Maternal alcohol intake up to and during pregnancy and risk of adverse birth outcomes: evidence from a British cohort

Background High maternal alcohol consumption has been linked to adverse birth outcomes such as small for gestational age and preterm birth, which in turn have been linked to increased risk of development of cardiovascular diseases and type 2 diabetes in adulthood. The UK Department of Health (DH) recommends that pregnant women and those trying to conceive should avoid alcohol and never drink more than 1GÇô2 units once or twice a week. This study aimed to investigate the association between alcohol intake before and during different stages of pregnancy with both birthweight and gestational age. Methods Data were used from the Caffeine and Reproductive Health Study (CARE), a prospective birth cohort that included 1303 low-risk pregnant women aged 18GÇô45 years, recruited from September, 2003, to June, 2006. Questionnaires administered in the first and second trimester and postpartum assessed alcohol consumption before pregnancy and for the three trimesters. Frequency of weekly alcohol consumption was analysed by categories of intake to accord with DH guidelines (Gëñ2 units per week, >2 units per week, and a non-drinking category as the referent) and was related to preterm birth and size at birth, measured as grams and as customised birthweight centile, which takes into account maternal prepregnancy weight, height, parity, ethnicity, gestation, and baby's sex in multivariable linear and logistic regression models. We also adjusted for maternal age, caffeine intake, education, energy intake, and salivary cotinine as a biomarker of smoking status. Only participants with complete data for all variables were included in the analyses, which excluded just under 10% of the sample. All women provided informed consent and the study was approved by the Leeds West Local Research Ethics Committee (ref 03/054). Findings 1153, 1135, 793, and 377 women, respectively, had data available for birth outcomes and alcohol consumption before pregnancy and during the three trimesters. 74% of women before pregnancy and 53% in the first trimester reported alcohol intakes above the DH recommendation. For intakes above 2 units per week compared with non-drinkers, the adjusted differences in birth centile were GêÆ7-À7 (95% CI GêÆ12-À8 to GêÆ2-À6; ptrend=0-À009), GêÆ8-À2 (GêÆ12-À6 to GêÆ3-À7; ptrend=0-À002), and GêÆ6-À4 (GêÆ11-À8 to GêÆ1-À1, ptrend=0-À06) before pregnancy and during trimesters 1 and 2, respectively. The association with small for gestational age and preterm birth was strongest in trimester 1, with adjusted odds ratios of 2-À0 (95% CI 1-À2GÇô3-À4; ptrend=0-À03) and 3-À5 (1-À1GÇô11-À2, ptrend=0-À04), respectively. Women who adhered to the recommendations in the first trimester of 2 units or fewer per week were also at a significantly higher risk of having babies born with lower birthweight (adjusted difference GêÆ98-À5, 95% CI GêÆ170-À9 to GêÆ26-À1; ptrend=0-À007), birth centile (GêÆ5-À8, GêÆ10-À8 to GêÆ0-À7; ptrend=0-À002), and preterm birth (adjusted odds ratio 4-À6, 95% CI 1-À4GÇô14-À7; ptrend=0-À04) compared with non-drinkers. Interpretation The first trimester was the most sensitive period for the association of alcohol with restricted fetal growth. However, this finding could be explained by under-reporting of alcohol intake. Our small sample size in the third trimester did not allow us to detect a change in birthweight, and larger prospective studies that take into account timing of exposure to alcohol are needed. We showed no evident safe level of alcohol consumption in pregnancy, and the safe advice should be to abstain from alcohol when planning to conceive and during pregnancy, particularly during its early stages. Funding The CARE study was supported by a grant from the Food Standards Agency, UK (T01033)