| BackgroundBirth weight,as a sensitive indicator of neonatal health outcome,reflects the mother’s intrauterine environment,fetal intrauterine growth and nutritional status.It is an important factor affecting the normal development of childhood and adolescence and the health status of adults.There are many factors affecting birth weight,such as parental factors.Among them,the impact of maternal exposure to atmospheric particulate matter during pregnancy on neonatal birth weight is receiving extensive attention.However,the results of studies on the relationship between maternal exposure to atmospheric particulate matter during pregnancy and neonatal birth weight are inconsistent.On the one hand,it may be related to the influence of unmeasured unknown confounding factors（researchers only adjusted the known confounding factors to pay attention to the relationship between the two,and did not consider the influence of unknown confounding factors on the research results）;On the other hand,it may be related to the birth weight（dependent variable）of all newborns including low birth weight and macrosomia,and the range of dependent variables is too broad.In addition,the key window period and nature of maternal particulate matter exposure on neonatal birth weight are not clear.Negative control exposures（NCEs）analysis is a method to evaluate whether there are unknown confounding factors in association analysis and determine their causal effects.It has been applied to observational epidemiological studies.Therefore,this study introduced NCEs analysis to analyze the causal relationship between maternal PM10 and PM2.5 exposure during pregnancy and full-term normal birth weight based on the baseline study of neonatal birth cohort in Jinan and single full-term newborns with normal birth weight,to determine the key window period of exposure and its nature.ObjectivesBased on the baseline study of neonatal birth cohort in Jinan,this study takes single full-term newborns with normal birth weight as the research object,analyzes the causal relationship between maternal PM10 and PM2.5 exposure during pregnancy and fullterm normal birth weight based on GLM and NCEs analysis,and discusses the key window period and nature of exposure,so as to improve people’s attention to air pollution.Materials and methods1.Research design This study is based on the baseline study of neonatal birth cohort in Jinan.2.Research subjects The subjects of this study were full-term singleton newborns with normal birth weight.3.Research methods（1）Dependent variables and covariates:①dependent variables:the dependent variables of this study are the birth weight of full-term normal birth weight newborns;②information collection methods:The relevant data of the subjects and their mothers came from the face-to-face interview survey conducted by the investigators.（2）Evaluation of independent variables:the atmospheric particulate matter data comes from the dynamic monitoring data of 17 monitoring stations in Jinan City;The inverse distance weighting（IDW）method was used to evaluate the daily exposure concentration of each mother during pregnancy（including first trimester,second trimester and third trimester）and the exposure period of negative control;（3）Negative control exposure（NCEs）analysis:In this study,the negative control exposure period of each subject was designed as the corresponding pregnancy period one year after the mother’s current pregnancy;The daily PM10 and PM2.5 exposure concentrations of each subject’s mother during the negative control exposure period were NCE variables.（4）Fitting of generalized linear model（GLM）:GLM fitting method was used to establish a model to study the linear relationship between air pollution exposure and full-term normal birth weight.（5）Generalized propensity score:the generalized propensity score is used to adjust the known confounding factors,and the propensity score after adjusting the confounding factors is included in the linear model for regression analysis to evaluate the impact of model setting errors on the research results.4.Sensitivity analysis Firstly,the effect of selection bias on the relationship between maternal PM10 and PM2.5 exposure and full-term normal birth weight was evaluated.Secondly,evaluate the impact of model setting errors on the research results.5.Statistical analysis All analyses were performed using R version 3.6.3（The R Project for Statistical Computing,Vienna,Austria;http://www.r-project.org/）.Numerical variables satisfying normal distribution are expressed as mean ± standard deviation（x±s）,and classified variables are expressed as frequency and percentage.The quantitative data were tested by Kolmogorov smienov test.If the data obeyed the normal distribution,the differences of birth body mass among different subgroups were compared by student’s t-test;If the data do not obey the normal distribution,the Wilcoxon rank sum test is used to compare the differences of birth weight between groups.Analysis of variance or Kruskal Wallis test were used to compare the exposure concentrations of PM10 and PM2.5 in the first,second and third trimester of pregnancy,as well as the number of people exceeding the standard and the number of days exceeding the standard in different pregnancy periods.GLM combined with NCEs analysis was used to study the relationship between air pollution exposure and full-term normal birth weight.Inspection standard α=0.05.Results1,Basic information of research subjects:the 4114 full-term newborns with normal birth weight were included in this study.Among them,2094 boys（50.90%）and 2317（56.32%）were born naturally.The average gestational age of the subjects was 275.58 days（39 weeks）,and the average birth weight was 3363.98g（s=331.21g,2510g～4000g）.The birth weight of boys,newborns born by cesarean section and newborns born at older gestational age was significantly higher（P<0.05）.2.The mothers of the subjects:the average age of the mothers of the subjects was 30.58 years（s=4.42 years,18-50 years）,831（20.20%）were housewives,and 97.16%were Han mothers.The birth weight of newborns was higher when the mother was older（≥30 years old）and the mother was Han nationality（P<0.05）.When the family population is small,the birth weight of newborns is small（P=0.035）.During this pregnancy,2908（70.69%）took exercise during pregnancy,108（2.63%）suffered from pregnancy induced hypertension,655（13.92%）smoked passively during pregnancy,and 2570（62.47%）insisted on working during pregnancy.The birth weight of newborns delivered by mothers with pregnancy induced hyperten sion was significantly lower（P=0.008）.3.Exposure to particulate matter of the study mother:the average individual exposure concentrations of PM10 and PM2.5 of the study mother during the whole pregnancy were 114.15 and 54.71μg/m3,respectively.The average PM10 exposure concentrations in the first,second and third trimesters were 102.18,115.80 and 124.97μg/m3,respectively,the average PM2.5 exposure concentrations were 46.65,56.48 and 61.26μg/m3,respectively.The average exposure concentrations of PM10 and PM2.5 in the negative control exposure period were 101.02μg/m3 and 51.83μg/m3,respectively.4.Effects of PM10 and PM2.5 on full-term normal birth weight:①After preliminary control of known confounding factors,the results show that,for every 10μg/m3 increase in PM10 concentration durin the whole pregnancy,the birth weight of the subjects decreased by 17.36g（P=0.012）.And for every 10μg/m3 increase in the concentration of PM2.5 during the whole pregnancy,the average birth weight of the subjects decreased by 28.92g（P=0.043）.②NCEs analysis showed that the PM10 exposure concentration of mothers in the whole pregnancy and negative control exposure period increased by 10μg/m3,the changes of neonatal birth weight were26.93g（P=0.003）and 13.76g（P=0.108）.PM2.5 every 10μg/m3,the changes of neonatal birth weight were-34.41g（P=0.047）and 8.00g（P=0.653）.There were no unknown confounding factors in this association analysis.②the results of multivariate model showed that every 10μ/m3 increase in PM10 exposure concentration in the first and third trimester of pregnancy,the birth weight of the subjects decreased by 10.59g（P=0.002）and 9.60g（P=0.001）,respectively,and there was no significant correlation between the second trimester of pregnancy and birth weight（P=0.547）.PM2.5 exposure concentration increased by 10μg/m3 in the first and third trimester of pregnancy,the birth weight of the subjects decreased by 18.43g（P=0.002）and 16.28g（P=0.001）,respectively,and there was no significant correlation between the second trimester of pregnancy and birth weight（P=0.381）.5.Sensitivity analysis:the results of evaluating the impact of selection bias and model setting on the research results show that the results of sensitivity analysis are consistent with the research results.Research conclusions1.High levels of PM10 and PM2.5 exposure during pregnancy significantly reduced full-term normal birth weight.For PM2.5,for every 10μg/m3 increase in maternal exposure concentration during the whole pregnancy and mirror period,the change of neonatal birth weight was-34.41g（P=0.047）and 8.00g（P=0.653）,respectively.2.The exposure levels of PM10 and PM2.5 during pregnancy of the mothers of the subjects far exceeded the national standards,and the reduction of full-term normal birth weight may be due to toxic effects.3.Early and late pregnancy may be the key window period of full-term normal birth weight loss caused by maternal exposure to PM10 and PM2.5 during pregnancy,and there is no significant correlation in the second trimester. |
PDF Full Text Request