Maternal urinary concentrations of organophosphate ester metabolites: associations with gestational weight gain, early life anthropometry, and infant eating behaviors among mothers-infant pairs in Rhode Island
Kathryn A. Crawford,Nicola Hawley,Antonia M. Calafat,Nayana K. Jayatilaka,Rosemary J. Froehlich,Phinnara Has,Lisa G. Gallagher,David A. Savitz,Joseph M. Braun,Erika F. Werner,Megan E. Romano
DOI: https://doi.org/10.1186/s12940-020-00648-0
2020-09-11
Environmental Health
Abstract:Abstract Background Organophosphate esters (OPEs)—used as flame retardants and plasticizers—are associated with adverse pregnancy outcomes such as reduced fecundity and live births and increased preterm delivery. OPEs may interfere with growth and metabolism via endocrine-disruption, but few studies have investigated endocrine-related outcomes. The objective of this pilot study ( n = 56 mother-infant pairs) was to evaluate associations of OPEs with gestational weight gain (GWG), gestational age at delivery, infant anthropometry, and infant feeding behaviors. Methods We quantified OPE metabolites (bis-2-chloroethyl phosphate [BCEP], bis (1,3-dichloro-2-propyl) phosphate [BDCPP], diphenyl phosphate [DPHP]) in pooled maternal spot urine collected throughout pregnancy (~ 12, 28, and 35 weeks’ gestation). We obtained maternal sociodemographic characteristics from questionnaires administered at enrollment and perinatal characteristics from medical record abstraction. Trained research assistants measured infant weight, length, head and abdominal circumferences, and skinfold thicknesses at birth and 6 weeks postpartum. Mothers reported infant feeding behavior via the Baby Eating Behavior Questionnaire (BEBQ). Using multiple linear regression, we assessed associations of log 2 -transformed maternal urinary OPE metabolites with GWG, gestational age at delivery, infant anthropometry at birth, weekly growth rate, and BEBQ scores at 6 weeks postpartum. We used linear mixed effects (LME) models to analyze overall infant anthropometry during the first 6 weeks of life. Additionally, we considered effect modification by infant sex. Results We observed weak positive associations between all OPE metabolites and GWG. In LME models, BDCPP was associated with increased infant length (β = 0.44 cm, 95%CI = 0.01, 0.87) and weight in males (β = 0.14 kg, 95%CI = 0.03, 0.24). BDCPP was also associated with increased food responsiveness (β = 0.23, 95%CI = 0.06, 0.40). DPHP was inversely associated with infant abdominal circumference (β = − 0.50 cm, 95%CI = − 0.86, − 0.14) and female weight (β = − 0.19 kg, 95%CI = − 0.36, − 0.02), but positively associated with weekly growth in iliac skinfold thickness (β = 0.10 mm/wk., 95%CI = 0.02, 0.19). Further, DPHP was weakly associated with increased feeding speed. BCEP was associated with greater infant thigh skinfold thickness (β = 0.34 mm, 95%CI = 0.16, 0.52) and subscapular skinfold thickness in males (β = 0.14 mm, 95%CI = 0.002, 0.28). Conclusions Collectively, these findings suggest that select OPEs may affect infant anthropometry and feeding behavior, with the most compelling evidence for BDCPP and DPHP.
environmental sciences,public, environmental & occupational health