Introduction

Pesticides are used in agriculture to control pests, diseases, and weeds and also for the post-harvest treatment of agricultural commodities. Insecticides, fungicides, herbicides, rodenticides, and plant growth regulators are typical examples of these chemicals (Townson 1992; Nicolopoulou-Stamati et al. 2016). The intensive use of pesticides in fruits and vegetables has resulted in dumping of residues both in the fruits and vegetables and also in the surrounding environment which happens due to unintended and overdose application without following pre-harvest intervals (PHI). These residues may cause a potential risk for human health the residue when dumping exceeds MRL (Fenik et al. 2011).

Fruits and vegetables are good sources of vitamins, minerals, fiber, and antioxidants. Despite this fact, these products can also be a source of toxic substances, i.e., pesticide residues (FAO 2007; Szpyrka et al. 2014; Mermer et al. 2020). Although pesticides provide mankind with many benefits, it poses risks to human health such as neurobehavioral and neuropsychological disorders, respiratory symptoms or diseases, and sperm DNA damage because of widespread use and the subsequent high biological activity (Scott et al. 2009) (Cooper and Dobson 2007). Many of the pesticides have been associated with health and environmental issues (Hayes et al. 2006; Sanborn et al. 2007; Mnif et al. 2011; Pimentel and Burgess 2014; Simon-Delso et al. 2015).

The pesticide use and application in fruits and vegetables are strictly regulated in many countries including Turkey to ensure the food quality. In addition, many countries carry out regular monitoring programs to assess pesticide residues, and such monitoring campaigns are conducted in accredited laboratories which routinely monitor the amount of pesticide residues in the environment (Turgut et al. 2010).

Worldwide, fruit orchards and vegetable fields cover 50 and 62 million ha of land, respectively (FAO 2018), whereas these areas cover 1.12 and 1.124 million ha, respectively, in Turkey (TUIK 2018) According to Turkish statistics data, 17.5 million tons of fruits and 29.5 million tons of vegetables were produced in 2015 in Turkey (TUIK 2018). Meanwhile, a number of pesticides have been applied either on the aerial portion of the crop or under the soil. The persistence of the residue in the produced fruits and vegetables depends on the surrounding conditions and crop culture. Kumari and John (2019) recently surveyed crop samples from the market and also from farmers and reported that more pesticide residue were observed in the samples obtained from the farmers.

Turkey is the fourth largest vegetable producer in the world after China, India, and the USA, and yet, more than 22 million tons of vegetables were produced in 2013 (Sector Report 2017). Production of fruits in 2013 amounted approx. 21 million tons in Turkey, and it placed as the fifth largest fruit producer in the world after China, India, Brazil, and the USA (Sector Report 2017).

In Turkey, farms located in the Aegean region produce substantial amount of fruits and vegetables which are also exported to the neighboring countries especially to those of the Europe (Anonymous 2018). The risk assessment of pesticide residues in the harvest or environment is necessary to ensure secure human health through maintaining food quality. In this manner, the aim of the current study is to investigate pesticide residues in fruits and vegetables samples collected between 2012 and 2016 from the Aegean region and to conduct a health risk assessment to human based on exposure to detected residue concentrations determined in the samples.

Materials and methods

Chemicals and reagents

All solvents were in chromatographic purity. Analytical standards for pesticide analysis were obtained from Supelco (Bellenfonte, PA, USA), Labinstruments (Castellana Grotte BA, Italy), Chem-Lab (Zedelgem, Belgium), and Dr. Ehrenstorfer GmbH (Ausburg, Germany). The purity degree of each analytical standard is given in Supporting Information Table S1.

Sample collection and sample preparation

Although the variety of farm-grown fruits and vegetables is much more in the Aegean region, in the current study, dried fruits (n=1149) including dried fig, dried apricot, and fresh fruits (n=1843) including strawberry, grapes, peach, cherry, sour cherry, pear, quince, apple, apricot, tangerine, pomegranate, nectarine, olives, and vegetables (n=52) including green pepper, cucumber, tomato, and pepper (green and red) were subjected to pesticide residue screening. Number of samples for each commodity is given in Table 1.

Table 1 Pesticide residues observed in different commodities
Full size table

Samples were collected from wholesalers between 2012 and 2016 and brought to the laboratory for analysis. The analyses were carried out as soon as possible after the sample collection. In the case of unavoidable delay, the samples were preserved following pre-shredding process and were kept at −20°C in freezer until sample extraction. A total of 3044 samples were analyzed for 395 numbers of pesticides routinely.

All samples were processed and analyzed following QuEChERS method which was easy, cheaper, effective, rugged, safe, and extensively used for pesticide residue analysis and also for validation of the pesticide residues (Anastassiades et al. 2003). Briefly, 1 kg of vegetable or fruit sample was homogenized for analysis, and a 10 g homogenized sub-sample of was transferred into a 50-mL centrifuge tube. A total of 10 mL of acetonitrile:citric acid (99:1 v:v ratio) was added on to the sample and was vortexed for 5 min at 4000 rpm. After vortexing, 4 g MgSO4, 1 g NaCl, 1 g trisodium citrate dehydrate (C6H9Na3O9), and 0.5 g disodium hydrogen citrate sesquihydrate (C12H18Na4O17) were added and shaken immediately for 1 min. The mixture was further mixed by vortexing for 1 min at 4000 rpm. The supernatant was collected and transferred into a 15-mL tube containing 150 mg MgSO4, 25 mg primary secondary amine (PSA) which then vortexed again for 1 min. Finally, the upper layer (1 mL) was filtered using 0.25-μm PTFE filter and stored in 2-mL vials until analysis (Anastassiades et al. 2003).

Pesticide residues were analyzed on a Schimadzu liquid chromatography-tandem mass spectrometry (LC/MS-MS) and on a Schimadzu gas chromatography mass spectrometry (GC-MS). The pesticide results were confirmed in accordance with European Commission Guidelines (Pihlström 2011). LC-MS/MS analysis was performed using a C18 column (Purospher® STAR RP-18). The injection volume was 20 μL with a flow rate of 0.4 mL/min. The auto sampler temperature was set up at 5°C, and the column temperature was selected 40°C.

GC-MS analysis was performed on a BP5 column (30-m length, 0.25-mm diameter, and 0.25-μm film thickness) (Agilent, DB 5 MS) with an injector temperature of 270°C. The temperature program was initial temperature 70°C (2.0 min hold), ramped by 30°C min−1 up to 180°C, and then was increased by 10°C min−1 to 290°C and held for 15 min. Helium was used as the carrier gas at a flow rate of 1.0 mL min−1. Detector and injector were set at 300°C and 260°C, respectively. Samples were injected splitless (2-μL injection volume, split opened after 2.0 min). Other instrument conditions were transfer line temperature at 280°C, ion source at 230°C, and quadrupole at 150°C. Target peaks were accepted if the target/qualifying ion ratio was within 20% of standards. Calibration was based on the area given using external standards for MS calibration.

Quality assurance/quality control

List of the targeted pesticides and Limit of Quantification (LOQ, mg/kg) of each analyte is given in Supporting Information Table S1. Limit of quantification is the minimum concentration of the analyte that can be quantified with acceptable accuracy and precision (Sanco 2010).

Calculation of risk assessment

The average residue of pesticides in commodities as calculated using the following formula of Chen et al. (2011) and Poulsen et al. (2005)

$$ {C}_{p,f}=\frac{C_{\mathrm{avg},\mathrm{pos},p,f}x{N}_{\mathrm{pos},p,f}}{N_{p,f}} $$
(1)

where, Cp,f is the average content of pesticide p (mg kg−1 ) in a particular commodity f; Cavg,pos,p,f is the average content (mg kg−1) of pesticide p in commodity f with detected residues, and Np,f is the number of commodities analyzed for the pesticide.

Data base collecting the average consumption cluster from G diets for commodities in Turkey compiled by World Health Organization (WHO) was used to derive the food consumption data for calculations (WHO/GEMS/FOOD 2006).

The EDI (Estimated Daily intake) was calculated with the following formula of equation 2;

$$ {\mathrm{EDI}}_{p,f}={C}_{p,f}\times {K}_f $$
(2)

where the EDI is the estimated daily intake (μg/kg bw/day) for pesticide p, and Kf is the average consumption rate of that commodity (g−1 bw day−1 ). EDIp,f can be expressed as sum of EDIp,f values calculated for individual commodities or pesticides or for combinations of pesticides and commodities.

The long-term chronic dietary risk or hazard quotient (HQ) was calculated using EDI and ADI (acceptable daily intake) values (EFSA 2007). The HQ is expressed as follows (Chen et al. 2011)

$$ \mathrm{HQ}=\frac{\mathrm{EDI}}{\mathrm{ADI}} $$
(3)

HQ ≤ 1 indicates that adverse effects are not likely to occur and thus can be considered to have negligible hazard.

Another important concept related to HQ is hazard index (HI), and it is the sum of hazard quotients for substances that affect the same target organ or organ system.

$$ \mathrm{HI}=\sum \limits_{n=1}^i{\mathrm{HQ}}_n $$
(4)

If HI exceeds a value of 1, this could indicate an unacceptable health risk.

Results and discussion

Concentrations and residue patterns detected in samples

Pesticides are applied in crop plants to protect the crop from pest in the aim of obtaining higher yields. The dose and methods of application differ based on the principles adopted which again varies among the countries. Food safety rules implemented in an individual country has a direct control over the application principles of that individual countries. However, residual occurrence of the applied pesticides in the harvest is a must, but it should not exceed the approved limit.

In this study, a total of 3044 samples of the 16 commodities were analyzed for a total of 395 pesticides (Table S1 and Table 1).

Figure 1 shows the number of residues detected in each commodity. As it is seen in Fig. 1, grapes showed the maximum number of residues followed by strawberry, dried apricot, dried fig, peach, pomegranate, cherry, pepper, tomato, pear and nectarine, quince, olives, fresh apricot, apple, and tangerine.

Fig. 1
figure 1

Number of pesticide residues in each commodity type

Full size image

The overall results of pesticide residues in commodities are given in Fig. 2 and Table S2. This study identified 64 different pesticides in 3044 samples of vegetables, and fruits having a total of 1571 samples showed one or more pesticide residue. Out of 64 different pesticides, only 23 of them showed a DF greater than 1 (Fig. 2a), while 36 of pesticides showed a DF smaller than 1 (Fig. 2b). Among the pesticides with DF>1, azoxystrobin was the most frequently detected analyte (detected in 145 samples, DF=9.23%), while 13 chemicals showed a DF of 0.6% (detected in only 1 sample).

Fig. 2
figure 2

Detection frequency of pesticides with concentration >LOQ (a DF>1, b 1>DF>0)

Full size image

The concentration range of azoxystrobin, which is a broad spectrum systemic fungicide widely used in agriculture to protect plants from fungal diseases and the most frequently detected pesticide in the current study, showed a concentration range between 0.011 and 0.758 mg/kg. Other the most frequently detected pesticides were triadimenol (3.8%, in 112 samples) with a concentration range between 0.011 and 0.235 mg/kg, carbendazim (3.4%, in 103 samples) with a concentration range between 0.012 and 0.605 mg/kg, chlorpyrifos (3.2%, in 98 samples) with a concentration range between 0.006 and 0.187 mg/kg), pyrimethanil (3.10% in 94 samples) with a concentration range between 0.012 and 0.771 mg/kg), cyprodinil (2.9%, in 90 samples) with a concentration range between 0.011 and 0.778 mg/kg), fludioxonil (2.50%, in 76 samples) with a concentration range between 0.011 and 0.743 mg/kg), indoxacarb (2.40%, in 75 samples) with a concentration range between 0.010 and 0.086 mg/kg), imidacloprid (2.10%, in 66 samples) with a concentration range between 0.012 and 0.300 mg/kg), and boscalid (1.90%, in 60 samples) with a concentration range between 0.011 and 0.314 mg/kg). Detection frequencies of the rest of the target chemicals were lower than 1.99% in all of the commodities (Fig. 2; Table 1).

Among fruit samples, the most pesticide residues were found in peach (76.1%), pear (66.7%), cherry (52.2%), nectarine (50%), tomato (47.7%), grapes (43%),dried apricot (38.2%), fresh apricot (33.3%), strawberry (29.8%), quince (26.7%), olive (26.1%), apple (20%), pepper (12.5%), pomegranate (6.4%), dried fig (6%), and tangerine (3.6%). Among these results, commodities showed 5 or more residues were grapes (n=40), strawberries (n=4), dried fig (n=3), peach (n=2), and cherry (n=2).

The number of samples showing single pesticide residue or pesticide residue combination is given in Table 2. A total 42% of the samples with residues (352 out of 838) contained 2 or more residues, while 58% of the samples with residues (486 out of 838) contained a single residue.

Table 2 Individual commodity with single or multiple pesticide residues
Full size table

In the current study, of 3044 samples, 11.6% of the samples (a total of 354 samples; 179 fruits and 175 vegetable) exceeded MRL levels of 25 pesticides which was set by Turkish authorities (Table S2) (Anonymous 2016).

Results also showed that 11.6% of the samples showed multiple pesticide residues that were higher than the approved MRL values, whereas 15.5% of the samples (a total of 473 samples; 443 fruits and 30 vegetables) had pesticide concentration below the approved MRL values (Table 1).

In a study carried out in Poland (Szpyrka et al. 2014), it was observed that 66.5% of the collected apple samples had pesticide residues, and 3% exceeded maximum residue level (MRL). In the same study, multiple residues were present in 35% of the samples with two to six pesticides, and one sample contained even seven pesticide residue compounds.

As stated earlier, in the current study, 11.6% of the samples were found to exceed the approved MRL levels of analyzed pesticides (Table S2). Such finding agrees well with the report of Chen et al. (2011) where 3009 samples were analyzed and was reported that 11.7% samples contained pesticide residues above the approved MRL levels. In another Turkish study, it was reported that 8.4% of fruits contained pesticide residues above MRL level, and likewise, 9.8% of vegetables had residue above MRL level (Bakırcı et al. 2014). Practically, the occurrence of pesticide residues depended on the dose of application and degradation rate of the pesticide (Helbling et al. 2014). In a study in Argentina (Mac Loughlin et al. 2018), pesticides were detected in 65% of the total 135 samples. In that study, 44% of samples showed residue levels either equal or below the MRLs, whereas and 56% of samples showed residue levels higher than the MRL levels.

In Colombia, in a study carried out with tomatoes, 24 pesticides were analyzed, and one sample contained carbendazim with residue level exceeded the MRL level. At least one pesticide was detected in 70.5% of the samples, and out of the detected residues, the most detected residues were of pyrimethanil, carbendazim, dimethomorph, and acephate (Arias et al. 2014).

Previous studies of 2010–2012 using 1026 vegetable and fruit samples were analyzed and 36.6% of samples contained pesticide residues in Poland (Szpyrka et al. 2014). In a study conducted in Turkey in the same area of the current study, a total of 1423 samples were analyzed, and it was observed that 50% of samples contained detectable pesticide residues (Bakırcı et al. 2014). Compared with that study, current study showed almost doubled pesticide residues.

Risk assessment

For risk assessment of pesticide residues in each commodity, estimated daily intake (EDI) and HQ (hazard quotient) of each commodity was calculated and is given in Table S3. The EDI values ranged from 3.57×10−3 (tangerine) to 8.98 (tomato). The acceptable daily intake (ADI) values are given in Table S4, and the calculation of EDI and HQ from all commodities are also listed in Table S4. The HQ of many pesticides were close to zero, and they were indicated as <0.001 in the relevant table. The higher HQ values were 0.120 for chlorothalonil, 0.071 for famoxadone, 0.027 for pyridaben, 0.024 for chlorpyrifos, and 0.020 for indoxacarb.

In a study in Poland (Szpyrka et al. 2015), pesticide residues were found in 376 samples (36.6% of tested samples). In 18 samples (1.8% of the total), residues exceeded MRLs. In 28 (2.7%) samples, substances not recommended for a given crop were detected. The highest values of long-term exposure were found for dimethoate residue in apples (1.7% ADI, adults; 6.8% ADI, children). For most detected pesticides, long-term exposures were below the values of 1% ADI for adults and 3% ADI for children. The highest values of short-term exposure were obtained in the case of consumption of apples with azoxystrobin (4.5% acute reference dose (ARfD), adults; 13.3% ARfD, children).

A study on the health risk for children, adults, and the general population consuming apples with various pesticides was performed by Szpyrka et al. 2015. In that study, the pesticide residue data were combined with the consumption of apples in the 97.5 percentile and the mean diet. In certain cases, the total dietary pesticide intake calculated from the residue levels observed in apples exceeded the toxicological criteria. Children were the group most exposed to the pesticides, and the greatest short-term hazard stemmed from flusilazole at 624%, dimethoate at 312%, tebuconazole at 173%, and chlorpyrifos methyl and captan with 104% ARfD each. In the cumulative chronic exposure, among the 17 groups of compounds studied, organophosphate insecticides constituted 99% acceptable daily intake (ADI).

Bakırcı et al. (2014) reported chlorpyrifos as the most detected pesticides in fruits and vegetables. Azoxystrobin methoxyacrylate is a widely used fungicide which is a protectant, curative, eradicant translaminar with systemic properties effective to control pathogens in cereals, rice, vines, and potato. The detection frequency of azoxystrobin methoxyacrylate in the current study was 4.76%. This organophosphate insecticide works in a non-systemic mode of action when in contact of stomach and respiratory system. It is used to control a number of pests in more than 100 crops (Zhang et al. 2015). Triadimenol, which had a detection frequency of 3.67% in the current study, is a triazole fungicide with systemic mode of action used to control powdery mildews, rusts, gunts, and smuts in cereals along with many other crops (Anonymous 2013). Pyrimethanil, with a detection frequency of 3.08% in this study, is an anilinopyrimidine fungicide which is used to control grey molds in vines and fruits along with vegetables and ornamentals. It is also used to control leaf scab on pomegranate fruits (Nerya et al. 2016). Carbendazim which showed a detection frequency of 3.38% in the current study is a benzimidazole fungicide has systemic properties. It is widely used to control many pathogens in cereals, oilseed rape, tomato, pomegranate, and stone fruits (Singh et al. 2016). In a recent survey aiming to detect pesticide residues in Chile (Elgueta et al. 2019), it was revealed that the HQs and HIs for the pesticides evaluated in different models decreased in the order of methamidophos>lambda-cyhalothrin>chlorpyrifos, and it was concluded that from a food safety perspective, the investigated samples presented a greater health risk to consumers

The lower values of HQ were determined in dried apricot, grape, and strawberry with the HQ value of 0.01. and the other commodities had HQ values ranging between 0.025 and 0.27. In this study, the HQ value in 32 out of 62 pesticides tested was found to be close to 0 (Table S4) which for 17 pesticides were between 0.01 and 0.05. The values of 8 pesticides were in the range 0.006 to 0.027. Five pesticides with the highest HQ value were chlorothalonil (HQ=0.120), famoxadone (HQ=0.071), pyridaben (HQ=0.027), chlorprifos (HQ=0.024), and indoxacarb (HQ=0.020). Despite frequent occurrence of pesticide residues in vegetable and fruits, HQ values indicates no significant public health concern

The present study shows that despite the high occurrence rate of pesticide residues in fruits and vegetables from Xiamen City from 2006 to 2009 (Chen et al. 2011), the contamination level could not be considered a serious public health problem.

MRLs are established to protect consumers, and the exceedance of MRLs may lead to possible exceedance of the health-based guidance values in particular for younger people such as children and toddlers who are consuming more on a body weight basis. Additionally, as most of these pesticides belong to the same/similar chemical classes, they are expected to have similar modes of action and target organs. Therefore, combined exposure to these chemicals with similar health endpoint occurring in several commodities or co-occurring in the same sample can cause more severe negative effects. Similar to the current study, studies conducted in other regions of the world reported presence of multiple pesticide residues in fruits and vegetables (i.e., two or more of the OCPs pesticide residues (Al-Shamary et al. 2016)) or residues of multiple pesticides in commodities such as tomatoes, cucumber, bell peppers, and watermelons, and yet, the most frequent combinations of two pesticides detected in the same sample were deltamethrin and either imidacloprid or cypermethrin (Jallow et al. 2017). Therefore, an accurate regional exposure assessment to pesticide mixtures (WHO 2020) would be more particularly relevant, and results of the current study are an example of regional exposure assessment to pesticide residues.

Conclusion

Results of the current study shows that although there exist residue levels close or higher compared MRL levels of a given pesticide and yet the results are similar to those reported for European countries, pesticide residues do not pose a serious health risk to the public. Risk might occur due to pesticide residues detected in the commodities are found lower compared with those in Asian countries. Results of the present study shows that 27.5% of dried fruit, fruit, and vegetable samples collected from the Aegean region of Turkey contained at least one pesticide residue, and yet, 42% of the samples showed 2 or more pesticide residue.

According to the published literature, among the several issues regarding pesticide use in fruits and vegetables, the most significant one is the use of pesticides in many commodities and above the authorized MRLs as this could lead to consumer’s exposure exceeding the reference dose. Exceedance of MRLs was observed in 11.6% of the samples concerning carbendazim, chlorpyrifos, acetamiprid, thiophanate-methyl, propargite, thiacloprid, famoxadone, dichlorvos, cypermethrin, lambda cyhalothrin, clofentezine, α-cypermethrin, captan, deltamethrin, thiamethoxam, malathion, bromopropylate, α-endosülfan, β-cyfluthrin, dimethoate, permethrin, dinocap, lufenuron, procymidone, and tau-fluvalinate. Risk assessment for long-term exposure was done for all targeted chemicals detected. However, chlorothalonil (8%) and fanoxadone (7.1%) exposure to pesticides from dried fruits, fruits ,and vegetables was below 1% of the ADI. The current study was limited to only 16 types of commodities and 64 pesticides, which is insufficient to assess a total exposure to pesticides through food. Moreover, the study was conducted only in the Aegean region of Turkey. Therefore, to be able to integrate the results of the study, monitoring of more pesticide residues in a greater variety of crops and in a wider study area should be developed in order to guarantee food intake according to international food safety standards for the Turkish consumers. Additionally, an implementation and strict enforcement of existing regulations and international agreements is the key activity to protect populations; however, strengthening farmer education and knowledge transfer as well as encouraging the producers to implement good agricultural practices and integrated pesticide management would be the most important activity to improve the situation in a sustainable way. Results also call for improved pesticide application practices, measures to prevent consecutive application of same/similar active ingredient formulations, and education of the farmers to ensure application of pesticides at advised doses.