Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis

EI-03611; No of Pages 8 Environment International xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Environment International journal homepage: www.elsevier.com/locate/envint

Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis Hui Gao a, Yuan-duo Zhu a, Yuan-yuan Xu a,b, Yun-wei Zhang a, Hui-yuan Yao a, Jie Sheng b, Zhong-xiu Jin b, Ling-ling Ren b, Kun Huang a, Jia-hu Hao a,b, Fang-biao Tao a,b,⁎ a b

Department of Maternal, Child and Adolescent Health, School of Public Health, Anhui Medical University, Hefei, Anhui, China Anhui Provincial Key Laboratory of Population Health & Aristogenics, Hefei, Anhui, China

a r t i c l e

i n f o

Article history: Received 26 November 2016 Received in revised form 21 February 2017 Accepted 24 March 2017 Available online xxxx Keywords: Phthalate metabolites Pregnant women Birth cohort Repeated measure analysis China

a b s t r a c t In utero exposure to phthalates may have adverse effects on pregnant women and their offsprings. Therefore, the exposure level of these substances among individuals, particularly among sensitive population, is of concern. The objective of the present study is to characterize urinary concentrations of phthalate metabolites at multiple time points during pregnancy in Chinese women. A total of 3455 pregnant women were included from Ma'anshan Birth Cohort in China. Spot urine samples in the morning (8:00–10:00) and questionnaires were obtained at three separate visits (approximately in 10, 26, and 34 gestational weeks). Seven phthalate metabolites from urine samples were analyzed, including monomethyl phthalate (MMP), monoethyl phthalate (MEP), monobutyl phthalate (MBP), mono benzyl phthalate (MBzP), mono-2-ethylhexyl phthalate (MEHP), mono (2-ethyl-5oxohexyl) phthalate (MEOHP) and mono (2-ethyl-5-hydroxyhexyl) phthalate (MEHHP). Geometric means of concentrations were ranged from 0.05 to 41.0 ng/mL for all the metabolites mentioned above. No individual exposure level was above the 95th percentiles for all the seven phthalates. On the three separate visits, 0.5%, 0.9% and 1.2% of the participants had coexposure to above the 75th percentiles for all metabolites. Taken these visits together, a total of 29 urine samples had concentrations above the 95th percentiles, while 3.0%–5.6% of urine levels were above 75th percentiles for at least one specific phthalate metabolite. We observed moderate intraclass correlation coefficients (ICCs) ranging from 0.44 to 0.56 for MBzP, MEHP and MEP, and lower ICCs, from 0.28 to 0.32, for MMP, MBP, MEOHP and MEHHP. Sampling season was associated with concentrations of all phthalate metabolites, showing heavier exposure was more likely to occur during summer. In summary, phthalate exposure is prevalent in Chinese pregnant women. However, throughout pregnancy coexposure to multiple phthalates at the upper percentile of exposure is infrequent. Mild to moderate temporal stability indicates that a single measurement in spot urine collected in the morning (8:00–10:00) seems not enough to describe throughout pregnancy phthalate exposure. Urinary levels vary by sampling seasons, which should be taken into consideration in future analyses. © 2017 Published by Elsevier Ltd.

1. Introduction Phthalates are a large class of synthetic compounds and highly produced for over 50 years. Phthalates are extensively used as plasticizers in daily life polyvinyl chloride (PVC) products, such as food packaging materials, personal care products, children's toys, medical and pharmaceutical articles (Dodson et al., 2012; Wittassek et al., 2011). The amount of PVC products used worldwide annually is around 35 million tonnes, and about 40% of which is consumed in Asia where China accounts for the majority (Andrady and Neal, 2009). Phthalates can easily ⁎ Corresponding author at: Department of Maternal, Child and Adolescent Health, School of Public Health, Anhui Medical University, No. 81 Meishan Road, Hefei 230032, Anhui, China. E-mail addresses: [email protected], [email protected] (F. Tao).

leach out into surrounding environment, because they are not chemically bound to the polymer of the products. Therefore, humans are constantly exposed to phthalates through ingestion, inhalation and dermal absorption (Wittassek et al., 2011). And their metabolites are detectable in various biospecimens, including urine, serum, breast milk, placental tissue, amniotic fluid and umbilical blood (Frederiksen et al., 2007; Hines et al., 2009; Jensen et al., 2012; Yan et al., 2009). Phthalates are well-known environmental endocrine disruptors. It has been shown that in utero exposure to phthalates potentially increase adverse health risks in pregnant women and their offsprings. For instance, epidemiological studies revealed that in utero exposure to some phthalates was associated with pregnancy complications (e.g. spontaneous abortion and increased blood pressure) (Jukic et al., 2016; Mu et al., 2015; Toft et al., 2012; Werner et al., 2015), adverse birth outcomes (e.g. preterm birth, low birth weight and decreased

http://dx.doi.org/10.1016/j.envint.2017.03.021 0160-4120/© 2017 Published by Elsevier Ltd.

Please cite this article as: Gao, H., et al., Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.03.021

2

H. Gao et al. / Environment International xxx (2017) xxx–xxx

birth length) (Ferguson et al., 2014; Huang et al., 2009; Latini et al., 2003; Meeker et al., 2009; Minatoya et al., 2017; Philippat et al., 2012; Wolff et al., 2008; Zhang et al., 2009), and impaired children growth (Valvi et al., 2015) and neurodevelopment (e.g. lower IQ, problems with attention, hyperactivity and poorer social communication) (Miodovnik et al., 2014). Furthermore, during the critical period of foetal/child development such as in utero period and the first few years of life, exposure to environmental chemicals will increase disease risk later in life (Barouki et al., 2012). Therefore, much more attention should be paid on exposure levels of phthalates in pregnant women. Currently, the urinary concentrations of phthalate metabolites in pregnant women have been reported across several countries and areas. For instance, the Health Outcomes and Measures of the Environment study (HOME) in the United States (Yolton et al., 2011), the Polish Mother and Child Cohort study (REPRO_PL) in Poland (Polanska et al., 2014), the Mothers and Children's Environmental Health study (MOCEH) in Korea (Kim et al., 2011), the Early Life Exposure in Mexico to Environmental Toxicants cohort study (ELEMENT) (Téllez-Rojo et al., 2013), and the Taiwan Maternal and Infant Cohort study (Lien et al., 2015) have analyzed urinary concentrations of several phthalate metabolites in pregnant women. In the mainland of China, limited studies have investigated urinary phthalate levels in this population. The Healthy Baby Cohort (HBC) study measured six phthalate metabolites in 293 urine samples from pregnant women at delivery (Zhu et al., 2016). Our previous Ma'anshan Birth Cohort (MABC) study analyzed seven phthalate metabolites in 3103 urine samples of pregnant women during early pregnancy (5–14 weeks gestation) (Gao et al., 2016). However, the exposure level was described by single urine sample measurement in the two studies mentioned above. The aim of our present study is to further investigate the concentrations and profiles of phthalate metabolites using urine samples from multiple time points during pregnancy. 2. Method

extraction followed by high-performance liquid chromatography-tandem mass spectrometry (6410LC-MS, Agilent Technologies Co., Santa Clara, CA, USA) analysis was used. Seven phthalate metabolites, including monomethyl phthalate (MMP), monoethyl phthalate (MEP), monobutyl phthalate (MBP), mono benzyl phthalate (MBzP), mono-2ethylhexyl phthalate (MEHP), mono (2-ethyl-5-oxohexyl) phthalate (MEOHP) and mono (2-ethyl-5-hydroxyhexyl) phthalate (MEHHP) were measured. Concentrations below the limit of detection (LOD) were assigned with LOD/√2. Creatinine adjustment was used to correct for urine dilution. Urine creatinine was measured using a creatinine assay kit through picric kinetic method (Jiancheng Bioengineering Institute, Nanjing, China). 2.3. Statistical analysis Geometric means (GMs) and their 95% confidence intervals (95% CIs) were used to describe the distributions of both unadjusted and creatinine-adjusted concentrations of urinary phthalate metabolites. Using percentile values, the coexposure patterns of the seven phthalate metabolites at each study visit, and the single exposure to a specific metabolite across all the three visits were evaluated. Spearman's rank correlations were performed to assess relationships between unadjusted and creatinine-adjusted concentrations of phthalate metabolites. Intraclass correlation coefficients (ICCs) and their 95% CIs were used to estimate between- and within-person variability (i.e., temporal reliability). ICCs N 0.75 suggested strong reproducibility, 0.4 to 0.75 suggested moderate reproducibility, and b 0.4 suggested weak reproducibility (Rosner, 2011). Finally, the metabolite concentrations of all urine samples were compared between categories of demographic variables and seasons by linear mixed models with random effects. All statistical analyses were performed with SPSS software (Version 13.0), and p b 0.05 was considered statistically significant with the twotailed tests.

2.1. Study participants and sampling

3. Results

This study was based on MABC study. The cohort study was a population-based prospective study, aimed to evaluate the impacts of maternal phthalate exposure on adverse pregnancy outcomes, child health and development. In brief, 3474 pregnant women, who were N18 years old and b14 gestational weeks, were consecutively recruited from May 2013 to September, when they came to Ma'anshan Maternal and Child Health Care Center for their first antenatal care visit. Gestational weeks were calculated using last menstrual period data, or estimated by ultrasound if their menstruation were irregular (Carey et al., 2000). At baseline (mean 10 gestational weeks), questionnaires and spot urine samples in the morning (8:00–10:00; within 7 days after completing questionnaires) were obtained. Afterwards, women were followed up during their second (mean 26 gestational weeks) and third (mean 34 gestational weeks) trimesters of pregnancy to complete followed-up questionnaires and provide urine samples. Three questionnaires were used to collect demographic data and three urine samples were provided to measure phthalate metabolites. Among 3474 women, 19 participants were excluded because of unavailable urine sample. 3455 women with at least one urinary sample were ultimately enrolled. Of these, 243 (7.0%), 350 (10.1%) and 2862 (82.4%) women provided one, two and three urine samples, respectively. The sampling seasons were defined as follows: spring (March–May); summer (June–August); fall (September–November); and winter (December–February) (Riala et al., 2009).

Demographic characteristics of the participants are shown in Table 1. The means of maternal age and pre-pregnancy BMI were 26.2 years and 20.9 kg/m2, respectively. Most women were highly educated (79.4% above high school), primiparas (88.2%), urban residents (60.3%), employed (57.9%) and nondrinkers (92.0%). 73.6% of the women had household income above 2500 yuan per month capita. 47.6% of the participants were passive smokers. 90% of the 9477 investigations reported no application of sunscreen products. 83.2% reported spending more time indoors. The proportions of sample collections in four seasons were almost equal. Statistical analyses were performed for both unadjusted and creatinine-adjusted concentrations of urinary phthalate metabolites, and the results were highly consistent (Spearman's r: 0.61–0.92, p b 0.05). The distributions of the seven phthalate concentrations were shown in Table 2. The detectable frequency ranged from 49.9% to 99.9%. The highest concentration was MBP, followed by MMP, MEOHP, MEHHP, MEP, MEHP and MBzP. Mann-Whitney U tests found significant differences of metabolites' concentrations across three visits (all p b 0.001, data not shown). Figs. 1 and 2 showed the percentages of individuals whose exposures were greater than certain percentile of the exposure distribution. For all the seven metabolites, most women were exposed to concentrations below the 25th percentiles at each study visit (63%, 61% and 60%, respectively) (Fig. 1). At any visit, no individual was exposed to concentrations above the 95th percentiles (data not shown). Regarding the urines that were collected from the three visits, 45%–50% of all metabolite (except MBzP) concentrations were above the 25th percentile. 29 (0.8%) samples were above the 95th percentile for at least one specific phthalate metabolite (Fig. 2).

2.2. Phthalate metabolites analyses Urine samples were analyzed with previous methodology with modifications (Gao et al., 2015; Wang et al., 2013). Solid phase

Please cite this article as: Gao, H., et al., Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.03.021

H. Gao et al. / Environment International xxx (2017) xxx–xxx Table 1 Characteristics of 3455 pregnant women (9529 urine samples) enrolled in the MABC study (2013/05–2014/09). Variables

Mean ± SD or n (%)

Maternal age at enrollment (years) Pre-pregnancy BMI (kg/m2) Underweight (b18.5) Normal (18.5–23.9) Overweight (N24.0)

26.2 ± 3.7 20.9 ± 2.9 612(17.7) 2404 (69.6) 439 (12.7)

Education Middle school or below High school Junior college University or above

713 (20.6) 782 (22.6) 1056 (30.6) 904 (26.2)

Gravidity (pregnanciesa) 1 2 N3

1906 (55.2) 984 (28.5) 565 (16.4)

Parity (live birthsb) 0 N1

3046 (88.2) 409 (11.8)

Household monthly income per capita (yuan) 2499 or less 2500–3999 4000 or more

913 (26.4) 1484 (43.0) 1058 (30.6)

Residence during the last half year City Rural

2085 (60.3) 1370 (39.7)

Employment Unemployed Employed

1456 (42.1) 1999 (57.9)

Passive smoking during pregnancy No Yes

1812 (52.4) 1643 (47.6)

Alcohol use during pregnancy No Yes

3179 (92.0) 276 (8.0)

Using sunscreen products at the time of the urine collection (N = 9477)c Frequently 237 (2.5) Sometimes 710 (7.5) Never 8530 (90.0) Time spent indoors (N = 9477)d More time indoors Same time in- and outdoors More time outdoors

7886 (83.2) 1392 (14.7) 205 (2.2)

Season at the time of the urine collection (N = 9517)e Spring Summer Fall Winter

1929 (20.3) 2503 (26.3) 2472 (26.0) 2613 (27.5)

Abbreviations: BMI, body mass index; SD, standard deviation. a Gravidity is the number of pregnancies. b Parity is the number of times a female has given live birth. c Data for 52 urine samples were unavailable. d Data for 46 urine samples were unavailable. e Data for 12 urine samples were unavailable.

Both unadjusted and creatinine-adjusted concentrations of urinary MEP, MBzP and MEHP varied slightly within persons, with ICCs ≥0.37 (Table 3). Unadjusted concentrations of MMP (ICC = 0.28), MBP (ICC = 0.30), MEOHP (ICC = 0.32) and MEHHP (ICC = 0.32) were relatively more, but moderately, variable throughout the collection period. Results of their creatinine-adjusted concentrations were consistent. Analyses of creatinine-adjusted concentrations of phthalate metabolites were shown in Table 4. Levels of MEHP, MEOHP and MEHHP were positively associated with parity. MMP and MBP were related to frequency of sunscreen products application. MMP, MEP and MBP increased in proportion to amount of time indoors. All metabolites'

3

concentrations were associated with sampling season. Phthalates exposure was more likely to be heavier during summer, except for creatinine-adjusted MEHP and MEOHP. The box plots (Fig. 3A and B) illustrated the comparisons of metabolites' concentrations between different seasons. Besides, some of the phthalate metabolites were found to be related to other factors, including maternal age, education, income, employment and passive smoking. However, prepregnancy BMI, gravidity, residence and alcohol consumption were not associated with any phthalate exposure.

4. Discussion Repeated measurements data on phthalate exposure levels in Chinese pregnant women are limited. In the present study, we collected and analyzed urine samples at three separate visits. Detectable metabolites suggested that phthalate exposure was prevalent among Chinese pregnant women. However, neither coexposure to multiple phthalates at the upper percentile of exposure nor high level exposure to a specific phthalate across all the visits existed. These results indicated that simultaneous exposure to multiple phthalates at low-doses across pregnancy was likely to occur in the participants. ICCs showed weak (0.14, MEOHP) to moderate (0.56, MEP) temporal reproducibility, suggesting that repeated measurements of phthalates were necessary within a certain period of time. Season-dependent concentrations of phthalate metabolites in our study provided a novel insight into controlling this potential confounder in future studies. The pattern of phthalates exposure in our study was similar to previous studies in Mainland China. In Chinese pregnant women from the HBC study, median concentrations of total MBP (53.0 ng/mL) were the highest, followed by secondary metabolites of di(2-ethylhexyl) phthalate (DEHP, including MEHHP (5.59 ng/mL) and MEOHP (4.70 ng/mL)), MEP (4.60 ng/mL) and MEHP (1.73 ng/mL) (Zhu et al., 2016). However, the characteristics of other cohorts (Kim et al., 2011; Lien et al., 2015; Polanska et al., 2014; Téllez-Rojo et al., 2013; Yolton et al., 2011)were different from ours. Comparing with data from the HOME, MOCEH and REPRO_PL cohort study (Kim et al., 2011; Polanska et al., 2014; Yolton et al., 2011), concentration of MBP in our study was much higher than that in USA (16 week: GM = 24.0 ng/mL; 26 week: GM = 20.3 ng/mL), Korea (median = 16.6 ng/mL) and Polish (GM = 29.4 ng/mL). But DEHP secondary metabolites' (MEOHP and MEHHP) levels in the present study were lower. In REPRO_PL study (Polanska et al., 2014), the concentrations of MEP (mean = 81.3 ng/mL) and MBzP (mean = 0.2 ng/mL) were higher and the concentration of MEHP (mean = 0.4 ng/mL) was lower, compared with our data. The concentrations of the all metabolites (MMP, MEP, MBP, MBzP, MEHP, MEOHP and MEHHP) in this study were lower than those from Taiwan Maternal and Infant cohort studies (GM = 54.53, 61.71, 66.88, 15.88, 16.93, 13.59 and 7.91 ng/mL, respectively) (Lien et al., 2015) and ELEMENT cohort study, Mexico (except MMP unreported, GM = 138, 85.61, 3.54, 6.56, 14.23 and 22.08 ng/mL, respectively) (Téllez-Rojo et al., 2013). In a word, the exposure levels of phthalates are inconsistently according to these international cohorts. A possible explanation for those discrepancies is different legislative actions across different countries. For instance, in the United States, more toxic phthalates with 4–6 carbons (e.g., di-n-butyl phthalate) have been largely replaced (e.g., by diethyl phthalate). In the EU legislation, some substances including DEHP, di-n-butyl phthalate, di-iso-butyl phthalate and butyl benzyl phthalate are banned in cosmetic products (Wittassek et al., 2011). There is no simple approach to characterize coexposure patterns for multiple chemicals (Qian et al., 2015). The NHANES data showed that no individual was exposed to concentrations above the 95th percentiles for all the six phthalates. For 75% of individuals, no single substance level was above the 95th percentile. These data suggested that heavy exposure to multiple phthalates was rare in the NHANES population (Qian et al., 2015). Similar analyses were performed in the present study.

Please cite this article as: Gao, H., et al., Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.03.021

4

H. Gao et al. / Environment International xxx (2017) xxx–xxx

Table 2 Geometric mean (95% confidence intervals) of unadjusted (ng/mL) and creatinine-adjusted (μg/g creatinine) urinary concentrations of phthalate metabolites of 3455 Chinese pregnant women. Phthalate metabolite

MMP MEP MBP MBzP MEHP MEOHP MEHHP

Unadjusted (ng/mL) Creatinine-adjusteda Unadjusted (ng/mL) Creatinine-adjusteda Unadjusted (ng/mL) Creatinine-adjusteda Unadjusted (ng/mL) Creatinine-adjusteda Unadjusted (ng/mL) Creatinine-adjusteda Unadjusted (ng/mL) Creatinine-adjusteda Unadjusted (ng/mL) Creatinine-adjusteda

NLOD (%)

98.7 99.5 99.9 49.4 99.4 99.9 99.8

Geometric mean (95% confidence intervals)

Spearman's rank correlation

Overallb

1st trimesterc

2nd trimesterd

3rd trimestere

12.9(0.7227.3) 18.0(1.7374.0) 5.1(0.3,83.4) 7.1(0.6113.6) 41.0(5.1339.8) 57.4(7.6477.7) 0.05(0.01,1.5) 0.06(0.01,2.0) 3.0(0.4,22.6) 4.2(0.5,46.4) 6.4(1.1,37.1) 8.9(2.0,61.3) 5.2(0.9,33.4) 7.3(1.4,61.1)

13.8(1.8129.9) 12.8(2.7112.6) 9.7(1.2135.3) 9.0(1.3110.7) 52.6(6.5410.4) 48.7(7.8330.7) 0.08(0.01,2.0) 0.08(0.01,1.8) 2.7(0.3,20.2) 2.5(0.3,23.5) 7.5(1.4,42.7) 6.9(2.1,34.9) 5.5(1.2,31.8) 5.1(1.3,30.8)

19.9(0.1438.4) 30.6(0.2902.2) 4.4(0.4,71.5) 6.8(0.5158.1) 41.6(5.2355.7) 64.2(6.1784.0) 0.04(0.01,1.4) 0.07(0.01,2.7) 4.1(0.6,28.4) 6.3(0.7,68.9) 7.2(1.2,46.0) 11.2(1.3110.7) 6.3(0.8,47.3) 9.8(1.0,98.9)

7.6(0.7121.4) 15.0(2.2197.8) 2.9(0.2,44.4) 5.8(0.5,75.9) 30.8(4.1205.4) 60.9(9.8376.6) 0.03(0.01,0.9) 0.05(0.01,1.5) 2.5(0.3,20.2) 5.0(0.7,36.2) 4.7(0.9,21.0) 9.3(3.1,37.6) 4.0(0.9,24.3) 7.9(2.1,44.7)

0.78–0.91 0.85–0.90 0.80–0.86 0.71–0.92 0.75–0.80 0.61–0.76 0.67–0.79

Abbreviations: LOD, limit of detection. a The unit was μg/g creatinine. b 9529 urine samples. c 3294 urine samples. d 3186 urine samples. e 3049 urine samples.

Our results indicated that throughout pregnancy coexposure to multiple phthalates at low-doses existed in Chinese pregnant women. Several studies have investigated the temporal reproducibility of phthalate metabolites in pregnant women populations over periods ranging from days to months (Adibi et al., 2008; Cantonwine et al.,

Fig. 1. Visit-specific percentages of the women had coexposure to seven metabolites above the 25th, 50th and 75th percentile levels.

2014). Adibi et al. (2008) collected 2–4 urine samples over a 6-week period during the third trimester in 28 pregnant Dominican and AfricanAmerican women. According to their report, the ICCs for unadjusted MBP (0.62) and MBzP (0.66) were higher, and the ICCs of unadjusted MEP (0.30) and MEHP (0.35) were lower, compared with our results (MBP = 0.30, MBzP = 0.44, MEP = 0.56 and MEHP = 0.49). More recently, Cantonwine et al. (2014) recruited 139 pregnant women in Northern Puerto Rico and determined phthalate metabolite concentrations using urines collected at three separate visits (averagely in 18, 22 and 26 weeks of gestation). In this study, the ICCs for SG-adjusted MBzP, MEHP and MEHHP were 0.41, 0.36 and 0.24, which were similar to our findings (0.43, 0.37 and 0.23), indicating moderate to weak robustness. The ICC of unadjusted MEP (0.43) was lower and that of MBP (0.41) was higher, compared with our results (MEP = 0.56 and MBP = 0.30). The ICCs of creatinine-adjusted MEHHP were comparable within these studies (Adibi et al. = 0.23, Cantonwine et al. = 0.24 and our study = 0.23). Because we collected three urine samples over approximately 5 months, we predicted that all ICCs in this study might be lower than those in other studies. These unexpected results raise issues. Is a single urine sample enough to assess phthalate exposure for a short period of exposure? How long and how often should we collect urine samples during pregnancy? In pregnant women population, there are no conclusive evidences on relationships between phthalate metabolites and socio-demographic and maternal factors. Regarding socio-demographic characteristics, Arbuckle et al. (2014) found that the majority of the target phthalate metabolites including MBzP, MEOHP and MEHHP were associated

Table 3 Intraclass correlation coefficients (ICCs) and 95% confidence intervals (95% CIs) for lntransformed concentrations of urinary phthalate metabolites. Phthalate metabolites

Fig. 2. Metabolite-specific percentages of the samples above the 25th, 50th, 75th and 95th percentile levels.

MMP MEP MBP MBzP MEHP MEOHP MEHHP

Unadjusted

Creatinine-adjusted

ICC

95% CI

ICC

95% CI

0.28 0.56 0.30 0.44 0.49 0.32 0.32

0.23,0.32 0.53,0.59 0.26,0.35 0.41,0.48 0.46,0.52 0.27,0.36 0.28,0.36

0.23 0.51 0.22 0.43 0.37 0.14 0.23

0.18,0.27 0.48,0.54 0.17,0.27 0.39,0.46 0.33,0.41 0.08,0.19 0.18,0.28

Please cite this article as: Gao, H., et al., Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.03.021

H. Gao et al. / Environment International xxx (2017) xxx–xxx

5

Table 4 Associations between creatinine-adjusted urinary phthalate concentrations and demographic/season categories by linear mixed model (N = 9529). Variables

β estimates and 95% confidence intervals MMP

MEP

MBP

MBzP

MEHP

MEOHP

MEHHP

0.73(−3.14,4.61)

−0.17(−6.70,6.35)

0.10(−0.15,0.34)

0.69(−0.64,2.02)

−0.66(−1.71,0.39)

−0.09(−1.31,1.13)

3.29(−1.68,8.26)

−1.56(−9.98,6.86)

−0.11(−0.43,0.20)

0.65(−1.07,2.36)

−0.13(−1.48,1.23)

0.19(−1.39,1.76)

−1.34(−5.53,2.86)

−1.90(−9.00,5.20)

0.25(−0.01,-0.52)

0.15(−1.30,1.60)

−0.23(−1.38,0.91)

0.48(−0.85,1.81)

−2.35(−7.23,2.54)

−1.46(−9.76,6.84)

0.11(−0.20,0.42)

−0.76(−2.45,0.93)

0.65(−0.68,1.99)

−0.53(−2.08,1.02)

4.84(−1.53,11.22)

−8.99(−15.00,-2.98)

0.70(−9.49,10.88)

0.14(−0.25,0.52)

0.83(−1.24,2.91)

0.28(−1.36,1.92)

0.76(−1.14,2.66)

1.51(−4.19,7.22) −4.50(−9.16,0.16) Ref

−6.78(−12.17,-1.39) −1.60(−5.99,2.79)

1.69(−7.42,10.81) 3.55(−3.90,10.99)

0.10(−0.24,0.45) 0.06(−0.22,0.34)

−1.09(−2.95,0.76) 0.05(−1.49,1.56)

−0.63(−2.09,0.84) 0.27(−0.93,1.46)

−0.55(−2.25,1.16) 0.72(0.67,2.11)

Gravidity (pregnanciesa) 1 Ref 2 −0.28(−4.31,3.375) N3 −2.11(−7.74,3.52)

3.05(−0.76,4.68) 2.58(−2.72,7.88)

4.65(−1.78,11.08) 3.31(−5.68,12.30)

0.04(−0.20,0.28) 0.13(−0.21,0.47)

0.58(−0.73,1.89) −0.12(−1.95,1.71)

0.50(−0.53,1.54) −0.50(−1.95,0.94)

0.69(−0.51,1.90) −0.21(−1.89,1.47)

Parity (live birthsb) 0 Ref N1 −2.35(−8.60,3.91)

−1.22(−7.13,4.68)

5.42(−4.58,15.42)

0.21(−0.17,0.58)

2.88(0.84,4.91)

1.97(0.36,3.58)

2.04(0.17,3.91)

Household monthly income per capita (yuan) 2499 or less 0.56(−4.10,5.21) −3.16(−7.56,1.23) 2500–3999 −1.07(−5.07,2.93) −0.07(−3.84,3.69) 4000 or more Ref

5.74(−1.70,13.18) 6.27(−0.13,12.67)

−0.11(−0.39,0.18) 0.09(−0.15,0.33)

1.28(−0.23,2.79) 0.20(−1.10,1.50)

1.11(−0.09,2.30) 0.32(−0.71,1.35)

1.71(0.33,3.10) 0.34(−0.86,1.53)

−0.58(−4.36,3.19)

−0.19(−6.59,6.22)

−0.10(−0.34,0.15)

0.30(−1.01,1.60)

0.59(−0.44,1.62)

0.48(−0.71,1.68)

2.03(−1.75,5.81)

6.62(0.22,13.03)

0.02(−0.22,0.26)

0.00(−1.30,1.30)

0.31(−0.72,1.34)

0.10(−1.10,1.29)

Passive smoking during pregnancy No Ref Yes −0.11(−3.58,3.36)

−1.31(−4.59,1.97)

−6.60(−12.14,-1.05)

−0.14(−0.35,0.07)

−0.71(−1.84,0.42)

0.11(−0.78,1.00)

0.00(−1.04,1.04)

Alcohol use during pregnancy No Ref Yes −1.34(−7.56,4.88)

−5.17(−10.98,0.65)

1.67(−8.29,11.63)

−0.07(−0.45,0.30)

−0.14(−2.16,1.88)

−0.08(−1.68,1.51)

−0.35(−2.20,1.50)

−0.01(−0.47,0.46) 0.14(−0.12,0.39)

−0.89(−4.37,2.59) 0.28(−1.77,2.32)

0.65(−2.15,3.45) −0.37(−2.00,1.26)

0.39(−2.88,3.65) 0.07(−1.76,1.91)

Maternal age at enrollment (years) b25 4.11(0.03,8.20) 25–29 Ref N30 1.95(−3.32,7.22) Pre-pregnancy BMI (kg/m2) 0.27(−4.18,4.71) Underweight (b18.5) Normal Ref (18.5–23.9) Overweight 0.70(−4.49,5.89) (N24.0) Education Middle school or below High school Junior college University or above

Residence at last half year City Ref Rural −0.53(−4.53,3.48) Employment Unemployed Employed

0.09(−3.91,4.10) Ref

c

Using sunscreen products at the time of the urine collection (N = 9477) Frequently 0.40(−10.40,11.20) 5.66(−4.61,15.93) 16.17(−1.74,34.07) Sometimes 7.69(1.47,13.91) 2.64(−2.67,7.95) 24.42(13.69,35.15) Never Ref Time spent indoors (N = 9483)d More time 2.69(−1.96,7.34) indoors Same time inRef and outdoors More time −2.21(−13.86,9.45) outdoors

0.76(−3.46,4.97)

11.65(3.82,19.49)

−0.02(−0.21,0.17) −0.27(−1.79,1.25)

−1.15(−2.37,0.07)

−1.03(−2.43,0.36)

−4.77(−15.51,5.98)

−2.93(−22.45,16.60)

0.39(−0.09,0.86)

0.93(−2.93,4.79)

1.27(−1.80,4.33)

−0.47(−3.98,3.05)

−0.95(−2.51,0.61) 2.27(0.79,3.75) 0.72(−0.78,2.22)

−1.40(−2.65,-0.16) −0.15(−1.33,1.03) −0.51(−1.73,0.71)

−0.88(−2.34,0.58) 2.20(0.84,3.57) 0.55(−0.87,1.98)

Season at the time of the urine collection (N = 9517)e Spring −4.98(−9.68,-0.28) 0.48(−4.25,5.21) Summer 20.41(16.07,24.74) 8.53(4.22,12.84) Fall 7.48(2.85,12.10) 0.10(−4.44,4.63) Winter Ref

21.14(14.03,28.26) 0.07(−0.13,0.07) 122.64(115.91,129.36) 0.17(0.01,0.32) 65.49(58.58,72.40) 0.11(−0.09,0.30)

Bold font indicates p value b 0.05. a Gravidity is the number of pregnancies. b Parity is the number of times a female has given live birth. c Data for 52 urine samples were unavailable. d Data for 46 urine samples were unavailable. e Data for 12 urine samples were unavailable.

Please cite this article as: Gao, H., et al., Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.03.021

6

H. Gao et al. / Environment International xxx (2017) xxx–xxx

Fig. 3. Comparisons of unadjusted (A) and creatinine-adjusted (B) concentrations between sampling seasons. The bottom and top of the box indicated the first and third quartiles, while the line inside the box represented the median value. The whisker markers represented the maximum and minimum values. When the levels of MBzP were below limit of detection (LOD), the minimum and the first quartile value of the unadjusted concentration was replaced by its LOD / √2 (0.01 ng/mL) for distribution.

with maternal age, prepregnancy BMI and income. However, we only found one relationship between MEHHP and income. In addition, a recent study by Zhu et al. (2016) suggested that there was no significant association of any phthalate metabolite with maternal age, prepregnancy BMI and income. Regarding maternal factors, all DEHP metabolites were positively related to parity in our study, which was partly consistent with Zhu et al. (2016). Furthermore, Zhu et al. (2016) also found significant relationships of parity with some low molecular metabolites (MEP and MBP). Despite these positive correlations, the only negative relationship was found between parity and MEP, according to Arbuckle et al. (2014). These conflicting results may be partly attributed to different sample sizes and designs of urine collection. We obtained spot urine per trimester of pregnancy from 3455 women, while one spot urine sample was collected from 293 women at delivery by Zhu et al. (2016) and from 2000 women during the first trimester by Arbuckle et al. (2014). Further scientific evidences are warranted. Our results showed that urinary concentrations of all phthalate metabolites varied by sampling seasons. The majority of the phthalate metabolites' levels, especially the low weight molecular metabolites, were highest in summer. However, Peck et al. (2010) and Arbuckle et al. (2014) did not observe this phenomenon. The reason was not well

understood. We speculated that sunscreens were possibly the main source of phthalate exposure in summer, because sunscreen application was correlated with higher level of MMP and MBP, according to our results. Besides, because of positive trends between time spending indoors and MMP, MEP and MBP concentrations, we also hypothesized that spending more time indoors might lead to higher phthalate exposure in summer. Some phthalates have been reported to be detectable in indoor air and dust samples (He et al., 2016; Weschler et al., 2015). In addition, poor ventilation in summer when houses are sealed off from the hot possibly influenced the concentrations of phthalates in room (Huo et al., 2016). Considering the two observations, it seems that more indoor activity, particular during summer, may be plausible to cause heavier phthalates exposure. Of note, Chi-square tests (Supplementary material, Table S1) demonstrated that the frequencies of sunscreen application and the amount of time spending indoors were greater in summer, which would further support our proposals. Unfortunately, we were unable to identify the direct origins of phthalate exposure, although the frequency of using sunscreen products and the time spent indoors were collected. This hampers the identification of the potential factors that may directly influence exposure or coexposure pattern, prediction and prevention.

Please cite this article as: Gao, H., et al., Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.03.021

H. Gao et al. / Environment International xxx (2017) xxx–xxx

To our knowledge, it is the first time a prospective cohort study was conducted in China and phthalate exposure was analyzed in pregnant women by collection of urinary biomarker measurements at up to three time points. However, a limitation of the study is the potential exposure misclassification. For instance, women who filled questionnaires and provided urine samples at their first visits were during 13–14 gestational weeks (9.7%), their urines might be erroneously recorded as the “first trimester” of pregnancy, according to the time order of follow-up visits. In other words, urine sample which was actually collected in the “second trimester” was possibly misclassified as a urine of the “first trimester”. In addition, we collected information regarding the time range (8:00–10:00), other than exact time (e.g., 8:30) of urine collection. Future studies should take this into consideration and make more detailed analyses. Finally, another limitation is that we were unable to adopt a “summary measure” to estimate coexposure to multiple phthalates and over multiple trimesters of pregnancy. We used percentiles of exposure distribution to evaluate phthalate exposure levels (e.g., below 25th percentile means low-dose), according to the study of Qian et al. (2015). 5. Conclusion In conclusion, exposure to phthalates is prevalent among Chinese pregnant women. Coexposure to multiple phthalates at high-doses and throughout pregnancy exposure to specific phthalate at highdoses are infrequent. Mild to moderate temporal stabilities indicate that a single measurement is insufficient to describe throughout pregnancy exposure to phthalates. Urinary levels are associated with sampling season and other factors, which suggests that future analyses should take them into consideration. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.envint.2017.03.021. Conflict statement The authors declare that they have no actual or potential competing financial interests in this manuscript. Acknowledgements Funding was provided by the National Natural Science Foundation of China (No. 81330068). References Adibi, J.J., Whyatt, R.M., William, P.L., Calafat, A.M., Camann, D., Herrick, R., et al., 2008. Characterization of phthalate exposure among pregnant women assessed by repeat air and urine samples. Environ. Health Perspect. 116 (4), 467–473. Andrady, A.L., Neal, M.A., 2009. Applications and societal benefits of plastics. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 364 (1526), 1977–1984. Arbuckle, T.E., Davis, K., Marro, L., Fisher, M., Legrand, M., LeBlanc, A., et al., 2014. Phthalate and bisphenol A exposure among pregnant women in Canada—results from the MIREC study. Environ. Int. 68, 55–65. Barouki, R., Gluckman, P.D., Grandjean, P., Hanson, M., Heindel, J.J., 2012. Developmental origins of non-communicable disease: implications for research and public health. Environ. Health 11, 42. Cantonwine, D.E., Cordero, J.F., Rivera-González, L.O., Anzalota Del Toro, L.V., Ferguson, K.K., Mukherjee, B., et al., 2014. Urinary phthalate metabolite concentrations among pregnant women in Northern Puerto Rico: distribution, temporal variability, and predictors. Environ. Int. 62, 1–11. Carey, J.C., Klebanoff, M.A., Hauth, J.C., Hillier, S.L., Thom, E.A., Ernest, J.M., et al., 2000. Metronidazole to prevent preterm delivery in pregnant women with asymptomatic bacterial vaginosis. N. Engl. J. Med. 342, 534–540. Dodson, R.E., Nishioka, M., Standley, L.J., Perovich, L.J., Brody, J.G., Rudel, R.A., 2012. Endocrine disruptors and asthma-associated chemicals in consumer products. Environ. Res. 120 (7), 935–943. Ferguson, K.K., McElrath, T.H., Meeker, J.D., 2014. Environmental phthalate exposure and preterm birth. JAMA Pediatr. 168 (1), 61–67. Frederiksen, H., Skakkebaek, N.E., Andersson, A.M., 2007. Metabolism of phthalates in humans. Mol. Nutr. Food Res. 51 (7), 899–911. Gao, H., Xu, Y., Sun, L., Jin, Z., Hu, H., Sheng, J., et al., 2015. Determination of seven phthalate metabolites in human urine by high performance liquid chromatography-tandem mass spectrometry. Se Pu 33 (6), 622–627 (Chinese).

7

Gao, H., Xu, Y.Y., Huang, K., Ge, X., Zhang, Y.W., Yao, H.Y., 2016. Cumulative risk assessment of phthalate in pregnant Chinese women associated with birth outcomes: a prospective cohort study. Environ. Pollut. 222, 549–556. He, R., Li, Y., Xiang, P., Li, C., Zhou, C., Zhang, S., et al., 2016. Organophorus flame retardants and phthalate esters in indoor dust from different microenvironments: bioaccessibility and risk assessment. Chemosphere 150, 528–535. Hines, E.P., Calafat, A.M., Silva, M.J., Mendola, P., Fenton, S.E., 2009. Concentrations of phthalate metabolites in milk, urine, saliva, and serum of lactating North Carolina women. Environ. Health Perspect. 117 (1), 86–92. Huang, P.C., Kuo, P.L., Chou, Y.Y., Lin, S.J., Lee, C.C., 2009. Association between prenatal exposure to phthalates and the health of newborns. Environ. Int. 35 (1), 14–20. Huo, C.Y., Liu, L.Y., Zhang, Z.F., Ma, W.L., Song, W.W., Li, H.L., et al., 2016. Phthalate esters in indoor window films in a northeastern Chinese urban center: film growth and implications for human exposure. Environ. Sci. Technol. 50 (14), 7743–7751. Jensen, M.S., Norgaard-Pedersen, B., Toft, G., Hougaard, D.M., Bonde, J.P., Cohen, A., et al., 2012. Phthalates and perfluorooctanesulfonic acid in human amniotic fluid: temporal trends and timing of amniocentesis in pregnancy. Environ. Health Perspect. 120 (6), 897–903. Jukic, A.M., Calafat, A.M., McConnaughey, D.R., Longnecker, M.P., Hoppin, J.A., Weinberg, C.R., 2016. Urinary concentrations of phthalate metabolites and bisphenol A and associations with follicular-phase length, luteal-phase length, fecundability, early pregnancy loss. Environ. Health Perspect. 124 (3), 321–328. Kim, Y., Ha, E.H., Kim, E.J., Park, H., Ha, M., Kim, J.H., et al., 2011. Prenatal exposure to phthalates and infant development at 6 months: prospective Mothers and Children's Environmental Health (MOCEH) study. Environ. Health Perspect. 119 (10), 1495–1500. Latini, G., De Felice, C., Presta, G., Del Vecchio, A., Paris, I., Ruggieri, F., et al., 2003. In utero exposure to di-(2-ethylhexyl)phthalate and duration of human pregnancy. Environ. Health Perspect. 111 (14), 1783–1785. Lien, Y.J., Ku, H.Y., Su, P.H., Chen, S.J., Liao, P.C., Chen, W.J., et al., 2015. Prenatal exposure to phthalate esters and behavioral syndromes in children at 8 years of age: Taiwan Maternal and Infant Cohort Study. Environ. Health Perspect. 123 (1), 95–100. Meeker, J.D., Hu, H., Cantonwine, D.E., Lamadrid-Figueroa, H., Calafat, A.M., Ettinger, A.S., et al., 2009. Urinary phthalate metabolites in relation to preterm birth in Mexico city. Environ. Health Perspect. 117 (10), 1587–1592. Minatoya, M., Araki, A., Miyashita, C., Sasaki, S., Goto, Y., Nakajima, T., et al., 2017. Prenatal di2-ethylhexyl phthalate exposure and cord blood adipokine levels and birth size: the Hokkaido study on environment and chilren's health. Sci. Total Environ. 579, 606–611. Miodovnik, A., Edwards, A., Bellinger, D.C., Hauser, R., 2014. Developmental neurotoxicity of ortho-phthalate diesters: review of human and experimental evidence. Neurotoxicology 41, 112–122. Mu, D., Gao, F., Fan, Z., Shen, H., Peng, H., Hu, J., 2015. Levels of phthalate metabolites in urine of pregnant women and risk of clinical pregnancy loss. Environ. Sci. Technol. 49 (17), 10651–10657. Peck, J.D., Sweeney, A.M., Symanski, E., Gardiner, J., Silva, M.J., Calafat, A.M., et al., 2010. Intra- and inter-individual variability of urinary phthalate metabolite concentrations in Hmong women of reproductive age. J. Expo. Sci. Environ. Epidemiol. 20 (1), 90–100. Philippat, C., Mortamais, M., Chevrier, C., Petit, C., Calafat, A.M., Ye, X., et al., 2012. Exposure to phthalates and phenols during pregnancy and offspring size at birth. Environ. Health Perspect. 120 (3), 464–470. Polanska, K., Ligocka, D., Sobala, W., Hanke, W., 2014. Phthalate exposure and child development: the Polish Mother and Child Cohort Study. Early Hum. Dev. 90 (9), 477–485. Qian, H., Chen, M., Kransler, K.M., Zaleski, R.T., 2015. Assessment of chemical coexposure patterns based upon phthalate biomonitoring data within the 2007/2008 National Health and Nutrition Examination Survey. J. Expo. Sci. Environ. Epidemiol. 25 (3), 249–255. Riala, K., Hakko, H., Taanila, A., Rasanen, P., 2009. Season of birth and smoking: findings from the Northern Finland 1996 Birth Cohort. Chronobiol. Int. 26 (8), 1660–1672. Rosner, B., 2011. Fundamentals of Biostatistics. Brooks/Cole, Boston, MA. Téllez-Rojo, M.M., Cantoral, A., Cantonwine, D.E., Schnaas, L., Peterson, K., Hu, H., et al., 2013. Prenatal urinary phthalate metabolites levels and neurodevelopment in children at two and three years of age. Sci. Total Environ. 461-462, 386–390. Toft, G., Jönsson, B.A., Lindh, C.H., Jensen, T.K., Hjollund, N.H., Vested, A., et al., 2012. Association between pregnancy loss and urinary phthalate levels around the time of conception. Environ. Health Perspect. 120 (3), 458–463. Valvi, D., Casas, M., Romaguera, D., Monfort, N., Ventura, R., Martinez, D., et al., 2015. Prenatal phthalate exposure and childhood growth and blood pressure: evidence from the Spanish INMA-Sabadell birth cohort study. Environ. Health Perspect. 123 (10), 1022–1029. Wang, H.X., Wang, B., Zhou, Y., Jiang, Q.W., 2013. Rapid sensitive analysis of phthalate metabolites, bisphenol A, and endogenous steroid hormones in human urine by mixedmode solid-phase extraction, dansylation, and ultra-performance liquid chromatography coupled with triple quadruple mass spectrometry. Anal. Bioanal. Chem. 405 (12), 4313–4319. Werner, E.F., Braun, J.M., Yolton, K., Khoury, J.C., Lanphear, B.P., 2015. The associated between maternal urinary phthalate concentrations and blood pressure in pregnancy: the HOME study. Environ. Health 14, 75. Weschler, C.J., Bekö, G., Koch, H.M., Salthammer, T., Schripp, T., Toftum, J., et al., 2015. Transdermal uptake of diethyl phthalate and di(n-butyl) phthalate directly from air: experimental verification. Environ. Health Perspect. 123 (10), 928–934. Wittassek, M., Koch, H.M., Angerer, J., Brüning, T., 2011. Assessing exposure to phthalates the human biomonitoring approach. Mol. Nutr. Food Res. 55 (1), 7–31. Wolff, M.S., Engel, S.M., Berkowitz, G.S., Ye, X., Silva, M.J., Zhu, C., et al., 2008. Prenatal phenol and phthalate exposures and birth outcomes. Environ. Health Perspect. 116 (8), 1092–1097.

Please cite this article as: Gao, H., et al., Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.03.021

8

H. Gao et al. / Environment International xxx (2017) xxx–xxx

Yan, X., Calafat, A., Lashley, S., Smulian, J., Ananth, C., Barr, D., et al., 2009. Phthalates biomarker identification and exposure estimates in a population of pregnant women. Hum. Ecol. Risk. Assess. 15 (3), 565–578. Yolton, K., Xu, Y., Strauss, D., Altaye, M., Calafat, A.M., Khoury, J., 2011. Prenatal exposure to bisphenol A and phthalates and infant neurobehavior. Neurotoxicol. Teratol. 33 (5), 558–566.

Zhang, Y., Lin, L., Cao, Y., Chen, B., Zheng, L., Ge, R.S., 2009. Phthalate levels and low birth weight: a nested case-control study of Chinese newborns. J. Pediatr. 155 (4), 500–504. Zhu, Y., Wan, Y., Li, Y., Zhang, B., Zhou, A., Cai, Z., et al., 2016. Free and total urinary phthalate metabolite concentrations among pregnant women from the Healthy Baby Cohort (HBC), China. Environ. Int. 88, 67–73.

Please cite this article as: Gao, H., et al., Season-dependent concentrations of urinary phthalate metabolites among Chinese pregnant women: Repeated measures analysis, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.03.021