Exposure to ambient carbon monoxide and risk of low birth weight in Western Washington State
MetadataShow full item record
Background . Low birth weight is an adverse pregnancy outcome that has been inconsistently associated with exposure to ambient carbon monoxide (CO). Objective . To prospectively examine the relationship between ambient CO exposure and the risk of low birth weight. Methods . A prospective cohort study was performed using data from 3,401 Western Washington women who delivered infants between 1996 and 2006. Ambient CO exposure was predicted using regression models based on regional air pollutant monitoring data. CO exposure models contained predictor terms for year, month, weather and land use characteristics. Based on previous literature, ambient CO exposure windows included each trimester of pregnancy and total ambient CO exposure over the entire pregnancy. The risk of low birth weight within each quartile of CO exposure was compared, with the lowest quartile in each exposure window serving as the referent category. In addition, CO exposure was analyzed as a continuous variable (in parts per million). The primary outcome was low birth weight, defined as any birth weighing less than 2500 grams, regardless of gestational age. Low birth weight was identified using abstracted maternal medical record data and was analyzed as a dichotomous variable adjusted for continuous gestational age at delivery. Secondary analyses analyzed the effect of CO exposure on birth weight (in grams) as a continuous variable. Model 1 adjusted for covariate information obtained from interview-administered maternal questionnaires and abstracted medical records. Model 2 adjusted for the covariates in Model 1 and fine particulate matter. Results . In the primary analysis, after adjustment for covariates (age, race, education, parity, smoking, household income, pre-pregnancy BMI, gestational diabetes, gestational age at delivery, marital status, employment status, year of conception and month of conception), the highest quartile of CO exposure in the second trimester of pregnancy was statistically significantly associated with a more than doubling of the risk of low birth weight (aOR 2.34; 95% CI 1.04–5.25). Further adjustment for fine particulate matter exposure had little effect on the association (aOR 2.40; 95% CI 0.97–5.92). Elevated odds ratios for low birth weight were observed with increasing exposure to CO in the first trimester, but the results were not significant. No statistically significant association between increased CO exposure and low birth weight was observed in the first or third trimesters, or over the entire pregnancy period. However, a significant trend was noted in each trimester. When CO was analyzed as a continuous variable, elevated odds ratios were observed for each part per million increase in ambient CO exposure in the second trimester (aOR 1.55; 95% CI 0.55–4.22), and third trimester (aOR 1.36; 95% CI 0.57–3.26), but the results were not significant. In the secondary analysis evaluating birth weight as a continuous variable, no association was observed between increased ambient CO exposure and a reduction in birth weight in grams. Conclusion . These data are consistent with previous literature suggesting that increased CO exposure may adversely impact birth weight, particularly exposures earlier in gestation. Additional, larger studies are warranted to further examine and more precisely quantify the relationship between ambient CO exposure and low birth weight.