PFAS levels and exposure determinants in sensitive population groups

BACKGROUND

Per- and polyfluoroalkyl substances (PFAS) are persistent organic pollutants. The first exposure to PFAS occurs in utero, after birth it continues via breast milk, food intake, environment, and consumer products that contain these chemicals. Our aim was to identify determinants of PFAS concentrations in sensitive population subgroups- pregnant women and newborns.

METHODS

Nine European birth cohorts provided exposure data on PFAS in pregnant women (INMA-Gipuzkoa, Sabadell, Valencia, ELFE and MoBa; total N = 5897) or newborns (3xG study, FLEHS 2, FLEHS 3 and PRENATAL; total N = 940). PFOS, PFOA, PFHxS and PFNA concentrations were measured in maternal or cord blood, depending on the cohort (FLEHS 2 measured only PFOS and PFOA). PFAS concentrations were analysed according to maternal characteristics (age, BMI, parity, previous breastfeeding, smoking, and food consumption during pregnancy) and parental educational level. The association between potential determinants and PFAS concentrations was evaluated using multiple linear regression models.

RESULTS

We observed significant variations in PFAS concentrations among cohorts. Higher PFAS concentrations were associated with higher maternal age, primipara birth, and educational level, both for maternal blood and cord blood. Higher PFAS concentrations in maternal blood were associated with higher consumption of fish and seafood, meat, offal and eggs. In cord blood, higher PFHxS concentrations were associated with daily meat consumption and higher PFNA with offal consumption. Daily milk and dairy consumption were associated with lower concentrations of PFAS in both, pregnant women and newborns.

CONCLUSION

High detection rates of the four most abundant PFAS demonstrate ubiquitous exposure of sensitive populations, which is of concern. This study identified several determinants of PFAS exposure in pregnant women and newborns, including dietary factors, and these findings can be used for proposing measures to reduce PFAS exposure, particularly from dietary sources.

Determination of biomarkers of exposure to boscalid, captan, folpel, mancozeb and tebuconazole in urine and hair samples

In order to develop a tiered approach to identify relevant biomarkers and matrices for assessing pesticide exposure in residents living close to vineyards, five priority pesticides (boscalid, captan, folpel, mancozeb and tebuconazole) and their metabolites were analyzed in urine and hair samples from the biobank of a French national prevalence study conducted between 2014 and 2016. To do this, several analytical methods based on gas chromatography coupled with tandem mass spectrometry (GC/MS/MS) were developed by relying on the expertise of the laboratory and the scientific literature, in particular on a paper describing the use of gas chromatography-mass spectrometry for the determination in human urine samples of ethylene thiourea (ETU), a metabolite of mancozeb, after a supported liquid extraction followed by a derivatization step [1]. The main adaptations carried out as part of this study concerned:•

the determination of ethylene urea (EU), another metabolite of mancozeb, at the same time as ETU in urine samples

•

the determination of all substances of interest including boscalid, EU and ETU, folpel and one of its metabolite (phthalimide), tebuconazole and one of its metabolite (hydroxytebuconazole), and tetrahydrophthalimide (metabolite of captan) in organic hair extracts by GC/MS/MS after a derivatization step

Identification of pesticides exposure biomarkers for residents living close to vineyards in France

Biomonitoring can be relevant for assessing pesticides exposure of residents living close to vineyards (LCTV). However, because xenobiotics are generally present at low levels in human biological matrices and the sources of pesticide exposure are multiple, several challenges need to be overcome to reliably assess exposure in residents LCTV. This includes particularly identifying the most appropriate exposure biomarkers, the biological matrices in which they should be measured, and analytical methods that are sufficiently sensitive and specific to quantify them. The aim of the present study was to develop a tiered approach to identify relevant biomarkers and matrices for assessing pesticide exposure in residents LCTV. We used samples from a biobank for 121 adults and children included in a national prevalence study conducted between 2014 and 2016 who lived near or far from vineyards. We analyzed five priority pesticides (folpet, mancozeb, tebuconazole, glyphosate, and copper) and their metabolites in urine and hair samples. We identified relevant biomarkers according to three criteria related to: i) the detection frequency of those pesticides and metabolites in urine and hair, ii) the difference in concentrations depending on residence proximity to vineyards and, iii) the influence of other environmental and occupational exposure sources on pesticide levels. This tiered approach helped us to identify three relevant metabolites (two metabolites of folpet and one of tebuconazole) that were quantified in urine, tended to be higher in residents LCTV than in controls, and were not significantly influenced by occupational, dietary, or household sources of pesticide exposure. Our approach also helped us to identify the most appropriate measurement strategies (biological matrices, analytical methods) to assess pesticide exposure in residents LCTV. The approach developed here was a prerequisite step for guiding a large-scale epidemiological study aimed at comprehensively measuring pesticides exposures in French residents LCTV with a view to developing appropriate prevention strategies.

Hair versus urine for the biomonitoring of pesticide exposure: Results from a pilot cohort study on pregnant women

BACKGROUND/AIM

The quantitative assessment of human exposure to contaminants such as pesticides is a crucial step in the characterization of exposure-associated risk. For this purpose, biomonitoring is often privileged as it presents the advantage of integrating all the possible sources and routes of exposure and of being representative of the internal dose resulting from exposure. Although biological fluids such as urine and blood have been used to date for this purpose, increasing interest has also been observed over the past decade for hair analysis. The present work aimed at comparing the information obtained from the analysis of urine versus hair regarding exposure to pesticides in a pilot cohort of pregnant women.

METHODS

In ninety-three pregnant women included in the pilot of the French cohort ELFE, one urine and one hair sample were collected simultaneously from each subject at the maternity. Samples were analyzed using GC-MS/MS analytical methods allowing for the detection of both parent pesticides and metabolites, and designed to be as similar as possible between urine and hair for reliable inter-matrix comparison. Fifty-two biomarkers of exposure were targeted, including parents and metabolites of organochlorines, organophosphates, pyrethroids, carbamates, phenylpyrazoles and other pesticides.

RESULTS

The number of different compounds detected ranged from 16 to 27 (median = 22) in hair, and from 3 to 22 (median = 12) in urine. In hair, 24 compounds were found in > 40% of the individuals, whereas only 12 compounds presented the same frequency of detection in urine. Among the chemicals detected in > 80% of both hair and urine samples, only one (pentachlorophenol) showed a signification correlation between hair and urine concentrations.

CONCLUSIONS

The present results highlight the multiple exposure of the pregnant women included in this cohort and suggest that hair provides more comprehensive information on pesticide exposure than urine analysis. This study thus supports the relevance of hair analysis in future epidemiological studies investigating association between exposure and adverse health effects.

Chlordecone and organochlorine compound levels in the French West Indies population in 2013-2014

Agricultural activities in the Caribbean, especially banana cropping, are known for their significant use of pesticides. In particular is chlordecone, which was used between 1972 and 1993 against the banana root borer, Cosmopolites sordidus (Germar, 1824). In this context, "Kannari study: Health, Nutrition and Exposition to Chlordecone in French West Indies" was put in place in 2013-2014 to supplement knowledge about the exposure of the population to chlordecone and other organochlorine pollutants. The data collected comprised a dietary intake description, data from biological samples (blood sample), socioeconomic and demographic information, and data from complementary specific items relative to life habits. A total of 742 subjects (292 in Guadeloupe and 450 in Martinique) were included in the impregnation component of the Kannari study. In this study, chlordecone and organochlorine compounds were detected in almost all participants. This result suggests that exposure to chlordecone is widespread, but also to other organochlorine pesticides. Chlordecone impregnation of the majority of the population appears to have decreased between 2003 and 2013, but various subgroups of the population remain highly exposed. The levels of impregnation are determined by dietary exposure and environmental contamination. However, total consumption of fresh fish (all species combined), especially from informal channels, is the main source of exposure to chlordecone. The serum PCB concentrations measured in the French Caribbean Islands of Guadeloupe and Martinique are lower than those observed in metropolitan France in 2007 (French Nutrition and Health Survey (ENNS)). In contrast, the French West Indies population seems more exposed to lindane than the French mainland population, and this exposure also seems more recent.

Pesticide exposures for residents living close to agricultural lands: A review

BACKGROUND

Residents living close to agricultural lands might be exposed to pesticides through non-occupational pathways including spray drift and volatilization of pesticides beyond the treated area.

OBJECTIVE

This review aimed to identify and analyze scientific literature measuring pesticide exposure in non-farmworker residents living close to agricultural lands, and to suggest practical implications and needs for future studies.

METHODS

A review was performed using inclusion criteria to identify original articles of interest published between 2003 and 2018.

RESULTS

From the 29 articles selected in this review, 2 belonged to the same study and were grouped, resulting in a total of 27 studies. Seven studies assessed exposure to pesticides using environmental samples, 13 collected biological samples and 7 analyzed both. Nine studies included a reference group of residents living far from agricultural lands while 11 assessed the influence of the spraying season or spray events on pesticide exposures. Studies included in this review provide evidence that residents living near to agricultural lands are exposed to higher levels of pesticides than residents living further away.

DISCUSSION AND CONCLUSION

This review highlights that the following study design characteristics may be more appropriate than others to measure pesticide spray drift exposure in non-farmworker residents living close to agricultural lands: inclusion of a non-agricultural control group, collection of both biological and environmental samples with repeated sampling, measurements at different periods of the year, selection of numerous study sites related to one specific crop group, and measurements of pesticides which are specific to agricultural use. However, few studies to date incorporate all these characteristics. Additional studies are needed to comprehensively measure non-occupational pesticide exposures in this population in order to evaluate health risks, and to develop appropriate prevention strategies.

Urinary levels of pyrethroid pesticides and determinants in pregnant French women from the Elfe cohort

Pyrethroids are insecticides which are widely used for agricultural and domestic purposes. The general population can be exposed to them. Given the suspected effects of pesticides on the development of the foetus, exposure to pyrethroids during pregnancy is a major public health concern. The objective of this paper is to describe the urinary levels of the following five pyrethroid metabolites and their associated determinants among pregnant French women in 2011 enrolled in the Elfe cohort (n = 1077): a) 3-phenoxy benzoic (3-PBA), b) 4‑fluoro‑3‑phenoxy benzoic acid (4-F-3-PBA), c) Cis‑3‑(2,2dibromovinyl)‑2,2‑dimethyl cyclopropane-carboxylic acid (Cis-DBCA); d) Cis‑3‑(2,2dichlorovinyl)‑2,2‑dimethyl cyclopropane-carboxylic acid (Cis-DCCA) and e) Trans-3‑(2,2dichlorovinyl)‑2,2‑dimethyl cyclopropane-carboxylic acid (Trans-DCCA). The distribution levels were estimated for each pyrethroid metabolite. Multivariable analyses helped determine the predictors of these levels. All metabolites except 4-F-3-PBA were detected in all the urine samples. The mean urinary concentration of the sum of the metabolites (3-PBA, Cis-DBCA, Cis-DCCA, Trans-DCCA) was 1.18 μg/L, with the highest concentrations observed for 3-PBA. A comparison of these levels with other studies showed that pregnant French women tend to be more exposed to pyrethroids than their American counterparts, less exposed than Chinese and Caribbean mothers, and have similar exposure to Japanese mothers. In our study, urinary levels of pyrethroid metabolites were positively related to smoking during pregnancy, consuming fish and alcohol, domestic pesticide use and living in the vicinity of crops during pregnancy. These findings highlight the importance of non-dietary pathways when evaluating exposure to pyrethroids.

Prenatal exposure to lead in France: Cord-blood levels and associated factors: Results from the perinatal component of the French Longitudinal Study since Childhood (Elfe)

BACKGROUND

As a result of the ban on lead in gasoline on 2nd January 2000, the French population's exposure to lead has decreased in recent years. However, because of the acknowledged harmful cognitive effects of lead even at low levels, lead exposure remains a major public health issue. In France, few biomonitoring data are available for exposure to lead in pregnant women and newborn. The purpose of the perinatal component of the French human biomonitoring (HBM) program was to describe levels of various biomarkers of exposure to several environmental pollutants, including lead, among mother-baby pairs. In this paper, we aimed to describe the distribution of cord blood lead levels (CBLL) in French mother-baby pairs, and to estimate the contribution of the main lead exposure risk factors to these levels.

METHOD

A total of 1968 mother-baby pairs selected from the participants of the perinatal component of the French HBM program were included in the study on lead. Lead levels were analyzed in cord blood collected at child delivery by inductively coupled plasma-mass spectrometry (ICP-MS). The data collected included biological sample, socio-demographic characteristics, environmental and occupational exposure, and information on dietary factors.

RESULTS

CBLL were quantified for 99.5% of the sample. The CBLL geometric mean was 8.30 μg/l (95% CI [7.94-8.68]) with a 95th percentile of 24.3 μg/l (95% CI [20.7-27.1]). Factors significantly associated with CBLL were tap water consumption, alcohol consumption, shellfish consumption, vegetable consumption, bread consumption, smoking, and the mother being born in countries where lead is often used.

CONCLUSION

This study provides the first reference value for CBLL in a random sample of mother-baby pairs not particularly exposed to high levels of lead (24.3 μg/l). A substantial decrease in CBLL over time was observed, which confirms the decrease of exposure to lead among the general population. CBLL observed in this French study were in the range of those found in recent surveys conducted in other countries.

Biomarkers of exposure to environmental contaminants in French pregnant women from the Elfe cohort in 2011

BACKGROUND

As part of the perinatal component of the French Human Biomonitoring (HBM) program, biomarkers levels of various chemicals have been described among pregnant women having given birth in continental France in 2011 and who have been enrolled in the Elfe cohort (French Longitudinal Study since Childhood). This paper describes the design of the study and provides main descriptive results regarding exposure biomarkers levels.

METHODS

Exposure biomarkers were measured in biological samples collected at delivery from pregnant women randomly selected among the participants in the clinical and biological component of the Elfe cohort (n=4145). The geometric mean and percentiles of the levels distribution were estimated for each biomarker. The sampling design was taken into account in order to obtain estimates representative of the French pregnant women in 2011.

RESULTS

Results provide a nation-wide representative description of biomarker levels for important environmental contaminants among pregnant women who gave birth in France in 2011. Bisphenol A (BPA), and some metabolites of phthalates, pesticides (mainly pyrethroids), dioxins, furans, polychlorobiphenyls (PCBs), brominated flame retardants (BFRs), perfluorinated compounds (PFCs) and metals (except uranium) were quantified in almost 100% of the pregnant women. Some compounds showed a downward trend compared to previous studies (lead, mercury), but others did not (pyrethroids) and should be further monitored.

CONCLUSION AND PERSPECTIVES

The present results show that French pregnant women are exposed to a wide variety of pollutants, including some that have been banned or restricted in France.