Maternal exposure to polycyclic aromatic hydrocarbons during pregnancy and timing of pubertal onset in a longitudinal mother–child cohort in rural Bangladesh

Background: In experimental studies, several polycyclic aromatic hydrocarbons (PAHs) have shown endocrine disrupting properties, but very few epidemiological studies have examined their impact on pubertal development and results have been heterogenous. Objective: To explore if maternal PAH exposure during pregnancy was associated with the offspring’s timing of pubertal onset. Methods: We studied 582 mother-daughter dyads originating from a population-based cohort in a rural setting in Bangladesh. Maternal urinary samples, collected in early pregnancy (on average, gestational week 8), were analyzed for monohydroxylated metabolites of phenanthrene (1-OH-Phe, Σ2-,3-OH-Phe, and 4-OH-Phe), fluorene (Σ2-,3-OH-Flu), and pyrene (1-OH-Pyr) using liquid chromatography with tandem mass spectrometry (LC-MS/MS). The girls were interviewed on two separate occasions concerning date of menarche, as well as breast and pubic hair development according to Tanner. Associations were assessed using Kaplan-Meier analysis and multivariable-adjusted Cox proportional hazards regression or ordered logistic regression. Results: In early pregnancy, the mothers’ median urinary concentrations of Σ1-,2-,3-,4-OH-Phe, Σ2-,3-OH-Flu, and 1-OH-Pyr were 3.25 ng/mL, 2.0 ng/mL, and 2.3 ng/mL respectively. At the second follow-up, 78 % of the girls had reached menarche, and the median age of menarche was 12.7 ± 0.81 years. Girls whose mothers belonged to the second and third quintiles of ΣOH-Phe metabolites had a higher rate of menarche, indicating a younger menarcheal age (HR 1.39; 95 % CI 1.04, 1.86, and HR 1.41; 95 % CI 1.05, 1.88, respectively), than girls of mothers in the lowest quintile. This trend was not observed in relation to either breast or pubic hair development. None of the other maternal urinary PAH metabolites or the sum of all thereof in early pregnancy were associated with age at menarche or pubertal stage. Conclusions: Indications of non-monotonic associations of prenatal phenanthrene exposure with the daughters’ age of menarche were found, warranting further investigation.