Perera, FP, V Rauh, RM Whyatt, WY Tsai, JT Bernert, YH Tu, H Andrews, J Ramirez, L Qu, D Tang. 2004. Molecular Evidence of an Interaction Between Prenatal Environmental Exposures on Birth Outcomes in a Multiethnic Population. Environmental Health Perspectives 112:626-630.

This study provides evidence that exposure to common environmental pollutants at levels currently encountered in New York City can undermine fetal development. Birth weight and head circumference are lower with higher exposures to specific pollutants. This is important because reductions in birth weight and head circumference correlate with lower IQs later in life, reduced cognitive functioning and diminished school performance.

Moreover, the specific pollutants studied, environmental tobacco smoke (ETS) and polycyclic aromatic hydrocarbons (PAHs), had an interactive effect, with ETS increasing susceptibility to the impact of PAHs. The effect of PAHs in this group of subjects would not have been detected without simultaneous consideration of ETS.

What did they do?

Perera et al. surveyed a group of nonsmoking, New York-area, African-American and Dominican mothers and their newborns, collecting maternal and cord blood samples and conducting a 45-minute questionnaire.

The questionnaire consisted of lifestyle questions about places of residence, places of travel, smoking, alcohol use, and types of food intake. Information from medical files gave researcher information on the length of pregnancy, infant birth weight, length, head circumference, deformations, APGAR scores, any complications, etc.

The Centers for Disease Control and Prevention tested the blood samples. Analysts tested specifically for cotenine as a biomarker of exposure to ETS (environmental tobacco smoke) and B[a]P (benzo[a]pyrene) as a measure of environmental PAH (polycyclic aromatic hydrocarbons). Exposure to ETS was also assessed independently in the questionnaire, asking whether the subject lived in residence with other smokers. Those reporting that they lived with smokers (or had smoking visitors) were scored ETS+. Actual smokers were excluded from the study, as were subjects whose cotinene levels were high enough to indicate they smoked.

The research team analyzed the relationship between birth outcome (birth weight and head circumference) and exposures to B[a]P and cotenine using multivariate regressions, including data on possible data on confounding variables gathered via the questionnaire.

Perera et al. also split the subjects into high/low groups for measurements of both B[a]P and cotenine, using the median value of each measurement to determine the dividing line between groups, high vs. low B[a]P and high vs. low cotenine.

What did they find?

49% of newborns and 46% of mothers has cotinine levels indicative of environmental tobacco smoke exposure (between 0.05 and 15 ng/mL). African-Americans were more commonly exposed to ETS with 78.1% having detectable levels of continene as opposed to 31.9% of Domincans. B[a]P was detected in 38% of cord blood samples and 39% of maternal blood samples.

The multiple regression analysis found ETS exposure (presence or absence of a smoker in the residence) to be associated with smaller head circumference (p = 0.04), and measured cotinine was associated with birth length (p = 0.05).

Overall, B[a]P was not significantly correlated with either cotinine, or dietary PAH.

However, among those cases with self-reported ETS exposure (ETS+), high B[a]P was associated with lower birth weight and head circumference. Individuals with high B[a]P and ETS+ showed a 233 gram (6.8%) reduction in birth weight (graph to right) and 1 cm (2.9%) reduction in head circumference (graph below) compared to cases of ‘low’ B[a]P and ETS -.

In other words, in the presence of environmental tobacco smoke (ETS+), exposure to polycyclic aromatic hydrocarbons can further decrease head circumference and birth weight.

What does this mean?

Prior work by Perera's research team in Poland had found significant associations between polycyclic aromatic hydrocarbons and birth outcome, including lower birth weight and reduced head circumference. The Polish newborns, however, had been exposed to PAH levels 10 times higher than those observed in New York. These new results demonstrate that the lower ambient air pollution levels experienced by minority residents in New York City can nonetheless undermine fetal growth and development.

The changes in birth weight and head circumference observed by Perera et al. are of concern because several prior studies have noted associations between those two birth characteristics and impaired functional abilities later in life, specifically lower IQ, poorer cognitive functioning and poorer school performance.

Their observation of heightened impact of B[a]P in the presence of environmental tobacco smoke reveals a pitfall for epidemiological work. Had Perera et al. looked only at B[a]P, they would have seen no effect, because their analysis overall, pooling ETS+ and ETS- subjects showed no significant impact of B[a]P. B[a]P's impact was only revealed by separating their data into ETS+ and ETS- subsets.