Urinary concentrations and profiles of organophosphate and pyrethroid pesticide metabolites and phenoxyacid herbicides in populations in eight countries

Abstract

Concentrations of nine metabolites of organophosphate and pyrethroid insecticides, as well as two phenoxy herbicides, were determined in 322 urine samples collected from eight countries during 2006–2014 by high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS). The target pesticides were found ubiquitously, indicating widespread exposure of humans to pesticides in these countries. The highest sum concentrations of 11 pesticides were found in urine collected from Vietnam (median, 28.9 ng/mL), followed in decreasing order by samples from India (14.2 ng/mL), China (13.6 ng/mL), Korea (12.5 ng/mL), Greece (12.3 ng/mL), Saudi Arabia (11.3 ng/mL), the USA (7.9 ng/mL), and Japan (7.1 ng/mL). Organophosphorus compounds accounted for 62–77% of the total urinary pesticide concentrations. Para-nitrophenol (metabolite of parathion and methyl parathion) and 3,5,6-trichloro-2-pyridinol (metabolite of chlorpyrifos and chlorpyrifos-methyl) were the major metabolites, especially in India (72%), China (69%), and Greece (66%). Differences in urinary pesticide concentrations between genders (male vs. female), age groups (categorized as ≤20, 21–49, and ≥50 years), and cities (Guangzhou, Shanghai, and Qiqihar) were examined. On the basis of the concentrations measured in urine, total daily intakes (DIs) of pesticides were estimated. The DIs of chlorpyrifos were found to be higher for populations in Vietnam, Greece, India, China, and Korea (≥9.6 μg/day) than those estimated for the other countries (<5 μg/day). The DIs of parathion (≥9.6 μg/day) in populations of China, India, and Korea were higher than those estimated for the other countries (5.7–9.3 μg/day). This is the first study to establish baseline levels of exposure of a variety of pesticides in several Asian countries.

Graphical abstract

1. Introduction

The global consumption of pesticides increased by 20% from 2008 to 2012, with an estimated production of 5.8 billion pounds of active ingredients (EPA, 2017). Organophosphorus (OP), pyrethroid (PYR) insecticides and phenoxy acid (PA) herbicides are the most commonly used pesticides in agricultural, domestic, and gardening applications (CDC, 2013; Oulhote and Bouchard, 2013). OPs inhibit acetylcholinesterase that normally breaks down the neurotransmitter acetylcholine; PYRs act upon voltage-gated sodium channels on nerves; and PAs mimic the plant growth hormone, auxin, resulting in uncontrolled growth and subsequent death of the plant (McKelvey et al., 2013; Jayakody et al., 2015).

Given the magnitude of their global use, residues of OPs, PYRs, and PAs have been frequently detected in various environmental media, including soil, water, and sediment, and they enter through food chains, raising concerns for human health (Alonso et al., 2012; Corcellas et al., 2012; Jayakody et al., 2015; Rippy et al., 2017; Rousis et al., 2017a; Tang et al., 2018). Several epidemiologic studies have shown that chronic, low-dose exposure to OPs and PYRs was associated with neurological outcomes (e.g., neurodevelopment in children, Parkinson’s disease), endocrine-disrupting effects (e.g., metabolic syndrome, obesity, diabetes), and reproductive effects (e.g., reduced semen quality, reduced gestational age, reduced birth weight) (Furlong et al., 2014; González-Alzaga et al., 2014; Saillenfait et al., 2015; Coker et al., 2018). Exposure to 2,4-dichlorophenoxyacetic acid (2,4-D) and 2,4,5-trichlorophenoxyacetic acid (2,4,5-T), the most commonly used PAs, is linked to wheeze, non-Hodgkin’s lymphoma, soft-tissue sarcoma, and age-related macular degeneration (Jayakody et al., 2015; Hoppin et al., 2017; Montgomery et al., 2017).

The general population is exposed to pesticides mainly through diet, with other contributions from inhalation of air, dermal absorption, and unintentional ingestion (Lu et al., 2006; McKelvey et al., 2013; Oulhote and Bouchard, 2013; Glorennec et al., 2017; Gracia-Lor et al., 2017). In human bodies, OPs and PYRs are typically metabolized and excreted in urine as conjugates within 4–48 h after exposure (Glorennec et al., 2017; Garí et al., 2018), whereas PAs are excreted in urine as unchanged parent compounds (80–95%) or as a conjugate, with urinary half-lives of 12–72 h (Aylward et al., 2010; Jurewicz et al., 2012). Thus, urinary levels of OPs, PYRs, and PAs are used as biomarkers of exposure to these pesticides (McKelvey et al., 2013; Oulhote and Bouchard, 2013; Glorennec et al., 2017). The US Centers for Disease Control and Prevention (CDC) has monitored the presence of pesticides and their metabolites in urine from a representative US population since 2003 (Swan et al., 2003; Eskenazi et al., 2004; Lu et al., 2006; McKelvey et al., 2013; Calafat et al., 2017). Biomonitoring studies on the measurements of pesticides in urine from populations in Spain (Roca et al., 2014; Garí et al., 2018), France (Viel et al., 2015; Glorennec et al., 2017), and China have been reported (Qi et al., 2012; Wu et al., 2013). Most of these studies focused on either OP or PYR metabolites, with measurements in urine of pregnant women and children, from a limited geographic location. Little is known on the exposure of pesticides in the general population, especially in rapidly developing countries in Asia.

The objectives of this study were to assess the exposure of the general populations in eight countries to a wide range of pesticides (Table S1) and to elucidate the determinants of exposure and potential risks. The urinary metabolites of 11 compounds (Table S1) referred to as “universal pesticides” by the CDC (2013) were analyzed in 322 urine samples collected from the general populations in eight countries between 2006 and 2014 (Table S2). By investigating the geographic distribution in concentrations and profiles of pesticide exposure, we establish baseline levels needed for determining future trends in exposures. On the basis of the concentrations measured in urine, daily exposure doses to pesticides were calculated.

2. Materials and methods

2.1. Standards and reagents

Information regarding standards and reagents used in this study is shown in the Supplementary Information.

2.2. Sample collection

Spot urine samples (n = 322) were collected from eight countries: the USA (n = 35; number of samples of females/males/unknown gender: 20/15/0), Greece (n = 40; 20/20/0), China (n = 86; 30/25/31), India (n = 35; 3/11/21), Saudi Arabia (n = 35; 14/21/0), Japan (n = 34; 26/8/0), Korea (n = 35; 17/9/9), and Vietnam (n = 22; 11/11/0) (Table S2). The samples originated from 13 cities in eight countries i.e., Albany (New York State), Athens, Shanghai/Guangzhou/Qiqihar, Mettupalayam, Jeddah, Ehime/Kumamoto, Seoul/Busan/Yeosu, and Hanoi. Samples from Korea were collected during 2006–2007, whereas the rest of the samples were collected during 2010–2014. The age of the donors ranged from 1 to 83 years (Table S2). Donors were provided with clean 50-mL polypropylene (PP) containers, and urine was collected directly into these containers. Information regarding age and gender was also collected. These samples were collected as a part of previous studies conducted 4-8 years ago and were archived in a freezer and were de-identified. Institutional Review Board approval was obtained from the New York State Department of Health for the analysis of urine. All samples were kept at −20° C until analysis.

2.3. Sample preparation and analysis

Details of analytical standards, internal standards, and reagents used in the study are presented in the Supplementary Information. The target pesticides were analyzed in urine samples after enzymatic deconjugation, followed by solid phase extraction, similar to that described by the CDC (2013). Briefly, 0.5 mL of urine was transferred into a 15-mL PP tube, and 1 ng each of labeled internal standard mixture (corresponding labeled internal standards were available for nine target compounds, except for cis-DCCA and cis-DBCA) was added. The urine samples were mixed with 400 μL of 0.2 M sodium acetate buffer, containing 745 units/mL of β-glucuronidase and 56 units/mL of sulfatase, and were incubated at 37 °C for at least 6 h. The samples were then passed through Oasis^® HLB 3 mL solid phase extraction cartridges that were conditioned with acetone (2 mL) and 1% acetic acid in water (2 mL). After loading samples, cartridges were washed with acetic acid/methanol/water (1:5:94, v/v/v) and vacuum dried for 5 min. Analytes were recovered by elution with 3 mL of acetone and 3 mL of hexane. Combined eluates were evaporated to dryness under nitrogen and re-dissolved in 100 μL of acetonitrile:water (1:1, v/v) and transferred into glass vial inserts for instrumental analysis.

Chromatographic separation was performed with a Waters ACQUITY Class I HPLC system (Waters, Milford, MA, USA) using a Betasil C18 column (100 × 2.1 mm, 5 μm; Thermo Electron Corp., Waltham, MA, USA), serially connected to a Javelin Betasil C18 guard column (20 × 2.1 mm, 5 μm; Thermo Electron Corp.). The injection volume was 3 μL, and the mobile phase flow rate was 350 μL/min. The mobile phase comprised HPLC-grade water with 0.2% (v/v) acetic acid (A) and acetonitrile (B). The gradient started with 95% A, held for 1 min, and linearly decreased to 50% within 1 min, and then to 5% in 1 min and held for 3 min. The mobile phase composition was reverted to initial conditions in 2 min, followed by an equilibration time of 2 min, for a total run time of 10 min.

Mass spectrometric analysis was performed on an Applied Biosystems API 5500 electrospray triple quadrupole mass spectrometer (ESI-MS/MS; Applied Biosystems, Foster City, CA, USA). The turbo ion spray source settings in the positive ionization mode were: curtain gas (CUR) 45 psi; collision gas (CAD) 6 psi; source temperature 450 °C; ion source gas 1 (GS1) 60 psi; ion source gas 2 (GS2) 65 psi; and ion spray voltage 4500 V. For the negative ionization mode, the mass spectrometric parameters were similar except that the ion spray voltage was −4500 V. Nitrogen was used as CUR and CAD. Further details of instrumental analysis and compound specific parameters are shown in Table S3.

2.4. Quality assurance and quality control

Analytes were quantified using an internal standard method (Table S3). The most abundant precursor/product ion transition was used for quantification. An 11- to 16-point standard calibration curve, with concentrations ranging from 0.01 to 200 ng/mL, was used for the quantification of the target analytes. The regression coefficients (r) were ≥0.99 for all calibration curves. The method limits of quantitation (MLOQs) were determined based on the lowest point of the calibration standard with a signal-to-noise ratio of ≥10, volume of urine taken for analysis, and the sample concentration factor (Table S4). As a check for instrumental drift in response factors, a midpoint calibration standard was injected after every 20 samples. Sample-to-sample carryover of target analytes was monitored by the injection of a pure solvent (acetonitrile) after every 10 samples. For each batch of 20 samples analyzed (within one day), one procedural blank was analyzed simultaneously to determine contamination arising from the laboratory materials and solvents. For every 20 samples, three pre-extraction matrix spikes (in urine) at 1, 5, and 25 ng/mL and one water spike at 10 ng/mL were prepared and passed through the entire analytical procedure.

2.5. Method performance

Typical chromatograms of standards and samples are shown in Fig. S1. The MLOQs for the 11 target analytes ranged from 0.002 to 0.019 ng/mL (Table S4). Trace concentrations of target compounds (0.016–0.607 ng/mL) were found in procedural blanks, which were subtracted from concentrations measured in urine samples. Relative recoveries of target chemicals were calculated based on the ratio of the analyte signal to that of the corresponding internal standard in spiked water and urine samples. The relative recoveries of all target compounds were in the ranges of 94.6–114% and 84.0–114% in water and urine, respectively (Table S4). The relative standard deviation (RSD) of repeated analysis of fortified samples was <15% (Table S4).

2.6. Data analysis

Data were analyzed using SPSS 19.0 (SPSS Inc., Chicago, IL, USA). The concentrations below the MLOQ were assigned a value of MLOQ divided by the square root of 2 and log-transformed (χ + 1) for further analysis. Creatinine was analyzed in urine, and creatinine-adjusted values were provided, as appropriate. Differences in urinary levels of chemicals among the eight countries were examined by one-way ANOVA if the data followed a normal distribution; otherwise, a non-parametric Kruskal-Wallis H test was applied. Values of p < 0.05 denoted statistical significance.

3. Results and discussion

3.1. Total pesticide concentrations

Pesticides were found in all urine samples from the eight countries, except for 2,4,5-T, which was not detected in any sample from Vietnam (Table S5). The concentrations of ∑11 pesticides differed significantly among the eight countries (one-way ANOVA, p < 0.001) in the decreasing order of Vietnam (median, 28.9 ng/mL), India (14.2 ng/mL), China (13.6 ng/mL), Korea (12.5 ng/mL), Greece (12.3 ng/mL), Saudi Arabia (11.3 ng/mL), the USA (7.9 ng/mL), and Japan (7.1 ng/mL) (Fig. 1; Table 1). The median concentration of ∑11 pesticides found in urine samples from Vietnam was approximately 4 times higher than that from Japan and the USA (p < 0.001). The highest concentration of ∑11 pesticides of 302 ng/mL was also found in a sample from Vietnam. Several urine samples from China and India showed elevated concentrations of pesticides (Fig. 1). No significant difference in the urinary concentrations of ∑11 pesticides was found among samples from India, China, Korea, Greece, and Saudi Arabia (post hoc test; p > 0.05).

Fig. 1.

Concentrations (ng/mL and creatinine-adjusted in μg/g) of urinary pesticides (∑11) from eight countries. The vertical lines represent the minimum, 50th percentile, and maximum, and the boxes represent the 25th and 75th percentiles. Standard and extreme outliers are denoted by circles and stars, respectively, indicating values greater than 1.5 and 3 times the interquartile range away from the 25th or 75th percentiles. Extreme outlier concentrations of “∑11” pesticides from Vietnam (at 302 ng/mL or 189 μg/g creatinine), and India (at 371 μg/g creatinine) are not shown in the figure.

Table 1.

Concentrations (ng/mL and creatinine-adjusted in μg/g) of urinary pesticides in eight countries.

		IMPY	MDA	PNP	TCPY	2,4-D	2,4,5-T	3-PBA	4F-3PBA	trans- DCCA	cis-DCCA	cis-DBCA	∑11
Country
USA (n = 35)	Median	0.3	0.4	1.6	1.4	0.3	0.02	0.7	0.01	0.4	0.5	nda	7.9
	GMb	0.3	0.4	1.2	1.4	0.3	0.02	0.6	0.02	0.4	0.5	0.03	7.1
	Medianc	0.2	0.5	1.3	1.3	0.3	0.02	0.6	0.02	0.5	0.5	nd	8.9
	GM	0.3	0.5	1.3	1.4	0.4	0.02	0.7	0.02	0.4	0.5	0.03	7.5
Greece (n = 40)	Median	0.3	0.4	1.6	6.1	0.2	nd	0.5	0.01	0.6	0.8	0.02	12.3
	GM	0.3	0.4	1.6	5.5	0.3	nd	0.6	0.01	0.5	0.6	0.1	12.2
	Median	0.3	0.3	1.2	4.8	0.2	nd	0.5	0.01	0.4	0.6	0.03	10.6
	GM	0.3	0.3	1.4	4.9	0.3	nd	0.5	0.01	0.4	0.6	0.1	10.9
China (n = 86)	Median	0.2	0.6	5.0	3.4	0.4	nd	0.6	nd	0.5	0.4	nd	13.6
	GM	0.2	0.5	5.0	3.7	0.4	0.003	0.7	0.004	0.6	0.4	0.02	14.2
	Median	0.2	0.6	5.4	4.2	0.4	nd	0.7	nd	0.5	0.4	nd	15.4
	GM	0.2	0.6	5.8	4.2	0.4	0.003	0.8	0.004	0.6	0.4	0.02	16.5
India (n = 35)	Median	0.2	0.2	4.7	4.3	0.3	0.01	0.9	0.01	0.9	0.6	0.03	14.2
	GM	0.2	0.2	3.7	4.4	0.2	0.01	0.7	0.01	0.8	0.5	0.04	13.3
	Median	0.4	0.5	9.0	9.7	0.5	0.03	1.4	0.01	1.7	1.2	0.07	27.4
	GM	0.4	0.5	9.1	10.9	0.5	0.03	1.8	0.02	2.1	1.3	0.1	32.9
Saudi Arabia (n = 35)	Median	1.1	0.2	1.7	1.5	0.1	0.01	0.6	0.02	0.9	0.5	0.05	11.3
	GM	1.3	0.3	2.0	1.7	0.1	0.01	0.7	0.02	0.7	0.6	0.04	10.9
	Median	1.0	0.2	1.4	1.2	0.1	0.01	0.6	0.01	0.6	0.5	0.03	9.2
	GM	1.0	0.2	1.6	1.4	0.1	0.01	0.5	0.02	0.6	0.5	0.03	9.0
Japan (n = 34)	Median	0.6	0.4	2.1	1.0	0.1	nd	0.5	nd	0.3	0.1	nd	7.1
	GM	0.5	0.5	2.1	1.0	0.1	nd	0.6	0.002	0.2	0.1	nd	6.5
	Median	0.4	0.4	1.7	0.7	0.1	nd	0.3	nd	0.2	0.1	nd	5.6
	GM	0.4	0.4	1.6	0.8	0.1	nd	0.4	0.002	0.2	0.1	nd	5.1
Korea (n = 35)	Median	0.9	0.4	2.8	3.3	0.2	0.003	1.0	0.004	0.8	0.5	0.07	12.5
	GM	0.9	0.4	3.2	3.3	0.2	0.01	1.0	0.01	0.6	0.4	0.1	12.5
	Median	1.0	0.4	3.9	3.6	0.2	0.01	1.0	0.01	0.6	0.4	0.06	13.6
	GM	1.0	0.4	3.5	3.5	0.2	0.01	1.1	0.01	0.7	0.4	0.1	13.5
Vietnam (n = 22)	Median	0.2	0.4	2.6	9.3	0.1	nd	2.1	nd	2.2	2.2	nd	28.9
	GM	0.2	0.5	2.6	9.3	0.05	nd	2.1	0.002	2.0	2.0	0.02	27.0
	Median	0.2	0.4	2.6	7.3	0.05	nd	1.9	nd	1.8	1.6	nd	19.6
	GM	0.2	0.4	2.3	8.3	0.04	nd	1.9	0.002	1.8	1.8	0.02	24.1
Gender
Male (n = 141)	Median	0.4	0.4	2.5	2.9	0.2	nd	0.6	0.004	0.5	0.4	nd	11.9
	GM	0.4	0.4	2.7	2.8	0.2	0.004	0.7	0.01	0.5	0.4	0.03	11.5
	Median	0.3	0.4	2.4	3.1	0.2	nd	0.6	0.004	0.4	0.4	nd	11.3
	GM	0.3	0.4	2.5	2.6	0.2	0.004	0.6	0.01	0.5	0.4	0.03	10.6
Female (n = 120)	Median	0.3	0.4	2.4	3.3	0.2	nd	1.0	0.01	0.8	0.6	0.01	13.1
	GM	0.4	0.4	2.4	3.1	0.2	0.01	0.9	0.01	0.7	0.6	0.03	12.3
	Median	0.4	0.4	2.4	4.0	0.2	nd	0.9	0.01	0.7	0.6	0.03	12.5
	GM	0.4	0.4	2.7	3.4	0.2	0.01	1.0	0.01	0.8	0.6	0.04	13.7
Age
≤20 (n = 27)	Median	0.4	0.3	2.3	1.6	0.2	0.02	0.5	0.01	0.4	0.3	nd	7.7
	GM	0.4	0.2	2.6	2.0	0.2	0.02	0.5	0.01	0.3	0.3	0.02	9.3
	Median	0.5	0.3	2.8	3.6	0.2	0.02	0.6	0.01	0.5	0.5	nd	15.6
	GM	0.6	0.3	3.5	2.7	0.2	0.02	0.7	0.01	0.4	0.3	0.03	12.7
21–49 (n = 173)	Median	0.4	0.5	3.0	3.2	0.2	nd	0.8	0.004	0.6	0.5	nd	12.3
	GM	0.4	0.5	2.9	2.9	0.2	0.004	0.8	0.01	0.6	0.4	0.03	11.9
	Median	0.3	0.5	2.7	3.0	0.2	nd	0.7	0.01	0.5	0.5	nd	11.5
	GM	0.4	0.4	2.7	2.8	0.2	0.004	0.7	0.01	0.5	0.4	0.03	11.4
≥50 (n = 60)	Median	0.3	0.3	1.7	3.1	0.2	nd	0.9	0.005	0.7	0.7	0.02	13.5
	GM	0.3	0.3	1.8	3.3	0.2	0.004	0.9	0.01	0.8	0.8	0.04	12.8
	Median	0.3	0.3	1.4	3.6	0.2	nd	0.8	0.01	0.6	0.7	0.03	11.8
	GM	0.3	0.3	1.9	3.4	0.2	0.005	0.9	0.01	0.8	0.8	0.04	13.2
All (n = 322)	Median	0.3	0.4	2.7	3.2	0.2	0.002	0.7	0.004	0.6	0.5	0.01	12.5
	GM	0.3	0.4	2.7	3.0	0.2	0.005	0.7	0.006	0.6	0.4	0.03	11.8
	Median	0.3	0.4	2.9	3.6	0.2	0.004	0.8	0.005	0.6	0.5	0.03	12.8
	GM	0.4	0.4	3.0	3.2	0.2	0.005	0.8	0.007	0.6	0.5	0.03	12.9

^and = not detectable;

^bGM = geometric mean;

^cbold italic: descriptive statistics of creatinine-adjusted concentrations (μg/g);

IMPY, 2-isopropyl-4-methyl-6-hydroxypyrimidine; MDA, malathion dicarboxylic acid; PNP, para-nitrophenol; TCPY, 3,5,6-trichloro-2-pyridinol; 2,4-D, 2,4-dichlorophenoxyacetic acid; 2,4,5-T, 2,4,5-trichlorophenoxyacetic acid; 3-PBA, 3-phenoxybenzoic acid; 4F-3PBA, 4-fluoro-3-phenoxybenzoic acid; trans/cis-DCCA, trans/cis-3-(2,2-dichlorovinyl)-2,2-dimethyl-cyclopropane-1-carboxylic acid; cis-DBCA, cis-3-(2,2-dibromovinyl)-2,2-dimethyl-cyclopropane-1-carboxylic acid.

Pesticides are widely used in many countries. A recent study showed that fruits and vegetables imported from Vietnam into European countries had high prevalence of exceeding maximum residue limits set for pesticides (Schreinemachers et al., 2017). Pesticide use in the USA accounts for 23% of the global consumption during 2008–2012 (EPA, 2017). However, the rate of pesticide use (kg/ha) during 2010–2014 in Japan was the greatest (18.9), followed by China (10.5), USA (3.9) and India (0.3) (Zhang, 2018; Sharma et al., 2014).

Urinary concentrations of pesticides were normalized for the creatinine content. Pearson correlation analysis showed a significant positive correlation between unadjusted and creatinine-adjusted concentrations of pesticides, regardless of individual pesticides or the sum concentration of 11 pesticides from the eight countries (p < 0.05), with the exception of the sum concentration of 11 pesticides from India (p = 0.052) (Table S6). Significant positive correlations were also found between concentrations of each pesticide and creatinine concentration in urine samples across the eight countries (r range, 0.704–0.820; p < 0.01). A significant positive correlation was also found between the sum concentration of 11 pesticides and creatinine-adjusted concentrations in the entire sample set (r = 0.609, p < 0.01). These findings suggest that the urine excretion volume at sampling (i.e., dilution factor) did not affect the concentrations of pesticides (Fig. 1). Further discussions on urinary pesticide concentrations were based on unadjusted values, unless specified otherwise.

3.2. Pesticide profiles

The overall distribution of pesticide concentrations in urine was similar among the eight countries studied (Fig. 2). The sum of four OPs, 2-isopropyl-4-methyl-6-hydroxypyrimidine (IMPY, metabolite of diazinon), malathion dicarboxylic acid (MDA, metabolite of malathion), para-nitrophenol (PNP, metabolite of parathion and methyl parathion), and 3,5,6-trichloro-2-pyridinol (TCPY, metabolite of chlorpyrifos and chlorpyrifos-methyl) (Table S1) collectively accounted for 62–77% of the total urinary pesticide concentrations. Pyrethroid metabolites accounted for 17–33% of the total pesticide concentrations. Such a pattern of higher OP concentrations than those of PYRs in urine has been reported for US and Australian preschool children (Babina et al., 2012; Calafat et al., 2017) and Spanish adults (Garí et al., 2018) (Table S7). In particular, PNP and TCPY were the dominant OPs (accounting for 43–72% of the total pesticide concentrations), especially in India (72%), China (69%), and Greece (66%), suggesting extensive use of parathion, methyl parathion, and chlorpyrifos in these countries. The use of parathion and methyl parathion is banned in several countries due to their high mammalian toxicity (Hoppin et al., 2017; Wu et al., 2018). The proportion of IMPY in total pesticide concentrations in urine from Japan, Korea, and Saudi Arabia (10–20%) was 3–18 times higher than those in the other countries, suggesting high exposure to diazinon, which is commonly used in indoor pest control (EPA, 2006). Although the use of OPs for indoor and garden pest control was banned in the USA and several European countries in 2000 (Barr et al., 2010; Roca et al., 2014), they are still commonly used in agriculture (Wang et al., 2016), accounting for 33% (20 million pounds) of all insecticides used in the USA in 2012 (EPA, 2017). Human exposure to OPs is still a cause for concern (Swan et al., 2003; Eskenazi et al., 2004; González-Alzaga et al., 2014).

Fig. 2.

Composition profile of urinary pesticides (to the sum ∑11 concentrations) from eight countries. IMPY, 2-isopropyl-4-methyl-6-hydroxypyrimidine; MDA, malathion dicarboxylic acid; PNP, para-nitrophenol; TCPY, 3,5,6-trichloro-2-pyridinol; 2,4-D, 2,4-dichlorophenoxyacetic acid; 2,4,5-T, 2,4,5-trichlorophenoxyacetic acid; 3-PBA, 3-phenoxybenzoic acid; 4F-3PBA, 4-fluoro-3-phenoxybenzoic acid; trans/cis-DCCA, trans/cis-3-(2,2-dichlorovinyl)-2,2-dimethyl-cyclopropane-1-carboxylic acid; cis-DBCA, cis-3-(2,2-dibromovinyl)-2,2-dimethyl-cyclopropane-1-carboxylic acid.

Trans- and cis-isomers of 3-(2,2-dichlorovinyl)-2,2-dimethyl-cyclopropane-1-carboxylic acid (trans/cis-DCCA, metabolites of cyfluthrin, cypermethrin and permethrin) were the most abundant pyrethroid insecticide metabolites (8–21% of the total pesticide concentrations), which were followed by 3-phenoxybenzoic acid (3-PBA, a common metabolite of 18 synthetic PYRs) (Barr et al., 2010) (accounting for 6–16% of the total pesticide concentrations), a pattern similar to what was reported previously (Saillenfait et al., 2015; Han et al., 2017). Urine samples from Japan contained the highest percentage of 3-PBA (16.3%), whereas the proportion of trans/cis-DCCA was relatively higher in urine samples from Vietnam, Saudi Arabia, and the USA (18–21%).

2,4-D accounted for only 0.3–6.6% of the total pesticide concentrations in urine from the eight countries. Other pesticides, 2,4,5-T, 4-fluoro-3-phenoxybenzoic acid (4F-3PBA, metabolite of cyfluthrin and flumethrin), and cis-3-(2,2-dibromovinyl)-2,2-dimethyl-cyclopropane-1-carboxylic acid (cis-DBCA, a specific metabolite of deltamethrin), accounted for less than 2% of the total concentrations in urine among the eight countries studied.

The measured concentrations of urinary pesticides in this study were similar to those reported in other countries previously (Table S7). This profile suggests that the populations in the eight countries are predominantly exposed to chlorpyrifos and parathion. Chlorpyrifos is the topmost OP (5–8 million pounds of active ingredient) used in the USA in 2012, whereas parathion is frequently used on fruits, vegetables, and nut crops (Yan et al., 2015; Wu et al., 2018). Diazinon, cyfluthrin, cypermethrin, and permethrin are the second most abundant group of compounds with their metabolites found in urine from the eight countries studied. The high correlation between 3-PBA and trans/cis-DCCA (r = 0.73–0.88, p < 0.01; Table S8) suggests that the major PYRs used in the eight countries were permethrin and cypermethrin, which is consistent with what was reported for the US (Barr et al., 2010) and Chinese populations earlier (Ding et al., 2012).

3.3. Gender and age differences in pesticides concentrations

Gender-related differences in urinary concentrations of 11 pesticides were examined for individual compounds and the sum of 11 pesticides across the eight countries. The median concentration of trans-DCCA in females (n = 120; 0.8 ng/mL and 0.7 μg/g creatinine) was significantly higher than that in males (n = 141; 0.5 ng/mL and 0.4 μg/g creatinine) (p < 0.05; Table 1). A similar gender-related difference in urinary trans-DCCA concentrations was found for the general US population (Barr et al., 2010). This may be related to females’ preference for vegetables/fruits, which were shown to contain cypermethrin, one of the top four most frequently detected pesticides (Quijano et al., 2016). However, no significant differences were found in the concentrations of other urinary pesticides between females and males (p > 0.05). The overall difference in sum concentrations of 11 pesticides between females and males (13.1 ng/mL and 12.5 μg/g creatinine vs. 11.9 ng/mL and 11.3 μg/g creatinine) was not significant when the entire data set from all eight countries was collectively analyzed (p > 0.05). Creatinine adjustment of urinary concentrations did not affect our observations on pesticide levels between genders.

We categorized the samples into three age groups, namely, ≤20, 21–49, and ≥50 years, for the examination of the relationship between age and urinary pesticide concentrations. Among the three age groups, for the sum of 11 urinary pesticides, no significant difference was found across the eight countries (p > 0.05), although an increasing trend in concentrations with age was observed (median: 7.7, 12.3, and 13.5 ng/mL, respectively; Table 1). When each pesticide was analyzed individually, we found significant differences in MDA, PNP, 2,4,5-T, and cis-DCCA concentrations among the three age groups (p < 0.05). The highest concentrations of MDA and PNP in urine samples were found in the age group of 21–49 years (n = 173), with median concentrations of 0.5 and 3.0 ng/mL, respectively, followed by ≤20 years (n = 27; median: 0.3 and 2.3 ng/mL), and ≥50 years (n = 60; median: 0.3 and 1.7 ng/mL). 2,4,5-T was detected in very few of samples (32.9%; Table S5) to allow for reliable evaluation of its distribution across three age groups. Urinary cis-DCCA concentrations were the highest in the age group of ≥50 years (median: 0.7 ng/mL and 0.7 μg/g creatinine), followed by 21–49 years (median: 0.5 ng/mL and 0.5 μg/g creatinine) and ≤20 years (median: 0.3 ng/mL and 0.5 μg/g creatinine). Children were reported to have significantly higher urinary concentrations of 3-PBA than did adolescents and adults in the USA (Barr et al., 2010). Such age-related variations in urinary pesticide concentrations may be attributed to the differences in dietary sources and metabolism (Barr et al., 2010; McKelvey et al., 2013; Oulhote and Bouchard, 2013; Saillenfait et al., 2015).

3.4. Location-specific variations

Differences in urinary concentrations of pesticides for samples collected from Guangzhou, Shanghai, and Qiqihar, China, were examined. For the entire data set, the highest sum concentration of 11 urinary pesticides was found in samples from Shanghai (n = 23, median: ng/mL and 17.9 μg/g creatinine), followed by Guangzhou (n = 25, median: 13.7 ng/mL and μg/g creatinine) and Qiqihar (n = 38, median: 11.6 ng/mL and 11.0 μg/g creatinine) (p < 0.05). TCPY and trans/cis-DCCA concentrations in urine samples from Shanghai (median: 7.7, 0.4 and 0.3 ng/mL) and Guangzhou (median: 5.1, 0.8 and 0.7 ng/mL) were significantly higher than those of Qiqihar samples (median: 2.5, 0.4 and 0.3 ng/mL) (p < 0.05). Urinary concentrations of 2,4-D in Qiqihar samples (median: 0.7 ng/mL and 0.9 μg/g creatinine) were relatively higher than those of Shanghai (median: 0.3 ng/mL and 0.2 μg/g creatinine) and Guangzhou samples (median: 0.2 ng/mL and 0.2 μg/g creatinine) (p < 0.05). Significant differences in pesticides metabolite concentrations and profiles also were reported among six Italian cities and across eight European cities (Rousis et al., 2017a; 2017b). This may inform location-specific application of pesticides.

3.5. Human exposure to pesticides

Urinary biomonitoring studies can be used to assess total daily exposure of environmental chemicals, including pesticides. Based on the high frequency of detection and concentrations, we estimated exposure to TCPY, PNP, IMPY, 3-PBA, and trans/cis-DCCA, as shown in eq. 1 (Guo et al., 2011; 2013) for the eight countries:

$$DI=CV\times \frac{M1}{M2}\times \frac{1}{f}$$ (1)

where DI is the daily intake of pesticide (μg/day), C is the median urinary pesticide metabolite concentration (ng/mL), V is the human daily excretion volume of urine (L/day; 24 h average urine excretion volume of 1.7 L/day for adults) (Guo et al., 2011), M1 and M2 are the respective molecular weights of parent pesticide and its metabolite (g/mol), and f is the ratio of metabolite excreted in urine relative to the total exposure dose of the parent compound (f = 1) (Glorennec et al., 2017; Garí et al., 2018). The f value of 1 is non-conservative, as only a small fraction is expected to be excreted in urine. Further, some of the pesticide metabolites analyzed are non-specific, as many related environmental chemicals can yield the same metabolite following biological transformation. Nevertheless, we assumed that TCPY was metabolized only from chlorpyrifos, PNP from parathion, 3-PBA, and trans/cis-DCCA from cypermethrin (one of the most commonly used PYRs) (Glorennec et al., 2017). Therefore, this exposure estimate should be considered a crude assessment.

The estimated daily exposure dose to pesticides by the populations in the eight countries are shown in Table 2. Vietnamese (59.7 μg/day) had the highest DI for total pesticides, followed by the populations in India, China, Greece, and Korea (30.6–38.4 μg/day). The DIs of total pesticides in Saudi Arabia, the USA, and Japan were in the range of 15.5–21.0 μg/day. The DIs of chlorpyrifos and parathion were 1.6–9.4 times higher than those of diazinon and cypermethrin. The DIs of chlorpyrifos were higher for populations in Vietnam, Greece, India, China, and Korea (≥9.6 μg/day) than those estimated for the other three countries (<5 μg/day). The DI values for parathion were higher for populations in China, India, and Korea (≥9.6 μg/day) than those for the other countries studied (5.7–9.3 μg/day). The DI of cypermethrin estimated for the Vietnamese population (21.8 μg/day) was much higher than that calculated for the other seven countries (3.0–8.1 μg/day). The US EPA’s provisional reference doses (RfDs) for chlorpyrifos (EPA, 2011), parathion (EPA, 2000), diazinon (EPA, 2006), and cypermethrin (EPA, 1998) are 0.0003, 0.006, 0.0002, and 0.01 mg/kg/day, respectively. Assuming an average body weight of 60 kg for adults (Honda et al., 2018), the estimated median DIs of these pesticides in the eight countries studied were below the RfDs by 0.6–6.0-fold for chlorpyrifos (EPA, 2011), 20–63-fold for parathion (EPA, 2000), 3.2–18-fold for diazinon (EPA, 2006), and 28–200-fold for cypermethrin (EPA, 1998). Nevertheless, 28% and 4.3% of the samples exceeded the RfDs of chlorpyrifos and diazinon, respectively. The estimated DIs of chlorpyrifos were above the chronic RfD values in 73% of the samples from Vietnam, followed in decreasing order by samples from Greece (53%), India (34%), China (33%), Korea (20%), Saudi Arabia (14%), and the USA and Japan (both 2.9%) (Fig. 3).

Table 2.

Median daily intake (DI) of pesticides estimated from urinary metabolite concentrations measured for eight countries (μg/day).

DI(μg/day)	Chlorpyrifos	Parathion	Diazinon	Cypermethrin	Totala
USA	4.2	5.7	1.0	5.4	16.3
Greece	18.3	5.7	1.0	6.4	31.4
China	10.2	17.8	0.7	5.0	33.7
India	12.9	16.7	0.7	8.1	38.4
Saudi Arabia	4.5	6.1	3.7	6.7	21.0
Japan	3.0	7.5	2.0	3.0	15.5
Korea	9.9	10.0	3.1	7.7	30.6
Vietnam	27.9	9.3	0.7	21.8	59.7
Allb	9.6	9.6	1.0	6.0	26.3

^aTotal refers to sum DIs of chlorpyrifos, parathion, diazinon and cypermethrin;

^bAll refers to DIs of pesticides estimated from urinary metabolite concentrations for the entire dataset from the eight countries.

Fig. 3.

Frequency distribution of daily intake (mg/kg/day) of chlorpyrifos in eight countries, estimated from urinary concentrations of 3,5,6-trichloro-2-pyridinol (TCPY).

4. Conclusions

This is the first study to describe exposure to pesticides in general populations of eight countries. Such an assessment provides insight into exposure patterns and doses of exposure. Nevertheless, our study has several limitations. Our sample size for individual countries is small. However, the study was intended to establish baseline concentrations, as the chronic exposure dose to pesticides in many of the countries studied here was not known previously. Spot urine sample may not represent exposures over time. Some of the pesticide metabolites analyzed are non-specific metabolites, and, therefore, exposure calculations may be overestimated (Ding et al., 2012). However, we assumed that 100% of the pesticide is excreted in urine (usually, this fraction can be much lower), which may result in underestimation of exposure. In general, this study has several merits of reporting exposure concentrations to pesticides in several countries for which no such information existed, and these exposure estimates will lay the foundation for designing future studies to understand potential risks from pesticide exposures. We elucidated the profiles and potential exposure doses of pesticides and highlighted the high exposures in Vietnam, India, China, Greece, and Korea.

Highlights.

• Organophosphate, pyrethroid and phenoxy acid pesticides were measured in urine.

• General populations in Asia are ubiquitously exposed to multiple pesticides.

• Vietnam had the highest sum concentrations of 11 pesticides (median: 28.9 ng/mL).

• Daily intake of chlorpyrifos was high (≥9.6 μg/day) in Asian countries.

• This is the first study to establish baseline levels of pesticide exposure in Asia.