Environmental Pollution 234 (2018) 667e676

Contents lists available at ScienceDirect

Environmental Pollution
journal homepage: www.elsevier.com/locate/envpol

Public health risk of trace metals in fresh chicken meat products on

the food markets of a major production region in southern China*
Yuanan Hu a, Wenfeng Zhang b, Gang Chen c, Hefa Cheng d, *, Shu Tao d
a
MOE Laboratory of Groundwater Circulation and Evolution, School of Water Resources and Environment, China University of Geosciences (Beijing), Beijing
100083, China
b
State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China
c
Department of Civil & Environmental Engineering, Florida A&M University-Florida State University, Tallahassee, FL 32310, United States
d
MOE Key Laboratory for Earth Surface Processes, College of Urban and Environmental Sciences, Peking University, Beijing 100871, China

a r t i c l e i n f o a b s t r a c t

Article history: Because most chickens are reared in intensive farms, where a range of feed additives are used routinely,
Received 18 August 2017 concerns have been raised on the potential public health risk of chicken product consumption. This study
Received in revised form was conducted to characterize the contents of trace metals in fresh chicken tissues (354 samples) on the
30 November 2017
food markets in Guangdong province of southern China, a major region of chicken production with heavy
Accepted 3 December 2017
per capita chicken consumption, and to assess the public health risk from chronic dietary exposure to the
trace metals through chicken consumption. With the exception of Cr, Ni, and Pb, the contents of trace
metals were generally higher in the chicken giblets (livers, gizzards, hearts, and kidneys) compared to
Keywords:
Trace metal
muscles (breasts and drumsticks). Chicken tissues from the urban markets generally contained higher
Inorganic arsenic levels of As, Cu, Mn, and Zn than those from the rural markets, while the contents of Pb were typically
Chicken tissue higher in the chicken muscles from the rural markets. Results of statistical analyses indicate that Cu, Zn,
Chronic dietary exposure and As in the chicken tissues derived mainly from the feeds, which is consistent with the widespread use
Cancer risk of Cu, Zn, and phenylarsenic compounds as feed supplements/additives in intensive poultry farming. No
Phenylarsenic feed additive non-carcinogenic risk is found with the consumption of fresh chicken meat products on the food mar-
kets, while approximately 70% of the adult population in Guangzhou and 30% of those in Lianzhou have
bladder and lung cancer risk above the serious or priority level (10
4), which arises from the inorganic
arsenic contained in the chicken tissues. These findings indicate that the occurrence of inorganic arsenic
at elevated levels in chicken tissues on the food markets in Guangdong province poses a significant
public health risk, thus the use of phenylarsenic feed additives in China's poultry farming should be
significantly reduced and eventually phased out.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction achieved primarily through the transition from family backyard

According to the Food and Agriculture Organization, the global
consumption of poultry meat per capita has increased by 76.6%
between 1990 and 2009 (Henchion et al., 2014). Chicken meat is
relatively inexpensive because chickens can be produced faster and
with higher feed efficiency than other commercial meats. Poultry
meat has been the second most consumed type of meat products
(>20%) in China and its production has increased by 10.2 times
between 1996 and 2015 (Hu et al., 2017a). Such a rapid increase was
*
This paper has been recommended for acceptance by Charles Wong.
* Corresponding author.
E-mail address: hefac@umich.edu (H. Cheng).
and small farms to intensive farms (Hu et al., 2017a). Approximately
80% of the chickens in China were raised in intensive poultry farms
(with annual production capacities of greater than 50 thousand
heads) by 2012 (Liu et al., 2013b).
Besides a range of environmental pollution and public health
problems associated with livestock and poultry production in
intensive animal farms, food safety concerns have also been raised
regarding the potential residues (e.g., antimicrobials and trace
metals) of veterinary drugs and feed supplements/additives in
animal food products, including meats, eggs, and milk (Hu and
Cheng, 2016; Hu et al., 2017a). The use of formulated animal
feeds and feed additives in intensive animal farming has turned out
to be a double-edged sword: they improve the production

https://doi.org/10.1016/j.envpol.2017.12.006
0269-7491/© 2017 Elsevier Ltd. All rights reserved.
668 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676
efficiency of animal food products (Poulsen, 1998; Visek, 1978),
while the occurrence of their residues at levels above the safety
thresholds in foods of animal origin can pose health risks to con-
sumers (Kan and Meijer, 2007; Nachman et al., 2013, 2017;
Silbergeld and Nachman, 2008). Besides the health risks from
foodborne residues, the overuse and misuse of antimicrobials in
animal farms also cause selection and spread of antimicrobial
resistance (AMR), which poses a significant public health risk
worldwide (Marshall and Levy, 2011; Mole, 2013; Singer et al.,
2016). Although heavy metals are not the main cause for the pro-
liferation of AMR in animal farms, some of them could drive co-
selection of AMR (Baker-Austin et al., 2006; Seiler and
Berendonk, 2012; Singer et al., 2016), which is beyond the scope
of this study.
Various trace metals, such as copper and zine, and nickel and
selenium to a lesser extent, are supplemented extensively in
commercial feeds for the purpose of boosting animal health and
performance (Richards et al., 2010). Zinc, copper, and manganese
are essential trace elements, and play multiple biological and
physiological roles in the development and health of all animals
(Richards et al., 2010; Underwood and Suttle, 1999). Various min-
erals, primarily those of iron, copper, zinc, manganese, and cobalt
compounds, are allowed as feed ingredients for food-producing
animals (AAFCO, 2017; MOA, 2009; 2013), even in organic live-
stock production (USDA, 2017), and are commonly found at
elevated levels in animal feeds (e.g., Granados-Chinchilla et al.,
2015; Nicholson et al., 1999; Shurson et al., 2011). In addition,
phenylarsenic compounds, primarily roxarsone and p-arsanilic
acid, used to be widely used as feed additives in intensive swine
and poultry farming for the purposes of enhancing feed conversion
efficiency, improving animal weight gain, and controlling intestinal
parasites (Liu et al., 2013a; Mangalgiri et al., 2015).
While most of the trace metals ingested by animals are excreted
via urine and feces, small fractions (e.g., approximately 10e20% for
Cu and Zn) are absorbed by the bodies of animals to meet their
nutritional requirements (Carlson et al., 2004; Veum et al., 2004).
The absorption rates of trace metals are influenced by a range of
factors, including species and chemical forms of the metal, its re-
actions with other metals and other dietary components in the
feed, its supplementation dose, and the growth stages of the ani-
mals (Ashmead and Graff, 1985; Creech et al., 2004; Vieira, 2008).
As a potential public health concern, residues of trace metals in
chicken meat products have been studied in various countries and
regions (Demirbaş, 1999; Iwegbue et al., 2008; Liu et al., 2016;
Oforka et al., 2012; Sinkakarimi et al., 2017; Tahvonen and
Kumpulainen, 1994; Uluozlu et al., 2009). In particular, the use of
phenylarsenic feed additives could result in elevated residues of
arsenic in chicken tissues. Results of a 35-day feeding experiment
showed that the breast of chickens fed with roxarsone had higher
concentrations of arsenic species, with <10% of the total arsenic
occurred in inorganic form, compared to those in the breast of
controls fed with a roxarsone-free diet, even after a 7-day with-
drawal period (Liu et al., 2016). A market basket study in the U.S.
conducted between 2010 and 2011 showed that the meat of
chickens produced by conventional farms without policies pro-
hibiting the use of roxarsone had higher contents of inorganic
arsenic (i-As) compared to that of chickens raised by antibiotic-free
producers, which had mixed policies on roxarsone use, and the
organic chicken meat (produced without using roxarsone)
(Nachman et al., 2013). As a result of the additional i-As ingested,
lifetime consumption of conventional chicken would bring 37
additional cases of bladder and lung cancer per 1 million persons
compared to organic chicken (Nachman et al., 2013). Similarly, due
to the significantly elevated residues of i-As, lifetime consumption
of the turkey meat produced by conventional farms not prohibiting
the use of nitarsone, another important phenylarsenic feed additive
that was allowed in the U.S. until 2014, would result in 3.1 addi-
tional bladder and lung cancer cases per 1 million persons
compared to the meat of turkey raised without using nitarsone
(Nachman et al., 2017). Although already phased out in many
developed countries, including the E.U. member states and the U.S.
(EC, 1999; FDA, 2009), the use of phenylarsenic feed additives is still
allowed in many countries, including China.
Exposures (through ingestion, inhalation, and injection) to non-
essential trace metals, such as As, Cd, and Pb, can cause carcino-
genic and non-carcinogenic adverse effects to human beings, even
at low concentrations, while the essential ones, including Co, Cr, Cu,
Mn, Se, and Zn, can also be toxic when occurring at concentrations
above the safety thresholds in food products (CAC, 2016; NHFPC
and SFDA, 2017). Concerns have been raised on the potential
accumulation of trace metals in chicken meat products due to the
use of formulated animal feeds and feed additives in intensive
farming. Despite of few investigations on exposure to trace metals
based on the total dietary study (TDS) approach (Wu et al., 2016;
Zheng et al., 2007), assessment of the public health risk of dietary
exposure to trace metals in chicken meat products is lacking in
China. This study was conducted to evaluate systematically the
contents of trace metals in fresh chicken meat products on the food
markets in Guangdong province, a major chicken production and
consumption region in southern China, and assess the potential
health risk associated with dietary intake of chicken meat products
for the local populations.
2. Materials and methods
2.1. Study area and sample collection
Guangdong province, which is one of the most economically
developed regions in southern China, ranks among the top in both
chicken production and per capita consumption of chicken meat
products in the country (NBS, 2017). As the largest city in Guang-
dong, Guangzhou has a total official population of approximately 14
million and a gross domestic product (GDP) per capita of 1.39  105
Yuan RMB in 2016 (Statistical Office of Guangzhou, 2017). Lianzhou,
which is located in the northern part of Guangdong and is domi-
nated by rural areas, has a total official population of approximately
0.55 million and a GDP per capita of 2.52  104 Yuan RMB in 2016
(Statistical Office of Lianzhou, 2017). Consumers in Guangdong are
well known for their love for chicken, and they prefer live poultry
sold in wet markets over frozen meat in grocery stores as the
former is much more flavorful. A total of 9 wet markets in
Guangzhou and 5 wet markets in Lianzhou, which were selected to
represent the developed urban centers and the developing rural
areas of Guangdong province, respectively, were visited multiple
times over the period of January 2013 through October 2014.
Multiple samples of whole chickens (slaughtered on site by the
vendors) and/or pieces of freshly slaughtered chickens, which were
generally representative of those sold in the markets, were pur-
chased during each visit. Although no specific labels existed for
chickens sold in wet markets, information on their breeds and lo-
cations of production (when known) was collected through inter-
viewing the vendors. A total of 354 chicken tissue samples, which
could well represent the chicken meat products consumed by local
populations, were obtained. According to the poultry vendors,
chickens sold in Guangzhou markets were mostly supplied by
intensive farms, while a large portion of those on Lianzhou markets
still came from family backyards or small farms. Details on sample
collection and transfer have been described in our previous work
(Hu et al., 2017b). Once received in the laboratory, all chicken
samples were carefully rinsed with distilled water and sliced into
 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676
small pieces. They were then freeze-dried, with the moisture
contents determined based on the difference between the fresh and
dried samples. The dried samples were powderized and stored
at
20  C before being digested for analysis of trace metal contents.
To survey the contents of heavy metals in chicken feeds, a total of 61
commercial feed samples, covering the major types and brands
used in intensive poultry farms in the province (Hu et al., 2017b),
were also included in this study.
2.2. Sample preparation and chemical analysis
The dry chicken tissue and feed samples (0.20 g) were pre-
digested in open vessels (with TFM liners) by a mixture of HNO3
(5.0 mL)/H2O2 (1.0 mL)/water (1.0 mL) for 6 h at room temperature.
They were then digested in sealed vessels using a MARS microwave
digestion system (CEM, U.S.). The mixtures were heated sequen-
tially to 120  C in 10 min (held for 5 min), to 180  C in 6 min (held
for 6 min), and then to 190  C in 5 min (held for 10 min). After
cooling, the digestion solutions were transferred quantitatively to
PTFE tubes and heated on an electric heating plate to evaporate the
acid. The residual solutions (~0.5 mL) were diluted to 10 mL by 2%
HNO3, and then filtered by 0.22 mm membrane filters before
instrumental analysis. All reagents used were of trace pure grade
and were checked for possible contamination by heavy metals,
while triple-distilled water was used in the preparation of all so-
lutions. For quality assurance and quality control (QA/QC) purposes,
procedural blanks, certified reference materials (CRMs), and du-
plicates of the samples (10% of the load) were processed in each
batch of samples digested. Three CRMs, including a lobster hepa-
topancreas (TORT-2, National Research Council of Canada), a pig
liver (GBW10051, Chinese Academy of Geological Sciences), and a
chicken muscle (GBW10018, Chinese Academy of Geological Sci-
ences), were used in this study. Contents of trace metals in the
diluted solutions were measured on a 7700X inductively coupled
plasma mass spectrometer (ICP-MS, Agilent, U.S.). The recovery
rates for the targeted metal species were within the range of
107.4 ± 12.9% based on comparison with the certified values of the
CRMs.
T-tests were applied to identify whether the contents of trace
metals in the chicken tissues from Guangzhou and Lianzhou had
significant difference. Due to the wide use of feed additives in
intensive poultry farming, chicken meat products on Guangzhou
markets, which came mostly from intensive farms, were expected
to contain heavy metals at higher levels compared to those from
the rural markets in Lianzhou, where a large portion of the chickens
were still raised in backyard and “free-range” farms. Cluster anal-
ysis was carried out to gain insight on the probable sources of trace
metals in the chicken feeds based on their occurrence patterns
(Supplementary Data).
2.3. Health risk assessment
The non-carcinogenic risk is typically characterized based on the
hazard quotient (HQ), which is defined as the ratio of the daily
intake dose of a specific metal to its reference dose (RfD) (USEPA,
1993). Meanwhile, the carcinogenic risk is generally represented
by the incremental lifetime cancer risk (ILCR), which is the incre-
mental probability of an individual developing cancer over a life-
time period from the exposure to a given carcinogenic metal
(USEPA, 2005). It is widely estimated with the cancer slope factor
(CSF), which quantitatively defines the relationship between the
intake dose of a carcinogenic chemical and the adverse effects
(USEPA, 2005).
To estimate the level of human exposure to a trace metal, the
average daily dose (ADD, mg/kg/day) is calculated as:
669
Cmetal  IR  DT
ADD ¼ (1)
BW  AT
where Cmetal is the content of the trace metal in chicken tissue (mg/
kg, wet weight), IR is the ingestion rate of chicken tissue per capita
(kg/day, wet weight), DT is the duration of exposure (day), AT is the
time period over which the dose is averaged (day) and is generally
treated to be equal to DT for non-carcinogenic effects, and BW is the
average body weight (kg) of the exposed population. Based on the
expenditures on poultry products of urban households and the
average consumption rate of the rural population in Guangdong
province, the average ingestion rates of chicken muscles were 48.8
and 25.2 g/person/day, and those of chicken giblets were 0.56 and
0.29 g/person/day for the adults in Guangzhou and Lianzhou,
respectively (Supplementary Table S1). According to the results of a
national survey (MEP, 2013), the mean body weights were 59.5 and
57.2 kg for the adults in Guangzhou and Lianzhou, respectively
(Supplementary Tables S2-S3). The fraction of i-As in the total As
contents of the chicken tissues was set to be 30%, based on the
findings of our speciation analysis and literature results (Hu et al.,
2017b). The lifetime average daily dose (LADD, mg/kg/day) is
calculated similarly.
The non-carcinogenic health risk of a trace metal from oral
exposure is estimated as:
ADD
HQ ¼ (2)
RfD
where HQ is the hazard quotient, RfD is the metal's oral reference
dose, which represents the maximum permissible daily exposure
(mg/kg/day). The following RfD values (mg/kg/day) were adopted:
3.0  10
4 for i-As, 1.0  10
3 for Cd, 3.0  10
4 for Co, 1.5 for Cr,
4  10
2 for Cu, 1.4  10
1 for Mn, 2.0  10
2 for Ni, 3.5  10
3 for
Pb, 5.0  10
3 for Se, and 3  10
1 for Zn, respectively
(Supplementary Table S4). It is worth noting that Cr can exist in two
major oxidation states: Cr(VI) and Cr(III). Cr(VI) has potentially
toxic and carcinogenic effects only by inhalation, while Cr(III), at
low doses, is essential for human health. Cr(VI) ingested would be
rapidly reduced to Cr(III) by gastric juices in human bodies (ATSDR,
2000; IRIS, 1998a,c). Therefore, the RfD of Cr(III) was used for
establishing the HQ of total Cr in this study. To evaluate the po-
tential health risk of exposure to multiple trace metals, the hazard
index (HI) is calculated as the sum of HQs of the major metal species
evaluated, assuming that they have similar working mechanisms
and linearly affect the target organ (OERR, 1989; Stara et al., 1986).
In general, the exposed population is considered safe when HI is
less than 1.
The cancer risk of carcinogenic metal species is estimated as:
ILCR ¼ LADD  CSF (3)
where CSF is the cancer slope factor for the metal ((mg/kg/day)
1).
In general, LADD is set to a lifetime of 70 years for carcinogenic
effect. The cancer risk was only estimated for the two trace metals
(As and Pb) that have carcinogenic effects through oral exposure.
The Integrated Risk Information System (IRIS) of the U.S. Environ-
mental Protection Agency (USEPA) suggested a CSF value of 1.5 (mg/
kg/day)
1 for i-As, which corresponds to skin cancer (IRIS, 1988). In
2010, the USEPA proposed a CSF value of 25.7 (mg/kg/day)
1 for
bladder and lung cancer based on epidemiologic literature (USEPA,
2010). Exposure to lead and lead compounds could contribute to
stomach and lung cancer (IRIS, 1998b), and a CSF of 0.0085 (mg/kg/
day)
1 has been proposed (OEHHA, 2009). According to the USEPA,
cancer risk is considered “acceptable” when the ILCR is below
1.0  10
6, while it becomes “serious” or “high priority” when the
670 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676

value exceeds 10
4 (USPEA, 2000). sampling of the values from these probability distributions, was
To account for the uncertainty of the impact factors used in performed for 10,000 runs, thus the results obtained could well
health risk assessment, including contents of trace metals, inges- account for the uncertainty of the estimated HQ and ILCR.
tion rate of chicken meat, body weight, and human variability in
genetic polymorphism, Monte Carlo simulations were performed to
3. Results and discussion
estimate the probability distributions of HI and ILCR using Matlab
software. The measured contents of heavy metals in chicken tissues
3.1. Contents of trace metals in chicken tissues
apparently observed log-normal distributions, thus their geometric
means (GMs) and corresponding 95% confidence intervals (CIs)
Table 1 summarizes the geometric means and the correspond-
were used in the Monte Carlo simulations. Because arsenic is highly
ing confidence limits at 95% level for the contents of major trace
toxic and carcinogenic when present in the form of i-As, health risk
metals (As, Cd, Co, Cr, Cu, Mn, Ni, Pb, Se, and Zn) in the chicken
was assessed for i-As instead of total As. Based on the results of our
tissue samples collected from the food markets in Guangzhou and
speciation analysis and the literature findings, a normal distribu-
Lianzhou. The large variability in the contents of trace metals
tion (mean: 30%; standard deviation: 10%) was assumed for the
among the chicken tissues could be caused by their varying metal
fraction of i-As in the measured total As contents (Hu et al., 2017b).
absorption and accumulation capabilities, and the difference in the
According to the findings of a recent diet study in Guangdong,
feeds, drinking water, and farming environment of the chickens as
normal distributions with standard deviations of 70% of the means
well. In general, Zn, Cu, and Mn were the most abundant ones
were assumed for the chicken consumption rates of adults in
among the trace metals investigated, and their contents were
Guangzhou and Lianzhou (Huang et al., 2013). Based on the data
approximately 2e4 orders of magnitude higher than those of As,
from a national survey (MEP, 2013), the body weights of adults in
Cd, and Pb. The contents of all trace metals measured in this study,
Guangzhou and Lianzhou were assumed to observe log-normal
with the exception of As and Pb, were below the maximum limits
distributions, with standard deviations (for log-transformed body
allowed in the national food safety standards (MOH, 2012). The
weights) of 0.15 and 0.16, respectively. Following the approach used
highest levels of Pb occurred in chicken muscle samples (breasts
in a recent study on estimating the global lung cancer risk from
and drumsticks), particularly in those from the rural food markets:
pollutant exposure (Shen et al., 2014), the CSF values were adjusted
11 of 47 chicken muscle samples from Lianzhou contained Pb at
by a factor of 0.86 for the Guangdong population, while the vari-
levels above the safety threshold of 0.2 mg/kg (fresh weight), while
ability in genetic polymorphisms was accounted for by an adjust-
only 2 of 63 muscle samples from Guangzhou had Pb contents
ment factor, which follows log-normal distribution (mean: 1.0;
exceeding this limit. High levels of As were generally observed in
deviation: 0.65). Monte Carlo simulation, which is based on random
the liver samples, with 10% of those from both Guangzhou and

Table 1
Summary of the contents of major trace metals (mg/kg, wet weight) in the chicken tissue samples collected from the food markets in Guangzhou and Lianzhou of Guangdong
province.

Metal Breast Drumstick Gizzard Heart Kidney Liver

Guangzhou Lianzhou Guangzhou Lianzhou Guangzhou Lianzhou Guangzhou Lianzhou Guangzhou Lianzhou Guangzhou Lianzhou
(n¼33) (n¼32) (n¼30) (n¼15) (n¼21) (n¼21) (n¼41) (n¼19) (n¼7) (n¼11) (n¼91) (n¼33)

As GMa 0.026 0.018 0.029 0.012 0.038 0.045 0.031 0.034 0.018 0.035 0.049 0.023
LCLMb 0.021 0.013 0.023 0.009 0.029 0.031 0.025 0.024 0.013 0.018 0.036 0.014
UCLMc 0.032 0.024 0.036 0.016 0.050 0.065 0.040 0.049 0.025 0.065 0.066 0.037
Cd GM 0.002 0.002 0.001 0.001 0.007 0.010 0.002 0.002 0.004 0.004 0.015 0.019
LCLM 0.001 0.002 0.001 0.001 0.005 0.008 0.002 0.001 0.002 0.003 0.013 0.016
UCLM 0.002 0.003 0.002 0.002 0.010 0.013 0.003 0.002 0.006 0.007 0.019 0.023
Co GM 0.007 0.008 0.009 0.005 0.012 0.015 0.018 0.019 0.010 0.013 0.030 0.030
LCLM 0.006 0.006 0.008 0.004 0.010 0.012 0.015 0.017 0.008 0.010 0.026 0.027
UCLM 0.008 0.011 0.010 0.007 0.015 0.017 0.022 0.022 0.012 0.017 0.035 0.034
Cr GM 0.093 0.093 0.111 0.061 0.124 0.110 0.117 0.158 0.067 0.152 0.086 0.092
LCLM 0.081 0.068 0.095 0.048 0.095 0.093 0.095 0.117 0.052 0.080 0.077 0.072
UCLM 0.107 0.127 0.130 0.077 0.161 0.129 0.145 0.213 0.086 0.287 0.097 0.116
Cu GM 0.445 0.362 0.757 0.378 1.028 0.927 3.19 3.20 0.858 0.781 3.56 2.84
LCLM 0.407 0.279 0.695 0.307 0.912 0.830 2.80 2.94 0.805 0.732 2.97 2.65
UCLM 0.486 0.471 0.825 0.465 1.16 1.04 3.63 3.48 0.916 0.833 4.27 3.05
Mn GM 0.429 0.504 0.533 0.237 0.781 0.897 1.03 1.103 0.784 0.877 3.05 2.37
LCLM 0.344 0.301 0.452 0.189 0.678 0.737 0.934 0.865 0.671 0.722 2.62 2.16
UCLM 0.534 0.845 0.630 0.297 0.901 1.093 1.130 1.406 0.917 1.07 3.56 2.59
Ni GM 0.051 0.055 0.069 0.036 0.066 0.063 0.062 0.056 0.046 0.076 0.059 0.051
LCLM 0.043 0.042 0.054 0.029 0.055 0.046 0.049 0.048 0.035 0.044 0.049 0.043
UCLM 0.060 0.071 0.088 0.045 0.081 0.086 0.078 0.066 0.059 0.133 0.071 0.060
Pb GM 0.071 0.116 0.073 0.055 0.059 0.083 0.055 0.075 0.042 0.077 0.068 0.093
LCLM 0.060 0.080 0.059 0.037 0.045 0.065 0.042 0.057 0.027 0.062 0.054 0.075
UCLM 0.084 0.169 0.090 0.082 0.078 0.106 0.070 0.098 0.064 0.096 0.087 0.115
Se GM 0.133 0.164 0.169 0.200 0.308 0.355 0.424 0.447 0.604 0.608 0.476 0.546
LCLM 0.121 0.141 0.156 0.184 0.279 0.330 0.390 0.415 0.568 0.520 0.392 0.488
UCLM 0.147 0.191 0.183 0.218 0.341 0.381 0.461 0.482 0.642 0.711 0.579 0.611
Zn GM 5.25 4.46 16.0 10.1 25.3 19.3 20.6 16.8 19.2 11.8 26.7 22.2
LCLM 4.66 3.27 14.3 8.64 22.4 18.0 18.8 14.8 17.7 9.75 22.8 19.3
UCLM 5.93 6.10 17.9 11.7 28.6 20.7 22.6 19.1 20.8 14.2 313 25.4
a
Geometric mean.
b
Lower confidence limit for the mean at 95% confidence level.
c
Upper confidence limit for the mean at 95% confidence level.
 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676 671

Lianzhou had total As contents above the limit (0.5 mg/kg, fresh
weight) allowed for meat products. Table 2 compares the contents
of total As in the chicken tissues on the food markets in Guangdong
with those reported in other countries. For the chicken tissues from
Guangdong province, the mean contents of total As decreased in
the order of livers (0.55 mg/kg) > gizzards (0.22 mg/kg) > hearts
(0.18 mg/kg) > kidneys (0.15 mg/kg) > breasts (0.11 mg/
kg) > drumsticks (0.10 mg/kg), while the difference among giz-
zards, hearts, kidneys, breasts, and drumsticks was not statistically
significant (Supplementary Table S5). In general, the contents of As
in chicken tissues, including livers, on Guangdong markets were
within the ranges found in the poultry products in other regions of
the world before the phase out of phenylarsenic feed additives
(Lasky et al., 2004; Kawalek et al., 2011; Morrison, 1969; Uluozlu
et al., 2009).
In this study, a total of 41 pairs of liver and muscle samples from
the same chickens were collected. The liver-to-muscle ratios for
trace metals depicted on Fig. 1 indicate that livers had the highest
overall accumulation potential for Cd, with a median ratio of 8.59,
followed by Mn (median: 6.18), Cu (median: 6.09), Co (median:
4.39), Se (median: 3.98), Zn (median: 3.16), and As (median: 1.1). In
contrast, chicken livers showed lower tendency of accumulating Cr
(median: 0.85), Ni (median: 0.91), and Pb (median: 0.91) compared
to muscles. It has been demonstrated that Cd and As were not easily
metabolized and would be accumulated in the livers of layers and
broilers, while Cr and Ni exhibited slightly higher concentrations in
muscles than livers (Okoye et al., 2015). Investigation on metal
contents in the tissues of “free-range” chickens raised near a lead-
zinc mine in Kabwe, Zambia revealed greater accumulation of Pb,
Cd, and Zn in livers relative to muscles (Yabe et al., 2013). It has also
been reported that the liver-to-muscle ratio of As in chickens
administered with roxarsone decreased from 11 to 4.2 and 2.9 after
3 and 5 days of drug withdrawal (Lasky et al., 2004). The overall low
liver-to-muscle ratios of As observed in the present study resulted
partially from the samples collected from Lianzhou, where the
chickens mainly came from family backyard farms and small farms
with much less common use of phenylarsenic feed additives
compared to intensive poultry farms. Meanwhile, the occurrence of
multiple cases of extremely high ratios for As (e.g., >10) could be
attributed to the failure of chicken producers on following the
required withdrawal period for phenylarsenic feed additives.
Results of t-tests indicate that the chicken tissue samples
collected from Guangzhou markets generally had higher contents
of Cu and Zn than those from Lianzhou markets, and the difference
was statistically significant in the drumsticks, kidneys, and livers
(Supplementary Table S6). As, Co, Cr, Mn, and Ni also occurred at
significantly higher levels in some tissues from Guangzhou mar-
kets, while the occurrence patterns of Pb, Se, and Cd did not exhibit
such a trend. The contents of Pb were higher in all types of chicken
tissues, with the exception of drumsticks, from the food markets in
Lianzhou than those from Guangzhou markets. Similarly, higher Se
levels were mainly found in the chicken breasts, drumsticks, and
gizzards from the food markets in Lianzhou. No significant differ-
ence in the contents of Cd was detected between Guangzhou and
Lianzhou for all types of chicken tissues. The above distribution
patterns of trace metals in the chicken tissues from the food mar-
kets of Guangzhou and Lianzhou likely resulted from the difference
in chicken feeds and the farming environment. Chickens produced
from intensive farms dominated the tissue samples from the food
markets in Guangzhou, while small farms and family backyard
farms still contributed largely to those collected from Lianzhou
markets. Cu and Zn are widely supplemented to the commercial
feeds used in intensive poultry farming. In contrast, Cd and Pb are
well known toxic elements and are unlikely to be introduced
intentionally into the chicken feeds. Compared to those confined in
intensive poultry farms, chickens raised in backyard and “free-
range” farms have plenty of access to soils, plants, and insects. As a
result, trace metals present in the soils can be bio-transferred to the
chickens through the food chain (Zhuang et al., 2009). In addition,
chickens raised in backyard and “free-range” farms peck in dirt, and
eat grit to help grind up food in their gizzards, thus they can also
absorb some of the metals present in the ingested soil and grit
particles directly.
Additional statistical analyses were conducted to investigate the
Fig. 1. Box and whisker plots for the ratios of metal contents in the liver and muscle
tissues of the same chickens (n ¼ 41). The horizontal line indicates the reference ratio
of 1.

Table 2
Comparison of the contents of total As (mg/kg, wet weight) in various types of fresh chicken tissues in Guangdong with those in other countries reported in the literature.

Chicken tissue This study Morrisonc Lasky et al.d Uluozlu et al.e Kawalek et al.f
a b
n GM AM AM AM AM AM (0-day) AM (5-day)

Liver 124 41.7 143 390 390 60 2836 1062

Gizzard 42 41.1 57.2 e e 100 e e
Heart 60 32.2 46.8 e e 60 e e
Kidney 18 26.8 39.0 130 e 90 e e
Drumstick 46 21.5 28.6 <100i e 70i 71g 28g
Breast 64 21.4 26.0
a
Geometric mean.
b
Arithmetic mean.
c
Roxarsone-treated chickens after a 5-day withdrawal period in Arkansas, U.S. (Morrison, 1969).
d
Whether the chicken samples in the study had been treated with roxarsone or not could not be definitively determined (Lasky et al., 2004). The arsenic contents of livers
from young chickens in the U.S. between 1994 and 2000 ranged from 0.33 to 0.43 mg/kg, with a mean of 0.39 mg/kg.
e
Chickens and chicken meat products from Tokat City, Turkey in 2006 (Uluozlu et al., 2009).
f
Roxarsone-treated chickens with a 0-day or 5-day withdrawal period (Kawalek et al., 2011).
g
Chicken muscle without differentiation between breast and drumstick.
672 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676
distribution patterns of trace metals in the chicken tissue samples
(Supplementary Tables S7-S10 and Figure S1). Although correlation
does not imply causation, the results consistently suggest that the
main sources of Cu and Zn, Cr and Ni, Cd, and As in the chicken
tissues were different (Supplementary data). Combined with the
common knowledge on the use of feed supplements/additives, it is
reasonable to attribute the occurrence of Cu, Zn, and As at elevated
levels in the chicken tissues on Guangdong markets to their wide
supplementation in commercial chicken feeds (which is confirmed
by results in the following section).
3.2. Trace metals in commercial chicken feeds
Fig. 2a shows the box and whisker plots for the contents of
major trace metals detected in the 61 commercial chicken feeds
(more details are presented in Supplementary Table S11). The mean
contents of trace metals in the feeds increased in the order of
Cd < Co < Se < Pb < As < Ni < Cr < Cu < Zn < Mn. As shown on the
dendogram (Fig. 2b), the trace metals could be clustered as five
distinct groups: the first group was made up of Se, Co, Cd, and Mn,
and the second one consisted of Cr, Ni, and Pb, while the distri-
butions of Zn, As, and Cu were significantly different from the
others. Cu, Zn, and As are widely supplemented to commercial
chicken feeds, while the non-essential, toxic metal species, such as
Cd and Pb, would not be purposely added to them. Therefore, Se, Co,
Cd, Mn, Cr, Ni, and Pb probably originated from the raw materials of
the feeds, such as soy and maize, while Cu, Zn, and As were
contributed mostly by the supplements and additives of the com-
mercial feeds.
Cu and Zn are essential trace elements for poultry growth, and
the amounts required are 4 and 50 mg/kg diet for Cu and Zn,
respectively (ARC, 1975). However, Cu (GM: 17.1 mg/kg; range
3.45e231 mg/kg) and Zn (GM: 84.8 mg/kg; range: 22.8e1250 mg/
Fig. 2. Contents of major trace metals in the 61 commercial chicken feed samples from
Guangdong province: (a) Box and whisker plots; and (b) Dendogram of cluster
analysis.
kg) were actually present in the commercial feeds in Guangdong
province at levels far exceeding what are necessary. In particular,
the Zn contents in some compound feeds were even an order of
magnitude higher than that required for healthy poultry growth.
These results are consistent with the findings of previous studies,
which showed that Cu and Zn were typically present at levels 2e8
times higher than the required ones in poultry and livestock feeds
in China (Cang et al., 2004; Zhang et al., 2012). The mean (5.07 mg/
kg) and range (0.30e33.3 mg/kg) of As contents in the chicken
feeds in our study were relatively close to the measurement results
(mean: 3.6 mg/kg; range: 0.1e17.7 mg/kg) of 70 commercial
chicken feeds in Guangdong province (Yao et al., 2013). The mean
As contents of chicken feeds collected in Jiangsu province of
southern China was reported to be 0.13 mg/kg (Cang et al., 2004),
while the total contents of As in poultry feeds in northeastern China
varied from 0.02 to 6.42 mg/kg (Zhang et al., 2012). It appears that
the mean and range of As contents in chicken feeds from Guang-
dong province were much higher than those found in the feed
samples from other parts of China. Plant-derived feeds, including
cereals, soya, and corns, have relatively low contents of As
(Nicholson et al., 1999). On the other hand, phenylarsenic com-
pounds, primarily roxarsone, are widely used as feed additives in
intensive poultry production in the country (Fu et al., 2016a,b,c; Hu
et al., 2017b; Liu et al., 2013a). Based on the standards for feed
additives in China, roxarsone could be supplemented to chicken
feeds at contents up to 50 mg/kg, or 14.2 mg-As/kg (MOA, 2001).
Thus, the apparently elevated levels of total As in the commercial
chicken feeds from Guangdong province were caused primarily by
the widespread supplementation of phenylarsenic feed additives,
which has been confirmed by arsenic speciation measurements (Hu
et al., 2017b).
According to the national standards for animal feeds issued in
2001, the upper limits of trace metals in poultry feeds are 2.0 mg/kg
for i-As, 0.5 mg/kg for Cd, 10.0 mg/kg for Cr, and 5.0 mg/kg for Pb
(GAQSIQ, 2001). The contents of As, Cd, Cr, and Pb in the com-
mercial chicken feed samples measured in this study were in the
ranges of 0.30e33.3, 0.021e0.11, 1.36e37.6, and 0.22e4.34 mg/kg,
with GMs of 1.55, 0.038, 2.89, and 0.66 mg/kg, respectively. It ap-
pears that the contents of these trace metals in the commercial
chicken feeds were generally in compliance with the respective
feed standards (most of the arsenic in the feeds was present in
organic form). Nonetheless, due to the non-degradability of trace
metals and their potential accumulation in animal tissues, efforts
should be stepped up to reduce the over-supplementation of trace
metals in animal feeds (Hu et al., 2017a,b).
3.3. Public health risk of trace metal exposure from chicken
consumption
Fig. 3a depicts the estimated HQs for exposure to individual
trace metal from consumption of chicken muscles and giblets on
the food markets in Guangzhou and Lianzhou. Although the con-
tents of most trace metals in the giblets were higher than those in
the muscles, the potential health risk from giblet consumption was
negligible compared to that resulting from muscle consumption
due to their much lower ingestion rates. Also, the health risk for the
adult population in Lianzhou was generally lower than that for the
adult population in Guangzhou because of the former's lower
ingestion rates of chicken meat products. The HQ values of all the
trace metals investigated were below 1, which is indicative of
insignificant health risk. Among all the trace metals, Cr posed the
lowest potential health risk, while Se had the highest potential
health risk for both Guangzhou and Lianzhou adults. The non-
carcinogenic risk posed by Zn, As, and Co was comparable to that
of Se for Guangzhou adults, while Pb had the second highest risk for
 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676 673

Fig. 3. Non-carcinogenic and carcinogenic risks from dietary exposure to the trace metals in chicken meat on local food markets for the adult populations of Guangzhou and
Lianzhou in Guangdong province: (a) Estimated HQ of individual trace metal; (b) Cumulative probability distribution of HI of major trace metals for the adult populations in
Guangzhou and Lianzhou; (c) Cumulative probability distribution of ILCR of Pb and i-As for the adult population in Guangzhou; and (d) Cumulative probability distribution of ILCR of
Pb and i-As for the adult population in Lianzhou.

Lianzhou adults. by Monte Carlo simulations. As expected, i-As in the chicken

The combined non-carcinogenic risk from exposure to the
multiple trace metals present in the chicken meat products on the
food markets in Guangzhou and Lianzhou was represented by the
HI. Fig. 3b shows the probability distribution of HI obtained by
Monte Carlo simulations. The HI values for Guangzhou and Lianz-
hou populations were in the ranges of 0.09e0.39 (median: 0.19)
and 0.04e0.29 (median: 0.10), respectively. These results suggest
that dietary exposure to trace metals from consumption of the
chicken meat products on the food markets in Guangzhou and
Lianzhou did not pose any significant non-carcinogenic risk.
Table 3 summarizes the estimated ILCR of As and Pb exposure
from lifetime consumption of the chicken meat products on the
food markets in Guangzhou and Lianzhou. Due to its rather low CSF,
the cancer risk from Pb is insignificant for both Guangzhou and
Lianzhou adults. Meanwhile, i-As is the dominant contributor to
cancer risk from lifetime consumption of chicken meat products,
especially for the adult population in Guangzhou. Although chicken
livers had much higher arsenic contents than muscles, the
maximum ILCR for bladder and lung cancer from consumption of
livers (in the scenario of all giblets consumed were livers) was
much lower than that of muscles (e.g., 3.06  10
6 vs. 1.41  10
4
for Guangzhou adults), primarily because of large difference in
their ingestion rates (0.56 vs. 48.8 g/person/day). Dietary exposure
to i-As in the chicken meat products on the food markets would
lead to 141 additional cases of bladder and lung cancer and 8
additional cases of skin cancer per 1,000,000 adults in Guangzhou.
In comparison, consumption of the chicken meat products on the
food markets in Lianzhou could result in 46 additional cases of
bladder and lung cancer per 1,000,000 adults.
Fig. 3c and d show the cumulative probability distributions of
ILCR of Pb and As exposure from lifetime consumption of chicken
muscles on the food markets of Guangzhou and Lianzhou obtained
muscles posed the most significant cancer risk (bladder and lung
cancer). Approximately 70% of the adult population in Guangzhou
and 30% of the adult population in Lianzhou had risk of bladder and
lung cancer above the serious or priority level of 10
4. In compar-
ison, the cancer risk from dietary exposure to Pb through lifetime
consumption of the chicken muscles on the food markets was
below 10
5 for essentially the entire adult populations in both
Guangzhou and Lianzhou. Together, the above non-carcinogenic
and carcinogenic risk assessment results indicate that i-As in
chicken muscles is the major public health concern for the con-
sumers in Guangdong province. These findings indicate the sig-
nificant need for controlling and even banning the use of
phenylarsenic feed additives in China's chicken production to
protect the health of consumers.
4. Conclusions
The contents of major trace metals (As, Cd, Co, Cr, Cu, Mn, Ni, Pb,
Se, and Zn) in fresh chicken muscles and giblets on the food mar-
kets of a major urban center (Guangzhou) and a primarily rural
region (Lianzhou) in Guangdong province were determined.
Chicken tissues from Guangzhou markets had generally higher
contents of Cu, Zn, As, Co, Mn, and Ni than those from Lianzhou
markets, while similar trend was not observed for Cd, Pb, Se, and
Cd. Results of statistical analyses indicate that the elevated levels of
Cu, Zn, and As in the chicken tissues were mainly caused by the
widespread use of feed supplements/additives containing these
metals (in inorganic and organic forms) in intensive chicken
farming, which is the primary supply source of the chicken on the
food markets in Guangzhou. Cu, Zn, and As were found to be pre-
sent at relatively high levels in the commercial chicken feeds from
Guangdong province. Cluster analysis results indicate that Cu, Zn,
674 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676

Table 3
Estimated ILCR from dietary exposure to i-As and Pb through consumption of the chicken meat products on local food markets in Guangzhou and Lianzhou of Guangdong
province.

Metal CSF Adjustment factora Site Body weight (kg) Chicken tissue consumed Metal content Intake rate ILCR
(mg/kg BW/day)
1 (mg/kg, wet wt.) (g/day, wet wt.)

i-As 25.7b 0.86 Guangzhou 59.5 Muscle 7.80 48.8 1.41  10
4
Giblet 12.5 0.56 2.60  10
6
Liverd 14.7 0.56 3.06  10
6
Lianzhou 57.2 Muscle 4.68 25.2 4.56  10
5
Giblet 9.36 0.29 6.87  10
7
Liverd 6.93 0.29 7.77  10
7
1.5c 0.86 Guangzhou 59.5 Muscle 7.80 48.8 8.25  10
6
Giblet 12.5 0.56 5.14  10
8
Liverd 14.7 0.56 1.78  10
7
Lianzhou 57.2 Muscle 4.68 25.2 2.66  10
6
Giblet 9.36 0.29 4.01  10
8
Liverd 6.90 0.29 4.51  10
8
Pb 0.0085 0.86 Guangzhou 59.5 Muscle 71.6 48.8 4.29  10
7
Giblet 61.8 0.56 4.25  10
9
Liverd 68.2 0.56 4.69  10
9
Lianzhou 57.2 Muscle 91.6 25.2 2.95  10
7
Giblet 83.8 0.29 3.11  10
9
Liverd 92.9 0.29 3.44  10
9
a
Because no specific data on cancer susceptibility from exposure to i-As and Pb, which is dependent on ethnicity and region, is available for the Chinese population, the CSF
values established from the U.S. population are adjusted with a factor of 0.86, a value that has been used previously in assessing the risk of lung cancer from exposure to
polycyclic aromatic hydrocarbons for the Asian populations (Shen et al., 2014).
b
CSF for bladder and lung cancer.
c
CSF for skin cancer.
d
The maximum risk from liver consumption was calculated based on the assumption that livers account for 100% of the giblets consumed.

and As in the commercial feeds originated predominately from the
feed additives, while Se, Co, Cd, Mn, Cr, Ni, and Pb came mostly from
the raw feed materials. The trace metals present in the chicken
tissues posed insignificant non-carcinogenic risk to the adult pop-
ulations in both Guangzhou and Lianzhou. Dietary exposure to Pb
through lifetime consumption of the chicken tissues on the food
markets had insignificant cancer risk to both Guangzhou and
Lianzhou adults, while cancer risk arose from exposure to the
inorganic arsenic contained in the chicken muscles. Results of
Monte Carlo simulations suggest that the risk of bladder and lung
cancer was above the serious or priority level (10
4) for around 70%
and 30% of the adult populations in Guangzhou and Lianzhou,
respectively, due to dietary exposure to the elevated levels of
inorganic arsenic present in the chicken meat products on local
food markets. These findings suggest that the widespread practice
of supplementing commercial feeds with metal-based additives,
particularly phenylarsenic compounds, in intensive chicken
farming has significant public health consequence. To protect
consumers' health, the contents of essential trace metals supple-
mented to commercial feeds should be restricted to the levels just
sufficient to maintain the healthy growth of farmed animals. In
particular, China's policy of allowing the continued use of phenyl-
arsenic feed additives has serious public health implications. Im-
mediate measures should be taken to reduce the contents of arsenic
in chicken meat products to lower the arsenic exposure of the
general population, which necessitates reduction and eventually
phase-out of the use of phenylarsenic feed additives in the coun-
try's poultry industry.
Acknowledgement
The authors gratefully acknowledge the constructive comments
and suggestions from the anonymous reviewers. This work was
supported in parts by the Natural Science Foundation of China
(Grant Nos. 41673089, 41725015, and 41472324), the National Key
Research and Development Program of China (2016YFD0800302),
and the 111 Project (B14001) of the Ministry of Education.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.envpol.2017.12.006.
References
Agency for Toxic Substances & Disease Registry (ATSDR), 2000. Toxicological Profile
for Chromium. Available at: https://www.atsdr.cdc.gov/toxprofiles/tp.asp?
id¼62&tid¼17.
Agricultural Research Council (ARC), 1975. The Nutrient Requirements of Farm
Livestock, No. 1 Poultry, second ed. Agricultural Research Council, London.
Technical Reviews and Summaries.
Ashmead, H.D., Graff, D.J., 1985. Intestinal Absorption of Metal Ions and Chelates.
C.C. Thomas, Springfield, IL.
Association of American Feed Control Officials (AAFCO), 2017. 2017 Official Publi-
cation. Champaign, IL.
Baker-Austin, C., Wright, M.S., Stepanauskas, R., McArthur, J.V., 2006. Co-selection of
antibiotic and metal resistance. Trends Microbiol. 14 (4), 176e182.
Cang, L., Wang, Y., Zhou, D., Dong, Y., 2004. Heavy metals pollution in poultry and
livestock feeds and manures under intensive farming in Jiangsu Province, China.
J. Environ. Sci. 16, 371.
Carlson, M.S., Boren, C.A., Wu, C., Huntington, C.E., Bollinger, D.W., Veum, T.L., 2004.
Evaluation of various inclusion rates of organic zinc either as polysaccharide or
proteinate complex on the growth performance, plasma and excretion of
nursery pigs. J. Anim. Sci. 82, 1359e1366.
CODEX Alimentarius Commission (CAC), 2016. General Standard for Contaminants
and Toxins in Food and Feed (CODEX STAN 193-1995, Amended in 2016). Joint
FAO/WHO Food Standards Programme, Geneva, Switzerland.
Creech, B.L., Spears, J.W., Flowers, W.L., Hill, G.M., Lloyd, K.E., Armstrong, T.A.,
Engle, T.E., 2004. Effect of dietary trace mineral concentration and source
(inorganic vs. chelated) on performance, mineral status, and fecal mineral
excretion in pigs from weaning through finishing. J. Anim. Sci. 82, 2140e2147.
Demirbaş, A., 1999. Proximate and heavy metal composition in chicken meat and
tissues. Food Chem. 67, 27e31.
European Commission (EC), 1999. Council directive 1999/29/EC of 22 April 1999 on
the undesirable substances and products in animal nutrition. Off. J. Eur. Union L
115, 32e46.
Fu, Q., He, J., Blaney, L., Zhou, D., 2016a. Roxarsone binding to soil-derived dissolved
organic matter: insights from multi-spectroscopic techniques. Chemosphere
155, 225e233.
Fu, Q., He, J., Blaney, L., Zhou, D., 2016b. Sorption of roxarsone onto soils with
different physicochemical properties. Chemosphere 159, 103e112.
Fu, Q., He, J., Gong, H., Blaney, L., Zhou, D., 2016c. Extraction and speciation analysis
of roxarsone and its metabolites from soils with different physicochemical
properties. J. Soil. Sediment. 16, 1557e1568.
General Administration of Quality Supervision Inspection and Quarantine
 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676
(GAQSIQ), 2001. Hygienical Standard for Feeds (GB 13078-2001). China Stan-
dard Press, Beijing.
Granados-Chinchilla, F., Mena, S.P., Arias, L.M., 2015. Inorganic contaminants and
composition analysis of commercial feed grade mineral compounds available in
Costa Rica. Int. J. Food Conta 2, 8.
Henchion, M., McCarthy, M., Resconi, V.C., Troy, D., 2014. Meat consumption: trends
and quality matters. Meat. Sci. 98, 561e568.
Hu, Y., Cheng, H., 2016. Health risk from veterinary antimicrobial use in China's food
animal production and its reduction. Environ. Pollut. 219, 993e997.
Hu, Y., Cheng, H., Tao, S., 2017a. Environmental and human health challenges of
industrial livestock and poultry farming in China and their mitigation. Environ.
Int. 107, 111e130.
Hu, Y., Zhang, W., Cheng, H., Tao, S., 2017b. Public health risk of arsenic species in
chicken tissues from live poultry markets of Guangdong province, China. En-
viron. Sci. Technol. 51, 3508e3517.
Huang, L., Luo, X., Tang, Y., Zhang, B., Shu, Y., 2013. Cross-sectional investigation on
diet in Guangzhou City residents aged from 40 to 65 years old. Zhonghua Jibing
Kongzhi Zazhi 17 (11), 942e945 (in Chinese).
Integrated Risk Information System (IRIS), 1988. Arsenic, Inorganic; CASRN 7440-
38-2. Available at: https://cfpub.epa.gov/ncea/iris/iris_documents/documents/
subst/0278_summary.pdf.
Integrated Risk Information System (IRIS), 1998a. Chromium(VI); CASRN 18540-29-
9. U.S. Environmental Protection Agency, Washington, DC. Available at: https://
cfpub.epa.gov/ncea/iris/iris_documents/documents/subst/0144_summary.pdf.
Integrated Risk Information System (IRIS), 1998b. Lead and Compounds (Inorganic);
CASRN 7439-92-1. U.S. Environmental Protection Agency, Washington, DC.
Available at: https://cfpub.epa.gov/ncea/iris/iris_documents/documents/subst/
0277_summary.pdf.
Integrated Risk Information System (IRIS), 1998c. Toxicological Review of Hex-
avalent Chromium, in Support of Summary Information on the IRIS. U.S. Envi-
ronmental Protection Agency, Washington, DC.
Iwegbue, C.M.A., Nwajei, G.E., Iyoha, E.H., 2008. Heavy metal residues of chicken
meat and gizzard and Turkey meat consumed in southern Nigeria. Bulg. J.
Veterinary Med. 11, 275e280.
Kan, C.A., Meijer, G.A.L., 2007. The risk of contamination of food with toxic sub-
stances present in animal feed. Anim. Feed Sci. Technol. 133 (1), 84e108.
Kawalek, J.C., Carson, M., Conklin, S., Lancaster, V., Howard, K., Ward, J., Farrell, D.,
Myers, M., Swain, H., Jeanettes, P., Frobish, S., Matthews, S., McDonald, M., 2011.
Final Report on Study 275.30. Provide Data on Various Arsenic Species Present
in Broilers Treated with Roxarsone: Comparison with Untreated Birds. Food and
Drug Administration, Laurel, MD.
Lasky, T., Sun, W., Kadry, A., Hoffman, M.K., 2004. Mean total arsenic concentrations
in chicken 1989-2000 and estimated exposures for consumers of chicken. En-
viron. Health Persp 112, 18e21.
Liu, Q., Peng, H., Lu, X., Zuidhof, M.J., Li, X., Le, X.C., 2016. Arsenic species in chicken
breast: temporal variations of metabolites, elimination kinetics, and residual
concentrations. Environ. Health Persp 124, 1174.
Liu, X., Zhang, W., Hu, Y., Cheng, H., 2013a. Extraction and detection of organo-
arsenic feed additives and common arsenic species in environmental matrices
by HPLC-ICP-MS. Microchem. J. 108, 38e45.
Liu, Z., Li, X., Chen, W., 2013b. China Animal Industry Yearbook Year: 2012. China
Agriculture Press, Beijing, China.
Mangalgiri, K.P., Adak, A., Blaney, L., 2015. Organoarsenicals in poultry litter:
detection, fate, and toxicity. Environ. Int. 75, 68e80.
Marshall, B.M., Levy, S.B., 2011. Food animals and antimicrobials: impacts on human
health. Clin. Microbiol. Rev. 24 (4), 718e733.
Ministry of Agriculture (MOA), 2001. Bulletin No. 168: Regulations on the Use of
Feed Additives, Beijing, China.
Ministry of Agriculture (MOA), 2009. Safe Use Specifications for Animal Feeds,
Announcement No. 1224 of the Ministry of Agriculture, Beijing, China.
Ministry of Agriculture (MOA), 2013. Lists of Additives Permitted to Use in Animal
Feeds (2013), Announcement No. 2045 of the Ministry of Agriculture, Beijing,
China.
Ministry of Environmental Protection (MEP), 2013. Exposure Factors Handbook of
Chinese Population. China Environmental Science Press, Beijing, China.
Ministry of Health (MOH), 2012. National Food Safety Standard: Limits of Con-
taminants in Foods (GB 2762-2012), Beijing.
Mole, B., 2013. MRSA: farming up trouble. Nature 499 (7459), 398e400.
Morrison, J.L., 1969. Distribution of arsenic from poultry litter in broiler chickens,
soil, and crops. J. Agr. Food Chem. 17 (6), 1288e1290.
Nachman, K.E., Baron, P.A., Raber, G., Francesconi, K.A., Navas-Acien, A., Love, D.C.,
2013. Roxarsone, inorganic arsenic, and other arsenic species in chicken: a US-
based market basket sample. Environ. Health Persp 121 (7), 818e824.
Nachman, K.E., Love, D.C., Baron, P.A., Nigra, A.E., Murko, M., Raber, G.,
Francesconi, K.A., Navas-Acien, A., 2017. Nitarsone, inorganic arsenic, and other
arsenic species in Turkey meat: exposure and risk assessment based on a 2014
U.S. market basket sample. Environ. Health Perspect. 125 (3), 363e369.
National Bureau of Statistics (NBS), 2017. China Statistical Yearbook 2016. China
Statistics Press, Beijing, China.
National Health and Family Planning Commission (NHFPC) and State Food and Drug
Administration (SFDA), 2017. National Food Safety Standard Maximum Levels of
Contaminants in Food (GB 2762-2017), Beijing, China.
Nicholson, F.A., Chambers, B.J., Williams, J.R., Unwin, R.J., 1999. Heavy metal con-
tents of livestock feeds and animal manures in England and Wales. Bioresour.
Technol. 70, 23e31.
675
Office of Environmental Health Hazard Assessment (OEHHA), 2009. 2009 Technical
Support Document for Cancer Potency Factors. Appendix B. Chemical-specific
Summaries of the Information Used to Derive Unit Risk and Cancer Potency
Values. Available at: https://oehha.ca.gov/air/crnr/technical-support-docume
nt-cancer-potency-factors-2009.
Office of Emergency and Remedial Response (OERR), 1989. Risk Assessment Guid-
ance for Superfund, Volume I Human Health Evaluation Manual (Part a) Interim
Final. U.S. Environmental Protection Agency, Washington, D.C.
Oforka, N.C., Osuji, L.C., Onwuachu, U.I., 2012. Assessment of heavy metal pollution
in muscles and internal organs of chickens raised in Rivers State, Nigeria.
J. Emerg. Trends Eng. Appl. Sci. 3, 406e411.
Okoye, P.A.C., Ajiwe, V.I.E., Okeke, O.R., Ujah, I.I., Asalu, U.B., Okeke, D.O., 2015.
Estimation of heavy metal levels in the muscle, gizzard, liver and kidney of
broiler, layer and local (Cockerel) chickens raised within awka metropolis and
its environs, anambra state, south eastern Nigeria. J. Environ. Prot. 6, 609e613.
Poulsen, H.D., 1998. Zinc and copper as feed additives, growth factors or unwanted
environmental factors. J. Anim. Feed Sci. 7 (1), 135e142.
Richards, J.D., Zhao, J., Harrell, R.J., Atwell, C.A., Dibner, J.J., 2010. Trace mineral
nutrition in poultry and swine. Asian. Austral. J. Anim. 23, 1527e1534.
Seiler, C., Berendonk, T.U., 2012. Heavy metal driven co-selection of antibiotic
resistance in soil and water bodies impacted by agriculture and aquaculture.
Front. Microbiol. 3, 399.
Shen, H., Tao, S., Liu, J., Huang, Y., Chen, H., Li, W., Zhang, Y., Chen, Y., Su, S., Lin, N.,
2014. Global lung cancer risk from PAH exposure highly depends on emission
sources and individual susceptibility. Sci. Rep. 4, 6561.
Shurson, G.C., Salzer, T.M., Koehler, D.D., Whitney, M.H., 2011. Effect of metal specific
amino acid complexes and inorganic trace minerals on vitamin stability in
premixes. Anim. Feed Sci. Technol. 163 (2e4), 200e206.
Silbergeld, E.K., Nachman, K., 2008. The environmental and public health risks
associated with arsenical use in animal feeds. Ann. N. Y. Acad. Sci. 1140,
346e357.
Singer, A.C., Shaw, H., Rhodes, V., Hart, A., 2016. Review of antimicrobial resistance
in the environment and its relevance to environmental regulators. Front.
Microbiol. 7, 1728.
Sinkakarimi, M.H., Mansouri, B., Azadi, N.A., Maleki, A., Davari, B., 2017. Assessment
of heavy metals in chicken meat distributed in sanandaj, Iran, and calculating
the food consumption risk. J. Mazandaran Univ. Med. Sci. 26, 128e138.
Stara, J., Patterson, J., Blackburn, K., Hertzberg, R., DeRosa, C., 1986. Guidelines for
the Health Risk Assessment of Chemical Mixtures, EPA/600/D-89/028 (NTIS
PB89222640). U.S. Environmental Protection Agency, Washington, D.C.
Statistical Office of Guangzhou, 2017. Statistical Report of National Economic and
Social Development of Guangzhou in 2016. The Government of Guangzhou,
Guangzhou.
Statistical Office of Lianzhou, 2017. Statistical Bulletin on National Economic and
Social Development in 2016 from Statistical Office of Lianzhou. The Govern-
ment of Lianzhou, Lianzhou.
Tahvonen, R., Kumpulainen, J., 1994. Lead and cadmium contents in pork, beef and
chicken, and in pig and cow liver in Finland during 1991. Food Addit. Contam.
11, 415e426.
Uluozlu, O.D., Tuzen, M., Mendil, D., Soylak, M., 2009. Assessment of trace element
contents of chicken products from Turkey. J. Hazard. Mater 163, 982e987.
Underwood, E.J., Suttle, N.F., 1999. The Mineral Nutrition of Livestock, third ed. CABI
Publishing, New York, USA.
U.S. Department of Agriculture (USDA), 2017. The National List of Allowed and
Prohibited Substances. Available at: https://www.ams.usda.gov/rules-regulatio
ns/organic/national-list.
U.S. Environmental Protection Agency (USEPA), 1993. Reference Dose (RfD):
Description and Use in Health Risk Assessments, Background Document 1A.
Available at: https://www.epa.gov/iris/reference-dose-rfd-description-and-use-
health-risk-assessments.
U.S. Environmental Protection Agency (USEPA), 2000. Methodology for Deriving
Ambient Water Quality Criteria for the Protection of Human Health. U.S.
Environmental Protection Agency, Washington, DC.
U.S. Environmental Protection Agency (USEPA), 2005. Guidelines for Carcinogen
Risk Assessment (EPA/630/P-03/001F). Available at: https://www.epa.gov/sites/
production/files/2013-09/documents/cancer_guidelines_final_3-25-05.pdf.
U.S. Environmental Protection Agency (USEPA), 2010. IRIS Toxicological Review of
Inorganic Arsenic (Cancer) (2010 External Review Draft), EPA/635/R-10/001. U.S.
Environmental Protection Agency, Washington, DC.
U.S. Food and Drug Administration (FDA), 2009. FDA's Response to the Citizen
Petition (FDA-2009-p-0594). Available at: http://www.fda.gov/downloads/Ani
malVeterinary/SafetyHealth/ProductSafetyInformation/UCM370570.pdf.
Veum, T.L., Carlson, M.S., Wu, C.W., Bollinger, D.W., Ellersieck, M.R., 2004. Copper
proteinate in weanling pig diets for enhancing growth performance and
reducing fecal copper excretion compared with copper sulfate. J. Anim. Sci. 82,
1062e1070.
Vieira, S.L., 2008. Chelated minerals for poultry. Rev. Bras. Cienc. Avic. 10 (2), 73e79.
Visek, W.J., 1978. The mode of growth promotion by antibiotics. J. Anim. Sci. 46 (5),
1447e1469.
Wu, Y., Zhang, H., Liu, G., Zhang, J., Wang, J., Yu, Y., Lu, S., 2016. Concentrations and
health risk assessment of trace elements in animal-derived food in southern
China. Chemosphere 144, 564e570.
Yabe, J., Nakayama, S.M., Ikenaka, Y., Muzandu, K., Choongo, K., Mainda, G.,
Kabeta, M., Ishizuka, M., Umemura, T., 2013. Metal distribution in tissues of
free-range chickens near a leadezinc mine in Kabwe. Zamb. Environ. Toxicol.
676 Y. Hu et al. / Environmental Pollution 234 (2018) 667e676

Chem. 32, 189e192.
Yao, L., Huang, L., He, Z., Zhou, C., Li, G., 2013. Occurrence of arsenic impurities in
organoarsenics and animal feeds. J. Agr. Food Chem. 61, 320e324.
Zhang, F., Li, Y., Yang, M., Li, W., 2012. Content of heavy metals in animal feeds and
manures from farms of different scales in northeast China. Int. J. Environ. Res.
Public Health 9, 2658e2668.
Zheng, N., Wang, Q., Zhang, X., Zheng, D., Zhang, Z., Zhang, S., 2007. Population
health risk due to dietary intake of heavy metals in the industrial area of
Huludao city, China. Sci. Total. Environ. 387, 96e104.
Zhuang, P., Zou, H., Shu, W., 2009. Biotransfer of heavy metals along a soil-plant-
insect-chicken food chain: field study. J. Environ. Sci. 21, 849e853.