Effects of seasons and parts on volatile N-nitrosamines and their exposure and risk assessment in raw chicken and duck meats

Abstract

The N-nitrosamine (NA) concentrations and types in raw chicken and duck meats of different parts and seasons were estimated by headspace solid-phase micro-extraction with gas chromatography-mass spectrometry (HS–SPME–GC–MS). The exposure level and hazard quotient of each detected volatile N-nitrosamine (VNA) were conducted. The selected chicken and duck samples were contaminated by VNAs to some extent. The major types and contents of VNAs in different parts of chicken and duck meats varied seasonally. For chicken samples, the order of the total VNA concentrations was as follows: autumn > spring > winter > summer. For duck samples, the order was changed as follows: winter > autumn > summer > spring (thigh samples) and autumn > spring > winter > summer (breast samples). The estimated exposure levels for adults caused by duck consumption were slightly higher than those by chickens, which was consistent with the tendency in 2–3 years old children. According to the linear regression correlation between the 10% benchmark dose limit (BMDL₁₀) and subtriplicate of median lethal dose (LD₅₀), BMDL₁₀ values of each VNA were calculated. Due to this hypothesis, the risk assessments of each detected VNA and total VNAs posed by consuming chicken and duck meats in Tianjin, China were of low concern.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13197-021-05195-1.

Introduction

N-nitrosamines (NAs), a group of carcinogenic chemicals, are one of the four contaminants (3, 4-Benzopyrene, Acrylic amide, N-nitrosamines, Aflatoxin) in foods based on their carcinogenic potential and versatility, which have been promulgated by International Agency for Research on Cancer (IARC 1978). Normally, VNA formation in meat products is a thermally-induced process that does not occur until the cooking temperature is reached. However, one cannot exclude the possibility that NAs can find a way to animal digestive system via its feed and water. As it is well known, the nitrogen substances (nitroso compounds, nitrates etc.), which accumulate in plants from many inconsiderate human activities, such as excessive fertilization, increased soil acidification, and global warming, might be NA precursors which are much easier for the formation of NAs (Liao et al. 2019; Kim and Hur 2019). In addition, nitrogen-loving plants are consumed by animals on pastures. Therefore, NAs may reach animal organisms (cattle, pigs, chicken, ducks, etc.) and cause immediate danger to their health and indirectly hazard to their consumers’ health as well. To date, more attention has been paid to NAs in red meats because of their high consumption and the use of food additives (nitrite or nitrate) (Lopez-Moreno et al. 2016). Current evidence shows that no NA has been detected above the limit of detection (LOD) in raw poultry meat except in processed poultry meat (Lee 2019), which ranges from 1.60 to 22.4 µg/kg for total VNAs. However, Juszkiewicz et al. studied the absorption, tissue deposition and passage into eggs of N-nitrosodimethylamine (NDMA) of hens, and the results showed that a single oral dose of 0.1 mg NDMA/kg produced tissue levels up to 0.06 mg/kg and resulted in excretion of detectable amounts of the carcinogen in eggs for up to 6 days (Juszkiewicz et al. 1978). Additionally, poultry meat is one of the most popular foods all over the world, which is surpassing beef consumption in most parts of the world over the last few decades (Poultry information 2020). Therefore, it’s very important to monitor NA levels in raw poultry meats to assess the potential risks posed by their consumption for human health. Moreover, as far as we know, there are few reports on NA levels in raw poultry meat affected by different parts and seasons. In short, there is an urgent need to determine the detailed distribution of NAs in raw poultry meat.

Up to now, reports on risk assessment for VNAs have been conducted in drinking water (Fan and Lin 2018; Luo et al. 2020; Uzun et al. 2015; Krasner et al. 2013), beer (Fan and Lin 2018), and processed meat (Herrmann et al. 2015), which involved methodologies of risk assessment such as Margin of exposure(MOE), Threshold of Toxicological Concern (TTC), Linear extrapolation and Relative potency factor(RPF) and so on. In contrast, few studies have been conducted on the health risks of VNAs posed by the consumption of poultry meat. Generally, the point risk assessment model is one of the most commonly used methods, which estimates human exposure assessment of a contaminant through oral ingestion using daily intake and uptake of the contaminant. The exposure level of each VNA in chicken and duck samples, which is represented as the estimated daily intake (EDI), was estimated for different groups of people in this study. Then, based on the exposure assessment, the hazard quotient (HQ, the ratio of EDI to acceptable daily intake (ADI)) was used to assess the potential human health risks posed by the consumption of raw poultry. However, there is no limitation of ADI for each or total NAs recommended by any authority, which makes assessment impossible. In this study, the Benchmark dose (BMD) model is introduced for extrapolating the ADI values of each studied VNA and its application in risk assessment (Chen et al 2021; Filipsson and Victorin 2003; Slob et al. 2005; Muri et al. 2009). However, the BMDL₁₀ values of some studied NAs could not obtain from the toxicology database. Hwang et al (2019) investigated the carcinogenic risk assessment of several cooking-related carcinogens (heterocyclic amines, acryalmide, furan, polycyclic aromatic hydrocarbons and nitrosamines) according to the oral slope factor and BMDL₁₀ (Hwang et al. 2019). In this reference, the BMDL₁₀ values of two types of VNAs were calculated based on the median toxic dose (TD₅₀) (Hwang et al. 2019). Nowadays, the only available parameter for each NA was LD₅₀ data. According to these available experimental toxicological data, in this study, correlations were observed between BMDL₁₀ and LD₅₀ for extrapolating BMDL₁₀ values of some VNAs which have no published ones. Then these speculated BMDL₁₀ values were employed to calculate ADI values in the subsequent study. According to the above hypothesis, hazard quotient (HQ) of each detected VNA in chicken and duck meats which affected by parts and seasons was conducted. To the best of our knowledge, this is the first report on the health risk assessment of each detected VNA and total VNAs in chicken and duck samples based on different parts and seasons.

In this paper, HS–SPME–GC–MS was employed for the quantification of VNAs in chicken and duck samples; then VNA types and levels affected by parts and seasons in raw chicken and duck samples obtained from Tianjin, China were performed. Secondly, the exposure levels of each VNA and total VNAs based on the data mentioned above, that is, the exposure level variation caused by parts and seasons, were calculated. Finally, due to the correlation between BMDL₁₀ and LD₅₀, risk assessment was performed by estimating the HQ values of each detected VNA and total VNAs to assess the potential risks for human health.

Materials and methods

Reagents and materials

Methanol was obtained from Tianjin Concord Science & Technology Co., Ltd, NaCl from Tianjin Fengchuan Chemical Reagent Technologies Co., Ltd; deionized water from Hangzhou Wahaha Group Co., Ltd,. Chicken and duck were purchased from a local retail market in Tianjin, China for sample preparation.

The standard VNAs (abbreviations of VNAs are shown in Table S1 as follows: N-nitrosodimethylamine (NDMA), N-nitrosodiethylamine (NDEA), N-nitrosodipropylamine (NDPA), N-nitrosodibutylamine (NDBA), N-nitrosopiperidine (NPIP), N-nitrosopyrrolidine (NPYR), N-nitrosomorpholine (NMOR), N-nitrosodiphenylamine (NDPheA) and N-nitrosomethylethylamine (NMEA) in methanol are commercial products used as standard solutions from Sigma, USA. The solutions were diluted with methanol to 40.00 µg/mL and stored in brown reagent bottles in the dark before the subsequent analysis.

Sample preparation

The chicken and duck samples were prepared according to a method described in a previous study (Sun et al. 2020). All samples were purchased from local wholesales in Tianjin, China. Samples were obtained during spring, summer, autumn and winter. A total of 140 samples (five different brands of chicken), consisting of 70 chicken (thigh and breast) and 70 duck samples (thigh and breast), were included. The samples were treated similarly to the method in a previous study (Sun et al. 2020). In general, chicken and duck in feeding farm are ready for market in about 50 days, which the weight is more than 2 kg. In this study, the samples that weighed above 1.5 kg were grown and then collected in the same season. The months for samples growing and collecting are shown in Table S2.

Extraction method

The HS-SPME extraction process followed a previously described work (Sun et al. 2020) with few modifications. Briefly, approximately 2.000 g samples were mixed with saturated salt water (9.0 mL) and added to 15 mL amber glass vials. Then the rest of the extraction process was similar to the previous study (Sun et al. 2020).

GC–MS analysis

The GC–MS analysis was performed based on a previous work (Sun et al. 2020). The series of solutions containing each VNA with concentrations in the range of 0.10–20.00 µg/mL were used for the calibration curves (using seven calibration levels) in methanol. The GC–MS conditions for detection of VNAs is quite similar to our another paper (Sun et al. 2020).

Validation of the method

The method (HS–SPME–GC–MS) was optimized according to a previous study (Sun et al. 2020). Linearity, LOD, limit of quantification (LOQ), accuracy, and precision were determined in a validation study based on a previous study (Sun et al. 2020). The LOD and LOQ were determined using spiked samples. The LOD was calculated as follows:

$$\text{LOD}=3\times \delta$$ 1

$$\text{LOQ}=10\times \delta$$ 2

where $\delta$ is the signal to noise ratio.

Methodology of exposure and risk assessment

ADI speculation by BMD modeling

In this study, the BMD approach is used to speculate ADI based on Eq. (3).

$$\text{ADI}=\frac{{\text{BMDL}}_{10}}{\text{UF}}$$ 3

where BMDL₁₀ is the benchmark dose lower-bound confidence limit of 10% and is obtained from public databases (SCCS 2012). The uncertainty factor (UF), which is 100 (10 for interspecies variation and 10 for intra-species variation) was used in this study. According to the results shown in Fig. S1, S2 and Table S3, the BMDL₁₀ values of individual VNAs (NDMA, NDEA, NPYR and NMOR) and their LD₅₀ values were significantly correlated with each other. Therefore, there was a 95% probability that the BMDL₁₀ values of individual VNAs (NDMA, NDEA, NPYR and NMOR) and subtriplicate of LD₅₀ were fitted using linear regression. The formula,$y=0.053x-0.257$ ₍$x=\sqrt[3]{{\text{LD}}_{50}}$_, $y={\text{BMDL}}_{10}$₎, was then employed to calculate BMDL₁₀ values of each detected VNA in this study. The ADI values calculated using BMDL₁₀ (Eq. (3)) are listed in Table S4.

Exposure assessment

The possible exposure of consumers to VNAs hinges on chicken and duck consumption. Consumption data on chicken and duck meats among the population in Tianjin, China, from the survey report of “Reports of Nutrition and Health Status of Chinese Residents (10): The 2002 nutrition and health data set”, was employed to estimate VNA exposure level (shown in Table S5). Several groups were employed during the estimation, such as ages (years of 2–3; 4–6; 7–13; 14–17; 18–59; ≥ 60), sex (male or female) and regions (urban and rural).

The EDI was used during the estimation process and was calculated using Eq. (4).

$$\text{EDI}=\frac{\sum {\text{CR}}_i\times C_i}{B_W}$$ 4

where EDI is the estimated daily intake of sample i (ng/kg bw d), $C_i$ is the mean residual content of VNAs in sample i (ng/kg), ${\text{CR}}_i$ is the mean consumption of sample i (ng/kg) (shown in Table S5), and $B_w$ is body weight (shown in Table S5).

Risk assessment

In this study, Point Risk Assessment was employed for the risk assessment of NAs proposed by the consumption of poultry meat, which is based on the approach suggested by the United States Environmental Protection Agency (US EPA 2019). The following Eq. (5) was used for the estimation and calculation processes.

$$\text{HQ}=\frac{\text{EDI}}{\text{ADI}}$$ 5

where HQ is the hazard quotient and ADI is the acceptable daily intake (µg/kg). In this study, HQ (hazard quotient) is compared with 1 as recommended by a previous study (Gad et al. 2015). There is a risk to consumers when HQ > 1; otherwise, it is considered to be of low concern.

Statistical analysis

Microsoft Office Excel 2007, SPSS 19.0 software and Origin 9.1 were used for data analysis and exposure assessment in this study. Values that were not detected (N.D.) were represented as “0” during the calculation of the exposure and risk assessment process in this study. By repeating the analysis three times, the most reliable evaluation of contamination levels for each of the animal parts and for each of the seasons of the year in the investigated materials could be achieved.

Results and discussion

VNAs concentrations in poultry meat samples

The analytical method was validated according to a previous report (Sun et al. 2020), which is shown in Table 1. Good linear ranges were obtained for the nine analytes with coefficient (R²) higher than 0.999. The LODs and LOQs obtained with the described method were in the ranges of 0.53–1.52 µg/kg and 1.77–5.07 µg/kg, respectively. The results indicated that HS–SPME–GC–MS is an applicable method to determine VNA concentration in chicken and duck samples.

Table 1.

Linear equation and recovery rate of 9 VNAs

VNAs	Linear equation	Relevance R2	LOD(μg/L)	LOQ(μg/L)	Recovery rate/%
NDMA	Y = 105.45x − 4.79	0.9999	1.15	3.83	90.21
NMEA	Y = 201.01x − 2.65	0.9995	0.96	3.20	95.15
NDEA	Y = 66.65x − 6.00	0.9996	1.21	4.03	91.26
NDPA	Y = 48.83x − 2.68	0.9999	0.89	2.97	90.89
NDBA	Y = 36.21x − 3.32	0.9998	1.52	5.07	93.53
NPIP	Y = 55.90x − 2.51	0.9999	0.75	2.50	91.54
NPYR	Y = 52.47x − 9.34	0.9990	0.99	3.30	92.82
NMOR	Y = 0.48x − 4.86	0.9994	0.53	1.77	91.68
NDPheA	Y = 4.85x − 10.18	0.9992	0.85	2.83	90.13

VNAs Volatile N-nitrosamines; NDMA N-nitrosodimethylamine; NDEA N-nitrosodiethylamine; NDPA N-nitrosodipropylamine; NDBA N-nitrosodibutylamine; NPIP N-nitrosopiperidine; NPYR N-nitrosopyrrolidine; NMOR N-nitrosomorpholine; NDPheA N-nitrosodiphenylamine; NMEA N-nitrosomethylethylamine; R² Coefficient; LOD Limit of detection; LOQ Limit of quantification

The concentrations and types of nine VNAs in the chicken and duck samples from different parts and seasons are summarized in Table S5. VNAs have been detected in some raw poultry meats. For chicken samples (mean detectable rate is 35.14%), the mean contents of VNAs ranged from N.D. to 1.83 ± 1.17 µg/kg, with the highest VNA, namely NMEA, observed in chicken breast sample in autumn. On the other hand, for duck samples (mean detectable rate is 37.13%), the concentrations of each VNA ranged from N.D. to 3.51 ± 0.82 µg/kg, where the NMEA was also the highest VNA in duck breast sample in autumn. Notably, the most detected VNAs were NMEA, NPYR and NDEA in almost all chicken samples; for the duck samples, NDPA, NMEA and NDEA were found. In addition, there was an interesting founding that no NDPA was detected in chicken samples but was detected in duck samples. The same phenomenon was observed for NPYR, which was detected only in chicken samples. The reason for this phenomenon is not clear, and further research on this field is necessary. Taken together, the possible explanation for this phenomenon might be due to the precursor for the formation of each VNA that changes with the type and content of major components, such as proteins, peptides, amino acids and amines, in various species (Ryszard 2003a, b). For example, a precursor of NDMA is an amino acid glycine, which may originate from quaternary amino salts, such as choline, acetylocholine, and betaine. NDEA is formed from alanine. Both NDMA and NDEA may also originate from mono- di- and tri-methylamines and respective methylamines. Therefore, the reasons for the different contents of each VNA in different parts and chicken or duck were not elucidated. It is worthwhile to systematically monitor the VNA contents in animal parts and food of animal origin for preventive actions aiming at protecting the population against exposure to toxic substance. In addition to the endogenic synthesis of NAs, polluted earth, water, vegetables, feed, and food environments are also of concern. It is difficult to determine the role of different synthesis factors of NAs in the raw meat of slaughtered animals. Current research reveals that nitroso compounds play important roles in the initiation of carcinogenesis in humans and animals, depending on the ecological and geophysical conditions of the environment. Therefore, investigations concerning the environment and feed should be considered to explain the origin or formation of NAs in animals depending on the chemical content of the feed, sources of their production, and type of feeding. Therefore, further research should be conducted.

In China, the limitation for VNAs for meat samples is very scarce, and the only regulated NA is NDMA, with a limit of 3 µg/kg set by the National Standards of the People’s Republic of China (GB2762-2017, China). Fortunately, the NDMA content in all studied samples was below the criterion (NDMA, 3 µg/kg). Indeed, NDEA, NMEA and NDPA were commonly found in almost all the meat samples, while other NAs were not detected above the LOD. The highest total VNA content in samples was 9.05 µg/kg, which was lower than 10 µg/kg (Li et al. 2018) but higher than 2 µg/kg (Li et al. 2018). In previous reports, no VNAs were detected above the LOD in uncooked poultry. Only cooked poultry had some, which ranged from 1.60 to 22.4 µg/kg (Bara et al. 2011; Lee 2019). As acidified soil and water have become more seriously all over the world, intensive formation of nitro, nitroso and other compounds was observed; these might be the formation precursors of NAs (Ryszard 2003b; Hill 1988). Moreover, the earlier results also proved that NAs concentrations in venison (such as deer, boar, hare and so on) from animals characterised by non-ecological and ecological breeding factors were significantly different (Ryszard 2003a). Despite the low contents for each NA, the concentrations of total NAs were high. Therefore, it is important to monitor the occurrence of NAs to avoid the potential human health risks posed by consuming poultry meat.

In light of the data in Table S5, the radar chart and statistical analysis were employed, and the results are shown in Fig. 1. As shown in Fig. 1, for chicken samples (thigh and breast), the order of the total VNA concentrations was as follows: autumn > spring > winter > summer. In contrast, the order of the total VNA concentrations for duck samples was as follows: winter > autumn > summer > spring (thigh samples), autumn > spring > winter > summer (breast samples), respectively, which was not consistent with the chicken samples. Therefore, the major types and VNA contents in different parts of the studied samples varied seasonally. Normally, the formation of residual NAs in meat samples depends on air conditions, environmental temperature and time, and presence of nitrosamine precursors, catalysts, inhibitors and storage conditions et al. (Krasner et al. 2013; Fan and Lin 2018; Bei et al. 2016; Ryszard 2003b). The reason why the content variation of each VNA depending on the season is not clear according to the current state of knowledge. Thus, further research about this is needed.

Fig. 1.

Radar chart comparison of VNA concentrations in different seasons and parts in poultry

The mean detectable rate for all investigated samples was 36.14% in this study through GC–MS, which was the main method for detecting VNA concentrations stipulated by the National Standard of the People’s Republic of China (GB 5009.26–2016). Using the CG–MS detection method, in the data above LOD, 74.41% was higher than LOD but lower than LOQ, and 25.59% was higher than LOQ, which indicates that the contents of some NAs can be accurately quantified. Obviously, as the standard analytical method, other methods with higher sensitivity still need to be developed in the future for NA detection. According to the data discussed above; the studied sample was contaminated by VNAs to some extent. Hence, more attention should be paid to NAs in chicken or duck samples. Herein, the potential health risks associated with exposure to VNAs via food intake were evaluated in the subsequent research.

Exposure assessment

Assessing the impacts and risks posed by VNAs requires the estimation of potential exposure levels to these chemicals. During exposure assessment, consumption data of chicken and duck meats and body weight were obtained from “Reports of Nutrition and Health Status of Chinese Residents (10): 2002 nutrition and health data set” (shown in Table S6). Several groups were selected during the estimation exposure assessment process, including different ages (years of 2–3; 4–6; 7–13; 14–17; 18–59; ≥ 60), sex (male and female), and regions (urban and rural) and the exposure level was calculated due to Eqs. (3 and 4).

The estimated mean exposure levels (long-term and short-term) of each detected VNA and total VNAs for the chicken thigh in spring are shown in Table S6 to S37. For the children (2–3 years old), the estimated exposure levels posed by consuming duck samples were notably higher than those of the other groups. For the duck samples, the highest EDI of total VNAs was found in the 2–3 years old female of urban (autumn, breast sample), which was 7.99 ng/(kg bw d) (mean 5.69 ng/(kg bw d)), and the ones for summer, winter and spring were 4.60, 5.08 and 6.34 ng/(kg bw d), respectively.

As for the adults (18–59 years old), the estimated exposure levels caused by duck samples were slightly higher than those caused by chicken, which was consistent with 2–3 years old children. The highest exposure level of total VNAs, for the adults (18–59 years old) was found in the urban male group (Autumn, Breast sample), which was 3.17 ng/(kg bw d). Additionally, the exposure levels were 1.83, 2.02 and 2.52 ng/(kg bw d) for summer, winter and spring, respectively.

In the literature, the exposure level to volatile NAs in West Germany was estimated to be 300 ng/day and 200 ng/day for males and females, respectively (Tricker et al. 1991). In this study, the estimated exposure levels posed by VNAs were of the same order of magnitude as those reported by other researchers (Herrmann et al. 2015; Keszei et al. 2013; Tricker et al. 1991).

Risk assessment

The ADI is used to assess the possibility of the lifetime intake of the chemical to cause any measurable harm to human health or not at this dose. ADI is the ratio of the no observed adverse effect level (NOAEL) to UF. Unfortunately, according to limited toxicological data, the NOAEL values of some contaminants in food could not be obtained, which restricts the development of food safety assessment. Therefore, it’s very important to speculate unknown toxicological parameters according to experimental toxicological data. Indeed, the published BMDL₁₀ values of NDPA, NDBA, NPIP, NMEA and NDPheA were not available from toxicology database, and the only LD₅₀ data of each VNA could be obtained. Therefore, the correlations between BMDL₁₀ and LD₅₀ were observed according to the toxicological data of NDMA, NDEA, NPYR and NMOR. The correlations between BMDL₁₀ and LD₅₀ after a series of fittings are shown in Fig. S1 and Table S4. As shown in Fig. S1 and Table S4, a formula linked to BMDL₁₀ and LD₅₀ was established. In this study, a 95% probability was shown that the correlation between published BMDL₁₀ values of four VNAs (NDMA, NDEA, NPYR and NMOR) and their LD₅₀ values were fitted using linear regression. More details are provided in the methodology of the exposure assessment. The BMDL₁₀ value of NPIP in this study was similar to that reported in a previous study (Hwang et al. 2019). Therefore, using this method, each detected VNA in the studied samples was assessed during the risk assessment process.

The HQ values calculated according to Eq. (5), are shown in Fig. 2. If the HQ value is higher than 1, it is considered dangerous; otherwise, it is of low concern. Fortunately, the HQ values obtained in this study for VNA contents in different seasons and parts, were all lower than 1, as shown in Fig. 2. Hence, the risk posed by VNAs from chicken and duck samples might be considered of low concern.

Fig. 2.

Risk Assessment of Total VNAs for Chicken in Long-term and Short-term: a 2–3 years old group; b 4–6 years old group; c 7–13 years old group; d 14–17 years old group; e 18–59 years old group; f ≥ 60 years old group

However, the VNAs are genotoxic compounds, and small exposure levels may still have a genotoxic effect (Dybing et al. 2008). Furthermore, it should also be noted that the results from this study were based on fairly limited number of results on the occurrence of VNAs in raw chicken and duck meats. For total VNAs, the HQ values caused by breast meat in chicken were slightly higher than those in thigh samples for the female populations, which is a concern. The same conclusion was drawn in the long-term risk assessment.

Although the present study indicated that the VNA exposure level posed by consumption of raw chicken and duck meats was of low concern, the dietary exposure to VNAs might still pose a health risk due to the increased consumption of them to some degree. Additionally, chicken and duck are only one of the food materials which pose a health risk to human health through NAs; other sources should be evaluated. Moreover, the carcinogenic potential of most NAs is largely unknown. Meanwhile, in this study, risk assessment caused only by exogenous VNAs was employed, and VNAs of endogenous origin were not considered, which should also be included in the total risk assessment. On the other hand, endogenous NAs could occur at much higher concentrations than the exogenous ones, and further toxicological studies are needed to fully assess the risk of NA exposure from chicken and duck samples.

Conclusion

Given the increase in chicken and duck consumption, more attention has been paid to the health risks associated with them. The studied samples were contaminated by VNAs to some degree. There was a seasonal variation in the VNA contents in the chicken, with the highest content occurring in autumn and the lowest in summer. In all seasons, the contents of total VNAs in the duck samples were higher than those in the chicken samples. Moreover, the major types and contents of VNAs in various parts of the studied samples varied with seasons. According to the results discussed above, chickens might be a better choice for the human diet in summer.

The highest exposure level caused by duck consumption was found in the 2–3 years old urban females (Autumn, Breast sample) with 7.99 ng/(kg bw d) (mean 5.69 ng/(kg bw d)). As for the adults (18–59 years old), the estimated exposure level posed by duck consumption was slightly higher than the one by chicken consumption, which was consistent with the tendency observed in 2–3 years old children. The highest exposure level of total VNAs for adults (18–59 years old) was found in the urban male group (Autumn, Breast sample) with 3.17 ng/(kg bw d). Additionally, for summer, winter and spring, the exposure levels in this group were presented as 1.83, 2.02 and 2.52 ng/(kg bw d), respectively. Based on a 95% probability, the correlation between published BMDL₁₀ values of four VNAs (NDMA, NDEA, NPYR and NMOR) and their LD₅₀ values were fitted using linear regression, which was employed to estimate the ADI values of other VNAs. According to above hypothesis, all HQ values in the risk assessment process were lower than 1. Thus, the risk assessment results showed that the chicken and duck samples should be of low concern. As VNAs are genotoxic compounds and even small exposure levels may still have a genotoxic effect, more studies on NAs in raw chicken and duck are suggested for better estimation and control of the risks associated with their consumption.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

HS–SPME–GC–MS

Headspace solid-phase micro-extraction with gas chromatography-mass spectrometry

VNA

Volatile N-nitrosamine

NAs

N-nitrosamines

LOD

Limit of detection

EDI

Estimated daily intake

HQ

Hazard quotient

ADI

Acceptable daily intake

BMD

Benchmark dose

BMDL₁₀

Benchmark dose limit 10% values

LD₅₀

Median lethal dose

TD₅₀

Median toxic dose

NDMA

N-Nitrosodimethylamine

NDEA

N-Nitrosodiethylamine

NDPA

N-Nitrosodipropylamine

NDBA

N-Nitrosodibutylamine

NPIP

N-Nitrosopiperidine

NPYR

N-Nitrosopyrrolidine

NMOR

N-Nitrosomorpholine

NDPheA

N-Nitrosodiphenylamine

NMEA

N-Nitrosomethylethylamine

LOQ

Limit of quantification

UF

Uncertainty factor

N.D.

Not detected

R²

Coefficient

NOAEL

No observed adverse effect level

MOE

Margin of exposure

TTC

Threshold of Toxicological Concern

RPF

Relative potency factor

Author contributions

Kexin Li Conceptualization, methodology, validation, formal analysis, data curation, investigation and writing. Rui Wang Methodology, validation, formal analysis and data curation. Xiaoxu Wang Conceptualization, methodology and resources. Changxia Sun and Qiang Li Conceptualization, methodology and resources.

Funding

The authors gratefully acknowledge the financial support provided by the Fundamental Research Funds for the Central Universities (2015ZCQ-LY-03).

Declarations

Conflict of interest

The authors declare that they have no known competing financial interest or personal relationships that could have appeared to influence the work reported in this paper.

Ethical approval

Compliance with ethics approval.

Consent to participate

All authors have seen and agreed with the contents of the manuscript.

Consent for publication

All authors are aware of its submission to JFST.