Health Risk Assessment from Organophosphate Insecticides Health Risk Assessment from Organophosphate Insecticides Residues in Commonly Consumed Vegetables of Local Markets, Residues in Commonly Consumed Vegetables of Local Markets, Northern Thailand Northern Thailand

Background : Commonly consumed vegetables from local markets are considered at a higher risk for organophosphate insecticides (OPs) contamination. This study aimed to investigate the contamination of OP residues in commonly consumed vegetable samples from local markets in 5 provinces, i.e., Nan, Chiang Rai, Lamphun, Lampang, and Chiang Mai provinces located in Northern Thailand. A risk assessment of OPs among adult Thai population was also evaluated by calculating the estimated daily intake (EDI) and the acceptable daily intake (ADI). Method : The multi cross-sectional studies were conducted in 5 provinces. Twelve vegetables were sampled from main local markets in every district of sampling sides. All samples have analyzed the level of 19 OP residues using GC-FPD. The estimated daily intake and health risk of OP residues were estimated by using hazard index. Results : Twelve OP residues were detected in 42.8% of the samples, and 16.0% had residue levels that exceeded EU maximum residue levels. The most frequently detected OP residue in the vegetable samples was chlorpyrifos. The percentage of OP contaminated samples slightly increased throughout sampling time, 40.6%, 43.4%, and 47.9%, respectively. Hazard indices (HIs) were calculated based on data from National Health Examination Survey Of ﬁ ce and the result showed that HIs of diazinon residue in Pakchoi and dimethoate residue in cucumber were over 1, hence posing a serious threat to human health. Conclusion : The unacceptable levels of OP residues were observed in all types of vegetables collected from local markets, leading to hazardous exposure among consumers.


Health Risk Assessment from Organophosphate Insecticides Residues in Commonly Consumed
Vegetables of Local Markets, Northern Thailand 1. Introduction I n Thailand, the consumption of organic products is increasing because of the awareness of the dangers of contaminated foodstuffs [1]. However, farming in Thailand is still using chemical pesticides to protect crops [2]. In 2018, more than 18,000 tonnes of insecticides were imported into the country [3]. This number implies the high use of insecticides which might cause the accumulation in the environment and contamination in foods sold in domestic markets.
Organophosphate (OP) insecticides are insecticides with highly toxic and the most widely used insecticides available. OPs are considered a vital component of modern farming, playing a significant role in maintaining high agricultural productivity. OPs are widely used in Thailand because of their high efficiency against the target pest. However, unsuspecting use, uncontrolled and overuse of OPs may cause high levels of toxic residue. According to a previous study in Payao Province in Northern Thailand in 2014, OP residues such as diazinon, chlorpyrifos, and profenofos were detected in 12 commonly consumed vegetable samples [4]. The study in central Thailand in 2016 also found the residues of insecticides, including OPs, a synthetic pyrethroid, and carbamate in fruit and vegetable samples from local markets and supermarkets [5]. Another study in 2017 reported the OP residues in vegetable samples from the local fresh market and supermarket of Chiang Mai Province, northern Thailand. This study showed that the most commonly detected residue was chlorpyrifos in both fruits and vegetable samples [6].
OPs can inhibit the activity of cholinesterase, leading to overstimulation of neurotransmission. Exposure to OPs was associated with a range of adverse health effects, including neurological function disorder [7] and the male reproductive system [8]. Some studies reported that OPs exposure could be affected by childhood IQ [9] and brain anomalies [10]. Moreover, they can affect the endocrine system and prostate cancer [11]. The health effects of OPs can occur on occupationally exposed people, like farmworkers and applicators, and non-occupationally exposed people, as consumers. Although organic agriculture is currently being supported and controlled by the Thai government, chemical insecticide like OPs is still widely used in Thailand [6]. It might pose a health risk to Thai people. Therefore, information about the contamination status of OP residues is needed for food safety policy.
This study aimed to observe the contamination of OP residues in common consumable vegetable samples from local markets of 5 provinces in Northern Thailand and assess the health risk associated with OP residue in the vegetable.

Study design, study area, and samples collection
This study investigated OP residues in sampling common consumable vegetables from local markets in 5 provinces, i.e., Nan, Chiang Rai, Lamphun, Lampang, and Chiang Mai province located in Northern Thailand during 2014e2018.

Ethical issue
The study was approved by the Human Experimentation Committee (HEC) of Research Institute for Health Sciences, Chiang Mai University (Certificated number 64/2013, 56/2014-6.

Preparation of samples for organophosphate pesticide analyzing
One kilogram of each type of vegetable was purchased from the market (from one seller was one sample). All samples without washing were chopped into small pieces, and then the final processed sample weighed 500 g were randomly taken for analysis. The processing procedure was performed according to Codex protocol 21 [12] within a day after sampling. All samples were transferred in an icebox to the Laboratory of Environmental and Occupational Health Sciences, Research Institute for Health Sciences, Chiang Mai University, and then kept in a freezer À20 C before analysis.

Sample extraction and clean-up
The extraction and clean-up method used was followed the method from a previous study [4]. An aliquot of 5 g of sample was placed in a 50-mL centrifuge tube. Then, 250 mL of 5 mg/mL triphenylphosphate was added as an internal standard. Fifteen mL of acetonitrile were added twice, subsequently centrifuged for 5 min at 2500 rpm. The supernatant was transferred to a cleaned 50-mL centrifuge tube with the addition of 6 g of MgSO4 and 3 g of NaCl and then centrifuged again for 5 min at 2500 rpm for clean-up. The extract solution was filtered through Na2SO4 and evaporated to complete dryness using a vacuum rotary evaporator (Buchi, Switzerland) with a water bath at 30e35 C and then reconstituted with 1 mL of ethyl acetate three times. One mL of ethyl acetate phase was pipetted to dispersive solid-phase extraction tube and centrifuged for 3 min at 2000 rpm. Finally, the upper supernatant was filtered by a 0.2-mm syringe filter and then inject into GC-FPD for analysis.

Analysis of organophosphate pesticide residue in vegetable samples
Sample preparation and extraction steps have followed the method of Sapbumrer and Hongsibsong [4]. The GC analysis consisted of an Agilent 7890B equipped with a flame photometric detector (GC-FPD), a capillary column (DB-5MS, 0.25 mm I.D.,30 m length, 0.25 1 m film thickness (J&W column; Agilent Technologies, USA), and a computerized data handling system (Open Lab CDS; Agilent Technologies, USA).
The temperature was 220 C for the injection port (splitless mode). Temperature programming of the oven was as follows: initial temperature of 100 C, first ramp at 15 C/min to 180 C (2 min), second ramp at 5 C/min to 245 C (10 min), and final temperature maintained at 280 C for 4 min. The carrier gas was helium 99.999%.

Analytical quality assurance
Pooled vegetables were prepared and spiked with OP standard solutions for establishing the standard curve at concentrations 0, 20, 40, 80, 160, and 320 mg/ L. Quality control samples at three concentration levels at low (20 mg/L), medium (40 mg/L), and high (160 mg/L) OP concentrations were prepared by spiking working standard solutions into pooled vegetables for recovery, accuracy, and precision testing. All samples for the calibration curve and quality control were processed and analyzed with the same method as described above.
Limit of detection (LOD) and Limit of quantitation (LOQ) were defined as 3 and 10 times of standard deviation of the lowest concentrations, respectively [12]. Recoveries were calculated by comparing the concentrations of the extracted compounds with the actual concentration [13]. The method's precision was evaluated using different batches of quality control samples at 3 levels of the calibration curve. The method's precision was reported through the relative standard deviations (%RSD) with the acceptable range at below 20%. For every 15 samples, quality control samples prepared at 80 ng/mL were prepared and passed through the entire analytical procedure.

Health risk assessment
Health risk assessments were done based on OPs residues detected in a vegetable sample from the local market of study areas. The arithmetic mean of OPs residue in each vegetable was used to evaluate the risk from intake in the adult population. The weight 60 kg body weight was used as the bodyweight of the adult Thai population and 155.2 g vegetables/person as average daily vegetable consumption from the national data survey by the National Health Examination Survey Office (NHESO) [14]. The acceptable daily intake (ADI) was used from literature studies and estimated daily intake (EDI) was calculated for the pesticide residues and daily vegetables consumable.
The results of the hazard index (HI) were considered as a risk to human health; the HI more than 1 is considered as not safe for human health [15]: EDI ¼ estimated daily intake (mg/kg bw/day) : [(concentration of detected pesticide x food consumption)/body weight of an adult] ADI ¼ Acceptable daily intake (mg/kg bw/day) based on International Programme on Chemical Safety (IPCS)

Statistical analysis
The statistical analysis was conducted using SPSS software. Descriptive statistics, including frequency (n), percentage (%), arithmetic mean, and standard deviation (SD), were used to explain the characteristics of samples. The levels of detected OP residues in samples will be compared with established European commission regulation-maximum residue level [16], which set MRLs covering all types of vegetables in this study.

Results
This study was conducted in 5 provinces in the northern part of Thailand. Twelve types of commonly consumed vegetable samples were randomly purchased from local markets in those provinces. One thousand one hundred thirty-six samples were analyzed for OP residues between 2014 and 2018 ( Table 1) at three different crosssectional samplings. Nineteen OP residues were analyzed in vegetable samples using GC-FPD with the same analytical method, validated for recovery, linearity, precision, and accuracy ( Table 2). Limit of detection (LOD) and the limit of quantification (LOQ) for all OP residues ranged from 1.0 to 3.0 and 2.0e6.0 mg/kg, respectively. For the three levels of spiked standards in pooled samples, the recovery values ranged from 82.47% to 106.53%. The calibration curves for all the pesticides analyzed were linear with R 2 ¼ 0.999.
Four hundred and eighty-six (42.8%) out of 1136 samples were detected at least 1 OP residue. Approximately thirty percent were detected 1 OP residue. The highest number of detected residues in one sample is 6 (Fig. 1). The percent of OP contaminated samples slightly increased throughout sampling time, 40.6%, 43.4%, and 47.9%, respectively. The maximum and minimum contamination of OP residues were found in Lamphun (58.4%) and Nan (34.3%), respectively. The data was shown in Table 1.
Twelve out of 19 OP residues, including chlorpyrifos, diazinon, dimethoate, ethion, fenitrothion, malathion, methidathion, parathion-methyl, pirimiphos-methyl, profenophos, prothiophos, and triazophos, were detected in samples of this study (Table 3). OP residues were detected in samples of vegetables from every sampling market. The most frequently detected OP residues in the vegetable samples among all sampling sites were chlorpyrifos, diazinon, profenophos, dimethoate, and ethion. These residues were also the most frequently detected residues above the MRLs. However, only dimethoate shows a significant decrease of the median concentrations across the sampling times (data not shown). Table 4 shows the results of OP contamination among 12 commonly consumed vegetables in three samplings. Coriander was the crop with the highest percentages of residue-detected samples (74%), while morning glory was the lowest residue-detected crop (21%). However, coriander was the crop with the lowest percentages of MRL-exceeded samples (5%), while chili was the highest MRL-exceeded crop (25%) (data not shown). In each type of vegetable, there was more than one OP-residue detected. Chlorpyrifos was the most frequently detected residue in three sampling periods. Other 11 detected OP residues had a low percentage of detection. Only the concentration of dimethoate contamination was significantly decreased while other residues slightly changed but not significantly. There was one sample that detected the varying concentration of ethion in the second sampling batch. Chlorpyrifos and ethion were detected in all 12 types of vegetables, followed by diazinon, dimethoate, and malathion. The vegetable samples in this study were mainly leafy vegetables except chili, cucumber and yard long bean. However, there was no obvious difference of OP contamination among vegetable characteristics (Fig. 2). Table 5 showed the results of assessing the degree of risk that may arise from the residue of organophosphate pesticides in commonly consumed vegetables and widely available vegetables in fresh markets. The risk assessment was calculated using EDI and ADI. EDI compared to the average daily showed the estimated daily intake values of the OPs residues and health risk index in the commonly consumed vegetable samples. They are showed EDI and HI had no risk to health except diazinon residue in Pakchoi and dimethoate residue in cucumber with HI over 1 (see Table 6).

Discussions
This study shows the results of contamination status of OP residues in commonly consumed vegetable samples in Northern Thailand during 2014e2018 from three cross-sectional samplings. We sampled similar types of vegetables in each sampling time.
In our study, 19 OP residues were analyzed with good sensitivity using a multi-residue analysis method validated by evaluating recovery, linearity, and precision. Twelve out of 19 OP residues, including chlorpyrifos, diazinon, dimethoate, ethion, malathion, methidathion, parathionmethyl, pirimiphos-methyl, profenophos, prothiophos, and triazophos, were detected in samples of this study. These pesticides were in the lists of Thailand imported pesticides in 2018 except methidathion and parathion-methyl, which were in the watched and banned lists, respectively. Other residues which were not detected in this study were in the watched list, including dicrotophos and EPN and banned lists including azinphos-ethyl  [17]. However, chlorpyrifos will be decided for banning in the next year. This result may indicate that the success of the pesticide ban policy, which was set by the legal body, such as the Department of Agriculture (DOA), in Thailand [18]. OP residues were detected in all 12 types of vegetables over the year 2014e2018 except broccoli. However, only six out of twelve, including chili, Chinese cabbage, cucumber, green onion, kale, and morning glory, were detected OP residues that exceeded the MRLs. The percentages of contamination exceeded the MRLs in coriander, cucumber, Pakchoi, and yard long bean was decreased across the different periods while those in green onion, kale, and morning glory were increased. This result shows the different trends of OP use in the different crops. The percentage of OP residue detection in each type of vegetable sample in this study was similar to previously reported residue [19]. These mean OPs were used in crops production for a long time, and still, residue in vegetables sold in market and consumers always had in their body. However, the unacceptable levels of OP residues found in this study imply unsafe consumption for consumers. This result is also related to other previous studies that reported the high exposure to OP insecticides among farmers and consumers living in Northern Thailand [6,20]. From the result, in every sampling batch which from different provinces, there were samples which had OP residues exceeded MRLs. This might be caused by the intensive use of OPs in all farming areas.
Considering the contaminated level of OP residues detected in vegetable samples, the ranges of median concentrations in each type of vegetables were broad. Cabbage had found the highest median concentration of chlorpyrifos and the highest percentage of MRL-exceeded samples. The high median concentrations of other commonly found OP residues such as diazinon, dimethoate, and ethion were found in cabbage, cauliflower, and Chinese cabbage. Although the highest OP residue detection was found in coriander, the low number of samples was detected residue above MRLs because of the high MRLs set for coriander (Table 6).
According to sampling sites and sampling period, a significant difference was not found. This may imply that OP residue use in every area in Northern Thailand was not varied. All detected OP residues are currently permitted for agricultural use.
Reasonably, there was no change of OP use observed over the period. Although there is a plan for banning chlorpyrifos soon, it is still available in the commercial market.
According to sampling sites and sampling period, the significant difference was not found. This may imply that the patterns of OP residue use in every area in Northern Thailand were not varied. All detected OP residues are currently permitted for  Among the 19 OPs, there were the unacceptable health risk due to diazinon and dimethoate highly contaminated Pakchoi and cucumber samples, respectively [5]. Both diazinon and dimethoate can pose health risk to human even at low levels, and the continuous intake could cause health hazards in the long term [21]. However, EDIs which were calculated based on consumption rate and average body weight of population showed that individual EDIs were mostly lower than ADIs, thus no risk by calculation should be concerned. However, we calculated EDIs and HIs individually. The cumulative risk of pesticides should be more studied.
The limitation of this study is the different sampling sites and periods. This led to the difficulty of OP contamination comparison among sites and time. However, in all sampling areas, the types of vegetables selected were similar, and all sampling sites were in Northern Thailand. They shared similar patterns of crop cultivation. Food consumption is also one of the weak points of this study, and we had no data specific to people of the Northern part of Thailand. Thus, we carefully analyzed and discussed the results considering the mentioned limitation.

Conclusions
This study reported the contamination status of OP residues in commonly consumed vegetables monitored in local markets in Northern Thailand. The results show similar patterns of OP use among 5 provinces and different years. The unacceptable level of OP residues was observed in all types of vegetable samples collected from local markets, leading to hazardous exposure among consumers in these areas.
Residues of diazinon and dimethoate detected in the vegetable samples showed a potential threat to human health in the adult Thai population. It is recommended to continue monitoring of OPs residue in commonly consumed vegetables in local markets. The result from this study showed the high rate of detection of OP residues in vegetables samples although estimated health risks of consumption of vegetables contaminated with OP residues were mostly in acceptable range. We suggested that consumers should always wash fruits and vegetables before eating or cooking appropriately. At the same time, the policy on agricultural pesticide management and practical education program for farmers and consumers should be implemented to protect people's health.