Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Time-trends in human urinary concentrations of phthalates and substitutes DEHT and DINCH in Asian and North American countries (2009–2019)

Time-trends in human urinary concentrations of phthalates and substitutes DEHT and DINCH in Asian... Journal of Exposure Science & Environmental Epidemiology www.nature.com/jes ARTICLE OPEN Time-trends in human urinary concentrations of phthalates and substitutes DEHT and DINCH in Asian and North American countries (2009–2019) 1✉ 1 1 2 3,4 3,4,5 Elena Domínguez-Romero , Klára Komprdová ,Jiří Kalina , Jos Bessems , Spyros Karakitsios , Dimosthenis A. Sarigiannis and Martin Scheringer © The Author(s) 2022 BACKGROUND: Many phthalates are environmental pollutants and toxic to humans. Following phthalate regulations, human exposure to phthalates has globally decreased with time in European countries, the US and Korea. Conversely, exposure to their substitutes DEHT and/or DINCH has increased. In other countries, including China, little is known on the time-trends in human exposure to these plasticizers. OBJECTIVE: We aimed to estimate time-trends in the urinary concentrations of phthalates, DEHT, and DINCH metabolites, in general population from non-European countries, in the last decade. METHODS: We compiled human biomonitoring (HBM) data from 123 studies worldwide in a database termed “PhthaLit”. We analyzed time-trends in the urinary concentrations of the excreted metabolites of various phthalates as well as DEHT and DINCH per metabolite, age group, and country/region, in 2009–2019. Additionally, we compared urinary metabolites levels between continents. RESULTS: We found solid time-trends in adults and/or children from the US, Canada, China and Taiwan. DEHP metabolites decreased in the US and Canada. Conversely in Asia, 5oxo- and 5OH-MEHP (DEHP metabolites) increased in Chinese children. For low-weight phthalates, the trends showed a mixed picture between metabolites and countries. Notably, MnBP (a DnBP metabolite) increased in China. The phthalate substitutes DEHT and DINCH markedly increased in the US. SIGNIFICANCE: We addressed the major question of time-trends in human exposure to phthalates and their substitutes and compared the results in different countries worldwide. IMPACT: Phthalates account for more than 50% of the plasticizer world market. Because of their toxicity, some phthalates have been regulated. In turn, the consumption of non-phthalate substitutes, such as DEHT and DINCH, is growing. Currently, phthalates and their substitutes show high detection percentages in human urine. Concerning time-trends, several studies, mainly in Europe, show a global decrease in phthalate exposure, and an increase in the exposure to phthalate substitutes in the last decade. In this study, we address the important question of time-trends in human exposure to phthalates and their substitutes and compare the results in different countries worldwide. Keywords: Phthalate plasticizers; Phthalate substitutes; Time-trends; Human biomonitoring; Asia; North America. Journal of Exposure Science & Environmental Epidemiology; https://doi.org/10.1038/s41370-022-00441-w INTRODUCTION low molecular weight phthalates (e.g. diethyl phthalate, DEP; di- Phthalates are the most widely used plasticizers worldwide [1, 2]. isobutyl phthalate, DiBP; and di-n-butyl phthalate, DnBP) are used Between countries, the consumption of plasticizers is led by as solvents and/or plasticizers in adhesives, paints, lacquers, China (approximately 50% of the world consumption in 2020), printing inks, and personal care products [2, 3]. In the plastics, followed by other Asia/Pacific countries, Europe and North phthalates are not chemically bound and can migrate to the America [1]. Phthalates and major phthalate replacements such environment, thus becoming environmental pollutants. From the as 1,2-cyclohexanedicarboxylic acid, 1,2-diisononyl ester (DINCH) environment, humans are exposed to these substances through and di(2-ethylhexyl) terephthalate (DEHT) are principally used as different exposure routes (oral, dermal, inhalation) and sources plasticizers in polymers, principally in flexible PVC. Furthermore, [4–6]. Exposure to phthalates, depending on factors such as the 1 2 3 RECETOX, Faculty of Science, Masaryk University, Kotlarska 2, Brno 611 37, Czech Republic. VITO (Flemish Institute for Technological Research), BE-2400 Mol, Belgium. Aristotle Univ Thessaloniki, Dept Chem Engn, Environm Engn Lab, Univ Campus,Bldg D,Rm 201, Thessaloniki 54124, Greece. HERACLES Res Ctr Exposome & Hlth, Ctr Interdisciplinary Res & Innovat, Balkan Ctr, Bldg B,10thkm Thessaloniki Thermi Rd, Thessaloniki 57001, Greece. Sch Adv Study IUSS, Sci Technol & Soc Dept, Environm Hlth Engn, Piazza Vittoria 15, I-27100 Pavia, Italy. email: elena.dominguez-romero@recetox.muni.cz Received: 8 October 2021 Revised: 1 April 2022 Accepted: 6 April 2022 1234567890();,: E. Domínguez-Romero et al. exposure level and duration, may result in toxic outcomes in We found significant time-trends for all the main phthalates in humans and animals. Notably, toxic effects of phthalates on the China, Taiwan, Korea, the US, and Canada, and for DEHT and development and function of male and female reproductive DINCH metabolites in the US. systems [7–9], neurodevelopment [10–12], and metabolism [13] have been reported. Phthalate replacements such as DINCH and DEHT are also toxic [14–16]. Considering the toxicological MATERIAL AND METHODS properties of these substances, some phthalate uses have been Literature research We aimed to conduct a non-biased literature review to prepare a human regulated in different countries [17]. Importantly in the EU, biomonitoring (HBM) data compilation. In this sense, our literature research fourteen phthalates have been classified as substances of very was in agreement with general recommendations for time-trend reviews high concern (SVHC) in the Regulation (EC) No 1907/2006 on the [38]. We conducted the literature research on the Web of Science. The Registration, Evaluation, Authorisation and Restriction of Chemi- dates, research keywords, exclusion criteria, and results of the literature cals (REACH), Annex XIV, by reason of their reproductive toxicity research are summarized in the first Supplementary Information file (SI 1), [18]. For each of these substances, a “sunset date” has been Table S1. established, after which its placing on the market and use must be prohibited in general. Phthalates listed in Annex XIV include “PhthaLit” database structure notably: di(2-ethylhexyl) phthalate (DEHP), butyl benzyl phthalate We created the “PhthaLit” database as a detailed compilation of literature (BBP), DnBP, and DiBP, for which the sunset date was in February HBM data for phthalates, DEHT, and DINCH in the general population from 2015. For these four substances (DEHP, DBP, BBP and DIBP), their different countries worldwide. The main columns in the database included endocrine disrupting properties were also included in REACH, variables considered of interest for the analysis of HBM data for phthalates. Annex XIV, on November 23, 2021 [19, 20]. Moreover, dioctyl These variables are described below. General variables for traceability phthalate (DOP), diisononyl phtalate (DINP), and diisodecyl (reference, specific biomonitoring plan, and study identification number) are described in the section entitled: Data traceability in the “PhthaLit” phthalate (DIDP) have been included in REACH, Annex XVII [21], database. and are subject to specific restrictions on the manufacture, placing on the market, and use. Substances and abbreviations. In the literature, a diversity of definitions Phthalate regulations help to reduce environmental as well as and abbreviations exist for phthalates and their metabolites. The main human exposure to these substances. Nonetheless, phthalates are abbreviations and definitions found in the literature for these substances presently still ubiquitous environmental pollutants. For example, and their metabolites are shown in SI 1, Table S2. We here used one phthalates have been found at high detection percentages (> abbreviation per substance. Among phthalate replacements, our priority 90%) in urine from the general population in Asian, European, and was for DEHT and DINCH [39]. North American countries [22–24]. Considering the relatively short half-lives of this group of substances in humans, generally lower Phthalate concentrations in urine. In the current manuscript, we work on than 24-h in blood and urine [25], the high levels of detection in the central values of urinary concentrations of phthalates and phthalate different countries are the result of recent exposure to these replacements. As central values, for publications where medians and pollutants. Nevertheless, human exposure to phthalates seems to geometric means (GM) were reported, both values were compiled, due to be globally decreasing. In this sense, the urinary concentrations of the generally lognormal distribution of urinary phthalate concentrations metabolites formed from low-weight phthalates and DEHP have [23, 27, 34]. However, only one of these metrics (priority for median as a more robust estimate of central tendency, followed by GM) per source was decreased in the latest years in Germany [26, 27], Belgium [28, 29], used for data analysis. In the current manuscript, the number of central Denmark [23], Sweden [30], Italy [31], the United States [32], values that we used in our analyses is identified by the abbreviation “n”. Canada [33], and Korea [34]. Among high-weight phthalates, the This abbreviation should not be confounded with “N”, which refers to the urinary concentrations of DINP metabolites have diminished in study-specific sample size (cf. Sample size (N)). Denmark [23] and Germany [27], and increased in Sweden [30]. Concerning the units, unadjusted urinary concentrations (µg/L) were Concerning phthalate replacements, metabolites from DEHT and/ the most frequent data found in the literature, followed by data or DINCH increased in Denmark [23], Germany [35, 36], and the US corrected for urinary dilution, in µg/g creatinine. Although we compiled [32]. Importantly, new time-trend analyses for European countries data in the different units, we work on unadjusted urinary concentra- are conducted within the European Human Biomonitoring tions (µg/L) in the current manuscript. Besides, for one study which reported only creatinine-corrected data and showed the geometric Initiative HBM4EU. More information on the time-trends of mean of measured creatinine concentrations in the study population phthalates and phthalate replacements in the general population [40], we back-calculated the unadjusted concentrations (SI 1, section would be needed and would be helpful to interpret the S3.1. Equation S1) and included them in the analysis. Additionally, for effectiveness of various regulatory risk measurements in various few publications which reported exclusively specific-gravity (SG) parts of the world, like the authorisation and restriction in the EU adjusted data, we considered the SG-adjusted data to be equivalent to as mentioned earlier, the concentration limits for children’s toys unadjusted concentrations. In this sense, Runkel et al. (2020) [41] and childcare products and for food packaging in the US. Notably, showed little differences between SG-adjusted and unadjusted phtha- there is a gap on the time-trends of these substances in the late concentrations in adults and children. Chinese population [37]. In the literature, some publications reported the central values of urinary concentrations for specific substances to be below the limits of Our objective was to analyse time-trends in the urinary detection (LOD, µg/L) or quantification (LOQ, µg/L). In those sources, concentrations of phthalates, DINCH, and DEHT in the general censored data were generally corrected through substitution methods. population from non-European countries (including China), in the Specifically, the values <LOD or <LOQ were generally replaced by LOD/2 last decade. We conducted a literature review and compiled or LOQ/2, respectively [23, 24]. Similarly, other authors replaced human biomonitoring (HBM) data for phthalates and phthalate censored data by LOD/√2orLOQ/√2[32, 42]. Following the same replacements for different age groups and countries worldwide. In substitution methods, we replaced the central values <LOD or <LOQ by summary, data from 123 publications for metabolites from all LOD/2 and LOQ/2, respectively. More details are given in the SI 1, section phthalates, DEHT and DINCH in more than 30 countries were S3.2. Substances with > 50% of censored central values in the database compiled within the “PhthaLit” database. Based on the compiled (see Fig. S1 for adults) were excluded from further data analyses. central values (medians, geometric means), we analysed time- trends in urinary concentrations per metabolite, age group, and Sampling years. We compiled data for sampling years from 2009 onwards. country/region, in non-European countries. Additionally, we In the literature, most studies reported data for one sampling campaign (in other words, for one time-point). Additionally, some “time-trend” studies compared phthalate urinary concentrations in adults and children, (including cross-sectional studies with multiple time-points and longitudinal between continents (including Asia, North America, and Europe). Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. studies), which reported data for consecutive sampling campaigns number was used for data analysis (e.g. to calculate the median (consecutive time-points), were also found. We compiled data from both value if the same study reported separate data in male and female) types of studies. To be precise, in the reviewed HBM studies, one sampling and for the identification, in the Figures, of data from the same source campaign (i.e. one time-point) did not necessarily refer to one specific year (same colour). but to one period (e.g. Aug. 2015–Sep. 2016; 2013–2015…). In our database, the average year per sampling time point was determined and used for our analyses. However, for studies with long sampling campaigns, Statistical methods we considered that an average sampling year may not be representative Time-trend analysis. In order to calculate time-trends between 2009 and enough for the purposes of this study and could introduce an error in the 2019, the central values in the database were aggregated per country, age time-trend analyses. In this sense, during data compilation, we decided to group, metabolite, year, and study (Study ID number, see Data traceability exclude data reported for sampling time points (specific periods) longer in the “PhthaLit” database), providing one median value for each unique than 4 years. combination of these variables. This principle was applied to all data, including for studies which reported only separated data for variables not Population and age groups. We aimed to compile HBM data which may included in the analysis (e.g. separated data for men and women, or for be representative of the exposure levels for the general population in a different cities/areas), for which the median value per country, age group, country. Therefore, we did not collect data from more highly exposed metabolite, year, and study/survey was calculated from all data and used in groups, such as a) occupationally exposed workers, b) population from the time-trend analysis. Each median was also characterised by the sum of reported “hotspot” locations (i.e. including phthalate manipulation sites), sample sizes used for its computation. The study-specific sample size (N) and c) transfused patients. We also excluded data from subjects with was used to assign an importance to each median value within the analysis confirmed or suspected reproductive pathology (e.g. couples under- (weighting). A minimum sample size of 120 has been proposed for going infertility treatment) or chronic disease (e.g. obesity). For studies derivation of European Reference Values in a population, in order to enable where a specific group of population, not representative for the a proper estimation of the 95% CI of the 95th percentile [48–51]. Taking general population, was compared with a control group, data from the this value as a reference, we considered that central values (medians or controls were compiled. Finally, we did not compile data for pregnant geometric means) computed from 120 or more subjects from the general women, since pregnancy may have an impact on phthalate pharmaco- population per study group were fully representative. Therefore, depend- kinetics [43]. ing on N, we assigned a weight to each central value: We classified the data by age group [27, 44]. Specifically, the following groups were defined, depending on the average age (age) of the subjects: ● for studies with N >=120, the weight of the central values was 1; “adults” (18 years < age < 60 years), “children” (4 years <= age <= ● for studies with N < 120 (i.e., 20 <= N < 120), the weight of the central 12 years), “teenagers” (12 years < age <= 18 years), “seniors” (60 years <= values was ⌈N/120⌉, where the half-brackets ⌈⌉ represent the ceiling age), and “younger children” (age < 4 years). The main age groups for which operation (rounding up). HBM data for phthalates were found were those of adults, followed by children (cf. Results, PhthaLit database). Therefore, we considered both age Assuming an approximately log-normal distribution of the data, we groups as a priority for our study. logarithmically (ln) transformed the urinary concentrations, similarly to the approach by Frederiksen et al. [23]. Next, we looked for linear time-trends Countries and continents. In the database, the continent, country, and on these log-transformed data. Time-trends were investigated using the region (if it was shown), were specified for each datum. non-parametric Kendall test and the Theil-Sen trend estimator [48]. These In Asia, several studies published HBM data specifically for the robust methods are not prone to be biased by possible extreme values and Taiwanese population [40, 45–47]. As a consequence the number of data provide robust estimates of the presence and slope of the trend. available in Taiwan was relatively high (SI 1, Tables S3 and S4). For these Specifically, we applied these methods for a) time series including 5 or reasons, we analysed Taiwan separately. more central values (“n” ≥ 5), and for b) time-series with lower n (n = 3–4) In Europe, time-trend analyses for these plasticizers are conducted only if data from the national human biomonitoring plan in the country within the HBM4EU project. Thus, we excluded European countries from were available at the earliest and the latest years in the time-series. Results our time-trend analyses (cf. Statistical methods, Time-trend analysis). with p ≤ 0.10 were considered significant. Nevertheless, data from European countries were included in the “PhthaLit” database, and Europe was included for geographic comparisons Additional analyses. We analysed the possible impact of the gender, area, of phthalate levels between continents (cf. Statistical methods, additional and sample type on the estimated trends per metabolite, country, and age analyses). group. Specifically, all time-trend analyses were conducted separately per gender, area, and sample type, and compared to the original trends using Sample size (N). In the database, per study, we specified the number of Fisher z-scoring [52]. None of these variables had a significant impact in subjects who provided urine for phthalate measurements (i.e., any trend. The p-value of the trend difference test [52] was in all cases study-specificsamplesize, “N”). This variable (N) was used for data higher than 0.10. Further details are shown in the SI 1 (section S3.3). weighting in our time-trend analysis (cf. Statistical methods, Time-trend Furthermore, geographic comparisons in the urinary phthalate concen- analysis). No data were compiled for N < 20. trations between continents were conducted. Specifically, the central values (i.e., median, GM) of urinary concentrations per metabolite, continent, and age group, for two different periods (i.e., 2009–2014 and Other variables. Major variables explained above (i.e. measured sub- 2015–2019) were compared. Further details are given in the SI 1 stance, age group, country, and sampling year) created a high data (section S3.4). categorization. For this reason, other variables which were considered comparatively less important for our analysis, for example the gender, area (urban, suburban), or the sample type were not included as variables for time-trend calculations per country or geographic comparisons per RESULTS continent. Nevertheless, these variables were specified in the database “PhthaLit” database and we analysed the possible impact of the gender, sample type, and area The “PhthaLit” database, generated and analysed during this on the estimated trends (cf. Statistical methods, additional analyses). study, can be found as an Excel file of Supplementary Information (SI 2) with the current manuscript. Data traceability in the “PhthaLit” database. In the database, general Data from 123 publications were compiled, including variables for traceability were included, particularly: principally peer-reviewed publications, as well as data reported Reference short name and digital object identifier (DOI). by the US Centers for Disease Control and Prevention [42], Human biomonitoring (HBM) survey name. Health Canada [24], and Santé Publique France [53]. In Study identification number (Study ID number). This number identifies summary, 2287 compiled central values of unadjusted urinary data from the same source. Generally, one Study ID number identifies concentrations (µg/L) of phthalates and phthalate replacements data from one individual study. Nevertheless, for some studies from the “PhthaLit” database were available and used in the which reported data from the same HBM survey in a population, we time-trend analysis. Concerning the distribution of data among used the same Study ID number (see SI 1, Tables S5 and S6). This Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. –1 age groups, more unadjusted central values were found for 3–10% yr (Fig. S6a, c). In contrast, MiBP increased in US adults –1 adults (n = 1208) and children (n = 859), than for teenagers and children (+4.8% yr )(Fig. S6b). (n = 160) or other age groups. Between countries, a high number of data were found principally in China, followed by High-weight phthalates (DINP, DIDP), and MCPP (non-specific Germany (also Denmark for adults) and the United States, both metabolite). For DINP and DIDP metabolites, specifically for cx- –1 in adults and children. Among substances, more data were MINP and cx-MIDP, respectively, decreasing trends by 2–9% yr available for DEHP metabolites and low-weight phthalates than were found in the US population (Table 1). for high-weight phthalates and phthalate replacements Concerning MCPP, this is a non-specific metabolite that can be (Tables S3 and S4). formed through metabolism of several high- and low-weight phthalates [26]. We found decreasing trends for MCPP in the US, Time-trends Canada, and China (Fig. S7). Significant time-trends per substance, country/region, and age group, in percentages of change per year, are summarized in Tables 1 and 2. In these Tables, the specific periods studied and DISCUSSION data sources are reported. Only trends including <33% of We analysed time-trends in the urinary concentrations of censored data, one or more data points after 2015, for a total phthalates in general population from Asian and North American period >= 4 years are shown. countries, following a literature review and data compilation. Time-trends were found principally in adults and/or children Naturally, analysing data from multiple sources presented a from five countries/regions, including three in Asia (China, Taiwan, challenge based on the significant inter-study variability under- Korea) and two in North America (Canada and the US). Only 4 lying the data sets. To deal with this matter, it was essential to a) trends in US children, out of 58 significant trends, had p-values conduct a non-biased literature review and a thorough data between 0.05 and 0.10. The other 54 trends were significant at p < compilation, and b) select the main variables for the time-trend 0.05 (see Tables 1 and 2). Time-trends included generally 5 or analyses. As major variables, we included the metabolite, age more central values. Exceptionally, trends in Canada and some group, country, sampling year, and sample size. By including the trends in the US with a lower (n = 3–4) were also included, since major variables in our analysis, on the one hand we found globally data from the national human biomonitoring plans in those solid time-trends. On the other hand, this created a high data countries [24, 42] were available at the earliest and the latest categorization. As an illustration, significant trends per metabolite, sampling years per trend. Importantly, the trends in both North age group, and country were found “just” in 5 non-European American countries were globally consistent. countries/regions, and a similar number of European countries (not shown), out of approximately 30 countries for which data DEHP metabolites. Significant time-trends were found for the were compiled. To avoid further data categorization, other main DEHP metabolites (Table 1). variables (i.e., gender, sample type, area) were not included in In the US, Canada, and Korea, DEHP metabolites decreased with the analyses. Importantly, we tested the possible impact of those time. Notably in the US population and Canadian adults, the excluded variables (gender, area, sample type) on the estimated urinary concentrations of secondary DEHP metabolites consis- time-trends. None of these variables had a significant impact in tently decreased by approximately 8–15% per year (Fig. 1a). any trend (SI 1, section S3.3). Nevertheless, we found a bias for the In China, time-trends for DEHP metabolites diverged between time-trends in Korea, for different reasons. Specifically in Korea, age groups and metabolites. On the one hand, decreasing trends phthalate concentrations reported in 2016 by Lee I. et al. and Lee –1 were found for 5cx-MEPP in children (–13.7% yr ; Fig. 1b) and G. et al. [54, 55] were low, as compared to KoNEHS data [56] (see –1 adults (–5.7% yr ) (Fig. S2a). On the other hand, 5oxo- and 5OH- Fig. S3b). Importantly, Lee I. and Lee G. et al. measured phthalates MEHP increased by 11–15% per year in children (Figs. 1c and S3a). in spot urine collected in control women, following a fasting of at Data for 5oxo-MEHP and 5OH-MEHP in Chinese adults did not least 8 h [54, 55]. Fasting prior to the collection of spot urine seem to follow a linear time-trend pattern, and no significant (which was not first morning urine) was relatively rare in the trends were found (Fig. S2b). The results for DEHP in China are reviewed literature for phthalates. For instance, no other studies further discussed below. with this design were included in the “PhthaLit” database. Importantly, fasting may reduce the exposure to these substances. DINCH and DEHT metabolites. In the US population, phthalate In fact, the diet is considered as a major exposure source for replacements markedly increased with time. Specifically, signifi- certain phthalates [4–6]. Thus, the use of spot urine data following –1 cant increases were found for OH-MINCH in children (+62.9% yr ) fasting may have biased the trends in Korea, which were –1 and 5cx-MEPTP in adults (+51.0% yr ) (Fig. 2). overestimated, as compared to less pronounced decreasing trends suggested by KoNEHS data [34, 56]. Importantly, no bias Low-weight phthalates (DMP, DEP, DnBP, DiBP, BBP). For low- was found for the time-trends in China, Taiwan, the US, and weight phthalates, contrasting results between countries/regions Canada. and substances were found (Table 2). In China, Taiwan, and the US, A question remains concerning some inconsistent time-trends metabolites from some low-weight phthalates decreased while between DEHP metabolites in China. To clarify this question, we others increased. In Canada and Korea, low-weight phthalates need to mention a particularity of the data in China. Specifically in decreased with time. this country, we found more HBM data for DEHP and low-weight To begin with Asian countries/regions, MMP increased in phthalates, from a higher number of peer-reviewed studies, than in –1 –1 children from Taiwan (+35.8% yr )and China(+8.7% yr ) any other country. Conversely, we found no data from cross- (Figs. 3c and S4a). Importantly, in China, MnBP consistently sectional national human biomonitoring studies in China. This is in –1 –1 increasedbothinchildren (+34.4% yr ) and adults (+9.1% yr ) opposition to other countries, for which phthalate data from (Fig. 3a, b). Other low-weight phthalates decreased. Notably, national HBM studies were found (e.g. the US National Health and –1 MBzP consistently decreased in children from China (–33.3% yr ) Nutrition Examination Survey, NHANES; the Canadian Health –1 and Taiwan (–13.6% yr ) (Fig. S4b, c), MiBP and MEP decreased Measures Survey, CHMS; and the Korean National Environmental –1 –1 in Chinese adults (–13.4% yr and –6.9% yr ,respectively) Health Survey, KoNEHS) and included in the time-trends estimations. (Fig. S5). Importantly, the China National Human Biomonitoring (CNHBM) In the US population and in Canadian adults, low-weight survey started in 2017–2018 [57], and data from this survey may phthalate biomarkers generally decreased by approximately facilitate future time-trend analyses in this country. In our study, data Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. Table 1. Significant time-trends for metabolites formed from DEHP, high-weight phthalates, DEHT and DINCH. % Data sources Parent Meta- Country change (included in p-value subs- Age group n n period bolite /region per the PhthaLit tance year database) adults –5.7 0.003 11 8.8 2010–2019 (65-75) China (70, 76-82) children –13.7 0.010 7 5.5 2012–2017 5cx-MEPP adults –11.9 <0.001 11 9.8 2009–2016 (32, 42, 83) (42, 84) US children –9.1 <0.001 5 5 2010–2016 teenagers –14.8 <0.001 4 4 2010–2016 (42) (24, 59) Canada adults –11.4 0.044 3 3 2010–2017 (70, 76-82, 85- China children 10.9 <0.001 11 8.7 2011–2018 87) South adults –18.6 <0.001 7 6.7 2009–2016 (54-56, 88-91) 5OH-MEHP Korea adults –9.4 0.001 8 7.0 2009–2016 (32, 42) US children –8.3 <0.001 5 5 2010–2016 (42, 84) DEHP teenagers –14.4 <0.001 4 4 2010–2016 (42) Canada adults –10.3 0.044 3 3 2010–2017 (24, 59) (70, 76-82, 85- China children 14.9 <0.001 11 8.7 2011–2018 87) South (54-56, 88-91) adults –14.1 <0.001 7 6.7 2009–2016 5oxo-MEHP Korea (42) adults –11.4 <0.001 4 4 2010–2016 US children –8.2 <0.001 5 5 2010–2016 (42, 84) (42) teenagers –14.1 <0.001 4 4 2010–2016 Canada adults –10.1 0.044 3 3 2010–2017 (24, 59) adults –6.1 <0.001 4 4 2010–2016 (42) MEHP US children –4.0 0.081 5 5 2010–2016 (42, 84) teenagers –8.4 <0.001 4 4 2010–2016 (42) (42, 83) DINP cx-MINP US adults –2.2 0.016 7 6.8 2009–2016 adults –8.8 <0.001 4 4 2010–2016 (42) (42, 84) DIDP cx-MIDP US children –1.8 0.081 5 5 2010–2016 teenagers –3.3 <0.001 4 4 2010–2016 (42) Canada adults –12.8 <0.001 3 3 2010–2017 (24, 59) (65-67, 71, 72, China adults –27.3 <0.001 7 5.5 2010–2016 74, 92) Several MCPP a adults –17.2 <0.001 4 4 2010–2016 (42) (42, 84) US children –12 <0.001 5 5 2010–2016 teenagers –14.3 <0.001 4 4 2010–2016 (42) a a DINCH OH-MINCH US children 62.9 <0.001 4 4 2012–2016 (42, 84) DEHT 5cx-MEPTP US adults 51 <0.001 5 4 2009–2016 (32, 42) Green: decreasing trends; red: increasing trends. Only significant trends (p-value <= 0.10) including <33% of censored data, with one or more data points after 2015, for a total period >= 4 years are shown. n = number of central values aggregated per country, age group, metabolite, year, and Study ID number, per time-trend; n = number of “full-weight” data per trend, calculated as the sum of the number of aggregated central values (n) multiplied by their weight (Statistical methods, time-trend analysis). For trends with n < 5 (i.e., trends in Canada and some trends in the US), data from national human biomonitoring plans [24, 42] were used both at the earliest and the latest years per trend. For studies where a specific group of population was compared with a control group, data from controls were compiled. Trends in Korea were overestimated (see Discussion). Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. Table 2. Significant time-trends for metabolites formed from low-weight phthalates. % Data sources Parent Metaboli Country change (included in Age group p-value n n period substance te /region per the PhthaLit year database) Canada adults –5.4 0.044 3 3 2010–2017 (24, 59) (70, 76, 78-82, DMP MMP China children 8.7 <0.001 9 8.2 2011–2018 85-87) Taiwan children 35.8 <0.001 6 6 2009–2016 (40, 45, 47, 93) Canada adults –10.4 <0.001 3 3 2010–2017 (24, 59) (22, 65-75, 87, China adults –6.9 0.016 15 12.2 2010–2019 94, 95) DEP MEP adults –7.9 <0.001 4 4 2010–2016 (42) US children –3.9 0.081 5 5 2010–2016 (42, 84) teenagers –4.4 0.013 4 4 2010–2016 (42) Canada adults –2.6 <0.001 3 3 2010–2017 (24, 59) (65-75, 87, 94, China adults –13.4 0.001 14 11.2 2010–2019 95) South DiBP MiBP adults –26.3 <0.001 3 2.7 2009–2016 (54, 55, 88) Korea adults 4.8 0.013 4 4 2010–2016 (42) US children 4.8 <0.001 5 5 2010–2016 (42, 84) Canada adults –4.9 0.044 3 3 2010–2017 (24, 59) (22, 65-75, 87, China adults 9.1 <0.001 15 12.2 2010–2019 94, 95) (70, 76-82, 85- China children 34.4 <0.001 11 8.7 2011–2018 87) South DnBP MnBP adults <0.001 6 5.7 2009–2016 (54-56, 88-90) –18.5 Korea Taiwan children –2.4 0.008 6 6 2009–2016 (40, 45, 47, 93) children –4.6 0.081 5 5 2010–2016 (42, 84) US teenagers –7.4 <0.001 4 4 2010–2016 (42) Canada adults –8.1 <0.001 3 3 2010–2017 (24, 59) (70, 77-82, 86, China children –33.3 0.007 8 5.7 2013–2018 87) South adults –36.8 <0.001 5 4.7 2009–2016 (54-56, 88, 89) BBP MBzP Korea Taiwan children –13.6 <0.001 6 6 2009–2016 (40, 45, 47, 93) adults –5.7 <0.001 4 4 2010–2016 (42) US teenagers –8.3 <0.001 4 4 2010–2016 (42) Green: decreasing trends; red: increasing trends. Only significant trends (p-value < = 0.10) including <33% of censored data, with one or more data points after 2015, for a total period >= 4 years are shown. n = number of central values aggregated per country, age group, metabolite, year, and Study ID number per time-trend; n = number of “full-weight” data per trend, calculated as the sum of the number of aggregated central values (n) multiplied by their weight (Statistical methods, time-trend analysis). For trends with n < 5 (i.e., trends in Canada and some trends in the US), data from national human biomonitoring plans [24, 42] were used both at the earliest and the latest years per trend. For studies where a specific group of population was compared with a control group, data from controls were compiled. Trends in Korea were overestimated (see Discussion). in China presented quite large inter-study variability. Hence, for the whole study period and for the most recent years (i.e. since the number of data available per time-trend was particularly 2015), per trend. In this sense, among DEHP metabolites, we found important in this country. Specifically, to identify the most reliable more data for 5oxo-MEHP and 5OH-MEHP than for 5cx-MEPP, both time-trends in China, we analysed the number of central values (n) in adults and children (cf. Figs. 1b, c and S2). In adults, log- Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. a) 5cx-MEPP (DEHP metabolite) in US adults Sampling year Annual decrease (weighted) of 11.9% (p < 0.001) b) 5cx-MEPP (DEHP metabolite) in Chinese children c) 5oxo-MEHP (DEHP metabolite) in Chinese children Sampling year Sampling year Annual decrease (weighted) of 13.7% (p = 0.01) Annual increase (weighted) of 14.9% (p < 0.001) Fig. 1 Time-trends in the urinary concentrations of DEHP metabolites (µg/L). a 5cx-MEPP in US adults. b 5cx-MEPP in Chinese children. c 5oxo-MEHP in Chinese children. Each symbol is a central value (i.e. median or geometric mean). The size of the symbols is related to the study-specific sample size (N) and its weight in the time-trend calculation (see Statistical methods, time-trend analysis). Data from the same data source (Study ID number: see Data traceability in the “PhthaLit” database) are identified by symbols with the same colour in all panels and figures. Grey area: 90% confidence interval. a) b) OH-MINCH (DINCH metabolite) in US children 5cx-MEPTP (DEHT metabolite) in US adults Sampling year Sampling year Annual increase (weighted) of 63% (p < 0.001) Annual increase (weighted) of 51% (p < 0.001) Fig. 2 Time-trends in the urinary concentrations of phthalate replacement metabolites (µg/L). a OH-MINCH (a DINCH metabolite) in US children. b 5cx-MEPTP (a DEHT metabolite) in US adults. Each symbol is a central value (i.e. median or geometric mean). The size of the symbols is related to the study-specific sample size (N) and its weight in the time-trend calculation (see Statistical methods, time-trend analysis). Data from the same data source (Study ID number: see Data traceability in the “PhthaLit” database) are identified by symbols with the same colour in all panels and figures. Grey area: 90% confidence interval. a) b) MnBP (DnBP metabolite) in Chinese children MnBP (DnBP metabolite) in Chinese adults Sampling year Sampling year Annual increase (weighted) of 34.4% (p < 0.001) Annual increase (weighted) of 9.1% (p < 0.001) c) MMP (DMP metabolite) in Taiwanese children Sampling year Annual increase (weighted) of 35.8% (p < 0.001) Fig. 3 Time-trends in the urinary concentrations of low-weight phthalate metabolites (µg/L). a MnBP (DnBP metabolite) in Chinese children. b MnBP in Chinese adults. c MMP (DMP metabolite) in Taiwanese children. Each symbol is a central value (i.e. median or geometric mean). The size of the symbols is related to the study-specific sample size (N) and its weight in the time-trend calculation (see Statistical methods, time-trend analysis). Data from the same data source (Study ID number: see Data traceability in the “PhthaLit” database) are identified by symbols with the same colour in all panels and figures. Grey area: 90% confidence interval. Journal of Exposure Science & Environmental Epidemiology Concentration (µg/L) Concentration (µg/L) Concentration (µg/L) Concentration (µg/L) C) oncentration (µg/ L Concentration (µg/L) Concentration (µg/L) Concentration (µg/L) Concentration (µg/L) E. Domínguez-Romero et al. transformed data for 5oxo- and 5OH-MEHP did not follow a linear be noted that the increases for MnBP in Chinese population, MMP time-trend pattern and no significant trends were found. Conversely and 5OH-MEHP in Asian children occurred in a context where the in children, we found significant time-trends for 5OH- and 5oxo- average concentrations of these substances in the last years seemed MEHP (p < 0.001), which were more solid than those for 5cx-MEPP to be relatively high in Asia, as compared to data in other continents (p= 0.010). To summarize our results for DEHP metabolites in China, (SI 1, section S3.4). In conclusion, our results in China raise some we found that 5OH-MEHP and 5oxo-MEHP increased over time, concern and seem globally consistent with the leading position of while 5cx-MEPP seemed to decrease over time, which is unexpected Asia in the plasticizer global market, and the increasing importance since the three of them are metabolites of DEHP. Notably, this of China in this market (i.e. China represented approximately 42% of apparent decreasing time trend for 5cx-MEPP could be due to the worldwide consumption of plasticizers in 2017, and 50% in chemical-analytical flaws. It has been observed that 5cx-MEPP may 2020) [1, 62]. co-elute with OH-MINP (oral communication Dr Koch). OH-MINP To conclude with few recommendations for future research, more being a metabolite of the high molecular weight phthalate DINP HBM data for DEHT and DINCH would be needed worldwide. The might be decreasing and so cause confusion where the quantified quite large increase per year of DEHT and DINCH metabolites in peaks may be attributed (partly) mistakenly to 5cx-MEPP instead of the US is quite troublesome because of the recent knowledge on the OH-MINP. Concerning the trends for low-weight phthalates in China, toxicological properties that also these phthalate substitutes exhibit. a relatively high number of central values per metabolite and age Furthermore, the world consumption of phthalate replacements is group (n = 11–15) were available for the time-trend calculations for predicted to increase markedly in the following years (2021–2025) MnBP (adults and children), MEP and MiBP in adults, which supports [1]. For high-weight phthalates, more data would be desirable too, the solidity of these results (i.e. increasing trends for MnBP and with a focus on the secondary metabolites from these substances decreasing trends for MEP and MiBP). Importantly, MnBP consis- [63]. To finalize, environmental pollutants such as the studied tently increased in adults and children. This consistency between plasticizers are a global concern. In this context, as previously age groups highlights the reliability of these results, since the suggested [64], the scarcity of HBM data for phthalates, DEHT, and majority of data sources were independent between adults and DINCH in general population in some continents (Africa, South children (i.e., for MnBP, out of 11–15 data sources per age group, America, Oceania) and some countries worldwide was striking. only 2 sources were common to both age groups). For MMP and MBzP in Chinese children, a lower number of central values was found (n = 8–9). Interestingly for these two metabolites, we found DATA AVAILABILITY The “PhthaLit” database, generated and analysed during this study, can be found as consistently increasing trends for MMP and decreasing trends for an Excel file of Supplementary Information (SI 2) with the current manuscript. MBzP both in Chinese and Taiwanese children. To sum up, in the general populations from the US and Canada, phthalate metabolites generally decreased with time. Conversely, REFERENCES metabolites from phthalate replacements (DEHT, DINCH) increased 1. IHS_Markit. Plasticizers. Chemical Economics Handbook. Published May 2021. in the US. Globally, these results were in agreement with decreasing 2021 [Available from: https://ihsmarkit.com/products/plasticizers-chemical- trends for phthalate metabolites, and increasing trends for economics-handbook.html. metabolites of the substitutes DEHT and DINCH in population from 2. Rodríguez-Bernaldo-de-Quirós A, Lestido-Cardama A, Sendón R, García-Ibarra V. Puerto Rico [58], Germany [26, 27, 35, 36], and Denmark [23]. In Phthalates and alternative plasticisers. Food Contamination by Packaging: Canadian population, phthalate time-trends have been recently Migration of Chemicals from Food Contact Materials. Berlin: Walter De Gruyter reported by Pollock et al., 2021 [33]. Although there is a partial Gmbh; 2019. p. 45–94. overlap in data sources used by Pollock et al. (2021) and those used 3. Danish_EPA. Hazardous substances in plastics. Survey of chemical substances in consumer products No. 132, 2014. The Danish Environmental Protection Agency. in the current study, we additionally used data by Albert et al. 2018 Erik Hansen (COWI A/S), Nils Nilsson (Danish Technological Institute), and Kristine [59]. Other differences between our data analysis in Canada and the Slot Ravnholt Vium (Danish Technological Institute). ISBN no. 978-87-93283-31-2. study by Pollock et al. (2021) are for the time periods considered, the The Danish Environmental Protection Agency; 2014. age ranges, and the selection of metabolites. The upshot is that 4. Giovanoulis G, Bui T, Xu FC, Papadopoulou E, Padilla-Sanchez JA, Covaci A, et al. Pollock et al. [33] showed decreasing trends for DEHP metabolites, Multi-pathway human exposure assessment of phthalate esters and DINCH. low-weight phthalate metabolites (specifically, MEP, MnBP, and Environ Int. 2018;112:115–26. MBzP), and MCPP in Canadian population, which is consistent with 5. Wormuth M, Scheringer M, Vollenweider M, Hungerbuhler K. What are the our results. In addition, we found slight decreasing trends for MMP sources of exposure to eight frequently used phthalic acid esters in Europeans? and MiBP in this country (these metabolites were not assessed in the Risk Anal. 2006;26:803–24. 6. Koch HM, Lorber M, Christensen KLY, Palmke C, Koslitz S, Bruning T. Identifying study by Pollock et al. [33]). Interestingly, for low-weight phthalates, sources of phthalate exposure with human biomonitoring: Results of a 48 h we found diverging time-trends between substances in China, fasting study with urine collection and personal activity patterns. Int J Hyg Taiwan, and the US. In other words, in these countries/regions some Environ Health. 2013;216:672–81. low-weight phthalates increased while others decreased. These 7. Hlisnikova H, Petrovcova I, Kolena B, Sidlovska M, Sirotkin A. Effects and results might be related to a substitution in the uses of certain low- mechanisms of phthalates’ action on reproductive processes and reproductive weight phthalates by others. Notably, we found contrasting trends health: A literature review. Int J Environ Res Public Health. 2020;17:6811. between DnBP and DiBP monoesters in the US and China. For 8. Radke EG, Braun JM, Meeker JD, Cooper GS. Phthalate exposure and male example, in the US, MiBP increased while MnBP decreased with reproductive outcomes: A systematic review of the human epidemiological evi- time. Similarly, in men attending a fertility clinic in Boston between dence. Environ Int. 2018;121:764–93. 9. Radke EG, Glenn BS, Braun JM, Cooper GS. Phthalate exposure and female 2000 and 2017, MiBP increased while other phthalates generally reproductive and developmental outcomes: a systematic review of the human decreased [60]. In China, we globally found solid trends, excepted epidemiological evidence. Environ Int. 2019;130:19. for 5cx-MEPP for which chemical analytical flaws could not be 10. Hlisnikova H, Petrovicova I, Kolena B, Sidlovska M, Sirotkin A. Effects and excluded. The results in Chinese population suggested notably an mechanisms of phthalates’ action on neurological processes and neural health: a increasing exposure to DEHP, DnBP and DMP, and decreasing literature review. Pharmacol Rep. 2021;73:386–404. exposure to DEP and DiBP in China in the last decade. The increases 11. Radke EG, Braun JM, Nachman RM, Cooper GS. Phthalate exposure and neuro- in the urinary concentrations of DEHP metabolites in Chinese development: A systematic review and meta-analysis of human epidemiological population contrasted with decreasing trends in North American evidence. Environ Int. 2020;137:105408. and European countries. Interestingly, in Japanese children, the 12. Merced-Nieves FM, Dzwilewski KLC, Aguiar A, Musaad S, Korrick SA, Schantz SL. Associations of prenatal exposure to phthalates with measures of cognition in concentrations of phthalate metabolites (including DEHP metabo- 4.5-Month-old infants. Int J Environ Res Public Health. 2021;18:1838. lites) were stable between 2012 and 2017 [61]. To finalize, it should Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. 13. Radke EG, Galizia A, Thayer KA, Cooper GS. Phthalate exposure and metabolic 35. Kasper-Sonnenberg M, Koch HM, Apel P, Ruther M, Palmke C, Bruning T, et al. effects: a systematic review of the human epidemiological evidence. Environ Int. Time trend of exposure to the phthalate plasticizer substitute DINCH in Germany 2019;132:104768. from 1999 to 2017: Biomonitoring data on young adults from the Environmental 14. Danish_EPA. Alternatives to classified phthalates in medical devices. Environ- Specimen Bank (ESB). Int J Hyg Environ Health. 2019;222:1084–92. mental Project No. 1557. Danish Ministry of the Environment - Danish Environ- 36. Lessmann F, Kolossa-Gehring M, Apel P, Ruther M, Palmke C, Harth V, et al. mental Protection Agency. ISBN no. 978-87-93178-27-4. 2014. German Environmental Specimen Bank: 24-hour urine samples from 1999 to 15. Campioli E, Lee S, Lau M, Marques L, Papadopoulos V. Effect of prenatal DINCH 2017 reveal rapid increase in exposure to the para-phthalate plasticizer di(2- plasticizer exposure on rat offspring testicular function and metabolism. Sci Rep. ethylhexyl) terephthalate (DEHTP). Environ Int. 2019;132:105102. 2017;7:14. 37. Wang W, Leung AOW, Chu LH, Wong MH. Phthalates contamination in China: 16. Campioli E, Lau M, Papadopoulos V. Effect of subacute and prenatal DINCH Status, trends and human exposure-with an emphasis on oral intake. Environ plasticizer exposure on rat dams and male offspring hepatic function: The role of Pollut. 2018;238:771–82. PPAR-alpha. Environ Res. 2019;179:108773. 38. Sharma BM, Kalina J, Whaley P, Scheringer M. Towards guidelines for time-trend 17. Eichler CMA, Hubal EAC, Little JC. Assessing Human Exposure to Chemicals in reviews examining temporal variability in human biomonitoring data of pollu- Materials, Products and Articles: The International Risk Management Landscape tants. Environ Int. 2021;151:106437. for Phthalates. Environ Sci Technol. 2019;53:13583–97. 39. ChemSec. Replacing phthalates - Why and how to substitute this hard-to-spell 18. ECHA. Consolidated version of the Regulation (EC) No 1907/2006 of the European chemical group. International Chemical Secretariat (ChemSec), September Parliament and of the Council on the Registration, Evaluation, Authorisation and 2019; 2019. Restriction of Chemicals (REACH). European Chemicals Agency (ECHA) 2022 40. Su PH, Chen JY, Lin CY, Chen HY, Liao PC, Ying TH, et al. Sex steroid [Available from: https://echa.europa.eu/regulations/reach/legislation. hormone levels and reproductive development of eight-year-old children 19. EC. Harmful chemicals – updated list of endocrine disrupting substances (REACH following in utero and environmental exposure to phthalates. Plos One. 2014;9: Regulation). European Commission (EC). 2021 [Available from: https://ec.europa. e102788. eu/info/law/better-regulation/have-your-say/initiatives/12550-Harmful-chemicals 41. Runkel AA, Snoj-Tratnik J, Mazej D, Horvat M. Urinary phthalate concentrations in -updated-list-of-endocrine-disrupting-substances-REACH-Regulation-_en. the slovenian population: An attempt to exposure assessment of family units. 20. Chemical_Watch. Four phthalates lose authorisation exemptions after REACH Environ Res. 2020;186:13. Annex XIV entry amendment. 30 November 2021 2021 [Available from: https:// 42. CDC. Fourth National Report on Human Exposure to Environmental Chemicals, chemicalwatch.com/381308/four-phthalates-lose-authorisation-exemptions- Updated Tables, January 2019, Volume One. U.S. Department of Health and after-reach-annex-xiv-entry-amendment. Human Services. Centers for Disease Control and Prevention (CDC). 2019. 21. ECHA. Substances restricted under REACH 2021 [Available from: https://echa. 43. Zhao HZ, Li JF, Zhou YQ, Zhu L, Zheng YY, Xia W, et al. Investigation on Meta- europa.eu/substances-restricted-under-reach. bolism of Di(2-Ethylhexyl) Phthalate in Different Trimesters of Pregnant Women. 22. Xu HL, Wu XY, Liang CM, Shen J, Tao SM, Wen X, et al. Association of urinary Environ Sci Technol. 2018;52:12851–8. phthalates metabolites concentration with emotional symptoms in Chinese 44. Choi J, Knudsen LE, Mizrak S, Joas A. Identification of exposure to environ- university students. Environ Pollut. 2020;262:10. mental chemicals in children and older adults using human biomonitoring 23. Frederiksen H, Nielsen O, Koch HM, Skakkebaek NE, Juul A, Jorgensen N, et al. data sorted by age: Results from a literature review. Int J Hyg Environ Health. Changes in urinary excretion of phthalates, phthalate substitutes, bisphenols and 2017;220:282–98. other polychlorinated and phenolic substances in young Danish men; 45. Hou JW, Lin CL, Tsai YA, Chang CH, Liao KW, Yu CJ, et al. The effects of phthalate 2009–2017. Int J Hyg Environ Health. 2020;223:93–105. and nonylphenol exposure on body size and secondary sexual characteristics 24. Health_Canada. Fifth report on human biomonitoring of environmental during puberty. Int J Hyg Environ Health. 2015;218:603–15. chemicals in Canada. Results Can Health Measures Surv Cycle 5 (2016–2017). 46. Huang HB, Siao CY, Lo YTC, Shih SF, Lu CH, Huang PC. Mediation effects of thyroid 2019;2019:429. function in the associations between phthalate exposure and glucose metabo- 25. Domínguez-Romero E, Scheringer M. A review of phthalate pharmacokinetics in lism in adults. Environ Pollut. 2021;278:10. human and rat: what factors drive phthalate distribution and partitioning? Drug 47. Liao KW, Chang WH, Chou WC, Huang HB, Waits A, Chen PC, et al. Human Metab Rev. 2019;51:314–29. biomonitoring reference values and characteristics of Phthalate exposure in the 26. Koch HM, Ruther M, Schutze A, Conrad A, Palmke C, Apel P, et al. Phthalate general population of Taiwan: Taiwan Environmental Survey for Toxicants 2013- metabolites in 24-h urine samples of the German Environmental Specimen Bank 2016. Int J Hygiene. Environ Health. 2021;235:113769. (ESB) from 1988 to 2015 and a comparison with US NHANES data from 1999 to 48. HBM4EU. Statistical Analysis Plan. Deliverable Report D10.5. WP 10 – Data 2012. Int J Hyg Environ Health 2017;220:130–41. management and analysis. Version 6.0. M. Vrijheid, P. Montazeri, L. Rambaud, N. 27. Schwedler G, Rucic E, Lange R, Conrad A, Koch HM, Palmke C, et al. Phthalate Vogel, J. Vlaanderen, S. Remy, E. Govarts, G. Schoeters. HORIZON 2020 Pro- metabolites in urine of children and adolescents in Germany. Human biomoni- gramme, Contract No. 733032 HBM4EU. 2019. toring results of the German Environmental Survey GerES V, 2014-2017. Int J Hyg 49. Solberg HE. International Federation of Clinical Chemistry (IFCC), Scientific Environ Health 2020;225:11. Committee, Clinical Section. Approved Recommendation (1986) on the Theory of 28. Schoeters G, Govarts E, Bruckers L, Den Hond E, Nelen V, De Henauw S, et al. Reference Values. Part 1. The concept of Reference Values. J Clin Chem Clin Three cycles of human biomonitoring in Flanders - Time trends observed in the Biochem. 1987;25:337–42. Flemish Environment and Health Study. Int J Hyg Environ Health. 2017;220:36–45. 50. Solberg HE, Petitclerc C. International Federation of Clinical Chemistry (IFCC), 29. Koppen G, Franken C, Den Hond E, Plusquin M, Reimann B, Leermakers M, et al. Scientific Committee, Clinical Section. Approved recommendation (1988) on the Pooled analysis of genotoxicity markers in relation to exposure in the Flemish Theory of Reference Values. Part 3. Preparation of individuals and collection of Environment and Health Studies (FLEHS) between 1999 and 2018. Environ Res specimens for the production of Reference Values. J Clin Chem Clin Biochem. 2020;190:13. 1988;26:593–8. 30. Larsson K, Lindh CH, Jonsson BAG, Giovanoulis G, Bibi M, Bottai M, et al. 51. Solberg HE. The IFCC recommendation on estimation of reference intervals. The Phthalates, non-phthalate plasticizers and bisphenols in Swedish preschool dust RefVal Program. Clin Chem Lab Med. 2004;42:710–4. in relation to children’s exposure. Environ Int. 2017;102:114–24. 52. Fisher RA. Frequency Distribution of the Values of the Correlation Coefficient in 31. Tranfo G, Caporossi L, Pigini D, Capanna S, Papaleo B, Paci E. Temporal trends of Samples from an Indefinitely Large Population. Biometrika 1915;10:507–21. urinary phthalate concentrations in two populations: Effects of REACH author- 53. Santé_publique_France. Imprégnation de la population française par les ization after five years. Int J Environ Res Public Health. 2018;15:1950. phtalates. Programme national de biosurveillance, Esteban 2014-2016. Saint- 32. Silva MJ, Wong LY, Samandar E, Preau JL, Calafat AM, Ye XY. Exposure to di-2- Maurice: Santé publique France, septembre 2019. 51 p. Available in URL: www. ethylhexyl terephthalate in a convenience sample of US adults from 2000 to santepubliquefrance.fr. 2019. 2016. Arch Toxicol 2017;91:3287–91. 54. Lee I, Kim S, Park S, Mok S, Jeong Y, Moon HB, et al. Association of urinary 33. Pollock T, Karthikeyan S, Walker M, Werry K, St-Amand A. Trends in environ- phthalate metabolites and phenolics with adipokines and insulin resistance mental chemical concentrations in the Canadian population: Biomonitoring related markers among women of reproductive age. Sci Total Environ. data from the Canadian Health Measures Survey 2007–2017. Environ Int 2019;688:1319–26. 2021;155:10. 55. Lee G, Kim S, Bastiaensen M, Malarvannan G, Poma G, Casero NC, et al. Exposure 34. Park C, Hwang M, Baek Y, Jung S, Lee Y, Paek D, et al. Urinary phthalate meta- to organophosphate esters, phthalates, and alternative plasticizers in association bolite and bisphenol A levels in the Korean adult population in association with with uterine fibroids. Environ Res. 2020;189:109874. sociodemographic and behavioral characteristics: Korean National Environmental 56. Lim S. The associations between personal care products use and urinary con- Health Survey (KoNEHS) 2012-2014. Int J Hyg Environ Health. 2019;222:903–10. centrations of phthalates, parabens, and triclosan in various age groups: The Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. Korean National Environmental Health Survey Cycle 3 2015-2017. Sci Total ventilation habits. 10th Int Symp Heat, Vent Air Conditioning, Ishvac2017. Environ. 2020;742:140640. 2017;205:1146–51. 57. Cao ZJ, Lin SB, Zhao F, Lv YB, Qu YL, Hu XJ, et al. Cohort profile: China National 80. Liao CX,Liu W, Zhang JL,Shi WM,Wang XY, CaiJ,et al. Associations of Human Biomonitoring (CNHBM)-A nationally representative, prospective cohort urinary phthalate metabolites with residential characteristics, lifestyles, and in Chinese population. Environ Int. 2021;146:10. dietary habits among young children in Shanghai, China. Sci Total Environ. 58. Rodríguez-Carmona Y, Ashrap P, Calafat AM, Ye XY, Rosario Z, Bedrosian LD, et al. 2018;616:1288–97. Determinants and characterization of exposure to phthalates, DEHTP and DINCH 81. Yao Y, Chen DY, Wu Y, Zhou L, Cheng JQ, Ma YY, et al. Urinary phthalate among pregnant women in the PROTECT birth cohort in Puerto Rico. J Exposure metabolites in primary school starters in Pearl River Delta, China: Occurrences, Sci Environ Epidemiol. 2020;30:56–69. risks and possible sources. Environ Res. 2019;179;108853;1–8. 59. Albert O, Huang JY, Aleksa K, Hales BF, Goodyer CG, Robaire B, et al. Exposure to 82. Yao Y, Chen DY, Yin JW, Zhou L, Cheng JQ, Lu SY, et al. Phthalate exposure linked polybrominated diphenyl ethers and phthalates in healthy men living in the to high blood pressure in Chinese children. Environ Int. 2020;143:8. greater Montreal area: A study of hormonal balance and semen quality. Environ 83. Silva MJ, Jia T, Samandar E, Preau JL, Calafat AM. Environmental exposure to the Int. 2018;116:165–75. plasticizer 1,2-cyclohexane dicarboxylic acid, diisononyl ester (DINCH) in US 60. Minguez-Alarcon L, Williams PL, Chiu YH, Gaskins AJ, Nassan FL, Dadd R, et al. adults (2000-2012). Environ Res. 2013;126:159–63. Secular trends in semen parameters among men attending a fertility 84. Hammel SC, Levasseur JL, Hoffman K, Phillips AL, Lorenzo AM, Calafat AM, et al. center between 2000 and 2017: Identifying potential predictors. Environ Int. Children’s exposure to phthalates and non-phthalate plasticizers in the home: 2018;121:1297–303. The TESIE study. Environ Int. 2019;132:10. 61. Ketema RM, Bamai YA, Ikeda-Araki A, Saito T, Kishi R. Secular trends of urinary 85. Zhang YH, Cao Y, Shi HJ, Jiang XX, Zhao Y, Fang X, et al. Could exposure to phthalate metabolites in 7-year old children and association with building phthalates speed up or delay pubertal onset and development? A 1.5-year characteristics: Hokkaido study on environment and children’s health. Int J Hyg follow-up of a school-based population. Environ Int. 2015;83:41–9. Environ Health. 2021;234:9. 86. Yu YX, Peng MM, Liu YL, Ma JJ, Wang N, Ma ST, et al. Co-exposure to polycyclic 62. IHS_Markit. Plasticizers. Chemical Economics Handbook. Published May aromatic hydrocarbons and phthalates and their associations with oxidative 2018 2018 [Available from: https://learning.ihsmarkit.com/courses/plasticizers- stress damage in school children from South China. J Hazard Mater. 2021; chemical-economics-handbook. 401:123390. 63. Calafat AM, Wong LY, Silva MJ, Samandar E, Preau JL, Jia LT, et al. Selecting 87. Zhang X, Tang S, Qiu T, Hu XJ, Lu YF, Du P, et al. Investigation of phthalate adequate exposure biomarkers of diisononyl and diisodecyl phthalates: Data metabolites in urine and daily phthalate intakes among three age groups in from the 2005-2006 National health and nutrition examination survey. Environ Beijing, China. Environ Poll. 2020;260:114005. Health Perspect. 2011;119:50–5. 88. Lee KM, Kho Y, Kim PG, Park SH, Lee JH. Urinary levels of phthalate metabolites 64. Wang Y, Zhu HK, Kannan K. A review of biomonitoring of phthalate exposures. and associations with demographic characteristics in Korean adults. Environ Sci Toxics 2019;7:28. Pollut Res. 2017;24:14669–81. 65. Gao CJ, Liu LY, Ma WL, Ren NQ, Guo Y, Zhu NZ, et al. Phthalate metabolites in 89. Choi W, Kim S, Baek YW, Choi K, Lee K, Seung DY. Exposure to environmental urine of Chinese young adults: Concentration, profile, exposure and cumulative chemicals among Korean adults-updates from the second Korean National risk assessment. Sci Total Environ. 2016;543:19–27. Environmental Health Survey (2012–2014). Int J Hyg Environ Health. 66. Guo Y, Alomirah H, Cho HS, Minh TB, Mohd MA, Nakata H, et al. Occurrence of 2017;220:29–35. phthalate metabolites in human urine from several asian countries. Environ Sci 90. Park C, Hwang M, Kim H, Ryu S, Lee K, Choi K, et al. Early snapshot on exposure to Technol. 2011;45:3138–44. environmental chemicals among Korean adults-results of the first Korean 67. Guo Y, Wu Q, Kannan K. Phthalate metabolites in urine from China, and impli- National Environmental Health Survey (2009-2011). Int J Hyg Environ Health. cations for human exposures. Environ Int. 2011;37:893–8. 2016;219:398–404. 68. Sun J, Chen B, Zhang LQ, Zhao D, Li SG. Phthalate ester concentrations in blood 91. Song NR, On JW, Lee J, Park JD, Kwon HJ, Yoon HJ, et al. Biomonitoring of urinary serum, urine and endometrial tissues of Chinese endometriosis patients. Int J Clin di(2-ethylhexyl) phthalate metabolites of mother and child pairs in South Korea. Exp Med. 2016;9:3808–19. Environ Int. 2013;54:65–73. 69. Chen JS, Zhou XF, Zhang H, Liu YM, Cao C, Dong RH, et al. Association between 92. Ding MY, Kang QY, Zhang SY, Zhao FR, Mu D, Zhang HF, et al. Contribution of urinary concentration of phthalate metabolites and impaired renal function in phthalates and phthalate monoesters from drinking water to daily intakes for the Shanghai adults. Environ Pollut. 2019;245:149–62. general population. Chemosphere 2019;229:125–31. 70. Dong RH, Zheng JH, Zhang MR, Chen JS, Zhang H, Gao X, et al. The concentra- 93. Huang PC, Chang WH, Wu MT, Chen ML, Wang IJ, Shih SF, et al. Characterization tions and cumulative risk assessment of phthalates in general population from of phthalate exposure in relation to serum thyroid and growth hormones, and Shanghai: The comparison between groups with different ages. Sci Total Environ. estimated daily intake levels in children exposed to phthalate-tainted products: A 2018;637:871–80. longitudinal cohort study. Environ Pollut. 2020;264:11. 71. Chen Q, Yang H, Zhou NY, Sun L, Bao HQ, Tan L, et al. Phthalate exposure, even 94. Yang LY, Yin WJ, Li P, Hu C, Hou J, Wang L, et al. Seasonal exposure to phthalates below US EPA reference doses, was associated with semen quality and repro- and inflammatory parameters: A pilot study with repeated measures. Ecotoxicol ductive hormones: Prospective MARHCS study in general population. Environ Int. Environ Saf. 2021;208:10. 2017;104:58–68. 95. Li XJ, Zhong Y, He WY, Huang SY, Li Q, Guo CS, et al. Co-exposure and health risks 72. Zhang B, Zhang T, Duan YS, Zhao Z, Huang XF, Bai XY, et al. Human exposure to of parabens, bisphenols, triclosan, phthalate metabolites and hydroxyl polycyclic phthalate esters associated with e-waste dismantling: Exposure levels, sources, aromatic hydrocarbons based on simultaneous detection in urine samples from and risk assessment. Environ Int. 2019;124:1–9. Guangzhou, south China. Environ Poll. 2021;272:115990. 73. Li JF, Zhao HZ, Xia W, Zhou YQ, Xu SQ, Cai ZW. Nine phthalate metabolites in human urine for the comparison of health risk between population groups with different water consumptions. Sci Total Environ. 2019;649:1532–40. 74. Zhang XM, Lou XY, Wu LH, Huang C, Chen D, Guo Y. Urinary phthalate meta- ACKNOWLEDGEMENTS bolites and environmental phenols in university students in South China. Environ The authors acknowledge Dr. Holger Koch for discussions concerning the chemical Res. 2018;165:32–9. analysis of phthalate metabolites; Prof. Jana Klánová (RECETOX) for support 75. Liu XT, Peng CF, Shi YM, Tan HL, Tang SQ, Chen D. Beyond phthalate diesters: concerning the publication of this work within HBM4EU project; Eva Govarts, MSc Existence of phthalate monoesters in South China house dust and implications (VITO), Laura Rodríguez Martín, MSc (VITO), and Dr. Nina Vogel (UBA) for discussions for human exposure. Environ Sci Technol. 2019;53:11675–83. concerning the work on phthalate time-trends in European population, which is 76. Wang HX, Zhou Y, Tang CX, He YH, Wu JG, Chen Y, et al. Urinary phthalate currently ongoing within the HBM4EU project. This project has received funding from metabolites are associated with body mass index and waist circumference in the European Union’s Horizon 2020 research and innovation programme under grant Chinese school children. Plos One. 2013;8:e56800. agreement No 733032, HBM4EU (www.HBM4EU.eu). This work was supported by the 77. Gong M, Weschler CJ, Liu L, Shen H, Huang L, Sundell J, et al. Phthalate meta- RECETOX research infrastructure (the Czech Ministry of Education, Youth and Sports: bolites in urine samples from Beijing children and correlations with phthalate LM2018121), CETOCOEN PLUS (CZ.02.1.01/0.0/0.0/15_003/0000469), and the Czech levels in their handwipes. Indoor Air. 2015;25:572–81. ministry of Education, Youth and Sports (CZ.02.1.01/0.0/0.0/17_043/0009632) and 78. Wu W, Zhou F, Wang Y, Ning Y, Yang JY, Zhou YK. Phthalate levels and related from the European Union’s Horizon 2020 research and innovation programme under factors in children aged 6–12 years. Environ Pollut. 2017;220:990–6. grant agreement No 857560, CETOCOEN Excellence. This publication reflects only the 79. Liao CX, Liu W, Zhang JL, Shi WM, Wang XY, Cai J, et al. Urine metabolites authors’ view and the European Commission is not responsible for any use that may of phthalate esters in 434 shanghai children and their associations with be made of the information it contains. Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. AUTHOR CONTRIBUTIONS Reprints and permission information is available at http://www.nature.com/ EDR and MS: study design, literature research and data compilation, contribution to reprints the design of the statistical analyses, interpretation of results, manuscript writing, manuscript editing and validation. Kl.K and JK: design and implementation of Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims statistical analyses, feedback on the interpretation of results, manuscript writing in published maps and institutional affiliations. (Statistical methods and related Figures), manuscript editing/validation. JB: validation of the study within HBM4EU project, feedback on the interpretation of results and discussion, manuscript editing and validation. SK and DS: validation of the study within HBM4EU project, manuscript editing/validation. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give COMPETING INTERESTS appropriate credit to the original author(s) and the source, provide a link to the Creative The authors declare no competing interests. Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory ADDITIONAL INFORMATION regulation or exceeds the permitted use, you will need to obtain permission directly Supplementary information The online version contains supplementary material from the copyright holder. To view a copy of this license, visit http://creativecommons. available at https://doi.org/10.1038/s41370-022-00441-w. org/licenses/by/4.0/. Correspondence and requests for materials should be addressed to Elena © The Author(s) 2022 Domínguez-Romero. Journal of Exposure Science & Environmental Epidemiology http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Exposure Science & Environmental Epidemiology Springer Journals

Time-trends in human urinary concentrations of phthalates and substitutes DEHT and DINCH in Asian and North American countries (2009–2019)

Loading next page...
 
/lp/springer-journals/time-trends-in-human-urinary-concentrations-of-phthalates-and-h7svuaEslS

References (117)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
ISSN
1559-0631
eISSN
1559-064X
DOI
10.1038/s41370-022-00441-w
Publisher site
See Article on Publisher Site

Abstract

Journal of Exposure Science & Environmental Epidemiology www.nature.com/jes ARTICLE OPEN Time-trends in human urinary concentrations of phthalates and substitutes DEHT and DINCH in Asian and North American countries (2009–2019) 1✉ 1 1 2 3,4 3,4,5 Elena Domínguez-Romero , Klára Komprdová ,Jiří Kalina , Jos Bessems , Spyros Karakitsios , Dimosthenis A. Sarigiannis and Martin Scheringer © The Author(s) 2022 BACKGROUND: Many phthalates are environmental pollutants and toxic to humans. Following phthalate regulations, human exposure to phthalates has globally decreased with time in European countries, the US and Korea. Conversely, exposure to their substitutes DEHT and/or DINCH has increased. In other countries, including China, little is known on the time-trends in human exposure to these plasticizers. OBJECTIVE: We aimed to estimate time-trends in the urinary concentrations of phthalates, DEHT, and DINCH metabolites, in general population from non-European countries, in the last decade. METHODS: We compiled human biomonitoring (HBM) data from 123 studies worldwide in a database termed “PhthaLit”. We analyzed time-trends in the urinary concentrations of the excreted metabolites of various phthalates as well as DEHT and DINCH per metabolite, age group, and country/region, in 2009–2019. Additionally, we compared urinary metabolites levels between continents. RESULTS: We found solid time-trends in adults and/or children from the US, Canada, China and Taiwan. DEHP metabolites decreased in the US and Canada. Conversely in Asia, 5oxo- and 5OH-MEHP (DEHP metabolites) increased in Chinese children. For low-weight phthalates, the trends showed a mixed picture between metabolites and countries. Notably, MnBP (a DnBP metabolite) increased in China. The phthalate substitutes DEHT and DINCH markedly increased in the US. SIGNIFICANCE: We addressed the major question of time-trends in human exposure to phthalates and their substitutes and compared the results in different countries worldwide. IMPACT: Phthalates account for more than 50% of the plasticizer world market. Because of their toxicity, some phthalates have been regulated. In turn, the consumption of non-phthalate substitutes, such as DEHT and DINCH, is growing. Currently, phthalates and their substitutes show high detection percentages in human urine. Concerning time-trends, several studies, mainly in Europe, show a global decrease in phthalate exposure, and an increase in the exposure to phthalate substitutes in the last decade. In this study, we address the important question of time-trends in human exposure to phthalates and their substitutes and compare the results in different countries worldwide. Keywords: Phthalate plasticizers; Phthalate substitutes; Time-trends; Human biomonitoring; Asia; North America. Journal of Exposure Science & Environmental Epidemiology; https://doi.org/10.1038/s41370-022-00441-w INTRODUCTION low molecular weight phthalates (e.g. diethyl phthalate, DEP; di- Phthalates are the most widely used plasticizers worldwide [1, 2]. isobutyl phthalate, DiBP; and di-n-butyl phthalate, DnBP) are used Between countries, the consumption of plasticizers is led by as solvents and/or plasticizers in adhesives, paints, lacquers, China (approximately 50% of the world consumption in 2020), printing inks, and personal care products [2, 3]. In the plastics, followed by other Asia/Pacific countries, Europe and North phthalates are not chemically bound and can migrate to the America [1]. Phthalates and major phthalate replacements such environment, thus becoming environmental pollutants. From the as 1,2-cyclohexanedicarboxylic acid, 1,2-diisononyl ester (DINCH) environment, humans are exposed to these substances through and di(2-ethylhexyl) terephthalate (DEHT) are principally used as different exposure routes (oral, dermal, inhalation) and sources plasticizers in polymers, principally in flexible PVC. Furthermore, [4–6]. Exposure to phthalates, depending on factors such as the 1 2 3 RECETOX, Faculty of Science, Masaryk University, Kotlarska 2, Brno 611 37, Czech Republic. VITO (Flemish Institute for Technological Research), BE-2400 Mol, Belgium. Aristotle Univ Thessaloniki, Dept Chem Engn, Environm Engn Lab, Univ Campus,Bldg D,Rm 201, Thessaloniki 54124, Greece. HERACLES Res Ctr Exposome & Hlth, Ctr Interdisciplinary Res & Innovat, Balkan Ctr, Bldg B,10thkm Thessaloniki Thermi Rd, Thessaloniki 57001, Greece. Sch Adv Study IUSS, Sci Technol & Soc Dept, Environm Hlth Engn, Piazza Vittoria 15, I-27100 Pavia, Italy. email: elena.dominguez-romero@recetox.muni.cz Received: 8 October 2021 Revised: 1 April 2022 Accepted: 6 April 2022 1234567890();,: E. Domínguez-Romero et al. exposure level and duration, may result in toxic outcomes in We found significant time-trends for all the main phthalates in humans and animals. Notably, toxic effects of phthalates on the China, Taiwan, Korea, the US, and Canada, and for DEHT and development and function of male and female reproductive DINCH metabolites in the US. systems [7–9], neurodevelopment [10–12], and metabolism [13] have been reported. Phthalate replacements such as DINCH and DEHT are also toxic [14–16]. Considering the toxicological MATERIAL AND METHODS properties of these substances, some phthalate uses have been Literature research We aimed to conduct a non-biased literature review to prepare a human regulated in different countries [17]. Importantly in the EU, biomonitoring (HBM) data compilation. In this sense, our literature research fourteen phthalates have been classified as substances of very was in agreement with general recommendations for time-trend reviews high concern (SVHC) in the Regulation (EC) No 1907/2006 on the [38]. We conducted the literature research on the Web of Science. The Registration, Evaluation, Authorisation and Restriction of Chemi- dates, research keywords, exclusion criteria, and results of the literature cals (REACH), Annex XIV, by reason of their reproductive toxicity research are summarized in the first Supplementary Information file (SI 1), [18]. For each of these substances, a “sunset date” has been Table S1. established, after which its placing on the market and use must be prohibited in general. Phthalates listed in Annex XIV include “PhthaLit” database structure notably: di(2-ethylhexyl) phthalate (DEHP), butyl benzyl phthalate We created the “PhthaLit” database as a detailed compilation of literature (BBP), DnBP, and DiBP, for which the sunset date was in February HBM data for phthalates, DEHT, and DINCH in the general population from 2015. For these four substances (DEHP, DBP, BBP and DIBP), their different countries worldwide. The main columns in the database included endocrine disrupting properties were also included in REACH, variables considered of interest for the analysis of HBM data for phthalates. Annex XIV, on November 23, 2021 [19, 20]. Moreover, dioctyl These variables are described below. General variables for traceability phthalate (DOP), diisononyl phtalate (DINP), and diisodecyl (reference, specific biomonitoring plan, and study identification number) are described in the section entitled: Data traceability in the “PhthaLit” phthalate (DIDP) have been included in REACH, Annex XVII [21], database. and are subject to specific restrictions on the manufacture, placing on the market, and use. Substances and abbreviations. In the literature, a diversity of definitions Phthalate regulations help to reduce environmental as well as and abbreviations exist for phthalates and their metabolites. The main human exposure to these substances. Nonetheless, phthalates are abbreviations and definitions found in the literature for these substances presently still ubiquitous environmental pollutants. For example, and their metabolites are shown in SI 1, Table S2. We here used one phthalates have been found at high detection percentages (> abbreviation per substance. Among phthalate replacements, our priority 90%) in urine from the general population in Asian, European, and was for DEHT and DINCH [39]. North American countries [22–24]. Considering the relatively short half-lives of this group of substances in humans, generally lower Phthalate concentrations in urine. In the current manuscript, we work on than 24-h in blood and urine [25], the high levels of detection in the central values of urinary concentrations of phthalates and phthalate different countries are the result of recent exposure to these replacements. As central values, for publications where medians and pollutants. Nevertheless, human exposure to phthalates seems to geometric means (GM) were reported, both values were compiled, due to be globally decreasing. In this sense, the urinary concentrations of the generally lognormal distribution of urinary phthalate concentrations metabolites formed from low-weight phthalates and DEHP have [23, 27, 34]. However, only one of these metrics (priority for median as a more robust estimate of central tendency, followed by GM) per source was decreased in the latest years in Germany [26, 27], Belgium [28, 29], used for data analysis. In the current manuscript, the number of central Denmark [23], Sweden [30], Italy [31], the United States [32], values that we used in our analyses is identified by the abbreviation “n”. Canada [33], and Korea [34]. Among high-weight phthalates, the This abbreviation should not be confounded with “N”, which refers to the urinary concentrations of DINP metabolites have diminished in study-specific sample size (cf. Sample size (N)). Denmark [23] and Germany [27], and increased in Sweden [30]. Concerning the units, unadjusted urinary concentrations (µg/L) were Concerning phthalate replacements, metabolites from DEHT and/ the most frequent data found in the literature, followed by data or DINCH increased in Denmark [23], Germany [35, 36], and the US corrected for urinary dilution, in µg/g creatinine. Although we compiled [32]. Importantly, new time-trend analyses for European countries data in the different units, we work on unadjusted urinary concentra- are conducted within the European Human Biomonitoring tions (µg/L) in the current manuscript. Besides, for one study which reported only creatinine-corrected data and showed the geometric Initiative HBM4EU. More information on the time-trends of mean of measured creatinine concentrations in the study population phthalates and phthalate replacements in the general population [40], we back-calculated the unadjusted concentrations (SI 1, section would be needed and would be helpful to interpret the S3.1. Equation S1) and included them in the analysis. Additionally, for effectiveness of various regulatory risk measurements in various few publications which reported exclusively specific-gravity (SG) parts of the world, like the authorisation and restriction in the EU adjusted data, we considered the SG-adjusted data to be equivalent to as mentioned earlier, the concentration limits for children’s toys unadjusted concentrations. In this sense, Runkel et al. (2020) [41] and childcare products and for food packaging in the US. Notably, showed little differences between SG-adjusted and unadjusted phtha- there is a gap on the time-trends of these substances in the late concentrations in adults and children. Chinese population [37]. In the literature, some publications reported the central values of urinary concentrations for specific substances to be below the limits of Our objective was to analyse time-trends in the urinary detection (LOD, µg/L) or quantification (LOQ, µg/L). In those sources, concentrations of phthalates, DINCH, and DEHT in the general censored data were generally corrected through substitution methods. population from non-European countries (including China), in the Specifically, the values <LOD or <LOQ were generally replaced by LOD/2 last decade. We conducted a literature review and compiled or LOQ/2, respectively [23, 24]. Similarly, other authors replaced human biomonitoring (HBM) data for phthalates and phthalate censored data by LOD/√2orLOQ/√2[32, 42]. Following the same replacements for different age groups and countries worldwide. In substitution methods, we replaced the central values <LOD or <LOQ by summary, data from 123 publications for metabolites from all LOD/2 and LOQ/2, respectively. More details are given in the SI 1, section phthalates, DEHT and DINCH in more than 30 countries were S3.2. Substances with > 50% of censored central values in the database compiled within the “PhthaLit” database. Based on the compiled (see Fig. S1 for adults) were excluded from further data analyses. central values (medians, geometric means), we analysed time- trends in urinary concentrations per metabolite, age group, and Sampling years. We compiled data for sampling years from 2009 onwards. country/region, in non-European countries. Additionally, we In the literature, most studies reported data for one sampling campaign (in other words, for one time-point). Additionally, some “time-trend” studies compared phthalate urinary concentrations in adults and children, (including cross-sectional studies with multiple time-points and longitudinal between continents (including Asia, North America, and Europe). Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. studies), which reported data for consecutive sampling campaigns number was used for data analysis (e.g. to calculate the median (consecutive time-points), were also found. We compiled data from both value if the same study reported separate data in male and female) types of studies. To be precise, in the reviewed HBM studies, one sampling and for the identification, in the Figures, of data from the same source campaign (i.e. one time-point) did not necessarily refer to one specific year (same colour). but to one period (e.g. Aug. 2015–Sep. 2016; 2013–2015…). In our database, the average year per sampling time point was determined and used for our analyses. However, for studies with long sampling campaigns, Statistical methods we considered that an average sampling year may not be representative Time-trend analysis. In order to calculate time-trends between 2009 and enough for the purposes of this study and could introduce an error in the 2019, the central values in the database were aggregated per country, age time-trend analyses. In this sense, during data compilation, we decided to group, metabolite, year, and study (Study ID number, see Data traceability exclude data reported for sampling time points (specific periods) longer in the “PhthaLit” database), providing one median value for each unique than 4 years. combination of these variables. This principle was applied to all data, including for studies which reported only separated data for variables not Population and age groups. We aimed to compile HBM data which may included in the analysis (e.g. separated data for men and women, or for be representative of the exposure levels for the general population in a different cities/areas), for which the median value per country, age group, country. Therefore, we did not collect data from more highly exposed metabolite, year, and study/survey was calculated from all data and used in groups, such as a) occupationally exposed workers, b) population from the time-trend analysis. Each median was also characterised by the sum of reported “hotspot” locations (i.e. including phthalate manipulation sites), sample sizes used for its computation. The study-specific sample size (N) and c) transfused patients. We also excluded data from subjects with was used to assign an importance to each median value within the analysis confirmed or suspected reproductive pathology (e.g. couples under- (weighting). A minimum sample size of 120 has been proposed for going infertility treatment) or chronic disease (e.g. obesity). For studies derivation of European Reference Values in a population, in order to enable where a specific group of population, not representative for the a proper estimation of the 95% CI of the 95th percentile [48–51]. Taking general population, was compared with a control group, data from the this value as a reference, we considered that central values (medians or controls were compiled. Finally, we did not compile data for pregnant geometric means) computed from 120 or more subjects from the general women, since pregnancy may have an impact on phthalate pharmaco- population per study group were fully representative. Therefore, depend- kinetics [43]. ing on N, we assigned a weight to each central value: We classified the data by age group [27, 44]. Specifically, the following groups were defined, depending on the average age (age) of the subjects: ● for studies with N >=120, the weight of the central values was 1; “adults” (18 years < age < 60 years), “children” (4 years <= age <= ● for studies with N < 120 (i.e., 20 <= N < 120), the weight of the central 12 years), “teenagers” (12 years < age <= 18 years), “seniors” (60 years <= values was ⌈N/120⌉, where the half-brackets ⌈⌉ represent the ceiling age), and “younger children” (age < 4 years). The main age groups for which operation (rounding up). HBM data for phthalates were found were those of adults, followed by children (cf. Results, PhthaLit database). Therefore, we considered both age Assuming an approximately log-normal distribution of the data, we groups as a priority for our study. logarithmically (ln) transformed the urinary concentrations, similarly to the approach by Frederiksen et al. [23]. Next, we looked for linear time-trends Countries and continents. In the database, the continent, country, and on these log-transformed data. Time-trends were investigated using the region (if it was shown), were specified for each datum. non-parametric Kendall test and the Theil-Sen trend estimator [48]. These In Asia, several studies published HBM data specifically for the robust methods are not prone to be biased by possible extreme values and Taiwanese population [40, 45–47]. As a consequence the number of data provide robust estimates of the presence and slope of the trend. available in Taiwan was relatively high (SI 1, Tables S3 and S4). For these Specifically, we applied these methods for a) time series including 5 or reasons, we analysed Taiwan separately. more central values (“n” ≥ 5), and for b) time-series with lower n (n = 3–4) In Europe, time-trend analyses for these plasticizers are conducted only if data from the national human biomonitoring plan in the country within the HBM4EU project. Thus, we excluded European countries from were available at the earliest and the latest years in the time-series. Results our time-trend analyses (cf. Statistical methods, Time-trend analysis). with p ≤ 0.10 were considered significant. Nevertheless, data from European countries were included in the “PhthaLit” database, and Europe was included for geographic comparisons Additional analyses. We analysed the possible impact of the gender, area, of phthalate levels between continents (cf. Statistical methods, additional and sample type on the estimated trends per metabolite, country, and age analyses). group. Specifically, all time-trend analyses were conducted separately per gender, area, and sample type, and compared to the original trends using Sample size (N). In the database, per study, we specified the number of Fisher z-scoring [52]. None of these variables had a significant impact in subjects who provided urine for phthalate measurements (i.e., any trend. The p-value of the trend difference test [52] was in all cases study-specificsamplesize, “N”). This variable (N) was used for data higher than 0.10. Further details are shown in the SI 1 (section S3.3). weighting in our time-trend analysis (cf. Statistical methods, Time-trend Furthermore, geographic comparisons in the urinary phthalate concen- analysis). No data were compiled for N < 20. trations between continents were conducted. Specifically, the central values (i.e., median, GM) of urinary concentrations per metabolite, continent, and age group, for two different periods (i.e., 2009–2014 and Other variables. Major variables explained above (i.e. measured sub- 2015–2019) were compared. Further details are given in the SI 1 stance, age group, country, and sampling year) created a high data (section S3.4). categorization. For this reason, other variables which were considered comparatively less important for our analysis, for example the gender, area (urban, suburban), or the sample type were not included as variables for time-trend calculations per country or geographic comparisons per RESULTS continent. Nevertheless, these variables were specified in the database “PhthaLit” database and we analysed the possible impact of the gender, sample type, and area The “PhthaLit” database, generated and analysed during this on the estimated trends (cf. Statistical methods, additional analyses). study, can be found as an Excel file of Supplementary Information (SI 2) with the current manuscript. Data traceability in the “PhthaLit” database. In the database, general Data from 123 publications were compiled, including variables for traceability were included, particularly: principally peer-reviewed publications, as well as data reported Reference short name and digital object identifier (DOI). by the US Centers for Disease Control and Prevention [42], Human biomonitoring (HBM) survey name. Health Canada [24], and Santé Publique France [53]. In Study identification number (Study ID number). This number identifies summary, 2287 compiled central values of unadjusted urinary data from the same source. Generally, one Study ID number identifies concentrations (µg/L) of phthalates and phthalate replacements data from one individual study. Nevertheless, for some studies from the “PhthaLit” database were available and used in the which reported data from the same HBM survey in a population, we time-trend analysis. Concerning the distribution of data among used the same Study ID number (see SI 1, Tables S5 and S6). This Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. –1 age groups, more unadjusted central values were found for 3–10% yr (Fig. S6a, c). In contrast, MiBP increased in US adults –1 adults (n = 1208) and children (n = 859), than for teenagers and children (+4.8% yr )(Fig. S6b). (n = 160) or other age groups. Between countries, a high number of data were found principally in China, followed by High-weight phthalates (DINP, DIDP), and MCPP (non-specific Germany (also Denmark for adults) and the United States, both metabolite). For DINP and DIDP metabolites, specifically for cx- –1 in adults and children. Among substances, more data were MINP and cx-MIDP, respectively, decreasing trends by 2–9% yr available for DEHP metabolites and low-weight phthalates than were found in the US population (Table 1). for high-weight phthalates and phthalate replacements Concerning MCPP, this is a non-specific metabolite that can be (Tables S3 and S4). formed through metabolism of several high- and low-weight phthalates [26]. We found decreasing trends for MCPP in the US, Time-trends Canada, and China (Fig. S7). Significant time-trends per substance, country/region, and age group, in percentages of change per year, are summarized in Tables 1 and 2. In these Tables, the specific periods studied and DISCUSSION data sources are reported. Only trends including <33% of We analysed time-trends in the urinary concentrations of censored data, one or more data points after 2015, for a total phthalates in general population from Asian and North American period >= 4 years are shown. countries, following a literature review and data compilation. Time-trends were found principally in adults and/or children Naturally, analysing data from multiple sources presented a from five countries/regions, including three in Asia (China, Taiwan, challenge based on the significant inter-study variability under- Korea) and two in North America (Canada and the US). Only 4 lying the data sets. To deal with this matter, it was essential to a) trends in US children, out of 58 significant trends, had p-values conduct a non-biased literature review and a thorough data between 0.05 and 0.10. The other 54 trends were significant at p < compilation, and b) select the main variables for the time-trend 0.05 (see Tables 1 and 2). Time-trends included generally 5 or analyses. As major variables, we included the metabolite, age more central values. Exceptionally, trends in Canada and some group, country, sampling year, and sample size. By including the trends in the US with a lower (n = 3–4) were also included, since major variables in our analysis, on the one hand we found globally data from the national human biomonitoring plans in those solid time-trends. On the other hand, this created a high data countries [24, 42] were available at the earliest and the latest categorization. As an illustration, significant trends per metabolite, sampling years per trend. Importantly, the trends in both North age group, and country were found “just” in 5 non-European American countries were globally consistent. countries/regions, and a similar number of European countries (not shown), out of approximately 30 countries for which data DEHP metabolites. Significant time-trends were found for the were compiled. To avoid further data categorization, other main DEHP metabolites (Table 1). variables (i.e., gender, sample type, area) were not included in In the US, Canada, and Korea, DEHP metabolites decreased with the analyses. Importantly, we tested the possible impact of those time. Notably in the US population and Canadian adults, the excluded variables (gender, area, sample type) on the estimated urinary concentrations of secondary DEHP metabolites consis- time-trends. None of these variables had a significant impact in tently decreased by approximately 8–15% per year (Fig. 1a). any trend (SI 1, section S3.3). Nevertheless, we found a bias for the In China, time-trends for DEHP metabolites diverged between time-trends in Korea, for different reasons. Specifically in Korea, age groups and metabolites. On the one hand, decreasing trends phthalate concentrations reported in 2016 by Lee I. et al. and Lee –1 were found for 5cx-MEPP in children (–13.7% yr ; Fig. 1b) and G. et al. [54, 55] were low, as compared to KoNEHS data [56] (see –1 adults (–5.7% yr ) (Fig. S2a). On the other hand, 5oxo- and 5OH- Fig. S3b). Importantly, Lee I. and Lee G. et al. measured phthalates MEHP increased by 11–15% per year in children (Figs. 1c and S3a). in spot urine collected in control women, following a fasting of at Data for 5oxo-MEHP and 5OH-MEHP in Chinese adults did not least 8 h [54, 55]. Fasting prior to the collection of spot urine seem to follow a linear time-trend pattern, and no significant (which was not first morning urine) was relatively rare in the trends were found (Fig. S2b). The results for DEHP in China are reviewed literature for phthalates. For instance, no other studies further discussed below. with this design were included in the “PhthaLit” database. Importantly, fasting may reduce the exposure to these substances. DINCH and DEHT metabolites. In the US population, phthalate In fact, the diet is considered as a major exposure source for replacements markedly increased with time. Specifically, signifi- certain phthalates [4–6]. Thus, the use of spot urine data following –1 cant increases were found for OH-MINCH in children (+62.9% yr ) fasting may have biased the trends in Korea, which were –1 and 5cx-MEPTP in adults (+51.0% yr ) (Fig. 2). overestimated, as compared to less pronounced decreasing trends suggested by KoNEHS data [34, 56]. Importantly, no bias Low-weight phthalates (DMP, DEP, DnBP, DiBP, BBP). For low- was found for the time-trends in China, Taiwan, the US, and weight phthalates, contrasting results between countries/regions Canada. and substances were found (Table 2). In China, Taiwan, and the US, A question remains concerning some inconsistent time-trends metabolites from some low-weight phthalates decreased while between DEHP metabolites in China. To clarify this question, we others increased. In Canada and Korea, low-weight phthalates need to mention a particularity of the data in China. Specifically in decreased with time. this country, we found more HBM data for DEHP and low-weight To begin with Asian countries/regions, MMP increased in phthalates, from a higher number of peer-reviewed studies, than in –1 –1 children from Taiwan (+35.8% yr )and China(+8.7% yr ) any other country. Conversely, we found no data from cross- (Figs. 3c and S4a). Importantly, in China, MnBP consistently sectional national human biomonitoring studies in China. This is in –1 –1 increasedbothinchildren (+34.4% yr ) and adults (+9.1% yr ) opposition to other countries, for which phthalate data from (Fig. 3a, b). Other low-weight phthalates decreased. Notably, national HBM studies were found (e.g. the US National Health and –1 MBzP consistently decreased in children from China (–33.3% yr ) Nutrition Examination Survey, NHANES; the Canadian Health –1 and Taiwan (–13.6% yr ) (Fig. S4b, c), MiBP and MEP decreased Measures Survey, CHMS; and the Korean National Environmental –1 –1 in Chinese adults (–13.4% yr and –6.9% yr ,respectively) Health Survey, KoNEHS) and included in the time-trends estimations. (Fig. S5). Importantly, the China National Human Biomonitoring (CNHBM) In the US population and in Canadian adults, low-weight survey started in 2017–2018 [57], and data from this survey may phthalate biomarkers generally decreased by approximately facilitate future time-trend analyses in this country. In our study, data Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. Table 1. Significant time-trends for metabolites formed from DEHP, high-weight phthalates, DEHT and DINCH. % Data sources Parent Meta- Country change (included in p-value subs- Age group n n period bolite /region per the PhthaLit tance year database) adults –5.7 0.003 11 8.8 2010–2019 (65-75) China (70, 76-82) children –13.7 0.010 7 5.5 2012–2017 5cx-MEPP adults –11.9 <0.001 11 9.8 2009–2016 (32, 42, 83) (42, 84) US children –9.1 <0.001 5 5 2010–2016 teenagers –14.8 <0.001 4 4 2010–2016 (42) (24, 59) Canada adults –11.4 0.044 3 3 2010–2017 (70, 76-82, 85- China children 10.9 <0.001 11 8.7 2011–2018 87) South adults –18.6 <0.001 7 6.7 2009–2016 (54-56, 88-91) 5OH-MEHP Korea adults –9.4 0.001 8 7.0 2009–2016 (32, 42) US children –8.3 <0.001 5 5 2010–2016 (42, 84) DEHP teenagers –14.4 <0.001 4 4 2010–2016 (42) Canada adults –10.3 0.044 3 3 2010–2017 (24, 59) (70, 76-82, 85- China children 14.9 <0.001 11 8.7 2011–2018 87) South (54-56, 88-91) adults –14.1 <0.001 7 6.7 2009–2016 5oxo-MEHP Korea (42) adults –11.4 <0.001 4 4 2010–2016 US children –8.2 <0.001 5 5 2010–2016 (42, 84) (42) teenagers –14.1 <0.001 4 4 2010–2016 Canada adults –10.1 0.044 3 3 2010–2017 (24, 59) adults –6.1 <0.001 4 4 2010–2016 (42) MEHP US children –4.0 0.081 5 5 2010–2016 (42, 84) teenagers –8.4 <0.001 4 4 2010–2016 (42) (42, 83) DINP cx-MINP US adults –2.2 0.016 7 6.8 2009–2016 adults –8.8 <0.001 4 4 2010–2016 (42) (42, 84) DIDP cx-MIDP US children –1.8 0.081 5 5 2010–2016 teenagers –3.3 <0.001 4 4 2010–2016 (42) Canada adults –12.8 <0.001 3 3 2010–2017 (24, 59) (65-67, 71, 72, China adults –27.3 <0.001 7 5.5 2010–2016 74, 92) Several MCPP a adults –17.2 <0.001 4 4 2010–2016 (42) (42, 84) US children –12 <0.001 5 5 2010–2016 teenagers –14.3 <0.001 4 4 2010–2016 (42) a a DINCH OH-MINCH US children 62.9 <0.001 4 4 2012–2016 (42, 84) DEHT 5cx-MEPTP US adults 51 <0.001 5 4 2009–2016 (32, 42) Green: decreasing trends; red: increasing trends. Only significant trends (p-value <= 0.10) including <33% of censored data, with one or more data points after 2015, for a total period >= 4 years are shown. n = number of central values aggregated per country, age group, metabolite, year, and Study ID number, per time-trend; n = number of “full-weight” data per trend, calculated as the sum of the number of aggregated central values (n) multiplied by their weight (Statistical methods, time-trend analysis). For trends with n < 5 (i.e., trends in Canada and some trends in the US), data from national human biomonitoring plans [24, 42] were used both at the earliest and the latest years per trend. For studies where a specific group of population was compared with a control group, data from controls were compiled. Trends in Korea were overestimated (see Discussion). Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. Table 2. Significant time-trends for metabolites formed from low-weight phthalates. % Data sources Parent Metaboli Country change (included in Age group p-value n n period substance te /region per the PhthaLit year database) Canada adults –5.4 0.044 3 3 2010–2017 (24, 59) (70, 76, 78-82, DMP MMP China children 8.7 <0.001 9 8.2 2011–2018 85-87) Taiwan children 35.8 <0.001 6 6 2009–2016 (40, 45, 47, 93) Canada adults –10.4 <0.001 3 3 2010–2017 (24, 59) (22, 65-75, 87, China adults –6.9 0.016 15 12.2 2010–2019 94, 95) DEP MEP adults –7.9 <0.001 4 4 2010–2016 (42) US children –3.9 0.081 5 5 2010–2016 (42, 84) teenagers –4.4 0.013 4 4 2010–2016 (42) Canada adults –2.6 <0.001 3 3 2010–2017 (24, 59) (65-75, 87, 94, China adults –13.4 0.001 14 11.2 2010–2019 95) South DiBP MiBP adults –26.3 <0.001 3 2.7 2009–2016 (54, 55, 88) Korea adults 4.8 0.013 4 4 2010–2016 (42) US children 4.8 <0.001 5 5 2010–2016 (42, 84) Canada adults –4.9 0.044 3 3 2010–2017 (24, 59) (22, 65-75, 87, China adults 9.1 <0.001 15 12.2 2010–2019 94, 95) (70, 76-82, 85- China children 34.4 <0.001 11 8.7 2011–2018 87) South DnBP MnBP adults <0.001 6 5.7 2009–2016 (54-56, 88-90) –18.5 Korea Taiwan children –2.4 0.008 6 6 2009–2016 (40, 45, 47, 93) children –4.6 0.081 5 5 2010–2016 (42, 84) US teenagers –7.4 <0.001 4 4 2010–2016 (42) Canada adults –8.1 <0.001 3 3 2010–2017 (24, 59) (70, 77-82, 86, China children –33.3 0.007 8 5.7 2013–2018 87) South adults –36.8 <0.001 5 4.7 2009–2016 (54-56, 88, 89) BBP MBzP Korea Taiwan children –13.6 <0.001 6 6 2009–2016 (40, 45, 47, 93) adults –5.7 <0.001 4 4 2010–2016 (42) US teenagers –8.3 <0.001 4 4 2010–2016 (42) Green: decreasing trends; red: increasing trends. Only significant trends (p-value < = 0.10) including <33% of censored data, with one or more data points after 2015, for a total period >= 4 years are shown. n = number of central values aggregated per country, age group, metabolite, year, and Study ID number per time-trend; n = number of “full-weight” data per trend, calculated as the sum of the number of aggregated central values (n) multiplied by their weight (Statistical methods, time-trend analysis). For trends with n < 5 (i.e., trends in Canada and some trends in the US), data from national human biomonitoring plans [24, 42] were used both at the earliest and the latest years per trend. For studies where a specific group of population was compared with a control group, data from controls were compiled. Trends in Korea were overestimated (see Discussion). in China presented quite large inter-study variability. Hence, for the whole study period and for the most recent years (i.e. since the number of data available per time-trend was particularly 2015), per trend. In this sense, among DEHP metabolites, we found important in this country. Specifically, to identify the most reliable more data for 5oxo-MEHP and 5OH-MEHP than for 5cx-MEPP, both time-trends in China, we analysed the number of central values (n) in adults and children (cf. Figs. 1b, c and S2). In adults, log- Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. a) 5cx-MEPP (DEHP metabolite) in US adults Sampling year Annual decrease (weighted) of 11.9% (p < 0.001) b) 5cx-MEPP (DEHP metabolite) in Chinese children c) 5oxo-MEHP (DEHP metabolite) in Chinese children Sampling year Sampling year Annual decrease (weighted) of 13.7% (p = 0.01) Annual increase (weighted) of 14.9% (p < 0.001) Fig. 1 Time-trends in the urinary concentrations of DEHP metabolites (µg/L). a 5cx-MEPP in US adults. b 5cx-MEPP in Chinese children. c 5oxo-MEHP in Chinese children. Each symbol is a central value (i.e. median or geometric mean). The size of the symbols is related to the study-specific sample size (N) and its weight in the time-trend calculation (see Statistical methods, time-trend analysis). Data from the same data source (Study ID number: see Data traceability in the “PhthaLit” database) are identified by symbols with the same colour in all panels and figures. Grey area: 90% confidence interval. a) b) OH-MINCH (DINCH metabolite) in US children 5cx-MEPTP (DEHT metabolite) in US adults Sampling year Sampling year Annual increase (weighted) of 63% (p < 0.001) Annual increase (weighted) of 51% (p < 0.001) Fig. 2 Time-trends in the urinary concentrations of phthalate replacement metabolites (µg/L). a OH-MINCH (a DINCH metabolite) in US children. b 5cx-MEPTP (a DEHT metabolite) in US adults. Each symbol is a central value (i.e. median or geometric mean). The size of the symbols is related to the study-specific sample size (N) and its weight in the time-trend calculation (see Statistical methods, time-trend analysis). Data from the same data source (Study ID number: see Data traceability in the “PhthaLit” database) are identified by symbols with the same colour in all panels and figures. Grey area: 90% confidence interval. a) b) MnBP (DnBP metabolite) in Chinese children MnBP (DnBP metabolite) in Chinese adults Sampling year Sampling year Annual increase (weighted) of 34.4% (p < 0.001) Annual increase (weighted) of 9.1% (p < 0.001) c) MMP (DMP metabolite) in Taiwanese children Sampling year Annual increase (weighted) of 35.8% (p < 0.001) Fig. 3 Time-trends in the urinary concentrations of low-weight phthalate metabolites (µg/L). a MnBP (DnBP metabolite) in Chinese children. b MnBP in Chinese adults. c MMP (DMP metabolite) in Taiwanese children. Each symbol is a central value (i.e. median or geometric mean). The size of the symbols is related to the study-specific sample size (N) and its weight in the time-trend calculation (see Statistical methods, time-trend analysis). Data from the same data source (Study ID number: see Data traceability in the “PhthaLit” database) are identified by symbols with the same colour in all panels and figures. Grey area: 90% confidence interval. Journal of Exposure Science & Environmental Epidemiology Concentration (µg/L) Concentration (µg/L) Concentration (µg/L) Concentration (µg/L) C) oncentration (µg/ L Concentration (µg/L) Concentration (µg/L) Concentration (µg/L) Concentration (µg/L) E. Domínguez-Romero et al. transformed data for 5oxo- and 5OH-MEHP did not follow a linear be noted that the increases for MnBP in Chinese population, MMP time-trend pattern and no significant trends were found. Conversely and 5OH-MEHP in Asian children occurred in a context where the in children, we found significant time-trends for 5OH- and 5oxo- average concentrations of these substances in the last years seemed MEHP (p < 0.001), which were more solid than those for 5cx-MEPP to be relatively high in Asia, as compared to data in other continents (p= 0.010). To summarize our results for DEHP metabolites in China, (SI 1, section S3.4). In conclusion, our results in China raise some we found that 5OH-MEHP and 5oxo-MEHP increased over time, concern and seem globally consistent with the leading position of while 5cx-MEPP seemed to decrease over time, which is unexpected Asia in the plasticizer global market, and the increasing importance since the three of them are metabolites of DEHP. Notably, this of China in this market (i.e. China represented approximately 42% of apparent decreasing time trend for 5cx-MEPP could be due to the worldwide consumption of plasticizers in 2017, and 50% in chemical-analytical flaws. It has been observed that 5cx-MEPP may 2020) [1, 62]. co-elute with OH-MINP (oral communication Dr Koch). OH-MINP To conclude with few recommendations for future research, more being a metabolite of the high molecular weight phthalate DINP HBM data for DEHT and DINCH would be needed worldwide. The might be decreasing and so cause confusion where the quantified quite large increase per year of DEHT and DINCH metabolites in peaks may be attributed (partly) mistakenly to 5cx-MEPP instead of the US is quite troublesome because of the recent knowledge on the OH-MINP. Concerning the trends for low-weight phthalates in China, toxicological properties that also these phthalate substitutes exhibit. a relatively high number of central values per metabolite and age Furthermore, the world consumption of phthalate replacements is group (n = 11–15) were available for the time-trend calculations for predicted to increase markedly in the following years (2021–2025) MnBP (adults and children), MEP and MiBP in adults, which supports [1]. For high-weight phthalates, more data would be desirable too, the solidity of these results (i.e. increasing trends for MnBP and with a focus on the secondary metabolites from these substances decreasing trends for MEP and MiBP). Importantly, MnBP consis- [63]. To finalize, environmental pollutants such as the studied tently increased in adults and children. This consistency between plasticizers are a global concern. In this context, as previously age groups highlights the reliability of these results, since the suggested [64], the scarcity of HBM data for phthalates, DEHT, and majority of data sources were independent between adults and DINCH in general population in some continents (Africa, South children (i.e., for MnBP, out of 11–15 data sources per age group, America, Oceania) and some countries worldwide was striking. only 2 sources were common to both age groups). For MMP and MBzP in Chinese children, a lower number of central values was found (n = 8–9). Interestingly for these two metabolites, we found DATA AVAILABILITY The “PhthaLit” database, generated and analysed during this study, can be found as consistently increasing trends for MMP and decreasing trends for an Excel file of Supplementary Information (SI 2) with the current manuscript. MBzP both in Chinese and Taiwanese children. To sum up, in the general populations from the US and Canada, phthalate metabolites generally decreased with time. Conversely, REFERENCES metabolites from phthalate replacements (DEHT, DINCH) increased 1. IHS_Markit. Plasticizers. Chemical Economics Handbook. Published May 2021. in the US. Globally, these results were in agreement with decreasing 2021 [Available from: https://ihsmarkit.com/products/plasticizers-chemical- trends for phthalate metabolites, and increasing trends for economics-handbook.html. metabolites of the substitutes DEHT and DINCH in population from 2. Rodríguez-Bernaldo-de-Quirós A, Lestido-Cardama A, Sendón R, García-Ibarra V. Puerto Rico [58], Germany [26, 27, 35, 36], and Denmark [23]. In Phthalates and alternative plasticisers. Food Contamination by Packaging: Canadian population, phthalate time-trends have been recently Migration of Chemicals from Food Contact Materials. Berlin: Walter De Gruyter reported by Pollock et al., 2021 [33]. Although there is a partial Gmbh; 2019. p. 45–94. overlap in data sources used by Pollock et al. (2021) and those used 3. Danish_EPA. Hazardous substances in plastics. Survey of chemical substances in consumer products No. 132, 2014. The Danish Environmental Protection Agency. in the current study, we additionally used data by Albert et al. 2018 Erik Hansen (COWI A/S), Nils Nilsson (Danish Technological Institute), and Kristine [59]. Other differences between our data analysis in Canada and the Slot Ravnholt Vium (Danish Technological Institute). ISBN no. 978-87-93283-31-2. study by Pollock et al. (2021) are for the time periods considered, the The Danish Environmental Protection Agency; 2014. age ranges, and the selection of metabolites. The upshot is that 4. Giovanoulis G, Bui T, Xu FC, Papadopoulou E, Padilla-Sanchez JA, Covaci A, et al. Pollock et al. [33] showed decreasing trends for DEHP metabolites, Multi-pathway human exposure assessment of phthalate esters and DINCH. low-weight phthalate metabolites (specifically, MEP, MnBP, and Environ Int. 2018;112:115–26. MBzP), and MCPP in Canadian population, which is consistent with 5. Wormuth M, Scheringer M, Vollenweider M, Hungerbuhler K. What are the our results. In addition, we found slight decreasing trends for MMP sources of exposure to eight frequently used phthalic acid esters in Europeans? and MiBP in this country (these metabolites were not assessed in the Risk Anal. 2006;26:803–24. 6. Koch HM, Lorber M, Christensen KLY, Palmke C, Koslitz S, Bruning T. Identifying study by Pollock et al. [33]). Interestingly, for low-weight phthalates, sources of phthalate exposure with human biomonitoring: Results of a 48 h we found diverging time-trends between substances in China, fasting study with urine collection and personal activity patterns. Int J Hyg Taiwan, and the US. In other words, in these countries/regions some Environ Health. 2013;216:672–81. low-weight phthalates increased while others decreased. These 7. Hlisnikova H, Petrovcova I, Kolena B, Sidlovska M, Sirotkin A. Effects and results might be related to a substitution in the uses of certain low- mechanisms of phthalates’ action on reproductive processes and reproductive weight phthalates by others. Notably, we found contrasting trends health: A literature review. Int J Environ Res Public Health. 2020;17:6811. between DnBP and DiBP monoesters in the US and China. For 8. Radke EG, Braun JM, Meeker JD, Cooper GS. Phthalate exposure and male example, in the US, MiBP increased while MnBP decreased with reproductive outcomes: A systematic review of the human epidemiological evi- time. Similarly, in men attending a fertility clinic in Boston between dence. Environ Int. 2018;121:764–93. 9. Radke EG, Glenn BS, Braun JM, Cooper GS. Phthalate exposure and female 2000 and 2017, MiBP increased while other phthalates generally reproductive and developmental outcomes: a systematic review of the human decreased [60]. In China, we globally found solid trends, excepted epidemiological evidence. Environ Int. 2019;130:19. for 5cx-MEPP for which chemical analytical flaws could not be 10. Hlisnikova H, Petrovicova I, Kolena B, Sidlovska M, Sirotkin A. Effects and excluded. The results in Chinese population suggested notably an mechanisms of phthalates’ action on neurological processes and neural health: a increasing exposure to DEHP, DnBP and DMP, and decreasing literature review. Pharmacol Rep. 2021;73:386–404. exposure to DEP and DiBP in China in the last decade. The increases 11. Radke EG, Braun JM, Nachman RM, Cooper GS. Phthalate exposure and neuro- in the urinary concentrations of DEHP metabolites in Chinese development: A systematic review and meta-analysis of human epidemiological population contrasted with decreasing trends in North American evidence. Environ Int. 2020;137:105408. and European countries. Interestingly, in Japanese children, the 12. Merced-Nieves FM, Dzwilewski KLC, Aguiar A, Musaad S, Korrick SA, Schantz SL. Associations of prenatal exposure to phthalates with measures of cognition in concentrations of phthalate metabolites (including DEHP metabo- 4.5-Month-old infants. Int J Environ Res Public Health. 2021;18:1838. lites) were stable between 2012 and 2017 [61]. To finalize, it should Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. 13. Radke EG, Galizia A, Thayer KA, Cooper GS. Phthalate exposure and metabolic 35. Kasper-Sonnenberg M, Koch HM, Apel P, Ruther M, Palmke C, Bruning T, et al. effects: a systematic review of the human epidemiological evidence. Environ Int. Time trend of exposure to the phthalate plasticizer substitute DINCH in Germany 2019;132:104768. from 1999 to 2017: Biomonitoring data on young adults from the Environmental 14. Danish_EPA. Alternatives to classified phthalates in medical devices. Environ- Specimen Bank (ESB). Int J Hyg Environ Health. 2019;222:1084–92. mental Project No. 1557. Danish Ministry of the Environment - Danish Environ- 36. Lessmann F, Kolossa-Gehring M, Apel P, Ruther M, Palmke C, Harth V, et al. mental Protection Agency. ISBN no. 978-87-93178-27-4. 2014. German Environmental Specimen Bank: 24-hour urine samples from 1999 to 15. Campioli E, Lee S, Lau M, Marques L, Papadopoulos V. Effect of prenatal DINCH 2017 reveal rapid increase in exposure to the para-phthalate plasticizer di(2- plasticizer exposure on rat offspring testicular function and metabolism. Sci Rep. ethylhexyl) terephthalate (DEHTP). Environ Int. 2019;132:105102. 2017;7:14. 37. Wang W, Leung AOW, Chu LH, Wong MH. Phthalates contamination in China: 16. Campioli E, Lau M, Papadopoulos V. Effect of subacute and prenatal DINCH Status, trends and human exposure-with an emphasis on oral intake. Environ plasticizer exposure on rat dams and male offspring hepatic function: The role of Pollut. 2018;238:771–82. PPAR-alpha. Environ Res. 2019;179:108773. 38. Sharma BM, Kalina J, Whaley P, Scheringer M. Towards guidelines for time-trend 17. Eichler CMA, Hubal EAC, Little JC. Assessing Human Exposure to Chemicals in reviews examining temporal variability in human biomonitoring data of pollu- Materials, Products and Articles: The International Risk Management Landscape tants. Environ Int. 2021;151:106437. for Phthalates. Environ Sci Technol. 2019;53:13583–97. 39. ChemSec. Replacing phthalates - Why and how to substitute this hard-to-spell 18. ECHA. Consolidated version of the Regulation (EC) No 1907/2006 of the European chemical group. International Chemical Secretariat (ChemSec), September Parliament and of the Council on the Registration, Evaluation, Authorisation and 2019; 2019. Restriction of Chemicals (REACH). European Chemicals Agency (ECHA) 2022 40. Su PH, Chen JY, Lin CY, Chen HY, Liao PC, Ying TH, et al. Sex steroid [Available from: https://echa.europa.eu/regulations/reach/legislation. hormone levels and reproductive development of eight-year-old children 19. EC. Harmful chemicals – updated list of endocrine disrupting substances (REACH following in utero and environmental exposure to phthalates. Plos One. 2014;9: Regulation). European Commission (EC). 2021 [Available from: https://ec.europa. e102788. eu/info/law/better-regulation/have-your-say/initiatives/12550-Harmful-chemicals 41. Runkel AA, Snoj-Tratnik J, Mazej D, Horvat M. Urinary phthalate concentrations in -updated-list-of-endocrine-disrupting-substances-REACH-Regulation-_en. the slovenian population: An attempt to exposure assessment of family units. 20. Chemical_Watch. Four phthalates lose authorisation exemptions after REACH Environ Res. 2020;186:13. Annex XIV entry amendment. 30 November 2021 2021 [Available from: https:// 42. CDC. Fourth National Report on Human Exposure to Environmental Chemicals, chemicalwatch.com/381308/four-phthalates-lose-authorisation-exemptions- Updated Tables, January 2019, Volume One. U.S. Department of Health and after-reach-annex-xiv-entry-amendment. Human Services. Centers for Disease Control and Prevention (CDC). 2019. 21. ECHA. Substances restricted under REACH 2021 [Available from: https://echa. 43. Zhao HZ, Li JF, Zhou YQ, Zhu L, Zheng YY, Xia W, et al. Investigation on Meta- europa.eu/substances-restricted-under-reach. bolism of Di(2-Ethylhexyl) Phthalate in Different Trimesters of Pregnant Women. 22. Xu HL, Wu XY, Liang CM, Shen J, Tao SM, Wen X, et al. Association of urinary Environ Sci Technol. 2018;52:12851–8. phthalates metabolites concentration with emotional symptoms in Chinese 44. Choi J, Knudsen LE, Mizrak S, Joas A. Identification of exposure to environ- university students. Environ Pollut. 2020;262:10. mental chemicals in children and older adults using human biomonitoring 23. Frederiksen H, Nielsen O, Koch HM, Skakkebaek NE, Juul A, Jorgensen N, et al. data sorted by age: Results from a literature review. Int J Hyg Environ Health. Changes in urinary excretion of phthalates, phthalate substitutes, bisphenols and 2017;220:282–98. other polychlorinated and phenolic substances in young Danish men; 45. Hou JW, Lin CL, Tsai YA, Chang CH, Liao KW, Yu CJ, et al. The effects of phthalate 2009–2017. Int J Hyg Environ Health. 2020;223:93–105. and nonylphenol exposure on body size and secondary sexual characteristics 24. Health_Canada. Fifth report on human biomonitoring of environmental during puberty. Int J Hyg Environ Health. 2015;218:603–15. chemicals in Canada. Results Can Health Measures Surv Cycle 5 (2016–2017). 46. Huang HB, Siao CY, Lo YTC, Shih SF, Lu CH, Huang PC. Mediation effects of thyroid 2019;2019:429. function in the associations between phthalate exposure and glucose metabo- 25. Domínguez-Romero E, Scheringer M. A review of phthalate pharmacokinetics in lism in adults. Environ Pollut. 2021;278:10. human and rat: what factors drive phthalate distribution and partitioning? Drug 47. Liao KW, Chang WH, Chou WC, Huang HB, Waits A, Chen PC, et al. Human Metab Rev. 2019;51:314–29. biomonitoring reference values and characteristics of Phthalate exposure in the 26. Koch HM, Ruther M, Schutze A, Conrad A, Palmke C, Apel P, et al. Phthalate general population of Taiwan: Taiwan Environmental Survey for Toxicants 2013- metabolites in 24-h urine samples of the German Environmental Specimen Bank 2016. Int J Hygiene. Environ Health. 2021;235:113769. (ESB) from 1988 to 2015 and a comparison with US NHANES data from 1999 to 48. HBM4EU. Statistical Analysis Plan. Deliverable Report D10.5. WP 10 – Data 2012. Int J Hyg Environ Health 2017;220:130–41. management and analysis. Version 6.0. M. Vrijheid, P. Montazeri, L. Rambaud, N. 27. Schwedler G, Rucic E, Lange R, Conrad A, Koch HM, Palmke C, et al. Phthalate Vogel, J. Vlaanderen, S. Remy, E. Govarts, G. Schoeters. HORIZON 2020 Pro- metabolites in urine of children and adolescents in Germany. Human biomoni- gramme, Contract No. 733032 HBM4EU. 2019. toring results of the German Environmental Survey GerES V, 2014-2017. Int J Hyg 49. Solberg HE. International Federation of Clinical Chemistry (IFCC), Scientific Environ Health 2020;225:11. Committee, Clinical Section. Approved Recommendation (1986) on the Theory of 28. Schoeters G, Govarts E, Bruckers L, Den Hond E, Nelen V, De Henauw S, et al. Reference Values. Part 1. The concept of Reference Values. J Clin Chem Clin Three cycles of human biomonitoring in Flanders - Time trends observed in the Biochem. 1987;25:337–42. Flemish Environment and Health Study. Int J Hyg Environ Health. 2017;220:36–45. 50. Solberg HE, Petitclerc C. International Federation of Clinical Chemistry (IFCC), 29. Koppen G, Franken C, Den Hond E, Plusquin M, Reimann B, Leermakers M, et al. Scientific Committee, Clinical Section. Approved recommendation (1988) on the Pooled analysis of genotoxicity markers in relation to exposure in the Flemish Theory of Reference Values. Part 3. Preparation of individuals and collection of Environment and Health Studies (FLEHS) between 1999 and 2018. Environ Res specimens for the production of Reference Values. J Clin Chem Clin Biochem. 2020;190:13. 1988;26:593–8. 30. Larsson K, Lindh CH, Jonsson BAG, Giovanoulis G, Bibi M, Bottai M, et al. 51. Solberg HE. The IFCC recommendation on estimation of reference intervals. The Phthalates, non-phthalate plasticizers and bisphenols in Swedish preschool dust RefVal Program. Clin Chem Lab Med. 2004;42:710–4. in relation to children’s exposure. Environ Int. 2017;102:114–24. 52. Fisher RA. Frequency Distribution of the Values of the Correlation Coefficient in 31. Tranfo G, Caporossi L, Pigini D, Capanna S, Papaleo B, Paci E. Temporal trends of Samples from an Indefinitely Large Population. Biometrika 1915;10:507–21. urinary phthalate concentrations in two populations: Effects of REACH author- 53. Santé_publique_France. Imprégnation de la population française par les ization after five years. Int J Environ Res Public Health. 2018;15:1950. phtalates. Programme national de biosurveillance, Esteban 2014-2016. Saint- 32. Silva MJ, Wong LY, Samandar E, Preau JL, Calafat AM, Ye XY. Exposure to di-2- Maurice: Santé publique France, septembre 2019. 51 p. Available in URL: www. ethylhexyl terephthalate in a convenience sample of US adults from 2000 to santepubliquefrance.fr. 2019. 2016. Arch Toxicol 2017;91:3287–91. 54. Lee I, Kim S, Park S, Mok S, Jeong Y, Moon HB, et al. Association of urinary 33. Pollock T, Karthikeyan S, Walker M, Werry K, St-Amand A. Trends in environ- phthalate metabolites and phenolics with adipokines and insulin resistance mental chemical concentrations in the Canadian population: Biomonitoring related markers among women of reproductive age. Sci Total Environ. data from the Canadian Health Measures Survey 2007–2017. Environ Int 2019;688:1319–26. 2021;155:10. 55. Lee G, Kim S, Bastiaensen M, Malarvannan G, Poma G, Casero NC, et al. Exposure 34. Park C, Hwang M, Baek Y, Jung S, Lee Y, Paek D, et al. Urinary phthalate meta- to organophosphate esters, phthalates, and alternative plasticizers in association bolite and bisphenol A levels in the Korean adult population in association with with uterine fibroids. Environ Res. 2020;189:109874. sociodemographic and behavioral characteristics: Korean National Environmental 56. Lim S. The associations between personal care products use and urinary con- Health Survey (KoNEHS) 2012-2014. Int J Hyg Environ Health. 2019;222:903–10. centrations of phthalates, parabens, and triclosan in various age groups: The Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. Korean National Environmental Health Survey Cycle 3 2015-2017. Sci Total ventilation habits. 10th Int Symp Heat, Vent Air Conditioning, Ishvac2017. Environ. 2020;742:140640. 2017;205:1146–51. 57. Cao ZJ, Lin SB, Zhao F, Lv YB, Qu YL, Hu XJ, et al. Cohort profile: China National 80. Liao CX,Liu W, Zhang JL,Shi WM,Wang XY, CaiJ,et al. Associations of Human Biomonitoring (CNHBM)-A nationally representative, prospective cohort urinary phthalate metabolites with residential characteristics, lifestyles, and in Chinese population. Environ Int. 2021;146:10. dietary habits among young children in Shanghai, China. Sci Total Environ. 58. Rodríguez-Carmona Y, Ashrap P, Calafat AM, Ye XY, Rosario Z, Bedrosian LD, et al. 2018;616:1288–97. Determinants and characterization of exposure to phthalates, DEHTP and DINCH 81. Yao Y, Chen DY, Wu Y, Zhou L, Cheng JQ, Ma YY, et al. Urinary phthalate among pregnant women in the PROTECT birth cohort in Puerto Rico. J Exposure metabolites in primary school starters in Pearl River Delta, China: Occurrences, Sci Environ Epidemiol. 2020;30:56–69. risks and possible sources. Environ Res. 2019;179;108853;1–8. 59. Albert O, Huang JY, Aleksa K, Hales BF, Goodyer CG, Robaire B, et al. Exposure to 82. Yao Y, Chen DY, Yin JW, Zhou L, Cheng JQ, Lu SY, et al. Phthalate exposure linked polybrominated diphenyl ethers and phthalates in healthy men living in the to high blood pressure in Chinese children. Environ Int. 2020;143:8. greater Montreal area: A study of hormonal balance and semen quality. Environ 83. Silva MJ, Jia T, Samandar E, Preau JL, Calafat AM. Environmental exposure to the Int. 2018;116:165–75. plasticizer 1,2-cyclohexane dicarboxylic acid, diisononyl ester (DINCH) in US 60. Minguez-Alarcon L, Williams PL, Chiu YH, Gaskins AJ, Nassan FL, Dadd R, et al. adults (2000-2012). Environ Res. 2013;126:159–63. Secular trends in semen parameters among men attending a fertility 84. Hammel SC, Levasseur JL, Hoffman K, Phillips AL, Lorenzo AM, Calafat AM, et al. center between 2000 and 2017: Identifying potential predictors. Environ Int. Children’s exposure to phthalates and non-phthalate plasticizers in the home: 2018;121:1297–303. The TESIE study. Environ Int. 2019;132:10. 61. Ketema RM, Bamai YA, Ikeda-Araki A, Saito T, Kishi R. Secular trends of urinary 85. Zhang YH, Cao Y, Shi HJ, Jiang XX, Zhao Y, Fang X, et al. Could exposure to phthalate metabolites in 7-year old children and association with building phthalates speed up or delay pubertal onset and development? A 1.5-year characteristics: Hokkaido study on environment and children’s health. Int J Hyg follow-up of a school-based population. Environ Int. 2015;83:41–9. Environ Health. 2021;234:9. 86. Yu YX, Peng MM, Liu YL, Ma JJ, Wang N, Ma ST, et al. Co-exposure to polycyclic 62. IHS_Markit. Plasticizers. Chemical Economics Handbook. Published May aromatic hydrocarbons and phthalates and their associations with oxidative 2018 2018 [Available from: https://learning.ihsmarkit.com/courses/plasticizers- stress damage in school children from South China. J Hazard Mater. 2021; chemical-economics-handbook. 401:123390. 63. Calafat AM, Wong LY, Silva MJ, Samandar E, Preau JL, Jia LT, et al. Selecting 87. Zhang X, Tang S, Qiu T, Hu XJ, Lu YF, Du P, et al. Investigation of phthalate adequate exposure biomarkers of diisononyl and diisodecyl phthalates: Data metabolites in urine and daily phthalate intakes among three age groups in from the 2005-2006 National health and nutrition examination survey. Environ Beijing, China. Environ Poll. 2020;260:114005. Health Perspect. 2011;119:50–5. 88. Lee KM, Kho Y, Kim PG, Park SH, Lee JH. Urinary levels of phthalate metabolites 64. Wang Y, Zhu HK, Kannan K. A review of biomonitoring of phthalate exposures. and associations with demographic characteristics in Korean adults. Environ Sci Toxics 2019;7:28. Pollut Res. 2017;24:14669–81. 65. Gao CJ, Liu LY, Ma WL, Ren NQ, Guo Y, Zhu NZ, et al. Phthalate metabolites in 89. Choi W, Kim S, Baek YW, Choi K, Lee K, Seung DY. Exposure to environmental urine of Chinese young adults: Concentration, profile, exposure and cumulative chemicals among Korean adults-updates from the second Korean National risk assessment. Sci Total Environ. 2016;543:19–27. Environmental Health Survey (2012–2014). Int J Hyg Environ Health. 66. Guo Y, Alomirah H, Cho HS, Minh TB, Mohd MA, Nakata H, et al. Occurrence of 2017;220:29–35. phthalate metabolites in human urine from several asian countries. Environ Sci 90. Park C, Hwang M, Kim H, Ryu S, Lee K, Choi K, et al. Early snapshot on exposure to Technol. 2011;45:3138–44. environmental chemicals among Korean adults-results of the first Korean 67. Guo Y, Wu Q, Kannan K. Phthalate metabolites in urine from China, and impli- National Environmental Health Survey (2009-2011). Int J Hyg Environ Health. cations for human exposures. Environ Int. 2011;37:893–8. 2016;219:398–404. 68. Sun J, Chen B, Zhang LQ, Zhao D, Li SG. Phthalate ester concentrations in blood 91. Song NR, On JW, Lee J, Park JD, Kwon HJ, Yoon HJ, et al. Biomonitoring of urinary serum, urine and endometrial tissues of Chinese endometriosis patients. Int J Clin di(2-ethylhexyl) phthalate metabolites of mother and child pairs in South Korea. Exp Med. 2016;9:3808–19. Environ Int. 2013;54:65–73. 69. Chen JS, Zhou XF, Zhang H, Liu YM, Cao C, Dong RH, et al. Association between 92. Ding MY, Kang QY, Zhang SY, Zhao FR, Mu D, Zhang HF, et al. Contribution of urinary concentration of phthalate metabolites and impaired renal function in phthalates and phthalate monoesters from drinking water to daily intakes for the Shanghai adults. Environ Pollut. 2019;245:149–62. general population. Chemosphere 2019;229:125–31. 70. Dong RH, Zheng JH, Zhang MR, Chen JS, Zhang H, Gao X, et al. The concentra- 93. Huang PC, Chang WH, Wu MT, Chen ML, Wang IJ, Shih SF, et al. Characterization tions and cumulative risk assessment of phthalates in general population from of phthalate exposure in relation to serum thyroid and growth hormones, and Shanghai: The comparison between groups with different ages. Sci Total Environ. estimated daily intake levels in children exposed to phthalate-tainted products: A 2018;637:871–80. longitudinal cohort study. Environ Pollut. 2020;264:11. 71. Chen Q, Yang H, Zhou NY, Sun L, Bao HQ, Tan L, et al. Phthalate exposure, even 94. Yang LY, Yin WJ, Li P, Hu C, Hou J, Wang L, et al. Seasonal exposure to phthalates below US EPA reference doses, was associated with semen quality and repro- and inflammatory parameters: A pilot study with repeated measures. Ecotoxicol ductive hormones: Prospective MARHCS study in general population. Environ Int. Environ Saf. 2021;208:10. 2017;104:58–68. 95. Li XJ, Zhong Y, He WY, Huang SY, Li Q, Guo CS, et al. Co-exposure and health risks 72. Zhang B, Zhang T, Duan YS, Zhao Z, Huang XF, Bai XY, et al. Human exposure to of parabens, bisphenols, triclosan, phthalate metabolites and hydroxyl polycyclic phthalate esters associated with e-waste dismantling: Exposure levels, sources, aromatic hydrocarbons based on simultaneous detection in urine samples from and risk assessment. Environ Int. 2019;124:1–9. Guangzhou, south China. Environ Poll. 2021;272:115990. 73. Li JF, Zhao HZ, Xia W, Zhou YQ, Xu SQ, Cai ZW. Nine phthalate metabolites in human urine for the comparison of health risk between population groups with different water consumptions. Sci Total Environ. 2019;649:1532–40. 74. Zhang XM, Lou XY, Wu LH, Huang C, Chen D, Guo Y. Urinary phthalate meta- ACKNOWLEDGEMENTS bolites and environmental phenols in university students in South China. Environ The authors acknowledge Dr. Holger Koch for discussions concerning the chemical Res. 2018;165:32–9. analysis of phthalate metabolites; Prof. Jana Klánová (RECETOX) for support 75. Liu XT, Peng CF, Shi YM, Tan HL, Tang SQ, Chen D. Beyond phthalate diesters: concerning the publication of this work within HBM4EU project; Eva Govarts, MSc Existence of phthalate monoesters in South China house dust and implications (VITO), Laura Rodríguez Martín, MSc (VITO), and Dr. Nina Vogel (UBA) for discussions for human exposure. Environ Sci Technol. 2019;53:11675–83. concerning the work on phthalate time-trends in European population, which is 76. Wang HX, Zhou Y, Tang CX, He YH, Wu JG, Chen Y, et al. Urinary phthalate currently ongoing within the HBM4EU project. This project has received funding from metabolites are associated with body mass index and waist circumference in the European Union’s Horizon 2020 research and innovation programme under grant Chinese school children. Plos One. 2013;8:e56800. agreement No 733032, HBM4EU (www.HBM4EU.eu). This work was supported by the 77. Gong M, Weschler CJ, Liu L, Shen H, Huang L, Sundell J, et al. Phthalate meta- RECETOX research infrastructure (the Czech Ministry of Education, Youth and Sports: bolites in urine samples from Beijing children and correlations with phthalate LM2018121), CETOCOEN PLUS (CZ.02.1.01/0.0/0.0/15_003/0000469), and the Czech levels in their handwipes. Indoor Air. 2015;25:572–81. ministry of Education, Youth and Sports (CZ.02.1.01/0.0/0.0/17_043/0009632) and 78. Wu W, Zhou F, Wang Y, Ning Y, Yang JY, Zhou YK. Phthalate levels and related from the European Union’s Horizon 2020 research and innovation programme under factors in children aged 6–12 years. Environ Pollut. 2017;220:990–6. grant agreement No 857560, CETOCOEN Excellence. This publication reflects only the 79. Liao CX, Liu W, Zhang JL, Shi WM, Wang XY, Cai J, et al. Urine metabolites authors’ view and the European Commission is not responsible for any use that may of phthalate esters in 434 shanghai children and their associations with be made of the information it contains. Journal of Exposure Science & Environmental Epidemiology E. Domínguez-Romero et al. AUTHOR CONTRIBUTIONS Reprints and permission information is available at http://www.nature.com/ EDR and MS: study design, literature research and data compilation, contribution to reprints the design of the statistical analyses, interpretation of results, manuscript writing, manuscript editing and validation. Kl.K and JK: design and implementation of Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims statistical analyses, feedback on the interpretation of results, manuscript writing in published maps and institutional affiliations. (Statistical methods and related Figures), manuscript editing/validation. JB: validation of the study within HBM4EU project, feedback on the interpretation of results and discussion, manuscript editing and validation. SK and DS: validation of the study within HBM4EU project, manuscript editing/validation. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give COMPETING INTERESTS appropriate credit to the original author(s) and the source, provide a link to the Creative The authors declare no competing interests. Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory ADDITIONAL INFORMATION regulation or exceeds the permitted use, you will need to obtain permission directly Supplementary information The online version contains supplementary material from the copyright holder. To view a copy of this license, visit http://creativecommons. available at https://doi.org/10.1038/s41370-022-00441-w. org/licenses/by/4.0/. Correspondence and requests for materials should be addressed to Elena © The Author(s) 2022 Domínguez-Romero. Journal of Exposure Science & Environmental Epidemiology

Journal

Journal of Exposure Science & Environmental EpidemiologySpringer Journals

Published: Mar 1, 2023

Keywords: Phthalate plasticizers; Phthalate substitutes; Time-trends; Human biomonitoring; Asia; North America.

There are no references for this article.