Artigos

Introduction
Like in most tropical countries. Colombia has an important Agricola economic sector. as it is characterized by having rich thermal floors and abundant water sources (1). However. agricultural systems are mainly monocultures. performed on a small scale that limits the agroecosystem to a unique plant. producing an extreme imbalance in homeostasis and favoring the appearance of pests (2). The technical response to this production model is the intensive use of pesticides (2,3). In previous research. we found that pesticide application practices include the use of “cocktails” which contain up to seven pesticides (4). In addition. we found that the most sold pesticides were organophosphates (OP) (chlorpyrifos. profenophos. methamidophos). synthetic pyrethroid insecticides (cypermethrin and permethrin). and fungicides (mancozeb. propined. and metalaxyl) (4,5). As a result. Colombia is one of the countries with the highest pesticide use per hectare of cultivated land in Latin America (6).
The exposure to pesticides represents an important problem to public health because it may increase the risk of adverse health effects (7). Previous studies have suggested that children are exposed to pesticides as a result of child labor in agricultural activities. spraying of pesticides in. or near. the home and school environment. the consumption of contaminated agricultural products or drinking water. and para-occupational exposure from parental pesticide use (8,9,10,11). Studies performed in both agricultural communities and general populations. have assessed children’s exposure to different pesticides. using general and specific urinary metabolites as biomarkers of exposure (12,7). However. few studies in Latin-America have used pesticide-specific metabolites to evaluate the multiple exposures to the parent compounds (13,14,15).
The measurement of urinary pesticide metabolites typically reflects recent exposure because of their short biological half-lives. Therefore. measuring these urinary biomarkers can provide useful information on recent pesticide exposure (1–3 days) (16). To our knowledge. to date. no studies have been performed to quantify children’s pesticide exposures in rural Colombia. For these reasons. we aimed to: 1. evaluate urinary pesticide metabolite concentrations among school-aged children in rural area of Bogotá. Colombia. and. 2. investigate what exposure determinants were associated with these concentrations.

Method
Study design and population
The study design was cross-sectional. During 2019. children from 7 to 10 years of age were recruited from the town of Sumapaz and Usme. which is located southeast of Bogotá – Colombia. mainly in rural areas with a high rate of agricultural production. (Figure 1).

Fig. 1
Figure 1. Map of the territories of Usme and Sumapaz. Bogota. Colombia (5).
In previous publication we described the characteristics of the territories and population. This research applied the following election criteria: being 7 to 10 years old at enrollment. studying at one of the primary schools in the study are. being born from a mother who resided in the rural areas of Usme and Sumapaz. Finally. 279 complied the inclusion criteria. and 231 (84%) children provided urine samples (Figure 2). This research was approved by the Ethics Committee of the EAN University and the Integrated South Health Services Subnetwork. All mothers gave written informed consent prior to their children’s participation. and children gave verbal informed consent.

Fig.2
Characterization of children’s exposure determinants
We geographically referenced each household and applied a structured questionnaire to the children’s mothers to obtain information about possible determinants of exposure; we modified a questionnaire previously used in a study in Costa Rica. and validated the adapted version before applying it (16). (5). The instrument included 100 questions divided into the following sections (Supplementary Material S1): sociodemographic data. pregnancy information. occupational information. eating habits. occupational exposure (during pregnancy and current). and housing conditions. The questions about pesticide use were (5):
¿Do you (mother) currently apply pesticides at work? ¿During the past year. have you (mother) or anyone used pesticides in your home? ¿During the past year. do you know if pesticides have been applied to crops near your child's school? Do you currently live with people who work in a place where pesticides are applied or used?
Biomonitoring of pesticide
Urinary pesticide metabolites sampling and analysis
Urine samples were collected in 50mL beakers (Kramer®. sterile). and transferred to 15mL tubes (BRIXCO®. sterile). These tubes were labeled with specific codes that included information about the territory. the ID of the family. and the date of sample collection. Finally. the samples were stored at –20°C until shipment (4°C) to Lund University. Sweden. for analysis.
All samples were analyzed using a modified method described by Norén et al 2020. Briefly. for the analysis. the urine samples were de-conjugated using ?-glucuronidase/arylsulfatase and diluted prior to quantitative analysis using liquid chromatography - triple quadrupole linear ion trap mass spectrometry (LC/MS/MS; QTRAP 5500; AB Sciex. Framingham. MA. USA). Norén et al 2020, in the urine samples 13 pesticide metabolites or parent compounds were reported. 2-isopropyl-4-methyl-6-hydroxypyrimidine (IMPy; a metabolite of the OP diazinon). 3.5.6-trichloro-2-pyridinol (TCPy; a metabolite of the OP chlorpyrifos). 4-Bromoo-2-chlorophenol (BCP; a metabolite of the OP Profenophos) Malathion dicarboxylic acid (MaD; a metabolite of the OP Malathion). 3-phenoxybenzoic acid (3PBA; an unspecific metabolite to several pyrethroids). 4-fluoro-3-phenoxybenzoic acid (4F3PBA; a metabolite to cyfluthrin). cis and trans 3-(2.2-dichlorovinyl)-2.2-dimethylcyclopropanecarboxylic acid (cis/trans DCCA; an unspecific metabolite to several pyrethroids). chloro-3.3.3-trifluoro-1-propen-1-yl]?2.2-dimethylcyclopropanecarboxylic acid) (CFCA; a metabolite to Bifenthrin). 2.4-dichlorophenoxyacetic acid (2.4-D). 5-hydroxy-tiabendazole (OH-T; a metabolite to thiabendazole). 3-hydroxy-pyrimethanil (OH-P; a metabolite to pyrimethanil). hydroxy tebuconazole (TEB-OH; a metabolite to tebuconazole). 2.4-Dichlorophenol (DCP; a metabolite from multiple sources e.g. metabolite of dichlorobenzene. the pesticide 2.4D) (17).
Urinary pesticide metabolite concentrations that were below the LOD but above LOD/2 were imputed with the value indicated by the analytical equipment; samples below LOD/2 were set at the value of LOD/2 (18). In addition. the specific gravity (kg/L) (sg) of each urine sample was measured with a digital refractometer. and pesticide metabolite concentrations were corrected for dilution by Msg = M * [(1.016–1)/(SG-1)]. where Msg is the specific gravity corrected metabolite concentration (?g/L). M is the observed metabolite concentration (?g/L). SG is the specific gravity of the urine sample. and 1.016 kg/L is the average specific gravity for all urine samples included in this study (n = 231).
Statistical analysis
Nominal and ordinal variables were described using proportions. The continuous variables. according to the Kolmogorov-Smirnov normality test. were described with medians and interquartile ranges.
Bivariate associations were investigated between exposures. covariates. and pesticide metabolite level with Chi-square and Mann Whitney tests. and using simple linear regression models. To establish associations between pesticide variables and their concentrations. a separate multiple linear regression was used with log10-transformed metabolite levels as the dependent variables and variables associated with this metabolite at p < 0.1. In the simple regression model, since in this range for multivariate models, biases in the results are avoided. In this analysis, Regression analyses of pesticides whose geometric measurements were below the LOD were excluded (see table 2). The models were fitted with factors that predict pesticide concentrations in children's urine samples, such as sex and age. They are also important factors when defining the sample collection for this work.
The results of the adjusted and unadjusted models were compared. and we evaluated the robustness of the results of those models with a statistically significant association between determinants of pesticide exposure with the outcome (log-10 transformed exposure biomarker concentrations). We evaluated Cook's distance and studentized residuals by two standard deviations. We already evaluated if residuals followed a normal distribution by plotting histograms of normality and applying the Kolmogorov-Smirnov test on standardized residual values. Finally. multicollinearity was evaluated considering a variance inflation factor of less than 10. and the independence of the variables was evaluated with the Durbin-Watson statistic. Associations were considered significant if p<0.05. Data analyses were run using the software SPSS version 25th.
Results
The median age of the children was 8.7 years (p25-75 = 8.0–9.6), and it was similar for boys and girls (Table 1). Almost a quarter (24%) of mothers had a low educational level. In approximately half of the cases (53%), the family income was below the current legal minimum wage (~ 225 USD per month). Regarding the use of pesticides, the families were exposed due to the work that the mother does in the local agri-food crops and were also exposed environmentally due to the proximity of the homes to the crops in the study area. Based on these variables, A large percentage of the study showed that both boys and girls were exposed to pesticides through their family environment or through crops near their schools (see Table 1).

Tab.1

The results presented in Table 2 show that all 13 pesticide metabolites were detected in part of the children's urine samples. Chlorpyrifos and 2,4-D were detected in all samples, followed by profenofos. 3-PDA. OH-pyrimethanil and DCP were detected in 99% of the samples. Unlike malathion, OH-thiabendazole and IMPy were detected in less than 30% of the children's samples (Table 2). These results show that even in Colombia, these pesticides are used regularly, which, as a rule in the SANTE guide (regulation of pesticides), are controlled and even prohibited. We observed the highest median (p25–75) for TCPy 4.3 (2.1–8.1), followed by profenofos 2.77 (0.99–7.19). and OH-pyrimetamil 1.37 (0.4–3.52).
We found para-occupational exposure variables. pesticide use at home. and pesticide use at school were associated with increased urinary pesticide biomarker concentrations. Regarding the regression models. The main findings of the adjusted models were that Chlorpyrifos concentrations increased significantly due to the mother's participation in activities with pesticides (? = 0.13. CI95% 0.04 - 0.21). In addition. this increase would be also observed if other relative in the home (? = 0.13. CI95% 0.05 - 0.20) used pesticides. Finally. the use of pesticides around the house increased urinary metabolite concentrations in children. Profenophos was increased significantly due to work activities related to the mother in the agribusiness (?: 0.14 CI95% -0.0006 - 0.28) and if she used pesticides during her work (?: 0.16 CI95% 0.0006 - 0.33) (Table 3).

Tab.2

Tab.3

Regarding urinary metabolites of pyrethroids. we found a higher concentration of 3-PBA in infants in whose mothers work in agro-industrial activities (? = 0.115. CI95% 0.010 - 0.219). Regarding the DCCA trans metabolite. we found an increase in its concentrations if someone who uses pesticides lives in the house or if the mother used pesticides. Statistically. significant differences were observed compared with those who did not. Finally. the concentrations of DCP increased with the use of pesticides around the school (? = 0.14. CI95% 0.024 - 0.25). The concentrations of other pesticides were not associated significantly with any determinants (Table 3). The variables ‘child participates in agricultural activities’ and ‘distance from home to nearest agricultural field’ show null associations with pesticide metabolite concentrations (Supplementary material 1).
Discussion
The present work analyzed the different levels of specific and non-specific pesticide metabolites in the urine of school-aged children from the rural area of Bogotá. Colombia. and their relationship with environmental and occupational determinants. The main finding was the high proportion of infants with measurable concentrations of pesticides. Organophosphate pesticides such as chlorpyrifos and profenophos presented the highest concentrations; which can be explained since these pesticides are widely used in agricultural production in the region (5). Additionally. others groups of pesticides. such as pyrethroids. conazoles. and herbicides were also frequently detected. but in lower concentrations. This may be because these are used less frequently in the study area.
Comparing our findings to research among child populations in rural areas of Latin America. our adjusted measure for creatinine showed considerable higher concentrations of TCPy (4.52 ng/g) than Nicaragua. which studies reported 3.7 ng/g (15). Additionally. when comparing the unadjusted mean concentrations. we also found concentrations higher than those reported for children in rural communities of Costa Rica (19,13). Likewise. recent publications showed concentrations of pesticides in children in several countries like Sweden. Germany. France. U.S. Spain and Thailand in which found lower urinary concentrations of TCPy than those founds in Colombia (21,15,9,17). Urinary Concentrations of TCPy only had higher mean levels in a Child Population in North Carolina and Rice farming communities in Thailand (22,23).
Regarding the levels of pyrethroids metabolites in rural areas ours results showed the highest concentrations. For example. the concentrations of 3-PDA (0.62?ng/mL) when compared with the levels observed in children from Canada. Germany. France. the US. Spain. and Thailand (0.02–0.29?ng/mL) (21,15,9). In contrast. Costa Rica reported the highest concentrations than our results (0.8 ng/ml). which may be for exposure the children to pesticides used to control vector-borne diseases. a condition that does not exist in the territories of our study (13).
Others finding of this study was maternal and family occupation in activities with pesticides predicted urinary metabolite concentrations of OPs (chlorpyrifos and profenophos) and pyrethroids. This result was similar to other studies. In Thailand. the children being a member of a rice farming family tended to increase the concentration of TCPy (23) . Furthermore. parental occupational exposure to pesticides has been considered as a predictive factor for higher exposure of their children (11). Also. higher concentrations of organophosphorus metabolites were found in children living with parents who are engaged in conventional agriculture activities compared to those who are not (23,24,25) . In Canada. a study also found that the children who lived in a family that reported any use of pesticides in the previous month had higher levels of pyrethroid metabolites in urine than those that reported no use (26).
Regarding the association between living near the crops and the increase of metabolites of pesticides. studies among Thai children determined that children living near rice farms had significantly higher levels of the pyrethroid metabolite DCCA and TCPy. especially in aquaculture participants during the high exposure season (27,11). In contrast. Van Wendel and co-workers. in a study conducted in Costa Rica determined that 3-PBA concentrations were similar in children living in conventional banana plantation villages than in organic villages may be for recent use of this pesticide in vector control (13).
Finally. we found the increase of DCP to result of its use around the school. Hernandez et al. 2019 proposed that children residing in agricultural areas may also be exposed to pesticides as a result of drifting from the application areas to household environments where children live.
Finally. it is necessary to highlight that this widespread exposure exists for the historical and social conditions generated by an agribusiness model of production that promotes the continuous use of pesticides (28). Additionally. in Colombia. talking about rurality means understanding the economic conditions of this population. which are evidenced by low economic income. structural elements related to families with low education. and factors of insecurity (29).
This study motivates. among other things. to generate a space for reflection. knowledge. and empowerment of communities. which implies processes of environmental education and health education in the community and school contexts. as well as promoting awareness and changing attitudes and practices through of the articulation and application of strategies from various disciplines in order to generate awareness and guide critical thinking regarding the care of life. health and the environment with a vision of sustainable development.
This study had several limitations. A cross-sectional study was carried out that did not permit the analysis of exposure over time and. our samples consisted of unique spot urine samples. Therefore. we were unable to study variability in and between the children from this study. This situation may be underestimated daily average concentrations (30). Even with these limitations. it was possible to detect concentrations of all of the studied pesticides.
In addition. it should be considered that the study investigated the determinants of exposure through the survey carried out on the mothers. a situation that could lead to underestimation of exposure due to undervaluation or concealment of information by the source. Nevertheless. data interpretation was performed cautiously so as not to have biased information. for this reason. the instrument was standardized for the application of the survey and collection.
On the other hand. the use of pesticides in Colombia had had practices how mixtures with various active ingredients of pesticides. a common practice to ensure the effectiveness of pest control. This study was limited to some pesticides. But mancozeb. the most widely used in Colombia. could not be analyzed.
Conclusion
In the school population of the rural area of Bogotá. specifically in the locality of Sumapaz and Usme where agricultural practices are carried out. widespread exposure to multiple pesticides was demonstrated. and one of the highest concentrations of pesticides were the pesticides of the family of organophosphates and pyrethroids.
In addition. there were determining factors that increased the practices of use by parents. The identification of pesticides by measuring proper biomarkers and its determinants are necessary to propose strategies to promote health and safety in the workplace and contribute to programs and policies that aim to reduce exposure and improve the quality of health care.
On the other hand. the study determines the relationship between the concentration of pesticides in urine and the environmental exposure to pesticides that schoolchildren have in the locality of Sumapaz and Usme in Bogotá.
Conflict of interest
The authors declare no conflict of interests.

Acknowledgments
The authors would like to thank the Ministry of Science. Technology and Innovation for the financial support provided through project code 1223-777-57906 - contract 619-2018. Additionally. The main author would like to thank the Coordination for the Improvement of Higher Education Personnel of Brazil (CAPES) for the doctoral scholarship.
References
1. OECD. Review of Agricultural Policies: Colombia. ; 2015.
2. Almeida J. A industrialização da agricultura na usina Nova América. In A extinção do arco-íris: ecologia e história. Rio de Janeiro: Centro Edelstein de Pesquisa Social; 2008. p. 35-56.
3. Gunier R. Bradman A. Harley K. Eskenazi B. Will buffer zones around schools in agricultural areas be adequate to protect children from the potential adverse effects of pesticide exposure? PLOS Biology. 2017; 15(12).
4. Benavides J. Lozada M. Práticas de pulverização entre os produtores de batata. conhecimentos e crenças sobre pesticidas e saúde. Tempus Actas De Saúde Coletiva. 2014; 8 (2): p. 93-108.
5. Benavides-Piracón J. Hernández-Bonilla D. Menezes-Filho J. van Wendel de Joode B. Lozada Y. Bahia TC. et al. Prenatal and postnatal exposure to pesticides and school-age children's cognitive ability in rural Bogotá. Colombia. Neurotoxicology. 2022 Mar ; 16(90): p. 112-120.
6. FAO. Plaguicidas comercio. [Online].; 2019 [cited 2020 08 10. Available from: http://www.fao.org/faostat/es/#data/RT.

7. González-Alzaga B. Romero-Molina D. López-Flores I. Giménez-Asensio M. Hernández A. M. L. Urinary levels of organophosphate pesticides and predictors of exposure in pre-school and school children living in agricultural and urban communities from south Spain. Environ Res. 2020 Jul.
8. Chetty-Mhlanga S. Fuhrimann S. Basera W. Eeftens M. Röösli M. Dalvie M. Association of activities related to pesticide exposure on headache severity and neurodevelopment of school-children in the rural agricultural farmlands of the Western Cape of South Africa. Environ Int. 2021; 146.
9. Panuwet P. Prapamontol . Chantara S. Barr DB. Urinary pesticide metabolites in school students from northern Thailand. Int J Hyg Environ Health. 2009 May; 212(3): p. 288-297.
10. Lu C. Fenske R. Simcox N. Kalman D. Pesticide exposure of children in an agricultural community: evidence of household proximity to farmland and take home exposure pathways. Environ Res. 2000; 84(3).
11. Rohitrattana J. Siriwong W. Tunsaringkarn T. Panuwet P. Ryan P. Barr D. et al. Organophosphate pesticide exposure in school-aged children living in rice and aquacultural farming regions of Thailand. J Agromedicine. 2014; 19(4): p. 406-16.
12. Bradman A. Castorina R. Barr D. Chevrier J. Harnly M. Eisen E. et al. Determinants of organophosphorus pesticide urinary metabolite levels in young children living in an agricultural community. Int J Environ Res Public Health. 2011; 8(4).
13. van Wendel de Joode B. Barraza D. Ruepert C. Mora AM. Córdoba L. Öbergd M. et al. Indigenous children living nearby plantations with chlorpyrifos-treated bags have elevated 3.5.6-trichloro-2-pyridinol (TCPy) urinary concentrations. Environmental Research. 2012; 117: p. 17-22.
14. Rodríguez T. Younglove L. Lu C. Funez A. Weppner S. D B. et al. Biological Monitoring of Pesticide Exposures among Applicators and Their Children in Nicaragua. Int J Occup Environ Health. 2006; 12(4).
15. Rodríguez T. van Wendel de Joode B. Lindh C. Rojas M. Lundberg I. Wesseling C. Assessment of long-term and recent pesticide exposure among rural school children in Nicaragua. Occup Environ Med. 2012; 69(2).
16. Hernández A. Lozano-Paniagua D. González-Alzaga B. Kavvalakis M. Tzatzarakis M. López-Flores I. et al. Biomonitoring of common organophosphate metabolites in hair and urine of children from an agricultural community. Environ Int. 2019 Oct; 131.
17. Norén E. Lindh C. Rylander L. Glynn A. Axelsson J. Littorin M. et al. Concentrations and temporal trends in pesticide biomarkers in urine of Swedish adolescents. 2000-2017. J Expo Sci Environ Epidemiol. 2020; 30(4): p. 756-767.
18. Corrales Vargas A. Peñaloza Castañeda J. Rietz Liljedahl E. Mora A. Menezes-Filho J. Smith D. et al. Exposure to common-use pesticides. manganese. lead. and thyroid function among pregnant women from the Infants' Environmental Health (ISA) study. Costa Rica. Sci Total Environ. 2022 Mar; 1.
19. Islam J. Hoppin J. Mora A. Soto-Martinez M. Córdoba L. Peñaloza Castañeda J. et al. Respiratory and allergic outcomes among 5-year-old children exposed to pesticides. Thorax. 2022; 0: p. 1-9.
20. Bravo N. Grimalt J. Mazej D. Tratnik J. Sarigiannis D. Horvat M. Mother/child organophosphate and pyrethroid distributions. Environ Int. 2020; 134.
21. Morgan M. Sheldon L. Croghan C. Jones P. Robertson G. Chuang J. et al. Exposures of preschool children to chlorpyrifos and its degradation product 3.5.6-trichloro-2-pyridinol in their everyday environments.. J Expo Anal Environ Epidemiol. 2005 Jul; 15(4).
22. Rohitrattana J. Siriwong W. Tunsaringkarn T. Panuwet P. Ryan P. Barr D. et al. Organophosphate Pesticide Exposure in School-Aged Children Living in Rice and Aquacultural Farming Regions of Thailand. Journal Of Agromedicine. 2014; 19(4): p. 406-416.
23. Butler-Dawson J. Galvin K. Thorne P. Rohlman D. Organophosphorus pesticide exposure and neurobehavioral performance in Latino children living in an orchard community. Neurotoxicology. 2016; 53: p. 165-172.
24. Gibbs J. Yost M. Negrete M. Fenske R. Passive Sampling for Indoor and Outdoor Exposures to Chlorpyrifos. Azinphos-Methyl. and Oxygen Analogs in a Rural Agricultural Community. Environmental Health Perspectives. 2017; 125(3): p. 333-341.
25. Oulhote Y. Bouchard MF. Urinary metabolites of organophosphate and pyrethroid pesticides and behavioral problems in Canadian children. Environmental health perspectives. 2013 Jan; 121((11-12)): p. 1378-1384.
26. Fiedler N. Rohitrattana J. Siriwong W. Suttiwan P. Strickland PO. Ryan PB. et al. Neurobehavioral effects of exposure to organophosphates and pyrethroid pesticides among Thai children. Neurotoxicology. 2015; 48: p. 90-99.
27. Benavides Piracón J. Bahia Coelho T. Agrotóxicos y Agroindustrias: Discursos para una vida tóxica. Análisis desde la determinación social de la salud. Ciênc. saúde coletiva. 2022; 27(09).
28. Rodríguez D. Benavides J. Salud y ruralidad en Colombia: análisis desde los determinantes sociales de la salud. Rev. Fac. Nac. Salud Pública. 2016; 34(3): p. 359-371.
29. Bradman A. Kogut K. Eisen E. Jewell N. L QA. Variability of Organophosphorous Pesticide Metabolite Levels in Spot and. Environmental Health Perspectives. 2013.