Organophosphate flame-retardant metabolite concentrations and pregnancy loss among women conceiving with assisted reproductive technology

Objective: To evaluate whether urinary concentrations of organophosphate flame retardant (PFR) metabolites are associated with pregnancy loss among women conceiving with assisted reproductive technology (ART). Design: Prospective preconception cohort of subfertile women. Setting: Academic hospital fertility center in Boston, Massachusetts. Patient(s): A total of 155 women conceiving 179 pregnancies with ART. Intervention(s): None. Mean exposure to each of five PFR metabolites was estimated by averaging the specific-gravity adjusted natural log concentrations from two urine samples collected during the ART cycle of conception. Main Outcome Measure(s): Adjusted risk ratios (RRs) and 95% confidence intervals (CIs) for biochemical and total pregnancy loss (all losses <20 weeks' gestation) by quartiles of PFR metabolite concentrations were estimated using a repeated measures log-binomial model, accounting for multiple pregnancies per woman. Result(s): Of the 179 pregnancies, 31% ended in pregnancy loss (12% in biochemical loss). Among the three metabolites with high detection frequency [bis(1,3-dichloro-2-propyl) phosphate (BDCIPP), diphenyl phosphate (DPHP), and isopropylphenyl phenyl phosphate (ip-PPP)], an increased risk of biochemical loss was observed for women with DPHP concentrations in the fourth vs. first quartile (RR 1.64; 95% CI 0.61-4.39). Also found was an elevated risk of biochemical pregnancy loss among women in the highest quartile of the molar sum of urinary PFR metabolites compared with the lowest (RR 1.89; 95% CI 0.64-5.58). Urinary concentrations of ip-PPP and BDCIPP were not associated with either outcome. Conclusion(s): Among subfertile women, urinary DPHP metabolite concentrations measured during the ART cycle of conception may be associated with early pregnancy loss. Although this study is uniquely designed to investigate early markers of pregnancy success and maintenance, the small sample size likely contributed to imprecision. Given their increasing use as replacement chemicals for traditional flame retardants, exposure to PFRs may increase, and more studies will be needed to investigate their potential to impact pregnancy and reproduction. (C) 2018 by American Society for Reproductive Medicine.