Urinary bisphenol S concentrations: Potential predictors of and associations with semen quality parameters among men attending a fertility center

Background: Bisphenol S (BPS) was introduced in the market as a potentially safer alternative to bisphenol A (BPA). However, there are limited studies on health effects of BPS and no epidemiologic studies on its relationship with male reproductive health outcomes, specifically semen quality. Objective: To investigate predictors of urinary BPS concentrations and its association with semen parameters among men attending a fertility center. Methods: This cross-sectional analysis included 158 men of couples seeking fertility treatment (2011-2017) contributing 338 paired semen and urine samples. At the time of sample collection, men completed a questionnaire on self-reported use of household products and food intake within the previous 24 h. Urinary concentrations of BPA, BPS and bisphenol F were quantified using isotope-dilution tandem mass spectrometry. Semen samples were analyzed following WHO guidelines. Multivariable mixed models were used to investigate predictors of urinary BPS concentrations and to evaluate associations between urinary BPS concentrations and semen parameters, using random intercept to account for correlation in outcomes across multiple observations per man and adjusting for abstinence time, specific gravity, age, body mass index (BMI), year of sample col- lection and BPA concentrations. Analyses were also stratified by BMI (>= 25 vs < 25 kg/m(2)). Results: Median (IQR) urinary BPS concentration was 0.30 (0.20, 0.90) mu g/L, and 76% of samples had detectable (> 0.1 mu g/L) concentrations. Self-reported fabric softener and paint/solvent use as well as intake of beef and cheese within 24 h before urine collection were positively associated with BPS concentrations. Men with higher BPS concentrations also had significantly higher BMI. Lower semen parameters were found among men with detectable BPS concentrations, compared to men with non-detectable BPS [2.66 vs. 2.91 mL for volume (p = 0.03), 30.7 vs. 38.3 mil/mL for concentration (p = 0.03), 76.8 vs. 90.0 mil for total count (p = 0.09), 43.7 vs. 47.0% for motility (p = 0.06), and 5.42 vs. 6.77% for morphologically normal sperm (p = 0.24)]. Some associations of BPS with lower semen parameters were only found among men with a BMI >= 25 kg/m(2). Conclusions: We identified dietary and lifestyle factors associated with BPS exposure, suggesting potential avenues for reducing exposures. We also observed negative associations between BPS and semen parameters, especially among overweight and obese men.