Associations between lifestyle factors and levels of per- and polyfluoroalkyl substances (PFASs), phthalates and parabens in follicular fluid in women undergoing fertility treatment

Background Concerns have been raised whether exposure to endocrine-disrupting chemicals (EDCs) can alter reproductive functions and play a role in the aetiology of infertility in women. With increasing evidence of adverse effects, information on factors associated with exposure is necessary to form firm recommendations aiming at reducing exposure. Objective Our aim was to identify associations between lifestyle factors including the home environment, use of personal care products (PCP), and dietary habits and concentrations of EDCs in ovarian follicular fluid. Methods April-June 2016, 185 women undergoing ovum pick-up for in vitro fertilisation in Sweden were recruited. Correlation analyses were performed between self-reported lifestyle factors and concentration of EDCs analysed in follicular fluid. Habits related to cleaning, PCPs, and diet were assessed together with concentration of six per- and polyfluoroalkyl substances (PFASs) [PFHxS, PFOA, PFOS, PFNA, PFDA and PFUnDA], methyl paraben and eight phthalate metabolites [MECPP, MEHPP, MEOHP, MEHP, cxMinCH, cxMiNP, ohMiNP, MEP, MOHiBP]. Spearman’s partial correlations were adjusted for age, parity and BMI. Results Significant associations were discovered between multiple lifestyle factors and concentrations of EDCs in ovarian follicular fluid. After correcting p values for multiple testing, frequent use of perfume was associated with MEP (correlation ρ = 0.41 (confidence interval 0.21–0.47), p < 0.001); hens’ egg consumption was positively associated with PFOS (ρ = 0.30 (0.15–0.43), p = 0.007) and PFUnDA (ρ = 0.27 (0.12–0.40), p = 0.036). White fish consumption was positively associated with PFUnDA (ρ = 0.34 (0.20–0.47), p < 0.001) and PFDA (ρ = 0.27 (0.13–0.41), p = 0.028). More correlations were discovered when considering the raw uncorrected p values. Altogether, our results suggest that multiple lifestyle variables affect chemical contamination of follicular fluid. Impact statement This study shows how lifestyle factors correlate with the level of contamination in the ovary by both persistent and semi-persistent chemicals in women of reproductive age. Subsequently, these data can be used to form recommendations regarding lifestyle to mitigate possible negative health outcomes and fertility problems associated with chemical exposure, and to inform chemical policy decision making. Our study can also help form the basis for the design of larger observational and intervention studies to examine possible effects of lifestyle changes on exposure levels, and to unravel the complex interactions between biological factors, lifestyle and chemical exposures in more detail.

Associations between lifestyle factors and levels of per-and polyfluoroalkyl substances (PFASs), phthalates and parabens in follicular fluid in women undergoing fertility treatment Corresponding author: Ida Hallberg, ida.hallberg@slu.se,+4618-671641

Performance characteristics bisphenols, phthalate metabolites and parabens 2
The recovery and repeatability of the analytical method were determined using fortified bovine follicular fluid at a level of 2 ng/ml (n=8).In parallel with the samples (n = 333) 22 blanks (MQ) were analyzed.The collected blank data were used for the determination of the limits of detection (LODs) and limits of quantification (LOQs).The LOD was defined as three times the standard deviation of the blank plus the average blank value.The LOQ was defined as 3.3 * LOD value.The fortified bovine follicular fluid samples were also analyzed as quality control sample included in each LC-MS/MS run.The reproducibility was determined with the values of the quality control samples (n=8 in the course of the analyses).Performance characteristics PFAS 3 LOD and LOQ were defined as the concentration corresponding to 3 and 10 times the standard deviation of the ratio of the peak at the same retention time as the analyzed compounds and the corresponding internal standard divided by the slope of the calibration line and determined in the chemical blank samples

Figure S2 .
Figure S2.Directed acyclic graph (DAG) diagram describing the potential effect of measured covariates on the association between lifestyle factors and chemical exposureAge and parity are potential confounders of the association while BMI may act as a potential confounder or be influenced by lifestyle factors and by this be on a causal pathway to chemical exposure.

Table S1 .
Self-administered questionnaire filled in by participating women Questionnaire

for women undergoing IVF treatment Environmental impact of female fertility 1. Personal data and home environment
1.1 For how long have you lived in Sweden?

THANK YOU FOR YOUR PARTICIPATION!
The native and isotopically labeled compounds were purchased from Merck, Toronto Research Chemicals, Cambridge Isotope laboratories, Chiron, CDN Isotopes and Wellington Laboratories.

Table S3 :
Performance characteristics, chemicals and internal standards used to quantify per-and polyfluoroalkyl substances (PFASs)

Table S5 .
Proportion of patients above level of detection for the analysed compounds in ovarian follicular fluid

Table S6 .
Quantification of chemicals in sample blanks quantified in sample blanks in concentrations of approximately 20% of the levels in ovarian follicular fluid.The correlation with the other metabolites of DEHP remained high and sensitivity-analysis excluding MEOHP from DEHP-metabolites generated similar results.**BPS was quantified in one sample blank and BPSIP in rinse media, but these analytes were not included in the analyses as the detection levels in women were <20%.

Table S7 .
Spearman's correlation between lifestyle factors and age, BMI and parity

Table S8 .
Spearman's correlation between lifestyle factors and chemical levels in follicular unadjusted for covariates (BMI, Age, Parity)

r(95% CI) unadjusted p-value (raw) p-corr a (MEFF) a
a corrected for multiple testing using MEFF.