products processed with grilling, frying, boiling, and roasting, compared to the raw products collected from Spain markets (data not shown). Even though an increase of heavy metal levels was recorded after processing in different studies, this increase could be due to the absorption phenomenon or environmental contamination as the culinary practices were not carried out in controlled close space. It could also be a concentration of the contaminants due to water loss during smoking and drying.
They could also be determined using atomic absorption spectrometry after microwave digestion and inductively coupled plasma mass spectrometry (ICP/MS) (Kabir et al., 2011;Uluozlu et al., 2009). The studies on the occurrence of heavy metals in smoked or grilled fish and meat reported in this section were mainly from Africa. Indeed, although some studies were from Turkey (Ersoy et al., 2006;Şireli et al., 2006), Spain (Perello et al., 2008) and Poland (Rajkowska-Myśliwiec et al., 2021), the majority of them were from Nigeria (Amos-Tautua et al., 2013;Anigboro et al., 2011;Aremu et al., 2014;Daniel et al., 2013;Ersoy et al., 2006;Ibanga et al., 2019) and other African countries such as Egypt (Abbas et al., 2021), Ghana (Kobia et al., 2016) and Burkina Faso (Bazié et al., 2021). Heavy metals contamination data (   Table 7. The presence of histamine was reported in smoked salmon at levels ranging between 2.5 and 171 mg/kg, in smoked Sardinella sp.

European Commission set maximal limits for histamine (100-
200 mg/kg) in fish and fishery products from fish species associated with a high amount of histidine (EC (European Commission), 2005). No maximal limit of histamine is available for meat products.

TA B L E 4
Examples of polycyclic aromatic hydrocarbon levels found in smoked or grilled fish and meat products in these past decades However, several authors reported the use of biogenic index (sum of putrescine, tyramine, cadaverine, and histamine levels) to assess the freshness and quality of pork (Cheng et al., 2016;Douny et al., 2019).
The highest histamine concentration in fish reported in this review was 44 times over the authorized European limit and resulted in histamine fish poisoning (HFP) (Marissiaux et al., 2018). Similar concentration (4,384.2 mg/kg) was also reported in smoked-dried fish from Benin (Table 8). Regarding the geographical location, the selected paper reported in the Tables 7 and 8

| Risk assessment methodology applicable to toxic compounds
The risk assessment is part of the risk analysis concept, which, as reported by Larsen (2006), includes risk assessment, risk evaluation, and risk communication. These three elements are separate tasks, performed by different actors, but should be part of an interactive TA B L E 5 Mean concentrations of heavy metals in smoked or grilled fish (a) and meat (b) products as reported from the literature process (Larsen, 2006;Stadler & Lineback, 2009  Extraction trichloroacetic acid ion-exchange chromatography Plahar et al. (1999) Cold-smoked salmon 30.9 ± 0.4 Extraction with perchloric acid high-performance liquid chromatography with a diode array detector Köse et al. (2012) Hot-smoked Bonito (Tuna fish) 98.7 ± 0.6 Grilled tuna 4,400 Not mentioned Marissiaux et al. (2018) Smoked fish from different species 11-63 Quantification colorimetrically at 495 nm using a spectrophotometer.
CSIR (2017) Smoked turkey breast fillets stored at 4°C after 30 days 32.9 ± 1.4 Extraction trichloroacetic acid With liquid chromatography. Quantification was performed coupled with a UV detector Ntzimani et al. (2008) Grilled pork <11.2-81.5 Extraction with perchloric acid and injection on UPLC coupled with a fluorescence detector Douny et al. (2019) dynamic and kinetic aspects, and how to establish an acceptable daily intake (ADI) or a tolerable daily intake (TDI) using a safety factor to consider for the intra-and inter-species variation.
Exposure assessment: To assess the exposure, consumption data and contamination data are needed to calculate the estimated daily intake (EDI) by multiplying the concentration of hazard by the daily consumption of food contaminated with this hazard. EDI can be calculated for several categories of population (i.e., babies, children, teenager, and adults). EDI can be calculated either following a deterministic approach using median, mean, or maximum of consumption or contamination data, or following a probabilistic approach using distributions of consumption and contamination data. Examples of risk assessments related to PAH ingestion through consumption of grilled and/or smoked fish and meat (Table 9) pointed out a health concern for consumers of several countries such as Cambodia , Benin (Iko Afé et al., 2020, Turkey (Sahin et al., 2020), Nigeria (Akpambang et al., 2009), and Latvia (Rozentale et al., 2018). Before 2015, the mean values of MoE associated with the consumption of smoked or grilled fish (Table 9a) and meat products (Table 9b) contaminated with PAHs including BaP and PAH4 were globally above 10,000, showing a very low concern for the consumers of these products. After 2015, the studies reported showed MoE globally below 10,000 for consumers of smoked or grilled fish and meat products from different countries such as Benin, Cambodia, Turkey, and Latvia (Table 9). MOE below 10,000 indicates a high concern (risk of cancer) for consumers for carcinogenic compounds such as PAHs.
Regarding consumers exposure to heavy metals from consumption of smoked or grilled fish and meat products, very few MOE = BMDL10 (mg per kg bw per day) EDI (mg per kg bw per day) , TA B L E 8 Mean (maximum) concentrations of histamine and tyramine in fish (a) and meat products (b) from different countries The cancer risk index linked to lead exposure calculated for consumers of braised and flamed chicken processed in Burkina Faso ranged between 7 × 10 −7 and 3 × 10 −6 ( Table 10). None of the index risk values was above the threshold set by US-EPA (IR > 10 −4 ). Similar TA B L E 9 Estimated daily intakes (EDI) and margin of exposure (MOE) for polycyclic aromatic hydrocarbons (PAH) through consumption of smoked or grilled fish (a) and meat (b) products, in different countries  (Table 11) was well below this ARfD.
Based on the limited published data, no adverse health effects have been observed in healthy volunteers exposed to a level of 25-50 mg of histamine per person per meal (EFSA, 2011). The mean histamine exposure reported in Table 11 revealed a health concern for Beninese consumers of smoked fish and smoked-dried fish . Although the mean histamine exposure reported for consumers of Cambodia, Spain, and Egypt showed an absence of intoxication risk, there is risk of histamine poisoning in case of extreme consumption of smoked or grilled fish and meat products during the same meal or for sensitive consumers.
TA B L E 1 0 Cancer and noncancer risks related to heavy metals through consumption of smoked or grilled fish and meat products reported from the literature

TA B L E 11
Histamine and tyramine exposure from consumption of fish and meat products Abbreviation: -, data not presented in the cited paper.
*Numbers in parentheses represent the maximum value.

| CON CLUS ION
Smoked fish and meat products may be contaminated by various toxic compounds including carcinogenic compounds. Most of the chemical hazards reported in this review are processing contaminants. Some of them can be formed when high temperature is reached inside the product (heterocyclic amines and nitrosamines) and others during pyrolysis of the fuel during processing (PAHs).
Biogenic amines are not related to the smoking process but can be present in raw or smoked fish due to decarboxylation of free amino acids occurring after microbial contamination. In case of heavy metals, they are environmental pollutants found in raw and processed food. In traditionally smoked fish or grilled meat, most of the chemical hazards mentioned in this review exceed the maximal limits established by EU. Several actions should be addressed to decrease them in smoked fish and meat as they are highly consumed products.

ACK N OWLED G M ENTS
This work was fully supported by QualiSani project through ARES CCD (Académie de Recherche et d'Enseignement Supérieur, Commission de la Coopération au Développement).

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare.

E TH I C A L A PPROVA L
This study does not involve any human or animal testing.

DATA AVA I L A B I L I T Y S TAT E M E N T
All the data used in this study can be made available upon reasonable request.