Associations Of In Utero Organochlorine Pesticides Exposure With Fetal Growth And Child Growth In The First 2 Years

Organochlorine pesticides(OCPs)are a kind of high-efficiency and broad-spectrum pesticides that have been widely used and also belong to persistent organic pollutants(POPs).Although OCPs have been banned for decades,they still can be detected in a variety of environmental media and human biological samples.Given that China used to be one the largest producers and users of OCPs,population exposure levels and health effects of OCPs deserve attention.Periods of the fetus and early childhood are critical ones in the course of individual growth.OCPs could be transferred into the fetus through the placental barrier and accumulate in human body,which could induce adverse health effects on individual growth.Animal studies have reported that OCPs have embryotoxicity and developmental toxicity,but findings from existing epidemiological studies for effects of OCPs on individual growth are mixed.Moreover,the existing research lacks the exploration of the dynamic characteristics of children’s growth and the mixed exposure of OCPs.Animal studies have suggested that OCPs have the properties of environmental endocrine disruptors.Known as a sensitive population,fetuses are more susceptible to the effects of OCPs on their endocrine hormone homeostasis.However,findings from epidemiological studies on the impacts of OCPs on the human endocrine system are still inconsistent.Endocrine hormones play an essential role in the normal growth of offspring,but it is unclear whether OCPs affect the growth of offspring by causing changes in endocrine hormone levels.Therefore,it is necessary to explore the associations of OCPs exposure with growth of the fetus and children as well as hormone levels of the fetus.As one class of POPs,OCPs could enter the fetus through the placental barrier and accumulate in human body.Many studies have shown that the levels of OCPs in maternal blood and cord blood are highly correlated,and OCPs have a long half-life in the human body.Therefore,this study applied cord blood OCPs levels to assess intrauterine exposure.Based on a birth cohort,this study detected the levels of OCPs in cord serum to evaluate intrauterine exposure to OCPs.The single-pollutant and multi-pollutant models were used to explore the relationships between OCPs exposure and the dynamic growth of fetuses and children.Additionally,this study evaluated the impacts of OCPs exposure on the levels of corticoids and sex hormones in cord serum,and applied the method of mediation effect analysis to further explore the potential roles of corticoids and sex hormones in the associations between OCPs and child growth.Part 1.Associations of in Utero Exposure to OCPs with Fetal GrowthObjective: To assess the levels of in utero exposure to OCPs and their associations with fetal growth in the study population.Method: This study was conducted based on a birth cohort in Wuhan City,with 1 046mother-infant pairs who joined the cohort in 2014-2015 and provided sufficient cord serum samples.Gas chromatography-triple quadrupole tandem mass spectrometry was used to quantify nine kinds of OCPs in cord serum.Serum total lipids were measured by gravimetric method.The fetal biparietal diameter,abdominal circumference,femoral length,and estimated fetal weight were obtained by prenatal ultrasonography in the first,second,third trimester and pre-delivery stages.The indicators above were standardized based on gestational age using generalized additive methods for location,scale and shape(GAMLSS)to obtain the standard deviation score(SD score).Then,evaluation indicators such as low weight status,slow growth status,and adverse growth status of fetus were also conducted.The concentrations of OCPs were corrected by serum total lipid content,and the natural logarithm(ln-)transformation of OCPs was performed to satisfy statistical normality.Correlations among OCPs were assessed by Spearman’s rank correlation coefficient.Multiple linear regression models were used to explore the influencing factors of cord serum OCPs concentrations,and the associations of OCPs with fetal growth indicators.Mixed linear models were used to analyze the mean effects of OCPs exposure on fetal growth.The weighted quantile sum(WQS)regression model was used to evaluate the effect of mixed exposure to OCPs on fetal growth.Gender-stratified analysis as well as interaction analysis of pollutants and gender were used to explore the potential impact of fetal gender on these associations.Results:(1)Among the nine OCPs detected in this study,the detection rates of 6 kinds of OCPs(α-HCH,β-HCH,γ-HCH,p,p’-DDE,p,p’-DDD,and p,p’-DDT)were above 50%.The median concentrations of β-HCH and p,p’-DDE were 8.43 ng/g and 31.92 ng/g respectively,which were higher than other OCPs.The influencing factors of OCPs levels in cord serum included maternal age,pre-pregnancy body mass index(BMI),weight gain during pregnancy,maternal education level,average annual family income,and passive smoking during pregnancy.There were differences in the influencing factors of OCPs.(2)OCPs exposure was mainly negatively associated with fetal growth indicators.One ln-unit increase of γ-HCH was in relation to 3.29%～7.68% decrease of fetal biparietal diameter,abdominal circumference,estimated body weight and femur length SD score in the first trimester,and4.23%～ 4.29% decrease of the fetal abdominal circumference and estimated body weight SD score in the third trimester.In the mixed linear models,each unit increase of ln-γ-HCH was associated with a 2.71%(95% CI:-5.02%,-0.40%)decrease in estimated body weight and a2.47%(95% CI: 95% CI:-4.74%,-0.19%)decrease of femur length SD score.In the WQS regression model,the negative WQS index of mixed OCPs exposure was significantly associated with estimated body weight SD score in the first trimester,third trimester and before delivery.In detail,γ-HCH contributed the largest weight in the associations with first trimester and third trimester,with 77% and 79%,respectively.(3)γ-HCH exposure may increase the risk of fetal low weight status.Each unit increase of ln-γ-HCH was related to a16% increase in risk of fetal low weight status in the third trimester(OR = 1.16;95% CI: 1.02,1.33).In the WQS regression model,mixed exposure to OCPs significantly increased the risk of low birth weight in the third trimester,with γ-HCH(61%)contributing the largest weight.Conclusion: The study populations in Wuhan may be widely exposed to a variety of OCPs,mainly β-HCH and p,p’-DDE.γ-HCH exposure was related to the reduction of fetal estimated body weight and femoral length,and may increase the risk of fetal low weight status.The results of the study suggested that in utero OCPs exposure may affect fetal growth.Part 2.Associations of in utero exposure to OCPs with child growth in the first 2 yearsObjective: To explore the relationships between exposure to OCPs and physical indicators,growth rate and growth trajectory of children in the first two years.Method: This part was conducted based on the 1 046 mothers-infants pairs whose cord serum had been determined for OCPs in the first part.In this part,7 infants with birth defects were excluded.This study followed up and collected physical information of children at birth,1,6,12,and 24 months of age.According to the child growth standards issued by the World Health Organization(WHO),the height,weight and BMI of children were standardized by gender and months of age to obtain the z-scores of the corresponding indicators.The growth rate of infants was defined as the difference of weight z-score between birth and other time points.The latent class growth model(LCGM)was used to fit the BMI z-score growth trajectories of children.Multiple linear regression models were used to explore the associations between the levels of OCPs in cord serum and the z-score indicators as well as the growth rate of children.Mixed linear models were employed to evaluate the total effects of OCPs exposure on child growth.Multinomial logistic regression models were used to assess the impacts of OCPs exposure on child growth trajectories.Linear-mixed growth curve models were used to estimate the difference of BMI z-score trajectories among children in groups of different OCPs exposure levels.WQS regression model was applied to assess the overall effect of mixed exposure to OCPs.Gender-stratified analysis as well as interaction analysis of pollutants and gender were used to explore the potential influences of neonatal gender on associations above.Results:(1)In mixed linear models,p,p’-DDE levels were negatively correlated with BMI z-scores of children in the first two years.In the WQS regression model,the negative WQS index of mixed exposure to OCPs was significantly associated with child BMI z-score at birth and 6 months,while the positive WQS index was significantly correlated with child BMI z-score at 6,12,and 24 months.(2)This study fitted five growth trajectories,including slow growth(7.42%),catch-up growth(16.40%),suitable growth(32.08%),faster growth(40.02%),and overgrowth(4.08%)trajectory in LCGM according to the BMI z-score of children.Compared to the suitable growth trajectory,exposure to β-HCH(OR = 1.32;95%CI: 1.09,1.60)and p,p’-DDE(OR = 1.21;95% CI: 1.04,1.42)may increase the possibility of catch-up growth of children;exposure to β-HCH may also increase the possibility of faster growth of children(OR = 1.16;95% CI: 1.00,1.34);exposure to p,p’-DDT increased the possibility of overgrowth trajectory of children(OR = 1.18;95% CI: 1.02,1.36).In the WQS regression model,compared with the suitable growth trajectory,the positive WQS index could increase the possibility of catch-up growth,faster growth and overgrowth trajectory of children.The substances with the largest contribution weight in the associations above are p,p’-DDE(38%),β-HCH(31%)and p,p’-DDT(51%).(3)In the linear-mixed growth curve model,compared with the low concentration group,the monthly BMI z-score increased by 2.75%(95% CI: 0.3%,5.19%)and 2.93%(95% CI: 0.49%,5.38%)of the children in high β-HCH group and the high γ-HCH group.Conclusion: Exposure to OCPs may affect the physical indicators of children in the first two years,and may increase the possibility of catch-up growth,faster growth and overgrowth trajectory of children.Part 3.The mediating roles of cord serum steroid hormones in the associations between in utero exposure to OCPs and child growth in the first 2 yearsObjective: To explore the associations of in utero exposure to OCPs with sex hormones and corticoids in cord serum,and evaluate the mediating role of cord serum steroid hormones in the relationships between exposure to OCPs and the child growth.Methods: This part was conducted based on the 1 046 mother-infant pairs whose cord serum OCPs were determined in the first part.Children with birth defects,insufficient cord serum,premature birth,and the ones whose mothers with gestational diabetes mellitus or hypertensive disease were further excluded.The concentrations of corticoids and sex hormones in cord serum of 874 infants were quantified by liquid chromatography-triple quadrupole tandem mass spectrometry.Multiple linear regression models were used to explore the relationships of levels of cord serum OCPs with levels of cord serum steroid hormones,and relationships of cord serum hormone levels with z-score indicators as well as the growth rate of children.Mixed linear model was used to estimate the average effects of cord serum steroid hormones on child growth in the first two years.Multinomial logistic regression model was used to explore the impacts of cord serum steroid hormones on child growth trajectories.Combined with the results above,we screened out the mediating pathways that satisfied the criteria of mediation effect analysis,and explored the mediating role of cord serum steroid hormones in the association between OCPs and child growth.Results:(1)The median concentrations of cortisol,cortisone,11-deoxycortisol,corticosterone,11-deoxycorticosterone,and cortisol/cortisone in cord serum were 33.99ng/m L,85.84 ng/m L,2.20 ng/m L,1.22 ng/m L,0.75 ng/m L,and 0.47,respectively.Exposure to OCPs was negatively correlated with cord serum corticoids.Specifically,one ln-unit increase of OCPs was related with 1.80%～4.33% decrease of cord serum corticoids.(2)The median concentrations of progesterone,17-OH progesterone,androstenedione,estrone,estradiol,and estriol in cord serum were 460.33 ng/m L,20.50 ng/m L,0.58 ng/m L,10.13 ng/m L,3.20 ng/m L,and 108.41 ng/m L,respectively.Exposure to OCPs was negatively correlated with cord serum estrogens.Each unit increase of ln-OCPs was related with 1.30%～3.84% decrease of the cord serum estrogens.But each unit increase of ln-γ-HCH was related with a 1.72% increase of the cord serum androstenedione.After sex stratification,the associations of OCPs with sex hormones were significant only in boys.(3)The results showed that each unit increase of ln-androstenedione was related with a 23.17%increase in growth rate of children in 0-6 months of age,and a 24.29% in BMI z-score in 6months of age in boys,and an 85% increase in possibility of rapid weight gain status in boys aged 0-6 months.The mediating effect analysis found that androstenedione could mediate the association between γ-HCH and the growth rate of children aged 0-6 months,and the percentage of mediating effect was 6.01%.In boys,androstenedione was found to mediate the associations of γ-HCH,p,p’-DDE and ∑DDTs with BMI z-score at 6 months of age,whose percentages of mediating effect were 11.91%,20.05% and 16.38%,respectively;androstenedione could also mediate the association between γ-HCH and the growth rate of0-6 months and rapid weight gain status in boys aged 0-6 months,and the percentages of mediating effect were 9.96% and 11.53%,respectively.Conclusion: Exposure to OCPs was negatively associated with cord serum levels of corticoids and estrogens,but positively associated with androstenedione.The mediation effect analysis found that cord blood androstenedione mediated the association between in utero OCPs exposure and child growth in the first two years.