Contexte : Les phtalates et les phénols sont des perturbateurs endocriniens. Les données concernant leurs effets sur la croissance fœtale et durant l'enfance sont limitées et suggèrent un effet dépendant du sexe de certains phénols sur le poids de naissance. Une des limites de ces études est l'estimation de l'exposition, basée sur la mesure de leurs concentrations dans un petit nombre d'échantillons d'urine maternelle. En raison de la faible persistance de ces composés chimiques dans l'organisme, les concentrations urinaires varient dans le temps. Les concentrations dosées dans d'autres matrices comme le liquide amniotique pourraient être pertinentes pour estimer l'exposition fœtale.Objectifs : Les objectifs de la thèse étaient : 1) d'étudier les impacts potentiels de l'exposition prénatale aux phénols et aux phtalates sur la croissance du fœtus et de l'enfant ; 2) de caractériser la variabilité des concentrations urinaires de phénols au long de la grossesse et de comparer les concentrations de phénols dosées dans le liquide amniotique et l'urine maternelle recueillis le même jour ; 3) de caractériser le biais et l'impact sur la puissance statistique de l'utilisation d'un faible nombre d'échantillons urinaires pour estimer les expositions. Méthodes : Les associations entre les phénols, les phtalates et la croissance ont été étudiées parmi un sous-effectif de la cohorte mère-enfant EDEN ayant accouché de garçons (n =520). La croissance fœtale a été estimée à l'aide d'échographies réalisées pendant la grossesse et de mesures à la naissance. La croissance postnatale a été modélisée à partir de mesures répétées normalisées, réalisées entre la naissance et 3 ans. Les biomarqueurs d'exposition aux phtalates et phénols ont été dosés dans les urines maternelles recueillies une fois pendant la grossesse. La variabilité des concentrations urinaires de phénols et la correspondance avec les concentrations mesurées dans le liquide amniotique ont été étudiées chez 71 femmes enceintes recevant une amniocentèse au centre médical Mount Sinaï (NY, États-Unis). Un échantillon d'urine maternelle a été recueilli le jour de l'amniocentèse, et à deux autres reprises pendant la grossesse. L'étude concernant les biais est basée sur des données simulées. Résultats : Les concentrations de triclosan étaient négativement associées à tous les paramètres de croissance mesurés à la troisième échographie (p ≤ 0,16) et avec la périmètre crânien à la naissance (β = - 1,4 mm, IC 95%; -2.8; 0.0). Les parabènes étaient associés positivement avec le poids à la naissance (p < 0,05). Le méthyle et propyle parabènes étaient aussi positivement associés au poids et à la circonférence abdominale à 3 ans (p-valeurs comprises entre 0,02 et 0,14). En ce qui concerne la variabilité des concentrations urinaires pendant la grossesse, les coefficients de corrélation intra-classe (ICC) variaient entre 0,48 et 0,62 pour l'ensemble des phénols sauf le bisphénol A (ICC = 0,11). Seuls la benzophénone-3 et le propyle parabène ont été détectés dans au moins 50 % des échantillons de liquide amniotique. Pour ces composés les concentrations dosées dans l'urine maternelle et le liquide amniotique, recueillis le même jour, étaient positivement associés. Dans le cadre d'une simulation, nous avons estimé que 5 échantillons d’urine étaient nécessaires pour estimer correctement l’exposition aux produits chimiques ayant un ICC de 0,6, tandis que pour des produits chimiques avec un ICC de 0,15, environ 25 échantillons étaient nécessaires.Conclusion : Un seul échantillon d'urine était disponible pour évaluer les expositions des femmes de la cohorte EDEN et nos résultats peuvent être affectés d'un biais résultant d'erreurs de classification des expositions, notamment pour le bisphenol A pour lequel nous avons observé une variabilité importante des concentrations. Néanmoins, notre étude suggérait un effet de l'exposition prénatale à certains phénols sur la croissance pré- et post-natale. / Background: Phthalates and phenols belong to the family of short half-life endocrine disruptors. Data regarding their effects on fetal and early post-natal growth in Human are sparse and suggest a sex-specific effect of some phenols on birth weight. One of the limitations of these studies is exposure assessment usually based on the measurement of their concentrations in a small number of maternal urine samples collected during pregnancy. Because of their low persistence in the organisms and the likely episodic nature of the exposures, urinary concentrations of these chemicals are likely to vary. Chemical concentrations measured in alternative matrix, such as amniotic fluid, might be a relevant dosimeter of fetal exposure.Objectives: Objectives of the thesis were: to study the potential effects of prenatal exposures to phenols and phthalates on pre- and early post-natal growth; to characterize variability in maternal urine concentrations of phenols throughout pregnancy and to compare phenol concentrations in amniotic fluid to those measured in maternal urine collected same day; to characterize the impact of increasing the number of measurements to estimate exposure on bias and statistical power of epidemiological studies.Methods: Associations between phenols, phthalates and growth were studied among a subsample of pregnant women of the French EDEN cohort delivered boys (n =520). We measured fetal growth with ultrasound (three times during pregnancy) and birth measurements. We used standardized measures acquired between birth and 3 years of age to model postnatal growth. We measured biomarkers of phthalates and phenols in maternal urines collected once during pregnancy: 191 women were assessed in 2008 and 410 other women in 2012 (ntot = 601). Variability in phenol urine concentrations and relationship between concentrations measured in amniotic fluid and maternal urine collected on the same day were studied among 71 pregnant women presenting for an amniocentesis at the Mount Sinai Medical Center (NY, USA). Maternal urine was collected at the time of the amniocentesis appointment, and on two subsequent occasions. Urine and amniotic fluid were analyzed for nine phenols.The study aiming at characterizing bias was based on simulated data. Results: Among the subsample of 191 pregnant women from the EDEN cohort, we observed a negative association between dichlorophenols and birth weight and a positive association between benzophenone-3 and birth weight. The associations with dichlorophenols were not replicated in the larger subsample of the EDEN cohort (n = 520). Triclosan concentration was negatively associated with all of the growth parameters measured at the third ultrasound examination (p ≤ 0.16) and with head circumference measured at birth (β = - 1.4 mm, 95% CI; -2.8; 0.0). All of the parabens were positively associated with weight at birth (p < 0.05). These associations remained in childhood for methyl- and propyl-parabens. Regarding the variability in phenol urinary concentrations during pregnancy, the intraclass correlation coefficients (ICC) ranged between 0.48 and 0.62 for all phenols except bisphenol A (ICC = 0.11). Only benzophenone-3 and propylparaben were detectable in at least half of the amniotic fluid samples; for these phenols, concentrations in maternal urine and amniotic fluid were positively associated. In a simulation study, we estimated that 5 samples will be needed to correctly estimate exposure to chemicals with ICC of 0.6, while for chemicals with ICC of 0.15 around 25 samples would be needed.Conclusion: We only had spot urine sample to assess exposure in the EDEN cohort and findings may be affected by exposure misclassification, especially for bisphenol A for which we observed high variability in urine concentrations. Nevertheless, our study lends support to a potential effect of prenatal exposure to some phenols on pre- and early post-natal growth.

Le bisphénol S (BPS) est largement utilisé comme substitut du Bisphénol A (BPA) et l’exposition humaine au BPS est désormais ubiquitaire. Or, le BPS, à l’instar du BPA, présente un potentiel perturbateur endocrinien, ce qui soulève la question du risque liée à une exposition fœtale au BPS pour la santé humaine. Dans ce contexte, l’objectif de cette thèse est d’évaluer l’exposition fœtale au BPS et de caractériser des biomarqueurs phénotypiques d’exposition fœtale et/ou d’effet du BPS à partir d’une signature stéroïdomique. Une étude toxicocinétique réalisée sur le modèle du fœtus ovin a montré que le transfert materno-fœtal du BPS est faible. Cependant, le BPS et son principal métabolite, le BPS glucuronide, sont lentement éliminés du compartiment fœtal en raison d’un passage placentaire fœto-maternel du BPS limité et de la faible vitesse de réactivation du BPSG en BPS. Il en résulte une exposition fœtale au BPS similaire à celle au BPA, lors d’exposition maternelle répétée. L’étude du transfert placentaire du BPS et du BPSG sur le modèle de placenta humain perfusé a conforté les résultats observés chez le mouton, avec des faibles transferts materno-fœtal et fœto-maternel du BPS, respectivement 10 et 3 fois inférieurs à ceux du BPA. L’exposition maternelle quotidienne au BPS au cours de la gestation chez la brebis n’a pas eu d’impact sur les voies de biosynthèse des androgènes dans l’unité materno-fœtoplacentaire pour les fœtus mâles. Bien que le potentiel d’exposition fœtale du BPS est similaire à celui du BPA, nous n’avons pas mis en évidence d’effets associés à cette exposition / Bisphenol S (BPS) is widely used as a substitute for Bisphenol A (BPA) and human exposure to BPS is now ubiquitous. However, BPS, like BPA, displays an endocrine disrupting potential, raising the issue of the risk of fetal exposure to BPS for human health. In this context, the objective of this thesis is to evaluate the fetal exposure to BPS and to characterize phenotypic biomarkers of fetal exposure and / or effect of BPS from a steroidal signature. A toxicokinetic study carried out on the model of the ovine fetus has shown that materno-fetal transfer of BPS is weak. However, BPS and its major metabolite, BPS glucuronide, are slowly eliminated from the fetal compartment due to the limited feto-maternal placental transfer of BPS and the low rate of reactivation of BPSG to BPS. This results in fetal exposure to BPS similar to BPA at repeated maternal exposure. The study of placental transfer of BPS and BPSG on the model of human perfused placenta reinforced the results observed in sheep, with low materno-fetal and feto-maternal transfers of BPS, respectively 10 and 3 times lower than those of BPA. Daily maternal exposure to BPS during pregnancy in ewes did not impact the androgen biosynthetic pathways in the materno-fetoplacental unit for male fetuses. Although the potential for fetal exposure of BPS is similar to that of BPA, we have not found any effects associated with this exposure.

O tabaco é uma das principais drogas consumidas mundialmente e seu uso por mulheres em idade reprodutiva, em particular, têm representado uma grande preocupação por parte de especialistas e da sociedade em geral. Apesar dos efeitos adversos associados ao ato de fumar durante a gestação serem bastante documentados e conhecidos, sabe-se que uma parcela de mulheres grávidas tem dificuldades em abandonar o hábito. A exposição fetal aos constituintes do tabaco tem sido associada com aumento do risco de aborto espontâneo e nascimentos prematuros, incidência de recém-nascidos com baixo peso, síndrome de morte súbita infantil e desordens cognitivas e neurocomportamentais. Entretanto, devido ao sentimento de culpa e medo de ações punitivas, mulheres raramente admitem terem utilizado tabaco durante a gestação. Como resultado, uma série de marcadores biológicos tem sido estudada para se diagnosticar a exposição fetal aos constituintes do tabaco. Dentre os marcadores utilizados estão a nicotina e o seu principal produto de biotransformação, a cotinina, que podem ser detectados em amostras de mecônio de recém-nascidos. No presente estudo, um método analítico foi desenvolvido visando à detecção desses marcadores em amostras de mecônio. Foi considerada neste projeto a técnica de extração acelerada por solvente (ASE) por ser uma técnica promissora para o preparo de amostras sólidas e/ou semi-sólidas. Os analitos foram identificados por cromatografia em fase-gasosa com detector de nitrogênio e fósforo (GC/NPD). Os limites de detecção foram de 3 ng/g e 30 ng/g e os de quantificação foram de 5 ng/g e 40 ng/g, para cotinina e nicotina respectivamente, e apresentaram boa linearidade na faixa de concentração estudada (5-500 ng/g), com coeficiente de correlação (r2) maior que 0,98. A precisão intra-ensaio, determinada pelo coeficiente de variação do método (CV%) foi menor que 15% e a precisão interensaio foram menor ou igual a 20% para nicotina e cotinina. A recuperação média foi de 77%. O método desenvolvido demonstrou ser rápido, preciso, prático e sensível e com uso de menores volumes de solventes orgânicos do que outros métodos descritos na literatura. O método desenvolvido e validado foi aplicado em amostras de mecônio de neonatos com suspeita ou não de exposição fetal aos constituintes do tabaco. / Tobacco is one of the main drugs consumed worldwide and its use by women of reproductive age in particular has played a major concern among experts and society in general. Despite the adverse effects associated with smoking during pregnancy are well documented, it is known that a significant proportion of pregnant women have difficulties to quit the habit. Fetal exposure to tobacco constituents has been associated with increased risk of spontaneous abortion (miscarriage) and premature births, incidence of newborns with low birth weight, sudden infant death syndrome and neurobehavioral and cognitive disorders. However, due to guilt and fear of punitive actions, women rarely admit to have used tobacco during pregnancy. As a result, a series of biological markers have been studied to diagnose fetal exposure to tobacco constituents. Nicotine and cotinine are some of these biomarkers which can be detected in meconium samples from newborns. In this study, an analytical method was developed to detect these biomarkers in meconium samples. Accelerated solvent extraction (ASE), a promising technique for solid or semi-solid sample preparation, was considered in this work the that is considered a. The analytes were identified by gas-chromatography with nitrogen phosphorus detector (GC / NPD). The limits of detection (LOD) were 3 ng/g and 30 ng/g and the limits of quantification (LOQ) were 5 ng/g and 40 ng/g for cotinine and nicotine, respectively. The method showed good linearity in the concentration range studied (5-500 ng/g), with a coefficient of correlation (r2) greater than 0.98. The intraday precision, determined by the coefficient of variation (CV %) was less than 15% and the interday precision was less or equal than 20% for nicotine and cotinine. The average recovery was 77%. The method proved to be fast, accurate, practical and sensitive and smaller volumes of organic solvents are necessary, compared to other methods published in the scientific literature. The developed and validated method was applied to meconium samples of suspected and nonsuspected neonates of having been exposed to tobacco constituents during gestation.

O consumo de drogas de abuso é grave problema de saúde pública em todo o mundo. No Brasil observa-se aumento no número de usuárias em idade fértil, levando a crescente preocupação com relação à exposição fetal. Efeitos deletérios como diminuição do peso e crescimento fetal, parto prematuro, déficits neurológicos e comportamentais estão associados ao uso de cocaína e cannabis durante a gestação. Portanto, é importante obter informações acerca do consumo de drogas durante a gravidez, permitindo intervenções médicas e psicológicas adequadas. Os relatos fornecidos pelas gestantes são relevantes, porém muitas vezes resultam em dados subestimados devido à omissão de informações por medo de ações punitivas. Desta forma, a confirmação por meio de análises toxicológicas em amostras biológicas se faz necessária. O mecônio, primeiras fezes do recém-nascido, tem sido proposto como matriz adequada para avaliação da exposição fetal, pois apresenta coleta não invasiva, fácil obtenção e fornece informações de longo prazo (2º e 3º trimestre de gestação). Entretanto, é uma matriz complexa, exigindo diversas etapas de purificação para posterior análise. No presente trabalho, métodos analíticos foram desenvolvidos visando à detecção dos biomarcadores da exposição fetal à cocaína e ao tetraidrocanabinol em amostras de mecônio por cromatografia gasosa acoplada à espectrometria de massas (GC-MS). Para ambos os métodos, a extração acelerada por solvente (ASE) foi utilizada para o isolamento dos analitos de interesse de amostras de mecônio, já que esta apresenta vantagens frente às técnicas convencionais, devido sua maior eficiência e menor manipulação da amostra. Associada à ASE, a extração em fase sólida (SPE) foi empregada para purificação e concentração dos analitos de interesse. Na etapa de desenvolvimento, inicialmente foram estabelecidos os procedimentos de derivatização e as condições cromatográficas a serem empregadas nas análises. Posteriormente, realizou-se a otimização dos procedimentos empregados na ASE, através de análise de superfície de resposta. Os métodos foram validados de acordo com o preconizado por referências internacionais, estabelecendo-se os limites de detecção e quantificação, recuperação, linearidade, precisão intra e interensaio e exatidão. O método para detecção dos biomarcadores da cocaína foi aplicado em 342 amostras de mecônio, provenientes do Hospital Universitário da Universidade de São Paulo (HU-USP). Destas, 19 (5,6%) apresentaram resultado positivo para um ou mais biomarcadores da exposição fetal à cocaína. Além disto, foi observada redução estatisticamente significante do peso ao nascimento, comprimento e perímetro cefálico entre os recém-nascidos expostos à cocaína durante a gestação. O método para a detecção da exposição fetal ao tetraidrocanabinol foi aplicado em 6 amostras positivas, obtidas do HU-USP, mostrando a aplicabilidade da técnica desenvolvida. / The use of illicit drugs is a relevant public health problem in the world. In Brazil, the number of women users in fertile age is increasing, which leads to a growing concern regarding fetal drug exposure. Adverse outcomes such as low birth weight, intra-uterine growth restriction, preterm birth, neurobehavioral and developmental deficits are associated with cocaine and cannabis use during pregnancy. Consequently, it would be important to obtain data related to drug misuse during gestation with the aim to plan medical and psychological interventions. Self-report drug use by pregnant women is often inaccurate due to feelings of guilt or fear of punitive actions. Therefore, confirmation by toxicological analysis in biological matrices must be accomplished. Meconium, the first stool of the newborn, has been proposed as a proper matrix to evaluate fetal exposure because it is collected by an easy and non-invasive way and enables the achievement of long-term information regarding fetal exposure. However, meconium is a complex matrix, which requires extensive sample cleaning previously to the analytes identification. In the present research, analytical methods were developed aiming the determination of cocaine and tetrahydrocannabinol biomarkers in meconium samples through gas chromatography-mass spectrometry (GC-MS). Accelerated solvent extraction (ASE) was used for analytes isolation due to its advantages over conventional techniques, such as greater extraction efficiency and minor sample handling. In order to achieve proper analytes selectivity and detectability, solid phase extraction (SPE) was employed for post-extraction clean-up. Initially, the derivatization procedure and chromatographic parameters were established for the development of the methods. Afterwards, ASE procedure was optimized through response surface methodology. The analytical methods were validated in accordance to international references. Limits of detection and quantification, recovery, linearity, precision and accuracy were obtained. The developed method for cocaine was applied in 342 meconium samples collected from the University Hospital of University of São Paulo (HU-USP). Among them, 19 (5.6%) showed positivity result for cocaine biomarkers. Additionally, newborns from mothers exposed to cocaine exhibited statistical significant lower birth weight, length and head circumference when compared with newborns from non-consumer mothers. For the method of tetrahydrocannabinol, applicability and importance was demonstrated by analyzing 6 positive samples from HU-USP.

L’exposition au bisphénol A (BPA) au cours du développement fœtal est suspectée d’être impliquée dans l’initiation d’effets biologiques. Dans ce contexte, l’objectif de cette thèse est d’évaluer l’exposition humaine fœtale au BPA. Une étude pharmacocinétique (PK) réalisée sur le modèle du fœtus ovin a montré que le BPA glucuronide (BPAG), métabolite majeur du BPA piégé dans le compartiment fœtal, est lentement éliminé via sa réactivation en BPA. A partir de ces données PK, le développement d’un modèle PK humanisé a permis de prédire des concentrations plasmatiques fœtales de BPAG maintenues autour de 40ng/L chez le fœtus humain en fin de grossesse, soit environ 1000 fois supérieures aux concentrations en BPA correspondantes. Une analyse protéomique de tissus fœtaux ovins a montré qu’une même dose molaire de BPAG ou de BPA affecte des voies physiologiques similaires, suggérant que l’hydrolyse du BPAG pourrait exposer les tissus à des concentrations en BPA compatibles avec l’expression d’un effet biologique. L’ensemble de ces résultats indique que les concentrations de BPAG dans le sang de cordon constituent un marqueur pertinent de l’exposition fœtale au BPA et qu’elles doivent être prises en compte dans l’évaluation du risque / Prenatal exposure to bisphenol A (BPA) is suspected to induce adverse effects later in life. In this context, the goal of this thesis was to evaluate the human fetal exposure to BPA. Using a pharmacokinetic (PK) approach, we showed that BPA glucuronide (BPAG), the main metabolite of BPA that remains trapped in the fetal compartment, is slowly eliminated through its back conversion into BPA. A humanized PK model developed from these data predicted steady BPAG fetal plasma concentrations of 40ng/L in late pregnancy, i.e. about 1000 folds higher than corresponding BPA concentrations. A proteomic analysis of fetal tissues has shown that equimolar BPAG or BPA fetal exposures trigger similar physiological shifts in the ovine model, suggesting that BPAG hydrolysis might expose fetal tissues to BPA at levels of functional significance. It is concluded that BPAG levels in human cord blood are relevant indicators of fetal exposure to BPA during late pregnancy and should be taken into account for risk assessment

Le Bisphénol A (BPA) est un perturbateur endocrinien dont les effets développementaux observés chez les rongeurs soulèvent la question du risque pour la santé humaine relatif à une exposition fœtale au BPA. L’objectif de cette thèse est de déterminer les mécanismes toxicocinétiques impliqués dans l’exposition fœtale au BPA. La caractérisation in vivo dans un modèle intégratif ovin des expositions maternelles et fœtales au BPA et à ses métabolites ont permis d’identifier le transfert placentaire et le métabolisme fœto-placentaire comme les déterminants majeurs de l’exposition fœtale au BPA. Le transfert bidirectionnel du BPA à travers le placenta humain se fait par diffusion passive conduisant à un rapport maximal des concentrations plasmatiques de BPA libre entre le fœtus et sa mère de 1. En revanche, la perméabilité placentaire du BPA-G est très limitée, en particulier dans le sens materno-fœtal. Les activités de conjugaison hépatique du BPA ont été faibles chez le fœtus ovin à un stade précoce de gestation et ont augmenté au cours du développement. Par ailleurs la réactivation des conjugués du BPA mise en évidence ex vivo dans les gonades fœtales ovines pourrait conduire à une exposition locale au BPA actif. L’ensemble de ces données suggère que le début de la gestation pourrait représenter une fenêtre critique d’exposition au BPA / Bisphenol A (BPA) an endocrine disruptor interfering with developmental processes in rodents, raises the question of risk for human health related to fetal exposure to BPA. The goal of this work was to determine the toxicokinetic mechanisms involved in fetal exposure to BPA. The disposition of BPA and its metabolites in the maternal-placental-fetal unit in an in vivo ovine model enabled us to identify the placental transfer and the fetal-placental metabolism as the major determining factors of fetal exposure to BPA. Bidirectional placental transfer of BPA occurs by passive diffusion leading to a ratio of free BPA between the fetal and maternal plasma concentrations of about 1. By contrast, the permeability of BPA-G is very limited, particularly in materno-to-fetal direction. The hepatic conjugation activities were very low in ovine fetus at an early stage of development and increased throughout pregnancy. Hydrolysis of BPA conjugates observed ex vivo into fetal ovine gonads could lead to local exposure to native BPA. Altogether, these results suggest that the early stage of pregnancy is a critical window of exposure for the developing fetus

O consumo de drogas de abuso é grave problema de saúde pública em todo o mundo. No Brasil observa-se aumento no número de usuárias em idade fértil, levando a crescente preocupação com relação à exposição fetal. Efeitos deletérios como diminuição do peso e crescimento fetal, parto prematuro, déficits neurológicos e comportamentais estão associados ao uso de cocaína e cannabis durante a gestação. Portanto, é importante obter informações acerca do consumo de drogas durante a gravidez, permitindo intervenções médicas e psicológicas adequadas. Os relatos fornecidos pelas gestantes são relevantes, porém muitas vezes resultam em dados subestimados devido à omissão de informações por medo de ações punitivas. Desta forma, a confirmação por meio de análises toxicológicas em amostras biológicas se faz necessária. O mecônio, primeiras fezes do recém-nascido, tem sido proposto como matriz adequada para avaliação da exposição fetal, pois apresenta coleta não invasiva, fácil obtenção e fornece informações de longo prazo (2º e 3º trimestre de gestação). Entretanto, é uma matriz complexa, exigindo diversas etapas de purificação para posterior análise. No presente trabalho, métodos analíticos foram desenvolvidos visando à detecção dos biomarcadores da exposição fetal à cocaína e ao tetraidrocanabinol em amostras de mecônio por cromatografia gasosa acoplada à espectrometria de massas (GC-MS). Para ambos os métodos, a extração acelerada por solvente (ASE) foi utilizada para o isolamento dos analitos de interesse de amostras de mecônio, já que esta apresenta vantagens frente às técnicas convencionais, devido sua maior eficiência e menor manipulação da amostra. Associada à ASE, a extração em fase sólida (SPE) foi empregada para purificação e concentração dos analitos de interesse. Na etapa de desenvolvimento, inicialmente foram estabelecidos os procedimentos de derivatização e as condições cromatográficas a serem empregadas nas análises. Posteriormente, realizou-se a otimização dos procedimentos empregados na ASE, através de análise de superfície de resposta. Os métodos foram validados de acordo com o preconizado por referências internacionais, estabelecendo-se os limites de detecção e quantificação, recuperação, linearidade, precisão intra e interensaio e exatidão. O método para detecção dos biomarcadores da cocaína foi aplicado em 342 amostras de mecônio, provenientes do Hospital Universitário da Universidade de São Paulo (HU-USP). Destas, 19 (5,6%) apresentaram resultado positivo para um ou mais biomarcadores da exposição fetal à cocaína. Além disto, foi observada redução estatisticamente significante do peso ao nascimento, comprimento e perímetro cefálico entre os recém-nascidos expostos à cocaína durante a gestação. O método para a detecção da exposição fetal ao tetraidrocanabinol foi aplicado em 6 amostras positivas, obtidas do HU-USP, mostrando a aplicabilidade da técnica desenvolvida. / The use of illicit drugs is a relevant public health problem in the world. In Brazil, the number of women users in fertile age is increasing, which leads to a growing concern regarding fetal drug exposure. Adverse outcomes such as low birth weight, intra-uterine growth restriction, preterm birth, neurobehavioral and developmental deficits are associated with cocaine and cannabis use during pregnancy. Consequently, it would be important to obtain data related to drug misuse during gestation with the aim to plan medical and psychological interventions. Self-report drug use by pregnant women is often inaccurate due to feelings of guilt or fear of punitive actions. Therefore, confirmation by toxicological analysis in biological matrices must be accomplished. Meconium, the first stool of the newborn, has been proposed as a proper matrix to evaluate fetal exposure because it is collected by an easy and non-invasive way and enables the achievement of long-term information regarding fetal exposure. However, meconium is a complex matrix, which requires extensive sample cleaning previously to the analytes identification. In the present research, analytical methods were developed aiming the determination of cocaine and tetrahydrocannabinol biomarkers in meconium samples through gas chromatography-mass spectrometry (GC-MS). Accelerated solvent extraction (ASE) was used for analytes isolation due to its advantages over conventional techniques, such as greater extraction efficiency and minor sample handling. In order to achieve proper analytes selectivity and detectability, solid phase extraction (SPE) was employed for post-extraction clean-up. Initially, the derivatization procedure and chromatographic parameters were established for the development of the methods. Afterwards, ASE procedure was optimized through response surface methodology. The analytical methods were validated in accordance to international references. Limits of detection and quantification, recovery, linearity, precision and accuracy were obtained. The developed method for cocaine was applied in 342 meconium samples collected from the University Hospital of University of São Paulo (HU-USP). Among them, 19 (5.6%) showed positivity result for cocaine biomarkers. Additionally, newborns from mothers exposed to cocaine exhibited statistical significant lower birth weight, length and head circumference when compared with newborns from non-consumer mothers. For the method of tetrahydrocannabinol, applicability and importance was demonstrated by analyzing 6 positive samples from HU-USP.

O tabaco é uma das principais drogas consumidas mundialmente e seu uso por mulheres em idade reprodutiva, em particular, têm representado uma grande preocupação por parte de especialistas e da sociedade em geral. Apesar dos efeitos adversos associados ao ato de fumar durante a gestação serem bastante documentados e conhecidos, sabe-se que uma parcela de mulheres grávidas tem dificuldades em abandonar o hábito. A exposição fetal aos constituintes do tabaco tem sido associada com aumento do risco de aborto espontâneo e nascimentos prematuros, incidência de recém-nascidos com baixo peso, síndrome de morte súbita infantil e desordens cognitivas e neurocomportamentais. Entretanto, devido ao sentimento de culpa e medo de ações punitivas, mulheres raramente admitem terem utilizado tabaco durante a gestação. Como resultado, uma série de marcadores biológicos tem sido estudada para se diagnosticar a exposição fetal aos constituintes do tabaco. Dentre os marcadores utilizados estão a nicotina e o seu principal produto de biotransformação, a cotinina, que podem ser detectados em amostras de mecônio de recém-nascidos. No presente estudo, um método analítico foi desenvolvido visando à detecção desses marcadores em amostras de mecônio. Foi considerada neste projeto a técnica de extração acelerada por solvente (ASE) por ser uma técnica promissora para o preparo de amostras sólidas e/ou semi-sólidas. Os analitos foram identificados por cromatografia em fase-gasosa com detector de nitrogênio e fósforo (GC/NPD). Os limites de detecção foram de 3 ng/g e 30 ng/g e os de quantificação foram de 5 ng/g e 40 ng/g, para cotinina e nicotina respectivamente, e apresentaram boa linearidade na faixa de concentração estudada (5-500 ng/g), com coeficiente de correlação (r2) maior que 0,98. A precisão intra-ensaio, determinada pelo coeficiente de variação do método (CV%) foi menor que 15% e a precisão interensaio foram menor ou igual a 20% para nicotina e cotinina. A recuperação média foi de 77%. O método desenvolvido demonstrou ser rápido, preciso, prático e sensível e com uso de menores volumes de solventes orgânicos do que outros métodos descritos na literatura. O método desenvolvido e validado foi aplicado em amostras de mecônio de neonatos com suspeita ou não de exposição fetal aos constituintes do tabaco. / Tobacco is one of the main drugs consumed worldwide and its use by women of reproductive age in particular has played a major concern among experts and society in general. Despite the adverse effects associated with smoking during pregnancy are well documented, it is known that a significant proportion of pregnant women have difficulties to quit the habit. Fetal exposure to tobacco constituents has been associated with increased risk of spontaneous abortion (miscarriage) and premature births, incidence of newborns with low birth weight, sudden infant death syndrome and neurobehavioral and cognitive disorders. However, due to guilt and fear of punitive actions, women rarely admit to have used tobacco during pregnancy. As a result, a series of biological markers have been studied to diagnose fetal exposure to tobacco constituents. Nicotine and cotinine are some of these biomarkers which can be detected in meconium samples from newborns. In this study, an analytical method was developed to detect these biomarkers in meconium samples. Accelerated solvent extraction (ASE), a promising technique for solid or semi-solid sample preparation, was considered in this work the that is considered a. The analytes were identified by gas-chromatography with nitrogen phosphorus detector (GC / NPD). The limits of detection (LOD) were 3 ng/g and 30 ng/g and the limits of quantification (LOQ) were 5 ng/g and 40 ng/g for cotinine and nicotine, respectively. The method showed good linearity in the concentration range studied (5-500 ng/g), with a coefficient of correlation (r2) greater than 0.98. The intraday precision, determined by the coefficient of variation (CV %) was less than 15% and the interday precision was less or equal than 20% for nicotine and cotinine. The average recovery was 77%. The method proved to be fast, accurate, practical and sensitive and smaller volumes of organic solvents are necessary, compared to other methods published in the scientific literature. The developed and validated method was applied to meconium samples of suspected and nonsuspected neonates of having been exposed to tobacco constituents during gestation.

Le méconium constitue les premières selles du nourrisson. Cette matrice complexe est analysée dans le cadre du projet « PENEW » (Pregnancy Environment and NEWborn malformations) afin de déterminer si le degré d'exposition fœtale aux xénobiotiques joue un rôle dans la survenue de malformations congénitales. Le but de cette étude était le développement analytique pour la détection et la quantification dans le méconium d’une quarantaine de composés de familles différentes (composés organiques volatils, pesticides, éthers de glycol et les métabolites associés). Trois techniques analytiques différentes et une préparation d’échantillon spécifiques ont été développées pour la détection et la quantification de ces composés dans le méconium. L’application des ces méthodes à 246 échantillons de méconium a montré une exposition fœtale à plusieurs des substances recherchées. / Meconium is the earliest stool of newborns. This complex matrix was analyzed through the "PENEW" project (Pregnancy and Newborn malformations Environment) to determine if the degree of fetal exposure to xenobiotics has an influence in the occurrence of birth defects. The objective of this thesis was developed analytical methods for the detection and quantification of several different families of compounds (volatile organic compounds, pesticides, glycol ethers and their metabolites) in meconium. Three different analytical methods and a specific sample preparation have been developed for the detection and quantification of these compounds in the meconium. The application of these methods to 246 meconium samples showed a fetal exposure to several target compounds.

O álcool é uma das substâncias psicoativas mais consumidas mundialmente e seu uso por mulheres em idade reprodutiva, em particular, tem representado grande preocupação por parte de especialistas e da sociedade em geral. Apesar dos efeitos adversos associados ao ato de ingerir bebidas alcoólicas durante a gestação ser bastante documentados e conhecidos, sabe-se que uma parcela de mulheres grávidas tem dificuldades em abandonar o hábito. O consumo excessivo de álcool durante a gravidez tem sido associado com a síndrome fetal pelo álcool (FAS), caracterizada por crianças com dificuldades comportamentais e de aprendizado. Entretanto, devido ao sentimento de culpa e medo de ações punitivas, mulheres raramente admitem terem utilizado álcool durante a gestação. Como resultado, uma série de marcadores biológicos tem sido estudada para se diagnosticar a exposição fetal ao etanol. Dentre os marcadores utilizados estão os ésteres etílicos de ácidos graxos (FAEE), que podem ser detectados em amostras de mecônio de recém-nascidos. No presente trabalho, um método analítico foi desenvolvido visando a detecção de oito FAEEs em amostras de mecônio e aplicada em amostras coletadas de recém-nascidos cujas mães admitiram ou não o uso de etanol durante a gestação. A microextração em fase sólida por Headspace (HS-SPME), uma técnica de preparação de amostras relativamente recente, foi utilizada para análise. Os FAEEs foram identificados e quantificados por cromatografia gasosa/espectrometria de massas (GC/MS), operado no modo de ionização química. Os correspondentes ésteres etílicos deuterados foram sintetizados e utilizados como padrão interno. Os limites de quantificação (LOQ) obtidos foram abaixo de 150 ng/g e limites de detecção (LOD) foram abaixo de 100ng/g para todos os analitos. O método mostrou boa linearidade na concentração estudada (LOQ-2000ng/g), com coeficiente (r2) melhor que 0.98. Os valores de precisão apresentaram coeficientes de variação menores que 15% para todos os FAEEs estudados. Quando o método foi aplicado em amostras de mecônio, foi possível detectar níveis de alguns FAEEs de recém-nascidos não suspeitos a exposição fetal ao etanol. / Alcohol is the main psychoactive drug consumed worldwide and its increasing use by young women has been a great problem point out by specialists in the subject. Although the adverse effects associated to the habit of drinking alcoholic beverages during gestation being very much documented, it is known that a considerable number of pregnant women have difficulties to abandon the habit. Excessive alcohol use during pregnancy has been associated with Fetal Alcohol Syndrome (FAS) characterized by children with cognitive and behavioral disorders. However, because of denial, embarrassment and fear, maternal reports of gestational use of alcohol are often inaccurate. Consequently, a series of biomarkers have been studied to diagnose fetal exposure to alcohol. Recently, fatty acid ethyl esters (FAEE) have been studied as biomarkers found in meconium of neonates exposed in utero. In the present work, an analytical method was developed aiming the detection of eight FAEEs in meconium samples and applied to real specimens collected from newborns whose mothers admitted or not the use of alcohol during pregnancy. Headspace solid-phase microextraction (HS-SPME), a relatively recent sample preparation technique, was used for analysis. FAEEs were identified and quantified by gas chromatography/mass spectrometry (GC/MS), operated in chemical ionization mode. The corresponding deuterated ethyl esters were synthesized and used as internal standards. The lower limits of quantification (LOQ) obtained were below 150ng/g and limits of detection (LOD) were bellow 100ng/g for all analytes. The method showed good linearity in the range of concentration studied (LOQ-2000ng/g), with coefficient of linearity better than 0.98. The precision assay, given by the relative standard deviation (RSD) of the method was lower than 15% for all FAEEs studied. When the method was applied to real samples, it was possible to detect trace levels of some FAEEs from non-suspected ethanol exposed newborns.