Pesticide Residues in Greenhouse Leafy Vegetables in Cold Seasons and Dietary Exposure Assessment for Consumers in Liaoning Province, Northeast China

Abstract

Excessive pesticide application in greenhouses leads to elevated levels of pesticide residues, especially in cold seasons, thereby posing a potential dietary exposure risk to the residents’ health. This study aimed to investigate 65 pesticides in 469 leafy vegetable samples collected from greenhouses in Liaoning province between October 2018 and May 2020. Overall, the pesticide levels in 96.4% (452/469) of the samples were below the MRLs established by the Codex Alimentarius Commission. The detection rate of 65 pesticides ranged from 0.2% to 11.9%. Multiple pesticide contamination was common, with dimethomorph being the most recurrent contaminant. The dietary risk assessment study revealed obvious differences in chronic and acute exposure risk values. The chronic risk quotients (RQcs) for leafy vegetable samples were 0.001%–3.993%, indicating an acceptable public health risk for the residents. As two leafy vegetables highly consumed in cold seasons in northeast China, the acute risk quotients (RQas) of Chinese cabbage samples were significantly lower than 100%; however, pakchoi samples exhibited RQa values ranging from 0.159% to 580.529%, showing an unacceptable acute dietary risk. This highlights that, compared to the chronic dietary risks, the potential acute dietary risks induced by the pesticides are higher in greenhouse-grown leafy vegetables during cold seasons.

1. Introduction

Vegetables are important for a healthy diet; they provide essential nutrients, including carbohydrates, multiple vitamins, and minerals to prevent diseases [1,2]. The increase in the population has a direct impact on the demand for global vegetable production, variety, and consumption, which have increased significantly [3]. Solar and plastic arch greenhouses have been used as protected facilities to create suitable environmental conditions in temperate climate zones in the cold seasons, where open-field planting is inappropriate for vegetable cultivation. These facilities improve the utilization of agricultural resources in the cold seasons to ensure the vegetable supply [4]. Compared with traditional vegetable cultivation, greenhouses can provide individuals with off-season vegetables while generating more employment opportunities and higher income [5]. Therefore, greenhouse vegetables have rapidly developed and become an essential pillar of the worldwide agricultural industry [6]. By the end of 2017, the global greenhouse vegetable planting area was about 3.91 million hm², and China’s total planting area accounted for about 80.4% of the world’s total [7]. Northeast China is the key development area of greenhouse vegetables with regional planting characteristics, where leafy vegetables, solanaceous vegetables, melon vegetables, and bulb vegetables are the main varieties of planted vegetables [8].

Compared with open-field vegetables, greenhouse vegetables, as a special ecosystem, are highly susceptible to diseases and pests because of the planting environment, the pathogens, and the host plant [9]. In agricultural production, pesticides are crucial to improve crop quality and quantity by actively preventing and controlling diseases, improving food growth efficiency, and promoting ripening [10,11]. Unfortunately, they are also the dominant route of pesticide exposure for the general population [12]. Exposure to pesticides may cause chronic or acute toxicity with harmful effects on human health. After application, some pesticides gradually persist in the environment, while others accumulate in the food chain [13,14]. Once absorbed through the human gastrointestinal tract, respiratory tract, and skin, pesticides can be distributed, stored, or accumulated in organs and tissues, which, when reaching a specific limit, can cause irreversible diseases of the human nervous, endocrine, urinary, and digestive systems, resulting in serious health risks [15,16,17,18]. International organizations and governments have set maximum residue limits (MRLs) for pesticides to ensure that residues are tolerable for humans.

Liaoning province, located in northeastern China, has a temperate continental monsoon climate, with an apparent temperature difference between morning and evening in cold seasons [19]. Compared with summer and autumn, the microenvironments of greenhouses with lower light, higher humidity, and no ventilation in winter and early spring are more likely to induce rapid occurrence and development of vegetable diseases and pests. Some farmers have resorted to indiscriminate use of pesticides, including single or multiple pesticide combinations, large and frequent applications, and applications throughout the growing season to reduce disease and insect damage to greenhouse vegetables. The microenvironments of greenhouses slow down the degradation of pesticides, which can lead to profound accumulation in vegetable plants. Compared to other vegetables, leafy vegetables have a significantly large leaf surface area, a thin waxy layer of cell epidermis, and an increased number of stomata on the epidermis. Consequently, leafy vegetables tend to readily absorb and translocate these chemicals, resulting in the accumulation of pesticide residues [20]. Jiang and Yao found that the pesticide detection rates of three leafy vegetables, including celery, spinach, and cabbage, were higher than those of other vegetables [21,22]. In an analysis of 58 pesticides in more than 15,000 vegetables in Liaoning province from 2013 to 2016, leafy vegetables accounted for 70% of the 10 vegetables with the lowest qualified rates [23]. Therefore, in temperate regions such as northeast China, pesticide residues in greenhouse vegetables, especially leafy vegetables, are not optimistic in the cold season and inevitably pose a particular risk to consumers’ diets. It is essential to monitor pesticide residues in leafy vegetables and assess their potential risk to human health.

Multi-residue methods integrate one sample preparation procedure with analytical equipment capable of identifying a wide range of compounds. At present, a multi-residue extraction procedure known as Quick, Easy, Cheap, Effective, Rugged, and Safe (QuEChERS) has gained extensive recognition and application. In the face of different classes of pesticides and complex agricultural sample matrices, ultrahigh-performance liquid chromatography (UPLC) coupled with tandem mass spectrometry (MS/MS) as well as gas chromatography (GC) coupled with MS/MS have emerged as prominent and effective analytical tools for the separation, identification, and quantitation of diverse compounds owing to their exceptional sensitivity, selectivity, and specificity. The above QuEChERS pretreatment and instrumental analysis methods have gained extensive utilization in the examination of pesticide residues in agricultural commodities, such as vegetables, fruits, and tea [21,24,25,26,27]. In this study, following GB 23200.113-2018 guidelines, the collected samples were pretreated by a QuEChERS-based method, followed by quantitative detection of 65 target pesticides using LC-MS/MS and GC-MS/MS through both internal standard and external standard methods.

In order to identify potential hazards in agricultural products and scientifically evaluate their health risks, researchers currently apply various risk assessment methods depending on specific research needs and data availability. These assessment methods determine the intake levels of chemical contaminants and the probability of adverse effects on human health based on the concentrations of pesticide residues in food, food consumption, relevant toxicological data, and exposure assessment models. In practical applications, dietary risk assessment can be categorized into chronic and acute dietary intake based on the time frame. It can also be classified into point assessment methods [28] and probabilistic assessment methods [29,30,31] according to quantitative assessment models. Furthermore, it can be conducted for a single pesticide residue or for cumulative risk assessment of a class of chemical substances with the same mechanism of action [32]. Currently, researchers have utilized hazard index (HI), relative potency factor (RPF), toxic equivalency factor (TEF), and cumulative risk index (CRI) in conjunction with either the point assessment method or probabilistic assessment method to evaluate the dietary risk exposure to pesticides in various agricultural products from many countries and regions [33,34,35,36,37,38]. However, to the best of our knowledge, there are few studies focusing on the dietary risk assessment of agricultural products susceptible to diseases and pests in specialized cultivation areas, such as pesticide residues in greenhouse leafy vegetables during cold seasons in temperate regions. According to the recommendation of the World Health Organization (WHO), the selection of exposure assessment methods should adhere to the distribution principle. This suggests that deterministic assessment methods (point assessment) with low data requirements, easy accessibility of data, and minimal time consumption should be preferred [39]. In this study, the number of samples is limited and the 65 pesticides monitored belong to different pest control categories with diverse mechanisms of action. Therefore, point assessmentmethod was employed to evaluate the dietary risk of each of 65 pesticides. More specifically, the objectives of the present study are as follows: (1) to determine concentrations of 65 pesticides in leafy vegetable samples using QuEChERS solid-phase extraction coupled with LC-MS/MS and GC-MS/MS; (2) to master the characteristics of pesticide residues, pesticides exceeding MRLs of the Codex, and mixed contamination; (3) to evaluate the risk of dietary exposure of the consumers based on pesticide toxicology and dietary consumption of the consumers through a deterministic assessment approach. The findings from this study would provide a scientific basis for ensuring the safe production of leafy vegetables in greenhouses and improving integrated pest management strategies in temperate regions. Simultaneously, it also offers a certain degree of reference value for greenhouse vegetable production in developing countries.

2. Materials and Methods

2.1. Sample Collection and Preparation

We collected a total of 469 leafy vegetable samples in October and December 2018, January and March 2019, and October and November 2020 in Liaoning province. Liaoning province experiences a temperate continental monsoon climate characterized by significant temperature fluctuations between morning and evening in winter and early spring [19]. The production of greenhouse vegetables reached 19.726 million tons, accounting for approximately 64% of the total vegetable production in Liaoning province [40]. The greenhouse vegetable production sites were located in eastern (Benxi and Dandong), southern (Liaoyang), western (Fuxin, Jinzhou, and Panjin), and northern (Shenyang, Tieling, and Fushun) Liaoning (Figure 1). Herein, nine categories of leafy vegetables commonly grown and consumed by residents of northeast China were collected as research subjects and picked before premarket sale. In each year of 2018, 2019, and 2020, nine kinds of vegetables were collected. These categories included Chinese cabbage (n = 109), pak choi cabbage (n = 108), chrysanthemum coronarium (C. coronarium) (n = 53), rape (n = 52), celery (n = 44), lettuce (n = 36), spinach (n = 29), leaf lettuce (n = 19), and chicory (n = 19) (

Table 1. The number of leafy vegetables samples.

Varieties	No. of Samples			Sum
	2018	2019	2020
Chinese cabbage	49	15	45	109
Pakchoi	41	49	18	108
C. coronarium	21	15	17	53
Rape	25	14	13	52
Celery	14	18	12	44
Lettuce	20	9	7	36
Spinach	12	4	13	29
Leaf lettuce	9	4	6	19
Chicory	6	6	7	19

). According to the Chinese Agricultural Standard NY/T 789-2004 [41], we removed rotten stems and leaves from each sample weighing ≥ 3 kg and containing at least 4–12 individuals. The samples were then packed into individual black polyethylene bags labeled accordingly, and transported directly to the laboratory, where they were transferred into polyethylene boxes after homogenization by the quarter method and stored in a refrigerator at −20 °C until further analysis within 30 days.

2.2. Chemicals and Materials

The pest control of greenhouse vegetables in northeast China was mainly based on chemical pesticides, supplemented by biological pesticides and inorganic compounds. The pesticides focused on in this study mainly included the following: pesticides registered in the production of high-yield vegetables and high toxicity in the “China Pesticide Information Network”, pesticides banned and restricted for use in vegetables as recommended by the Ministry of Agriculture and Rural Affairs of the People’s Republic of China, pesticides applied by farmers in greenhouse vegetables production research in Liaoning province, and pesticides listed in the “Agricultural Products Quality and Safety Supervision Program” of Liaoning province 2013–2016 [23]. This study focused only on the active ingredients of pesticides frequently used in greenhouses, including insecticides, fungicides, plant growth regulators, acaricides, and herbicides, excluding biological pesticides, rodenticides, insect attractants, and other organic and inorganic compounds. Ultimately, 65 pesticides were identified for analysis in this study.

Sixty-five pesticide reference standards (1000 mg/L) and the isotope-labeled internal standard heptachloride B (100 mg/L) were purchased from the Ago Environmental Protection Institute, Ministry of Agriculture and Rural Affairs (Tianjin, China). Analytical grade magnesium sulfate (MgSO₄), sodium chloride (NaCl), sodium citrate (Na₃C₆H₅O₇), and sodium citrate dibasic sesquihydrate (C₁₂H₁₈Na₄O₁₇) were acquired from Concord (Tianjin, China). Chromatographic-grade acetonitrile (CH₃CN), acetic ether (CH₃COOC₂H₅), ammonium acetate (NH₄CH₃CO₂), and formic acid (HCOOH) were obtained from Merck (Darmstadt, Germany). Primary–secondary amine (PSA, 40–60 μm particle size) and ceramic homogeneous proton were purchased from DIKMA (Tianjin, China) and Anpel Laboratory Technologies, Inc. (Shanghai, China), respectively.

2.3. Extraction and Cleanup

The pesticides in leafy vegetables were extracted and purified in accordance with the Chinese national standard GB 23200.113-2018 [42]. Briefly, 10 g of each sample was placed into a 50 mL polypropylene centrifuge tube to which 10 mL acetonitrile, 4 g MgSO₄, 1 g NaCl, 1 g Na₃C₆H₅O₇, 0.5 g C₁₂H₁₈Na₄O₁₇, and a ceramic proton were added. The centrifuge tube was vortexed vigorously for 1 min in the vortex genius 3 mixer (IKA, Germany) and centrifuged at 4200 rpm for 5 min, followed by transferring 6 mL of the supernatant into a 15 mL polypropylene centrifuge tube containing 900 mg MgSO₄ and 150 mg PSA and another round of vigorous shaking for 1 min. The supernatant (2 mL) was then accurately measured and transferred to two separate 10 mL polypropylene tubes and evaporated to near dryness under a nitrogen stream at 40 °C. In one tube, 20 μL of the internal standard solution (5 mg/L heptachloride B) was added to the obtained residue, re-dissolved in 1 mL of CH₃COOC₂H₅, and then filtered through a 0.22 μm polytetrafluoroethylene (PTFE) film for GC-MS/MS analysis (8890 GC equipped with MS/MS 7000D GC/TQ, Agilent, Shanghai, China). The residue obtained in the other tube was re-dissolved in 1 mL CH₃CN/water (3:1, v/v) and filtered through a 0.22 μm PTFE film before LC-MS/MS analysis (ExionLC coupled with MS/MS Qtrap 5500+, AB Sciex, Woodlands, Singapore).

2.4. Sample Analyses

We used GC-MS/MS to analyze the following 33 pesticides: phosalone, triazophos, fenitrothion, parathion-methyl, parathion, phosmet, procymidone, triadimefon, fenpropathrin, bifenthrin, tau-fluvalinate, cyhalothrin, cypermethrin, fenvalerate, deltamethrin, cyfluthrin, flucythrinate, chlorothalonil, dicofol, quintozene, vinclozolin, iprodione, pyrimethanil, metalaxyl, pendimethalin, chlorfluazuron, chlorfenapyr, pyridaben, permethrin, etofenprox, difenoconazole, chlorpyrifos, and isofenphos-methyl. We separated the compounds on a DB-5MS column (30 m × 0.25 mm, 0.25 µm particle size) purchased from Agilent (Santa Clara, CA, USA) using helium as the carrier gas at a flow rate of 1.0 mL/min. The injection mode was splitless and the injection volume was 1 μL. The solvent delay time was set to 5 min. The injector temperature was 280 °C. The oven temperature was set at 60 °C and held for 1 min, then increased to 120 °C at 40 °C/min, and finally increased to 300 °C at 5 °C/min and held for 3.5 min. The collision gas was nitrogen at a flow rate of 1.5 mL/min. The temperatures of the mass spectrometry detector transfer line and the ion source were 280 °C and 230 °C, respectively. The GC-MS/MS was operated in electron ionization mode and dynamic multiple reaction monitoring (dMRM) mode.

LC-MS/MS was used to determine the following 32 additional pesticides: chlorantraniliprole, abamectin, imidacloprid, carbaryl, carbofuran (including 3-hydroxy carbofuran), thiamethoxam, cyromazine, acetamiprid, emamectin benzoate, fipronil (including fipronil desulfinyl, fipronil-sulfone, and fipronil-sulfide), aldicarb (including aldicarb-sulfone and aldicarb-sulfide), methomyl, chlorbenzuron, isocarbophos, dichlorvos, methamidophos, phoate (including phoate-sulfone and phoate-sulfide), omethoate, dimethoate, malathion, profenofos, diazon, acephate, diflubenzuron, dimethomorph, carbendazim, prochloraz, azoxystrobin, propamocarb, pyraclostrobin, paclobutrazol, and forchlorfenuron. Liquid chromatographic separation was performed on an Atlantic T3 column (100 mm × 2.1 mm, 3 µm particle size) purchased from Waters (Taunton, MA, USA) at a flow rate of 400 µL/min. The mobile phase consisted of eluent A (water containing 5 mmol/L ammonium acetate and 0.1% formic acid) and eluent B (methanol). The gradient elution was performed as follows: 2% B was initially held for 0.5 min, then linearly ramped to 98% B from 0.5 to 3 min, held for 2 min, and finally returned to the initial condition from 5 to 5.1 min and maintained for 2.9 min. The injection volume was 10 μL. The electrospray ionization (ESI) source was operated in positive and negative ionization modes. The capillary voltages were 4.5 and 4.0 kV in positive and negative modes, respectively. The pressures of the curtain gas, atomizing gas, heating auxiliary gas, and collision gas were 30, 55, 55, and 8 psi, respectively. The temperature of ion source was set at 350 °C. The analyzer was operated in multiple reaction monitoring (MRM) mode.

2.5. Quality Assurance and Quality Control

This study involved the collection of samples and analysis of pesticide residues from 2018 to 2020. After the samples collected annually were delivered to the laboratory, in order to avoid the decomposition of pesticide residues, it was guaranteed that the pretreatment and instrumental analysis of samples would be completed within 30 days. Furthermore, in order to maintain data comparability across the three-year experimental period, identical manufacturers were chosen for purchasing pesticide standards, reagents, and consumables. Additionally, stringent quality control measures have been implemented to ensure compliance with SANTE/12682/2019 [43] and GB/T 27417-2017 [44] requirements for linearity, LOD, LOQ, accuracy, and precision. The calibration curves were established in the mixed blank supernatants of nine kinds vegetables by spiking mixed solvent standard solutions at eight different concentrations (0.001, 0.005, 0.01, 0.025, 0.05, 0.075, 0.1, and 0.2 mg/L for LC-MS/MS; 0.01, 0.025, 0.05, 0.1, 0.3, 0.5, 0.75, and 0.1 mg/L for GC-MS/MS), while heptachloride B was spiked before analysis for GC-MS/MS. The correlation coefficients were > 0.99. Accuracy was evaluated at two levels of quality control. The spiked concentrations of all pesticides were 15 and 50 µg/kg, except for the synthetic pyrethroid pesticides, which were 25 and 125 µg/kg. The average recoveries of 65 pesticides varied between 60% and 120%. The estimated minimum acceptable concentrations (5–15 μg/kg) of pesticides were added to the blank mixed vegetable matrices, and 10 replicate analyses were conducted with acceptable recoveries (60–120%) [43]. The standard deviation (SD) of the test results was calculated. The LOD and LOQ were 3 and 10, respectively, as the SD. The precision, expressed as relative standard deviation (RSD), was determined by analyzing three duplicate samples and was set 20% lower [44]. One procedural blank, consisting of 10 mL of HPLC-water throughout the preparation, was established during the analysis of every 24 samples. The target analytes were not detected in the blanks.

2.6. Food Consumption Data

Owing to its large population, China has few food consumption surveys. However, the relative data can be obtained from the Chinese National Nutrition and Health Survey conducted by the National Health and Family Planning Commission of China [45]. This report contains the most comprehensive, representative, systematic, and up-to-date data on food consumption. Using an individual 24 h recall method for three consecutive days, the investigators collected food consumption data by face-to-face questionnaire, and the anthropometric parameters were obtained by measurement. Ganjingzi in Dalian, Heping in Shenyang, Yuhong in Shenyang, Zhen’an in Dandong, Pulandian, Dengta, and Chaoyang are the representative districts of Liaoning province. The data from the above regions of Liaoning province, including the weight parameters of permanent residents aged ≥ 18 years and dietary vegetable consumption, were used in this study.

2.7. Risk Assessment

The concentrations of pesticide residues were compared with the MRLs established by the Codex Alimentarius Commission (WHO) [46] and the Chinese National Standard GB 2763-2021 [47], as listed in

Table 2. Maximum residue limits of 65 pesticides.

Pesticide	Leafy Vegetable	Spinach	Pakchoi	Lettuce	Celery	Chinese Cabbage	Leaf Lettuce	C. coronarium	Rape	Chicory
Acephate	0.02	/	/	/	/	/	/	/	/	/
Acetamiprid	1.5	5	1	5	3	1	/	/	/	/
Aldicarb	0.03	/	/	/	/	/	/	/	/
Avermectin	/	0.05	0.05	0.05	0.05	0.05	0.05	0.05	0.1	/
Azoxystrobin	/	/	/	3 (3)	5	/	/	/	/	/
Bifenthrin	/	/	/	/	/	/	/	/	/	/
Carbaryl	1	/	5	/	/	/	/	/	/	/
Carbendazim	/	/	/	/	/	/	/	/	/	/
Carbofuran	0.02	/	/	/	/	/	/	/	/	/
Chlorantraniliprole	20 (20)	/	/	/	7	/	5	/	/	/
Chlorbenzuron	/	30	30	/	/	/	/	/	/	/
Chlorfenapyr	/	/	10	/	/	2	/	/	/	/
Chlorfluazuron	/	10	7	/	/	2	/	/	/	/
Chlorothalonil	/	5	5	5	5	5	/	/	/	/
Chlorpyrifos	0.02	/	/	/	0.05	/	/	/	/	/
Cyfluthrin	/	0.5	0.5	/	0.5	0.5	10	/	/	/
Cyhalothrin	/	2	2	2	0.5	1	2	5	/	/
Cypermethrin	0.7 (0.7)	2	2	2	1	2	7	7	/	/
Cyromazine	/	/	/	4 (4)	4	/	20	/	/	/
Deltamethrin	/	0.5	0.5	2	2	0.5	2	2	/	/
Diazon	/	0.5 (0.5)	0.2	0.5 (0.5)	/	0.05 (0.05)	/	/	/	/
Dichlorvos	0.2	0.5	0.1	/	/	0.5	/	/	/	/
Dicofol	0.01	/	/	/	/	/	/	/	/	/
Difenoconazole	/	10	/	2 (2)	3	1	10	/	/	/
Diflubenzuron	/	1	1	1	/	1	/	/	/	/
Dimethoate	0.01	/	/	/	/	/	/	/	/	/
Dimethomorph	/	30 (30)	/	(9)	15	/	40	/	/	/
Emamectin benzoate	/	0.2	0.1	0.7 (0.7)	/	0.05	0.05	/	/	/
Etofenprox	/	1	1	/	1	1	/	/	/	/
Fenitrothion	0.5	/	/	/	/	/	/	/	/	/
Fenpropathrin	/	1	1	0.5 (1.5)	1	1	/	7	/	/
Fenvalerate	/	1	1	1	/	3	/	10	/	/
Fipronil	0.02	/	/	/	/	/	/	/	/	/
Flucythrinate	/	/	/	/	/	/	/	/	/
Forchlorfenuron	/	/	/	/	/	/	/	/	/	/
Imidacloprid	5	/	0.5	1	5	0.2	1	/	/	/
Iprodione	/	/	/	25 (25)	/	/	25	/	/	/
Isocarbophos	0.05	/	/	/	/	/	/	/	/	/
Isofenphos-methyl	0.01	/	/	/	/	/	/	/	/	/
Malathion	/	2 (3)	8	8	1	8	/	/	/	/
Metalaxyl	/	2 (0.02)	/	(1.5)	/	/	/	/	/	/
Methamidophos	0.05	/	/	/	/	/	/	/	/	/
Methomyl	0.2	/	/	(0.2)	/	/	/	/	/	/
Omethoate	0.02	/	/	/	/	/	/	/	/	/
Paclobutrazol	/	/	/	/	/	/	/	/	/	/
Parathion	0.01	/	/	/	/	/	/	/	/	/
Parathion-methyl	0.02	/	/	/	/	/	/	/	/	/
Pendimethalin	/	0.2	0.2	0.1 (4)	0.2	0.2	/	/	/	/
Permethrin	1	2 (2)	/	/	2	5 (5)	/	/	/	/
Phoate	0.01	/	/	/	/	/	/	/	/	/
Phosalone	/	1	1	1		1	/	/	/	/
Phosmet	/	/	/	/	/	0.5	/	/	/	/
Prochloraz	/	/	/	/	/	/	/	/	/	/
Procymidone	/	10	/	15	/	/	15	/	/	/
Profenofos	/	/	/	/	/	5	/	/	/	/
Propamocarb	/	100 (40)	/	(100)	/	10	/	/	/	/
Pyraclostrobin	/	20	/	2	30	5	20	5	/	/
Pyridaben	/	/	/	/	/	2	/	/	/	/
Pyrimethanil	/	/	/	/	/	/	20	/	/	/
Quintozene	/	/	/	/	/	/	/	/	/	/
Tau-fluvalinate	/	0.5	0.5	/	0.5	0.5	/	/	/	/
Thiamethoxam	3 (3)	5	/	10	1	/	/	/	/	/
Triadimefon	/	/	/	/	/	/	/	/	/	/
Triazophos	0.05	/	/	/	/	/	/	/	/	/
Vinclozolin	/	/	/	/	/	/	/	/	/	/

The aldicarb content is determined by the combined measurement of aldicarb, aldicarb-sulfone, and aldicarb-sulfide; the carbofuran content is determined by the combined measurement of carbofuran and 3-hydroxy carbofuran; the fipronil content is determined by the combined measurement of fipronil, fipronil desulfinyl, fipronil-sulfide and fipronil-sulfone; the phoate content is determined by the combined measurement of phoate, phoate-sulfide, and phoate-sulfone. The values within and outside the brackets represent MRLs defined by the Codex and GB 2763-2021, respectively.

. The pesticides with no specified limit values in the aforementioned standards were ignored. All risks originating from chronic and acute pesticide intake must be evaluated.

2.7.1. Long-Term (Chronic) Dietary Risk Assessment

The long-term dietary intake of pesticides is expressed as the National Estimated Daily Intake ($NEDI$) [48,49,50] and calculated using Formula (1):

$$NEDI=R\times F\times 1000/bw$$

(1)

where $NEDI$ is expressed in μg/kg bw/day; $R(\mathrm{mg}/\mathrm{kg})$ is the average value of the actual pesticide residue; $F(\mathrm{mg}/\mathrm{kg})$ is the average intake of leafy vegetables; and $bw$ is the average body weight of adult residents in Liaoning province (66.4 kg).

The dietary intake of leafy vegetables consumed by the residents of Liaoning province in the cold season is mainly from locally grown greenhouses production. Because of the lack of specific data on greenhouse-grown leafy vegetables consumption in Liaoning province, this study adopted the consumption data of Liaoning residents from the 2002 National Household Nutrition and Health Survey [45]. The average daily intakes of dark and light leafy vegetables of the Liaoning population were 34.8 and 216.3 g/day, respectively. Since all the nine types of leafy vegetables used in this study were dark and light, the total intake amounted to 251.1 g/day. In addition, the vegetable consumption and dietary structure of Chinese residents have significantly changed in recent years; with the increase in greenhouse vegetable planting areas in northeast China, the total annual per capita consumption of leafy vegetables in the cold seasons (winter and spring) increased as well. Therefore, the total amount of vegetables included in our calculation could compensate for the increased leafy vegetables consumption, thus reducing the possibility of underestimating the risk of dietary exposure.

The chronic risk quotient (${RQ}_c$) of pesticide intake was calculated using Formula (2):

$${RQ}_c\left(\mathrm{\%}\right)=NEDI\times 100\mathrm{\%}/\left(ADI\times 1000\right)$$

(2)

where $ADI$ is the acceptable daily intake (mg/kg bw/day); in this study, the $ADI$ values of the pesticides are in accordance with the requirements of GB 2763-2021 [47]. ${RQ}_c$ ≤ 100% indicates an acceptable chronic risk, and the smaller ${RQ}_c$ is, the lower the risk. ${RQ}_c$ > 100% indicates a certain dietary risk, and the higher ${RQ}_c$ is, the greater the risk [34].

2.7.2. Short-Term (Acute) Dietary Risk Assessment

Chinese cabbage and pak choi cabbage accounted for the largest number of collected samples in this study. Dietary intake of these two leafy vegetables was shown to be higher in northeast China in cold seasons. Therefore, a risk assessment of acute dietary exposure to Chinese cabbage and pak choi cabbage was conducted in this study. The short-term dietary intake of pesticides in leafy vegetables was expressed as the International Estimated Short-Term Intake ($IESTI$) in μg/kg bw. Based on the nature of product, the evaluation was performed in the following four modes, including case 1 (one unit weight of a product is < 25 g), case 2a (one unit weight of a product is ≥ 25 g but lower than a large portion per meal), case 2b (one unit weight of a product is ≥ 25 g but higher than a large portion per meal), and case 3 (a variety of mixed processed foods). Chinese cabbage and pak choi cabbage belong to the assessment scenario of case 2b (Formula (3)) and case 2a (Formula (4)), respectively, as follows [48,49,50]:

$$IESTI=LP\times HR\times V/bw$$

(3)

$$IESTI=\left(U_e\times HR\times V+\left(LP-U_e\right)\times HR\right)/bw$$

(4)

where $LP$ (g) is the large meal portion representing the 97.5th percentile of eaters. The acute dietary exposure risk assessment uses $LP$ recommended by the Food and Agriculture Organization of the United Nations (FAO). The $LP$ of Chinese cabbage is 668.4 g. The $LP$ of pak choi cabbage is 503.2 g, which is the average of shepherd’s purse (554.5 g) and lettuce (451.9 g) [51]. $HR$ (mg/kg) is the highest pesticide residue in the edible portion of the sample, and this research adopts the maximum for each pesticide residue in vegetables. $V$ is the variation factor, which represents the residue variation of different individuals in the same batch of products or different parts of the same individual. It is defined as the ratio of 97.5% location residue to the average residual amount. The default value is 3, as recommended by the Joint Meeting on Pesticide Residues (JMPR) [52]. $U_e$ (g) is the edible part of an individual product. In this study, the individual weight of Chinese cabbage recommended by FAO is 2162 g [51]. Due to the lack of individual product quality of pakchoi, the average value (275 g) of shepherd’s purse (245 g) and lettuce (305 g) recommended by FAO is used as the quality of pakchoi [51].

The acute dietary risk quotient (${RQ}_a$) was calculated using Formula (5):

$${RQ}_a\left(\%\right)=IESTI\times 100\%/ARf\mathrm{D}\times 1000$$

(5)

where $ARf\mathrm{D}$ (mg/kg bw) is an acute reference dose obtained from the JMPR database [52]. If $ARf\mathrm{D}$ ≤ 100%, it is likely that the residents would not experience health effects, whereas $ARf\mathrm{D}$ > 100% indicates an unacceptable acute risk. The higher $ARf\mathrm{D}$ is, the greater the risk.

2.8. Substitution of Undetected Results

Due to the limitations of current detection technology, this analytical monitoring study included some undetected data, indicating either the absence of pesticides or their presence at concentrations below LOD. In the dietary exposure assessment, accurate data processing of pesticide residues is crucial as it significantly affect the assessment results. The risk of dietary exposure to the pesticide may be under- or overestimated when the residue concentration is less than the LOD. To appropriately handle undetected residue values in samples, WHO has provided an empirical guideline. If the measured value of more than 60% of the contaminant in the food is lower than the LOD, all results are estimated as 0 (lower limit) and LOD (upper limit), respectively [53]. Moreover, these principles were also adopted by the National Food Safety Risk Assessment Expert Committee of China in 2010 [54]. Considering that the number of samples with pesticide residues lower than the LOD was more than 60% in this study, the undetected values were replaced by 0 and LOD for the calculation of the dietary exposure assessment of pesticides.

2.9. Data Analysis

Pesticide data on LC-MS/MS were collected and analyzed using Analyst Version 1.7.2 and OS Version 2.0.1 software (AB Sciex Company, Woodlands, Singapore). Pesticide data on GC-MS/MS were completed using MassHunter Workstation Version 10.0 software (Agilent Company, Shanghai, P.R. China). Statistical analyses were performed using Microsoft Excel version 2016 (Microsoft Corporation, Seattle, WA, USA). Plotting of collected samples regions and statistical graphs was performed using ArcGIS Pro 2.7.3 (ESRI Inc., Redlands, CA, USA) and Origin 2018 (Origin Inc., Northampton, MA, USA) software.

3. Results and Discussion

3.1. Method Validation

To evaluate the performance of the established method, the method was validated in terms of the linear relationships, precision, accuracy, LODs, and LOQs, which are shown in

Table 3. Linear equations, correlation coefficients, average recoveries, and relative standard deviations of 65 pesticides.

Pesticide	Linear Equation	Correlation Coefficient	Linear Range (mg/L)	LOD (μg/kg)	LOQ (μg/kg)	Average Recovery (%)	RSD (%)	Average Recovery (%)	RSD (%)
						15 μg/kg (n = 3)		50 μg/kg (n = 3)
Methamidophos	y = 8.39839 × 107 x − 1.40518 × 106	0.9942	0.005–0.2	10	30	60.3	6.4	73.3	2.7
Acephate	y = 6.35974 × 107 x + 3.77635 × 105	0.9954	0.005–0.2	10	30	78.5	8.4	82.3	3.1
Phoate	y = 4.72280 × 105 x + 8025.20407	0.9995	0.005–0.2	10	30	88.7	3.7	91.9	2.3
Phoate-sulfone	y = 1.30202 × 108 x − 1.02770 × 105	0.9941	0.005–0.2	10	30	86.0	11.1	87.7	5.2
Phoate-sulfide	y = 6.45845 × 108 x − 1.46708 × 106	0.9917	0.005–0.2	10	30	79.3	6.4	84.5	1.1
Omethoate	y = 1.20998 × 108 x -− 3.62998 × 105	0.9942	0.005–0.2	10	30	82.7	5.5	92.7	1.0
Carbofuran	y = 1.32702 × 109 x − 2.80931 × 106	0.9940	0.001–0.2	5	15	93.4	8.8	105.7	5.0
3-hydroxy carbofuran	y = 1.55582 × 108 x − 2.12487 × 105	0.9914	0.001–0.2	5	15	91.3	5.6	93.6	2.7
Aldicarb	y = 1.74744 × 108 x − 3.85052 × 105	0.9935	0.001–0.2	5	15	69.7	12.2	73.5	6.6
Aldicarb-sulfone	y = 1.25427 × 108 x − 2.66677 × 105	0.9935	0.001–0.2	5	15	83.7	10.2	87.5	4.5
Aldicarb-sulfide	y = 3.24361 × 107 x − 7.72354 × 104	0.9911	0.001–0.2	5	15	92.5	9.7	99.0	2.6
Fipronil	y = 3.77463 × 108 x + 5986.58779	0.9995	0.001–0.2	5	15	114.8	11.1	107.2	1.3
Fipronil desulfinyl	y = 1.38282 × 108 x − 1.37382 × 105	0.9977	0.001–0.2	5	15	109.0	3.6	99.5	1.9
Fipronil-sulfide	y = 3.04139 × 108 x − 1.70492 × 104	0.9998	0.001–0.2	5	15	112.3	5.0	98.9	5.0
Fipronil-sulfone	y = 7.52666 × 108 x − 6.27160 × 105	0.9979	0.001–0.2	5	15	109.7	6.3	98.0	2.2
Methomyl	y = 1.29386 × 108 x − 2.53419 × 105	0.9928	0.001–0.2	5	15	82.7	8.6	97.3	3.2
Carbaryl	y = 2.71699 × 108 x − 5.83809 × 105	0.9903	0.001–0.2	5	15	116.2	11.1	97.5	7.3
Chlorbenzuron	y = 5.53654 × 107 x − 2.86082 × 105	0.9901	0.001–0.2	5	15	114.3	4.8	102.0	2.2
Imidacloprid	y = 1.16293 × 108 x − 1.90029 × 105	0.9902	0.001–0.2	5	15	97.7	5.5	92.1	3.1
Acetamiprid	y = 8.65808 × 108 x − 1.63937 × 106	0.9902	0.001–0.2	5	15	97.3	8.6	82.7	3.2
Avermectin	y = 9.68231 × 106 x − 2.40343 × 104	0.9985	0.005–0.2	10	30	73.3	5.9	88.7	3.0
Carbendazim	y = 2.16731 × 109 x − 6.36224 × 106	0.9915	0.001–0.2	5	15	95.0	10.5	91.5	7.0
Difenoconazole	y = 3.82620 × 108 x − 8.04053 × 105	0.9925	0.001–0.2	5	15	71.7	8.0	86.0	6.0
Emamectin benzoate	y = 1.58396 × 108 x − 3.95905 × 105	0.9973	0.005–0.2	10	30	84.3	6.5	97.5	5.0
Dimethomorph	y = 6.42839 × 108 x − 4.02588 × 105	0.9956	0.001–0.2	5	15	78.0	5.3	80.0	1.6
Prochloraz	y = 5.19039 × 108 x − 1.30951 × 106	0.9930	0.001–0.2	5	15	80.3	6.4	93.3	2.7
Azoxystrobin	y = 1.24394 × 109 x − 4.37042 × 105	0.9982	0.001–0.2	5	15	83.7	5.6	85.7	4.1
Thiamethoxam	y = 2.83631 × 108 x + 4.13036 × 105	0.9952	0.001–0.2	5	15	74.7	5.3	85.5	2.7
Cyromazine	y = 5.34189 × 107 x − 1.82179 × 106	0.9916	0.001–0.2	5	15	73.5	7.6	91.1	6.4
Propamocarb	y = 1.19756 × 109 x − 3.35228 × 105	0.9934	0.005–0.2	10	30	68.5	6.7	74.9	5.5
Paclobutrazol	y = 2.19432 × 108 x − 5.65162 × 105	0.9951	0.001–0.2	5	15	116.3	6.0	103.5	1.7
Chlorantraniliprole	y = 1.91081 × 108 x − 3.83118 × 105	0.9912	0.001–0.2	5	15	110.7	12.0	95.7	3.4
Pyraclostrobin	y = 6.91635 × 108 x − 2.09781 × 106	0.9901	0.001–0.2	5	15	72.3	7.0	78.6	4.7
Forchlorfenuron	y = 2.73834 × 108 x − 3.56087 × 105	0.9930	0.001–0.2	5	15	118.7	2.3	111.9	3.7
Diflubenzuron	y = 1.23202 × 108 x − 2.91265 × 105	0.9904	0.001–0.2	5	15	113.0	7.1	93.2	1.7
Parathion-methyl	y = 9.219504 x − 0.502101	0.9996	0.01–1	10	30	82.0	4.6	82.3	4.3
Phosmet	y = 4.04178 × 108 x − 3.51979 × 105	0.9992	0.01–1	10	30	69.4	8.1	85.3	1.5
Chlorpyrifos	y = 4.171031 x − 0.124316	0.9997	0.01–1	10	30	118.3	6.9	106.4	3.1
Triazophos	y = 3.050311 x − 0.047997	0.9991	0.01–1	10	30	80.5	19.9	86.9	2.3
Dimethoate	y = 6.26674 × 108 x − 8.26187 × 105	0.9965	0.005–0.2	15	45	77.0	5.4	88.3	2.4
Dichlorvos	y = 1.21165 × 108 x − 2.99533 × 105	0.9957	0.005–0.2	10	30	78.0	10.7	83.0	3.1
Fenitrothion	y = 3.515750 x + 0.005005	0.9998	0.01–1	10	30	70.8	6.5	76.3	2.7
Profenofos	y = 3.55864 × 107 x − 1.08006 × 105	0.9999	0.005–0.2	10	30	119.0	3.4	117.9	4.1
Malathion	y = 2.74490 × 108 x − 6.73535 × 105	0.9926	0.005–0.2	10	30	74.4	14.7	91.8	4.4
Diazon	y = 6.53992 × 108 x − 1.93327 × 106	0.9915	0.005–0.2	10	30	70.4	7.1	81.5	3.6
Isocarbophos	y = 1.22967 × 108 x − 2.36372 × 105	0.9935	0.005–0.2	10	30	75.6	12.5	88.6	5.3
Parathion	y = 6.480895 x + 0.081211	0.9995	0.01–1	10	30	85.7	10.2	89.2	6.9
Phosalone	y = 7.524482x − 0.166662	0.9994	0.01–1	10	30	106.3	5.6	92.3	4.2
Isofenphos-methyl	y = 7.657783x + 0.011626	0.9998	0.01–1	10	30	77.3	11.4	86.2	7.3
Cyhalothrin	y = 4.252657 x − 0.146437	0.9993	0.025–1	15	45	105.8	7.0	104.3	5.3
Cypermethrin	y = 5.115630 x + 0.266442	0.9955	0.025–1	15	45	112.7	5.3	101.6	7.4
Fenvalerate	y = 1.682568 x + 0.078467	0.9997	0.025–1	15	45	117.5	9.2	113.2	6.3
Deltamethrin	y = 2.085506 x + 0.072699	0.9972	0.025–1	15	45	112.5	10.1	108.9	3.2
Cyfluthrin	y = 3.998447 x + 0.576045	0.9989	0.025–1	15	45	105.8	7.0	104.3	5.3
Flucythrinate	y = 10.236721 x − 0.377243	0.9962	0.025–1	15	45	91.8	5.4	98.3	5.8
Fenpropathrin	y = 3.041895 x + 0.088083	0.9992	0.01–1	10	30	117.3	10.6	111.4	7.1
Bifenthrin	y = 22.152456 x + 1.471447	0.9979	0.01–1	10	30	118.1	12.3	92.6	2.7
Tau-fluvalinate	y = 1.381301 x + 0.009669	0.9967	0.01–1	10	30	117.0	5.7	111.5	2.9
Permethrin	y = 2.870759 x − 0.094977	0.9993	0.01–1	10	30	113.3	9.1	105.1	4.6
Etofenprox	y = 2.290453 x − 0.075012	0.9999	0.01–1	15	45	85.3	7.8	94.2	5.3
Pyridaben	y = 16.226839 x − 0.433625	0.9999	0.01–1	10	30	112.3	9.2	100.7	1.2
Pendimethalin	y = 1.665961 x − 0.115475	0.9994	0.01–1	10	30	82.1	5.9	86.9	1.2
Chlorfenapyr	y = 0.491678 x − 0.026446	0.9994	0.01–1	10	30	86.2	9.0	93.2	6.1
Iprodione	y = 0.270364 x + 0.020345	0.9976	0.025–1	10	30	99.0	16.3	108.1	10.3
Pyrimethanil	y = 4.097693 x − 0.117433	0.9996	0.01–1	10	30	82.4	7.8	85.1	6.2
Chlorothalonil	y = 3.770007 x + 1.537023	0.9982	0.025–1	10	30	65.8	11.0	73.7	9.3
Dicofol	y = 11.134304 x + 0.181842	0.9994	0.01–1	10	30	88.6	9.0	93.4	6.3
Triadimefon	y = 2.202226 x + 0.075454	0.9996	0.01–1	10	30	83.8	4.6	92.5	3.5
Procymidone	y = 5.908395 x + 0.331752	0.9985	0.01–1	10	30	70.0	7.9	85.3	4.7
Chlorfluazuron	y = 0.487121 x − 0.035471	0.9990	0.025–1	10	30	85.3	5.7	103.2	1.2
Quintozene	y = 1.543211 x + 0.060027	0.9984	0.01–1	10	30	71.8	8.0	90.1	5.0
Vinclozolin	y = 2.047427 x + 0.062417	0.9990	0.01–1	10	30	70.8	14.3	86.7	9.2
Metalaxyl	y = 1.734433 x − 0.150437	0.9998	0.01–1	15	45	86.2	7.5	91.5	3.1

The concentrations of synthetic pyrethroid pesticides added were 25 and 125 µg/kg, respectively.

. Linear regression analysis was performed with the peak areas of the 65 pesticides as vertical coordinates and the mass concentration of each pesticide as horizontal coordinates. The results demonstrated excellent linearity within the range of 0.001(0.005)–0.2 mg/L (for liquid quality) and 0.01(0.025)–1 mg/L (for gas quality), exhibiting correlation coefficients exceeding 0.99. Two concentration levels of 15 and 50 µg/kg (the concentrations of synthetic pyrethroid pesticides to be added were 25 and 125 µg/kg, respectively) were spiked into the mixed blank matrix of nine kinds of vegetables, with three parallels for each concentration. As a whole, the results demonstrated that the recoveries and RSDs of the higher concentration (50 µg/kg) were better than those of the lower concentration (10 µg/kg). Specifically, the recoveries ranged from 60.3% to 119.0% with RSDs ranging from 2.3% to 19.9%. On the other hand, for the higher concentration, the recoveries varied between 73.3% and 119.9% with RSDs of 1.0–10.3%. It should be noted that these experimental results were in accordance with the requirements of GB/T 27417-2017 [44] for multi-residue analysis. The low concentration of 5–15 μg/kg was added to the mixed blank matrix of nine kinds of vegetables. Following the pretreatment described in Section 2.3 and the detection conditions for the target pesticide outlined in Section 2.4, a total of ten parallel determinations were performed. The method’s LODs and LOQs, calculated as three times the SD and ten times the SD, respectively, were found to be within the range of 5–15 μg/kg and 10–45 μg/kg accordingly (refer to Table 3). The obtained LOQs were not higher than the Codex and Chinese MRLs for pesticides in the collected nine kinds of leafy vegetables, indicating that the method is suitable for the comprehensive survey of the pesticide multi-residues in the vegetable samples.

3.2. Pesticide Residues in Leafy Vegetables

3.2.1. Residues of 65 Pesticides

Sixty-five different pesticides of various chemical groups were analyzed in the collected leafy vegetable samples. The occurrence and level of pesticide residues are presented in

Table 4. Pesticide residues in positive greenhouse leafy vegetables in the cold season of Liaoning province.

Pesticide	Category	Toxicity	No. of Positive Samples (%)	No. of Samples > MRL (%)	Min (mg/kg)	Max (mg/kg)
Dimethomorph	F	low	56 (11.9)	0 (0)	0.006	18.852
Carbendazim	F	low	36 (7.7)	0 (0)	0.007	5.354
Procymidone	F	low	24 (5.1)	0 (0)	0.009	2.482
Acetamiprid	I	low	24 (5.1)	0 (0)	0.009	0.640
Thiamethoxam	I	low	21 (4.5)	0 (0)	0.006	0.380
Imidacloprid	I	low	19 (4.1)	0 (0)	0.010	1.900
Difenoconazole	F	low	18 (3.8)	0 (0)	0.005	0.312
Pyridaben	A	moderate	17 (3.6)	0 (0)	0.006	1.361
Bifenthrin	I	low	17 (3.6)	0 (0)	0.007	2.201
Chlorantraniliprole	I	low	16 (3.4)	0 (0)	0.008	1.146
Emamectin benzoate	I	low	14 (3.0)	0 (0)	0.006	0.072
Profenofos	I	moderate	11 (2.3)	0 (0)	0.022	0.620
Paclobutrazol	P	low	11 (2.3)	0 (0)	0.009	0.204
Abamectin	I	high	11 (2.3)	3 (0.6)	0.012	0.348
Pyraclostrobin	F	low	9 (1.9)	0 (0)	0.008	0.135
Chlorpyrifos	I	moderate	8 (1.7)	5 (1.1)	0.012	0.920
Fenpropathrin	I	moderate	8 (1.7)	0 (0)	0.022	0.195
Propamocarb	F	low	8 (1.7)	0 (0)	0.037	1.446
Pyrimethanil	F	low	8 (1.7)	0 (0)	0.017	5.640
Triadimefon	F	low	7 (1.5)	0 (0)	0.009	4.495
Methamidophos	I	high	7 (1.5)	0 (0)	0.010	0.031
Carbofuran	I	high	6 (1.3)	6 (1.3)	0.090	0.760
Metalaxyl	F	low	5 (1.1)	0 (0)	0.027	0.068
Cyhalothrin	F	moderate	5 (1.1)	0 (0)	0.150	0.228
Azoxystrobin	F	low	5 (1.1)	0 (0)	0.005	0.410
Diflubenzuron	I	high	5 (1.1)	0 (0)	0.012	0.386
Cyromazine	I	low	4 (0.9)	0 (0)	0.011	0.320
Prochloraz	F	low	4 (0.9)	0 (0)	0.006	0.130
Cypermethrin	I	moderate	3 (0.6)	0 (0)	0.076	0.321
Fenvalerate	I	moderate	3 (0.6)	0 (0)	0.031	0.307
Triazophos	I	moderate	3 (0.6)	2 (0.4)	0.036	0.366
Pendimethalin	H	low	2 (0.4)	0 (0)	0.034	0.192
Chlorfluazuron	I	low	2 (0.4)	0 (0)	0.008	0.009
Iprodione	F	low	2 (0.4)	0 (0)	0.057	0.342
Dichlorvos	I	moderate	2 (0.4)	0 (0)	0.014	0.056
Chlorbenzuron	I	low	2 (0.4)	0 (0)	0.026	0.028
Forchlorfenuron	P	low	2 (0.4)	0 (0)	0.028	0.112
Dimethoate	I	moderate	1 (0.2)	1 (0.2)	/	0.021
Fenitrothion	I	moderate	1 (0.2)	0 (0)	/	0.250
Methomyl	I	high	1 (0.2)	0 (0)	/	0.016
Carbaryl	I	low	1 (0.2)	0 (0)	/	0.012

Category: I: insecticide; F: fungicide; A: acaricide; P: plant growth regulator; H: herbicide.

. Of the total samples, 41 active ingredients were detected, accounting for 63.1% of the analyzed pesticides, including 18 low-toxicity, 19 moderate-toxicity, and 4 high-toxicity pesticides. The detection frequency of the 41 detected pesticides ranged from 0.2% to 11.9%. The most frequently detected classes were insecticides (24/41) and fungicides (13/41). Dimethomorph (11.9% of the samples) was the most constantly detected pesticide with residues of 0.006–18.852 mg/kg. Carbendazim (7.7%) ranked second with residues ranging from 0.007 to 5.354 mg/kg. This was followed by procymidone and acetamiprid (5.1%) with residues of 0.009–2.482 and 0.009–0.640 mg/kg, respectively. The experimental data demonstrated that the four low-toxicity fungicides were commonly used for greenhouse vegetables in winter and spring in Liaoning province. The unique ecological environment of prolonged exposure to both low temperature and high temperature, which is a characteristic of greenhouse agriculture, is very suitable for the breeding and spread of pathogens. Consequently, dimethomorph, carbendazim, procymidone, and acetamiprid are frequently applied by farmers as low-toxicity, highly effective, broad-spectrum fungicides to effectively prevent and control gray mold and downy mildew in greenhouses.

The recommended MRLs for pesticide residues in leafy vegetable, as suggested by the Codex and China, range from 0.01 to 100 mg/kg (Table 2) [46,47]. In general, the MRLs of 65 pesticides in GB 2763-2021 are the same as those in the Codex, or the MRLs in GB 2763-2021 are more stringent. However, there are certain exceptional cases. For instance, while the Codex stipulates a MRL of 9 mg/kg for dimethomorph in lettuce, GB 2763-2021 does not provide any specific requirement for this pesticide. Similarly, propamocarb’s MRL in Spinach established by the Codex is lower than that established by GB 2763-2021 (40 mg/kg vs. 100 mg/kg), and an additional MRL of 100 mg/kg is specified for lettuce. In this study, we have adopted the stricter MRLs from the Codex and GB 2763-2021 to implement data statistics of pesticide residue exceeding MRLs. Overall, the vast majority of individual pesticide concentrations in the leafy vegetable samples were compliant with the Codex MRLs [46] and GB 2763-2021 [47] MRLs. Carbofuran (including 3-hydroxy carbofuran) (1.3%), chlorpyrifos (1.1%), abamectin (0.6%), triazophos (0.4%), and dimethoate (0.2%) were the pesticides exceeding the MRLs (For example, the mass spectrometric confirmation of chlorpyrifos in a positive celery sample is shown in Figure 2). It is noteworthy that carbofuran, chlorpyrifos, triazophos, and dimethoate have all been banned for use as pesticides in vegetables by the Ministry of Agriculture and Rural Affairs of China [55,56,57]. The similar detected results of banned pesticides have also been found in other regions of China [24]. The data results objectively indicated that farmers were actively or passively using the active ingredients of the above four banned pesticides for controlling pests and diseases in leafy vegetable cultivation. The presence of excessive levels of abamectin indicated the potential for repeated utilization of the same active ingredients in pesticide application.

The overall qualified rate of 469 samples was 96.4% (452/469), which was higher than the statistical results of vegetable quality risk monitoring in Liaoning province from 2013 (84.7%) to 2016 (96.3%) [23]. The total pesticide qualification rate of greenhouse vegetables was higher than that of Chinese national vegetable survey results in 2018 (87.3%–90.8%), lower than the results of Pingliang (97.34%) [58]. The detection rate of pesticide residues in leafy vegetables in winter and spring in Liaoning province is relatively high, which is mainly due to the following factors: First, the prevalence of repeated and continuous cultivation is a common phenomenon in Liaoning province. Cultivation facilities have 2–3 cultivation cycles per year, with only 1–2 months of downtime for shed maintenance. This continuous cultivation leads the accumulation of pathogens and pests, requiring large amounts and frequent application of pesticides. Furthermore, the continuous introduction of new vegetable varieties in greenhouses has brought a gradual escalation of disease and pest species, a succession of dominant pathogens, and the rapid emergence of pesticide resistance. In terms of plant disease and pest prevention, farmers tend to overlook the application of comprehensive measures such as ecological regulation and physical and biological control. Instead, they only rely on chemical pesticides with a wide range of pesticide types and higher doses to meet the needs of controlling plant diseases and pests. Third, vegetable production exhibits a significant economic advantage, which prompts farmers to occasionally abuse pesticides in pursuit of short-term high yields. Fourth, the lack of a safe interval after pesticide application has resulted in excessive residues of these chemicals in harvested produce that subsequently enters the market.

3.2.2. Pesticide Residues of the Nine Types of Leafy Vegetables

In this study, 48.6% (228/469) of the total leafy vegetable samples exhibited levels above the LOD for at least 1 out of the 65 pesticides analyzed. The detection rate of pesticides in greenhouse vegetables was obviously higher than that of Changchun (28.43%) [21] and Weifang (17.5% for leafy vegetables) [59]. This result should be related to the sampling time as well as the type and amount of pesticide monitored. Weifang is located in the northern temperate monsoon region. The sampling period was from April to December in 2018, during which vegetables showed lower incidence and severity of diseases and pests compared to those grown in winter and spring. Although vegetable samples in Changchun were collected in autumn and winter, when plant diseases and pests frequently occur, the detection rate of pesticides was relatively low due to the small number of pesticides monitored (only 18 kinds). Among the positive samples, apparent differences in disparate pesticide residues were observed (Figure 3). Chinese cabbage had the lowest percentage of pesticide residues (19.3%), whereas celery and chicory had the highest with 84.1% and 84.2%, respectively. C. coronarium, celery, and chicory had the highest proportion of pesticides with one, three, and four pesticides detected, respectively. In contrast, the other six leafy vegetables had the largest proportion of samples with no pesticides detected.

Specific cultivation modes (continuous cropping, annual production, and high input of agricultural fertilizers) and environmental characteristics of greenhouse (high temperature, high humidity, and low light) can lead to an increase in the severity of disease and pests. With the extensive use of pesticides, the phenomenon of multi-pesticide residues in single agricultural products has been gradually increasing. This study also found that there was a relatively common phenomenon of multiple pesticide contamination in leafy vegetables, as shown in Figure 3. A total of 26.4% of these positive vegetable samples contained two or more pesticide residues. Alternatively, three types of pesticides were simultaneously detected in Chinese cabbage samples, four in rapeseed, spinach, and lettuce samples, and five or more in celery, C. coronarium, leaf lettuce, chicory, and pakchoi samples. The results of pesticide residues showed that 34.1% of 1720 vegetable samples had two or more positive pesticides in Shandong province [60]. Furthermore, 300 leafy vegetables were collected from 20 regions in Italy for the analysis of 210 chemical pesticides; the number of samples with multiple residues of 7 pesticides exceeded 25% of the total samples [61]. Similar result was observed in north-central agricultural areas of Chile [62], where two or more pesticide residues were detected in 65% of the 118 leafy vegetable samples. These results were attributed to the frequent applications of different types of pesticides, the short time intervals between successive application, and the retention capacity of pesticides adhering to leafy vegetable surfaces.

As shown in Figure 4, we can conclude from the proportions of pesticide detection rate in different types of leafy vegetables that the detection rate of acetamiprid in chicory (31.6%) was the highest, followed by dimethomorph in chicory and leaf lettuce and paclobutrazol in leaf lettuce (26.3%) and emamectin benzoate and abamectin in leaf lettuce (25%). Except for Chinese cabbage, the detection rates of dimethomorph were either first or second among other leafy vegetables; however, the concentrations did not exceed the MRL. Consequently, it can be inferred that dimethomorph with low toxicity has been widely used in greenhouse vegetables in spring and winter in Liaoning province. This finding is similar to the results of Yao’s study on vegetable residues in Hubei province, China. His findings showed that although no samples exceeded the MRL for dimethomorph, the detection rates of dimethomorph were higher in winter (23.9%) and spring (11.8%) than in summer (4.9%) and autumn (4.6%) [22]. This phenomenon can be attributed to three primary factors: First, as a low-toxicity fungicide developed by BASF, dimethomorph has been widely available in the Chinese market for a long time. With its broad spectrum of sterilization compounds and increasing sales volume, dimethomorph has emerged as the leading morpholine fungicide in China, which is now the largest producer of morpholine worldwide. Second, this study mainly collected greenhouse leafy vegetables in the temperate monsoon climate of Northeast China in the cold seasons, when leafy vegetables are prone to downy mildew and other diseases due to low temperature and high humidity. The combination of dimethomorph and protectant fungicides, including metalaxyl, mancozeb, and chlorothalonil, has a synergistic effect and can delay resistance development. As a result, Chinese farmers frequently use it as their preferred choice for broad-spectrum pest and disease control. Third, because of the low toxicity, the MRL of dimethomorph specified in GB 2763-2021 [47] is higher. Applying a safe dose of this fungicide on vegetables can ensure that pesticides do not exceed the MRL through expected degradation before reaching the market. However, the low toxicity of dimethomorph makes it easy for vegetable growers to abuse its use, which correspondingly increases the detection frequency of this pesticide residue.

3.3. Dietary Risk Assessment

3.3.1. Chronic Dietary Risk Assessment

The national estimated long-term exposure and chronic dietary risk assessment of 41 pesticides detected in leafy vegetables (based on Formulas (1) and (2)) are shown in

Table 5. $NEDI$ and $RQc$ of the detected pesticides of greenhouses leafy vegetables in Liaoning province.

Pesticide	ADI (mg/kg bw/day)	ND = 0		ND = LOD
		NEDI (μg/kg bw/day)	RQc (%)	NEDI (μg/kg bw/day)	RQc (%)
Dimethomorph	0.20	0.470	0.235	0.474	0.237
Carbendazim	0.03	0.155	0.517	0.162	0.540
Procymidone	0.10	0.084	0.084	0.095	0.095
Acetamiprid	0.07	0.026	0.037	0.037	0.047
Thiamethoxam	0.08	0.012	0.015	0.020	0.024
Imidacloprid	0.01	0.046	0.463	0.054	0.536
Difenoconazole	0.06	0.009	0.014	0.014	0.024
Pyridaben	0.01	0.023	0.229	0.030	0.299
Bifenthrin	0.01	0.070	0.701	0.088	0.883
Chlorantraniliprole	0.01	0.027	0.271	0.034	0.344
Emamectin benzoate	0.03	0.003	0.011	0.011	0.036
Profenofos	0.0005	0.010	1.908	0.020	3.993
Paclobutrazol	0.03	0.005	0.018	0.012	0.038
Abamectin	0.1	0.007	0.007	0.015	0.015
Pyraclostrobin	0.001	0.006	0.628	0.014	1.370
Chlorpyrifos	0.03	0.021	0.071	0.030	0.101
Fenpropathrin	0.03	0.006	0.020	0.015	0.049
Propamocarb	0.4	0.026	0.006	0.030	0.008
Pyrimethanil	0.20	0.069	0.034	0.076	0.038
Triadimefon	0.03	0.061	0.204	0.072	0.241
Methamidophos	0.004	0.001	0.026	0.012	0.305
Carbofuran	0.001	0.015	1.508	0.023	2.254
Metalaxyl	0.08	0.002	0.002	0.009	0.012
Cyhalothrin	0.02	0.008	0.039	0.026	0.132
Azoxystrobin	0.20	0.006	0.003	0.014	0.007
Diflubenzuron	0.02	0.004	0.021	0.009	0.043
Cyromazine	0.06	0.004	0.007	0.009	0.015
Prochloraz	0.01	0.001	0.014	0.009	0.088
Cypermethrin	0.02	0.005	0.026	0.024	0.120
Fenvalerate	0.02	0.004	0.021	0.023	0.115
Triazophos	0.001	0.006	0.604	0.017	1.731
Pendimethalin	0.1	0.002	0.002	0.009	0.009
Chlorfluazuron	0.005	<0.001	0.003	0.008	0.153
Iprodione	0.06	0.003	0.005	0.018	0.029
Dichlorvos	0.004	0.001	0.014	0.012	0.296
Chlorbenzuron	1.25	<0.001	0.001	0.008	0.001
Forchlorfenuron	0.07	0.001	0.002	0.008	0.011
Dimethoate	0.002	<0.001	0.008	0.011	0.574
Fenitrothion	0.006	0.002	0.034	0.013	0.222
Methomyl	0.02	<0.001	0.001	0.008	0.038
Carbaryl	0.008	<0.001	0.001	0.008	0.095

. The $RQc$ values of 41 pesticides were low, ranging from <0.001% to 3.993%. When ND = 0, the long-term exposure level was observed to be <0.001–0.470 μg/kg bw/day. Profenofos and carbofuran exhibited higher $RQcs$ (>1%) at 1.908% and 1.508%, respectively. When ND = LOD, the $RQcs$ ranged from 0.008 to 0.474 μg/kg bw/day, and the $RQc$ values of profenofos, carbofuran, triazophos, and pyraclostrobin were relatively high at 3.993%, 2.254%, 1.731%, and 1.370%, respectively, but all values remained below the threshold of concern (<100%). The total $RQc$ values of all residues were 7.81% (ND = 0) and 15.17% (ND = LOD), which were within acceptable limits (<100%). This indicated that the risk of chronic dietary exposure of Liaoning residents to leafy vegetables in the cold seasons was low and within the acceptable range. The results of Yu’s study are consistent with our research. In Changchun, located in the same temperate region as Liaoning province, Yu et al. (2016) investigated the concentrations of organophosphorus pesticides (OPs) in fresh vegetables and estimated the potential health risks to local residents. The results showed that leafy vegetables had higher OP concentrations than non-leafy vegetables; however, the residents were not at risk of dietary exposure to OPs [4].

According to the published literature, the chronic risks of pesticide residues in vegetables are generally within acceptable levels and have no health effects. Sun et al. (2017) analyzed and evaluated 55 pesticide residues in six leafy vegetables in Guangdong province in 2014 and 2015. The detection rates of pesticides ranged from 1.7% to 36.1%, and the dietary exposure risks were all within the acceptable range [63]. Si et al. (2021) used a pesticide screening database to investigate the residues of multiple pesticides in raw vegetables in Shanghai, southern China. The hazard index of different groups ranged from 3.7% to 10.6%, indicating that the chronic dietary risk of raw vegetables was low and safe for consumers of different age groups [64]. A total of 2319 vegetable samples, collected from the distribution channel in Delhi and surrounding areas, were analyzed for 155 multi-class pesticides. The hazard quotient and hazard index did not exceed the unit value for both adults and children, indicating that the levels of pesticide residues in these vegetables did not pose any risk to consumers [65]. The pesticide residues in leafy vegetables collected from open-air markets, greenhouses, and wholesale vegetable markets in Turkey were analyzed, and the detection rate of pesticides was 57.6%. However, no long-term health risks to consumers were identified [66]. Similarly, risk assessments of pesticide residues in vegetables in many regions have indicated acceptable levels of chronic risks [67,68,69,70,71,72]. In this study, the chronic dietary risk of greenhouse leafy vegetables was acceptable even during the cold seasons in the temperate zone of northeast China. However, there are exceptional circumstances. Ssemugabo et al. (2022) used point estimation methods to assess health risks in different age groups in Uganda. Hazard coefficients for dichlorvos, alanycarb, fonofos, fenitrothion, dioxacarb, and benfuracarb were all greater than 1, suggesting that consumers may face chronic health risks. Notably, the risk was highest among the younger participants [73]. Furthermore, Ssemugabo et al. also explained in the article that there was no systematic monitoring of pesticide residues and no assessment of associated risks in fruits and vegetables in Uganda. In conclusion, the chronic dietary risk associated with pesticide residues in vegetables is generally low on a global scale, but is more likely to exist in some regions where the risk of pesticide residues has not been addressed at the national level.

3.3.2. Acute Dietary Risk Assessment of Two Major Leafy Vegetables

Compared with open-field cultivation, the relatively closed ecological environment of greenhouse cultivation leads to a higher temperature, higher humidity, and less ventilation in the greenhouse [74,75]. However, adopting the application methods of open-field cultivation for the control of plant diseases and pests is widespread. Research results have shown that most pesticides degrade more slowly and have longer half-lives in greenhouses, especially in the cold seasons [76]. Therefore, leafy vegetables cultivated at this time may contain higher pesticide residues, and consumption may correspond to higher short-term acute dietary exposures. In order to evaluate the acute dietary risk of pesticides, a relevant assessment was performed on Chinese cabbage and pakchoi, which are the main types of leafy vegetables consumed in the cold seasons in northeast China.

Based on the JMPR database [52], the $ARf\mathrm{D}$ values of chlorantraniliprole and azoxystrobin are unnecessary for the 12 pesticides detected in Chinese cabbage samples, whereas those of other pesticides are shown in

Table 6. Acute risk assessment for the detected pesticides of Chinese cabbage in Liaoning greenhouses.

Pesticide	ARfD (mg/kg bw)	HR (mg/kg)	IESTI (μg/kg bw)	RQa (%)
Methomyl	0.02	0.100	3.020	15.099
Fenpropathrin	0.03	0.131	3.956	13.187
Methamidophos	0.1	0.031	0.936	0.936
Propamocarb	2	0.380	11.476	0.574
Profenofos	1	0.141	4.258	0.426
Procymidone	0.10	0.010	0.302	0.302
Imidacloprid	0.40	0.028	0.846	0.211
Difenoconazole	0.30	0.016	0.483	0.161
Carbendazim	0.50	0.022	0.664	0.133
Pyraclostrobin	0.70	0.015	0.453	0.065

; the $RQas$ decreased in the following order: methomyl > fenpropathrin > methamidophos > propamocarb > profenofos > procymidone > imidacloprid > difenoconazole > carbendazim > pyraclostrobin. The $RQa$ values of fenpropathrin (13.187%) and methomyl (10.099%) were higher than those of the remaining eight positive pesticides (0.065%–0.936%) (Table 6). In summary, the $RQa$ values were significantly below 100%, indicating an acceptable level and low risk of pesticide residues in Chinese cabbage in the cold seasons in Liaoning province.

According to the JMPR database [52], among the 27 positive pesticides detected in pakchoi samples, the $ARf\mathrm{D}$ levels of chlorantraniliprole, diflubenzuron, metalaxyl, iprodione, pyrimethanil, and forchlorfenuron are unnecessary, whereas those of the other 21 pesticides are shown in

Table 7. Acute risk assessment of the detected pesticides of pakchoi in Liaoning greenhouses.

Pesticide	ARfD (mg/kg bw)	HR (mg/kg)	IESTI (μg/kg bw)	RQa (%)
Triazophos	0.001	0.366	5.805	580.529
Abamectin	0.005	0.27	4.283	85.652
Bifenthrin	0.01	0.33	5.234	52.343
Pyridaben	0.01	0.311	4.933	49.329
Chlorpyrifos	0.1	0.91	14.434	14.434
Carbendazim	0.1	0.73	11.579	11.579
Dimethomorph	0.5	2.944	46.696	9.339
Acetamiprid	0.6	2.706	42.921	7.154
Imidacloprid	0.1	0.24	3.807	3.807
Cyromazine	0.4	0.83	13.165	3.291
Fenvalerate	0.1	0.16	2.538	2.538
Prochloraz	0.2	0.307	4.869	2.435
Fenpropathrin	0.1	0.13	2.062	2.062
Propamocarb	0.03	0.03	0.476	1.586
Profenofos	2	1.446	22.936	1.147
Difenoconazole	1	0.62	9.834	0.983
Emamectin benzoate	0.3	0.15	2.379	0.793
Pyraclostrobin	0.03	0.008	0.127	0.423
Methamidophos	0.7	0.084	1.332	0.190
Paclobutrazol	0.1	0.01	0.159	0.159
Thiamethoxam	0.1	0.01	0.159	0.159

. The $RQas$ of pakchoi samples were higher than that of Chinese cabbage samples. Triazophos demonstrated an $RQa$ exceeding 100%, while those of abamectin, bifenthrin, pyridaben, and chlorpyrifos were 85.652%, 52.343%, 49.329%, and 14.434%, respectively (Table 7). Therefore, the acute dietary intake risk of pesticides in pakchoi samples in Liaoning province during the cold seasons is relatively high, which requires further investigation.

Fang et al. (2015) conducted an acute risk assessment on 300 celery samples from eight provinces in China. The most prominent $RQa$ values were those of triazophos (28.0%), followed by procymidone (24.4%), carbendazim (19.0%), abamectin (9.6%), and cyhalothrin (6.5%) [77]. The findings of their research are consistent with the results of our study, as both triazophos and abamectin ranked among the top five risks. This phenomenon can primarily be attributed to the following two main reasons: i) they are often sprayed in vegetable production in China, hence the residue phenomenon is common, and ii) compared to other detected pesticides, their lower $ARf\mathrm{D}$ (0.001 and 0.005 mg/kg bw, respectively) and relatively higher toxicity resulted in elevated $RQa$ values and greater associated risks after calculation.

From experimental results in the literature, the chronic exposure risks remain relatively consistent (all low), while acute risks exhibit variability with some being high and others being low. In Algeria, short-term exposure assessments of 160 samples showed that acute reference doses were exceeded in seven pesticide/commodity combinations, including three pesticides (chlorpyrifos, deltamethrin, and cyhalothrin) [78]. Gad Alla et al. (2015) analyzed 215 compounds from different pesticide chemical groups in 116 vegetable samples and estimated acute dietary exposure. The findings revealed that carbendazim and methomyl exhibited a potential risks to children, while methomyl posed a potential risk to adults [79]. An acute risk assessment study on Polish vegetables also showed potentially hazardous contingencies [80]. Zuo et al. (2021) analyzed 105 pesticide residues in 682 leafy vegetable samples from different sampling channels in Hebei province, China. The acute dietary exposure risks of the 12 pesticides detected ranged from 0.01% to 1.61%, which were well below 100% [81]. Acute risk assessments by Zentai et al. (2016) and Blaznik et al. (2016) also found no health concerns [82,83]. Our study showed that there was no acute risk of dietary exposure from Chinese cabbage, while pakchoi presented an acute risk of dietary exposure. In conclusion, the level of acute risk of pesticide residues in vegetables is not entirely consistent, ranging from acceptable to potential health risks. At the same time, acute risks of pesticide residues are more likely to occur than chronic risks.

3.4. Uncertainties in the Assessment of Dietary Exposure

Dietary exposure assessment is a relevant criterion for risk assessment, with differences due to pollution levels, dietary habits, and global exposures. The outcome of the risk assessment relies on several factors. Firstly, it is related to the samples collected, including the growing area and environment, as well as the type, frequency, and manner of pesticide application. Additionally, it encompasses the types and quantities of samples gathered. Secondly, laboratory analysis plays an important role, involving the examination of pesticides in terms of their types and quantities using analysis methods with defined detection limits. Thirdly, the dietary habits of different consumer groups (gender, age), including consumption amount and consumption pattern, can affect results. Finally, assessment considerations include the selection of appropriate calculation models and the substitution of undetected data.

In this study, we propose several uncertainty analyses to find out the main sources of uncertainty and to improve the credibility of our research results. Firstly, due to the limited amount of data in this study, only a point assessment method was employed for dietary risk assessment, instead of utilizing the probabilistic assessment method, which can provide confidence intervals for the results. When the point assessment outcomes indicate an unacceptable dietary risk associated with a pesticide (such as triazophos), further investigation of exposure variability through probabilistic assessment should be pursued; however, this necessitates additional data support. Secondly, the monitoring results of pesticide residues in this study have revealed that certain vegetable samples had combined contamination from multiple pesticide residues. It is noteworthy that when multiple pesticide residues share the same mechanism of action, they often exhibit additive effects in terms of toxicity. Currently, the US EPA has established five common mechanism groups (CMGs): organophosphates, N-methyl carbamates, triazines, chloroacetanilides, and pyrethrins/pyrethroids [84]. However, it should be recognized that the insecticides and fungicides with high detection rates in this study belong to different mechanisms of action; thus, the above five cumulative exposure groups cannot comprehensively cover them. Consequently, only single-pesticide dietary exposure assessment can be employed in our research which may lead to underestimation. However, at the national level, due to the limitations of the knowledge level of pesticide toxicology, the availability of data, and the actual economic significance, the grouping methods and evaluation criteria of compounds with the same mechanism of action are still under discussion. Therefore, further cumulative dietary exposure risk assessment becomes imperative.

The risk assessment of pesticide exposure through consumers’ diets is of paramount importance, because food consumption serves as the primary route for human exposure to environmental contaminants. However, it should be acknowledged that this process is neither straightforward nor highly accurate, and thus requires gradual improvement and refinement.

4. Conclusions

In this study, 65 pesticide residues in greenhouse leafy vegetables cultivated during cold seasons in Liaoning province were monitored for three years. Of the 469 samples, 48.6% were positive for at least one pesticide detected, while 3.6% exceeded the MRLs set by the Codex and GB 2763-2021. The pesticide residues in the different types of leafy vegetables varied from 0.2% to 11.9%, and the top four pesticides in terms of detection rate were dimethomorph, carbendazim, procymidone, and acetamiprid. Four banned pesticides in vegetables in China, namely, carbofuran, chlorpyrifos, triazophos, and dimethoate, were identified. The chronic dietary exposure risks of 41 detected pesticides were generally low, ranging from <0.001% to 3.993%. Among the two mainly consumed leafy vegetables during the cold seasons in northeast China, Chinese cabbage showed a low risk of acute dietary exposure to pesticide residues, whereas pakchoi showed a relatively higher level of potential acute toxicity to consumers. This study reflects the overall contamination of pesticide residues in greenhouse-grown leafy vegetables in cold seasons and potential dietary exposure for consumers.

This study represents a single-point assessment of pesticide residues in leafy vegetables, focusing on one particular vegetable variety, and there is an inherent uncertainty for the general population. However, the findings could also contribute to enhancing consumer awareness of food quality and safety and help pesticide suppliers and policy makers to take necessary action. First, it is critical for regulators to prioritize potential risks associated with pesticides that exceed MRLs and lack ADI reference values. This will require revising existing good agricultural practices (GAPs), especially for high-risk pesticides such as carbofuran, chlorpyrifos, triazophos, and dimethoate, which are prohibited in Chinese vegetable production. Furthermore, while our study suggests that the majority of pesticides found in greenhouse leafy vegetables posed a low risk to Chinese consumers, it is important to note that other studies have identified certain leafy vegetables with unacceptable acute and chronic risks associated with pesticide residues, particularly for infants and children. Therefore, future dietary exposure risk assessments should take into account different consumer groups (including children, women, and the elderly), evolving food consumption patterns, and the potential accumulation of pesticides in the human body. At the same time, based on the continuous development of assessment techniques, this can be achieved by adopting cumulative exposure assessments of mechanisms of action for a wider range of pesticides and by utilizing assessment models that can provide confidence intervals and uncertainty levels, such as probabilistic risk assessment models, to ensure accurate evaluations.