Prenatal Exposure Of Per-and Poly-Fluoroalkyl Substances And Its Effects On Birth Outcomes

Part Ⅰ Exposure Levels and Trend of PFAS in Different Time Points of PregnancyBackground: Pregnant women and foetuses are sensitive for per-and poly-fluorokyl substances(PFAS)exposure.Recently,studies have theoretically speculated the changes of perfluorooctanoic acid(PFOA)and perfluorooctane sulfonate(PFOS)during pregnancy,but few studies have tracked the actual changes of concentration during pregnancy.At the same time,there is a lack of research on the exposure of other PFAS and their change during pregnancy.Objectives: To describe the concentrations and trends of 10 PFAS during pregnancy;to establish the model of concentration change.Methods: In this study,255 pregnant women who successfully got pregnant and followed till delivery were selected from the pre-pregnancy birth cohort in Shanghai Birth Cohort(SBC).Their sociodemographic information was collected through faceto-face questionnaires before pregnancy,and the plasma concentrations of 10 PFAS were detected by high-performance liquid chromatography/tandem mass spectrometry(HPLC-MS /MS)in first,second and third trimester of pregnancy and cord blood.Trend test was used to analyze whether the concentration of PFAS had a trend during pregnancy.Multiple linear regression was used to establish the pattern of PFAS concentration change.Age,BMI,smoking and alcohol drinking,education level,family income,gestational age and fetal gender were considered aspotential confounders and controlled for.Results: Except for the low detection rate of perfluorooctane sulfonamide(PFOSA),which was excluded in the followed study,the detection rate of the other 9 PFAS in first,second and third trimester of pregnancy and cord blood was more than 80%.PFAS exposure was dominated by PFOA and PFOS.In early,middle and late pregnancy,the median concentration of PFOA was 15.20,13.84,11.22,7.02ng/m L,respectively;the median concentration of PFOS was 9.70,9.62,and 7.80,3.17ng/m L,respectively.The remaining PFAS,ordered in order of early pregnancy concentration,were perfluorononanoic acid(PFNA)(2.33 ng/m L),perfluorodecanoic acid(PFDA)(1.73 ng/m L),perfluoroundecanoic acid(PFUA)(1.43 ng/m L),perfluorohexanesulfonic acid PFHx S(0.64 ng/m L),pfdodecanoic acid(0.36 ng/m L),perfluoroheptanoic acid(PFHp A)(0.17 ng/m L),perfluorooctane sulfonamide,and perfluorobutane sulfonic acid(PFBS)(0.04 ng/m L).The concentrations of PFBS during pregnancy were 0.04,0.07,and 0.15 ng/m L,respectively,showing an upward trend,with a p value less than 0.05.The concentrations of PFHx S during pregnancy were 0.64,0.67 and 0.67 ng/m L,respectively.There was no change trend,and the p value was greater than 0.05 for PFHx S.The other 7 PFAS showed a downward trend with p values less than 0.05.The concentration change model has no practical significance.Conclusion: Pregnant women in Shanghai are exposed to a variety of PFAS,in which PFOA and PFOS were the main PFAS,and other PFAS such as PFNA,PFDA and PFUA also had high exposure concentrations.Most PFAS decreased during pregnancy,but PFBS increased and PFHx S showed no significant change.Part Ⅱ Correlations between concentrations of PFAS in Mixed ExposureBackground:High correlation between PFOA and PFOS concentrations has been reported in many studies.However,the correlation of other PFAS has received little attention.People exposed to PFAS were also exposed to a variety of other environmental endocrine disruptors(EEDs),and the correlations between different types of EEDs were under studied.Objectives: To evaluate the correlations between PFAS concentrations and between PFAS and other EEDs concentratins and to provide theoretical basis for the evaluation of mixed effects.Methods: This part was conducted in two populations.1.In the pregnant women in Shanghai,the correlation between PFAS was analyzed by Pearson linear correlation method.The original concentration of PFAS in different trimesters and the ln-transformed concentration were analyzed,respectively.2.EEDs concentration of the U.S.population was obtained in the NHANES study.The correlation between two EEDs was analyzed by Pearson linear correlation method.The correlation results were filled into the heat map,and the profile of correlation between different types of EEDs was reported.Results: Whether ln-transformation was performed or not,apart from PFBS,the other 8 PFAS concentrations were highly correlated in Shanghai pregnant women.The highest correlation coefficient was 0.96 between PFDA and PFUA in the third trimester.PFBS was weakly correlated with other PFAS.There were no significant changes in the correlation pattern between PFAS across different trimesters.PFAS with a strong correlation in the first trimester were still strongly correlated in the second and third.In the U.S.population,PFAS were also found to be correlated,having a maximum coefficient of 0.92 between the linear PFOS and the branched PFOS.At the same time,PFAS were also found to be associated with mercury compounds in the NHANES data,with maximum coefficient 0.56.Conclusion: There are a correlations between PFAS concentrations and between PFAS and mercury compounds concentrations.Similar structures or common sources of exposure might be responsible for their correlation.When examiningthe influence of a certain EED on a specific outcome,relevant EEDs should be controlled for in the analysis.Part Ⅲ Effect of Mixed Exposure to PFAS on Fetal Size in Multiple Time Points during PregnancyBackground: A number of animal studies have found effects of PFAS exposure during pregnancy on fetal growth and development,but conclusions from human studies were inconsistent.Meta-analysis suggested that PFAS had different effects in different trimesters during pregnancy.At the same time,previous studies have found that human were exposure to multiple PFAS,but there was a lack of researches on mixed effectObjectives: To find the sensitive trimester during pregnancy when PFAS affects fetal growth and development;and to assess the difference between mixed and individual exposure effects.Methods: In the first part of the study,we obtained the socio-demographic information,birth outcomes and PFAS exposure of pregnant women in the pre-pregnancy birth cohort of Shanghai.We further identified the participants in the Part I,including: 1)single birth without birth defects;2)no medical history of diabetes,hypertension or other diseases affecting fetal growth and development;3)No pregnancy complications such as gestational diabetes mellitus or pregnancy-induced hypertension.Finally,214 pregnant women and their child were included in the study.Generalized linear model was used to analyze the relationship between exposure to single PFAS and birth outcomes at the first,second,and third trimesters,respectively.Generalized estimating equations were used to analyze the relationship between multiple time points exposure of single PFAS and birth outcomes.By using principal component analysis and the sum of the PFAS concentrations,the difference between mixed and single exposures was evaluated.Outcome: In single-point,single-exposure analysis,exposure to PFAS in the first trimester was associated with decreased birth length.Per ln-PFOS increased reduced birth length-0.27 cm(95%CI,-0.51,-0.02),while PFNA-0.34 cm(-0.65,-0.03),PFDA-0.27 cm(-0.53,-0.01),PFUA-0.29 cm(-0.58,0.01),and PFDo A-0.54 cm(-0.00,-0.08).It was not obversed of PFAS effect on birth weight and birth head circumference in the first trimester.No effect of PFAS on birth weight,birth length and birth head circumference was found in the analysis of the second and third trimester and umbilical cord blood.Stratified analysis suggested that the effect of PFAS exposure in early pregnancy on birth length was mainly in pregnant women with normal prepregnancy BMI or female fetal sex.In the multi-point,single-exposure analysis,we also found that PFAS exposure in early pregnancy would affect fetal birth length,and the results were consistent with the single-time point analysis.So it did after stratification.In addition,it was found that PFOA,PFOS,PFHx S,PFNA,PFDA and PFUA exposure would increase the birth weight in the BMI<18.5 group after stratification.And in pregnant women with BMI>24,PFHx S exposure during pregnancy resulted in reduced birth weight.After principal component analysis(PCA),the mixed effects of PFAS were evaluated,and it was found that some of the results were contrary to the results of multipoint analysis.Evaluation of ∑ PFAS found that the results were similar with the multipoint analysis results.Conclusion: PFAS was negatively associated with the birth length and its sensitive period might be the first trimester.The effect of PFAS on birth weight varied with BMI before pregnancy,which may be related to the reproductive toxicity of PFAS.The effects of PFAS on birth weight were different with different pre-pregnancy BMI,which was related to the reproductive toxicity of PFAS.In the presence of highly correlated PFAS,the sum of PFAS concentrations was appropriate for assessing mixed effects.