Alcohol Use in Pregnancy and Miscarriage: A Systematic Review and Meta-Analysis

Abstract

To systematically review and critically evaluate studies reporting alcohol exposure during pregnancy and miscarriage. We searched PubMed, EMBASE, PsycINFO, and ProQuest Theses for publications from January 1970 to January 2019. We identified studies about alcohol exposure during pregnancy and miscarriage. Information about study population, alcohol exposure assessment, outcome definition, covariates, and measures of association was collected. We assessed study quality using an adapted Newcastle-Ottawa Scale. Data were abstracted by 2 investigators independently. We conducted a random-effects meta-analysis to calculate the association between alcohol exposure and miscarriage risk and performed subgroup analyses to determine robustness of results to study differences. For studies reporting dose-specific effects, a pooled dose-response association was estimated using generalized least squares regression with and without restricted cubic spline terms for number of drinks consumed per week. Of 2,164 articles identified, 24 were eligible for inclusion. Meta-analysis of data from 231,808 pregnant women finds those exposed to alcohol during pregnancy have a greater risk of miscarriage compared to those who abstained (odds ratio [OR] 1.19, 95% confidence intervals [CI] 1.12, 1.28). Estimates did not vary by study design, study country, or method of alcohol ascertainment. For alcohol use of 5 or fewer drinks per week, each additional drink per week was associated with a 6% increase in miscarriage risk (OR 1.06, 95% CI 1.01, 1.10). Common study limitations reflect challenges inherent to this research, including difficulty recruiting participants early enough in pregnancy to observe miscarriage and collecting and quantifying information about alcohol consumption during pregnancy that accurately reflects use. This review provides evidence that alcohol consumption during pregnancy is associated with a dose-mediated increase in miscarriage risk. Future studies evaluating change in alcohol use in pregnancy are needed to provide insight into how alcohol consumption prior to pregnancy recognition impacts risk.

Keywords: Alcohol; Drinking; Miscarriage; Pregnancy; Spontaneous Abortion.

Figure 1.

Flow diagram of studies identified for the systematic review.

Figure 2.

Forest plot for the association between alcohol exposure during pregnancy and risk of miscarriage with subgroup estimates by study design. Size of point estimate markers indicates weight in meta-analysis. Abbreviations: OR, odds ratio; CI, confidence interval.

Figure 3.

Dose-response trend for average number of alcoholic drinks per week during pregnancy and miscarriage risk, spline model. Dashed lines represent the 95% confidence interval, knots selected using Harrell’s recommended percentiles located at 0, 3.5, and 14 drinks per week.