| ObjectivesTo profile internal exposure and characteristics of 41 antibiotics and their two antibiotic metabolites during pregnancy,to assess the potential health risks of pregnant exposed to veterinary antibiotics(VAs)and preferred as veterinary antibiotics(PVAs),to identify predictors of urinary antibiotic exposure throughout the entire pregnancy.To investigate the associations of maternal low-doses antibiotics exposure during pregnancy and fetal growth and development and to further explore the mediating effects of maternal thyroid function in those abovementioned associations.MethodsBased on the Ma'anshan Birth Cohort,demographic characteristics and medical information were obtained during three trimesters,while maternal urine and serum samples were collected at same time points.Fetal serum samples were collected at delivery.Maternal antibiotics internal exposure characteristics were determined by liquid chromatography-triple quadrupole tandem mass spectrometry,the birth weight and length of the neonates were obtained through obstetric medical records,the ponderal index(PI)was calculated as birth weight divided by birth height cubed.The levels of four types of thyroid hormones in maternal or cord serum were determined by the electrochemiluminescence immunoassay.A total of 3 235 women who provided at least one urine sample during pregnancy were included.The detection frequencies and selected percentiles were employed to characterize maternal antibiotic internal-exposure levels and to evaluate the co-exposure pattern of antibiotics with different usages.Wilcoxon signed-rank tests were performed to investigate the difference of antibiotic concentration in different trimesters,Spearman's rank correlation coefficients were used to describes the correlation between different antibiotics during entire pregnancy and the same antibiotic during different trimesters,intraclass correlation coefficient(ICC)and their95%confidence intervals were calculated to evaluated the temporal variability of antibiotics over three trimesters.Hazard index(HI)were used to reflect the cumulative potential health risks of antibiotics exposure,and linear mixed models(LMMs)were used to analyze the associations between antibiotics exposure levels during pregnancy and demographic characteristics or urines sampling seasons.A total of 3 180 mother–neonate pairs who provided information on antibiotic exposure during pregnancy and fetal growth and development were included.Fetal growth and development indicators were described by mean±standard deviation,t test and?~2 test were performed to compare the differences of growth and development between fetuses of different genders and their parents'demographic sociological characteristics as well.LMMs were employed to analyze the association between repeated measurements of antibiotic concentration during pregnancy and fetal growth and development,multiple linear regression models were used to analyze the relationship between antibiotic exposure and fetal growth and development during different pregnancy,restrict cubic spline models to explore the non-linear associations of the selected antibiotics concentrations and outcome measurements.To explore the gender difference,the gender-stratified model was performed.After excluding 66 pregnant women with history of thyroid disease and 2 mother–neonate pairs who did not collect serum samples,a total of 3 112 mother–neonate pairs were finally included.LMMs were used to analyze the associations between repeated measurements of maternal thyroid hormone levels and fetal growth and development,multiple linear regression models were performed to the relationship between thyroid hormone levels in cord serum and fetal growth and development.Generalized estimating equation(GEE)was used to analyze the association between repeated measurements maternal antibiotic concentrations and repeated measurements of maternal thyroid hormone levels,and LMMs were used to analyze the relationship between repeated measurements of maternal antibiotic concentrations and thyroid hormone levels in cord serum.The mediating effect of maternal thyroid function on antibiotics exposure and neonatal birth weight were also examined.All analyses were conducted using SPSS version 23.0 for Windows(SPSS Inc.,Chicago,IL,USA)and R statistical software version 4.0.3 for Windows(R–4.0.3,R Core Team),all figures were made with Graph Pad Prism version 8.0 for Windows(Graph Pad Inc,Cfaliornia,USA)and R statistical software.ResultsThe total detection frequency of antibiotics in the urine of 3 235 pregnant women in Ma'anshan was 93.6%,with the highest detection rate of PVAs(93.6%),followed by VAs(61.8%),human antibiotics(HAs,25.0%)and preferred as human antibiotics(PHAs,13.7%).The total detection rate of urinary antibiotics in the first,second and third trimester was 97.1%,92.9%and 90.7%,respectively.The maternal urinary antibiotics concentrations were mainly ranged from 0 to 20 ng/m L,moderate or high temporal variability(ICC=0.03–0.63),weak or moderate correlations(r=0.02–0.75),and some antibiotics levels were posed statistically difference during three trimesters.The proportion of pregnant women co-exposed to all three usages antibiotic(VAs,PVAs and HAs/PHAs)in the first,second and third trimester were 31.0%,16.7%and13.4%,respectively.And the proportion of pregnant women co-exposed to two or more usage antibiotics were 78.2%,63.4%,and 59.6%during the first,second,and third trimester,respectively.Over the three trimesters,63.1%and 29.1%of pregnant women were consistently exposed to PVAs and VAs,and only 7.9%of pregnant women were consistently exposed to HAs/PHAs.The proportion of overall maternal urine samples with HI values>1 was 4.5%,and the rates of HI value>1 in the first,second trimester and third trimester were 3.6%,5.1%and 4.7%,respectively.The results of LMMs showed that there were statistical associations between maternal demographic characteristics or urines sampling seasons and urinary antibiotics concentrations.Neonatal birth weight,birth length and PI were 3365.96±447.21 g,50.03±1.79cm and 26.85±2.95 kg/m~3,respectively.LMMs showed that maternal low-doses sulfamethazine exposure was associated with increased birth weight.In the trimester-specific analysis,sulfamethazine and enrofloxacin were positively associated with birth weight,sulfamethazine was positively associated with PI during the first trimester.In the second trimester,other antibiotics,?-lactams,HAs/PHAs and all antibiotics were negatively associated with birth weight and birth length,and fluoroquinolones was negatively associated with birth weight.In the third trimester,florfenicol was positively associated with birth weight,fluoroquinolones was negatively associated with birth length,and enrofloxacin,ciprofloxacin,sulfonamides,fluoroquinolones were positively associated with PI.Restricted cubic spline model did not found any non-linear associations between antibiotics and outcomes indicators.In the gender-stratified models,maternal low-doses antibiotics exposure during entire pregnancy or each single trimester were associated with fetal growth and development indicators of different genders,and among them,the numbers of statistically significant associations in female neonate were higher than that in male neonate during second trimester.Additionally,abovementioned associations weren't change in sensitivity analysis.LMMs showed that maternal free thyroxine level(FT4)was negatively associated with birth weight.Multiple linear regression models showed that total thyroxine(TT4)and FT4 level in cord serum were positively associated with birth weight,birth length and PI.GEE showed that maternal urinary low-doses antibiotics levels were positively associated with maternal total triiodothyronine(TT3)and thyroid stimulating hormone(TSH),while negatively associated with maternal TT4 and FT4 level.LMMs show that there is no statistical associations between maternal antibiotics exposure and thyroid hormones level in cord serum.The maternal FT4 level played a positive mediating role in the associations between some antibiotics exposure and birth weight,and the mediating effect accounts for the proportion of the total effect in the range of 10.12%to 61.93%.Additionally,maternal FT4 levels play a negative mediating role in the association of other antibiotics exposure and birth weight,and the mediating effect accounts for the proportion of the total effect in the range of 10.25%to 53.59%.Conclusions1.Pregnant women are generally exposed to multiple low-dose antibiotics during pregnancy,which show moderate or high temporal variability,weak or moderate correlations,and pose statistically difference during three trimesters.Most pregnant women are continuously exposed to low-dose VAs and PVAs across pregnancy.More than half of pregnant women are co-exposed to two or more different usage antibiotics during each single trimester,while co-exposed to all three usage antibiotics is rare.Some pregnant women pose potential health risks due to multiple antibiotics exposure,and those antibiotics are affected by demographic characteristics and urines sampling seasons.2.Maternal low-doses antibiotics exposure(mainly VAs and PVAs)during pregnancy can affect the growth and development of the fetus.Antibiotics exposure in the first and third trimesters are mainly promotes fetal growth,while in the second trimester are inhibits fetal development.And female neonates are more sensitive to antibiotic exposure during the second trimester.3.Low-doses antibiotics exposure during pregnancy can affects maternal thyroid function and increases the TT3 and TSH level,decreases TT4 and FT4 level.Maternal low-doses antibiotics exposure may affect fetal growth and development by decreasing maternal serum FT4 level. |
PDF Full Text Request