Association between maternal urinary phthalate metabolites in early and late pregnancy and neonatal birth outcomes: A cohort analysis

This study aimed to investigate maternal urinary phthalate (PAE) metabolite concentrations during pregnancy and their associations with neonatal birth outcomes. A total of 3689 pregnant women were enrolled, among whom 866 provided early-pregnancy urine samples and 2833 provided late-pregnancy urine samples. Urinary PAE metabolite levels were significantly higher in the late pregnancy period compared with the early pregnancy period. Regression analyses demonstrated that mono-(2-ethyl-5-carboxypentyl) phthalate (MECPP) was associated with male birth weight in early pregnancy prior to sex stratification, while mono-butyl phthalate (MBP) and mono-iso-octyl phthalate (MOP) were correlated with female birth weight. Additionally, mono-(2-ethyl-5-hydroxyhexyl) phthalate (MEHHP) was associated with male birth length, and MBP was linked to female birth parity. In late pregnancy, unstratified analyses revealed associations of mono-ethyl phthalate (MEP) with both birth weight and birth length; sex-stratified results further indicated diverse associations between different PAE metabolites and female neonatal outcomes. Restricted cubic spline (RCS) curves identified non-linear relationships between PAE metabolites and birth outcomes. For instance, MBP exhibited an inverted U-shaped association with birth parity in early pregnancy before sex stratification. Bayesian kernel machine regression (BKMR) analyses showed that in early pregnancy, PAE metabolites collectively had negative associations with birth weight and birth parity, and positive associations with birth length. In late pregnancy, sex-stratified BKMR results revealed distinct gender differences in the associations between PAE metabolites and neonatal birth outcomes. Collectively, this study indicates that maternal urinary PAE metabolite concentrations are higher in late pregnancy than in early pregnancy, and these metabolites are associated with neonatal birth outcomes. These associations are characterized by exposure dose relationships, combined effects, and gender-specific differences.