Critical Reviews in Food Science and Nutrition

ISSN: 1040-8398 (Print) 1549-7852 (Online) Journal homepage: www.tandfonline.com/journals/bfsn20

Worldwide contamination of food-crops with

mycotoxins: Validity of the widely cited ‘FAO
estimate’ of 25%

Mari Eskola, Gregor Kos, Christopher T. Elliott, Jana Hajšlová, Sultan Mayar &
Rudolf Krska

To cite this article: Mari Eskola, Gregor Kos, Christopher T. Elliott, Jana Hajšlová, Sultan Mayar
& Rudolf Krska (2020) Worldwide contamination of food-crops with mycotoxins: Validity of
the widely cited ‘FAO estimate’ of 25%, Critical Reviews in Food Science and Nutrition, 60:16,
2773-2789, DOI: 10.1080/10408398.2019.1658570

To link to this article: https://doi.org/10.1080/10408398.2019.1658570

© 2019 The Author(s). Published with

license by Taylor & Francis Group, LLC.

Published online: 03 Sep 2019.

Submit your article to this journal

Article views: 34696

View related articles

View Crossmark data

Citing articles: 629 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=bfsn20
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION
2020, VOL. 60, NO. 16, 2773–2789
https://doi.org/10.1080/10408398.2019.1658570

REVIEW

Worldwide contamination of food-crops with mycotoxins: Validity of the widely

cited ‘FAO estimate’ of 25%
Mari Eskolaa , Gregor Kosb , Christopher T. Elliottc , Jana Haj
slovad, Sultan Mayarb , and
Rudolf Krskaa,c
a
Institute of Bioanalytics and Agro-Metabolomics, Department of Agrobiotechnology (IFA-Tulln), University of Natural Resources and Life
Sciences Vienna (BOKU), Tulln, Austria; bDepartment of Chemistry and Biochemistry, Concordia University, Montreal, QC, Canada; cInstitute
for Global Food Security, School of Biological Sciences, Queens University Belfast, Belfast, Northern Ireland, UK; dDepartment of Food
Analysis and Nutrition, Faculty of Food and Biochemical Technology, University of Chemistry and Technology, Prague 6, Czech Republic

ABSTRACT KEYWORDS
Prior to 1985 the Food and Agriculture Organization (FAO) estimated global food crop contamin- Chemical contaminant;
ation with mycotoxins to be 25%. The origin of this statement is largely unknown. To assess the occurrence; cereals; data;
rationale for it, the relevant literature was reviewed and data of around 500,000 analyses from the analysis; uncertainty
European Food Safety Authority and large global survey for aflatoxins, fumonisins, deoxynivalenol,
T-2 and HT-2 toxins, zearalenone and ochratoxin A in cereals and nuts were examined. Using dif-
ferent thresholds, i.e. limit of detection, the lower and upper regulatory limits of European Union
(EU) legislation and Codex Alimentarius standards, the mycotoxin occurrence was estimated.
Impact of different aspects on uncertainty of the occurrence estimates presented in literature and
related to our results are critically discussed. Current mycotoxin occurrence above the EU and
Codex limits appears to confirm the FAO 25% estimate, while this figure greatly underestimates
the occurrence above the detectable levels (up to 60–80%). The high occurrence is likely explained
by a combination of the improved sensitivity of analytical methods and impact of climate change.
It is of immense importance that the detectable levels are not overlooked as through diets,
humans are exposed to mycotoxin mixtures which can induce combined adverse health effects.

Introduction HT-2 toxin (HT-2), zearalenone (ZEN), fumonisins (FBs),

ochratoxin A (OTA), ergot alkaloids (EAs), patulin and cit-
Occurrence of mycotoxins is widespread throughout the
rinin are considered to be the mycotoxins of the greatest
world. Their acute and chronic dietary exposures can induce
importance from food safety and regulatory viewpoints. This
a variety of adverse health effects in humans and animals,
is evident not only because of the large number of published
making these chemically diverse substances highly relevant
research, and the many international and national toxicity
agricultural contaminants. They typically co-occur in agri-
cultural crops and through diversified diets, exposure to and exposure assessment reports, but also due to the exten-
mixtures of mycotoxins is common. Many fungal species sive standards and legislation, which have been in place for
produce them as a defense against their changing environ- many years to safeguard public and animal health at the
mental growing conditions; Aspergillus, Fusarium and world, European Union (EU) and national levels (Eskola
Penicillium regarded as the most important producers. et al. 2018; EUR-Lex 2018; WHO 2018).
Despite many years of research, and the introduction of AFs B1, B2, G1, G2 and M1 can be regarded as the most
good agricultural and manufacturing practices in the food important mycotoxins owing to their genotoxic carcinogenic
chain, mycotoxin occurrence remains to be a global prob- properties. In fact, AFs are amongst the most potent muta-
lem. Significant economic losses are associated with the genic and carcinogenic substances known. AFB1 is the most
impact of mycotoxins on human health, animal welfare and potent followed by AFG1 and AFM1 (JECFA 1999, 2017).
productivity, and both domestic and international trade (Pitt The chronic AF-exposure induces liver cancer, infections
and Miller 2017; Wu and Mitchell 2016). Combined toxico- and growth impairment in humans, while high exposures
logical effects from the mycotoxin co-exposure are increas- cause acute symptoms, even a death (EFSA 2004a, 2007,
ingly studied (Alassane-Kpembi et al. 2017; Grenier and 2012a, 2018c; IARC 1993, 2002, 2012, 2015; JECFA 1997,
Oswald 2011; EFSA 2016, 2017a, 2017b, 2017c, 2017d, 1998, 1999, 2002, 2008, 2017). Sterigmatocystin, a genotoxic-
2018a, 2018b; JECFA 2011; Lee and Ryu 2017). Currently, carcinogen, shares its biosynthetic pathway with aflatoxins
aflatoxins (AFs), deoxynivalenol (DON), T-2 toxin (T-2), and has the same target organ (liver) with AFB1 (JECFA

CONTACT Rudolf Krska [email protected] Institute of Bioanalytics and Agro-Metabolomics, Department of Agrobiotechnology (IFA-Tulln), University
of Natural Resources and Life Sciences Vienna (BOKU), Konrad Lorenzstraße 20, Tulln, A-3430, Austria.
ß 2019 The Author(s). Published with license by Taylor & Francis Group, LLC.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/),
which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
2774 M. ESKOLA ET AL.

2018). Cereals and different types of nuts and their products carcinogenicity in humans, although some data are available
are the main exposure sources. The contribution of maize, (EFSA 2011, 2017c; IARC 1993).
peanuts, rice, sorghum and wheat each make more than Although OTA is present in variety of foods (e.g. cereals,
10% to the global exposure to AFs (JECFA 2017). The pulses, coffee beans, grapes, cacao beans, nuts, spices and
AFB1-metabolite AFM1 may be present in milk (EFSA animal-based foods), 60% of its total exposure is from cere-
2004a, 2007; IARC 2002; JECFA 1997, 1999; Shephard als (EFSA 2006; IARC 1983, 1993; JECFA 1996, 2001, 2007,
2008). Co-exposure to genotoxic AFB1 and FBs, which may 2008; Marin et al., 2013). Kidney is the main target, OTA
induce regenerative cell proliferation, is of health concern, exposure exhibiting nephrotoxicity in animals (EFSA 2006;
but due to the marked differences in their modes of action IARC 1983, 1993; JECFA 1996, 2001, 2007, 2008). Species-
the combined effects are unlikely additive (IARC 2015; differences exist and pigs, poultry and dogs seem to be the
JECFA 2012, 2017). Of the FB-analogues, FB1 is the most most sensitive (EFSA 2004c, 2006; JECFA 1996, 2001, 2002,
predominant, comprising about 70% of the total FB-contam- 2007, 2008). The causality of OTA inducing human nephr-
ination. FB2 and FB3 co-occur with FB1 but with lower opathy remains unclear but OTA appears to be one cause of
concentrations (IARC 2015; JECFA 2012, 2017). Oral expos- the endemic Balkan nephropathy (Bui-Klimke and Wu 2014;
ure to FBs induces a wide range of adverse effects, with not- EFSA 2006, 2010; JECFA 1996, 2001, 2002, 2007, 2008;
able species differences (Deepa and Sreenivasa 2017; EFSA Stiborova et al. 2016). In rodents OTA is a potent renal car-
2005a, 2014, 2018a; IARC 2002, 2015; JECFA 2001, 2002, cinogen, and it is possibly carcinogenic to humans (IARC
2012, 2017; WHO 2000). Liver is the target organ for FB- 1993). Other regulated mycotoxins are patulin, EAs and cit-
mediated toxicity in mice and kidney in rats, while in horses rinin, but based on the published assessments and other
equine leukoencephalomalacia and in pigs porcine pulmon- relevant data they appear to be less relevant from the food
ary edema with signs of liver toxicity can occur. FB1, FB2, safety viewpoint (Diana Di Mavungu et al. 2011; EC 2006a;
FB3 and FB4 and their modified forms have similar toxico- EFSA 2005b, 2012b, 2012c, 2017e; FDA 2001a; IARC 1986;
logical profiles, but the potencies of the modified forms are Iqbal et al. 2018; JECFA 1996; Krska and Crews 2008; Lopez
lower (EFSA 2018a). FB1 is possibly carcinogenic to Sanchez et al. 2017; Mornar et al. 2013; Pleadin et al. 2018;
humans, and an association between chronic dietary expos- Puel et al. 2010; SCF 2000b; SCOOP 2002; Topi et al. 2017;
ure and human esophageal and liver cancer has been WHO 1990). There are also several mycotoxins not included
claimed, but causality has not been confirmed (EFSA 2018a;
in any legislation or standards due to their apparent minor
IARC 2002; JECFA 2001, 2012, 2017).
relevance (Eskola et al, 2018; JECFA 2017; WHO 1990).
DON occurs in cereal grains alone or in combination
These mycotoxins are excluded from this paper.
with its most relevant forms 3-acetyl-DON, 15-acetyl-DON
and DON-3-glucoside (10–20% of the DON-levels) (EFSA
2011, 2017d; JECFA 2001, 2011; McCormick et al. 2011; Widely cited estimate of 25% world’s food crops
Nathanail et al. 2015; Vanheule et al. 2014). The exposure is contaminated with mycotoxins
mainly from cereals and cereal-based foods, wheat making
the highest proportion (56–100%) (EFSA 2004b, 2011, Over the years, published mycotoxin papers and reports
2017d; JECFA 2001, 2011). High exposure mediates acute have aimed at estimating the overall global prevalence of
gastrointestinal symptoms in humans (e.g. vomiting) but the mycotoxins in agricultural crops. In this context, authors
effects from the chronic human exposure are unknown have often referred to an estimate of 25%, supposedly by
(EFSA 2004b, 2011, 2017d; IARC 1993; JECFA 2001, 2011; Food and Agriculture Organization (FAO), published by
SCF 1999). Weight gain suppression and anorexia are the Park and coauthors in the FAO Food, Nutrition and
most common chronic effects in animals (JECFA 2001, Agriculture Journal, number 23 in 1999. The second sen-
2011; EFSA 2017d). T-2 and HT-2 are also regularly found tence states that ‘According to FAO, at least 25 percent of
in cereal grains, but with lower frequency and levels than the world’s food crops are contaminated with mycotoxins, at
DON. Their chronic exposure induces general toxicity, hem- a time when the production of agricultural commodities is
ato- and immunotoxicity and high exposure vomiting in barely sustaining the increasing population (Boutrif and
animals (EFSA 2011, 2017b). Both DON and T-2 appear not Canet 1998)’. However, further evidence of the background
to be carcinogenic (EFSA 2011, 2017b, 2017d; IARC 1993; for the figure of 25% is not provided by Park et al. (1999).
JECFA 2002). ZEN is a common contaminant in maize and Looking more in detail into the paper by Boutrif and Canet
other grains making the cereal-based food its main source of (1998) published in Revue de medecine veterinaire, to which
exposure (EFSA 2011, 2014, 2016, 2017c; JECFA 2000; Park et al. (1999) refers, no figure or mention of 25% of
Nathanail et al. 2015; Vanheule et al. 2014). Female pigs are food crops contaminated with mycotoxins can be found in
the most sensitive species (including human) towards its this paper. It is also noted that on 28 February 1985 in the
estrogenic activity (EFSA 2011, 2014, 2016, 2017c; IARC journal of New Scientist, Mannon and Johnson write in
1993; JECFA 2000; SCF 2000a). Relative estrogenic potencies their article on mycotoxin contamination that according to
show that some modified ZEN forms are markedly more the FAO, mycotoxins contaminate some 25% of the world’s
estrogenic than ZEN itself (EFSA 2016). No definite link of food crop, and later in the article they state that ‘FAO has
a high ZEN-exposure with negative health effects in humans estimated that 25% of the world’s food crops are ruined by
has been concluded. Nor ZEN was classifiable as to its mycotoxins’ (Mannon and Johnson 1985). The authors do
 CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 2775

not give any other references, except FAO, to the informa- although in general prevalence is assumed to be higher in
tion of 25% contamination rate. feed crops than in food crops. Only those mycotoxins which
Thus, to best of our knowledge, it seems that there is no are included in the standards of Codex Alimentarius and/or
accurate published reference available for the figure of 25% in the regulations of the EU are addressed in this review.
world’s food crop mycotoxin-contaminated. Nor there is any The World Health Organization (WHO) collects global
information on which dataset this estimate was based on or food occurrence data on mycotoxins from the published lit-
how it was calculated. Following the exchange of views with erature and through its call for data. As these data are care-
the FAO officials in spring 2017, it turned out that neither fully scrutinized and cleaned by the WHO/FAO experts, the
them were able to trace back the origin of the estimate. authors considered that the risk assessments of the Joint
However, it could be reasonable to presume that the esti- FAO/WHO Expert Committee on Food Additives (JECFA)
mate of FAO stems from the proportion of the mycotoxin- represent currently the best available literature source for
contaminated food crops exceeding the Codex standards in the global-level estimates of mycotoxin prevalence in food
place at the time, rather than from the contaminated food crops. Therefore, the estimates calculated by JECFA for
crops above the limits of detection (LODs)/limits of quanti- mycotoxin-contaminated food crops are presented in this
fication (LOQs) of the analytical methods used back then. paper. In addition, the global estimates of prevalence from
The estimate of 25% presented in the FAO journal back in those individual research papers, which were considered of
1999 has been widely cited since then in many publications. providing added value, were included. Furthermore, myco-
Over the years it has therefore become ‘THE’ reference value toxin prevalence was assessed by the authors using two large
for the mycotoxin incidence in food crops at the global datasets. The first one comprised the occurrence data on
level. This can be evidenced for example by carrying out a food collected in 2010–2015 by European Food Safety
search for key words ‘FAO 25%, mycotoxins’ through the Authority (EFSA) from the EU Member States. These data
Google search engine (19 August 2019) resulting in thou- were requested from EFSA through the application of the
sands of citations varying from the citations in the docu- EU-regulation No 1049/2001 on public access to documents
ments of the public national and international food safety (EP and Council 2001). The second dataset on mycotoxins
organizations (including FAO) to the citations in research in feed was from Biomin Inc., Austria. Aspects causing vari-
papers and in industry documents. For these aforemen- ability in the occurrence data and impacting the uncertainty
tioned reasons, it is therefore highly important to discuss of the estimated mycotoxin prevalence are also discussed.
where this decades old FAO estimate for the world’s
mycotoxin-contaminated food crops stands today.
Variability and uncertainties associated with
Notably in the recent past, data on large surveys on
prevalence estimates of mycotoxins in food-crops
mycotoxin occurrence across the world have been released
suggesting higher mycotoxin prevalence than 25% stated by Many aspects cause variability in the data and therefore
FAO. Streit et al. (2013) reported that overall 72% of the impact the uncertainty of the estimated frequencies of myco-
around 17,300 feed samples origin from different parts of toxin-contamination. This should be kept in mind when
the world collected over eight years contained mycotoxins. estimating the prevalence of mycotoxins in food and feed
Also, Kovalsky et al. (2016) reported that mycotoxin con- crops at the global level. Therefore, it is clear that the esti-
tamination in feeds could be up to 79% or even above in mate for the overall mycotoxin-contaminated crops at the
about 2000 samples from 52 countries. The authors noted global level, based on the currently available data, can at its
that the observed frequency is considerably higher than best be an indicative rather than an exact figure. Before an
FAO’s 25% and concluded that the outcome of their study accurate estimate can be provided, statistically relevant
is a starting point for a more detailed investigation of myco- amount of high-quality data should be made available.
toxin contamination rates at the global level. They further Therefore, large occurrence data surveys on mycotoxin con-
emphasized that more data should be included in such an tamination in food-crops, with harmonized sampling strat-
investigation. Similar high frequencies of mycotoxins in feed egies and analytical performance criteria, across the world
were recently published by Biomin (2017) (see also below) over several growing seasons should be conducted.
in a large survey comprising 72 countries across the globe.
Although in these surveys mainly animal feed samples have
Different regulatory levels and standards impact the
been studied, they point to the direction that the prevalence
reported mycotoxin prevalence
of mycotoxin-contaminated food crops could also be higher
than the 25%-estimation of FAO. The applied regulatory levels and standards of mycotoxins
The aim of this paper is to provide a compilation of the vary in the different regions of the world. At the global level,
most recent and most relevant published information on the the Codex Alimentarius Commission of the Food, the joint
estimated global prevalence of mycotoxins in food crops, in inter-governmental body of FAO and WHO, is the main
particular. Mainly food-grade cereals and nuts are considered. operator in standard setting. The Codex standard CXS 193-
However, no large harmonized global-level mycotoxin occur- 1995 establishes maximum levels (MLs) for AFs (sum of
rence surveys for food crops were identified in the literature, AFB1, AFB2, AFG1 and AFG2, and, separately, AFM1 in
while for feed they are accessible. Therefore, the relevant data milk), DON, FBs (sum of FB1 and FB2) and OTA for differ-
on mycotoxin-contaminated feed are also noted in this paper ent foods (Codex Alimentarius 1995). The Codex standard
2776 M. ESKOLA ET AL.

Table 1. Maximum levels for most often regulated mycotoxins in food crops.
Mycotoxin Food crop Established levels (mg/kg)
Codex Alimentarius standard
Aflatoxins AFB1 Total
Almonds, Brazil nuts, hazelnuts, peanuts, pistachios for processing — 15
Almonds, Brazil nuts, hazelnuts, pistachios for direct human consumption — 10
Fumonisins B1 þ B2 Unprocessed maize 4000
Deoxynivalenol Cereal grains (wheat, maize and barley) for processing 2000
Ochratoxin A Unprocessed wheat, barley, rye 5
European Union maximum and guidance levels
Aflatoxins AFB1 Total
Groundnuts (peanuts), hazelnut, Brazil nuts, other oilseeds for processing 8 15
Almonds, pistachios, apricot kernels for processing 12 15
Tree nuts, other than the tree nuts above for processing 5 10
Groundnuts (peanuts), other oilseeds, other three nuts below for direct human consumption 2 4
Almonds, pistachios, apricot kernels for direct human consumption 8 10
Hazelnuts and Brazil nuts for direct human consumption 5 10
All cereals except maize and rice 2 4
Maize and rice for processing 5 10
Fumonisins B1 þ B2 Unprocessed maize 4000
Maize intended for direct human consumption 1000
Deoxynivalenol Unprocessed cereals other than durum wheat, oats, maize 1250
Unprocessed durum wheat, oats, maize 1750
Ochratoxin A Unprocessed cereals 5
Cereals intended for direct human consumption 3
Zearalenone Unprocessed cereals other than maize 100
Unprocessed maize 350
Cereals intended for direct human consumption 75
Maize intended for direct human consumption 100
T-2 þ HT-2 Unprocessed barley and maize 200
Unprocessed oats 1000
Unprocessed wheat, rye and other cereals 100
Maize intended for direct human consumption 100
Oats intended for direct human consumption 200
Other cereals intended for direct human consumption 50
USA: Action and guidance levels
Aflatoxin B1 All food crops 20
Fumonisins FB1 þ FB2 þ FB3 Maize 4000
Canada: Maximum and guidance levels
Aflatoxins Nuts 15
Deoxynivalenol Unprocessed soft wheat 2000
Australia: Maximum levels
Aflatoxins Peanuts, tree nuts 15
Japan: Maximum and provisional maximum levels
Aflatoxin B1 All food crops 10
Deoxynivalenol Wheat 1100a
China: Maximum and guidance levels
Aflatoxin B1 Maize 20
Rice (brown rice) 10
Wheat, barley, other cereals 5
Peanuts 20
Deoxynivalenol Maize, barley, wheat, other cereals 1000
Ochratoxin A Cereals 5
Beans 5
Zearalenone Wheat and maize 60
The guidance levels (the term may vary in legislation from country to country) are indicated in italics.
AFB1, aflatoxin B1; Total, total aflatoxins (sum of AFB1, AFB2, AFG1 and AFG2).
a
Provisional maximum level.

also defines sampling plans and analytical method perform- The enforceable MLs for AFs (AFB1 alone, sum of AFB1,
ance criteria for the mycotoxins (Codex Alimentarius 1995). AFB2, AFG1 and AFG2, and separately for AFM1), OTA,
The standards issued by Codex are voluntarily applied by DON, ZEN, FBs (sum of FB1 and FB2) in various foods are
the Codex member countries. They are considered as neutral stipulated in the Commission Regulation (EC) No 1881/
and generally accepted as global standards, and national 2006 and in its amendments (EC 2006a). The EU MLs for
mycotoxin legislation in many countries are based on them. food-crops are listed in Table 1. For the sum of T-2 and
The Codex MLs for food-crops are presented in Table 1. HT-2, the EC prescribes indicative levels for cereals and
The EU has its own established MLs for mycotoxins in cereal products (EC 2013) (Table 1). Furthermore, the EC
foods and feeds. The EU-legislation comprises several legally recognizes the need for simultaneous analysis of Fusarium-
binding regulations, and recommendations which are not toxins (EC 2013). As an analog to the Codex standards, also
binding but give a guidance for risk management. Since the EC specifies sampling methods and performance criteria
1999 harmonized ML for AFs have been in place in the EU. for the analytical methods (EC 2006b). While the Codex
 CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 2777

standard CXS 193-1995 specifies that its MLs are for feeds well established for many mycotoxins; currently liquid chro-
only when the mycotoxins transfer to animal-derived foods matography (LC) based methods coupled to tandem mass
and are relevant for the public health, the EU Directive spectrometry (MS) are probably the most widely applied
2002/32/EC regulates mycotoxin levels in animal feed (EP and preferred techniques (as also shown from the analyzed
and Council 2002). However, only MLs for AFB1 in feeds datasets below). The current trend in mycotoxin analysis
have been set. The acceptable guidance levels for DON, seems to be on the one hand to reach ever lower concentra-
ZEN, OTA, T-2 and HT-2 and the sum of FB1 and FB2 in tion levels and multi-mycotoxin and full metabolite profiles
feeds are covered by the Commission Recommendations (Arroyo-Manzanares et al. 2018; Brouwer-Brolsma et al.
2006/576/EC and 2013/165/EU (EC 2006c, 2013). Although 2017; de Lourdes Mendes de Souza et al. 2013; Doppler
several parts of the EU mycotoxin-legislation have been et al. 2016; Adekoya et al. 2019; Malachova et al. 2014, 2018;
harmonized with the Codex standards, the EU-legislation Njumbe Ediage et al. 2015; Monbaliu et al. 2010; Rofiat
appears to be more detailed and the MLs more stringent et al. 2015; Shephard et al. 2013; Simader et al. 2015; Uhlig
than the corresponding Codex ones. et al. 2013; Varga et al. 2013; Warth et al. 2012; 2015), and
In addition to the Codex standards and EU-legislation, on the other hand to determine the compliance at the statu-
many world countries have their own national mycotoxin tory levels by fast on-site techniques (Dorokhin et al. 2011;
legislations. For example, in USA, the US Food and Drug EC 2006b; Ediage et al. 2012; Joshi et al. 2016; Peters et al.
Administration (US FDA) has issued Compliance Policy 2017; Sieger et al. 2017). The current multi-analyte and
Guides on regulatory MLs for AFB1 and AFM1, and multi-class analytical methods are capable of a high
Guidance for Industry (non-binding) for guidance MLs of throughput of samples making larger amounts of results
DON and sum of FB1, FB2 and FB3 in certain foods and available than ever before in a short period of time. Today’s
feeds (FDA 1980a, 1980b, 1982, 1983, 2001a, 2001b, 2010). analytical methods generally have lower LODs and LOQs
In Canada, legislated MLs have been established for AFs, than the methods a decade or two ago. For example for the
and guidelines for DON and HT-2 in some foods and feeds determination of DON in foods, while the typical LODs for
(RG-8 Regulatory Guidance, CFIA 2017). Recommended a range of different methods were within 5–1600 mg/kg as
levels Canada has set for T-2, ZEN and OTA in many feeds reported by JECFA in 2001, they were within 1–60 lg/kg for
(CFIA 2017). In Australia, statutory MLs for AFs in nuts are the MS-based methods (mainly used) as reported by JECFA
in force (FSANZ 2017). Japan has established regulatory 10 years later in 2011 (JECFA 2001, 2011). This, no doubt
MLs for AFB1 in all foods and a provisional ML for DON has a direct impact on the presently reported apparently
in wheat (FAMIC 2011; Personal Communication, Ministry higher mycotoxin-contamination rates than during earlier
of Health, Labor and Welfare, Japan 2018). Regulatory ML years. Most importantly, the higher amount of available data
and advisory levels for AFB1, DON and ZEN for feeds are and more stringent analytical data quality requirements
also in place in Japan (FAMIC 2015). China has established decrease variability in the currently estimated mycotoxin fre-
statutory MLs for AFB1, AFM1 and OTA in many foods, quencies, thus justifying an update of occurrence data.
while for DON and ZEN guidance levels have been pre- The analysis at the low concentrations requires not only
scribed (GB 2017). The regulatory MLs and guidance levels advanced (and expensive) techniques but also experienced
in food-crops for these selected countries are presented in human resources and time. These reasons typically limit the
Table 1. routine control laboratories to analyze mycotoxins at their
Because only the compliance of the mycotoxin concentra- lowest possible LODs/LOQs the method is capable to. They
tions in food and feed with the regulatory MLs allows the rather operate with pragmatic fit-for-purpose methods in
food and feed business operators to place the commodities the region of the legal mycotoxin MLs. Thus, the adjusted
on the market, it is apparent that a large amount of the ana- sensitivity of the methods by the laboratories also affects
lytical occurrence data is generated for this purpose. The the found prevalence of mycotoxins and causes variability
mycotoxin MLs and guidance levels established for food- in the reported frequencies. It is also clear that between-
crops by Codex Alimentarius, EU and some selected indi- laboratory variation occurs independently from the
vidual countries differ as seen in Table 1. Therefore, when approach of the laboratory to generate the analytical results
considering the prevalence of mycotoxins across the globe, at the lowest capable level of LOD/LOQ or at the level of
the different MLs and guidance levels in place in the differ- MLs. Furthermore, the general practice how to report the
ent geographical regions undoubtedly influence the outcome positive results varies. The prevalence can be expressed
of this consideration. either considering the mycotoxin results (quantifiable)
above the LOD or above the LOQ of the respective labora-
tory. Thus, depending on the published report, the samples
Sampling, monitoring and performance of the
either above LODs or LOQs are considered positive or
analytical methods
sometimes a distinction between the two is not made, lead-
Across the world, the occurrence of mycotoxins in foods ing to an increased variability of the reported data on fre-
and feeds is actively monitored by food and feed business quencies of mycotoxin contamination. Even in the
operators, official control laboratories and research organiza- extensive mycotoxin reports of JECFA and EFSA, it is not
tions either to determine the regulatory compliance or the always clear whether the indicated mycotoxin-incidence is
quality of foods and feeds. Analytical methods are generally based on > LOD or > LOQ.
2778 M. ESKOLA ET AL.

It is very obvious that sampling influences the reported mycotoxin prevalence highly variable across the world.
mycotoxin-incidence values. Different sampling strategies Mycotoxins, such as trichothecenes, are typically regarded as
e.g. targeted and random, generate completely different out- toxins of the regions with temperate climate, while the pres-
comes. Sampling schemes can also alter, for example ence of others, such as aflatoxins, are associated with trop-
depending on the exporting destination of the food crops. ical and subtropical climate areas. A large amount of
Another sampling plan can be used when exporting to the information is available in the literature on the environmen-
EU with stringent mycotoxin regulations than when export- tal factors impacting the presence of mycotoxins in food
ing to countries with more liberal requirements (JECFA and feed crops (e.g. Van der Fels-Klerx et al. 2012a, 2012b;
2008). This certainly impacts the observed incidence data on 2016; Battilani et al. 2012, 2013, 2016; Paterson and Lima
mycotoxins. For the prevalence data presented in this paper, 2017). A year-to-year variation occurs. While on the other
it is expected that results from targeted sampling have been years high mycotoxin contamination rates are observed, on
omitted by the authors when reporting their observations on the other years the prevalence of mycotoxins is low
mycotoxin incidences. (Edwards and Jennings 2018; Kos et al. 2013; Meyer et al.
2019). Climate change is further believed to affect fungal
growth and agricultural practices and, subsequently to the
Amount of occurrence data varies across the world
mycotoxin incidences and levels on the agricultural crops
While generally a large quantities of mycotoxin occurrence (Battilani et al. 2013; Paterson and Lima 2017; Van der Fels-
data are currently available, there are still many geographical Klerx et al. 2012a, 2012b, 2016). It is also projected that
regions in the world from where the data are limited or changes in climate will shift the geographic distributions of
lacking. Typically, these are developing countries or coun- mycotoxin-producing fungi and hence the patterns of myco-
tries with ongoing crisis events. This lack of data is evident toxin occurrence in the world. The amount of mycotoxin-
only by looking into the many international mycotoxin contaminated crops are generally anticipated to increase
assessments issued by JECFA and IARC or the WHO Global with the global warming, but at the same time major varia-
Environment Monitoring System (GEMS)/Food contami- tions in mycotoxin contamination are also expected as well
nants database on mycotoxin occurrence data collected by as decreased occurrences of mycotoxins (Battilani et al.
WHO, where majority of the data are from Europe, 2013; Paterson and Lima 2017; Van der Fels-Klerx et al.
Americas and Western Pacific (WHO 2018). It is clear that 2012a, 2012b, 2016). Furthermore, most of the mycotoxins
limited and absence of data on mycotoxins lead to high are known to be chemically stable, with pre- and post-har-
uncertainties in the prevalence estimates or entirely prevents vest mitigation strategies, such as with food and feed proc-
their calculations. essing, only eliminating them to a certain extent (Leslie and
Contrary to the limited government and publicly avail- Logrieco 2014).
able data, a wealth of mycotoxin occurrence data is assumed
to be in the possession of the food and feed companies. To
Impact of global trade on mycotoxin prevalence data
produce safe and high-quality food and feed commodities
from agricultural crops, and to comply with the statutory Cereals are the core of human nutrition and their use for
levels, it is apparent that the largest amount of occurrence food and feed production will only increase during the
data is generated by the private food and feed business sec- coming decades (FAO 2015). World cereal production in
tor. Of these data, only a fraction is available for estimating 2017/2018 was estimated to be 2094 million tonnes by EC
the worldwide mycotoxin contamination rates as most of (EC 2018a). Globally, food remains as the most important
the companies do not publish their data or submit them to use of cereals but due to the higher demand for protein-
the organizations like WHO, FAO or EFSA. The occurrence based foods, the need of cereals for feed materials will
data used by these international food safety bodies are from increase. Global food and feed trade has grown rapidly in
the official control laboratories and from research organiza- the recent decades and is only expected to grow (FAO
tions. It is apparent that at the stage of the official food con- 2015). The countries in every region of the world have
trol, the food business operators have already applied their become more involved in the global trade than ever before,
in-house control systems and discarded the most contami- either as exporters or importers or both. For example, dur-
nated lots. Therefore, largely the data from the official con- ing the past decades the EU-imports from developing
trols represent the mycotoxin-results which are compliant countries have increased sharply (Bureau and Swinnen
with the MLs. This leads to a skewness of the dataset and a 2018). Currently, the EU exports mainly wheat and barley,
lower prevalence of mycotoxins than it likely would be in approximately 20% of the EU’s wheat crop is exported
the crops at the farm level. For example, majority of the annually, while large amounts of food-grade rice and feed
EFSA data are from the national control programs organized (e.g. maize) are imported by the EU (EC 2018b). A coun-
by the EU Member States (Eskola et al. 2018). try like Japan, where less than 40% of consumed foods
were domestic in 2016, relies heavily on imported foods,
agricultural foods comprising about 70% (MHLW 2018).
Climate influences mycotoxin prevalence
As agricultural food and feed crops are widely internation-
It is common knowledge that different mycotoxin producing ally traded, the estimates of mycotoxin occurrence and
fungal species favor different climate conditions making prevalence reported in the literature for specific country or
 CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 2779

region likely comprise results for both domestic and prevalence varied from 50% in Asia to 76% in Africa (Lee
imported crops. and Ryu 2017). These percentages support the contamin-
ation rates of 46–64% from the eight years when 17,300 feed
samples from America, Asia and Europe were analyzed
Previously reported global incidences for the (Streit et al. 2013). Biomin survey of around 15,100 feed
regulated mycotoxins samples sourced globally in 2016–2017 showed a lowest inci-
While JECFA did not provide global estimates for AF-conta- dence of 65% for DON in about 4700 European samples, of
minated foods in its reports, IARC (2012) reported that it 70% in about 650 Middle East and African samples and the
can vary from a few percent to 30% or more depending on highest DON-incidence of 82% in about 4800 South
the food product and geographical region. JECFA (2008), American samples (Biomin 2017). JECFA reported that the
however, suggests that about 20% of tree nuts from different incidence of T-2 was 11% and of HT-2 14% in grain sam-
regions are contaminated. According to EFSA the prevalence ples of barley, maize, oats, rice, rye and wheat for human
in nuts for total AF (sum of AF B1, B2, G1 and G2) above consumption (mainly from Europe) (JECFA 2002). About
LOD varied from 10% in cashew nuts to 40% in pistachio 30% of the feed samples from Europe and South-America
and Brazil nuts (EFSA 2009). Total AF contamination above were positive for T-2, while only a few percentages were
LOD/LOQ was 45% in dried figs (EFSA 2012a) and 13% in positive from other geographical regions in all nearly 8000
cereals (EFSA 2013). Recently, Lee and Ryu (2017) assessed samples analyzed by Biomin (2017).
the incidence of AFs in unprocessed food-grade cereals (bar- For the incidence of ZEN in contaminated food-crops
ley, maize, wheat, rice, and other cereals e.g. oats) in Africa, across the world, JECFA (2000) reported a large variation
America, Asia and Europe compiling data from papers pub- (0–100%), suggesting roughly 30–40% of contamination. In
lished over 10 years since 2006. Total AF-prevalence in these unprocessed food-grade cereals (barley, maize, wheat, rice
regions was 55%, varying from 15% in the American contin- and other cereals) an incidence of 46% of ZEN across the
ent to 63% in Asia. Similar global contamination rates from globe was estimated based on the data published in
16% up to 33% for AFs over eight years were reported for 2006–2016 (Lee and Ryu 2017). The ZEN-incidence in cere-
17,300 feed samples mainly from America (15%), Asia als was the lowest in Asia (15%) and the highest (59%) in
(40%) and Europe (38%) in 2013 (Streit et al. 2013). Africa (Lee and Ryu 2017). The range of 31–43% for ZEN-
Amongst the around 14,400 globally collected feed samples positive feed samples over the 8-year period in 17,300 sam-
in 2016–2017, the highest AF-incidence of 23% was found ples primarily from Asia, America and Europe (Streit et al.
in nearly 7000 samples from North America (Biomin 2017). 2013) was similar to those of JECFA (2000) and Lee and
In 2001, JECFA reported that over 60% of the data on Ryu (2017). The global data of around 15,600 feed samples
maize and maize-derived foods from nine different world indicated that the ZEN-incidence was at the region of
countries were FB-contaminated (above LOD) (JECFA 40–50% (Biomin 2017).
2001). In 2012, it reported that 70% of the results for maize Based on the worldwide data, an incidence of OTA-posi-
samples worldwide were above the LOD or LOQ for FB1 tive samples varied depending on the type of food crop but
(JECFA 2012). In 2017, the maize and maize-based foods was generally lower than 50% for cereals (JECFA 2001,
had the highest occurrence of FB1, with the frequency close 2007, 2008). Similar incidences, varying from 20% in Europe
to 50% exceeding the LOQ (data from Europe, Canada and to 42% in America, in unprocessed food-grade barley, maize,
Japan) (JECFA 2017). A similar global prevalence of 61% for wheat, rice, and other cereals were reported by Lee and Ryu
FBs was calculated by Lee and Ryu (2017) for unprocessed (2017) based on the published data from the 10 years. They
food-grade barley, maize, wheat, rice, and other cereals from estimated a global incidence of 29%. Only slightly lower
the published data from 10 years since 2006. The total inci- prevalence of 17–28% for OTA was found in the 8-year-
dence varied from 39% of Europe to 95% of America (Lee long study comprising 17,300 feed samples mainly from
and Ryu 2017). For FBs during a 8-year period, the preva- America, Asia and Europe (Streit et al. 2013). Similar preva-
lence was 50–63% in the 17,300 feed samples collected from lence of up to 27% was reported by Biomin (2017) for 6500
America (15%), Asia (40%) and Europe (38%) (Streit et al. global feed samples. In the three largest sets of samples (n ¼
2013). Within nearly 12,800 global feed-samples from 2016 1500 to 2500), the OTA-contamination rates were 5%
to 2017, 85% of about 2500 Asian samples were contami- for North America, 24% for Europe and 27% for Asia
nated with FBs, 75% of 5500 South American samples and (Biomin 2017).
63% of about 450 African samples (Biomin 2017).
JECFA (2011) reported that worldwide 50% of oats
Assessment of prevalence of mycotoxins in food
results, 70% of barley, wheat and rice results and 90% of
crops by using EFSA and Biomin data
maize exceeded the LODs for DON. In its earlier evaluation
(JECFA 2001), DON-contamination in cereal grains varied To assess mycotoxin occurrence frequency at different thresh-
from 30% in rice to 60% in wheat and barley up to 70% in olds two data sets were investigated. The first set was obtained
oats (levels above LOQs). A global DON-incidence of nearly from EFSA with commodities destined for human consump-
60% was reported by Lee and Ryu (2017) for unprocessed tion; the second set was commercial feed analysis data set pro-
food-grade barley, maize, wheat, rice, and other cereals vided by Biomin Inc. EFSA data were obtained through an
based on the published data in 10 years since 2006. The access to information request (EFSA Ref. 17238686; PAD
2780 M. ESKOLA ET AL.

Table 2. Summary statistics of the EFSA and Biomin data sets investigated for occurrence of six mycotoxin compounds, i.e. AFs (sum
of AFB1, AFB2, AFG1 and AFG2), DON, ZEN, OTA, FBs (sum of FB1 and FB2) and the sum of T-2 and HT-2.
All samples and analyses EFSA data Biomin data
Total number of analyses 536,013 (single toxin) 1942 (multi-toxin)
Total number of samples 175,270 1942
Countries with three largest contributions (n analyses) Germany (173,341) Germany (348)
Netherlands (87,662) South Africa (229)
France (33,087) Austria (234)
Methods with 3 largest contributions (n analyses) LC-MS (166,885) LC-MS (1942)
HPLC-FD (135,169)
ELISA (36,200)
Years with 3 largest contributions (n analyses) 2013 (88,025) 2013 (139)
2014 (126,419) 2014 (874)
2015 (126,179) 2015 (837)
Grain samples only
Number of grain analyses 74,696 (single-toxin) 560 (multi-toxin)
Number of grain samples 21,841 560
Countries with 3 largest contributions (n analyses) Germany (25,367) Germany (224)
Netherlands (10,878) Austria (143)
Romania (9025) Denmark (46)
Years with 3 largest contributions (n analyses) 2013 (13,428) 2013 (26)
2014 (17,814) 2014 (227)
2015 (16,220) 2015 (292)

2017 017) and was an extract of the EFSA chemical occurrence emerging toxins. In this study, we discuss outcomes of the
database concerning selected mycotoxin concentration data of statistical analysis of a very different subset, i.e. regulated
all EU Member States. For the sampling years from 2010 to mycotoxins and mycotoxins with guidance levels in grains of
2015 the following information was obtained from EFSA: (1) all kinds as reported to obtain an estimate of contamination
Year of sampling, (2) month of sampling, (3) country of origin frequencies and levels not provided previously. Table 2 pro-
of the sample, (3) commodity sampled, (4) method of analysis vides a summary of the datasets and subsets investigated.
employed, (5) limit of detection of the method employed, and It should be noted that for analyses reported in the EFSA
(6) co-occurring species with concentration levels. The data data set a large variety of methods was reported, including
set contained 536,013 analyses from 175,270 samples of EU several methods classified as LC-MS (with and without the
regulated mycotoxins with the ancillary data mentioned type of mass filter mentioned), high performance liquid
above. At the request of some national data providers, EFSA chromatography (HPLC) with several detectors (e.g. fluores-
provided a replacement of the country name by the indication cence, diode array) and enzyme-linked immunosorbent
‘EU’ due to ‘a risk for indirect identification of economic assay (ELISA). Furthermore, complete method information
operators, which may be country-specific and is to be consid- was not always provided, such as no classification at all for
ered as a commercial interest deserving protection’ as speci- 105,192 samples and ‘HPLC with standard detection meth-
fied by EFSA in its cover letter to the authors. In the dataset ods’ for 40,336 samples. Incomplete method information
31,651 samples were labeled as ‘EU’. The EFSA occurrence was not a reason to exclude an analysis, although a very
data on food and feed were collected from the European mar- small number of erroneously reported samples analysis
kets but were not necessarily cultivated or originated from methods was removed prior to analysis, e.g. samples ana-
Europe. Owing to the fact that the FAO estimate is for food lyzed for mycotoxin concentrations using atomic absorption
crops, to calculate the frequency of the contaminated samples spectroscopy (AAS; n ¼ 149). Because of this large number
in the EFSA dataset, results for processed food items were of different methods, an important consideration was to
excluded and occurrence in raw materials only was evaluated. choose suitable LOD and LOQ values for statistical analysis.
The dataset obtained from Biomin was described in detail Regarding the analytical method employed the Biomin data-
in an earlier report by Kovalsky et al. (2016). In brief, the ori- set was more consistent. Analysis was performed using a
ginal data set covered the years 2012–2015 for 1942 samples single LC-MS method by a single laboratory, thus eliminat-
originating from 46 countries around the globe with a subset ing a significant source of variability and allowing the use of
consisting of EU countries that was used for analysis. Samples a single LOD and LOQ value for a specific matrix.
were directly provided by farmers for analysis and commod- Considering that the FAO 25% estimate was most likely
ities analyzed were primarily destined for use in animal feed, calculated for samples exceeding the Codex MLs, one of the
although the ultimate fate of the commodities remains unclear. cutoff values for the data in this study was set at the current
Mycotoxin and metabolite concentrations were determined Codex MLs (Table 1) with the caveat that they might have
using a single multi-mycotoxin LC-MS method reported by changed from the (initial and unknown) levels, when the
Malachova et al. (2014). In the previous study by Kovalsky estimate of 25% was adopted over three decades ago. The
et al. focused on mycotoxin and metabolite concentrations (57 percentages of mycotoxin contaminated samples above the
compounds) in three selected matrices with a focus on the co- current EU MLs and reported LODs/LOQs were also calcu-
occurrence of regulated mycotoxins with masked and lated from the EFSA and Biomin occurrence data.
 CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 2781

Samples and mycotoxin concentrations in both data sets from the field without further prescreening. Both datasets,
were of different origin and were chosen to serve different therefore, provide a realistic range of mycotoxin occurrence
purposes for this study: The EFSA data set contained data from grain batches. For each of the samples selected above,
primarily analyzed and submitted by public food safety and subset by country and commodity, concentrations for the
research organizations from samples of food crops destined following toxins were determined for both EFSA and
for human consumption. The food crops likely underwent Biomin subsets: AFs (AFLAS, sum of AFB1, AFB2, AFG1
some form of prescreening by e.g., food business operators and AFG2), DON, ZEN, OTA, FBs (FUMS, sum of FB1 and
in their internal quality control prior to submission of sam- FB2) and the sum of T-2 and HT-2 (T2HT2).
ple for analysis. The Biomin dataset, on the other hand, pro- In order to find the percentage mycotoxin contamination
vided data from samples submitted by farmers around the and provide a detailed measure of contaminated grains in
globe for investigation of mycotoxin occurrence and com- samples, several thresholds were defined and the percentage
modities were primarily intended to be used as animal feed. samples within each category was reported: (1) < LOD, (2)
Consequently, results obtained from EFSA data likely pro- between LOD and lower legal ML/guidance level, (3)
vided a lower frequency cutoff for mycotoxin occurrence. between lower and upper legal ML/guidance levels and (4)
The Biomin data, on the other hand, illustrated the upper > upper legal ML/guidance level. While the definition of
EU MLs/guidance levels and Codex MLs for the above cate-
threshold of mycotoxin occurrence in grains. Samples were
gories was straightforward, the selection of a single LOD
intended for animal feed production and sampled directly
from the EFSA data posed some challenges. LODs for statis-
tical analysis were, therefore, defined as follows in order to
obtain a single value to be used for each toxin:
For the EFSA data sets a histogram of the different LODs
from all analyses of a specific toxin in each grain subset was
plotted. The LOD with the highest frequency was chosen for
subsequent analysis and plotting. This approach taken was
considered to be appropriate with the underlying rationale
that the few laboratories reporting higher LODs did so to
make the employed method suitable for the concentrations
reported (i.e. fit for purpose) without assessing the method’s
full potential. Since other analyses showed lower LODs than
the chosen value, reported results represent a conservative
estimate regarding the frequency of mycotoxins detected. On
the other hand, very few analyses had a higher than the chosen
LOD, suggesting a marginal influence of the non-detects asso-
Figure 1. Histogram of reported LODs for ZEN in the EFSA grain subset. ciated with higher LODs on the statistical analyses. Figure 1

Table 3. Thresholds employed for statistical analysis of EFSA and Biomin data.
Codex level (mg/kg) EU food level (mg/kg) EU feed level (mg/kg)
Used for dataset EFSA EFSA Biomin
Sum of AFs
LOD 0.1 0.1 1.5
Lower legal levela NA 2 5
Upper legal levela NA 10 20
DON
LOD 10 10 1.5
Lower legal levela 2000 1250 900
Upper legal levela NA 1750 12,000
ZEN
LOD 1 1 0.15
Lower legal levela NA 75 100
Upper legal levela NA 350 3000
Sum of T-2 and HT-2 toxins
LOD 5 5 10
Lower legal levela NA 50 50
Upper legal levela NA 1000 2000
Sum of FBs
LOD 20 20 4
Lower legal levela 4000 1000 5000
Upper legal levela NA 4000 60,000
OTA
LOD 0.1 0.1 1.5
Lower legal levela 5 3 100
Upper legal levela NA 5 250
Sum of AFs was defined as the sum of AFB1, AFB2, AFG1 and AFG2; the sum of FBs was defined as the sum of FB1 and
FB2. LODs were determined from reported data.
a
Legal level ¼ EU ML or EU guidance level or Codex ML.
2782 M. ESKOLA ET AL.

provides an example for ZEN with the final LOD chosen to be Employing the thresholds from Table 3, plots were cre-
1 mg/kg. For Biomin data, which were acquired employing a ated with the percentage of samples for each toxin falling
single analytical method in a single laboratory made the han- within one of the four categories: < LOD, between LOD
dling of the LOD straightforward. Table 3 provides a summary and the lower limit, between the lower and upper limit and,
of the thresholds and limits employed for statistical analysis. finally above the upper limit. Limits were chosen to be EU
MLs (for food) and guidance values (for feed) and Codex
MLs. Figure 3 depicts results for EU food MLs applied to
Determination of frequency of contaminated samples at the EFSA subset for grains; Figure 4 shows the result
different thresholds obtained for the Biomin dataset and EU feed ML and guid-
ance values.
Figure 2 shows an overview of the concentration distribution Quantitative plots confirmed the initial assessment as
for the six investigated mycotoxins in grain samples from (a) shown in Figure 4. While most samples showed concentra-
EFSA and (b) Biomin samples. Means and medians were very tions below the LOD (i.e. no toxin was detected), there is a
low due to the fact that most of the results were found to be certain percentage of samples that are contaminated (with
below the LOD. But there were occurrences of samples with concentrations between the LOD and the lower legal level)
high concentrations of especially DON and ZEN in both sam- with DON and ZEN showing the highest occurrence. Very
ple sets. While these numbers were to be expected for Biomin few samples were also found to contain elevated concentra-
samples as they originated directly from the fields, samples tions above the legal levels. Notable are the high frequencies
without further prescreening, EFSA data submitted by the of occurrence of DON, ZEN and FBs, with 40, 20 and 15%
public organizations throughout the EU also showed an of samples, respectively, containing detectable concentra-
important number of samples with high DON and ZEN con- tions, but below the EU lower ML (Figure 4a).
centrations, despite assumed prescreening prior to analysis. For the Codex MLs (Figure 3 b) applied to EFSA data it
The dense cluster of points with higher concentrations (often has to be noted that only one level is typically provided
above the legal and guidance levels listed in Table 3) in the (Table 3), reducing the number of categories to three
DON and ZEN boxplots suggested that an important number (< LOD, between LOD and the ML and above the ML). We
of samples featured high DON and ZEN concentrations, war- assume that the calculated percentages reported here are
ranting further quantitative analysis. appropriate for comparison with the 25% reported by FAO.

Figure 2. Boxplots of toxin concentration distribution in (a) EFSA and (b) Biomin grain samples.

Figure 3. Percentage of EFSA grain samples falling within one of the four categories for each toxin as follows: (1) <LOD, (2) between LOD and lower legal level, (3)
between lower and upper legal level and (4) above the upper legal level. (a) EU food MLs and (b) Codex MLs. For the respective limits see Table 3.
 CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 2783

Figure 4. Percentage of Biomin grain samples falling within one of the four categories for each toxin as follows (from left to right): (1) <LOD, (2) between LOD and
lower legal level, (3) between lower and upper legal level and (4) above the upper legal limit. (a) Samples of global origin with EU feed guidance values applied
and (b) Samples of EU origin with EU feed guidance values applied. For limits see Table 3.

The calculated percentages in Figure 3a and b are very com- but below the lower legal level or guidance value), was sig-
parable and the percentages above the lower EU ML and the nificantly higher. While no historical reference value for
Codex ML are in the single digits, supporting the prescreen- comparison was available for this category, its percentage
ing hypothesis. has likely increased over the last decades. Method develop-
Biomin data for samples originating directly from the ment progress has been significant, and sensitivity and
field without further elimination of highly contaminated selectivity of current state-of-the-art methods have greatly
samples by prescreening shows an entirely different picture. increased. These improvements provide the tools for further
Applied limits followed EU guidance values for feed. The investigation of human and animal low-level exposures to
application of EU and Codex food MLs were not considered mycotoxins, in particular exposures to mycotoxin-mixtures,
appropriate since samples were preliminary destined for the and their contributions to the associated health risks. On
production of animal feed. Furthermore, the MLs of DON the other hand, prevention strategies in the field and during
and ZEN in feed, which showed the highest percentages of storage would have helped to curb mycotoxin production
contamination above the lower legal/guidance level, are over the last decades (Jard et al. 2011; Magan and Aldred
similar for food and feed (see Table 3), thus making per- 2007). The number of agricultural crops infected with toxi-
centage contamination data comparable for at least the most genic fungi is indeed quite high as the data > LOD clearly
commonly found toxins. demonstrates with up to 80% of samples showing detectable
Comparing plots for samples of (a) global and (b) EU concentrations of ZEN and 60% of samples showing detect-
origin show a high similarity, indicating that contamination able DON concentrations. This constitutes a large pool of
percentages measured at the field levels are quite similar for infected crops potentially to produce even large quantities of
mycotoxins commonly found in temperate climates, such as mycotoxins in a warming climate. With climate change
DON and ZEN. While established food safety systems are favoring fungal growth and toxin production the global
able to significantly reduce the risk of contaminated samples mycotoxin problem is expected to worsen (Van der Fels-
entering the food chain (as show in Figure 3 for EFSA Klerx et al. 2012a, 2012b; 2016; Battilani et al. 2013;
data), the actual occurrence percentages (for EU countries Paterson and Lima 2017).
presented in Figure 4 (b)) are significantly higher when
looking at the data for feed. For ZEN the percentage of sam-
Summary and conclusion
ples with detected mycotoxin concentrations reached about
80% with similar values for global and EU data from the In this paper we strive to provide a rationale for a widely
Biomin dataset. Among these samples about 60% showed cited FAO estimate of 25% mycotoxin contamination in
detected concentrations and while more than 20% of sam- food crops around the globe. The exact origin and the basis
ples were found to be contaminated above the lower limit for the estimate are unknown today and little details are
(lower EU guidance value for feed). The situation for DON available apart from secondary sources. The presented
was similar with 40% detected concentrations and slightly study provides a review of the occurrence of mycotoxins in
less than 20% contaminated above the lower limit. grain from the literature with a focus on reports of JECFA.
Provided that the FAO 25% figure was established as per- It has conducted extensive global reviews of the occurrence
centage of samples above the Codex MLs and assuming that of known health-relevant and regulated mycotoxins, such
Codex MLs have not changed, the recent data for non-pre- as AFs, DON, ZEN, FBs, T-2 and HT-2 and OTA. Based
screened (i.e. feed) samples showed a similar level of con- on the literature data, the global mycotoxin prevalence in
tamination with around 20%. The percentage of samples food crops varies largely depending on many factors, such
with detected concentrations, on the other hand (> LOD, as the mycotoxin of concern, used analytical methods and
2784 M. ESKOLA ET AL.

reporting of the results, but it appears that the prevalence Acknowledgements

for the detected mycotoxins is up to 60–80%. Analyzing This research was supported by the European Union Horizon 2020
recent data on food grains for the years 2010–2015 pro- research and innovation programme under grant agreement no. 692195
vided by EFSA (from national food control programs and, (‘MultiCoop’). The authors thank Biomin Inc. for the use of the
therefore, most likely prescreened by the food business Biomin World Mycotoxin Survey 2017 data. In addition, we thank the
European countries who have submitted their mycotoxin data to EFSA.
operators for quality before sampling) and Biomin Inc. We also wish to thank the FAO officials for a fruitful email discussion.
(grains sampled at source at the farm-level) we found simi-
lar percentages in grain samples for DON (60%) and ZEN
(80%), the two most prevalent mycotoxins in temperate cli- ORCID
mates. In addition to the detected mycotoxins, the analysis Mari Eskola https://orcid.org/0000-0002-5298-8976
provides additional detail using different regulatory thresh- Gregor Kos http://orcid.org/0000-0001-8223-7308
olds from the current EU legislation and Codex standards Christopher T. Elliott https://orcid.org/0000-0003-0495-2909
Sultan Mayar https://orcid.org/0000-0002-2396-0114
for food and feed. Results showed in the region of 20% of
Rudolf Krska https://orcid.org/0000-0001-6843-9755
samples being contaminated above the lower regulatory EU
and Codex level for materials sampled at source, but lower
contamination levels (<10%) for food-grade grain samples. References
Comparing the FAO 25% estimate to values above EU and Adekoya, I., P. Njobeh, A. Obadina, S. Landschoot, K. Audenaert, S.
Codex limits (typically 20% for samples collected at source) Okoth, M. De Boevre, and S. De Saeger. 2019. Investigation of the
would appear to confirm the original estimate. However, a metabolic profile and toxigenic variability of fungal species occur-
comparison with the percentage of samples found to have ring in fermented foods and beverage from Nigeria and South
Africa using UPLC-MS/MS. Toxins 11 (2):E85. doi: 10.3390/
mycotoxins present (in 60–80% of all samples for DON toxins11020085.
and ZEN alone) these figures greatly underestimate their Alassane-Kpembi, I., G. Schatzmayr, I. Taranu, D. Marin, O. Puel, and
occurrence. While progress in method development likely I. P. Oswald. 2017. Mycotoxins co-contamination: Methodological
led to an increase in the percentage of detected mycotoxins aspects and biological relevance of combined toxicity studies.
Critical Reviews in Food Science and Nutrition 57 (16):3489–507.
(due to lower LODs), active mitigation strategies have most doi: 10.1080/10408398.2016.1140632.
likely contributed to a reduction of high contamination Arroyo-Manzanares, N., K. De Ruyck, V. Uka, L. Gamiz-Gracia, A. M.
batches, estimates that cannot be produced with the cur- Garcıa-Campa~ na, S. De Saeger, and J. Diana Di Mavungu. 2018. In-
rently available data, but will need to be considered in house validation of a rapid and efficient procedure for simultaneous
determination of ergot alkaloids and other mycotoxins in wheat and
future investigations. Climate change may have also had an maize. Analytical and Bioanalytical Chemistry 410 (22):5567–81. doi:
impact on the higher incidence and will only serve to 10.1007/s00216-018-1018-6.
increase this level of contamination in future years. Before Battilani, P., M. Camardo Leggieri, V. Rossi, and P. Giorni. 2013.
an accurate estimate can be provided, statistically relevant AFLA-maize, a mechanistic model for Aspergillus flavus infection
and aflatoxin B1 contamination in maize. Computer and Electronics
amount of high-quality data should be made available. in Agriculture 94:38–46. doi: 10.1016/j.compag.2013.03.005.
Therefore, large occurrence data surveys on mycotoxin Battilani, P., V. Rossi, P. Giorni, A. Pietri, A. Gualla, H. J. van der
contamination in food-crops, with harmonized sampling Fels-Klerx, C. J. H. Booij, A. Moretti, A. Logrieco, F. Miglietta, et al.
strategies and analytical performance criteria, across the 2012. Modelling, predicting and mapping the emergence of aflatox-
ins in cereals in the EU due to climate change. European Food
world over several growing seasons should be conducted. Safety Authority (EFSA) Supporting Publication 9 (EN-223):172.
Humans and animals are typically exposed to mycotoxin Battilani, P., P. Toscano, H. J. Van der Fels-Klerx, A. Moretti, M.
mixtures which can induce combined adverse health Camardo Leggieri, C. Brera, A. Rortais, T. Goumperis, and T.
effects. However, in those events where the concentration Robinson. 2016. Aflatoxin B1 contamination in maize in Europe
increases due to climate change. Scientific Reports 6 (1):24328. doi:
exceedance of the statutory limits (e.g. 20% or 25% of the 10.1038/srep24328.
samples) would lead to regular high dietary exposures to Biomin. 2017. Biomin World mycotoxin survey 2017. Annual Report
mycotoxins, and to their mixtures in particular, adverse No. 14. Accessed May 11, 2018. https://www.biomin.net/en/blog-
health consequences can be expected. It is also of immense posts/2017-biomin-mycotoxin-survey-results/.
Boutrif, E., and C. Canet. 1998. Mycotoxin prevention and control:
importance that the detectable levels (based on our estima- FAO programmes. Revue de Medecine Veterinaire 149:681–94.
tion up to 60–80% of the samples) are not overlooked as Brouwer-Brolsma, E. M., L. Brennan, C. A. Drevon, H. van Kranen, C.
through diets humans and animals are exposed to mixtures Manach, L. O. Dragsted, H. M. Roche, C. Andres-Lacueva, S. J. L.
of different mycotoxins. Many of these possibly additive Bakker, J. Bouwman, et al. 2017. Combining traditional dietary
assessment methods with novel metabolomics techniques: Present
and synergistic effects are yet to be toxicologically investi- efforts by the Food Biomarker Alliance. Proceedings of the Nutrition
gated and their influence on the public health to be Society 76 (4):619–27. doi: 10.1017/S0029665117003949.
assessed by applying harmonized methodologies (EFSA Bui-Klimke, T., and F. Wu. 2014. Evaluating weight of evidence in the
2019; Eskola et al. 2018). mystery of Balkan Endemic Nephropathy. Risk Analysis 34 (9):
1688–705. doi: 10.1111/risa.12239.
Bureau, J.-C., and J. Swinnen. 2018. EU policies and global food secur-
ity. Global Food Security 16:106–15. doi: 10.1016/j.gfs.2017.12.001.
Declaration of interest CFIA. 2017. RG-8 Regulatory guidance. Contaminants in feed (for-
merly RG-1, Chapter 7), Section 1: Mycotoxins in livestock feed.
The authors do not declare any conflict of interest. Canadian Food Inspection Agency (CFIA). Accessed March 15,
 CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 2785

2018. http://www.inspection.gc.ca/animals/feeds/regulatory-guidance/ related to deoxynivalenol (DON) as undesirable substance in animal

rg-8/eng/1347383943203/1347384015909. feed. EFSA Journal 72:1.
Codex Alimentarius. 1995. Codex Alimentarius international food EFSA. 2004c. Opinion of the Scientific Panel on Contaminants in the
standards, General standard for contaminants and toxins in food Food Chain (CONTAM Panel) related to ochratoxin A (OTA) as
and feed. CXS 193-1995. Adopted in 1995. Revised in 1997, 2006, undesirable substance in animal feed. European Food Safety
2008, 2009. Amended in 2010, 2012, 2013, 2014, 2015, 2016, 2017. Authority. EFSA Journal 2:101.
Food and Agriculture Organization of the United Nations (FAO), EFSA. 2005a. Opinion of the Scientific Panel on Contaminants in the
the World Health Organization (WHO). Food Chain (CONTAM Panel) related to fumonisins as undesirable
Deepa, N., and M. Y. Sreenivasa. 2017. Fumonisins: A review on its substances in animal feed. European Food Safety Authority. EFSA
global occurrence, epidemiology, toxicity and detection. Journal of Journal 3:235.
Veterinary Medicine and Research 4 (6):1093. EFSA. 2005b. Opinion of the Scientific Panel on Contaminants in the
de Lourdes Mendes de Souza, M., M. Sulyok, O. Freitas-Silva, S. Soares Food Chain (CONTAM Panel) related to ergot as undesirable sub-
Costa, C. Brabet, M. Machinski Junior, B. Leiko Sekiyama, E. stance in animal feed. European Food Safety Authority. EFSA
Azevedo Vargas, R. Krska, and R. Schuhmacher. 2013. Cooccurrence Journal 3:225.
of mycotoxins in maize and poultry feeds from Brazil by liquid EFSA. 2006. Scientific opinion related to ochratoxin A in food.
chromatography/tandem mass spectrometry. The Scientific World European Food Safety Authority. EFSA Journal 4:365.
Journal 2013:427369. doi: 10.1155/2013/427369. EFSA. 2007. Opinion of the Scientific Panel on Contaminants in the
Diana Di Mavungu, J., D. A. Larionova, S. V. Malysheva, C. Van Food Chain (CONTAM Panel) related to the potential increase of
Peteghem, and S. De Saeger. 2011. Survey on ergot alkaloids in cere- consumer health risk by a possible increase of the existing max-
als intended for human consumption and animal feeding. European imum levels for aflatoxins in almonds, hazelnuts and pistachios and
Food Safety Authority (EFSA) Supporting Publication 8:EN-214. derived products. European Food Safety Authority. EFSA Journal 5:
Doppler, M., B. Kluger, C. Bueschl, C. Schneider, R. Krska, S. 446.
Delcambre, K. Hiller, M. Lemmens, and R. Schuhmacher. 2016. EFSA. 2009. Statement of the Scientific Panel on Contaminants in the
Stable isotope-assisted evaluation of different extraction solvents for Food Chain (CONTAM Panel) on a request from the European
untargeted metabolomics of plants. International Journal of Commission on the effects on public health of an increase of the
Molecular Sciences 17 (7):1017. doi: 10.3390/ijms17071017. levels for aflatoxin total from 4 lg/kg to 10 lg/kg for tree nuts other
Dorokhin, D., W. Haasnoot, M. C. Franssen, H. Zuilhof, and M. W. than almonds, hazelnuts and pistachios. European Food Safety
Nielen. 2011. Imaging surface plasmon resonance for multiplex Authority. EFSA Journal 6:1168.
microassay sensing of mycotoxins. Analytical and Bioanalytical EFSA. 2010. Statement of the Scientific Panel on Contaminants in the
Chemistry 400 (9):3005–11. doi: 10.1007/s00216-011-4973-8. Food Chain (CONTAM Panel) on recent scientific information on
EC. 2006a. Commission Regulation (EC) No 1881/2006 of 19 the toxicity of ochratoxin A. European Food Safety Authority. EFSA
December 2006 setting maximum levels for certain contaminants in Journal 8:1626.
foodstuffs. European Commission. Official Journal of the European EFSA. 2011. Opinion of the Scientific Panel on Contaminants in the
Union L 364:5–24. Food Chain (CONTAM Panel) on the risks for public health related
EC. 2006b. Commission Regulation (EC) No 401/2006 of 23 February to the presence of T-2 and HT-2 toxins in food and feed. European
2006 laying down the methods of sampling and analysis for the offi- Food Safety Authority. EFSA Journal 9:2481.
cial control of the levels of mycotoxins in foodstuffs. European EFSA. 2012a. Effect on dietary exposure of an increase of the levels for
Commission. Official Journal of the European Union L 70:12–34. aflatoxin total from 4 lg/kg to 10 lg/kg for dried figs. European
EC. 2006c. Commission Recommendation of 17 August 2006 on the Food Safety Authority. EFSA Supporting Publications 9:EN-311.
presence of deoxynivalenol, zearalenone, ochratoxin A, T-2 and HT- EFSA. 2012b. Opinion of the Scientific Panel on Contaminants in the
2 and fumonisins in products intended for animal feeding (2006/ Food Chain (CONTAM Panel) on ergot alkaloids in food and feed.
576/EC). European Commission. Official Journal of the European European Food Safety Authority. EFSA Journal 10:2798.
Union L 229:7–9. EFSA. 2012c. Opinion of the Scientific Panel on Contaminants in the
EC. 2013. Commission Recommendation of 27 March 2013 on the Food Chain (CONTAM Panel) on the risks to human and animal
presence of T-2 and HT-2 toxin in cereals and cereal products health related to the presence of citrinin in food and feed. European
(2013/165/EU). European Commission. Official Journal of the Food Safety Authority. EFSA Journal 10:2605.
European Union L 91:12–5. EFSA. 2013. Aflatoxins (sum of B1, B2, G1, G2) in cereals and cereal-
EC. 2018a. European Commission, Directorate General Agri, EU Crops derived food products. European Food Safety Authority. EFSA
Market Observatory – Cereals, updated on 1.3.2018. Accessed Supporting Publications 10:EN-406.
March 10, 2018. https://circabc.europa.eu/sd/a/df783825-dc2c-4673- EFSA. 2014. Opinion of the Scientific Panel on Contaminants in the
a66e-0cbd5c5be336/cereals-dashboard_en.pdf. Food Chain (CONTAM Panel) on the risks for human and animal
EC. 2018b. European Commission, Agriculture and Rural health related to the presence of modified forms of certain mycotox-
Development, Cereals, oilseeds and protein crops, rice. Accessed ins in food and feed. European Food Safety Authority. EFSA Journal
March 10, 2018. https://ec.europa.eu/agriculture/cereals_en. 12:3916.
Ediage, E. N., J. D. Di Mavungu, I. Y. Goryacheva, C. Van Peteghem, EFSA. 2016. Opinion of the Scientific Panel on Contaminants in the
and S. De Saeger. 2012. Multiplex flow-through immunoassay for- Food Chain (CONTAM Panel) on the appropriateness to set a
mats for screening of mycotoxins in a variety of food matrices. group health-based guidance value for zearalenone and its modified
Analytical and Bioanalytical Chemistry 403:265–78. doi: 10.1007/ forms. European Food Safety Authority. EFSA Journal 14:4425.
s00216-012-5803-3. EFSA. 2017a. Opinion of the Scientific Panel on Contaminants in the
Edwards, S. G., and P. Jennings. 2018. Impact of agronomic factors on Food Chain (CONTAM Panel) on the appropriateness to set a
fusarium mycotoxins in harvested wheat. Food Additives & group health-based guidance value for nivalenol and its modified
Contaminants: Part A 35 (12):2443–54. doi: 10.1080/19440049.2018. forms. European Food Safety Authority. EFSA Journal 15:4751.
1543954. EFSA. 2017b. Opinion of the Scientific Panel on Contaminants in the
EFSA. 2004a. Opinion of the Scientific Panel on Contaminants in the Food Chain (CONTAM Panel) on the appropriateness to set a
Food Chain (CONTAM Panel) related to aflatoxin B1 as undesirable group health-based guidance value for T-2 and HT-2 toxin and its
substance in animal feed. European Food Safety Authority. EFSA modified forms 2017. European Food Safety Authority. EFSA
Journal 2:39. Journal 15:4655.
EFSA. 2004b. Opinion of the Scientific Panel on Contaminants in the EFSA. 2017c. Opinion the Scientific Panel on Contaminants in the
Food Chain (CONTAM Panel) on a request from the commission Food Chain (CONTAM Panel) on the risks for animal health related
2786 M. ESKOLA ET AL.

to the presence of zearalenone and its modified forms in feed. GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/

European Food Safety Authority. EFSA Journal 15:4851. ChemicalContaminantsMetalsNaturalToxinsPesticides/default.htm.
EFSA. 2017d. Opinion of the Scientific Panel on Contaminants in the FDA. 1983. Compliance Policy Guide Sec. 570.200 Brazil nuts –
Food Chain (CONTAM Panel) on the risks to human and animal Adulteration with aflatoxin (August 1, 1983). U.S. Food and Drug
health related to the presence of deoxynivalenol and its acetylated Administration. Accessed March 26, 2018. https://www.fda.gov/Food/
and modified forms in food and feed. European Food Safety GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/
Authority. EFSA Journal 15:4718. ChemicalContaminantsMetalsNaturalToxinsPesticides/default.htm.
EFSA. 2017e. Scientific report on human and animal dietary exposure FDA. 2001a. Compliance Policy Guidance Sec. 510.150 Apple juice, apple
to ergot alkaloids. European Food Safety Authority. EFSA Journal juice concentrates, and apple juice products – Adulteration with patu-
15:4902. lin (October 22, 2001). U.S. Food and Drug Administration. Accessed
EFSA. 2018a. Opinion of the Scientific Panel on Contaminants in the March 26, 2018. https://www.fda.gov/Food/GuidanceRegulation/
Food Chain (CONTAM Panel) on the appropriateness to set a GuidanceDocumentsRegulatoryInformation/ChemicalContaminants
group health-based guidance value for fumonisins and their modi- MetalsNaturalToxinsPesticides/default.htm.
fied forms. European Food Safety Authority. EFSA Journal 16:5172. FDA. 2001b. Guidance for Industry. Fumonisin levels in human foods
EFSA. 2018b. EFSA Scientific Colloquium 24 – ‘omics in risk assess- and animal feeds (November 9, 2001). U.S. Food and Drug
ment: State of the art and next steps. EFSA supporting publication. Administration. Accessed March 25, 2018. https://www.fda.gov/Food/
EFSA Journal 2018:EN-1512. GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/
EFSA. 2018c. Scientific statement on the effect on public health of a ChemicalContaminantsMetalsNaturalToxinsPesticides/default.htm.
possible increase of the maximum level for ‘aflatoxin total’ from 4 FDA. 2010. Guidance for Industry. Advisory levels for deoxynivalenol
to 10 mg/kg in peanuts and processed products thereof, intended for (DON) in finished wheat products for human consumption and grains
direct human consumption or use as an ingredient in foodstuffs. and grain by-products used for animal feed (June 29, 2010). U.S. Food
European Food Safety Authority. EFSA Journal 16:5175. and Drug Administration. Accessed March 25, 2018. https://www.fda.
EFSA. 2019. Guidance on harmonised methodologies for human gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatory
health, animal health and ecological risk assessment of combined Information/ChemicalContaminantsMetalsNaturalToxinsPesticides/
exposure to multiple chemicals. EFSA Journal 17 (3):5634, 77. default.htm.
EP and Council. 2002. Directive 2002/32/EC of the European FSANZ. 2017. Australia New Zealand Food Standards Code, 2017.
Parliament and of the Council of 7 May 2002 on undesirable sub- Schedule 19, Maximum levels of contaminants and natural toxicants.
stances in animal feed. European Parliament and Council of the Prepared by Food Standards Australia New Zealand (FSANZ) on 13
European Union. Official Journal of the European Communities L April 2017. Australian Government, Federal Register of Legislation.
140:10–21. Accessed March 25, 2018. https://www.legislation.gov.au/Details/
EP and Council. 2001. Regulation (EC) No 1049/2001 of the European F2017C00333.
Parliament and of the Council of 30 May 2001 regarding public GB. 2017. Maximum levels of mycotoxins in food, China Food Safety
access to European Parliament, Council and Commission docu- National Standard GB 2761-2017. Released March 17, 2017, imple-
ments. European Parliament and Council of the European Union. mented September 17, 2017. In Chinese.
Official Journal of the European Communities L 145:43–8. Grenier, B., and I. P. Oswald. 2011. Mycotoxin co-contamination of
Eskola, M., A. Altieri, and J. Galobart. 2018. Overview of the activities food and feed: Meta-analysis of publications describing toxicological
of the European Food Safety Authority on mycotoxins in food and interactions. World Mycotoxin Journal 4 (3):285–313. doi: 10.3920/
feed. World Mycotoxin Journal 11 (2):277–89. doi: 10.3920/ WMJ2011.1281.
WMJ2017.2270. IARC. 1983. IARC monographs on the evaluation of carcinogenic risk
EUR-Lex. 2018. An official website of European Union law and other of chemicals to humans. Some food additives, feed additives and
public documents of the European Union. Access to European naturally occurring substances. International Agency for Research
Union law. Accessed March 1, 2018. https://eur-lex.europa.eu/home- on Cancer, Volume 31, Lyon, France.
page.html. IARC. 1986. IARC monographs on the evaluation of carcinogenic risks
FAMIC. 2011. Aflatoxin. Food and Agricultural Materials Inspection to humans. Some naturally occurring and synthetic food compo-
Centre (FAMIC). Accessed March 4, 2018. http://www.famic.go.jp/ nents, furocoumarins and ultraviolet radiation. International Agency
ffis/oie/obj/hc_aflatoxin.pdf. for Research on Cancer, Volume 40, Lyon, France.
FAMIC. 2015. Regulation value of pesticides, heavy metals and myco- IARC. 1993. IARC monographs on the evaluation of carcinogenic risks
toxins (administrative guideline) (revised December 18, 2015). Food to humans. Some naturally occurring substances: Food items and
and Agricultural Materials Inspection Centre (FAMIC). Accessed constituents, heterocyclic aromatic amines and mycotoxins.
March 4, 2018. http://www.famic.go.jp/ffis/feed/r_safety/r_feeds_ International Agency for Research on Cancer, Volume 56, Lyon,
safety22.html. France.
FAO. 2015. The State of Agricultural Commodity Markets 2015–2016. IARC. 2002. IARC monographs on the evaluation of carcinogenic risks
Trade and food security: Achieving a better balance between to humans. Some traditional herbal medicines, some mycotoxins,
national priorities and the collective good. Food and Agriculture naphthalene and styrene. International Agency for Research on
Organization of the United Nations, Rome. Accessed April 19, 2018. Cancer, Volume 82, Lyon, France.
http://www.fao.org/publications/soco/the-state-of-agricultural-com- IARC. 2012. IARC monographs on the evaluation of carcinogenic risks
modity-markets-2015-16/en/. to humans. Chemical agents and related occupations. A review of
FDA. 1980a. Compliance Policy Guide Sec. 555.400 Foods – Adulteration human carcinogens. International Agency for Research on Cancer,
with aflatoxin (October 1, 1980). U.S. Food and Drug Administration. Volume 100 F, Lyon, France.
Accessed March 26, 2018. https://www.fda.gov/Food/Guidance IARC. 2015. Mycotoxin control in low- and middle-income countries.
Regulation/GuidanceDocumentsRegulatoryInformation/Chemical IARC Working Group report no. 9, eds. C. P. Wild, J. D. Miller, and
ContaminantsMetalsNaturalToxinsPesticides/default.htm. J. D. Groopman. Lyon, France: International Agency for Research
FDA. 1980b. Compliance Policy Guide Sec. 570.375 Aflatoxin in peanuts on Cancer.
and peanut products (October 1, 1980). U.S. Food and Drug Iqbal, S. Z., S. Malik, M. R. Asi, J. Selamat, and N. Malik. 2018.
Administration. Accessed March 26, 2018. https://www.fda.gov/Food/ Natural occurrence of patulin in different fruits, juices and
GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/ smoothies and evaluation of dietary intake in Punjab, Pakistan. Food
ChemicalContaminantsMetalsNaturalToxinsPesticides/default.htm. Control 84:370–4. doi: 10.1016/j.foodcont.2017.08.024.
FDA. 1982. Compliance Policy Guide Sec. 570.500 Pistachio nuts – Jard, G., T. Liboz, F. Mathieu, A. Guyonvarc’h, and A. Lebrihi. 2011.
Aflatoxin adulteration (October 1, 1982). U.S. Food and Drug Review of mycotoxin reduction in food and feed: From prevention
Administration. Accessed March 25, 2018. https://www.fda.gov/Food/ in the field to detoxification by adsorption or transformation. Food
 CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 2787

Additives & Contaminants: Part A 28:1590–609. doi: 10.1080/ JECFA. 2017. Evaluation of certain contaminants in food. Eighty-third
19440049.2011.595377. report of the Joint FAO/WHO Expert Committee on Food
JECFA. 1996. Toxicological evaluation of certain food additives and Additives (JECFA). Food and Agriculture Organization of the
contaminants. Prepared by the Forty-forth meeting of the Joint United Nations, World Health Organization, WHO Technical
FAO/WHO Expert Committee on Food Additives (JECFA). JECFA Report Series 1002, World Health Organization, Geneva.
Monographs. Food and Agriculture Organization of the United JECFA. 2018. Safety evaluation of certain contaminants in food.
Nations, World Health Organization, IPCS – International Prepared by the eighty-third meeting of the Joint FAO/WHO Expert
Programme on Chemical Safety, WHO Food Additive Series No. 35, Committee on Food Additives (JECFA). WHO Food Additives
World Health Organization, Geneva. Series No. 74, FAO JECFA Monographs 19 bis. World Health
JECFA. 1997. Evaluation of certain food additives and contaminants. Organization, Geneva, Food and Agriculture Organization of the
Forty-sixth report of Joint FAO/WHO Expert Committee on Food United Nations, Rome.
Additives (JECFA). Food and Agriculture Organization of the Joshi, S., R. M. Annida, H. Zuilhof, T. A. van Beek, and M. W. Nielen.
United Nations, World Health Organization, WHO Technical 2016. Analysis of mycotoxins in beer using a portable nanostruc-
Report Series 868, World Health Organization, Geneva. tured imaging surface plasmon resonance biosensor. Journal of
JECFA. 1998. Safety evaluation of certain food additives and contami- Agricultural and Food Chemistry 64 (43):8263–71. doi: 10.1021/acs.
nants. Prepared by the Forty-ninth meeting of the Joint FAO/WHO jafc.6b04106.
Expert Committee on Food Additives (JECFA). JECFA Monographs. Kos, J., J. Mastilovic, E. J. Hajnal, and B.
Saric. 2013. Natural occur-
Food and Agriculture Organization of the United Nations, World rence of aflatoxins in maize harvested in Serbia during 2009–2012.
Health Organization, IPCS – International Programme on Chemical Food Control 34 (1):31–4. doi: 10.1016/j.foodcont.2013.04.004.
Safety, WHO Food Additives Series No. 40, World Health Kovalsky, P., G. Kos, K. N€ahrer, C. Schwab, T. Jenkins, G. Schatzmayr,
Organization, Geneva. M. Sulyok, and R. Krska. 2016. Co-occurrence of regulated, masked
JECFA. 1999. Evaluation of certain food additives and contaminants. and emerging mycotoxins and secondary metabolites in finished
Forty-sixth report of Joint FAO/WHO Expert Committee on Food feed and maize – An extensive survey. Toxins 8 (12):363. doi: 10.
Additives (JECFA). Food and Agriculture Organization of the 3390/toxins8120363.
United Nations, World Health Organization, WHO Technical Krska, R., and C. Crews. 2008. Significance, chemistry and determin-
Report Series 884, World Health Organization, Geneva. ation of ergot alkaloids: A review. Food Additives & Contaminants.
JECFA. 2000. Safety evaluation of certain food additives and contami- Part A, Chemistry, Analysis, Control, Exposure & Risk Assessment 25
nants. Prepared by the Fifty-third meeting of the Joint FAO/WHO (6):722–31. doi: 10.1080/02652030701765756.
Expert Committee on Food Additives (JECFA). JECFA Monographs. Lee, H. J., and D. Ryu. 2017. Worldwide occurrence of mycotoxins in
Food and Agriculture Organization of the United Nations, World cereals and cereal-derived food products: Public health perspectives
of their co-occurrence. Journal of Agricultural and Food Chemistry
Health Organization, IPCS – International Programme on Chemical
65 (33):7034–51. doi: 10.1021/acs.jafc.6b04847.
Safety, WHO Food Additives Series No. 44, World Health
Leslie, J. F., and A. F. Logrieco. 2014. Mycotoxin reduction in grain
Organization, Geneva.
chains. Ames, IA: Wiley Blackwell, John Wiley & Sons, Inc.
JECFA. 2001. Safety evaluation of certain mycotoxins in food. Prepared
Lopez Sanchez, P., M. de Nijs, M. Spanjer, A. Pietri, T. Bertuzzi, A.
by the Fifty-sixth meeting of the Joint FAO/WHO Expert
Starski, J. Postupolski, M. Castellari, and M. Hort os. 2017.
Committee in Food Additives (JECFA). JECFA Monographs. Food
Generation of occurrence data on citrinin in food. European Food
and Agriculture Organization of the United Nations, World Health
Safety Authority. EFSA supporting publication, EN-1177.
Organization, IPCS – International Programme on Chemical Safety,
Magan, N., and D. Aldred. 2007. Post-harvest control strategies:
WHO Food Additives Series No. 47, FAO Food and Nutrition
Minimizing mycotoxins in the food chain. International Journal of
Paper 74. Food Microbiology 119 (1–2):131–9. doi: 10.1016/j.ijfoodmicro.2007.
JECFA. 2002. Evaluation of certain mycotoxins in food. Fifty-sixth
07.034.
report of the Joint FAO/WHO Expert Committee on Food Malachova, A., M. Stranska, M. Vaclavıkova, C. T. Elliott, C. Black, J.
Additives. Food and Agriculture Organization of the United Meneely, J. Haj
slova, C. N. Ezekiel, R. Schuhmacher, and R. Krska.
Nations, World Health Organization, WHO Technical Report Series 2018. Advanced LC-MS-based methods to study the co-occurrence
906, World Health Organization, Geneva. and metabolization of multiple mycotoxins in cereals and cereal-
JECFA. 2007. Evaluation of certain food additives and contaminants. based food. Analytical and Bioanalytical Chemistry 410 (3):801–25.
Sixty-eighth report of the Joint FAO/WHO Expert Committee on doi: 10.1007/s00216-017-0750-7.
Food Additives (JECFA). Food and Agriculture Organization of the Malachova, A., M. Sulyok, E. Beltran, F. Berthiller, and R. Krska. 2014.
United Nations, World Health Organization, WHO Technical Optimization and validation of a quantitative liquid chromatog-
Report Series 947, World Health Organization, Geneva. raphy-tandem mass spectrometric method covering 295 bacterial
JECFA. 2008. Safety evaluation of certain food additives and contami- and fungal metabolites including all regulated mycotoxins in four
nants. Prepared by the Sixty-eighth meeting of the Joint FAO/WHO model food matrices. Journal of Chromatography A 1362:145–56.
Expert Committee on Food Additives (JECFA). JECFA Monographs. doi: 10.1016/j.chroma.2014.08.037.
Food and Agriculture Organization of the United Nations, World Mannon, J., and E. Johnson. 1985. Fungi down on the farm. New
Health Organization, IPCS – International Programme on Chemical Scientist 1445:12–6.
Safety, WHO Food Additives Series No. 59, World Health Marin, S., A. J. Ramos, G. Cano-Sancho, and V. Sanchis. 2013.
Organization, Geneva. Mycotoxins: Occurrence, toxicology, and exposure assessment. Food
JECFA. 2011. Safety evaluation of certain contaminants in food. and Chemical Toxicology 60:218–37. doi: 10.1016/j.fct.2013.07.047.
Prepared by the Seventy-second meeting of the Joint FAO/WHO McCormick, S. P., A. M. Stanley, N. A. Stover, and N. J. Alexander.
Expert Committee in Food Additives (JECFA). JECFA Monographs. 2011. Trichothecenes: From simple to complex mycotoxins. Toxins 3
WHO Food Additives Series No. 63, FAO JECFA Monographs 8. (7):802–14. doi: 10.3390/toxins3070802.
World Health Organization, Geneva, Food and Agriculture Meyer, H., Z. D. Skhosana, M. Motlanthe, W. Louw, and E. Rohwer.
Organization of the United Nations, Rome. 2019. Long term monitoring (2014–2018) of multi-mycotoxins in
JECFA. 2012. Safety evaluation of certain food additives and contami- South African commercial maize and wheat with a locally developed
nants. Prepared by the Seventy-fourth meeting of the Joint FAO/ and validated LC-MS/MS method. Toxins 11 (5):271. doi: 10.3390/
WHO Expert Committee on Food Additives (JECFA). JECFA toxins11050271.
Monographs. Food and Agriculture Organization of the United MHLW. 2018. Ministry of Health, Labour and Welfare, Food Safety
Nations, World Health Organization, WHO Food Additives Series Standards and Evaluation Division, Pharmaceutical Safety and
No. 65, World Health Organization, Geneva. Environmental Health Bureau. Information about imported foods.
2788 M. ESKOLA ET AL.

Safety of imported foods. Accessed March 26, 2018. http://www. Control, Exposure & Risk Assessment 25 (10):1246–56. doi: 10.1080/
mhlw.go.jp/english/policy/health-medical/food/dl/leaflet.pdf. 02652030802036222.
Monbaliu, S., C. Van Poucke, C. Detavernier, F. Dumoulin, M. Van De Shephard, G. S., H. M. Burger, L. Gambacorta, Y. Y. Gong, R. Krska,
Velde, E. Schoeters, S. Van Dyck, O. Averkieva, C. Van Peteghem, J. P. Rheeder, M. Solfrizzo, C. Srey, M. Sulyok, A. Visconti, et al.
and S. De Saeger, 2010. Occurrence of mycotoxins in feed as ana- 2013. Multiple mycotoxin exposure determined by urinary bio-
lyzed by a multimycotoxin LC-MS/MS method. Journal of markers in rural subsistence farmers in the former Transkei, South
Agriculture and Food Chemistry 58 (1):66–71. Africa. Food and Chemical Toxicology 62:217–25. doi: 10.1016/j.fct.
Mornar, A., M. Sertic, and B. Nigovic. 2013. Development of a rapid 2013.08.040.
LC/DAD/FLD/MSn method for the simultaneous determination of Sieger, M., G. Kos, M. Sulyok, M. Godejohann, R. Krska, and B.
monacolins and citrinin in red fermented rice products. Journal of Mizaikoff. 2017. Portable infrared laser spectroscopy for on-site myco-
Agricultural and Food Chemistry 61 (5):1072–80. doi: 10.1021/ toxin analysis. Scientific Reports 7 (1):44028. doi: 10.1038/srep44028.
jf304881g. Simader, A. M., B. Kluger, N. K. N. Neumann, C. Bueschl, M. Lemmens,
Nathanail, A. V., J. Syv€ahuoko, A. Malachova, M. Jestoi, E. Varga, H. G. Lirk, R. Krska, and R. Schuhmacher. 2015. QCScreen: A software
Michlmayr, G. Adam, E. Sievil€ainen, F. Berthiller, and K. Peltonen. tool for data quality control in LC-HRMS based metabolomics. BMC
2015. Simultaneous determination of major type A and B trichothe- Bioinformatics 16 (1):341. doi: 10.1186/s12859-015-0783-x.
cenes, zearalenone and certain modified metabolites in Finnish cer- Stiborova, M., V. M. Arlt, and H. H. Schmeiser. 2016. Balkan endemic
eal grains with a novel liquid chromatography-tandem mass nephropathy: An update on its aetiology. Archives of Toxicology 90
spectrometric method. Analytical and Bioanalytical Chemistry 407 (11):2595–615. doi: 10.1007/s00204-016-1819-3.
(16):4745–55. doi: 10.1007/s00216-015-8676-4. Streit, E., K. Naehrer, I. Rodrigues, and G. Schatzmayr. 2013.
Njumbe Ediage, E., C. Van Poucke, and S. De Saeger. 2015. A multi- Mycotoxin occurrence in feed and feed raw materials worldwide:
analyte LC-MS/MS method for the analysis of 23 mycotoxins in dif- Long-term analysis with special focus on Europe and Asia. Journal
ferent sorghum varieties: The forgotten sample matrix. Food of the Science of Food and Agriculture 93 (12):2892–9. doi: 10.1002/
Chemistry 177:397–404. doi: 10.1016/j.foodchem.2015.01.060. jsfa.6225.
Park, D. L., H. Njapau, and E. Boutrif. 1999. Minimizing risks posed Topi, D., B. Jakovac-Strajn, K. Pav
si
c-Vrta
c, and G. Tav
car-Kalcher.
by mycotoxins utilizing the HACCP concept. FAO Food, Nutrition 2017. Occurrence of ergot alkaloids in wheat from Albania. Food
and Agriculture Journal 23:49–56. Additives & Contaminants. Part A, Chemistry, Analysis, Control,
Paterson, R. R. M., and N. Lima. 2017. Thermophilic fungi to dominate Exposure & Risk Assessment 34 (8):1333–43. doi: 10.1080/19440049.
aflatoxigenic/mycotoxigenic fungi on food under global warming. 2017.1307528.
Uhlig, S., G. Eriksen, I. Hofgaard, R. Krska, E. Beltran, and M. Sulyok.
International Journal of Environmental Research and Public Health
2013. Faces of a changing climate: Semi-quantitative multi-myco-
14 (2):199–210. doi: 10.3390/ijerph14020199.
toxin analysis of grain grown in exceptional climatic conditions in
Peters, J., R. van Dam, R. van Doorn, D. Katerere, F. Berthiller, W.
Norway. Toxins 5 (10):1682–97. doi: 10.3390/toxins5101682.
Haasnoot, and M. W. F. Nielen. 2017. Mycotoxin profiling of 1000
Van der Fels-Klerx, H., J. E. Olesen, M. Madsen, and P. Goedhart.
beer samples with a special focus on craft beer. PLoS One 12 (10):
2012a. Climate change increases deoxynivalenol contamination of
e0185887. doi: 10.1371/journal.pone.0185887.
wheat in North-Western Europe. Food Additives & Contaminants:
Pitt, J. I., and J. D. Miller. 2017. A concise history of mycotoxin
Part A 29 (10):1593–604. doi: 10.1080/19440049.2012.691555.
research. Journal of Agricultural and Food Chemistry 65 (33):
Van der Fels-Klerx, H., J. E. Olesen, L.-J. Naustvoll, Y. Friocourt, M.
7021–33. doi: 10.1021/acs.jafc.6b04494.
Mengelers, and J. Christensen. 2012b. Climate change impacts on natural
Pleadin, J., M. Zadravec, T. Le
sic, N. Vah
cic, J. Frece, M. Mitak, and K.
toxins in food production systems, exemplified by deoxynivalenol in
Markov. 2018. Co-occurrence of ochratoxin a and citrinin in unpro-
wheat and diarrhetic shellfish toxins. Food Additives & Contaminants:
cessed cereals established during a three-year investigation period.
Part A 29 (10):1647–59. doi: 10.1080/19440049.2012.714080.
Food Additives & Contaminants: Part B 11 (1):20–5. doi: 10.1080/ Van der Fels-Klerx, H. J., C. Liu, and P. Battilani. 2016. Modelling cli-
19393210.2017.1389994. mate change impacts on mycotoxin contamination. World
Puel, O., P. Galtier, and I. P. Oswald. 2010. Biosynthesis and toxico- Mycotoxin Journal 9 (5):717–26. doi: 10.3920/WMJ2016.2066.
logical effects of patulin. Toxins 2 (4):613–31. doi: 10.3390/ Vanheule, A., K. Audenaert, M. De Boevre, S. Landschoot, B. Bekaert,
toxins2040613. F. Munaut, M. Eeckhout, M. H€ofte, S. De Saeger, and G. Haesaert.
Rofiat, A. S., F. Fanelli, O. Atanda, M. Sulyok, G. Cozzi, S. Bavaro, R. 2014. The compositional mosaic of Fusarium species and their
Krska, A. F. Logrieco, and C. N. Ezekiel. 2015. Fungal and bacterial mycotoxins in unprocessed cereals, food and feed products in
metabolites associated with natural contamination of locally processed Belgium. International Journal of Food Microbiology 181:28–36. doi:
rice (Oryza sativa L.) in Nigeria. Food Additives & Contaminants: Part 10.1016/j.ijfoodmicro.2014.04.012.
A 32:950–9. doi: 10.1080/19440049.2015.1027880. Varga, E., T. Glauner, F. Berthiller, R. Krska, R. Schuhmacher, and M.
SCF. 1999. Opinion of the Scientific Committee on Food on Fusarium Sulyok. 2013. Development and validation of a (semi-)quantitative
toxins. Part 1: Deoxynivalenol (DON). Scientific Committee on UHPLC-MS/MS method for the determination of 191 mycotoxins
Food. Accessed February 15, 2018. https://ec.europa.eu/food/sites/ and other fungal metabolites in almonds, hazelnuts, peanuts and
food/files/safety/docs/cs_contaminants_catalogue_out44_en.pdf. pistachios. Analytical and Bioanalytical Chemistry 405 (15):
SCF. 2000a. Opinion of the Scientific Committee on Food on Fusarium 5087–104. doi: 10.1007/s00216-013-6831-3.
toxins. Part 2: Zearalenone (ZEA). Scientific Committee on Food. Warth, B., A. Parich, J. Atehnkeng, R. Bandyopadhyay, R.
Accessed February 15, 2018. https://ec.europa.eu/food/sites/food/ Schuhmacher, M. Sulyok, and R. Krska. 2012. Quantitation of myco-
files/safety/docs/cs_contaminants_catalogue_out65_en.pdf. toxins in food and feed from Burkina Faso and Mozambique using
SCF. 2000b. Minute statement on patulin. Expressed by the Scientific a modern LC-MS/MS multitoxin method. Journal of Agricultural
Committee on Food during the plenary meeting on 8 March 2000. and Food Chemistry 60 (36):9352–63. doi: 10.1021/jf302003n.
Scientific Committee on Food. Accessed March 5, 2018. https://ec. Warth, B., A. Parich, C. Bueschl, D. Schoefbeck, N. K. N. Neumann, B.
europa.eu/food/sites/food/files/safety/docs/cs_contaminants_cata- Kluger, K. Schuster, R. Krska, G. Adam, M. Lemmens, and R.
logue_patulin_out55_en.pdf. Schuhmacher. 2015. GC-MS based targeted metabolic profiling iden-
SCOOP. 2002. Assessment of dietary intake of patulin by the popula- tifies changes in the wheat metabolome following deoxynivalenol
tion of EU Member States, Reports on tasks for scientific cooper- treatment. Metabolomics 11 (3):722–38. doi: 10.1007/s11306-014-
ation, Report of experts participating in Task 3.2.8, March 2002. 0731-1.
Directorate-General Health and Consumer Protection. WHO. 1990. Selected mycotoxins: Ochratoxins, trichothecenes, ergot.
Shephard, G. 2008. Risk assessment of aflatoxins in food in Africa. Environmental Health Criteria Monographs. United Nations
Food Additives & Contaminants. Part A, Chemistry, Analysis, Environment Programme, International Labour Organisation, World
 CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 2789

Health Organization. International Programme on Chemical Safety, Database, GEMS/Food Contaminants database and the WHO
Environmental Health Criteria 105. Collaborating Centres Database. Accessed 13 March 2018. https://
WHO. 2000. Fumonisin B1. Environmental Health Criteria Monographs. extranet.who.int/sree/Reports?op=vs_html&path=/WHO_HQ_Reports/
United Nations Environment Programme, International Labour G7/PROD/EXT/chemical_overview&userid=G7_ro&password=
Organisation, World Health Organization. International Programme inetsoft123.
on Chemical Safety. Environmental Health Criteria 219. Wu, F., and N. J. Mitchell. 2016. How climate change and regulations
WHO. 2018. FOSCOLLAB Database. Chemical Overview Dashboard can affect the economics of mycotoxins. World Mycotoxin Journal 9
– Integrated summary elements from JECFA Evaluations (5):653–63. doi: 10.3920/WMJ2015.2015.