Associations Of Prenatal Exposure To Phthalates With Fetal Hormones And Early Childhood Growth

Phthalic acid esters（PAEs）,which are also known as phthalates,have been widely added as plasticizes to various consumer products.Generally,PAEs are physically embedded into plastic materials via hydrogen bonding and van der Waals forces,changes of the environment（temperature,pH,daylight,pressure,etc.）or contact with solvents can cause PAEs leakage from plastic materials.These PAEs further migrated to air,water and food,resulting in extensive human exposure.Pregnant women,as sensitive population,are more susceptible to PAEs.In addition,PAEs were detected in amniotic fluid,umbilical cord blood and neonatal urine,indicating that the fetuses were also exposed to PAEs.Many epidemiological studies have shown that prenatal phthalate exposure might have an adverse impact on maternal and infant health,such as the disruption of homeostasis and postnatal growth restriction,but the results were inconsistent.Different assessments of phthalate exposure might be one of the reasons for these inconsistent results.PAEs have a short half-life in the human body（less than 24 hours）.However,most of the studies have only measure the PAEs levels at one spot time,which only reflect the exposure in recent days or weeks.Therefore,in recent years,researchers suggested to collect samples from multiple time points to evaluate the phthalate exposure level of pregnant women throughout pregnancy.In addition,current studies only briefly discussed the effects of PAEs on a specific class of endocrine hormones,but lacked a comprehensive assessment of the endocrine disruption effects caused by PAEs.As we all know,hormone plays an important role in the maintenance of normal pregnancy and infant development.It is still unknown whether the hormone disruption by PAEs has an adverse impact on early childhood growth or not.Based on a prospective birth cohort,urine samples from pregnant women in the1^st,2ⁿdd and 3^rdd trimester of pregnancy were collected to assess the levels of phthalate exposure during pregnancy.The levels of hormones in heel blood samples and umbilical cord blood samples of newborns were measured to explore the relationships between PAEs exposure on fetal thyrotropin（TSH）,glucocorticoids and sex steroids.In addition,this study also explored the potential mediation role of fetal hormone in the associations between between phthalate exposure and the early childhood growth.Part 1:Profiles,variability,and predictors of urinary phthalate metabolites in pregnant womenObjective:We aimed to characterize the exposure profiles,temporal variability,and potential predictors of urinary phthalate metabolites during pregnancy.Methods:Based on a prospective birth cohort study,947 pregnant women were recruited from 2013 to 2015 who provided urine samples in the 1^st,2ⁿdd and 3^rdd trimester.Ultra-high performance liquid chromatography coupled with triple quadrupole mass spectrometry was used to measure nine phthalate metabolites in the2841 urine samples:MEHP,MECPP,MEHHP,MEOHP,MBzP,MiBP,MnBP and MMP.The correlation coefficients among phthalate metabolites were calculated by Spearman rank correlation analysis.We calculated the intraclass correlation coefficient（ICC）to assess the variability and investigated potential predictors of urinary phthalate metabolites by using the linear mixed models.Results:The detection rates of phthalate metabolites were more than 70%,except for MMP（less than 50%）.Geometric means（GM）of phthalate metabolites（ng/mL）in 1^st,2^ndand 3^rdtrimester were 0.09-47.41,0.06-29.90 and 0.07-43.85,respectively.Spearman rank correlation coefficients of phthalate metabolites were mostly above0.3,and the coefficients of DEHP metabolites were above 0.6（P<0.01）.ICC of urinary phthalate metabolites ranged from 0.21 to 0.44.Compared with mothers who were younger than 25 years old,the mothers aged 30-34 years had lower urinary MEOHP andΣDEHP levels[Percent changes=-23.37%（95%CI:-44.75%,-1.99%）and 21.43%（95%CI:-41.28%,-1.58%）,respectively].The pregnant women with pre-pregnancy BMI over 24 kg/m² had 38.55%（95%CI:5.46%,71.64%）higher concentrations of urinary MiBP than women with normal pre-pregcnany BMI.Compared with pregnant women who had lower education levels,the women with higher education levels exhibited 27.48%（95%CI:10.11%,44.85%）higher levels of urinary MEP.Across the four sampling seasons,the lowest levels of urinary phthalate metabolites were generally occurred in winter.Conclusion:In summary,exposure to PAEs was highly prevalent among the pregnant women in our study.Urinary concentrations of phthalate metabolites varied across trimesters during pregnancy.Factors such as age,BMI,education and sampling season were associated with the urinary concentrations of phthalate metabolites.Part 2.Associations between prenatal phthalate exposure and fetal hormone levelsObjective:To explore the associations between prenatal phthalate exposure and fetal thyrotropin（TSH）,glucocorticoid and sex hormone levels.Methods:（1）Based on Part 1,we selected the mother-infant pairs who have completed TSH screening test（n=718）.（2）Based on Part 1,we selected the population with cord blood samples（n=553）.The concentration of glucocorticoid（cortisol and cortisone）in umbilical cord blood was measured by liquid chromatography coupled with triple quadrupole mass spectrometry in positive-ion mode.（3）We selected the same population in（2）（n=553）.Liquid chromatography coupled with triple quadrupole mass spectrometry was used to detect cord blood progesterone,17-OH progestrone and androstendione in positive-ion mode,and estrone,estradiol and estriol in negative-ion mode.The distributions of nine phthalate metabolites,as well as hormone levels were skewed and the data were natural logarithm（ln-）.Generalized estimated equation（GEE）models were conducted to explore the relationships of phthalate metabolite concentrations at each trimester and TSH,glucocorticoid and sex hormone levels.We conducted sex-stratified analysis by fetal sex and calculated the P-values of the interaction terms between each phthalate metabolite and trimester and fetal sex in the models to explore the potential effect modification by fetal sex.Results:（1）The median of TSH concentration was 3.03μIU/mL.The concentrations of phthalate metabolites in 1^sttrimester were negatively associated with TSH levels.Each ln-unit increase in MiBP was associated with 4.90%（95%CI:-9.42%,-0.39%）decrease in TSH.This association was more pronounced in male infants.（2）The median concentrations of cortisol,cortisone and cortisol/cortisone ratio were 38.43ng/mL,79.17 ng/mL and 0.54.The associations between phthalates and glucocorticoids were varied by sex.Among the male infants,one ln-unit increase of MECPP,MEHHP and MEOHP in 3^rdtrimester was in relation to 7.27%-8.18%decrease in cortisol levels.However,among female infants,per ln-unit increase in MECPP,MEHHP,MEOHP and the sum molar concentrations of DEHP metabolites（ΣDEHP）was in relation to 5.45%-8.77%increment in cortisol/cortisone ratio.In addition,the urinary MEHP and MBzP concentrations in 1^sttrimeser were also positively associated with 2.79%-5.87%increase in glucocorticoid levels.（3）The median concentrations（ng/mL）of progesterone,17OH-progesterone,androstenedione,estrone,estradiol and estriol were 469.55,21.44,0.68,11.28,2.82 and 121.65,respectively.Urinary phthalate metabolites were negatively associated with androgen and positively associated with estrogen.In 1^stt trimester,each ln-unit increment of MECPP,MEOHP and MBzP was significantly correlated with 4.54%-4.82%reduction in androstenedione.Each ln-unit increase of MiBP and MnBP was associated with 4.13%and 5.08%increment in estradiol,respectively.In addition,there was also a positive association between urinary MEHHP in 3^rdtrimester and estriol.We also observed similar results in sex stratification analysis.Conclusions:Prenatal phthalate exposure was associated with fetal steroidogenesis.To our knowledge,this is the first study reported the sex difference on the association between phthalate and glucocorticoid.This study provides new epidemiological evidences for the impact of PAEs on fetal hormones.Part 3.Associations of early childhood growth with prenatal phthalate exposure and fetal hormonesObjective:To explore the associations of maternal urinary phthalate metabolites with early childhood growth and the potential mediating role of fetal hormones in these associations.Methods:We selected the population as same as Part 2.Using World Health Organization（WHO）child growth standards to normalize the offspring’s weights and heights at birth and 24 month GEE models were conducted to evaluate the association of PAEs exposure with early childhood growth.Multivariate regression models were used to explore the associations of fetal hormones with early childhood growth.We also conducted mediation analysis to explore the mediation role of hormones on these associations between phthalate exposure and early childhood growth.Result:（1）The means（standard deviation）of z-score of birth weight and length of newborns were-0.02（1.00）and-0.04（0.97）,respectively.Phthalate exposure was negatively correlated with birth length z-score.Each ln-unit increase in MBzP was associated with the birth length z-score（Percent changes=-7.44%;95%CI:-13.56%,-1.31%）.（2）The means（standard deviation）of z-score of weight and height of 24month children were 0.50（0.86）and 0.45（0.97）,respectively.Phthalate exposure was also negatively correlated with children’s weight z-score.One unit increase of ln-MEHHP in 2ⁿdd trimester and ln-MBzP in 3^rdd trimester was in relation to 9.07%（95%CI:-16.60%,-1.55%）and 7.72%（95%CI:-14.12%,-1.32%）decrease in weight z-score of 24 month children.We also observed similar results in sex stratification analysis.（3）Estradiol and 17OH-progesterone were negatively correlated with birth weight z-score.The mediation analysis suggested that 29.27%and 16.54%of the associations of urinary MiBP concentrations in 1^stt and 2ⁿdd trimester with birth weight z-score might be mediated by fetal estradiol levels.Conclusions:Prenatal phthalate exposure was negatively associated with lenth z-score of newborn and the weight z-score of 24 month children.In addition,the effect of PAEs on fetal estradiol might be a possible mechanism under the relationship between prenatal phtalate exposures with reduced early childhood growth.