Prenatal Alcohol Exposure and Risk of Birth Defects

Abstract

The term alcohol-related birth defects (ARBDs) refers to a wide range of birth defects included in the Institute of Medicine (IOM) diagnostic guidelines on prenatal alcohol exposure (PAE). Evidence for many of the ARBDs included in the IOM guidelines was derived from animal studies and high-risk populations, and many of the ARBDs occur rarely, and can have other potential causes. Information on dose, pattern, and timing of PAE is also important for the diagnosis of ARBDs, but is not always available. For these reasons, it can be difficult to confidently attribute a defect to PAE. Population-based evidence for the association between PAE and birth defects is limited. This population-based study investigated the associations between dose, pattern, and timing of PAE (before pregnancy, in the first trimester, and in late pregnancy) and ARBDs according to IOM criteria. Between 1995 and 1997, data from a randomly selected cohort of nonindigenous women in Western Australia (WA) (n = 4714), who gave birth to a live infant (n = 4714), were linked to birth information recorded in the WA Midwives' Notification System and the WA Birth Defects Registry. Information about alcohol use before and during pregnancy was collected retrospectively from the participating women within 3 months of giving birth. Alcohol use was categorized as low, moderate, or heavy; the comparison group consisted of women who abstained from drinking alcohol. Multivariate logistic regression analysis was used to estimate the associations between PAE in each time period and the birth defects. There was a low prevalence of birth defects classified as ARBDs according to IOM criteria. As compared to the abstinent group, heavy PAE in the first trimester was associated with an increased likelihood of having an infant with a birth defect classified as an ARBD; the adjusted odds ratio was 4.57, with a 95% confidence interval of 1.46–14.26). The association was similar (adjusted odds ratio: 4.56) after bootstrap analysis. In contrast, no association was found between low or moderate PAE and any ARBD classified according to IOM criteria. These findings show a greater than 4-fold increase in the risk of birth defects classified as ARBD after heavy PAE in the first trimester. Many of the individual birth defects classified as ARBDs according to IOM criteria were not present in the cohort or were not related to PAE.