Predictors of Third Trimester Blood Trihalomethanes and Urinary

Subscriber access provided by Loyola University Libraries

Article

Predictors of third trimester blood trihalomethanes and urinary trichloroacetic acid concentrations among pregnant women Qiang Zeng, Wen-Cheng Cao, Bin Zhou, Pan Yang, Yi-Xin Wang, Zhen Huang, Jin Li, and Wen-Qing Lu Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b05971 • Publication Date (Web): 20 Apr 2016 Downloaded from http://pubs.acs.org on April 20, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36

Environmental Science & Technology

1

Predictors of third trimester blood trihalomethanes and urinary trichloroacetic acid

2

concentrations among pregnant women

3

Qiang Zeng†,‡, Wen-Cheng Cao†,‡, Bin Zhou§, Pan Yang†,‡, Yi-Xin Wang†,‡, Zhen Huang†,‡,

4

Jin Li†,‡, Wen-Qing Lu†,‡*

5

†

6

Medical College, Huazhong University of Science and Technology, Wuhan, Hubei 430030,

7

PR China

8

‡

9

Environmental Protection, and State Key Laboratory of Environmental Health (incubating),

10

School of Public Health, Tongji Medical College, Huazhong University of Science and

11

Technology, Wuhan, Hubei 430030, PR China

12

§

13

*

14

Wen-Qing Lu

15

Department of Occupational and Environmental Health, School of Public Health,

16

Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei

17

Tel: +86-27-83610149; Fax: +86-27-83657765

18

E-mail: [email protected] (W.Q. Lu)

Department of Occupational and Environmental Health, School of Public Health, Tongji

Key Laboratory of Environment and Health, Ministry of Education & Ministry of

College of Public Health, University of South China, Hengyang, Hunan 421001, PR China Corresponding author:

19

1

ACS Paragon Plus Environment


20

Abstract

21

Prenatal exposure to disinfection by-products (DBPs) has been associated with a variety of

22

adverse birth outcomes. However, little is known about predictors of prenatal biomarkers of

23

exposure to DBPs among pregnant women. We aimed to identify predictors of third trimester

24

blood trihalomethanes (THMs) and urinary trichloroacetic acid (TCAA) concentrations, two

25

biomarkers of exposure to DBPs, among pregnant women. Blood samples, urine samples, and

26

questionnaires on individual characteristics and water-use activities were collected from 893

27

pregnant women in a Chinese cohort study. Maternal blood THM [chloroform (TCM),

28

bromodichloromethane (BDCM), dibromochloromethane (DBCM), and bromoform (TBM)]

29

and urinary TCAA concentrations were measured. We used multivariable linear regression to

30

identify the predictors of third trimester blood THM and creatinine-adjusted urinary TCAA

31

concentrations. The geometric mean of blood TTHM (sum of TCM, BDCM, DBCM, and

32

TBM) and creatinine-adjusted urinary TCAA concentrations were 51.90 ng/L and 9.66 µg/g

33

creatinine, respectively. Study city was the strongest significant predictors of blood THM and

34

creatinine-adjusted urinary TCAA concentrations. Prenatal body mass index (BMI) was

35

associated with decreased blood THM and decreased creatinine-adjusted urinary TCAA

36

concentrations. Age was associated with increased blood Br-THM (sum of BDCM, DBCM,

37

and TBM) concentrations. Intake of boiled water and passive smoking were associated with

38

lower blood THM concentrations. The predictors of blood THM and urinary TCAA

39

concentrations identified in this study provide potential health implications on how to reduce

40

DBP exposure during pregnancy.

41

Key words: Disinfection by-products; Predictors; Pregnant women; Trichloroacetic acid;

42

Trihalomethanes. 2


Page 2 of 36

Page 3 of 36


43

Introduction

44

Chlorine has been extensively used for disinfection of drinking water to reduce the

45

prevalence of water-borne communicable disease. However, chlorination generates hundreds

46

of disinfection by-products (DBPs) as a result of the reaction between the chlorine and

47

inorganic compounds and organic material. Trihalomethanes (THMs) and haloacetic acids

48

(HAAs) are the two most prevalent classes of DBPs found in chlorinated drinking water

49

worldwide including in China.1,

50

activities such as chlorinated water consumption, showering/bathing, and swimming.3 After

51

exposure, DBPs such as brominated THMs and HAAs are metabolized into mutagenic

52

intermediates through a conjugation with glutathione,4 resulting in potentially adverse health

53

effects. Epidemiological studies have suggested that exposure to DBPs is associated with

54

elevated risk of cancer5, 6 and adverse male reproductive effects.7-9 Prenatal exposure to DBPs

55

has also been reported to be associated with a variety of adverse birth outcomes such as

56

preterm delivery, small for gestational age, low birth weight, and fetal growth restriction.10-16

57

However, the findings remain inconclusive and inconsistent.3, 17

2

Human exposure to DBPs occurs through water-use

58

Assessment of exposure has become the main limitation in epidemiological studies of

59

DBP exposures and potential health effects.18 Compared with external surrogates of DBPs,

60

biomarkers can offer the enhanced assessment of exposure.19 Two valid biomarkers of

61

exposure to DBPs including blood THMs and urinary trichloroacetic acid (TCAA) have been

62

developed. Blood THMs can reflect an integrative exposure measure from multiple routes

63

and are generally more sensitive to low exposures.20,

64

correlated with ingestion of TCAA through chlorinated drinking water.22,

21

Urinary TCAA is significantly

3


23

Several


65

determinants of blood THM and urinary TCAA concentrations across various populations

66

have been reported including THM and TCAA concentrations in drinking water, recent

67

water-use activities (e.g., tap water consumption, showering/bathing, and swimming),

68

socio-demographic characteristics (e.g., age, smoking, household income, and educational

69

background), and genetic and physiologic characteristics (e.g., metabolic enzyme gene

70

polymorphisms).24-32

71

However, there is a paucity of data to date on the determinants of blood THM and urinary

72

TCAA concentrations among pregnant women. Pregnant women are a unique and vulnerable

73

population due to many physiological, metabolic, and behavioral changes. For example,

74

blood volume, cytochrome P450 (CYP) enzyme activity, and water consumption during

75

pregnancy increase to provide for the demands of the developing fetuses.33, 34 These changes

76

may differentially affect exposure to DBPs among pregnant women.28 Furthermore, the

77

maternal concentrations of THMs in blood and TCAA in urine during pregnancy are relevant

78

as indicator of the exposure experienced by the developing fetuses, which has been associated

79

with adverse birth outcomes.10, 11, 16 Therefore, it is essential to understand the predictors of

80

prenatal exposure to DBPs among pregnant women.

81

In the present study, we identified the predictors of third trimester blood THM and

82

urinary TCAA concentrations among pregnant women who participated in a cohort of

83

prenatal exposure to drinking water DBPs and birth outcomes in China. This information can

84

provide potential health implications on how to reduce DBP exposure during pregnancy and

85

further to reduce the risks of adverse birth outcomes.

4


Page 4 of 36

Page 5 of 36


86

Materials and Methods

87

Study population

88

The study population in this study was from pregnant women who participated in a

89

cohort of prenatal exposure to drinking water DBPs and birth outcomes in China, as has been

90

described previously.10 Briefly, a total of 1747 pregnant women who lived in Wuhan and

91

Xiaogan presenting to a local hospital to wait for delivery were invited to participate in the

92

study from July 2011 to June 2012. The water sources for domestic water supplies in the two

93

cities are surface waters, and the chlorination is applied during drinking water treatment. All

94

the pregnant women participated in the study during the third trimester of pregnancy (≥35

95

weeks, mean: 39 weeks, range: 35 to 47 weeks). Eligible pregnant women included in this

96

study were > 18 years old, lived in the local cities for at least 1 year, and were single

97

gestation viable fetuses. After exclusions, a total of 1184 pregnant women were retained in

98

this study. Of them, 893 pregnant women provided blood samples and urine samples for

99

determination, and thus were finally included in the current analysis. Written informed

100

consent was obtained from each participant on recruitment. The study was approved by the

101

Ethics Committee of Tongji Medical College.

102

Questionnaire

103

A face-to-face questionnaire was administered to each pregnant woman at enrollment

104

under the guidance of trained investigators. The questionnaire included socio-demographic

105

characteristics, lifestyles, water-use activities, medical histories, and occupational exposures.

106

The socio-demographic characteristics included ethnicity, study city, age, marital status,

107

parity, educational background, and household income. Maternal factors and lifestyles 5



108

included pre-pregnancy and prenatal body mass index (BMI), weight gain during pregnancy,

109

alcohol use, and passive smoking. Pregnant women were defined as passive smokers if they

110

were exposed to cigarette fumes over 15 min per day at home and/or at work. The

111

information on water-use activities included use of boiled water and filtered water (yes/no),

112

status of swimming during pregnancy (yes/no), total tap water consumption per day (glass

113

size multiplied by number), and minutes spent showering/bathing time per day (the duration

114

of showering/bathing multiplied by frequency).

115

Blood sample collection and analysis

116

On the first day of hospital admittance to a prenatal care visit, we collected 5-mL

117

peripheral blood from each pregnant woman, using a specially treated tube to reduce

118

background THM contaminations.35 After the blood sample draw, the tube was gently shaken

119

to dissolve the anticoagulant, sent to the laboratory with ice packs and stored at 4°C until

120

THM analysis within 2 weeks. The THM concentrations in blood, including chloroform

121

(TCM), bromodichloromethane (BDCM), dibromochloromethane (DBCM), and bromoform

122

(TBM), were measured following the method that has been described in detail in our previous

123

study.36 Briefly, volatiles from 3-mL blood sample headspace were extracted using a solid

124

phase micro-extraction (SPME) fiber on a magnetic stirrer. Then, the SPME fiber was

125

immediately inserted into the gas chromatography (GC) inlet and maintained it for 3 minutes

126

to desorb the analytes followed by detection using an electron capture detector (ECD). Each

127

analysis run (20-30 samples) included two blanks and two quality control samples. The limit

128

of detections (LOD) for the four individual THMs ranged from 0.45 to 2.00 ng/L.

129

Urine sample collection and analysis 6


Page 6 of 36

Page 7 of 36


130

We collected a spot first morning urine sample from each participant using a 100-mL

131

polyethylene cup. All the urine samples were sent to the laboratory with ice packs and stored

132

at -40°C until TCAA analysis within 6 months. The TCAA concentration in urine was

133

measured following the method that has been described in detail in our previous study.37

134

Briefly, a 10-mL urine sample was adjusted to pH < 0.5 with concentrated sulfate acid and

135

extracted with methyl-tert-butyl-ether containing the internal standard 1,2-dipropyl bromide.

136

After extraction, acidic methanol was added into organic extraction to convert TCAA into its

137

methyl ester. Then, the target analyte was detected using the GC coupled with an ECD. Each

138

analysis run (30-40 samples) included one blank and two quality control samples. The LOD

139

for urinary TCAA was 2.00 µg/L. Urinary creatinine was measured with an automatic

140

biochemical analyzer according to Jaffe’s colorimetric method. The creatinine-adjusted

141

TCAA concentration (µg/g creatinine) was calculated using the TCAA value divided by the

142

creatinine value.

143

Statistical analysis

144

The concentrations of target analytes below the LOD were assigned with LOD divided

145

by the square root of 2 for the analysis. Brominated THMs (Br-THMs) were defined as sum

146

of BDCM, DBCM, and TBM. Total THMs (TTHMs) were defined as sum of TCM and

147

Br-THMs. Descriptive statistics for individual characteristics, water-use activities, and blood

148

THM and urinary TCAA concentrations of the study population were calculated. Because

149

blood THM and urinary TCAA concentrations were positive-skewed distributions,

150

log10-transformed values were used in the following analysis. Spearman correlation

7



151

coefficients were used to examine the correlations between blood THMs and

152

creatinine-adjusted urinary TCAA concentrations.

153

Multivariable linear regression models were used to identify predictors of blood THM

154

and creatinine-adjusted urinary TCAA concentrations. DBCM and TBM were not modeled

155

individually because high proportions of non-detects were measured in blood samples. Based

156

on previous studies on predictors of human exposure to THMs and TCAA,26, 28-32 we selected

157

potential predictors including study city, ethnicity, age, pre-pregnancy and prenatal BMI,

158

marital status, parity, household income, educational background, alcohol use, passive

159

smoking, log total tap water consumption, boiled water, filtered water, showering/bathing

160

time, and swimming during pregnancy. We firstly examined the associations of potential

161

predictors with blood THM and urinary TCAA concentrations using bivariate regression

162

analyses. Potential predictors with p ≤ 0.20 in bivariate regression analyses were included in

163

an initial multivariable model that was subsequently reduced to retain all predictors with p ≤

164

0.10 in a final multivariable model using a stepwise procedure.

165

We finally modeled log total tap water consumption, age, and prenatal BMI as

166

continuous variables, study city (Wuhan vs. Xiaogan), ethnicity (other vs. Han), passive

167

smoking (yes vs. no), and boiled water (yes vs. no) as categorical variables in multivariable

168

models. For ease of interpretation, each estimated change in final multivariable models of

169

blood THM and creatinine-adjusted urinary TCAA concentrations was calculated according

170

to the following formula: 10β-1, where β is the regression coefficient of a given predictor.

171

Sensitivity analyses were also conducted to predictors of creatinine-adjusted urinary TCAA

172

concentrations when excluding very dilute urine samples (creatinine values < 0.3 g/L) and 8


Page 8 of 36

Page 9 of 36


173

very concentrated urine samples (creatinine values > 3.0 g/L). We used R2 values to estimate

174

the relative variation of the total variance. Statistical significance was considered as a p-value

175

< 0.05. All the data analysis was performed using the Predictive Analytics Suite Workstation

176

(PASW) version 18.0 (IBM corporation, Armonk, New York, USA).

9



177

Results

178

Characteristics of study population

179

Table 1 shows descriptive statistics of the study population. The study population was

180

predominantly Han (99.0%) and first marriage (97.1%), and had a mean age at delivery of

181

28.7 (±4.6) years. The mean pre-pregnancy and prenatal BMI were 20.4 (±3.4) and 26.8 (±3.4)

182

kg/m2, respectively. Four hundred seventy-nine pregnant women (53.6%) reported their

183

educational background less than college; and 429 pregnant women (49.3%) reported their

184

annual household income below 5500 $. For water-use activities, the mean total tap water

185

consumption per day were 1303.7 (±835.9) mL; the majority of pregnant women drank

186

boiled water (91.8%), but few of them used filtered water (16.6%) and swam during

187

pregnancy (0.4%).

188

Blood THM and urinary TCAA concentrations

189

Table 2 presents the distribution of blood THM and urinary TCAA concentrations among

190

the study population. Among the individual THM species, the maximum percent detection

191

was TCM (92.2%) followed by BDCM (58.8%), DBCM (30.9%), and TBM (22.6%), with

192

the geometric mean of 40.01, 1.56, 0.85, and 1.61 ng/L, respectively. The TCAA was detected

193

in 98.9% of the urine samples, and the geometric mean of creatinine-adjusted and unadjusted

194

urinary TCAA concentrations were 9.66 µg/g creatinine and 7.28 µg/L, respectively.

195

All correlation coefficients (rs) were statistically significant among the blood THMs,

196

except for TCM and Br-THMs (Table 3). There were the largest correlation coefficient

197

between TCM and TTHMs (rs = 0.99) and the smallest correlation coefficient between TCM

10


Page 10 of 36

Page 11 of 36


198

and BDCM (rs = 0.08). There were no significant associations between blood THMs and

199

urinary creatinine-adjusted TCAA.

200

Predictors of blood THM concentrations

201

Multivariable models for predicting third trimester blood THM concentrations among

202

pregnant women are showed in Table 4. The final models explained 19.5%, 19.3%, 19.2%,

203

and 15.9% of the total variance in blood concentrations of TCM, BDCM, Br-THMs, and

204

TTHMs, respectively. Study city was the strongest significant predictor of THM

205

concentrations. Compared with pregnant women from Xiaogan, pregnant women from

206

Wuhan had significantly lower concentrations of TCM and TTHMs but had significantly

207

higher concentrations of mean BDCM and Br-THMs (all p < 0.05). Prenatal BMI was

208

associated with decreased TCM and TTHM concentrations (p = 0.02 and 0.07, respectively);

209

whereas age was associated with increased BDCM and Br-THM concentrations (p = 0.05 and

210

0.03, respectively). Log total tap water consumption was associated with increased BDCM

211

concentrations. Intake of boiled water and passive smoking were associated with lower TCM

212

and TTHM concentrations. Other ethnicities of pregnant women had higher concentrations of

213

BDCM and Br-THMs than those Han (both p = 0.03).

214

Predictors of urinary TCAA concentrations

215

Multivariable model for predicting third trimester creatinine-adjusted urinary TCAA

216

concentrations among pregnant women are presented in Table 5. Prenatal BMI and study city

217

were the statistically significant predictors of creatinine-adjusted urinary TCAA

218

concentrations that explained 28.5% of the total variance. Pregnant women from Wuhan had

219

significantly lower concentrations of creatinine-adjusted urinary TCAA than those from 11



220

Xiaogan (p < 0.001). Prenatal BMI was associated with decreased creatinine-adjusted TCAA

221

concentrations (p = 0.05). Sensitivity analyses for predictors of creatinine-adjusted urinary

222

TCAA concentrations showed similar results (data not shown).

12


Page 12 of 36

Page 13 of 36

223


Discussion

224

This study identified predictors of third trimester blood THM and urinary TCAA

225

concentrations among pregnant women in a Chinese cohort. Consistent with previous

226

studies,28,

227

population. The median concentrations of blood TCM and TTHMs in this study population

228

were higher than those reported in a sample of U.S. postpartum women29 and in a U.S.

229

general population from 1999-2006 National Health and Nutrition Examination Survey

230

(NHANES),28 but were comparable to those reported in a male population from Wuhan,

231

China32 (Fig.1 A). The median concentrations of creatinine-adjusted urinary and unadjusted

232

TCAA in this study population were also higher than those reported in a U.S. general

233

population from 1988-1994 NHANES,38 but were comparable to those reported in a sample

234

of pregnant women from Wuhan, China.16 The median concentrations of unadjusted urinary

235

TCAA in this study population were also similar to those reported in a sample of UK

236

pregnant women31 (Fig.1 B).

29, 31

a low proportion of blood Br-THMs (

Predictors of Third Trimester Blood Trihalomethanes and Urinary

Recommend Documents