Investigating intergenerational differences in human PCB exposure due to variable emissions and reproductive behaviors

Abstract

Background: Reproductive behaviors--such as age of childbearing, parity, and breast-feeding prevalence--have changed over the same historical time period as emissions of polychlorinated biphenyls (PCB) and may produce intergenerational differences in human PCB exposure.

Objectives: Our goal in this study was to estimate prenatal, postnatal, and lifetime PCB exposures for women at different ages according to year of birth, and to evaluate the impact of reproductive characteristics on intergenerational differences in exposure.

Methods: We used the time-variant mechanistic model CoZMoMAN to calculate human bioaccumulation of PCBs, assuming both hypothetical constant and realistic time-variant emissions.

Results: Although exposure primarily depends on when an individual was born relative to the emission history of PCBs, reproductive behaviors can have a significant impact. Our model suggests that a mother's reproductive history has a greater influence on the prenatal and postnatal exposures of her children than it does on her own cumulative lifetime exposure. In particular, a child's birth order appears to have a strong influence on their prenatal exposure, whereas postnatal exposure is determined by the type of milk (formula or breast milk) fed to the infant.

Conclusions: Prenatal PCB exposure appears to be delayed relative to the time of PCB emissions, particularly among those born after the PCB production phaseout. Consequently, the health repercussions of environmental PCBs can be expected to persist for several decades, despite bans on their production for > 40 years.

Figure 1

PCB-153 lipid-normalized concentration profile for a default female under constant emissions. Prenatal exposure (i.e., the estimated PCB-153 lipid-normalized concentration at birth) is indicated by the arrow. Postnatal exposure (gray area) is based on 6 months of breast-feeding. The entire area under the curve (gray plus blue areas) represents lifetime exposure.

Figure 2

Predicted lifetime lipid-normalized concentration profiles of PCB-153 according to year of birth for default females born between 1920 and 2010 under time-variant conditions.

Figure 3

Comparison of prenatal (A), postnatal (B), and lifetime (C) lipid-normalized PCB-153 exposure predictions for default females according to year of birth (black line) based on estimated time-variant PCB-153 emissions (blue line). Note the difference in the y-axis scale in C.

Figure 4

Predicted lipid-normalized PCB-153 profiles and estimated prenatal, postnatal, and lifetime exposures, assuming constant PCB-153 emissions over time for a woman who gives birth to her first and only child at 20, 30, or 40 years of age (A); has none, two, or five children (with births at 2-year intervals and the last birth at 30 years of age) (B); and consumed breast milk (BM) or formula milk (FM) exclusively for 6 months after her birth and breast-fed (BF) or formula-fed (NBF) a single infant born when the woman was 30 years of age (C). In a steady-state simulation, every generation experiences the same exposure (i.e., results for mother and child are the same).

Figure 5

Predicted lipid-normalized prenatal, postnatal, and lifetime exposures to PCB-153 as a function of year of birth, assuming time-variant PCB-153 emissions for a woman who gives birth to her first and only child at 20, 30, or 40 years of age (A); has none, one, two, three, four, or five children (with childbirth at 2-year intervals and the fifth child born at 30 years of age for multiparous women) (B); and consumed breast milk (BM) or formula milk (FM) exclusively for 6 months after birth and breast-fed (BF) or formula fed (NBF) a single infant born when the woman was 30 years of age (C).