Association between Urinary Haloacetic Acid Concentrations and Liver Injury among Women: Results from the Tongji Reproductive and Environmental (TREE) Study.

BACKGROUND: Experimental studies have shown that disinfection byproducts (DBPs) including haloacetic acids (HAAs) can cause liver toxicity, but evidence linking this association in humans is sparse. OBJECTIVES: We aimed to explore the associations between HAA exposures and liver injury. METHODS: We included 922 women between December 2018 and January 2020 from the Tongji Reproductive and Environmental (TREE) cohort study in Wuhan, China. Urinary HAA concentrations including trichloroacetic acid (TCAA) and dichloroacetic acid (DCAA) and serum indicators of liver function, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma-glutamyltransferase (GGT) were measured. Liver injury was defined as if any of serum indicator levels were above the 90th percentile. Multivariate logistic and linear regression models were fitted to assess the associations of urinary HAA concentrations with the risk of liver injury and liver function indicators. Stratified analyses by age, body mass index (BMI), alcohol use, and passive smoking were also applied to evaluate the potential effect modifiers. RESULTS: There is little evidence of associations of urinary TCAA concentrations with liver injury risk and liver function indicators. However, urinary DCAA concentrations were associated with a higher risk of liver injury [odds ratios (OR) for 1-interquartile range (IQR) increase in natural log (ln) transformed DCAA concentrations: 1.45; 95% confidence interval (CI): 1.07, 1.98]. This association was observed only among nondrinkers (pinteraction=0.058). We also found that a 1-IQR increase in ln-transformed DCAA concentrations was positively associated with ALT levels (percentage change=6.06%; 95% CI: 0.48%, 11.95%) and negatively associated with AST/ALT (percentage change=-4.48%; 95% CI: -7.80%, -1.04%). In addition, urinary DCAA concentrations in relation to higher GGT levels was observed only among passive smokers (pinteraction=0.040). CONCLUSION: Our findings suggest that exposure to DCAA but not TCAA is associated with liver injury among women undergoing assisted reproductive technology. https://doi.org/10.1289/EHP13386.

Associations of exposure to disinfection by-products with blood coagulation parameters among women: Results from the Tongji reproductive and environmental (TREE) study.

BACKGROUND: Experimental studies have shown that disinfection byproducts (DBPs) induce coagulotoxicity, but human evidence is scarce. OBJECTIVE: This study aimed to explore the relationships of DBP exposures with blood coagulation parameters. METHODS: Among 858 women from the Tongji Reproductive and Environmental (TREE) study, urinary dichloroacetic acid (DCAA) and trichloroacetic acid (TCAA) were detected as internal biomarkers of DBP exposures. We measured activated partial thromboplastin time (APTT), fibrinogen (Fbg), international normalized ratio (INR), prothrombin time (PT), and thrombin time (TT) as blood coagulation parameters. Multivariable linear regression models were utilized to estimate the relationships between urinary DCAA and TCAA and blood coagulation parameters. The effect modifications by demographic and lifestyle characteristics were further explored. RESULTS: Elevated tertiles of urinary DCAA concentrations were associated with increased PT and INR (11.29%, 95% CI: 1.66%, 20.92% and 0.99%, 95% CI: 0.08%, 1.90% for the third vs. first tertile, respectively; both P for trends < 0.05). Stratification analysis showed that the positive associations were only observed among younger (< 30 years), leaner (body mass index < 24.0 kg/m2), and non-passive smoking women. Moreover, elevated tertiles of urinary TCAA concentrations in positive associations with PT and INR were observed among younger women (17.89%, 95% CI: 2.50%, 33.29% and 1.82%, 95% CI: 0.34%, 3.30% for the third vs. first tertile, respectively; both P for trends < 0.05) but not among older women (both P for interactions < 0.05). CONCLUSION: Higher levels of urinary DCAA and TCAA are associated with prolonged clotting time among women.

Individual and joint associations of urinary phthalate metabolites with polycystic ovary and polycystic ovary syndrome: Results from the TREE cohort.

Phthalates are widespread endocrine disrupting chemicals that adversely affect female reproductive health. We aimed to investigate the individual and joint associations of phthalate exposures measured by repeated urinary metabolites with polycystic ovary (PCO) and polycystic ovary syndrome (PCOS) (96 PCO cases, 96 PCOS cases and 370 controls). In single-pollutant analyses, mono-isobutyl phthalate (MiBP), monobenzyl phthalate (MBzP) and the sum of di(2-ethylhexyl) phthalate (∑DEHP) were associated with increased prevalence of PCO. Mono(2-ethylhexyl) phthalate (MEHP), MBzP and ∑DEHP were associated with elevated prevalence of PCOS. In multiple-pollutant analyses, one-quartile increase of weighted quantile sum index in phthalate metabolite mixtures was associated with increased prevalence of PCO and PCOS, and MBzP was the most major contributor. Our findings suggest a potential role for phthalate exposures, both individually and in mixtures, in the development of PCO and PCOS.

Oxidatively generated DNA damage mediates the associations of exposure to phthalates with uterine fibroids and endometriosis: Findings from TREE cohort.

BACKGROUND: Epidemiological studies on phthalate exposures in associations with uterine fibroids (UF) and endometriosis (EMT) are inconsistent. The underlying mechanisms are poorly understood. OBJECTIVES: To investigate the relationships of urinary phthalate metabolites with UF and EMT risks, and further to examine the mediating role of oxidative stress. METHODS: This study included 83 and 47 women separately diagnosed with UF and EMT, as well as 226 controls from the Tongji Reproductive and Environmental (TREE) cohort. Two spot urine samples from each woman were analyzed for two oxidative stress indicators and eight urinary phthalate metabolites. Unconditional logistic regression models or multivariate regression models were fitted to evaluate the associations among phthalate exposures, oxidative stress indicators, and the risks of UF and EMT. The potential mediating role of oxidative stress was estimated by the mediation analyses. RESULTS: We observed that each ln-unit increase in urinary mono-benzyl phthalate (MBzP) concentrations was associated with increased UF risk [adjusted OR (aOR): 1.56, 95% CI: 1.20, 2.02], and that each ln-unit increase in urinary MBzP (aOR: 1.48, 95% CI: 1.09, 1.99), mono-isobutyl phthalate (MiBP) (aOR: 1.83, 95% CI: 1.19, 2.82), and mono-2-ethylhexyl phthalate (MEHP) (aOR: 1.66, 95% CI: 1.19, 2.31) concentrations were associated with increased EMT risk (all FDR-adjusted P < 0.05). Moreover, we observed that all tested urinary phthalate metabolites were positively associated with two oxidative stress indicators [4-hydroxy-2-nonenal-mercapturic acid (4-HNE-MA) and 8-hydroxy-2-deoxyguanosine (8-OHdG)], in which 8-OHdG was associated with increased risks of UF and EMT (all FDR-adjusted P < 0.05). The mediation analyses showed that 8-OHdG mediated the positive relationships of MBzP with UF risk, and of MiBP, MBzP, and MEHP with EMT risk, with the estimated intermediary proportion ranging from 32.7% to 48.1%. CONCLUSIONS: Oxidatively generated DNA damage may mediate the positive associations of certain phthalate exposures with the risks of UF and EMT. However, further investigation is warranted to confirm these findings.

Within-day variability, predictors, and risk assessments of exposure to parabens among Chinese adult men.

BACKGROUND: Parabens, as suspected endocrine disruptors, are widely used in personal care products and pharmaceuticals. However, variability, predictors, and risk assessments of human exposure to parabens are not well characterized. OBJECTIVE: To evaluate within-day variability, predictors, and risk assessments of exposure to parabens among Chinese adult men. METHODS: We measured four parabens including methylparaben (MeP), ethylparaben (EtP), propylparaben (PrP), and butylparaben (BuP) in repeated urine samples from 850 Chinese adult men. We examined the variability by intraclass correlation coefficients (ICCs) and identified the predictors by multivariable linear mixed models. We assessed risks of paraben exposures based on the estimated daily intake (EDI). RESULTS: The four parabens were detected in >76% of urinary samples. We observed fair to good to high reproducibility (ICCs: 0.71 to 0.86) for urinary paraben concentrations within one day. Use of facial cleanser was associated with higher four urinary paraben concentrations. Increasing age, taking medicine, intravenous injection, and interior decoration in the workplace were related to higher urinary concentrations of specific parabens. Smoking and drinking were associated with lower urinary concentrations of specific parabens. The maximum EDIs for the four parabens ranged from 13.76 to 848.68 µg/kg bw/day, and 0.9% of participants had the hazard quotient values > 1 driven by PrP exposure. CONCLUSIONS: Urinary paraben concentrations were less variable within one day. Several lifestyle characteristics including use of facial cleanser and pharmaceuticals may contribute to paraben exposures.

Urinary concentrations of phenols, oxidative stress biomarkers and thyroid cancer: Exploring associations and mediation effects.

Phenols have been shown to influence the cellular proliferation and function of thyroid in experimental models. However, few human studies have investigated the association between phenol exposure and thyroid cancer, and the underlying mechanisms are also poorly understood. We conducted a case-control study by age- and sex-matching 143 thyroid cancer and 224 controls to investigate the associations between phenol exposures and the risk of thyroid cancer, and further to explore the mediating role of oxidative stress. We found that elevated urinary triclosan (TCS), bisphenol A (BPA) and bisphenol S (BPS) levels were associated with increased risk of thyroid cancer (all P for trends < 0.05), and the adjusted odds ratios (ORs) comparing the extreme exposure groups were 3.52 (95% confidence interval (CI): 2.08, 5.95), 2.06 (95% CI: 1.06, 3.97) and 7.15 (95% CI: 3.12, 16.40), respectively. Positive associations were also observed between urinary TCS, BPA and BPS and three oxidative stress biomarkers measured by 8-hydroxy-2'-deoxyguanosine (8-OHdG), 8-iso-prostaglandin F2α (8-isoPGF2α) and 4-hydroxy-2-nonenal-mercapturic acid (HNE-MA), as well as between urinary 8-isoPGF2α and HNE-MA and the risk of thyroid cancer. Mediation analysis showed that urinary 8-isoPGF2α mediated 28.95%, 47.06% and 31.08% of the associations between TCS, BPA and BPS exposures and the risk of thyroid cancer, respectively (all P < 0.05). Our results suggest that exposure to TCS, BPA and BPS may be associated with increased risk of thyroid cancer and lipid peroxidation may be an intermediate mechanism. Further studies are warranted to confirm the findings.

Serum multiple organochlorine pesticides in relation to testosterone concentrations among Chinese men from an infertility clinic.

BACKGROUND: Organochlorine pesticides (OCPs) are endocrine-disrupting chemicals and may alter male reproductive hormone concentrations. OBJECTIVE: To explore the associations between multiple OCP exposures and serum testosterone concentrations among Chinese men. METHODS: We investigated 421 men who provided serum samples from an infertility clinic in Wuhan, China. Each man completed a questionnaire concerning demographic characteristics and lifestyle habits. Serum concentrations of 18 OCPs and total testosterone were measured. Linear regression models were used to explore whether serum OCP levels were associated with altered testosterone concentrations, and potential effect modifications by age and body mass index (BMI) were also examined. RESULTS: After adjusting for potential confounders, elevated dieldrin and p,p'-DDD levels had monotonically negative and positive exposure-response associations with testosterone concentrations, respectively (-30.98 ng/dL, 95% CI: -72.34, 10.37; P for trend = 0.12 and 41.31 ng/dL, 95% CI: -0.32, 82.93; P for trend = 0.06 for the highest vs. non-detectable exposure category, respectively). After stratification by age and BMI, dieldrin in relation to decreased testosterone concentrations was more pronounced among men aged <30 years old, and p,p'-DDD in relation to increased testosterone concentrations was stronger among men aged ≥30 years old and among men with a BMI <24 kg/m2, though the differences were not statistically significant. CONCLUSION: The study found evidence that serum dieldrin and p,p'-DDD concentrations might be associated with altered serum testosterone concentrations.

Exposure to disinfection by-products and reproductive hormones among women: Results from the Tongji Reproductive and Environmental (TREE) study.

BACKGROUND: Disinfection by-products (DBPs) have been shown to impair female reproductive function. However, epidemiological evidence on reproductive hormones is scarce. OBJECTIVE: To investigate the associations between DBP exposures and reproductive hormones among women undergoing assisted reproductive technology. METHODS: We included 725 women from the Tongji Reproductive and Environmental (TREE) Study, an ongoing cohort conducted in Wuhan, China during December 2018 and January 2020. Urine samples collected at recruitment were quantified for dichloroacetic acid (DCAA) and trichloroacetic acid (TCAA) as biomarkers of DBP exposures. At day 2-5 of menstruation, serum reproductive hormones including luteinizing hormone (LH), estradiol (E2), total testosterone (T), progesterone (PRGE), and prolactin (PRL) were determined. Multivariate linear regression models were performed to assess the associations of urinary DCAA and TCAA concentrations with reproductive hormone levels. Dose-response relationships were investigated using natural cubic spline (NCS) and restricted cubic spline (RCS) models. RESULTS: After adjusting for relevant confounders, we observed that higher urinary DCAA levels were associated with increased serum PRGE (9.2%; 95% CI: -0.55%, 19.8% for the highest vs. lowest tertile; P for trend = 0.06). Based on NCS models, we observed U-shaped associations of urinary DCAA with serum PRGE and PRL; each ln-unit increment in urinary DCAA concentrations above 3.61 µg/L and 6.30 µg/L was associated with 18.9% (95% CI: 4.8%, 34.7%) and 23.3% (95% CI: -0.92%, 53.5%) increase in serum PRGE and PRL, respectively. The U-shaped associations were further confirmed in RCS models (P for overall association ≤0.01 and P for non-linear associations ≤0.04). We did not observe evidence of associations between urinary TCAA and reproductive hormones. CONCLUSION: Urinary DCAA but not TCAA was associated with altered serum PRGE and PRL levels among women undergoing assisted reproductive technology.

Serum concentrations of organochlorine pesticides, biomarkers of oxidative stress, and risk of breast cancer.

Studies have documented that exposure to organochlorine pesticides (OCPs) is linked with breast cancer, but the underlying biological mechanisms are still unknown. This study included 313 women diagnosed with breast cancer and 313 controls in Wuhan, China, and measured 18 OCPs in serum and 3 oxidative stress biomarkers in urine. Multivariable adjusted regression models were used to evaluate the associations among OCPs, oxidative stress biomarkers, and breast cancer. The mediating effect of oxidative stress was assessed by mediation analysis. We observed that most OCPs were positively associated with risk of breast cancer (all FDR-P values < 0.05 or 0.10). Moreover, we found that p,p'-DDT, p,p'-DDD, dieldrin, heptachlor, and heptachlor epoxide were positively associated with 4-hydroxy-2-nonenal-mercapturic acid (HNE-MA) and 8-iso-prostaglandin F2α (8-isoPGF2α), which in turn were positively associated with risk of breast cancer. Mediation analysis indicated that HNE-MA and 8-isoPGF2ɑ mediated the positive associations between these OCPs and risk of breast cancer, with mediating proportion ranging from 6.23% to 19.9%. Our results suggest that lipid peroxidation may mediate the positive associations between OCP exposures and risk of breast cancer.