| Objective:The purpose of this study is to explore the association between maternal exposure to ambient air pollutants(PM2.5,PM10,SO2,O3,NO2,CO)and the risk of preterm birth（PTB）and low birth weight（LBW）,and to estimate the sensitive time window,lag effect and cumulative effect of exposure.Methods:The data o f this study came from a large retrospective birth cohort in Chongqing from 2015 to 2020.A total of 598,018 mother-infant pairs were included in the analysis data sets according to different inclusion and exclusion criteria.The air pollutant data in Chongqing was obtained from the China Environmental Monitoring Station of the Ministry of Ecology.The daily concentrations of air pollutants for nearest monitoring stations（≤10 km）were used to estimate the exposure of each mother based on where the mother lived during pregnancy.The influence effects,including long-term effect,short-term effect,cumulative effect and lag effect,were analyzed by using Generalized Additive Model（GAM）,Distributed Lag Non-linear Model（DLNM）,DLNM nested Cox regression model,and attribution analysis,etc.The RR values and 95%CI for preterm birth,very preterm birth（VPTB）,low birth weight,and very low birth weight（VLBW）were measured per 10μg/m3（CO was 100μg/m3）increase in exposure to air pollutants.In this study,the single pollutant model adjusted covariates including mean temperature and mean humidity,parental age,birth weight,or gestational age;the multi-pollutant model adjusted the same covariates,and additionally adjusted other air pollutants.Results:（1）Results of association between ambient air pollutant exposure and PTB.1)The results showed that every 10μg/m3 increase for the exposure of PM2.5 and PM10had a statistically significant effect on PTB during the entire pregnancy and the third trimester of pregnancy.The RR values of PM2.5exposure in the third trimester for the single pollutant model and multi-pollutant model were 1.036（95%CI:1.021～1.050）and 1.028（95%CI:1.010～1.047）,respectively.The RR values of PM2.5 exposure in the entire trimester for the single pollutant model and multi-pollutant model were 1.101（95%CI:1.075～1.128）and 1.186（95%CI:1.147～1.226）,respectively.The results of PM10 were similar to those from PM2.5.For every 100μg/m3 increase in CO exposure during pregnancy,the relative risk of PTB in the first,second and third trimesters increased,and the RR values were statistically significant.The effect of CO exposure to PTB during the entire pregnancy was statistically significant in the single pollutant model（RR=1.009,95%CI:1.001～1.018）.For SO2,O3,and NO2,no statistically significant RR values were found between pregnancy exposure and PTB during the first,second,and third trimesters.2)There was a significant seasonal difference between ambient air pollutant exposure and PTB.Preterm births were more common in winter and less frequently in spring and summer.The effects of PM2.5 and PM10 exposure throughout the entire pregnancy and third trimester on preterm birth were statistically significant when conception occurred in summer and autumn.When conception time was in different seasons,every 100μg/m3increase in CO exposure in the third trimester and entire pregnancy was associated with a statistically significant increase in the risk of preterm birth.3)For every 10μg/m3 increase of PM2.5,PM10,and NO2,the exposure during the entire pregnancy and the first trimester had a statistically significant effect on VPTB.4)Daily exposure to ambient air pollutants PM2.5,PM10,SO2,NO2,and CO increased the risk of preterm birth from a lag effect.The RR values of PM2.5 exposure on lag0-lag3and lag10-lag21 days were statistically significant,the highest on the first day,and then decreased.The 7-day cumulative effect of PM2.5 was statistically significant when the concentration exceeded 100μg/m3,while the 30-day cumulative effect was statistically significant when the concentration exceeded 50μg/m3.The lag effect of PM10 on preterm birth was highly similar to that of PM2.5.The lag effect and cumulative effect of CO exposure were the strongest,and the impact on the first day was the strongest（RR=1.044,95%CI:1.018～1.069）.The cumulative effect of CO exposure showed that the RR values increased rapidly after 1000μg/m3.The HR values of PM2.5 exposure were statistically significant when the lag time was 0～4 weeks,9～22 weeks,and 28～36 weeks.The results of PM10were similar.The HR values of CO exposure were statistically significant on the lag of 0～3 weeks and 34～36 weeks.（2）Results of association between exposure to ambient air pollutants and LBW.1)The analysis results showed that the PM2.5 and PM10exposure had no statistically significant effect on LBW for every 10μg/m3 increase in exposure concentration.For every100μg/m3 increase in CO exposure in the third trimester,the RR value of LBW increased significantly.The RR value of CO exposure in the single pollutant model was statistically significant（RR=1.011,95%CI:1.003～1.018）.The RR value of the multi-pollutant model of the relationship between O3 exposure and LBW in the second trimester was statistically significant（RR=1.021,95%CI:1.004～1.038）.2)According to the season stratification of conception date,for every 100μg/m3increase in CO,the risk of LBW on the date of conception increased in spring and summer.The RR values for the single pollution model were 1.033（95%CI:1.012～1.055）and 1.033（95%CI:1.009～1.056）,respectively.3)For every 10μg/m3 increase in the concentration of PM2.5 and PM10,the exposure in the first trimester had a statistically significant effect on VLBW.For PM2.5 in early pregnancy,the RR value of the single pollutant model was 1.100（95%CI:1.012～1.195）.For every 10μg/m3 increase in PM10 exposure in the first trimester,the RR value of the single pollutant model was 1.129（95%CI:1.055～1.209）,and that of the multi-pollutant model was1.115（95%CI:1.024～1.213）.For NO2,first trimester and second trimester exposures were found to have statistically significant RR values for VLBW.For O3,a statistically significant RR values for VLBW were found during the entire pregnancy and the second trimester.4)The RR values of LBW in the daily lag effect and the cumulative effect of 0-30 days were not statistically significant for PM2.5 and PM10.Both NO2 and SO2 had different degrees of lag and accumulation effects.The lag effect of CO was the strongest with both lag0-lag4 and lag9-lag21 being statistically significant.The cumulative effect of CO exposure for 30 days showed that the RR values increased significantly after the concentration exceeded 1000μg/m3.For the lag effect analysis of LBW on gestational age,the results were highly consistent with PTB.The RR values of PM2.5 exposure were statistically significant when the lag time was 0～5 weeks and 25～35 weeks.The HR values were statistically significant for PM10 exposure on the lag of 0～5 weeks and 22～35weeks.The HR value of CO exposure was statistically significant on the lag of 0～1 week.Conclusions:The results indicate that maternal exposure to ambient air pollutants during pregnancy increases the risk of PTB and LBW.Exposure of pregnant women to high levels of PM2.5 and PM10 throughout pregnancy may increase the risk of preterm birth,especially in the third trimester.CO exposure in first,second,third trimester,and the entire pregnancy is positively correlated with the occurrence of preterm birth;CO exposure in third trimester increases the risk of LBW.In particular,exposure to PM2.5,PM10,and NO2increase the risk of VPTB and VLBW.The risk of PTB and LBW caused by exposure to ambient air pollutants was associated with conception season and had lag effect and cumulative effect.The cumulative effect increased with increasing exposure concentrations.O3 exposure may not be associated with PTB.Therefore,it is necessary for pregnant women to reduce exposure to high levels of ambient air pollutants.Pregnant women should be more aware of the potential health hazards of ambient air pollution,pay more attention to local air quality,and take appropriate protective measures to reduce the risk of air pollution. |
PDF Full Text Request