| Objectives(1)To establish land use regression(LUR)models of Wuhan city for estimating the long term nitrogen dioxide(NO2)exposure level of individual.(2)To examine the relationship between prenatal NO2 exposure and intrauterine fetal growth indexes or neonatal birth growth indexes.(3)To investigate the association between NO2 exposure during pregnancy and gestational diabetes mellitus(GDM)or blood glucose level of oral glucose tolerance test(OGTT).Methods(1)The NO2 monitoring data during Jan 1,2013 to Dec 31,2015 that provided by Wuhan environmental protection bureau was collected and used as dependent variable,and the information of land use,altitude,population density and total length of road in Wuhan were collected and used as predictive variables,the multiple stepwise regression method was used to build the annual and monthly LUR models.We used leave-one-out-cross-validation(LOOCV)method to evaluate the accuracy of LUR models.(2)Individual exposure levels of NO2 during the periods of weeks 0～22,0～32 and the entire pregnancy were assessed based on home address using established monthly LUR models.The biparietal diameter(BPD),head circumference(HC),abdominal circumference(AC),femur length(FL),humerus length(HL)and estimated fetal weight(EFW)were evaluated via ultrasonography,birth weight and birth length were measured at birth by a midwife in the delivery room.We used a semi-quantitative questionnaire to obtain the information of relevant co variates such as maternal age and educational level.The multiple linear regression model was used to examine the relationship between prenatal NO2 expoure and fetal growth indexes.The mixed-effect model was used to analyze the repeated measurement data.Stratification analysis was used to explore the potential effects modification.(3)Individual exposure levels of NO2 during the period of weeks 0～22 were assessed based on home address using established monthly LUR models.The information of GDM was obtained from OGTT at the second trimester of pregnancy.We used a semi-quantitative questionnaire to obtain the information of relevant covariates such as maternal age and educational level.The multiple linear regression model was used to examine the relationship between NO2 expoure and blood glucose level,the multiple Logistic regression model was used to assess the association between NO2 expoure and GDM.Stratification analysis was used to explore the potential effects modification.Results(1)The average daily NO2 monitoring concentrations of 10 air quality monitoring stations in Wuhan were collected.The annual LUR models for 2013,2014 and 2015 were built respectively,the models showed that NO2 concentration in Wuhan was mainly related to the area of vegetation,waterbody,bareland with 2 Km buffer and the area of cropland with 5 Km buffer.The adjusted R2 of the models were ranged from 0.84 to 0.86,suggesting that they can explain most of the variation.The adjusted R2 of the LOOCV were ranged from 0.45 to 0.67,suggesting that the models may fitted well.In addition,due to the big differences among NO2 concentration of each season,the annual LUR models were adjusted for temporal variations of monthly NO2 concentrations using the monitoring data,and 36 monthly LUR models were built to reduce exposure measurement error.The adjusted R2 of the models were ranged from 0.50 to 0.93,and the adjusted R2 of the LOOCV were ranged from 0.41 to 0.69,suggesting that the monthly LUR models may fitted well also.(2)A total of 1001 pregnant women were included in the study,and their average NO2 exposure level during the entire pregnancy was 61.16 |}Lg/m3.After adjusted for maternal age,height,pre-pregnancy weight,educational level,smoking,drinking,gestational age,ethnic group,average personal income,and folic acid supplementation use,with a 10 μg/m3 increase in week 0-22 NO2 exposure,the BPD,HC,AC,FL,HL and EFW in the second trimester decreased by 0.40(95%CI:-0.56,-0.24)mm,1.04(95%CI:-1.57,-0.52)mm,1.01(95%CI:-1.55,-0.47)mm,0.24(95%CI:-0.36,-0.12)mm,0.12(95%CI:-0.23,-0.01)mm,7.69(95%CI:-11.43,-3.95)g,respectively.With a 10 μg/m3 increase in week 0～32 NO2 exposure,the BPD,HC in the third trimester decreased by 0.28(95%CI-0.52.-0.04)mm,0.76(95%CI:-1.41,-0.10)mm,respectively.All the differences were statistically significant(all P<0.05).The stratified analyses showed that the effects of NO2 on the HC,FL and EFW in the second trimester were stronger among girls,and that the effect of NO2 on EFW in the third trimester was stronger among smoking mothers,all the interactions were statistically significant(all P<0.05).The mixed-effect model showed inverse associations of NO2 exposure with BPD,HC.AC and FL(all P<0.05).(3)A total of 1050 pregnant women were included in the study,and their average NO2 exposure level during the period of weeks 0-22 was 64.29 μg/m3.After adjusted for age,pre-pregnancy body mass index(BMI),educational level,smoking,drinking,gestational age,ethnic group,average personal income,family history of diabetes and folic acid supplementation use.with a 10 μg/m3 increase in week 0～22 NO2 exposure,the fasting,1 hour and 2 hours blood glucose of OGTT increased by 0.02(95%CI:0.01,0.03)mmol/L,0.23(95%CI:0.16,0.29)mol/L,0.14(95%CI:0.09,0.19)mmol/L,respectively;the risk of GDM increased by 41%(OR:1.41,95%CI:1.19,1.69).The stratified analyses showed that the effect of NO2 on GDM was stronger among women whose age ≥30 years,and was stonger among women with family history of diabetes,all the interactions were statistically significant(all P<0.05).In addition,we observed that there was a J-type positive correlation between NO2 exposure level during pregnancy and blood glucose,and the cutoff point value was about 80 μg/m3.When the NO2 exposure level was more than 80 μg/m3,the rising trend that blood glucose level increased with the growth of NO2 exposure level during pregnancy was more rapid.Conclusions(1)The LUR models that built in this study have strong accuracy to predict the long-term NO2 exposure level in Wuhan city.(2)Maternal NO2 exposure level during pregnancy was inversely associated with fetal growth in the second and third trimester.(3)NO2 exposure level during pregnancy was associated with higher fasting,1 hour and 2 hours blood glucose level,and higher risk of GDM. |
PDF Full Text Request