Background: Maternal urinary biomarkers are often used to assess fetal exposure to phenols and their precursors. Their effectiveness as a measure of exposure in epidemiological studies depends on their variability during pregnancy and their ability to accurately predict fetal exposure.
Objectives: We assessed the relationship between urinary and amniotic fluid concentrations of nine environmental phenols, and the reproducibility of urinary concentrations, among pregnant women.
Methods: Seventy-one women referred for amniocentesis were included. Maternal urine was collected at the time of the amniocentesis appointment and on two subsequent occasions. Urine and amniotic fluid were analyzed for 2,4- and 2,5-dichlorophenols, bisphenol A, benzophenone-3, triclosan, and methyl-, ethyl-, propyl-, and butylparabens using online solid phase extraction-high performance liquid chromatography-isotope dilution tandem mass spectrometry.
Results: Only benzophenone-3 and propylparaben were detectable in more than half of the amniotic fluid samples; for these phenols, concentrations in amniotic fluid and maternal urine collected on the same day were positively correlated (ρ = 0.53 and 0.32, respectively). Other phenols were detected infrequently in amniotic fluid (e.g., bisphenol A was detected in only two samples). The intraclass correlation coefficients (ICCs) of urinary concentrations in samples from individual women ranged from 0.48 and 0.62 for all phenols except bisphenol A (ICC = 0.11).
Conclusion: Amniotic fluid detection frequencies for most phenols were low. The reproducibility of urine measures was poor for bisphenol A, but good for the other phenols. Although a single sample may provide a reasonable estimate of exposure for some phenols, collecting multiple urine samples during pregnancy is an option to reduce exposure measurement error in studies regarding the effects of phenol prenatal exposure on health.