Examining Complex Relationships Between Alcohol Exposure and Miscarriage

Abstract

Half of women in the United States use alcohol during pregnancy and most exposure occurs in the first weeks of gestation. Although most women who use alcohol change their behavior after pregnancy recognition, information about how timing of alcohol exposure impacts miscarriage risk is scarce. I performed a systematic review and meta-analysis of the literature, which provided evidence that alcohol-exposed pregnancies are at higher risk of miscarriage than unexposed pregnancies. Almost all studies were limited by overly simplistic methods for modeling alcohol exposure. Using a simulation study, I selected two approaches for measuring the association between alcohol use and miscarriage in the prospective pregnancy cohort, Right from the Start: A Study of Early Pregnancy Health (2000–2012; n=5,424). Each additional week of alcohol use was associated with an 8% increase in miscarriage compared to women who abstained. Exposure in weeks five through ten of pregnancy conferred the most risk. These findings did not differ by alcohol type or dose. Genetic variants of ADH1C determine efficiency of alcohol dehydrogenase, the first enzyme in the alcohol metabolism pathway. In a substudy of 987 participants with genetic data, ADH1C haplotype was not associated with miscarriage risk and did not modify the association between alcohol use and miscarriage. Overall, this work highlights timing of alcohol exposure during pregnancy as an important determinant miscarriage risk.