Introduction

Persistent organic pollutants (POPs) are a group of chemical compounds of mainly synthetic origin, which have been widely used for various purposes. The problem surrounding these substances lies in their toxicity to both humans and biota, and their persistence in the environment, since due to their physical and chemical characteristics, their degradation is slow, causing them to remain for long periods of time in environmental compartments such as soil, sediment, air and water, as well as in the adipose tissue of living beings in which they can carry out bioaccumulation and biomagnification processes (Ashraf, 2017; Magulova & Priceputu, 2016; UNEP, 2020).

Generally, POPs include pesticide compounds for agricultural and residential use, compounds used in industrial chemical processes, while others are unintended by-products (UNEP, 2020). Exposure to these substances has been associated with various effects on human health (Guo et al., 2019), such as endocrine disruption (Schecter, 2012), decrease in birth weight (Cabrera-Rodríguez et al., 2019), neurological problems (Berghuis et al., 2015; Rocha-Amador et al., 2009) and an increased risk to other conditions such as cancer, diabetes, obesity, hypertension and cardiovascular diseases (Ennour-Idrissi et al., 2019; Ljunggren et al., 2014; Vafeiadi et al., 2015; Yang et al., 2017). Children are particularly susceptible to the effects of these types of environmental pollutants because they may be exposed through various routes, including specific routes such as breast milk. In addition, their state of development can lead to windows of vulnerability, so that children can be exposed at different stages of their growth, manifesting late effects (World Health Organization, 2006).

Organochlorine pesticides (OCPs) correspond to a group of semi-volatile and highly persistent POPs, characterized by having carbon and chlorine atoms united in their structure. These compounds have been widely used in past decades, in the agricultural sector and as controls for disease-transmitting vectors in tropical countries (Ali et al., 2020; Jayaraj et al., 2016; Sparling, 2016). Currently, the production and use of most OCPs is regulated under the Stockholm Convention agreements, whose objective since its launch in 2001 is to protect human health and the environment from POPs. Regarding OCPs, the regulatory instrument establishes the prohibition for their manufacture and use, with only dichlorodiphenyltrichloroethane (DDT) being part of the restricted use category for countries with prevalence of endemic vector-borne diseases (Magulova & Priceputu, 2016; UNEP, 2020).

OCPs can enter the environment in different ways, mainly by direct application, also through the discharge of waste contaminated with these substances into landfills, and through the discharge from the factories where they are synthesized (Jayaraj et al., 2016). Human exposure to these compounds can occur via ingestion, inhalation, or dermal absorption (Genuis et al., 2016; Jayaraj et al., 2016). Soil has been shown to be one of the most important exposure routes to POPs in general, as it serves as a reservoir from which these compounds can migrate to other environmental matrices and enter the food chain (Bidleman & Leone, 2004; Cai et al., 2008). Therefore, soil can be considered as one of the most important sources of exposure for the human population (Herrera-Portugal et al., 2005).

As part of biomonitoring in various places in Mexico, there are studies that have evidenced the presence of OCPs in different environmental matrices, biota and in human population, in relation to scenarios that involve agricultural and industrial activities, unregulated solid waste deposits, and endemic malaria zones (Cantu-Soto et al., 2011; Domínguez-Cortinas et al., 2013; Espinosa-Reyes et al., 2012; Flores-Ramírez et al., 2016, 2017; González-Mille et al., 2010; Martínez et al., 2012; Martínez-Salinas et al., 2011; Orta-García et al., 2016; Pérez-Maldonado et al., 2010, 2013, 2014; Perez-Vazquez et al., 2015; Sánchez-Osorio et al., 2017; Torres-Dosal et al., 2012; Trejo-Acevedo et al., 2009, 2012).

The city of Salamanca, located in the state of Guanajuato, in central Mexico, is characterized by its industrial and agricultural development. There is a particular case related to contamination by OCPs in the place: the facilities and surroundings of the former Tekchem Industrial Unit, in which organochlorine and organophosphate pesticides were manufactured, among other dangerous compounds. The company operated between 1943 and 2007; however, to date there is a problem of soil and groundwater contamination due to the inadequate management of processes and waste during most of the production stages (Albert & Jacott, 2015; Beltrán Hernández et al., 2019). An environmental risk assessment study at the site showed that the soil is mostly contaminated by OCPs, representing approximately 70% of the identified compounds, with DDT and its metabolites being found in the highest concentrations (SEMARNAT, 2018).

To date, the impact of the compounds present in the facilities of the former Tekchem Industrial Unit is still little known, in terms of their dispersion toward the surrounding areas, and the possible contamination of other environmental matrices, as well as the exposure that may occur in human population. In context with what has been previously described, the objective of this study is to quantify levels of OCPs in soil and in school-age children from the city of Salamanca, Mexico.

Materials and methods

Study setting and population

A cross-sectional observational study was performed in Salamanca city, Mexico, based on the history of industrial activity of the municipality, assessing environmental and biological samples in order to estimate exposure to OCPs. Salamanca city is in the central part of Guanajuato state, with an area of 774 km2 and a population of 135,874 people in the municipal seat. The climate of the place is characterized by being semi-warm with rain in summer, with an average temperature of 19.3 °C and a rainfall of 600–800 mm per year (Sistema Meteorologico Nacional, n.d.). For environmental monitoring, soil samples were taken covering the entire metropolitan area of the municipality, following the protocol established by the Official Mexican Standard NMX-AA-132-SCFI-2006 (SCFI, 2006). For biological monitoring, four elementary schools located in the metropolitan area were selected: “Eduardo Soto,” “Luis G. Araujo,” “Mártires de Cananea” and “Xidoo,” counting with a total participation of 88 children. Inclusion criteria were as follows: informed and voluntary consent signed by parents or guardians; apparently healthy children; residence in the city not less than one year; and 12-h fasting to avoid variability in blood lipid levels. The parents or guardians responded to a carefully designed questionnaire for health, habits and exposure data concerning children, which was conducted by trained personnel; their address was also requested to obtain the georeferencing of children’s homes. It is important to mention that of the schools sampled, the one attended by each child is the closest to his or her home. The children were measured to obtain weight and height, and blood samples were taken. The body mass index (BMI) for age was calculated as an anthropometric indicator for children, and this parameter was categorized based on the z-score cut-offs established by the World Health Organization (WHO) Growth Reference Data for 5–19 years (WHO n.d.). The study was approved by the Institutional Committee of Bioethics in Research of the University of Guanajuato.

Environmental sample collection

A detailed sampling was carried out according to the specifications of the Official Mexican Standard NMX-AA-132-SCFI-2006 (SCFI, 2006), to determine the volume of soil contaminated by OCPs. In this way, sampling areas of 600 m × 600 m were established, and the sampling points were selected with the support of Geographic Information Systems (GIS), thus recording the georeferencing of each one. The samples were taken in outdoor surface soils with a shovel, covering up to a depth of 5 cm and collecting a quantity of 1 kg of the material at each point. A quantity of 52 samples were obtained, which were stored in aluminum foil inside valve bags, at a temperature of − 20 °C until analysis.

Biological sample collection

Blood samples were taken from the antecubital vein in 8 mL vacuum tubes without anticoagulant, to obtain serum. The samples were centrifuged at 1200× g for 10 min, and serum was transferred to Eppendorf tubes and stored at − 80 °C until analysis.

Extraction of OCPs in soil

The extraction of the analytes was carried out based on the extraction method Soxhlet (3540C) of the Environmental Protection Agency of the United States (EPA, 1996). 0.2 g of soil was weighed and placed in extraction thimbles (Whatman) inside the chamber. 130 mL of dichloromethane (J.T. Baker) was used as a solvent in a flat-bottomed round flask, to which 25 ppb of PCB 141-C13, DDT-C13 and HCH-C13 were added as surrogate standards. The system was kept at reflux for 10 h, on the basis that every 40 min an extraction cycle is carried out. The obtained extract underwent two evaporation processes, first on a rotavapor at 30 ºC up to a volume of 3–4 mL, to subsequently clean with 2 mL of hexane, three times; and second using a nitrogen flow until obtaining an approximate volume of 2 mL. The analytes were solid phase extracted using florisil cartridges previously conditioned with a mixture of hexane/acetone 1:1 (Karal) and eluted with 6 mL of hexane (J. T. Baker). The eluate was recovered in a conical tube and then evaporated with a stream of nitrogen at 37 °C until a volume of 100 µL was obtained, which was transferred to a 150 mL insert and the final volume was adjusted with hexane (J.T. Baker). The insert was placed in a vial and stored at − 10 °C for further analysis by gas chromatography.

Extraction of OCPs in blood

First, the lipids were extracted from each blood sample according to method described by Ryan et al., (1997), for which 2 mL of separated plasma as above was placed in a previously weighed round tube. Then 25 ppb of surrogate standards PCB 141-C13, DDT-C13 and HCH-C13 (CAMBRIDGE ISOTOPE) was added to the sample, and it was stored for 48 h at − 20 °C. The mixture was thawed at room temperature, and 0.5 mL of formic acid (J. T. Baker) and 3 mL of isopropyl alcohol (J. T. Baker) were added. A 1: 1 ethyl ether/hexane (J. T. Baker) solution was added to the contents of the tube and then centrifuged at 3000 rpm for 3 min. The upper phase was recovered, to which 2 mL of a 1% KCl (J. T. Baker) solution was added and centrifuged to recover the upper phase again. The entire process was repeated two more times, and the three extracted phases were recovered in a conical glass tube previously weighed, to finally be brought to dryness under a stream of nitrogen at 37 °C, until obtaining a constant weight. The lipid content in the sample was determined by gravimetry.

The extraction of OCPs was carried out based on (Sugawara et al., 2002) and (García-Nieto et al., 2010). The lipid samples obtained as previously indicated were dissolved in 3 mL of hexane (J.T. Baker), and then the solution was transferred to a tube round to which 2 mL of H2SO4 (J. T. Baker) was added. The mixture was incubated for 10 min at room temperature and centrifuged at 3000 rpm for 3 min, to later recover the upper phase in a conical tube, while the lower phase was extracted two more times with 3 mL of hexane (J.T. Baker). The three extracts corresponding to each sample were pooled in the conical tube and then concentrated to a volume of 2 mL under a nitrogen stream at 37 °C. The analytes were extracted with florisil cartridges, exactly as described in paragraph 2.4.

Quantitative analysis of OCPs

The quantification of the analyzed compounds was carried out on a Perkin Elmer Clarus 500 gas chromatograph coupled to a PerkinElmer Clarus 560S mass spectrometer (Waltham, MA, USA). A 30 m × 0.32 mm ID Elite-5 MS column with 0.25 µm film thickness was used. The oven temperatures were as follows: initial at 100 °C (2 min), final at 310 °C (ramps: 20 °C/min to 200 °C, 7–270 °C/min, and 15 °C/min up to 310 °C for 6 min), the injector temperature at 270 °C with 30 mL/min splitless, the transfer line at 300 °C and the mass detector at 225 °C. Helium was used as carrier gas at a linear speed of 1.5 mL/min. Under these conditions, the retention times of each compound were determined. A standard mixture (Sigma-Aldrich) of 13 OCPs was used: α- hexachlorocyclohexane (α-HCH), β-hexachlorocyclohexane (β-HCH) and γ-hexachlorocyclohexane (γ-HCH), heptachlor, heptachlor epoxide, aldrin, dieldrin, endrin, 2,4′-dichlorodiphenyldichloroethane (2,4′-DDD), 4,4′-dichlorodiphenyldichloroethane (4,4′-DDD), 4,4′-dichlorodiphenyldichloroethylene (4,4′-DDE), 2,4′-dichlorodiphenyltrichloroethane (2,4′-DDT) and 4,4′-dichlorodiphenyltrichloroethane (4,4′-DDT) in octane starting from a concentration of 100 ppb to generate the calibration curve. Blood OCPs concentrations were adjusted to lipid content, determined as described above.

Limit of detection (LOD) is defined as the arithmetic mean of a hexane blank, added with 25 ppb of surrogated standard, adding three times the standard deviation. If no chemical was found in the blanks, the LOD was defined as the Instrumental Detection Limit (IDL), which was estimated by injecting low concentrations of target analytes until a small peak was obtained with a signal-to-noise ratio of ~ 3: 1. The values obtained were: 0.01 ppb for the compounds α-, β- and γ-HCH, heptachlor, heptachlor epoxide, aldrin, dieldrin and endrin for the compounds 2,4′-DDD, 4,4′-DDD, 4,4′-DDE, 2,4′-DDT and 4,4′-DDT was 0.4 ppb.

The concentrations of the compounds 2,4′-DDT, 4,4′-DDT and 4,4′-DDE were used to calculate the DDT/DDE ratio, both in soil and blood samples (only in samples with values above the LOD) (Tavares et al., 1999).

Development of concentration distribution maps of the total OCPs

Concentration distribution maps of the total sum of OCPs both in soil and blood in the Salamanca metropolitan area were elaborated through the program QGIS 3.16.6 (Alkmaar, The Netherlands), using the georeferencing data of the soil sampling points, and the corresponding to the homes of the children who participated in the study. The distribution is shown in percentiles, so that each color represents the concentration quartile to which the sample belongs (green: quartile 1; light green: quartile 2; orange: quartile 3; red: quartile 4). The maps included the locations of the schools, and of some industrial sites of possible influence for the presence of the pollutants, such as the Tekchem property, the thermoelectric power plant of the Federal Electricity Commission (CFE) and the refinery Ing. Antonio M. Amor (RIAMA) of PEMEX company.

Estimation of the exposure dose

The exposure dose was calculated for those compounds that were detected in more than 50% of the blood samples (heptachlor, heptachlor epoxide, 4,4′-DDT, 4,4′-DDE and 4,4′-DDD). The following equation was used (ATSDR, 2005):

$$ D = \frac{{C*{\text{IR}}*{\text{AF}}*{\text{EF}}}}{{{\text{BW}}}} $$

where C is the concentration of the compound in soil (mg/kg); IR is the intake rate of the contaminated media (kg/day; value of 0.00001–0.001) (Moya & Phillips, 2014); AF is the bioavailability factor (0.1–0.6 for 4,4′-DDT and its metabolites (Juhasz et al., 2016)) and 0.5 for heptachlor and heptachlor epoxide (ATSDR, 2007)); BW is body weight (kg); and EF is the exposure factor (obtained by multiplying the exposure frequency (F) by the duration of exposure (ED), and dividing between the time during which the dose should be averaged (AT)):

$$ {\text{EF}} = \frac{{F*{\text{ED}}}}{{{\text{AT}}}} $$

The exposure frequency (days/year; 2 days/week * 52 weeks/year), exposure duration (years; mean [SD]: 7.7 [3.38]), time averaged (ED [years] * 365 days/years; mean [SD]: 2803.33 [1220.12]), exposure factor (mean [SD]: 0.47 [6.14]) and body weight (kg; mean [SD]: 37.2 [11.4]) were estimated from the data obtained through the questionnaires.

The exposure dose was estimated using a Monte Carlo simulation model, with Oracle Crystal Ball software version 11.1.2.4, establishing a total of 100,000 iterations to obtain descriptive statistics (mean, standard deviation, percentiles). If the resulting values were greater than 1, it was considered a noncarcinogenic risk; if they did not exceed 1, it was not considered as risk. Reference doses for noncarcinogenic effects are only reported for compounds 4,4′-DDT, heptachlor, and heptachlor epoxide, through intake, and are based on liver lesions in mice (U.S. EPA, 1987a, 1987b, 1987c).

Statistics

A descriptive analysis of the data set for both soil and blood OCPs concentrations was performed, calculating mean, standard deviation, minimum, maximum, percentiles, frequencies and distribution. For analysis purpose, if the concentration value of a sample was less than LOD for any compound, then it was replaced by LOD/2 (Cohen & Ryan, 1989). OCPs concentration values did not show a normal or lognormal distribution (Kolmogorov–Smirnov test), so the results are described with the median and range. An exploratory analysis was also performed to determine the relation between the general characteristics of children and the concentrations of OCPs. Spearman correlation tests were performed to estimate the association between OCPs levels and age, weight, height, BMI for age and number of years living in the city. Mann–Whitney tests were performed to compare OCPs levels between grouped data according to BMI category and sex. All analyses were performed using GraphPad Prism 8.0.1 statistical package (San Diego, CA, USA).

Results

Table 1 shows the descriptive characteristics of the children participating in the study, in terms of their anthropometric data, as well as certain habits or risk activities. The mean age was 10 years, ranging from 6 to 15 years; 53% of children were female, while 47% were male. Regarding BMI for age, 60% of children had a normal or healthy value, while 24% were overweight, 15% obese, and 1% underweight (WHO n.d.). In addition, an average of approximately 8 years of residence of children in the city of Salamanca was reported, ranging from 1 to 15 years. With respect to risk activities, 92% of parents or guardians mentioned the use of household pesticides (insecticides or rodenticides), and 22% smoke inside the house.

Table 1 General characteristics of participating children
Full size table

Table 2 shows the statistical distribution parameters of both the individual soil OCPs concentrations and their total sum, in addition to the sum of total DDT, and DDT/DDE ratio. All the measured OCPs were detected to a certain extent, highlighting 2,4′-DDD that was detected in almost all the samples (96.2%). It is followed by 4,4′-DDT and 4,4′-DDD with detection percentages close to 90 and 85%, respectively. 2,4′-DDT, 4,4′-DDE, heptachlor epoxide and aldrin were also detected in significant percentages, all above 50%. The compounds γ-HCH, heptachlor and endrin were detected in less of 50% of the samples.

Table 2 OCPs concentrations in soil from Salamanca, Mexico
Full size table

For the total sum of OCPs in soil, a median concentration of 70.6 ng/g (6.93–3299) was obtained, while for the total sum of DDT it was 49.6 ng/g (6.93–3276). As for the individual OCPs, those that resulted in highest median levels were 4,4′-DDT (11.3 [< LOD—593] ng/g) and 2,4′-DDD (10.7 [< LOD—968] ng/g). 4,4′-DDD and 4,4′-DDE resulted with similar median levels (9.19 [< LOD—1389] ng/g and 8.19 [< LOD—1474] ng/g, respectively). 2,4′-DDT was found to have a median level of 7.32 ng/g (< LOD—241). The lowest median concentrations corresponded to aldrin (4.25 [< LOD—9.11] ng/g) and heptachlor epoxide (4.06 [< LOD—21.9] ng/g). γ-HCH, heptachlor and endrin median concentrations resulted below the LOD. With respect to the DDT/DDE ratio in soil, a median value of 1.04 (0.09–23.1) was obtained.

On the other hand, Table 3 shows the statistical distribution of the concentrations of OCPs measured in children’s blood and their total sum, as well as DDT total sum and DDT/DDE ratio. Like OCPs in soil, all the compounds measured in the children also showed positive detection levels. In this case, it is noteworthy that heptachlor was detected in 100% of the samples. Regarding DDT and its metabolites, only 4,4′-DDE, 4,4′-DDD and 4,4′-DDT were detected in significant percentages (82, 60 and 58%, respectively). Heptachlor epoxide was also detected at a significant level (57%). The compounds γ-HCH, aldrin, endrin, 2,4′-DDT and 2,4′-DDD were detected in less of 50% of the samples.

Table 3 Blood OCPs concentrations in children from Salamanca, Mexico
Full size table

Regarding the total sum of OCPs in children’s blood, the median concentration was 390 ng/g lipid (7.34–14,895), while for the total sum of DDT the value was 175 ng/g lipid (< LOD—14,802). Of the individual compounds analyzed, the one with the highest median concentration was 4,4′-DDE (70.6 [< LOD—9719] ng/g lipid), followed by heptachlor (47.1 [2.02–4975] ng/g lipid), 4,4′-DDT (20.2 [< LOD—7577] ng/g lipid), 4,4′-DDD (15.0 [< LOD—1841] ng/g lipid) and heptachlor epoxide (13.7 [< LOD—9576] ng/g lipid). γ-HCH, aldrin, endrin, 2,4′-DDT and 2,4′-DDD had median concentrations below the LOD. For the DDT/DDE ratio in children’s blood, the median was 1.72 (0.05–11.3).

Regarding the results of the exploratory analysis of the general characteristics of children and the concentrations of OCPs in blood, only the correlations between age and γ-HCH (Spearman r = 0.271; p = 0.01) and heptachlor epoxide (Spearman r = 0.298; p = 0.005) were statistically significant, indicating that older children exhibit higher concentrations of the compounds. Also, the following trends were observed: children who have lived more years in the city have higher concentrations of heptachlor (Spearman r = 0.198; p = 0.07) and heptachlor epoxide (Spearman r = 0.200; p = 0.07); children with normal weight have higher concentrations of γ-HCH, compared to those with overweight or obesity (Mann–Whitney test; p = 0.06).

The distributions of the estimated daily intake doses for those OCPs that were detected in more than 50% of the blood samples are depicted in Table 4. Based on the measured soil levels, the estimated mean daily intake doses were 3 × 10−6, 2 × 10−6, 1 × 10−6, 2 × 10−7 and 2 × 10−7 mg/kg/day, respectively, for 4,4′-DDT, 4,4′-DDE, 4,4′-DDD, heptachlor and heptachlor epoxide. None of the calculated doses resulted above the reference dose of compounds for which this value is reported in the literature.

Table 4 Estimated daily intake dose for OCPs
Full size table

Figure 1 shows the spatial distribution of total OCPs concentrations in soil, grouped by quartiles. It can be observed that the highest concentrations tended to be present around the site of the former industrial unit Tekchem, toward the east and southeast of the city of Salamanca, while lower concentrations predominated in the north and west. On the other hand, the spatial distribution of total OCPs concentrations in children is depicted in the map in Fig. 2. In this case, a more heterogeneous distribution of concentrations can be seen, resulting in higher levels in children living in the north and west of the city than in children living in the area around Tekchem.

Fig. 1
figure 1

Spatial distribution of total OCPs concentrations in soil. The points correspond to the georeferencing of the sample. Each color represents the concentration quartile to which the sample belongs (green: quartile 1 [6.93–41.8 ng/g]; light green: quartile 2 [41.9–70.6 ng/g]; orange: quartile 3 [70.7–160 ng/g]; red: quartile 4 [160.1–3299 ng/g])

Full size image
Fig. 2
figure 2

Spatial distribution of total OCPs concentrations in children’s blood. The points correspond to the georeferencing of each participant’s home. Each color represents the concentration quartile to which the sample belongs (green: quartile 1 [7.34–77.6 ng/g lipid]; light green: quartile 2 [77.7–390 ng/g lipid]; orange: quartile 3 [390.1–1282 ng/g lipid]; red: quartile 4 [1282.1–14,895 ng/g lipid])

Full size image

Discussion

The city of Salamanca, Mexico, is well known for its history of environmental pollution, as it is a center of industrial activity at the national level. Such activities, when not carried out based on regulatory standards, or when this does not exist, unfortunately lead to problems associated with the introduction of dangerous substances into the environment, with the risks involved. One of the most worrying cases, which so far has no definitive solution, is the presence of environmental liabilities within the vicinity of the former Tekchem Industrial Unit, which, when in operation, manufactured a wide range of OCPs and other synthetic chemical compounds. The environmental liabilities generated were the result of an accumulation of several years due to inconveniences in terms of waste disposal. Previous studies have characterized the polluting substances in the site, mainly reporting the presence of OCPs, which constitute 70% of the polluting mass (Beltrán Hernández et al., 2019; SEMARNAT, 2018).

In this study, levels of γ-HCH, heptachlor, heptachlor epoxide, aldrin, endrin, as well as DDT and its metabolites were detected and quantified in soil and in school-age children in the city of Salamanca. Said compounds correspond to those that have been identified in the environmental liabilities of Tekchem (SEMARNAT, 2018). DDT and its metabolites were the compounds that showed the highest levels of concentration in soil, followed by aldrin and heptachlor epoxide (Table 2). Comparing the total DDT concentration obtained in our study with the Canadian Environmental Quality Guideline, which establishes a maximum value of 700 ng/g of total DDT for residential soil (Canadian Council of Ministers of the Environment, 1999), we find that the level did not exceed this value. The California Human Health Screening Levels guide, on the other hand, establishes a maximum value of 1900 ng/g for 4,4′-DDT, 2000 ng for 4,4′-DDE, and 2300 ng/g for 4,4′-DDD (California Department of Toxic Substances Control, 2020), which were not exceeded in the soil samples analyzed. In the case of aldrin, the same guide establishes a maximum value of 3300 ng/g for residential soil, so that the levels obtained in this study did not exceed this limit.

In our country, monitoring studies have been carried out on OCPs in soil, mainly in agricultural and industrial areas, and in those with a high prevalence of vector-borne diseases such as malaria (Table 5). The total DDT concentration obtained in this study was higher than the levels reported in urban soil of San Luis Potosí (6.10 ng/g) (Perez-Vazquez et al., 2015), and in the Mexicali Valley (20 ng/g) and the Yaqui Valley (3.6 ng/g) in northwestern Mexico (Sánchez-Osorio et al., 2017). In this last work mentioned, lower heptachlor epoxide levels (Mexicali Valley: 0.013 ng/g; Yaqui Valley: 0.022 ng/g) (Sánchez-Osorio et al., 2017) were also reported compared to ours. Another study performed in soil from communities from Tabasco reported lower levels of total DDT compared to ours (median concentrations ranged from 4 to 38 ng/g), except for the community Cardenas whose level was higher (91 ng/g) (Torres-Dosal et al., 2012). In another monitoring carried out in urban soils of Chihuahua, the range of total DDT values (2–141.7 ng/g) was lower than that of our study (Martínez et al., 2012). On the other hand, Orta-García et al. (Orta-García et al., 2016) found higher median levels of total DDT (79.5 ng/g) compared to the present study, in urban soil from Monterrey. Likewise, the levels within the resulting range for total DDT in soil from various communities in Chiapas (2–26,980 ng/g) exceeded the maximum value obtained for total DDT in our study (Martínez-Salinas et al., 2011).

Table 5 Comparison of OCPs concentrations found in the present study compared with data from other monitoring in Mexico
Full size table

Cantu-Soto et al. (Cantu-Soto et al., 2011) reported ranges with very high levels of OCPs with respect to those obtained in the present work, in residential soils of the Yaqui Valley (4,4′-DDT: < LOD—679,700 ng/g; 4,4′-DDE: < LOD—621,300 ng/g; 4,4′-DDD: < LOD—197,300 ng/g; aldrin: < LOD—25,800 ng/g) and the Mayo Valley (4,4′-DDT: < LOD—301,200 ng/g; 4,4′-DDE: < LOD—226,300 ng/g; 4,4′-DDD: < LOD—39,300 ng/g; aldrin: < LOD—74,000 ng/g), in southern Sonora, Mexico. There is a site similar to Salamanca in terms of industrial activity: the city of Coatzacoalcos, in the state of Veracruz, where OCPs have also been quantified in different environmental matrices. Espinosa-Reyes et al. (2013) conducted a monitoring study in the area, reporting total DDT levels in soil in the following ranges and in different years: 2006 (< LOD—12.9 ng/g), 2008 (0.7–174.9 ng/g) and 2012 (18.6–805.7 ng/g). It can be observed that the range for total DDT in soil in our study exceeds the previous ones (up to 4 times for the range reported in 2012). In another study conducted in the area, Gonzalez-Mille et al. (2013) found the following ranges for OCPs quantified in soil: γ-HCH (2.8–88.2 ng/g), 4,4′-DDE (< LOD—6.7 ng/g) and 4,4′-DDT (< LOD—43.2 ng/g). With respect to our results, the above exceed the range we recorded for γ-HCH; however, for 4,4′-DDE and 4,4′-DDT, our results exceed the above maximum values by up to 200 and 14 times, respectively.

Comparing with soil monitoring for OCPs in other countries, we found that the ranges obtained in this study had higher levels than the resulting maximum concentrations in urban soils in Pristina, Serbia (DDT: 2.41–18.8 ng/g; DDE: 1.19–30.5 ng/g; DDD: 0.97–8.38 ng/g; aldrin: 1.94–3.28 ng/g) (Gulan et al., 2017). Likewise, the maximum total DDT value from our study was higher than that reported in agricultural soils from Wuhan, China (< LOD—1198 ng/g), while on the other hand aldrin range resulted in higher values than our maximum value obtained (< LOD—21.56 ng/g) (Zhou et al., 2013). Another study performed in agricultural soil from Shanghai, China, reported maximum values for total DDT and heptachlor epoxide (0.44–247.45 ng/g and < LOD—4.78 ng/g, respectively) lower than those from our study, while for aldrin was obtained a similar range (< LOD—6.62 ng/g) (Jiang et al., 2009).

With respect to the levels of OCPs in the infant population, the compounds that resulted in the highest concentrations were 4,4′-DDT and its metabolites, as well as heptachlor and heptachlor epoxide (Table 3). By comparing our results with the median concentrations obtained in the German Environmental Survey for Children and Adolescents (GerEs) 2014–2017 for 4,4′-DDT (1.94 ng/g lipid), 4,4′-DDE (26.7 ng/g lipid) and 4,4′-DDD (1.88 ng/g lipid) (Bandow et al., 2020), we can see that our levels exceed the previous ones by 10, 2.6 and 8 times, respectively. On the other hand, the Fourth National Report on Human Exposure to Environmental Chemicals (population aged 12 to 19) reports heptachlor epoxide and 4,4′-DDT as undetectable, while for 4,4′-DDE the median concentration was 93.6 ng/g lipid (CDC, 2009), slightly exceeding our found median value.

As well as in soil, in our country monitoring has also been carried out in children exposed to OCPs (Table 5). Flores-Ramírez et al. (Flores-Ramírez et al., 2017) reported a median value of 4,4′-DDE slightly lower than ours (58.77 ng/g lipid), in children living near contaminated sites in Mexico. Another study in children from pollution hot spots in Mexico showed median levels of DDE (387 ng/g lipid) and DDT (189 ng/g lipid) higher than ours, while heptachlor epoxide was not detected (Trejo-Acevedo et al., 2009). Studies conducted in children in malaria-endemic areas of the country show levels of DDT and its metabolites that far exceed those in this study. In Chiapas and Oaxaca, in southeastern Mexico, median levels of total DDT ranged from 11,353.8 ng/g lipid to 34,189 ng/g lipid between three assessed communities (Pérez-Maldonado et al., 2013). In Quintana Roo, in southern Mexico, median levels of DDT of 5,646 ng/g lipid and DDE of 39,934 ng/g lipid were found, while heptachlor epoxide was not detected (Trejo-Acevedo et al., 2012). In children from Chihuahua, ranges with higher levels than those in our study were reported, both for DDT (711–68, 669 ng/g lipid) and for DDE (271–170,596 ng/g lipid) (Martínez et al., 2012).

Regarding monitoring in other countries, OCPs levels in children from our study were higher than those found in children in South Korea (heptachlor: < LOD; heptachlor epoxide: < LOD; 4,4′-DDT: 3.2 ng/g lipid; 4,4′-DDE: 44.7 ng/g lipid; 4,4′-DDD: < LOD) (Park et al., 2016). The ranges we obtained in this study for 4,4′-DDT and 4,4′-DDE include values that exceed the maximum concentrations reported in children from Canada (4,4′-DDT: 2.3–62.7 ng/g lipid; 4,4′-DDE: 5.6–1864.4 ng/g lipid) (Turgeon O’Brien et al., 2012). On the other hand, Windham et al. (Windham et al., 2010) reported a median 4,4′-DDE value of 101.5 ng/g lipid in US girls, which is slightly higher than that obtained in our study.

The DDT/DDE ratio for soil resulted in a median value of 1.04, while in children it was 1.72. According to Tavares et al. (Tavares et al., 1999), a value of this ratio ≥ 1 indicates the recent incorporation of DDT into the environment, which is usually due to recent applications. The half-life time for DDT in terms of its persistence in the environment is 2 to 15 years, depending on the factors that are present in the site influencing its degradation (ATSDR, 2019; Jayaraj et al., 2016). According to the previous studies performed both in soil and children, the concentrations of OCPs obtained in our study were lower, in general, than those reported in endemic areas of malaria, where large amounts of these compounds were applied in the past, mainly DDT (Díaz-Barriga et al., 2003). However, in the case of OCPs in blood, they are higher compared with baseline levels from general population data, such as those from the GerEs (Bandow et al., 2020). In the case of Tekchem, DDT was discontinued in 1997, however, as mentioned above, to date residues are present within the facilities, which are likely dispersing to the soils of the urban area of Salamanca, and this would explain the value obtained for DDT/DDE ratios in both soil and children, indicating both present and past exposure to DDT (Tables 2, 3).

In our study, heptachlor was detected in 100% of children; however, in soil it was not detected significantly (less than 50% of the samples). It is important to mention that very few studies have reported levels of heptachlor and heptachlor epoxide in the human population, and in most cases, these compounds are not detected. Morgan et al. (Morgan et al., 2014) quantified heptachlor in various environmental media such as soil, dust, outdoor air and indoor air, finding the highest levels in air, mainly in indoor environments. The same study determined that the predominant route of exposure for this compound in preschool children was inhalation of air both outdoors and indoors (98%) (Morgan et al., 2014). This could explain our findings in terms that in the case of heptachlor the soil is not the main route of exposure.

About the spatial distribution of the total sum of OCPs quantified in soil (Fig. 1), it is noted that the highest concentrations were determined in the vicinity of Tekchem, so in this case there is a well-identified source for the emission of these substances into the environment, and thus for human exposure. However, it is important to consider the possible influence of other OCPs sources that may be contributing to the concentrations of OCPs found in Salamanca soil. On the other hand, this type of compounds has the characteristic that they can be transported by the wind and travel great distances, being later in places distant to their origin (Heidelore Fiedler, 2002; Sparling, 2016). In the case of the total sum of OCPs in blood, the spatial distribution is interesting (Fig. 2), since the highest concentrations were mainly identified in the northern area of Salamanca, however, there are also high concentration points in the vicinity of Tekchem. This variability in distribution could be explained by the different routes of exposure through which OCPs can be incorporated into the infant population. Although outdoor soil is one of the most important routes of exposure to OCPs (Herrera-Portugal et al., 2005), other routes such as food, breast milk and indoor soil and dust can also contribute significantly, which were not considered in this study. For example, it has been established that high concentrations of OCPs in indoor spaces, and longer residence times, can increase exposure to these pollutants up to 1000 times compared to outdoor exposure (Hwang et al., 2008). Another important point to consider in our study is the mobility of children within the city, which could also explain the variability in the spatial distribution of OCPs. Finally, as already mentioned, although the compounds studied are subject to either prohibition or restriction regulations (Stockholm Convention), unfortunately to date they are still dispersed globally and ubiquitously, regardless of their origin (Sparling, 2016).

For the present study, no biomarker of health effect was measured, only associations of the concentrations of OCPs were made with the general data obtained from the questionnaires (age, weight, height, sex, BMI for age, years living in the city). In this regard, several health effects associated with exposure to OCPs in children have already been reported (Berghuis et al., 2015; Cabrera-Rodríguez et al., 2019; Pérez-Maldonado et al., 2006; Rocha-Amador et al., 2009; Segal & Giudice, 2019; Vafeiadi et al., 2015; Vrijheid et al., 2016). We cannot know specific risks to the health of children with the results obtained in our work, because cut-off levels have not been established for the compounds studied. However, by using data of concentrations in soil and of frequency and duration of exposure obtained from the questionnaire, it was possible to calculate an exposure dose to establish a risk level (Table 4), for 4,4′-DDT and its metabolites, heptachlor, and heptachlor epoxide. None of the doses obtained by probabilistic simulation was greater than 1, so we can conclude that there is no risk through the intake of the compounds in soil, and for the effects associated with the reference doses. It is important to mention that the frequency of exposure, that is, the time spent by children outside their homes, was low (2 days a week * year) and for the present study no sampling points were made in the children’s homes/yards. In addition, other variables that could influence soil OCPs levels, such as wind direction, were not considered. Another important point to consider is that no other environmental matrices (air for example) were evaluated, which may constitute a risk through other routes such as inhalation of compounds.

Conclusion

The results of the present exploratory study demonstrated the presence of OCPs both in the soil and in school-age children in the city of Salamanca, Mexico. The spatial distribution showed that, in the case of soil, the highest concentrations of OCPs could be related to the presence of environmental liabilities in the Former Tekchem Industrial Unit. In the case of children, the fact that the highest concentrations were identified at different points in the city could suggest the influence of other routes of exposure—besides soil—and the mobility of children within the city inherent in their activities.

Our work has several limitations, such as sample size, and the fact that other exposure routes to quantify compounds were not considered. In addition, due to the cross-sectional nature of the study, there may be biases in terms of spatio-temporal variability of the concentrations of the compounds analyzed, both in soil and in children.

Despite the limitations, this study provides relevant information regarding the presence of OCPs in Salamanca, Mexico, while providing the basis for future research in which the problem can be addressed from a broader perspective. As perspective, we strongly recommend a large-scale monitoring study from the data generated in this work, including the study of other routes of exposure, as well as the presence of compounds in other environmental media that could serve as potential additional sources of OCPs to the population of the city, mainly to children, who are especially vulnerable toward exposure to environmental pollutants. It also reiterates the importance of continuing efforts to generate effective and precautionary measures with respect to the case of environmental liability Tekchem.