Associations Between Biomarkers Of Late Pregnancy Exposure To Drinking Water Disinfection By-products And Reproductive Outcomes

Reproductive outcomes are associated with many risk factors including maternal health, nutrition during pregnancy genetic background and exposure to environmental factors, which play an important role in the process of pregnancy outcomes. Exposure to environmental risk factors during pregnancy may result in adverse reproductive outcomes, such as preterm delivery, pregnancy loss, intrauterine growth retardation (IUGR), low birth weight (LBW) and birth defects. Chemical disinfectants can react with natural organic pollutant during the process of disinfect drinking water to form disinfection by-products (DBPs), an important environmental pollutant. Widespread exposure to DBPs can result from ingestion, inhalation and dermal absorption during routine water-use activities. Some kinds of DBPs have showed reproductive developmental toxicity. And exposure to DBPs can produce preterm birth, spontaneous abortion, decreased survival rate, reduction in body weight and birth defects. A large number of epidemiological studies have also examined the relationship between DBPs exposure and adverse reproductive outcomes, however, the results are inconclusive. Most of them used DBPs levels in public water supplies as a surrogate of exposure; however few studies took advantage of exposure biomarkers to conduct exposure risk assessment of DBPs.We conducted a cohort study to investigate the effects of exposure to DBPs on reproductive outcomes. We used blood trihalomethanes (THMs) and urinary trichloroacetic acid (TCAA) during late pregnancy as exposure biomarkers to assess the internal dose of exposure to DBPs, and used fetal growth measures as the index of reproductive outcomes to analyze the relationships between exposure to DBPs and reproductive outcomes. Based on the above study, we evaluated the feasibility of exposure biomarkers of DBPs in the epidemiological studies and their influencing factors. Our study could provide basic research data about female reproductive toxicity of DBPs.I.Associations between blood THM level during late pregnancy and reproductive outcomesObjective:To investigate the associations between blood THM concentrations as biomarkers of late pregnancy exposure to THMs in drinking water and reproductive outcomes.Methods:We conducted a prospective study to recruit1184mothers who labored during July2011to June2012in Wuhan, Hubei and during October2012to December2013in Xiaogan, Hubei. We measured the four THM concentrations [chloroform (TCM), bromodichloromethane (BDCM), dibromochloromethane (DBCM) and bromoform (TBM)] in blood by using headspace solid-phase micro-extraction gas chromatography (SPME-GC). The concentrations of blood TCM、Br-THMs (sum of BDCM、DBCM and TBM concentrations) and total THMs (TTHMs; sum of Br-THMs and TCM concentrations) were divided into tertiles. The concentrations of blood BDCM、DBCM and TBM were divided into a three level-ordinal variable according to the limit of detection (LOD) and median of detectable concentrations. Multivariable linear and logistic regression models were used to assess the associations between the biomarkers of late pregnancy exposure to THMs and reproductive outcomes. The indices of reproductive outcomes included birth weight, birth length, birth body mass index (BMI), gestational age, small for gestational age (SGA) and weight for length. According to the criteria of Chinese children health examination for birth BMI and weight for length, the subjects were dichotomized as below or at/above the median level.Results:TCM、BDCM、DBCM and TBM were detected in92.5%、57.4%.33.5%and22.6%. The median concentration (range) of TCM, Br-THMs and TTHMs was50.7ng/L [<LOD-480.3ng/L],5.6ng/L (<LOD-40.2ng/L) and57.7ng/L (<LOD-486.7), respectively. Of1184single live newborns,60neonates (5.1%) were SGA, birth BMI of the247neonates (20.9%) were below the median level, and weight for length of the240neonates (20.3%) were below the median level.Multivariable linear regression analysis have showed significant dose-response relationships between elevated TTHMs and reduced birth weight (P for trend=0.03), between elevated levels of BDCM and DBCM and decreased birth length (P for trend=0.04, P for trend=0.02), between elevated levels of TCM and TTHMs and decreased BMI (P for trend=0.03, P for trend=0.01). Multivariable logistic regression analysis have also found an increased risk of SGA for the second and third tertiles of TTHMs (OR=2.91,95%CI,1.32to6.42; OR=2.25,95%CI,1.01to5.03); however, the dose-response relationship was only suggestive(P for trend=0.08).Conclusions:We found that elevated maternal blood THM concentrations were associated with decreased birth weight, reduced birth length and increased risk of SGA, suggesting that elevated maternal THM exposure during late pregnancy may adversely affect reproductive outcomes. Ⅱ. Associations between the urinary TCAA level of late pregnancy and reproductive outcomesObjective:To investigate the associations between the urinary TCAA level as a biomarker of late pregnancy exposure to TCAA in drinking water and reproductive outcomes.Methods:We conducted a prospective study to recruit1306mothers who labored during July2011to June2012in Wuhan, Hubei, and during October2012to December2013in Xiaogan, Hubei. We measured TCAA level in urine by using liquid-liquid extraction-gas chromatography (LLE-GC). The urinary TCAA concentrations were divided into quartile. Multivariable linear and logistic regression models were used to examine the associations between the urinary TCAA concentrations of late pregnancy and reproductive outcomes. The indices of reproductive outcomes included birth weight, birth length, BMI, gestational age, SGA and weight for length. According to the criteria of Chinese children health examination for birth BMI and weight for length, the subjects were dichotomized as below or at/above median level.Results:The TCAA concentrations in urine were detected in97.9%, ranged from <LOD to82.5μg/L (median=7.3μg/L, mean=8.7μg/L). Of1306single live newborns,68neonates (5.2%) were SGA, birth BMI of the307neonates (23.5%) were below the median level, and weight for length of the256neonates (19.6%) were below the median level.Multivariable linear regression analysis have showed that newborn birth weight (P=-96.8g,95%CI:-158.23to-35.36) and birth BMI (β=-0.35kg/m2,95%CI:-0.57to-0.13) decreased at the highest quartile of urinary TCAA levels (>10.0μg/L), compared to the first quartile of urinary TCAA levels (<4.8μg/L). Multivariable logistic regression analysis has showed an increased risk of having below-median level of BMI (OR=1.83,95%CI:1.22to2.73) and below-median level of weight for length (OR=1.90,95%CI:1.23to2.94) at the fourth quartile of TCAA (>10.0μg/L), compared to the first quartile of urinary TCAA levels (<4.8μg/L).Conclusions:We found that elevated maternal urinary TCAA concentrations were associated with decreased birth weight and reduced BMI, suggesting that elevated maternal TCAA exposure during late pregnancy may adversely affect reproductive outcomes.III.The factors influencing internal dose of DBPs during late pregnancyObjectives:To identify the factors influencing the internal dose of DBPs during late pregnancy, and provide information for exposure assessment of DBPs in drinking waterMethods:A total of892blood samples and urine samples were both collected from the mothers who participated in a cohort study of exposure to DBPs and reproductive outcomes during2011to2013in Wuhan and Xiaogan, Hubei. General linear regressions were used to evaluate the factors influencing the internal dose of DBPs during late pregnancy. Among892mothers,354mothers who lived in the area with the same water supply system during pregnancy were selected. The concentrations of THMs and TCAA in tap water in this area were monitored. General linear regressions were used to evaluate the relationships of internal dose of DBPs with concentrations of DBPs in tap water, and other influencing factors.Results:(1) The concentrations of blood TCM was negatively associated with Wuhan city (β=-0.22, P<0.01), winter sampling (β=-0.30, P<0.01), prenatal BMI (β=-0.15, P<0.05), boiled drinking water (β=-0.15, P<0.05) and household income (>5000yuan)(P=-0.20, P<0.01).(2) The concentrations of blood Br-THMs (sum of BDCM, DBCM and TBM) was negatively correlated with Xiaogan city (P=-0.08, P<0.01), autumn sampling (β=-0.11, P<0.01) and diabetes (β=-0.10, P<0.05).(3) The concentrations of blood TTHMs (sum of Br-THMs and TBM) was negatively associated with Wuhan city (β=-0.11, P<0.01), winter sampling (β=-0.17,P<0.01), boiled drinking water (β=-0.12, P<0.05) and household income (≥5000yuan)(β=-0.14, P<0.05).(4) The TCAA concentrations in urine was negatively associated with winter sampling (β=-0.30, P<0.01), whereas positively correlated with TCAA levels in tap water (β=0.49, P<0.01).Conclusion:Our results suggest that study city, season for biological sample, DBPs concentrations in tap water, boiled drinking water, prenatal BMI, diabetes, and household income may affect the internal dose of DBPs during late pregnancy. These factors should be considered in the further epidemiology studies of DBPs exposure risk assessment among pregnant women.