Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine

Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine

EI-03560; No of Pages 9 Environment International xxx (2017) xxx–xxx Contents lists available at ScienceDirect Environment International journal hom...

1MB Sizes 0 Downloads 12 Views

EI-03560; No of Pages 9 Environment International xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

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

Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine Ruihua Dong a, Tong Zhou b, Shanzhen Zhao c, Han Zhang a, Meiru Zhang a, Jingsi Chen a, Min Wang c, Min Wu a, Shuguang Li a, Bo Chen a,⁎ a b c

Key Laboratory of Public Health Safety of the Ministry of Education, School of Public Health, Fudan University, Shanghai, China Shanghai Institutes of Preventive Medicine, China Shanghai Entry–Exit Inspection and Quarantine Bureau, Shanghai, China

a r t i c l e

i n f o

Article history: Received 4 August 2016 Received in revised form 10 January 2017 Accepted 10 January 2017 Available online xxxx Keywords: Phthalate metabolites in urine 24-hour recall survey Food Frequency Questionnaire Shanghai adults

a b s t r a c t Background: Diet is considered to be a significant exposure pathway for phthalates. In this study, we assessed the associations between food consumption and urinary concentrations of phthalate metabolites among Shanghai adults. Methods: A cross-sectional study involving 2418 participants was conducted in the fall of 2012. Recent food consumption was assessed by a 24-h dietary recall survey, and a Food Frequency Questionnaire (FFQ) characterized long-term dietary patterns. Urinary metabolites of six phthalates were measured. Results: Both the 24-h recall survey and FFQ identified wheat, dairy, and fruits as being positively associated with the excretion of phthalate metabolites. The 24-h recall data also showed positive associations with processed meats and alcohol. We evaluated the impact of reported consumption of multiple food categories simultaneously (wheat, fruits, meats, etc.) on metabolite excretion and found that, as more food types were consumed, the number of metabolites excreted, as well as their concentrations, increased with high significance (p values b 0.0001). We also evaluated the two survey instruments together. When both surveys reported consumption of fruits and dairy, the numbers of metabolites and their concentrations were significantly higher compared to when both surveys reported non-consumption, (p values b 0.000001). Rice consumption was found to be negatively associated with phthalate excretion; frequent and high levels of rice consumption were found to be associated with lower excretion of metabolites. Conclusion: Food consumption was associated with phthalate exposure in Shanghai adults. Both 24-h recall and FFQ identified significant associations between consumption of food types and phthalate exposure. © 2017 Elsevier Ltd. All rights reserved.

1. Introduction Public concern has been raised about phthalates due to associations with detrimental health effects, such as allergies, obesity, Abbreviations: BMI, body mass index; BBzP, butyl benzyl phthalate; CIs, confidence intervals; DBP, dibutyl phthalate; DEHP, bis(2-ethylhexyl) phthalate; DEP, diethyl phthalate; DiBP, di-iso-butyl phthalate; DiDP, di-iso-decyl phthalate; DiNP, di-iso-nonyl phthalate; DMP, dimethyl phthalate; DnBP, din-butyl phthalate; FCMs, food contact materials; FFQ, Food Frequency Questionnaires; HMW, high molecular weight; IQR, interquartile range; LC-MS/MS, liquid chromatography tandem mass spectrometry; LMW, low molecular weight; LOD, limit of detection; MBzP, mono-benzylphthalate; MCMHP, mono-2-carboxymethyl-hexyl phthalate; MECPP, mono-2-ethyl-5-carboxypentylphthalate; MEP, monoethyl phthalate; MEHP, mono-2-ethylhexylphthalate; MEHHP, mono2-ethyl-5-hydroxyhexylphthalate; MEOHP, mono-2-ethyl-5-oxohexyphthalate; MiBP, monoisobutylphthalate; MMP, monomethyl phthalate; MnBP, mono-nbutylphthalate; PVC, polyvinyl chloride; SHFCS, Shanghai Food Consumption Survey. ⁎ Corresponding author at: School of Public Health, Fudan University, 138 Yixueyuan Road, Shanghai 200032, China. E-mail address: [email protected] (B. Chen).

atherosclerosis, and asthma (Lind and Lind, 2011; Martino-Andrade and Chahoud, 2010; Hatch et al., 2008; Bornehag et al., 2004; Singh and Li, 2012; Caldwell, 2012). Phthalates, which are the diesters of 1,2-benzenedicarboxylic acid, are chemicals that are present in several commercial products. High-molecular-weight (HMW) phthalates (≥ 250 Da), such as butyl benzyl phthalate (BBzP), bis(2-ethylhexyl) phthalate (DEHP), di-iso-nonyl phthalate (DiNP), and di-iso-decyl phthalate (DiDP), are mostly used in the production of flexible vinyl plastics, flooring, and medical devices (Schettler, 2006; Cao, 2010; Wormuth et al., 2006). Low-molecular-weight (LMW) phthalates (b250 Da), such as dimethyl phthalate (DMP), diethyl phthalate (DEP), di-iso-butyl phthalate (DiBP), and din-butyl phthalate (DnBP), are commonly used in the production of varnishes, paints, lacquers, and personal care products (Cao, 2010; Wormuth et al., 2006; Sathyanarayana, 2008). The major use of phthalates is to impart durability and flexibility to plastics, such as polyvinyl chloride (PVC) (Wormuth et al., 2006). Phthalates are not chemically bound to PVC; therefore, they can easily leach from commercial or industrial products

http://dx.doi.org/10.1016/j.envint.2017.01.008 0160-4120/© 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008

2

R. Dong et al. / Environment International xxx (2017) xxx–xxx

into food or air, leading to the contamination of foods and the environment (Wormuth et al., 2006). Diet may be a significant pathway for exposure to phthalates. Certain foods have been associated with phthalate exposure (Schettler, 2006; Colacino et al., 2010; Fromme et al., 2007; Schecter et al., 2013; Martine et al., 2012; Bradley et al., 2013; Sioen et al., 2012; Gonzalez-Castro et al., 2011). For example, the presence of DEHP has been reported in poultry, meat carcasses, cream, eggs, and fish (Serrano et al., 2014). Cross-sectional studies have reported that DEP metabolites in urine are associated with vegetable consumption, and that the metabolites of HMW phthalates and DEHP were positively associated with meat and poultry consumption (Colacino et al., 2010; Trasande et al., 2013). Population-based dietary assessments are of considerable interest due to the presence of phthalates in different types of foods. However, results obtained from epidemiologic and food-monitoring studies have been inconsistent. Phthalates can migrate into food from plasticized PVC food contact materials (FCMs), such as plastic containers, lid gaskets, can linings, tubing, dishware and utensils, gloves, and conveyor belts used in the manufacturing, processing, storage, and transportation of foods (Cao, 2010; National Toxicology P, 2003). Phthalates are also found in the adhesives on food wrappers and printing inks as well as coatings on cookware (Serrano et al., 2014). Foods with high fat have been reported to be contaminated by HMW phthalates that are more lipophilic, such as DEHP (Cao, 2010). Phthalate concentrations may vary based on food production origin, processing, packaging, and lipid content (Schecter et al., 2013; Wittassek et al., 2011). Most phthalates in human toxicokinetic studies have been reported to have half-lives of b24 h (Koch et al., 2005; Hoppin et al., 2002). It seems that phthalates do not build up in human bodies over days. Based on this, previous studies have primarily used a 24-h recall survey or dietary record method to identify associations between food consumption and phthalate exposure (Zota et al., 2016; Jo et al., 2016; Hartle et al., 2016). In contrast, a Food Frequency Questionnaire (FFQ) is usually used to collect information on long-term dietary behavior and is generally thought not to be well correlated with phthalate excretions in urine. However, some types of food may be consumed at least once in a day, especially main staple foods. We hypothesized that contamination of these types of food with phthalates would result in phthalate excretions in urine would be associated with the long-term dietary behavior. However, such studies linking phthalate exposure to FFQ data are scarce. Phthalate production and importation in China has been reported to be dominated with more toxic phthalates such as DEHP and DBP than DiNP or DiDP of lesser toxicity (Zhang et al., 2006). DEHP exposure in China has been reported to be similar to that in the U.S. and Japan, but DBP exposure has been shown to be almost ten times higher than in the U.S. and Japan (Guo et al., 2012). Diet was the major source of exposure to DEHP (accounting for N50% of the total DEHP exposure) and an important source of DBP (sum of DiBP and DnBP, accounting for N 10% of the total DBP exposure) in China (Guo et al., 2013). The Chinese people have unique dietary habits, and the impact of Chinese diets on phthalate exposure is of special interest in better understanding phthalate exposure in China. To the best of our knowledge, only one study has discussed this topic, with phthalate exposure in school children linked to FFQ data (Guo et al., 2012). In the current study, we used both 24-h recall and FFQ data to identify associations between food consumption and phthalate exposure in Shanghai adults.

using a four-time 24-h dietary recall questionnaire to collect the seasonal data of food consumption in the community-based general population (fall 2012, spring and winter 2013, and summer 2014), and a one-time 24-h dietary recall survey to collect the food data in schoolbased students with ages ≤ 18 years. All participants in the SHFCS also completed the FFQ. However, not all school-based students provided the spot urine samples, and we measured the phthalate metabolites in only the community-based population with ages N 18 years at the time this paper was drafted. The study population in this analysis was therefore community residents of ages N 18 years. The community-based SHFCS used the multi-stage cluster random sampling method to draw samples from nine of the 18 districts/counties in Shanghai, including Huangpu, Xuhui, Putuo, Hongkou, Jinshan, Pudong, Qingpu, Baoshan, and Chongming. Based on population density, one to six residential communities were randomly selected from each district/county. A total of 25 communities were selected throughout the city. In the first interview (fall 2012), 3322 participants were asked to complete a 24-h dietary recall survey and FFQ in the presence of trained dietitians. Additionally, anthropometric measurements and sociodemographic characteristics were recorded. Spot urine samples were obtained from 3082 participants and stored at − 20 °C. After the exclusion of 89 participants lacking weight or height information, 326 with insufficient urine samples for the detection of phthalate metabolites, 25 for unreasonable creatinine concentration (b20 μmol/L or N30,000 μmol/L), and 224 with age ≤ 18 years, 2418 participants with ages N 18 years had complete information on anthropometric measurements, demographic characteristics, food consumption, and phthalate metabolites. Written informed consent was obtained from all participants. The study was approved by the local authorities and the Ethics Committee of the School of Public Health at Fudan University. 2.2. Dietary assessment Dietary intake was assessed by the 24-h dietary recall survey and FFQ by trained dietitians during face-to-face interviews. The 24-h dietary recall survey gathered information on the types and amounts of foods consumed in the 24 h prior to the spot urine collection. A total of 312 food types were consumed. We grouped these 312 types of foods into the following 19 food categories: rice and products (rice); wheat and products (wheat); other staples; vegetables and products (vegetables); legumes and products (legumes); fungi and products (fungi); fruits and products (fruits); eggs and products (eggs); pork; beef and mutton; poultry; processed meats; visceral products (viscera); fish; other aquatic products (other aquatic); dairy products; beverages; nuts and snacks; and alcohol. In the FFQ, participants reported the yearly consumption frequency (“never or seldom”, “less than once per month”, “one to three times per month”, “one to three times per week”, “four to six times per week”, “once per day”, “twice per day”, and “three or more times per day”) and the average mass of each meal for the following 23 food categories: rice and products (rice); wheat and products (wheat); other staples; vegetables and products (vegetables); legumes and products (legumes); fungi and products (fungi); fruits and products (fruits); pork; beef and mutton; viscera; eggs and products (eggs); fish; other aquatic products (other aquatic); processed meats; poultry; cow's milk, yogurt, soy milk; nuts and snacks; beverages; beer; white wine; red wine; and spirits. To simplify the data analysis, we combined cow's milk, yogurt, and soy milk into a “dairy products” category, and beer, white wine, red wine, and spirits into an “alcohol” category.

2. Methods 2.3. Measurement of phthalate metabolites in urine 2.1. Study population and sampling The study participants were Shanghai residents who participated in the Shanghai Food Consumption Survey (SHFCS). The SHFCS was performed by Fudan University from September 2012 to August 2014

One spot urine sample from each participant was collected in glass tubes capped with polypropylene lids. Both tubes and lids had been previously washed to remove the background phthalates. We measured 10 phthalate metabolites, including monomethyl

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008

R. Dong et al. / Environment International xxx (2017) xxx–xxx

3

Table 1 Parent phthalates, their metabolites, and the main exposure pathways. Parent phthalate

Phthalate metabolites

Main exposure pathways

Dimethyl phthalate, (DMP) Diethyl phthalate, (DEP)

Monomethyl phthalate, (MMP) Monoethyl phthalate, (MEP)

Di-n-butylphthalate, (DnBP)

Mono-n-butyl phthalate, (MnBP)

Diisobutyl phthalate, (DiBP)

Monoisobutyl phthalate, (MiBP)

Butyl-benzyl phthalate, (BBP)

Mono-benzyl phthalate, (MBzP)

Di-2-ethylhexyl phthalate, (DEHP)

Mono-2-ethylhexyl phthalate, (MEHP) Mono-2-ethyl-5-oxohexyl phthalate, (MEOHP) Mono-2-ethyl-5-hydroxyhexyl phthalate, (MEHHP) Mono-2-ethyl-5-carboxypentyl phthalate, (MECPP) Mono-2-carboxymethyl-hexyl phthalate, (MCMHP)

Personal care products, air fresheners (Schettler, 2006; Kim et al., 2015) Personal care products, dust, air fresheners (Guo et al., 2013; Kim et al., 2015; Duty et al., 2005; Guo and Kannan, 2011) Personal care products, food, dust, medications (Schettler, 2006; Guo et al., 2012; Guo et al., 2013; Guo and Kannan, 2011; Hernandez-Diaz et al., 2009) Personal care products, food, dust (Guo et al., 2012; Guo et al., 2013; Guo and Kannan, 2011) Personal care products, dust, food, air fresheners (Wormuth et al., 2006; Guo et al., 2013; Kim et al., 2015; Duty et al., 2005; Guo and Kannan, 2011) Food, children's toys (Wormuth et al., 2006; Guo et al., 2012; Miodovnik et al., 2011)

phthalate (MMP), monoethylphthalate (MEP), mono-n-butylphthalate (MnBP), monoisobutylphthalate (MiBP), mono-benzylphthalate (MBzP), mono-2-ethylhexylphthalate (MEHP), mono-2-ethyl-5oxohexyphthalate (MEOHP), mono-2-ethyl-5-hydroxyhexylphthalate (MEHHP), mono-2-ethyl-5-carboxypentylphthalate (MECPP), and mono-2-carboxymethyl-hexyl phthalate (MCMHP). Table 1 lists the 10 metabolites and their parent phthalates and the main exposure pathways reported in the literature. Ten metabolites and six isotopically labeled internal standards (13C4-MnBP, 13C4-MEHP, 13C4-MEOHP, 13C4MEHHP, 13C4-MECPP, and 13C4-MCMHP) were purchased from Cambridge Isotope Laboratories (Andover, MA, USA). Phthalate metabolites in urine were analyzed by liquid chromatography tandem mass spectrometry (API 4000, LC-MS/MS, Shimadzu, USA) according to Tranfo et al. (2013). Briefly, 1 mL of urine sample was incubated with β-glucuronidase (Helix pomatia; Sigma, Louis, MO, USA; Type HP-2, aqueous solution, ≥ 100,000 units/mL) at 37 °C for 120 min. The sample was subsequently acidified with 1 mL of aqueous 2% (v/v) acetic acid, mixed with 100 μL of internal standard (100 μg/L), and loaded into a PLS column (Dikma, China; 60 mg/3 mL) previously activated with 2 mL methanol and 2 mL of aqueous 0.5% (v/v) acetic acid. After sample loading, the column was washed with 2 mL of aqueous 0.5% (v/v) acetic acid and eluted with 1 mL of methanol. The eluate was passed through a 0.2-μm filter and analyzed (10 μL) by LC-MS/MS (Shimadzu, USA; API 4000 LC/MS/MS system) coupled to an AQUASIL C18 column (150 × 4.6 mm; Thermo Fisher Scientific, Inc., USA). For each batch of 30 samples analyzed, two procedural blanks and four matrix-spiked samples at two different spiking concentrations (10 and 25 ng/mL) were processed. The average recoveries and relative standard deviations (RSD) of target metabolites in spiked samples respectively ranged from 71.5% to 109.1% and from 1.2% to 7.4% at 10 ng/mL, and ranged from 58.5% to 139.2% and from 0.8% to 8.1% at 25 ng/mL. Trace concentrations of MEP, MnBP, MiBP, and MEHP were detected in procedural blanks with average concentrations and RSDs ranging from 0.05 to 0.8 μg/L and from 3.7% to 9.3%, respectively. Sample concentrations of these metabolites were determined after subtraction of the blank values. The limits of detection (LOD) were 0.02, 0.20, 0.04, 0.04, 0.20, 0.60, 0.10, 0.20, 0.03, and 0.50 μg/L for MMP, MEP, MnBP, MiBP, MBzP, MEHP, MEOHP, MECPP, MEHHP, and MCMHP, respectively. In addition to the 10 metabolites, two micromolar sums (μmol/L) were also calculated, the sum of DBP metabolites (ΣDBP, including MiBP and MnBP) and the sum of DEHP metabolites (ΣDEHP, including MEHP, MEOHP, MECPP, MEHHP, and MCMHP). The concentrations of 10 phthalate metabolites and two micromolar sums were adjusted using creatinine to correct for urine dilution. Urinary creatinine concentrations were analyzed with an enzymatic method on an Architect C8000 automatic biochemical analyzer (ARCHITECT C8000, Abbott Laboratories, Illinois, USA).

2.4. Statistical analyses Data were analyzed by SPSS version 21.0. Urinary concentrations of metabolites below the LOD were assigned a value of 1/2 LOD. To normalize the data, phthalate metabolites in urine were natural log-transformed. Two-sided p-values b 0.05 were considered to be statistically significant. We used a four-step strategy to explore the associations between phthalate exposure and food consumption, using both the 24-h recall survey and FFQ. Step 1. Initial Screening with 24-Hr Recall: We used the 24-h recall data to identify associations between urinary concentrations of individual metabolites (or the micromolar sums of grouped metabolites) and mass intakes of individual food categories. Results from this analysis are described below, and they were also used to further explore other associations. Multivariate linear regression models to conduct the initial screen. The potential covariates used in the regression models were the total food intake, sex (male, female), nationality, age, educational level (≤ primary school, high school/technical secondary school, college or higher), marriage (married, other), smoking status (never, current/past smokers), and body mass index (BMI). The estimated regression coefficient (β) and standard error (SE) of each regression model were used to calculate the percent difference in urinary concentrations of phthalate parameters following the increase of the extra intake of 100 g of food in the past 24 h. The percent difference was calculated by the equation (e(β) − 1) × 100%. Step 2. Continued Use of 24-Hr Recall by Expanding the Co-Variates: Step 1 identified likely candidates for further statistical analysis with the 24-h recall data. Specifically, we hypothesized that if one phthalate metabolite was associated with a food type, it might be associated with other food types. For example, if DEHP metabolites in Step 1 were found to be positively linked to individual candidates (e.g., wheat, fruits, processed meats, and dairy), the co-consumption of two or more candidates would further increase the phthalate exposure. Therefore, we tested the relationship between DEHP metabolite excretions and additional food categories. We used the univariate linear models to estimate the effects of the co-consumption of candidate food categories on phthalate exposure. The other potential covariates in the univariate models were total food intake, sex, nationality, age, educational level, marriage, smoking status, and BMI, as mentioned above. The calculated geometric means (GM) and 95% confidence intervals (CIs) were presented to directly compare phthalate excretions between different groups of co-consumption in the 24-h data. The p value in the univariate models represented the significance across each group.

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008

4

R. Dong et al. / Environment International xxx (2017) xxx–xxx

Step 3. Find Associations Using the Food Frequency Questionnaire: From Steps 1 and 2, we were able to identify associations between recent food consumption patterns and phthalate metabolite excretions. We then sought to determine whether similar associations could be identified when looking at long-term patterns of consumption instead of only recent consumption. We divided the food categories of the FFQ data into two groups (I and II): group I included food categories in which N5% of the participants had a consumption frequency of at least one meal per day, and group II included other categories (group II data were not used). Only the group I categories of FFQ data were used to further explore the effects of food consumption on phthalate exposure. For individual categories of group I foods (e.g. rice), we divided the population into two further groups (A and B): group A included participants who consumed less than one meal per day of the individual category (e.g. rice), and group B included participants who consumed food from the evaluated category in at least one meal per day. We recalculated the consumption volume of group A participants as “0”, but in group B participants it was calculated as a frequency (once per day, twice per day, three or more times per day) multiplied by the mass of each meal. In this way, we were able to explore the long-term pattern of a frequently consumed food category in its association with phthalate exposure. When the FFQ data were recalculated, we also used the multivariate linear regression models to estimate the dietary effects on phthalate excretions. The set of the covariates was the same as that used in the 24-h recall analyses, and the percent difference represented the increase of urinary concentrations of phthalate parameters following the intake of an extra 100 g of food each day. Step 4. Associations Found When Using Both Questionnaires Jointly: From Steps 1, 2 and 3, we identified increased excretions of phthalate metabolites in association with food types (e.g. wheat) from both 24-h recall and FFQ data. We then reasoned that if the participants consumed these types of food (e.g. wheat) in both short-term (within the past 24 h) and longterm (at least one meal per day in the past year) patterns, they would have higher phthalate excretions than those with no consumptions in both patterns. Univariate linear models were also used to identify the associations in step 4. 3. Results Table 2 lists the demographic characteristics of the study population. The median age in our sample was 53 years. The results of the four-step analyses were as follows. Step 1. Initial Screen with 24-Hr Recall: The multivariate linear regression analyses of the 24-h recall data identified positive associations between excretions of some phthalate metabolites with recent consumption of wheat, fruits, processed meats, dairy, and alcohol, and negative associations with rice and fish (Fig. 1). The significant effects of fruits and fish on phthalate excretions were limited to DEHP metabolites. However, the significant effects of rice, wheat, processed meats, dairy, and alcohol were not limited to DEHP metabolites. After the extra intake of 100 g wheat, processed meats, dairy or alcohol, most phthalate excretions in the urine increased by 10 to 30%. Step 2. Continued Use of 24-Hr Recall by Expanding the Covariates: Step 1 identified some positive associations between phthalate excretions and individual categories of food. The consumption of these identified categories simultaneously (wheat, fruits, processed meats, etc.) significantly increased the metabolite excretions (Fig. 2). A higher number of identified categories consumed by the participants consumed in the past 24 h

Table 2 Demographic characteristics of study population (n = 2418). Characteristic

Category

Result

Age, median (IQR), y Sex, n (%)

– Male Female Han Others ≤Primary school High school/technical Secondary school ≥College graduate Married Other Never smoked Current/past smoker – – – –

53 (41.64) 1154 (47.7) 1264 (52.3) 2396 (99.1) 22 (0.9) 543 (22.8) 1428 (59.9)

Nationality, n (%) Education, n (%)

Marriage, n (%) Smoke, n (%) Total food intake, median (IQR), g Height, median (IQR), m Weight, median (IQR), kg BMI, median (IQR), kg/m2

413 (17.3) 2012 (85.7) 335 (14.3) 1801 (74.8) 607 (25.2) 1233 (875, 1488) 1.65 (1.59, 1.70) 64.0 (55.0, 70.0) 23.5 (21.3, 25.7)

BMI: body mass index; IQR: interquartile range.

indicated higher urinary concentrations of phthalate metabolites. The p values for trend across zero to one or more consumed categories were b 0.000001 for all representative parameters (MMP, MEP, MBzP, ΣDBP, and ΣDEHP). Step 3. Find Associations Using the Food Frequency Questionnaire: In seven categories, large segments of the population (above 5% of the sample size) consumed individual categories of food at a frequency of at least one meal per day (Fig. 3). The multivariate linear regression analyses identified wheat, fruits, and dairy being positively associated with excretions of some phthalate metabolites and rice as being negatively associated with all phthalate metabolites. Compared to the results of 24-h recall data, fruits had significant impacts on phthalate excretions with more parameters and higher percent differences of increased concentrations, but wheat had fewer parameters and dairy had lower percent differences (Fig. 1 and Fig. 3). Step 4. Associations Found When Using Both Questionnaires Jointly: Both 24-h recall and FFQ data had similar patterns in identifying wheat, dairy, and fruits as being positively associated with phthalate excretions. For wheat, dairy, and fruits, the univariate analyses showed that the p values for trend across zero (no consumption in the past 24 h and the FFQ data of less than one meal per day) to both (have consumption in the past 24 h and the FFQ data of at least one meal per day) were b0.01 for the representative parameters of most phthalates (MMP, MEP, ΣDBP and ΣDEHP), especially in wheat and dairy with p values b 0.000001 (Fig. 4).

4. Discussion In this study, we used both 24-h recall survey and FFQ data to identify associations between food consumption and urinary phthalate metabolites in Shanghai adults. It is generally thought that phthalates do not build up in human bodies over days because of their short halflives (Genuis et al., 2012; Zeng et al., 2013). However, some types of food are consumed in at least one meal per day and high-frequency food consumption may reveal a pattern of association with phthalate excretion, even if the food was limited or absent in the previous day or two from urine sampling. It is possible that phthalate metabolites are still being excreted 2 or maybe even 3 days after an exposure event. In seven types of food with high-frequency consumption, we identified similar types of food that were related to phthalate metabolites in both FFQ (Fig. 3) and 24-h recall data (Fig. 1). In wheat, dairy,

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008

R. Dong et al. / Environment International xxx (2017) xxx–xxx

5

Fig. 1. Multivariate linear regression analyses of phthalate metabolites in urine in associations with food categories of the 24-h recall data. Data were adjusted for total food intake, age, sex, nationality, education, smoking, marriage, and BMI. Each cell represents the percent increase or decrease in urinary concentrations of phthalate metabolites following the consumption of 100 g of food in the past 24 h. Colored cells represent significant associations (p b 0.05); gray cells with no numbers represent non-significant associations.

and fruits, both data sets showed that phthalate metabolites increased following the increase of the consumption mass. Since the FFQ data usually represent the coherent habits of dietary behavior, the FFQ data probably provide the baseline exposure to phthalates if these are related to each other. Then, when the population has a coherent food frequency habit of certain types of food, we expect a “bump up” of exposure in this population if they also consumed the food in the past 24 h. As expected, we found the synergistic effects of FFQ and 24-h data in wheat, dairy, and fruits. Participants who had an ongoing pattern of wheat, dairy, or fruit consumption (at least one meal per day) and also consumed these types of food in the past 24 h had the highest concentrations of urinary phthalate metabolites. The use of both 24-h recall and FFQ data in our study is of special interest for better understanding the contribution of diet to phthalate exposure, especially when the FFQ data can be divided into different food types based on the frequency of consumption. To our knowledge, this study is the first to use both types of food surveys to identify these associations. Bai et al. (2015) used FFQ data to identify the association of fruit, soft drinks, and a Western dietary pattern with total phthalate levels. However, only the daily frequency of fruit (two servings or

more per day) and soft drinks (more than one can per day) showed significance when compared to other frequencies, and the Western dietary pattern remained the same on the day before urine sampling in Bai's study (Bai et al., 2015). Unlike the FFQ data, 24-h recall data have been relatively well used to identify phthalates associated with diets (Zota et al., 2016; Jo et al., 2016; Hartle et al., 2016). However, the correlation analyses using only 24-h data may miss some key information. Phthalate contamination in foods may be occasional, and therefore, when people try to link phthalate exposure to the diets in the past 24 h, such a link may be missed. In contrast, the FFQ data represent prolonged dietary behaviors, and in the situations in which the food types are largely consumed on a daily basis, the link between phthalates and the diet may represent the effects of prolonged dietary behavior on phthalate exposure. In this case, using both FFQ and 24-h recall data to identify positive or negative associations between food consumption and phthalate exposure could be a better representation of population exposure. The FFQ data in those food types with daily frequency of consumption, as mentioned above, may provide useful information for identifying dietary sources of phthalate exposure. However, for the FFQ data

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008

6

R. Dong et al. / Environment International xxx (2017) xxx–xxx

Fig. 2. Univariate linear analyses of representative phthalate exposure parameters in associations with dietary factors identified as being associated with increased phthalate exposure by 24-h recall data. Data were adjusted for total food intake, age, sex, nationality, education, smoking, marriage, and BMI. Each number of the horizontal axis represents the number of dietary factors the participant had consumed in the past 24 h. The calculated GMs and 95% CIs are presented. The p values represent the significant trend across zero to more consumed food categories of the 24-h recall data.

in those food types with a low frequency of consumption (less than one meal per day), the relationship between phthalates and diet may be unreliable because of the short half-lives of phthalates in urine. In this case, the 24-h recall data probably serve as a better tool to identify dietary sources of phthalate exposure. In this study, the linear regression analyses were first used to perform a preliminary screening for dietary candidates positively associated with phthalate excretions. Those candidates that co-occurred in a person in the past 24 h showed higher phthalate excretions than those that did not co-occur. The p-value for the trend across zero (no consumption of candidate types of food) to high numbers (consumed all candidates simultaneously) showed tremendous significance. In this study, we identified several dietary factors (wheat, fruits, processed meats, dairy products, and alcohol) to be positively

associated with phthalate exposure and two factors (rice and fish) with negative associations. The three positive factors of wheat, fruits, and dairy products showed concurrent correlations with most phthalate excretions in both 24-h recall and FFQ data. Most positive factors in preliminary regression analyses of 24-h data showed synergistic effects when combined with each other. All the above results indicate the extensive and consistent associations between recent consumption of some food types and phthalate exposure. Previous studies have reported higher dietary exposure to HMW phthalates than LMW phthalates, and many papers have reported DEHP in association with diets (Trasande et al., 2013; Koch et al., 2013). However, our data showed that positive associations were not limited to the DEHP metabolites, but also other metabolites, including DMP, DEP, DBP, and BBzP. Unlike in the U.S. and the European Union

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008

R. Dong et al. / Environment International xxx (2017) xxx–xxx

7

Fig. 3. Multivariate linear regression analyses of phthalate metabolites in urine in association with food categories of the FFQ data. Data were adjusted for total food intake, age, sex, nationality, education, smoking, marriage, and BMI. Each cell represented the percent increase or decrease in urinary concentrations of phthalate metabolites following the consumption of 100 g of food each day. Colored cells represent significant associations (p b 0.05); gray cells with no numbers represent non-significant associations. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

(Zota et al., 2016), phthalate production and importation in China are still dominated by DEHP and DBP, but not their substitutes DiNP or DiDP (Zhang et al., 2006). The mean estimated daily dietary intake of phthalates for DMP, DEP, DiBP, DnBP, BBzP, and DEHP were 0.092, 0.051, 0.505, 0.703, 0.022, and 1.60 μg/kg-bw/d, respectively, for Chinese adults (Guo et al., 2012). Therefore, phthalate exposure from dietary sources should not be limited to DEHP in China. In this study, we identified wheat, fruits, processed meats, dairy products, and alcohol as dietary factors having extensive and consistent associations with increased metabolite excretions of phthalates. Among them, wheat, meats, and dairy have also been consistently reported to be dietary sources of phthalate exposure in previous studies (Serrano et al., 2014; Dickson-Spillmann et al., 2009). Wheat is a grain. A Norwegian study reported grain as a significant source of DiNP and DEHP (Sakhi et al., 2014). One Belgian study and one German study identified bread as a significant source of DEHP within the adult and adolescent populations (Sioen et al., 2012; Heinemeyer et al., 2013). In these studies, wheat was not an isolated type of food being used to explore the associations. In European countries and America, wheat products are the dominantly consumed grain, serving as a staple food. However, in China, rice is the staple food. In this study, we divided grains into rice and wheat. The wheat category in our data mainly included pasta, buns, flour, pizza, and bread. The wheat products are commonly consumed as fast food, which are often packaged in flexible plastic bags or wrapping film for easy handling. In contrast to the wheat category, our results unexpectedly showed a negative association between rice and phthalate exposure. Unlike wheat products, rice is the staple food in Shanghainese and does not serve as typical fast food in China. Rice may be negatively associated with phthalate exposure because the increase of rice consumption leads to the decreased intake of other dietary sources, such as wheat. Dairy was another type of food consistently associated phthalates in our study. This finding is supported by previous studies (Petersen and Jensen, 2010). Mono-(3-carboxypropyl) phthalate (MCPP), a metabolite of multiple phthalates, was found to be significantly associated with total dairy consumption in the data from the 2003–2004 National Health and Nutrition Examination Survey (NHANES) (Colacino et al., 2010). In a Korean study, dairy consumption significantly increased urinary MEOHP and MEHHP (Ji et al., 2010). Norwegian samples of raw milk in a study published in 1994 had total phthalate levels between 0.12 and 0.28 mg/kg (Sharman et al., 1994). Phthalates have a lipophilic nature and tend to concentrate in the lipid phase of food stuffs (Fankhauser-Noti et al., 2006).

Dairy products, including whole fat milk, cream, butter, and cheese have high fat contents and may be contaminated with phthalates (Cao, 2010; Dickson-Spillmann et al., 2009; Casajuana and Lacorte, 2004). However, the strong, extensive, and consistent associations between dairy consumption and phthalate exposure in our data were surprising when compared to the previous studies with sporadic findings. It is a limitation that our data of both 24-h recalls and FFQ did not contain information on whether the milk or yogurt was whole fat or not. More comprehensive studies are needed to explore whether dairy is a dietary source of phthalate exposure and whether packaging (glass, paper, plastic) has an effect. As mentioned above, phthalates tend to be concentrated in the lipid phase of foodstuffs (Fankhauser-Noti et al., 2006). However, our data only showed sporadic associations of enhanced phthalate exposure to commonly consumed fatty foods, including poultry, beef, and mutton. Pork and fish in our 24-h data showed negative associations between food consumption and some metabolite excretions of DEHP. The processed meats in the 24-h recall data had extensive associations with phthalate exposure in our study, but they differed from other animal foods in the ways that they are produced, sold, and consumed. Previous studies have reported fatty animal foods, including poultry, meat and fish, to be dietary sources of phthalate exposure (Trasande et al., 2013; Mervish et al., 2014). A “Temple Stay” intervention study reported that the elimination of meat and dairy products decreased the estimated daily intake of DEP, DiBP, DnBP, and DEHP (Ji et al., 2010). It is difficult to explain the lack of positive associations between meats and phthalates in our study. Chinese people have different cooking and eating styles than those of Western cultures. Although Chinese cooking fries and smokes meat and fish, Chinese people more frequently cook these animal foods by boiling them in hot water. In this case, phthalates may have leached into the soup, but when people only consumed the meat or fish and not the soup, it is not a dietary source of exposure. Another possible reason may lie in the difference between open fairs, where meat and fish are unpackaged and supermarkets, where meat and fish are packaged in wrapping film. Unpackaged meat and fish sold in open fairs have less possibility of being contaminated with phthalates. It should be noted that the previous studies usually did not separate meats from processed meats. In our data, processed meats showed an extensive correlation with enhanced phthalate excretions. The processed meat in China can usually be purchased from supermarkets, roadside grocery stores, and fast food outlets. Most of the processed meats are packaged in flexible plastic bags or wrapping film, which could be the source of phthalate contamination.

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008

8

R. Dong et al. / Environment International xxx (2017) xxx–xxx

effect on the overall levels of chemicals, including DiBP (Bradley et al., 2014). Our data also showed some positive associations with alcohol consumption in metabolites, mainly in LMW phthalates. The alcohol molecule is a small polar molecule with both lipophilic and hydrophilic characteristics. Phthalates are lipophilic, and the LMW phthalates are slightly polar, but the HMW phthalates are non-polar (Cao, 2008). It could be easier for the LMW phthalates to be dissolved in alcoholic drinks than the HMW phthalates. However, previous studies have reported alcoholic drinks being contaminated with both LMW phthalates and DEHP, and it is possible that the non-polar DEHP can also be dissolved in alcoholic drinks, especially when produced using plastic containing devices (Tranfo et al., 2013; Huang et al., 2014; Wang et al., 2015). Overall, this study identified some dietary factors that were extensively and consistently associated with increased phthalate exposure. Each identified food category was carefully discussed for the possible reasons of phthalate pollution. One important reason is suspected to be food contact materials (FCMs). China has banned the use of PVC in FCMs (Zhou et al., 2015) and recommended the replacement of PVC with PE, PP, and PET, which do not require the use of phthalates (Enneking, 2006; Biedermann et al., 2013). However, some studies have detected phthalates in these materials (Schmid et al., 2008; Montuori et al., 2008; Farhoodi et al., 2008). The extensive associations between phthalates and diets in our study require further investigation, especially on the role of FCMs. Our study has some strengths and weaknesses. The main strength of this study is the use of both 24-h recall and FFQ data. The 24-h recall data were categorized into more types of food than the previous studies and the FFQ data of food consumed in at least one meal per day were carefully analyzed. However, our study also has some limitations. One weakness is the cross-sectional design. Phthalate exposure assessed by a single spot urine sample is also a concern, since human urine is easily affected by factors other than diet, such as other routes of exposure or the timing of sampling. The 24-h recall data estimated on one day was also not sufficient. In this case, the possibility exists that these data may not be truly representative of either dietary behavior or phthalate exposure. Another weakness of this study is that we did not get other lifestyle information on the study participants, such as personal care products, medications, floor coverings, etc. Phthalate exposure may be attributed to these lifestyle factors rather than dietary behavior. 5. Conclusions Fig. 4. Univariate linear analyses of representative phthalate exposure parameters in associations with dietary factors (A. wheat; B. dairy; C. fruits) identified as being associated with increased phthalate exposure by both 24-h recall and FFQ data. Data were adjusted for total food intake, sex, nationality, age, educational level, marriage, smoking status, and BMI. The log transformed values of calculated GMs and 95% CIs are presented. The p values represent the significant trend across zero (no consumption in the past 24 h and the FFQ data of less than one meal per day) for both (have consumption in the past 24 h and the FFQ data of at least one meal per day).

Our data also showed extensively positive associations of phthalate exposure with the consumption of fruits and alcohol. These food types in the previous studies had either inconsistent or no associations with phthalate exposure. The fruits in our study were positively associated with most investigated phthalates. Another study in China also found positive associations between fruit consumption and DEHP metabolites (MEHHP and MEOHP) (Shen et al., 2015). By contrast, the data of the 2003–2004 NHANES study in the US showed inverse associations between fruit consumption and investigated phthalates. It is difficult to explain the different results between China and the U.S. Roadside stands of fruits are popular in China. In these stands, fruits are usually overwrapped in plastic film or foam, suggesting a possible source of phthalate contamination. However, it has been reported that washing, peeling, or cooking fruits that had been in contact with paper had no

In this unique study of Shanghai adults, both 24-h recall and FFQ data identified dietary factors associated with increased phthalate exposure. Both data sets showed enhanced phthalate excretions in association with dietary consumption of wheat, dairy, and fruits. The 24-h recall data also showed positive associations with enhanced phthalate excretions in processed meats and alcohol. Both data sets showed that rice consumption was negatively associated with phthalate excretions. These associations were extensive and consistent across most metabolites. These findings indicated that certain types of food could be important sources of phthalate exposure. Acknowledgments We thank all participants for their participation and kind assistance. This work was supported by funding from the National Natural Science Foundation of China (No. 81202208) and the Major State Research Development Program of China (No. 2016YFD0400602). References Bai, P.Y., Wittert, G.A., Taylor, A.W., Martin, S.A., Milne, R.W., Shi, Z., 2015. The association of socio-demographic status, lifestyle factors and dietary patterns with total urinary phthalates in Australian men. PLoS One 10 (4), e0122140.

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008

R. Dong et al. / Environment International xxx (2017) xxx–xxx Biedermann, M., Ingenhoff, J.E., Zurfluh, M., Richter, L., Simat, T., Harling, A., Altkofer, W., Helling, R., Grob, K., 2013. Migration of mineral oil, photoinitiators and plasticisers from recycled paperboard into dry foods: a study under controlled conditions. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 30 (5), 885–898. Bornehag, C.G., Sundell, J., Weschler, C.J., Sigsgaard, T., Lundgren, B., Hasselgren, M., Hagerhed-Engman, L., 2004. The association between asthma and allergic symptoms in children and phthalates in house dust: a nested case-control study. Environ. Health Perspect. 112 (14), 1393–1397. Bradley, E.L., Burden, R.A., Leon, I., Mortimer, D.N., Speck, D.R., Castle, L., 2013. Determination of phthalate diesters in foods. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 30 (4), 722–734. Bradley, E.L., Castle, L., Speck, D.R., 2014. Model studies of migration from paper and board into fruit and vegetables and into Tenax as a food simulant. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 31 (7), 1301–1309. Caldwell, J.C., 2012. DEHP: genotoxicity and potential carcinogenic mechanisms-a review. Mutat. Res. 751 (2), 82–157. Cao, X.L., 2008. Determination of phthalates and adipate in bottled water by headspace solid-phase microextraction and gas chromatography/mass spectrometry. J. Chromatogr. A 1178 (1–2), 231–238. Cao, X.L., 2010. Phthalate esters in foods: sources, occurrence, and analytical methods. Compr. Rev. Food Sci. Food Saf. 9 (1), 21–43. Casajuana, N., Lacorte, S., 2004. New methodology for the determination of phthalate esters, bisphenol A, bisphenol A diglycidyl ether, and nonylphenol in commercial whole milk samples. J. Agric. Food Chem. 52 (12), 3702–3707. Colacino, J.A., Harris, T.R., Schecter, A., 2010. Dietary intake is associated with phthalate body burden in a nationally representative sample. Environ. Health Perspect. 118 (7), 998–1003. Dickson-Spillmann, M., Siegrist, M., Keller, C., Wormuth, M., 2009. Phthalate exposure through food and consumers' risk perception of chemicals in food. Risk Anal. 29 (8), 1170–1181. Duty, S.M., Ackerman, R.M., Calafat, A.M., Hauser, R., 2005. Personal care product use predicts urinary concentrations of some phthalate monoesters. Environ. Health Perspect. 113 (11), 1530–1535. Enneking, P.A., 2006. Phthalates not in plastic food packaging. Environ. Health Perspect. 114 (2), A89–A90. Fankhauser-Noti, A., Biedermann-Brem, S., Grob, K., 2006. PVC plasticizers/additives migrating from the gaskets of metal closures into oily food: Swiss market survey June 2005. Eur. Food Res. Technol. 223 (4), 447–453. Farhoodi, M., Emam-Djomeh, Z., Ehsani, M.R., Oromiehie, A., 2008. Effect of environmental conditions on the migration of di(2-ethylhexyl)pthalate from pet bottles into yogurt drinks: influence of time, temperature, and food simulant. Arab. J. Sci. Eng. 33 (2B), 279–287. Fromme, H., Gruber, L., Schlummer, M., Wolz, G., Bohmer, S., Angerer, J., Mayer, R., Liebl, B., Bolte, G., 2007. Intake of phthalates and di(2-ethylhexyl)adipate: results of the Integrated Exposure Assessment Survey based on duplicate diet samples and biomonitoring data. Environ. Int. 33 (8), 1012–1020. Genuis, S.J., Beesoon, S., Lobo, R.A., Birkholz, D., 2012. Human elimination of phthalate compounds: blood, urine, and sweat (BUS) study. TheScientificWorldJOURNAL 2012, 615068. Gonzalez-Castro, M.I., Olea-Serrano, M.F., Rivas-Velasco, A.M., Medina-Rivero, E., Ordonez-Acevedo, L.G., De Leon-Rodriguez, A., 2011. Phthalates and bisphenols migration in Mexican food cans and plastic food containers. Bull. Environ. Contam. Toxicol. 86 (6), 627–631. Guo, Y., Kannan, K., 2011. Comparative assessment of human exposure to phthalate esters from house dust in China and the United States. Environ. Sci. Technol. 45 (8), 3788–3794. Guo, Y., Zhang, Z., Liu, L., Li, Y., Ren, N., Kannan, K., 2012. Occurrence and profiles of phthalates in foodstuffs from China and their implications for human exposure. J. Agric. Food Chem. Guo, Y., Wang, L., Kannan, K., 2013. Phthalates and parabens in personal care products from China: concentrations and human exposure. Arch. Environ. Contam. Toxicol. Hartle, J.C., Navas-Acien, A., Lawrence, R.S., 2016. The consumption of canned food and beverages and urinary Bisphenol A concentrations in NHANES 2003–2008. Environ. Res. 150, 375–382. Hatch, E.E., Nelson, J.W., Qureshi, M.M., Weinberg, J., Moore, L.L., Singer, M., Webster, T.F., 2008. Association of urinary phthalate metabolite concentrations with body mass index and waist circumference: a cross-sectional study of NHANES data, 1999– 2002. Environ. Health 7, 27. Heinemeyer, G., Sommerfeld, C., Springer, A., Heiland, A., Lindtner, O., Greiner, M., Heuer, T., Krems, C., Conrad, A., 2013. Estimation of dietary intake of bis(2ethylhexyl)phthalate (DEHP) by consumption of food in the German population. Int. J. Hyg. Environ. Health 216 (4), 472–480. Hernandez-Diaz, S., Mitchell, A.A., Kelley, K.E., Calafat, A.M., Hauser, R., 2009. Medications as a potential source of exposure to phthalates in the U.S. population. Environ. Health Perspect. 117 (2), 185–189. Hoppin, J.A., Brock, J.W., Davis, B.J., Baird, D.D., 2002. Reproducibility of urinary phthalate metabolites in first morning urine samples. Environ. Health Perspect. 110 (5), 515–518. Huang, Y., Lu, W.W., Chen, B., You, J., Wu, M., Li, S.G., 2014. Phthalates in commercial Chinese rice wines: concentrations and the cumulative risk assessment to adult males in Shanghai. Biomed. Environ. Sci. 27 (10), 819–823. Ji, K., Lim Kho, Y., Park, Y., Choi, K., 2010. Influence of a five-day vegetarian diet on urinary levels of antibiotics and phthalate metabolites: a pilot study with “Temple Stay” participants. Environ. Res. 110 (4), 375–382. Jo, A., Kim, H., Chung, H., Chang, N., 2016. Associations between dietary intake and urinary bisphenol A and phthalates levels in Korean women of reproductive age. Int. J. Environ. Res. Public Health 13 (7).

9

Kim, S., Hong, S.H., Bong, C.K., Cho, M.H., 2015. Characterization of air freshener emission: the potential health effects. J. Toxicol. Sci. 40 (5), 535–550. Koch, H.M., Bolt, H.M., Preuss, R., Angerer, J., 2005. New metabolites of di(2ethylhexyl)phthalate (DEHP) in human urine and serum after single oral doses of deuterium-labelled DEHP. Arch. Toxicol. 79 (7), 367–376. Koch, H.M., Lorber, M., Christensen, K.L., Palmke, C., Koslitz, S., Bruning, T., 2013. Identifying sources of phthalate exposure with human biomonitoring: results of a 48 h fasting study with urine collection and personal activity patterns. Int. J. Hyg. Environ. Health. Lind, P.M., Lind, L., 2011. Circulating levels of bisphenol A and phthalates are related to carotid atherosclerosis in the elderly. Atherosclerosis 218 (1), 207–213. Martine, B., Marie-Jeanne, T., Cendrine, D., Fabrice, A., Marc, C., 2012. Assessment of adult human exposure to phthalate esters in the Urban Centre of Paris (France). Bull. Environ. Contam. Toxicol. Martino-Andrade, A.J., Chahoud, I., 2010. Reproductive toxicity of phthalate esters. Mol. Nutr. Food Res. 54 (1), 148–157. Mervish, N., McGovern, K.J., Teitelbaum, S.L., Pinney, S.M., Windham, G.C., Biro, F.M., Kushi, L.H., Silva, M.J., Ye, X., Calafat, A.M., et al., 2014. Dietary predictors of urinary environmental biomarkers in young girls, BCERP, 2004–7. Environ. Res. 133, 12–19. Miodovnik, A., Engel, S.M., Zhu, C., Ye, X., Soorya, L.V., Silva, M.J., Calafat, A.M., Wolff, M.S., 2011. Endocrine disruptors and childhood social impairment. Neurotoxicology 32 (2), 261–267. Montuori, P., Jover, E., Morgantini, M., Bayona, J.M., Triassi, M., 2008. Assessing human exposure to phthalic acid and phthalate esters from mineral water stored in polyethylene terephthalate and glass bottles. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 25 (4), 511–518. National Toxicology P, 2003. NTP-CERHR monograph on the potential human reproductive and developmental effects of butyl benzyl phthalate (BBP). NTP CERHR MON 5 (i-III90). Petersen, J.H., Jensen, L.K., 2010. Phthalates and food-contact materials: enforcing the 2008 European Union plastics legislation. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 27 (11), 1608–1616. Sakhi, A.K., Lillegaard, I.T., Voorspoels, S., Carlsen, M.H., Loken, E.B., Brantsaeter, A.L., Haugen, M., Meltzer, H.M., Thomsen, C., 2014. Concentrations of phthalates and bisphenol A in Norwegian foods and beverages and estimated dietary exposure in adults. Environ. Int. 73, 259–269. Sathyanarayana, S., 2008. Phthalates and children's health. Curr. Probl. Pediatr. Adolesc. Health Care 38 (2), 34–49. Schecter, A., Lorber, M., Guo, Y., Wu, Q., Yun, S.H., Kannan, K., Hommel, M., Imran, N., Hynan, L.S., Cheng, D., et al., 2013. Phthalate concentrations and dietary exposure from food purchased in New York State. Environ. Health Perspect. 121 (4), 473–494 (494e471–474). Schettler, T., 2006. Human exposure to phthalates via consumer products. Int. J. Androl. 29 (1), 134–139 (discussion 181-135). Schmid, P., Kohler, M., Meierhofer, R., Luzi, S., Wegelin, M., 2008. Does the reuse of PET bottles during solar water disinfection pose a health risk due to the migration of plasticisers and other chemicals into the water? Water Res. 42 (20), 5054–5060. Serrano, S.E., Braun, J., Trasande, L., Dills, R., Sathyanarayana, S., 2014. Phthalates and diet: a review of the food monitoring and epidemiology data. Environ. Health 13 (1), 43. Sharman, M., Read, W.A., Castle, L., Gilbert, J., 1994. Levels of di-(2-ethylhexyl)phthalate and total phthalate esters in milk, cream, butter and cheese. Food Addit. Contam. 11 (3), 375–385. Shen, Q., Shi, H., Zhang, Y., Cao, Y., 2015. Dietary intake and phthalates body burden in boys and girls. Arch. Public health = Archives belges de sante publique 73 (1), 5. Singh, S., Li, S.S., 2012. Epigenetic effects of environmental chemicals bisphenol A and phthalates. Int. J. Mol. Sci. 13 (8), 10143–10153. Sioen, I., Fierens, T., Van Holderbeke, M., Geerts, L., Bellemans, M., De Maeyer, M., Servaes, K., Vanermen, G., Boon, P.E., De Henauw, S., 2012. Phthalates dietary exposure and food sources for Belgian preschool children and adults. Environ. Int. 48, 102–108. Tranfo, G., Papaleo, B., Caporossi, L., Capanna, S., De Rosa, M., Pigini, D., Corsetti, F., Paci, E., 2013. Urinary metabolite concentrations of phthalate metabolites in Central Italy healthy volunteers determined by a validated HPLC/MS/MS analytical method. Int. J. Hyg. Environ. Health 216 (4), 481–485. Trasande, L., Sathyanarayana, S., Jo Messito, M., SG, R., Attina, T.M., Mendelsohn, A.L., 2013. Phthalates and the diets of U.S. children and adolescents. Environ. Res. 126, 84–90. Wang, J., Li, X., Zhang, Q., Xiong, J., Li, H., 2015. Determination of phthalate esters in Chinese spirits using isotope dilution gas chromatography with tandem mass spectrometry. J. Sep. Sci. 38 (10), 1700–1710. Wittassek, M., Koch, H.M., Angerer, J., Bruning, T., 2011. Assessing exposure to phthalates - the human biomonitoring approach. Mol. Nutr. Food Res. 55 (1), 7–31. Wormuth, M., Scheringer, M., Vollenweider, M., Hungerbuhler, K., 2006. What are the sources of exposure to eight frequently used phthalic acid esters in Europeans? Risk Anal. 26 (3), 803–824. Zeng, Q., Wei, C., Wu, Y., Li, K., Ding, S., Yuan, J., Yang, X., Chen, M., 2013. Approach to distribution and accumulation of dibutyl phthalate in rats by immunoassay. Food Chem. Toxicol. 56, 18–27. Zhang, Y.H., Zheng, L.X., Chen, B.H., 2006. Phthalate exposure and human semen quality in Shanghai: a cross-sectional study. Biomed. Environ. Sci. 19 (3), 205–209. Zhou, J.H., Cao, X.Y., Lin, H.L., Ni, Z.H., He, Y.N., Chen, M.H., Zheng, H.G., Chen, X.M., 2015. Safety and effectiveness evaluation of a domestic peritoneal dialysis fluid packed in non-PVC bags: study protocol for a randomized controlled trial. Trials 16. Zota, A.R., Phillips, C.A., Mitro, S.D., 2016. Recent fast food consumption and bisphenol A and phthalates exposures among the U.S. population in NHANES, 2003–2010. Environ. Health Perspect.

Please cite this article as: Dong, R., et al., Food consumption survey of Shanghai adults in 2012 and its associations with phthalate metabolites in urine, Environ Int (2017), http://dx.doi.org/10.1016/j.envint.2017.01.008