Considerations for the Analysis of Urinary Environmental Chemical Concentrations During Pregnancy

Pregnancy physiology may impact indicators of hydration, affecting exposure assessment in gestational studies with urinary chemical biomarkers.

We aimed to characterize hydration and demonstrate the impact of different methods for standardizing urinary chemical biomarker concentrations on exposure descriptives in the Human Placenta and Phthalates Study (n = 303, 2017-2020), a prospective pregnancy cohort with eight study visits between 12 and 38 weeks of gestation. We assessed trajectories and predictors of hydration using urine flow rate (UFR) and specific gravity (SG). Likewise, we examined trajectories and predictors of mono-n-butyl phthalate (MBP), for which concentrations were unstandardized or standardized via UFR (analyte excretion rates), SG alone (Boeniger method), or covariate-adjusted SG (O'Brien method). We used generalized additive mixed effects models to examine trajectories. We used linear mixed effects models to investigate participant demographic and pregnancy characteristics influencing UFR and SG, and agreement between unstandardized and standardized MBP concentrations.

As pregnancy progressed, SG declined linearly, whereas UFR varied in a nonlinear manner. Several demographic and pregnancy characteristics, notably race and ethnicity, were associated with UFR and SG. Unstandardized and standardized MBP concentrations showed good agreement, with lower agreement observed between concentrations standardized using UFR (analyte excretion rates) relative to SG (Boeniger and O'Brien methods). Nevertheless, trajectories and predictors of MBP concentrations were similar across all standardization approaches.

Despite systematic variations in hydration across gestation and by demographic and pregnancy characteristics, methods commonly used for standardizing urinary concentrations of chemical biomarkers were robust to differences in the hydration indicator as well as the standardization method employed.