Indoor Air 2015; 25: 572–581 wileyonlinelibrary.com/journal/ina Printed in Singapore. All rights reserved
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd INDOOR AIR doi:10.1111/ina.12179
Phthalate metabolites in urine samples from Beijing children and correlations with phthalate levels in their handwipes Abstract Little attention has been paid to dermal absorption of phthalates even though modeling suggests that this pathway may contribute meaningfully to total uptake. We have concurrently collected handwipe and urine samples from 39 Beijing children (5–9 years) for the purpose of measuring levels of five phthalates in handwipes, corresponding concentrations of eight of their metabolites in urine, and to subsequently assess the contribution of dermal absorption to total uptake. In summer sampling, DEHP was the most abundant phthalate in handwipes (median: 1130 lg/m2), while MnBP was the most abundant metabolite in urine (median: 232 ng/ml). We found significant associations between the parent phthalate in handwipes and its monoester metabolite in urine for DiBP (r = 0.41, P = 0.01), DnBP (r = 0.50, P = 0.002), BBzP (r = 0.48, P = 0.003), and DEHP (r = 0.36, P = 0.03). Assuming that no dermal uptake occurred under clothing-covered skin, we estimate that dermal absorption of DiBP, DnBP, BBzP, and DEHP contributed 6.9%, 4.6%, 6.9%, and 3.3%, respectively, to total uptake. Assuming that somewhat attenuated dermal uptake occurred under clothing-covered skin, these estimates increase to 19%, 14%, 17%, and 10%. The results indicate that absorption from skin surfaces makes a meaningful contribution to total phthalate uptake for children and should be considered in future risk assessments.
M. Gong1, C. J. Weschler1,2, L. Liu3, H. Shen3, L. Huang1, J. Sundell1, Y. Zhang1 1 Department of Building Science, Tsinghua University, Beijing, China, 2Environmental and Occupational Health Sciences Institute, Rutgers University, Piscataway, NJ, USA, 3Institute of Urban Environment, Key Lab of Urban Environment Health, Chinese Academy of Sciences, Xiamen, China
Key words: Indoor air quality; Contact transfer; Dermal absorption; Exposure pathway; Indoor exposure; Percutaneous absorption.
C. J. Weschler Environmental and Occupational Health Sciences Institute, Rutgers University, Piscataway, NJ 08854, USA Tel.: +01 848 445 2073 Fax: +01 732 445 0116 e-mail: [email protected] and Y. Zhang Department of Building Science, Tsinghua University Beijing 100084, China Tel.: +86 10 6277 2518 Fax: +86 10 6277 3461 e-mail: [email protected] Received for review 28 September 2014. Accepted for publication 21 December 2014.
Practical Implications
This study indicates that children’s hands acquire substantial amounts of various phthalates. The levels measured in handwipes, although higher, are somewhat representative of levels on other body locations. Via dermal absorption, as well as hand-to-mouth activity, phthalates on hands and other body locations contribute to the overall body burden of these compounds. Dermal absorption from air and contact transfer from surfaces is expected to occur for many SVOCs commonly found indoors (e.g. bisphenols, synthetic musks, organophosphates). However, the dermal pathway has often been neglected in exposure assessments of indoor pollutants. Knowledge regarding phthalates and other SVOCs in handwipes can facilitate our understanding of risks and aid in the mitigation of adverse health effects resulting from indoor exposures. To make progress toward these goals, further studies are necessary, including investigations of phthalate level variability in skinwipes collected at different locations on the body and the impact of clothing on dermal absorption from air.
Introduction
Over the past few decades, phthalate esters have been widely used in various industrial and consumer 572
applications. The long chain phthalates, such as di(2ethylhexyl) phthalate (DEHP), are primarily used as plasticizers in polyvinyl chloride (PVC) and other
Phthalates in handwipes; metabolites in urine polymers, while short chain phthalates, such as diethyl phthalate (DEP), di(isobutyl) phthalate (DiBP), and di(n-butyl) phthalate (DnBP), are often used as carriers in personal care products, paints, varnishes, and coatings (Dodson et al., 2012; Wittassek et al., 2011). Globally, 6650 kilotons of plasticizers were produced in 2006, 25% of which were consumed in China and 90% of which were phthalates (Wang et al., 2010). Certain phthalates are implicated in adverse health effects. For example, toxicological studies using animals have shown that exposures to some phthalates can induce disruption of reproductive development (Foster, 2006; National Research Council, 2008). Meanwhile, epidemiologic studies have found various relationships including associations between specific phthalates and abnormal male reproductive outcomes (Meeker et al., 2009), children’s neurodevelopmental issues (Whyatt et al., 2012), behavior issues (Braun et al., 2013), and asthma and allergies (Bornehag et al., 2004). Humans can be exposed to phthalates through inhalation; dermal absorption, including direct air-to-skin transport, contact with contaminated surfaces and use of personal care products (PCPs); and ingestion, including dietary ingestion and incidental ingestion. Phthalate metabolite concentrations in urine can serve as a measure of recent total exposure from multiple exposure pathways (Koch et al., 2005). Hence, many studies in Europe and the United States have reported phthalate metabolite concentrations in urine samples, with results indicating that humans are extensively exposed to phthalates (Boas et al., 2010; CDC, 2013; Koch et al., 2011; Langer et al., 2014). Children tend to have greater exposure than adults (Becker et al., 2009; CDC, 2013). However, despite the extensive use of phthalates in China, comparable measurements of phthalate metabolite concentrations in urine from individuals in China, especially children, are scarce. To date, there is only one such study in mainland China (Wang et al., 2013) and several studies in Taiwan (Hsu et al., 2012; Lin et al., 2011; Wang et al., 2014; Wu et al., 2013). Dietary ingestion appears to be the dominant exposure pathway for DEHP (Fromme et al., 2007; Itoh et al., 2007; Rudel et al., 2011; Shin et al., 2014), while for other common phthalates (e.g. DEP, DiBP, DnBP, and BBzP), other exposure pathways appear to make significant contributions (Adibi et al., 2008; Bek€ o et al., 2013; Langer et al., 2014). However, little attention has been paid to dermal absorption, even though some assessments have indicated that it may be important (Weschler and Nazaroff, 2012). Dermal application of personal care products was judged to be the major exposure pathway for DEP in an evaluation of sources of phthalate exposure among Europeans (Wormuth et al., 2006). Recent studies have suggested that dermal absorption from air may be important for
DEP, DiBP, and DnBP (Bek€ o et al., 2013; Gaspar et al., 2014; Gong et al., 2014b; Little et al., 2012; Weschler and Nazaroff, 2012, 2014). However, in these studies, the dermal absorption was indirectly estimated based on predicted or measured phthalate concentrations in dust or in air. It may be possible to use skin wipes to at least partially assess dermal absorption of phthalates. A recent study (Gong et al., 2014a) has estimated dermal absorption based on directly measured phthalate levels in skin wipes collected for adults; this study found that dermal absorption contributed significantly to total uptake. However, the comparisons depended on total uptakes derived for populations different from the group whose handwipes were sampled. We are aware of no study that has examined phthalate levels in children’s handwipes in China, even though children may have greater dermal exposure than adults as a consequence of crawling and objectto-hand activities (Stapleton et al., 2008). Additionally, we are aware of no study that has measured phthalate levels in handwipes and its metabolite concentrations in urine concurrently for the same study population and subsequently compared phthalate uptake through dermal absorption to total uptake on a person-byperson basis. The aims of this study were to: (1) determine the levels of five commonly used phthalates (DEP, DiBP, DnBP, BBzP, and DEHP) in handwipes and the concentrations of their metabolites (the monoesters MEP, MiBP, MnBP, MBzP, and MEHP, as well as additional oxidized metabolites of DEHP—MEHHP, MECPP, MEOHP) in urine from Beijing children, (2) examine the association between the phthalate levels in handwipes and the concentrations of their metabolites in urine, and (3) estimate the contribution of phthalate uptake through dermal absorption to total uptake. This investigation was part of a pilot field program for the Beijing component of the China Child Home Health (CCHH) study (Qu et al., 2013; Zhang et al., 2013).
Methods Study design
We recruited subjects from 39 households that participated in the questionnaire component of the CCHH study in Beijing. The children who participated in the study were 5–9 years old; 28 were boys and 11 were girls. We conducted field sampling twice at each home—once during the ‘summer’ (July to September 2013) and once during the ‘winter’ (December to April 2014). Basic personal information, for example, height and weight, were reported by the parents and recorded. Fudan University’s Ethical Review Board approved the study protocol prior to collection of all handwipe and urine samples (IRB00002408 and 573
Gong et al. FWA00002399), and all the parents or grandparents of the children gave their informed consent. Handwipe samples
Handwipe samples were collected when children were at home and at least 60 min after their last hand washing. In total, 38 handwipe samples were obtained during the summer and 15 during the winter periods. Handwipe samples were collected by the investigators using gauze pads. The wipe procedure was similar to the process described in a US Environment Protection Agency (EPA) study (Morgan et al., 2004) and our previous study (Gong et al., 2014a). Briefly, a pre-cleaned gauze pad (8 cm 9 8 cm) was wetted with 5 ml isopropanol and then used to wipe the hand surface. The handwipe sample was stored in a 60 ml pre-cleaned brown glass jar at 36°C until analysis. Palm and back-of-hand were sampled and analyzed separately. A field blank wipe was paired for each child by soaking a pre-cleaned gauze pad in isopropyl alcohol and placing it directly into a brown glass jar. Handwipe samples were analyzed for five phthalates by GC/EI-MS, using methods described previously (Gong et al., 2014a). Urine samples
Parents were asked to collect their children’s first morning urine in a 60 ml pre-cleaned brown glass jar on the day after handwipe sampling and then store the sample in a freezer before the investigators’ visit. In total, 37 urine samples were collected during summer and 30 during winter periods. After transporting the urine samples to the laboratory on ice, they were stored at 36°C until analysis. Eight phthalate metabolites were analyzed using a method modified from Silva et al. (2007) and described in our previous study (Liu et al., 2012). Briefly, an aliquot (1.0 ml) of urine sample was deconjugated using Escherichia coli b-glucuronidase, purified by solid-phase extraction, and analyzed by isotopic dilution HPLC/ESI-MS-MS. Urine specific gravity (SG) was measured by a handheld refractometer that was calibrated with distilled water between each measurement. Dermal absorption/total uptake estimation
The daily dermal absorption (DAdermal, lg/kg/day) of phthalates for each child was estimated using the following equation: kp DAdermal ¼
l
P
Cl;i SAi texp
i
X BW
ð1Þ
where kp_l is the skin permeability coefficient (lm/h) from skin surface lipids to dermal capillaries for 574
partially hydrated stratum corneum; the values derived in our previous study are used (0.076, 0.074, 0.04, and 0.004 lm/h for DiBP, DnBP, BBzP, and DEHP, respectively) (Gong et al., 2014a). Cl,i is the phthalate level in skin surface lipids at different exposed locations (lg/m2). We assume that in summer, the head, neck, hands, forearms, and lower legs of the children are bare skin locations, while in winter, only the head, neck, and hands are bare. The phthalate levels on the forearms and lower legs are assumed to be 1/2 of the levels on the back-of-hand, while the levels on head and neck are assumed to be 1/3 of the levels on the back-of-hand based on our previous analysis (Gong et al., 2014a). Conservatively, we assume that skin locations that are covered by clothing make no contribution to dermal absorption. Less conservatively, we somewhat arbitrarily assume that the phthalate levels on the clothing-covered body locations equal 1/2 of that on the back-of-hand (see 2nd paragraph of subsection Dermal Absorption/Total Uptake). SAi is the skin surface area at different exposed locations (m2), calculated using the proportion of children’s total surface area by body part listed in Table 7–8 of the USEPA exposure factors handbook (US EPA, 2011); the whole body surface area is calculated from the child’s weight and height using the formula of Du Bois and Du Bois (1916). texp is the exposed time per day (h/day), assumed to be 24 h/day. X is the thickness of the skin surface lipids (lm), assumed to be 0.88 lm (NazzaroPorro et al., 1979). The total daily uptake (DUtotal, lg/kg/day) of phthalates for each child was estimated using the following equation: DUtotal ¼
Cu Vu MWp Fue BW MWm
ð2Þ
where Cu is the specific gravity adjusted phthalate metabolite concentration in urine (ng/ml), calculated by the formula Cu = P[(SGm 1)/(SG 1)], where P is the observed phthalate metabolite concentration, SGm is the median SG value in the study population, and SG is the specific gravity of each urine sample; Vu is the daily excreted urine volume (l/day); Fue is the urinary excretion factor (dimensionless); BW is the child’s body weight (kg); MWp and MWm are the molecular weights of the parent phthalate and its metabolite (g/mol). Total daily uptake of DEHP was calculated as the average uptake derived from Equation (2) using MEOHP and MECPP concentrations, judged to be the most accurate among the measured metabolites (Supporting Information, subsection ‘Metabolite measurements . . .’). Vu is assumed to be 0.66 l/day (Perucca et al., 2007). Fue for DiBP is 0.71 from Koch et al. (2012), while values for other phthalates are from Table 1 of Wittassek et al. (2011).
Phthalates in handwipes; metabolites in urine Data analysis
Handwipe samples were corrected by subtracting the average mass in the field blanks. DEP, DnBP, DiBP, and DEHP were detected in field blanks and averaged 0.04 0.03, 0.29 0.08, 0.32 0.16, and 1.2 0.73 lg (mean s.d.), respectively. Phthalate metabolites were rarely detected in the procedural blanks, with the exception of MiBP and MnBP at very low concentrations; hence, urine samples were not blank-corrected. The method detection limits (MDLs) for handwipe samples were determined as three times the standard deviation of the field blanks, and the laboratory instrument detection limit (i.e. signal-to-noise ratio of three) was used to calculate the MDL when phthalates were not detected in the field blank. For DEP, DiBP, DnBP, BBzP, and DEHP, the MDLs are 0.09, 0.24, 0.48, 0.005, and 2.1 lg, respectively. The MDLs for urine samples were set as the laboratory instrument detection limit, which are 0.52 (MEP), 0.53 (MiBP), 0.45 (MnBP), 0.31 (MBzP), 0.27 (MEHP), 0.21 (MEOHP), 0.16 (MECPP), and 0.10 (MEHHP) ng/ml. For all samples, concentrations below the MDL were substituted with one-half the MDL. Statistical analyses were performed using the SPSS statistics software package, V. 17.0 (Shanghai, China), with statistical significance defined at the P = 0.05 level. Results Handwipes
Table 1 presents the levels of the targeted phthalates in handwipes for both the summer and winter periods. DEP is not included because its detection frequency was 0.05) while the BBzP levels were not, possibly due to its lower detection frequency. There was no significant difference between boys and girls for phthalate levels in handwipes. Levels in summer tended to be lower than levels in winter, with the median summer and winter levels of
DiBP and DEHP statistically different. DiBP, DnBP, and DEHP in handwipes from summer were significantly correlated with each other, suggesting common sources (Table S1). Urine
Table 2 presents the unadjusted and specific gravity (SG) adjusted concentrations of the targeted phthalate metabolites in urine from both the summer and winter periods. All phthalate metabolites except MBzP were detected in all samples. MnBP was the most abundant, followed by the sum of the DEHP metabolites (MEHP, MEOHP, MECPP, and MEHHP) and MiBP. The Shapiro–Wilk test showed that the concentrations of all targeted metabolites except MBzP were log-normally distributed (P > 0.05). There was no significant difference for phthalate metabolite concentrations between boys and girls. As shown in Supporting Information, Table S2, the metabolites of DEHP were significantly correlated with each other, but the correlation between the secondary metabolites was stronger than the correlation between the primary and secondary metabolites, perhaps due to the different half-lives of the primary and secondary metabolites. MiBP and MnBP were also significantly correlated with each other. The concentrations of MEP, MiBP, and MnBP in summer were significantly higher than those in winter, while the concentrations of the DEHP metabolites were not significantly different, with the exception of MEHHP (higher in winter). Handwipe/urine associations
Table 3 presents the Spearman correlation coefficients for the phthalate levels in handwipes and their metabolite concentrations in paired urine samples from both the summer and winter periods. For summer samples, statistically significant positive correlations were observed between DiBP and MiBP (r = 0.41, P = 0.01), DnBP and MnBP (r = 0.50, P = 0.002), BBzP and MBzP (r = 0.48, P = 0.003), and DEHP and MEHP (r = 0.36, P = 0.03). For winter samples, a statistically significant positive correlation was observed between DnBP and MnBP (r = 0.71, P = 0.003), while for DiBP and MiBP, the correlation coefficient was
Table 1 Levels (lg/m2) of phthalates in handwipes from Beijing children taken during the two sampling periods (summer/winter; n = 38/15) Phthalate
%Detect
Mean
Min
50% tile
75% tile
95% tile
Max
DiBP DnBP BBzP DEHP
97/100 89/93 64/33 100/100
57.1/68.9 99.8/203 1.2/1.0 1940/1918
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd INDOOR AIR doi:10.1111/ina.12179
Phthalate metabolites in urine samples from Beijing children and correlations with phthalate levels in their handwipes Abstract Little attention has been paid to dermal absorption of phthalates even though modeling suggests that this pathway may contribute meaningfully to total uptake. We have concurrently collected handwipe and urine samples from 39 Beijing children (5–9 years) for the purpose of measuring levels of five phthalates in handwipes, corresponding concentrations of eight of their metabolites in urine, and to subsequently assess the contribution of dermal absorption to total uptake. In summer sampling, DEHP was the most abundant phthalate in handwipes (median: 1130 lg/m2), while MnBP was the most abundant metabolite in urine (median: 232 ng/ml). We found significant associations between the parent phthalate in handwipes and its monoester metabolite in urine for DiBP (r = 0.41, P = 0.01), DnBP (r = 0.50, P = 0.002), BBzP (r = 0.48, P = 0.003), and DEHP (r = 0.36, P = 0.03). Assuming that no dermal uptake occurred under clothing-covered skin, we estimate that dermal absorption of DiBP, DnBP, BBzP, and DEHP contributed 6.9%, 4.6%, 6.9%, and 3.3%, respectively, to total uptake. Assuming that somewhat attenuated dermal uptake occurred under clothing-covered skin, these estimates increase to 19%, 14%, 17%, and 10%. The results indicate that absorption from skin surfaces makes a meaningful contribution to total phthalate uptake for children and should be considered in future risk assessments.
M. Gong1, C. J. Weschler1,2, L. Liu3, H. Shen3, L. Huang1, J. Sundell1, Y. Zhang1 1 Department of Building Science, Tsinghua University, Beijing, China, 2Environmental and Occupational Health Sciences Institute, Rutgers University, Piscataway, NJ, USA, 3Institute of Urban Environment, Key Lab of Urban Environment Health, Chinese Academy of Sciences, Xiamen, China
Key words: Indoor air quality; Contact transfer; Dermal absorption; Exposure pathway; Indoor exposure; Percutaneous absorption.
C. J. Weschler Environmental and Occupational Health Sciences Institute, Rutgers University, Piscataway, NJ 08854, USA Tel.: +01 848 445 2073 Fax: +01 732 445 0116 e-mail: [email protected] and Y. Zhang Department of Building Science, Tsinghua University Beijing 100084, China Tel.: +86 10 6277 2518 Fax: +86 10 6277 3461 e-mail: [email protected] Received for review 28 September 2014. Accepted for publication 21 December 2014.
Practical Implications
This study indicates that children’s hands acquire substantial amounts of various phthalates. The levels measured in handwipes, although higher, are somewhat representative of levels on other body locations. Via dermal absorption, as well as hand-to-mouth activity, phthalates on hands and other body locations contribute to the overall body burden of these compounds. Dermal absorption from air and contact transfer from surfaces is expected to occur for many SVOCs commonly found indoors (e.g. bisphenols, synthetic musks, organophosphates). However, the dermal pathway has often been neglected in exposure assessments of indoor pollutants. Knowledge regarding phthalates and other SVOCs in handwipes can facilitate our understanding of risks and aid in the mitigation of adverse health effects resulting from indoor exposures. To make progress toward these goals, further studies are necessary, including investigations of phthalate level variability in skinwipes collected at different locations on the body and the impact of clothing on dermal absorption from air.
Introduction
Over the past few decades, phthalate esters have been widely used in various industrial and consumer 572
applications. The long chain phthalates, such as di(2ethylhexyl) phthalate (DEHP), are primarily used as plasticizers in polyvinyl chloride (PVC) and other
Phthalates in handwipes; metabolites in urine polymers, while short chain phthalates, such as diethyl phthalate (DEP), di(isobutyl) phthalate (DiBP), and di(n-butyl) phthalate (DnBP), are often used as carriers in personal care products, paints, varnishes, and coatings (Dodson et al., 2012; Wittassek et al., 2011). Globally, 6650 kilotons of plasticizers were produced in 2006, 25% of which were consumed in China and 90% of which were phthalates (Wang et al., 2010). Certain phthalates are implicated in adverse health effects. For example, toxicological studies using animals have shown that exposures to some phthalates can induce disruption of reproductive development (Foster, 2006; National Research Council, 2008). Meanwhile, epidemiologic studies have found various relationships including associations between specific phthalates and abnormal male reproductive outcomes (Meeker et al., 2009), children’s neurodevelopmental issues (Whyatt et al., 2012), behavior issues (Braun et al., 2013), and asthma and allergies (Bornehag et al., 2004). Humans can be exposed to phthalates through inhalation; dermal absorption, including direct air-to-skin transport, contact with contaminated surfaces and use of personal care products (PCPs); and ingestion, including dietary ingestion and incidental ingestion. Phthalate metabolite concentrations in urine can serve as a measure of recent total exposure from multiple exposure pathways (Koch et al., 2005). Hence, many studies in Europe and the United States have reported phthalate metabolite concentrations in urine samples, with results indicating that humans are extensively exposed to phthalates (Boas et al., 2010; CDC, 2013; Koch et al., 2011; Langer et al., 2014). Children tend to have greater exposure than adults (Becker et al., 2009; CDC, 2013). However, despite the extensive use of phthalates in China, comparable measurements of phthalate metabolite concentrations in urine from individuals in China, especially children, are scarce. To date, there is only one such study in mainland China (Wang et al., 2013) and several studies in Taiwan (Hsu et al., 2012; Lin et al., 2011; Wang et al., 2014; Wu et al., 2013). Dietary ingestion appears to be the dominant exposure pathway for DEHP (Fromme et al., 2007; Itoh et al., 2007; Rudel et al., 2011; Shin et al., 2014), while for other common phthalates (e.g. DEP, DiBP, DnBP, and BBzP), other exposure pathways appear to make significant contributions (Adibi et al., 2008; Bek€ o et al., 2013; Langer et al., 2014). However, little attention has been paid to dermal absorption, even though some assessments have indicated that it may be important (Weschler and Nazaroff, 2012). Dermal application of personal care products was judged to be the major exposure pathway for DEP in an evaluation of sources of phthalate exposure among Europeans (Wormuth et al., 2006). Recent studies have suggested that dermal absorption from air may be important for
DEP, DiBP, and DnBP (Bek€ o et al., 2013; Gaspar et al., 2014; Gong et al., 2014b; Little et al., 2012; Weschler and Nazaroff, 2012, 2014). However, in these studies, the dermal absorption was indirectly estimated based on predicted or measured phthalate concentrations in dust or in air. It may be possible to use skin wipes to at least partially assess dermal absorption of phthalates. A recent study (Gong et al., 2014a) has estimated dermal absorption based on directly measured phthalate levels in skin wipes collected for adults; this study found that dermal absorption contributed significantly to total uptake. However, the comparisons depended on total uptakes derived for populations different from the group whose handwipes were sampled. We are aware of no study that has examined phthalate levels in children’s handwipes in China, even though children may have greater dermal exposure than adults as a consequence of crawling and objectto-hand activities (Stapleton et al., 2008). Additionally, we are aware of no study that has measured phthalate levels in handwipes and its metabolite concentrations in urine concurrently for the same study population and subsequently compared phthalate uptake through dermal absorption to total uptake on a person-byperson basis. The aims of this study were to: (1) determine the levels of five commonly used phthalates (DEP, DiBP, DnBP, BBzP, and DEHP) in handwipes and the concentrations of their metabolites (the monoesters MEP, MiBP, MnBP, MBzP, and MEHP, as well as additional oxidized metabolites of DEHP—MEHHP, MECPP, MEOHP) in urine from Beijing children, (2) examine the association between the phthalate levels in handwipes and the concentrations of their metabolites in urine, and (3) estimate the contribution of phthalate uptake through dermal absorption to total uptake. This investigation was part of a pilot field program for the Beijing component of the China Child Home Health (CCHH) study (Qu et al., 2013; Zhang et al., 2013).
Methods Study design
We recruited subjects from 39 households that participated in the questionnaire component of the CCHH study in Beijing. The children who participated in the study were 5–9 years old; 28 were boys and 11 were girls. We conducted field sampling twice at each home—once during the ‘summer’ (July to September 2013) and once during the ‘winter’ (December to April 2014). Basic personal information, for example, height and weight, were reported by the parents and recorded. Fudan University’s Ethical Review Board approved the study protocol prior to collection of all handwipe and urine samples (IRB00002408 and 573
Gong et al. FWA00002399), and all the parents or grandparents of the children gave their informed consent. Handwipe samples
Handwipe samples were collected when children were at home and at least 60 min after their last hand washing. In total, 38 handwipe samples were obtained during the summer and 15 during the winter periods. Handwipe samples were collected by the investigators using gauze pads. The wipe procedure was similar to the process described in a US Environment Protection Agency (EPA) study (Morgan et al., 2004) and our previous study (Gong et al., 2014a). Briefly, a pre-cleaned gauze pad (8 cm 9 8 cm) was wetted with 5 ml isopropanol and then used to wipe the hand surface. The handwipe sample was stored in a 60 ml pre-cleaned brown glass jar at 36°C until analysis. Palm and back-of-hand were sampled and analyzed separately. A field blank wipe was paired for each child by soaking a pre-cleaned gauze pad in isopropyl alcohol and placing it directly into a brown glass jar. Handwipe samples were analyzed for five phthalates by GC/EI-MS, using methods described previously (Gong et al., 2014a). Urine samples
Parents were asked to collect their children’s first morning urine in a 60 ml pre-cleaned brown glass jar on the day after handwipe sampling and then store the sample in a freezer before the investigators’ visit. In total, 37 urine samples were collected during summer and 30 during winter periods. After transporting the urine samples to the laboratory on ice, they were stored at 36°C until analysis. Eight phthalate metabolites were analyzed using a method modified from Silva et al. (2007) and described in our previous study (Liu et al., 2012). Briefly, an aliquot (1.0 ml) of urine sample was deconjugated using Escherichia coli b-glucuronidase, purified by solid-phase extraction, and analyzed by isotopic dilution HPLC/ESI-MS-MS. Urine specific gravity (SG) was measured by a handheld refractometer that was calibrated with distilled water between each measurement. Dermal absorption/total uptake estimation
The daily dermal absorption (DAdermal, lg/kg/day) of phthalates for each child was estimated using the following equation: kp DAdermal ¼
l
P
Cl;i SAi texp
i
X BW
ð1Þ
where kp_l is the skin permeability coefficient (lm/h) from skin surface lipids to dermal capillaries for 574
partially hydrated stratum corneum; the values derived in our previous study are used (0.076, 0.074, 0.04, and 0.004 lm/h for DiBP, DnBP, BBzP, and DEHP, respectively) (Gong et al., 2014a). Cl,i is the phthalate level in skin surface lipids at different exposed locations (lg/m2). We assume that in summer, the head, neck, hands, forearms, and lower legs of the children are bare skin locations, while in winter, only the head, neck, and hands are bare. The phthalate levels on the forearms and lower legs are assumed to be 1/2 of the levels on the back-of-hand, while the levels on head and neck are assumed to be 1/3 of the levels on the back-of-hand based on our previous analysis (Gong et al., 2014a). Conservatively, we assume that skin locations that are covered by clothing make no contribution to dermal absorption. Less conservatively, we somewhat arbitrarily assume that the phthalate levels on the clothing-covered body locations equal 1/2 of that on the back-of-hand (see 2nd paragraph of subsection Dermal Absorption/Total Uptake). SAi is the skin surface area at different exposed locations (m2), calculated using the proportion of children’s total surface area by body part listed in Table 7–8 of the USEPA exposure factors handbook (US EPA, 2011); the whole body surface area is calculated from the child’s weight and height using the formula of Du Bois and Du Bois (1916). texp is the exposed time per day (h/day), assumed to be 24 h/day. X is the thickness of the skin surface lipids (lm), assumed to be 0.88 lm (NazzaroPorro et al., 1979). The total daily uptake (DUtotal, lg/kg/day) of phthalates for each child was estimated using the following equation: DUtotal ¼
Cu Vu MWp Fue BW MWm
ð2Þ
where Cu is the specific gravity adjusted phthalate metabolite concentration in urine (ng/ml), calculated by the formula Cu = P[(SGm 1)/(SG 1)], where P is the observed phthalate metabolite concentration, SGm is the median SG value in the study population, and SG is the specific gravity of each urine sample; Vu is the daily excreted urine volume (l/day); Fue is the urinary excretion factor (dimensionless); BW is the child’s body weight (kg); MWp and MWm are the molecular weights of the parent phthalate and its metabolite (g/mol). Total daily uptake of DEHP was calculated as the average uptake derived from Equation (2) using MEOHP and MECPP concentrations, judged to be the most accurate among the measured metabolites (Supporting Information, subsection ‘Metabolite measurements . . .’). Vu is assumed to be 0.66 l/day (Perucca et al., 2007). Fue for DiBP is 0.71 from Koch et al. (2012), while values for other phthalates are from Table 1 of Wittassek et al. (2011).
Phthalates in handwipes; metabolites in urine Data analysis
Handwipe samples were corrected by subtracting the average mass in the field blanks. DEP, DnBP, DiBP, and DEHP were detected in field blanks and averaged 0.04 0.03, 0.29 0.08, 0.32 0.16, and 1.2 0.73 lg (mean s.d.), respectively. Phthalate metabolites were rarely detected in the procedural blanks, with the exception of MiBP and MnBP at very low concentrations; hence, urine samples were not blank-corrected. The method detection limits (MDLs) for handwipe samples were determined as three times the standard deviation of the field blanks, and the laboratory instrument detection limit (i.e. signal-to-noise ratio of three) was used to calculate the MDL when phthalates were not detected in the field blank. For DEP, DiBP, DnBP, BBzP, and DEHP, the MDLs are 0.09, 0.24, 0.48, 0.005, and 2.1 lg, respectively. The MDLs for urine samples were set as the laboratory instrument detection limit, which are 0.52 (MEP), 0.53 (MiBP), 0.45 (MnBP), 0.31 (MBzP), 0.27 (MEHP), 0.21 (MEOHP), 0.16 (MECPP), and 0.10 (MEHHP) ng/ml. For all samples, concentrations below the MDL were substituted with one-half the MDL. Statistical analyses were performed using the SPSS statistics software package, V. 17.0 (Shanghai, China), with statistical significance defined at the P = 0.05 level. Results Handwipes
Table 1 presents the levels of the targeted phthalates in handwipes for both the summer and winter periods. DEP is not included because its detection frequency was 0.05) while the BBzP levels were not, possibly due to its lower detection frequency. There was no significant difference between boys and girls for phthalate levels in handwipes. Levels in summer tended to be lower than levels in winter, with the median summer and winter levels of
DiBP and DEHP statistically different. DiBP, DnBP, and DEHP in handwipes from summer were significantly correlated with each other, suggesting common sources (Table S1). Urine
Table 2 presents the unadjusted and specific gravity (SG) adjusted concentrations of the targeted phthalate metabolites in urine from both the summer and winter periods. All phthalate metabolites except MBzP were detected in all samples. MnBP was the most abundant, followed by the sum of the DEHP metabolites (MEHP, MEOHP, MECPP, and MEHHP) and MiBP. The Shapiro–Wilk test showed that the concentrations of all targeted metabolites except MBzP were log-normally distributed (P > 0.05). There was no significant difference for phthalate metabolite concentrations between boys and girls. As shown in Supporting Information, Table S2, the metabolites of DEHP were significantly correlated with each other, but the correlation between the secondary metabolites was stronger than the correlation between the primary and secondary metabolites, perhaps due to the different half-lives of the primary and secondary metabolites. MiBP and MnBP were also significantly correlated with each other. The concentrations of MEP, MiBP, and MnBP in summer were significantly higher than those in winter, while the concentrations of the DEHP metabolites were not significantly different, with the exception of MEHHP (higher in winter). Handwipe/urine associations
Table 3 presents the Spearman correlation coefficients for the phthalate levels in handwipes and their metabolite concentrations in paired urine samples from both the summer and winter periods. For summer samples, statistically significant positive correlations were observed between DiBP and MiBP (r = 0.41, P = 0.01), DnBP and MnBP (r = 0.50, P = 0.002), BBzP and MBzP (r = 0.48, P = 0.003), and DEHP and MEHP (r = 0.36, P = 0.03). For winter samples, a statistically significant positive correlation was observed between DnBP and MnBP (r = 0.71, P = 0.003), while for DiBP and MiBP, the correlation coefficient was
Table 1 Levels (lg/m2) of phthalates in handwipes from Beijing children taken during the two sampling periods (summer/winter; n = 38/15) Phthalate
%Detect
Mean
Min
50% tile
75% tile
95% tile
Max
DiBP DnBP BBzP DEHP
97/100 89/93 64/33 100/100
57.1/68.9 99.8/203 1.2/1.0 1940/1918
