Intended for healthcare professionals


Low level alcohol consumption and the fetus

BMJ 2005; 330 doi: (Published 17 February 2005) Cite this as: BMJ 2005;330:375
  1. Raja A S Mukherjee, honorary lecturer, learning disability psychiatry (rmukherj{at},
  2. S Hollins, professor, learning disability psychiatry,
  3. Mohammed T Abou-Saleh, reader, addictive behaviour,
  4. Jeremy Turk, reader, developmental psychiatry
  1. Department of Mental Health, St George's Hospital Medical School, London SW17 ORE
  2. Department of Clinical Developmental Sciences

    Abstinence from alcohol is the only safe message in pregnancy

    Recently the media in the United Kingdom highlighted the messages presented by one of us (RASM) and researchers from other industrialised countries, that the only safe communication about alcohol consumption in pregnancy is that of abstinence. Unfortunately the scientific basis for this recommendation was not clarified in the media. We provide here examples of the evidence that has led the United States, Canada, Australia, and other countries to adopt the abstinence message.

    Fetal alcohol syndrome was first reported in the international literature by Smith and Jones in 1973.1 Before that, Lemoine published a series of 127 cases in France, highlighting the phenotypic presentation of people exposed to alcohol while pregnant.2 The full syndrome is characterised by a combination of short stature, neurocognitive deficits, and a specific triad of facial dysmorphology (short palpebral fissures, flat philtrum, and thin upper lip vermilion).

    Fetal alcohol spectrum disorder is an overarching term encompassing the behavioural diagnoses occur ring when the prenatal brain has been exposed to the teratogenic effects of alcohol, but the full phenotype of fetal alcohol syndrome has not developed. Fetal alcohol spectrum disorder is difficult to diagnose, but it is possible, with a positive history of alcohol consumption in the mother and neurocognitive deficits with or without the facial features.3

    The neurocognitive deficits in fetal alcohol syndrome and fetal alcohol spectrum disorder are pervasive. They include hyperactivity, impulsivity, difficulties with planning and mental organisation, concrete thinking, visuospatial problems, lack of awareness of social cues, and difficulties understanding the consequences of their own behaviour.4 They can cause immense distress and persistent difficulties for the affected individual. Furthermore, evidence shows overlap and comorbid presentation with conditions such as attention deficit hyperactivity disorder, autism, and personality disorder, requiring skilled clinical differentiation.5

    Over the last 30 years the opinions of professionals working in this area have changed. Previously fetal alcohol syndrome was considered to be a possible consequence of chronic alcohol consumption occurring in specific high risk populations such as the native American tribal groups.1 6 Views later changed to encompass moderate consumption in all populations.6 Evidence from animals and humans now provides confirmation that behavioural changes may be seen even at low doses of alcohol consumption.

    Ikonomidou et al report that exposure of the rat brain to ethanol for a period of hours during a specific developmental stage induces an apoptotic neurodegenerative reaction that deletes large neurones from several developing sites.7 Charness et al report that cell adhesion molecules are inhibited even at exposure to low concentrations of ethanol. These have subsequent effects on neuronal migration, fasciculation, and synaptogenesis, all vital to the developing brain.8

    Sood et al report a prospective study of 501 mother-child dyads in which the child's behaviour was adversely affected even at low levels of alcohol consumption—as low as one drink per week. They further report that a dose-response relation existed between the level of alcohol consumed and behaviours seen. Children exposed to any level of alcohol, compared with those not exposed, showed 3.2 times greater odds of delinquent behaviour. These relations continued even after other factors were controlled for.9 Knupfer et al argue that because of methodological problems, such as differing classifications of drinking practice and the imprecise definition of what low level consumption meant in early studies, more research is needed before conclusive evidence is available about the dose-response relation.10

    Evidence is currently emerging, but is as yet inconclusive, about the exact dose of alcohol that is safe in pregnancy. The likelihood is that individual differences in alcohol metabolism may protect most women when drinking small quantities. Currently we cannot predict who is and is not at risk.

    A small unpublished study carried out by one of the authors (MAS) at St George's Hospital in London showed that just under 50% of those attending the teenage antenatal clinic drank more than four units on a single occasion and 27% occasionally got drunk. A further 25% drank at least once a week. Kaskutas and Graves studied alcohol consumption in 321 pregnant women. They found that when self selecting drinks, the size of the drinks was up to 307% greater than standard measures. The authors concluded that risk based on current assumptions might actually be much higher than previously expected.11

    Unlike the position of the department of health—that one to two units a week in pregnancy are safe—the position adopted increasingly in other countries is that no level of alcohol consumption is known to be safe in pregnancy. A health promotion message about a safe amount of alcohol, although designed to protect the pregnant mother and the developing child, can be dangerous as it can be so easily misinterpreted. The uncertain level of individual risk to the developing fetus together with the possibility of misinterpreting a health promotion message mean that the only safe message in pregnancy is abstinence from alcohol.


    • Competing interests None declared.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    View Abstract