It has been more than 40 years since fetal alcohol syndrome was first recognized as a brain disorder leading to a wide range of learning and behavior problems – fetal alcohol spectrum disorders – in children prenatally exposed to alcohol.
Over that time, obstetric providers have played a key role in counseling patients, both preconception and during pregnancy, about the risks associated with various amounts and patterns of alcohol consumption. This advice is critical as about half of women of reproductive age in the United States consume some alcohol, and about half of pregnancies are not planned, leading to a high prevalence of exposure to alcohol prior to pregnancy recognition.
©Fuse/thinkstockphotos.com
But how much alcohol at what specific time in early pregnancy leads to a known risk of learning and behavior problems as children reach school age?
The U.S. Surgeon General’s Office and the Centers for Disease Control and Prevention recommend that alcohol be avoided entirely during all weeks of pregnancy, as there is no known safe amount, type of beverage, or timing in gestation that a woman can consume alcohol. However, in recent years, a number of publications have suggested that “low to moderate” alcohol consumption in pregnancy is not demonstrably harmful to the developing fetus, at least in terms of learning ability.
Three recently published studies exemplify the dilemma. Colleen M. O’Leary et. al. examined educational achievement in 8- to 9-year olds in Western Australia (Pediatrics 2013;132:e468-75). The sample was a population-based cohort of 4,056 infants randomly ascertained with births between 1995 and 1997 whose mothers had responded to a postnatal survey about health behaviors including alcohol consumption. Researchers linked these infants to a midwives database to obtain birth details and to an educational testing database to obtain measures of school achievement.
Dr. Christina D. Chambers
Children were not evaluated for the physical features or a diagnosis of FAS or something on the FASD spectrum. Low alcohol consumption was defined as 1-2 standard drinks (10 g alcohol per standard drink in Australia) per occasion and fewer than 7 drinks per week. Moderate alcohol consumption was defined as 3-4 standard drinks per occasion and no more than 7 drinks per week. Binge drinking was defined as 5 or more drinks per occasion less frequently than weekly, and heavy drinking was defined as more than 7 standard drinks per week including binge drinking weekly or more often.
Underachievement in reading and writing was significantly associated with either heavy first trimester or binge drinking in late pregnancy. However, achievement in numeracy, reading, spelling and writing was not significantly impaired with low to moderate prenatal alcohol exposure.
In a study of a sample derived from the Danish National Birth Cohort, 1,628 women and their children were sampled from the original cohort based on maternal alcohol drinking patterns reported in pregnancy (BJOG 2012;119:1191-1200). The child’s IQ was assessed at 5 years of age.
Children were not specifically evaluated for the physical features of FAS or a diagnosis of something on the FASD spectrum. Levels of alcohol consumption were categorized as none, average intake of 1-4 standard drinks per week (12 g alcohol per standard drink in Denmark), 5-8 standard drinks per week, and more than 8 standard drinks per week. There were no differences in the performance of children whose mothers consumed up to 8 standard drinks per week at some point in pregnancy compared to children whose mothers abstained.
In a subsequently published study in which researchers used the same sample, the parent and teacher versions of the Strengths and Difficulties Questionnaire, a standard behavioral screening tool, were completed by the mothers and the preschool teachers (BJOG 2013;120:1042-50). After adjustment for confounders, overall there were no significant associations found for any drinking category compared to abstainers.
Many experts asked to comment on these findings emphasized that these studies were limited to a few measures of learning and behavior in young children that may not be reflective of the range of alcohol-related developmental effects. They also pointed out the great difficulty in obtaining an accurate report of alcohol exposure in the absence of a sensitive and specific biomarker.
For example, recall of specific quantities, frequencies, and timing of alcohol consumption either after delivery (the Australian study) or in a single prenatal interview that was conducted sometime between 7 and 39 weeks’ gestation (the Danish study) may be inaccurate. This could be because of difficulty in remembering these details, as well as the influence of the social unacceptability of drinking during pregnancy.