March 10, 2014

expert reaction to early pregnancy, alcohol intake and birth outcomes

A study in the Journal of Epidemiology and Community Health, based on responses to questionnaires by 1,264 women at low risk of birth complications, suggested alcohol consumption in the first 3 months of pregnancy may heighten the risk of having a premature or small baby.

Dr Patrick O’Brien, RCOG (Royal College of Obstetrics and Gynaecology) spokesperson, said:

“This is an interesting study as it provides us with an insight into women’s drinking patterns both before and during pregnancy.

“The findings suggest that even small amounts of alcohol in the first three months of pregnancy may increase the risk of a small impairment to the baby’s growth and possibly premature birth.

“This is consistent with RCOG advice that if a woman falls pregnant she should abstain from alcohol during the first 12 weeks of pregnancy because this is a particularly sensitive time for the baby’s development.

“However, while the safest approach would be to choose not to drink at all, small amounts of alcohol, not more than one to two units once or twice a week, have not been shown to be harmful after 12 weeks of pregnancy.

“Pregnant women should always consult their midwives or doctors if they have any concerns about their alcohol intake.”

Prof Jean Golding, Emeritus Professor of Paediatric & Perinatal Epidemiology, University of Bristol, said:

“This study of alcohol intake in pregnancy would have been important in that it purported to analyse information on the amount drunk in the 4 weeks before pregnancy, and in each of the three trimesters of pregnancy. It is specifically testing the validity of the recommendation that up to two drinks a week is OK. The aim is excellent – but there are a few problems: (i) the numbers of pregnancies involved is relatively small; (ii) the assessment of the amount of alcohol is misleading – the participants were asked how often they had a drink per week – not how many drinks they had. Thus someone who went on one binge a week would be likely to answer ‘once a week’, but actually have drunk 4 glasses. (iii) Curiously although the authors found a link with preterm delivery for 1-2 times a week in the first trimester, the women who drank more often did not have a worse outcome. Thus, the data do not show convincing evidence that 1 or 2 drinks a week are unsafe. It would be helpful to know what the true associations are – but the findings in this paper, alas, are probably misleading.”

Catherine Collins, Principal Dietitian at St George’s Hospital NHS Trust, said:

“The current DH/ NHS recommendations for alcohol recommend all pregnant women to limit alcohol intake to less than 2 units a week, and to avoid getting drunk. This paper supports this statement, and doesn’t provide any real evidence for abstinence. The paper showed that first-time mums-to-be were more likely to avoid alcohol during pregnancy than mums who had done it all before.

“The authors quantified alcohol consumption across pregnancy trimesters using an established nutritional survey tool which offered 9 potential answers for alcohol intake, ranging from ‘I never drink’ to ‘I have more than 6 drinks a day’. Despite these 9 original categories, the authors chose to limit the alcohol response to three: ‘I never drink’, ‘drink less than 2 units a week’, and ‘I drink more than two units a week’ – the latter would encompass 6 of the 9 categories – which is a really large range, from an alcohol intake of less than 3 units per week to over 18 units a day….

“The largest flaw in the study was that women who perhaps had 3 units of alcohol a week were placed in a category that included intake up to 6 units a day. Adding the modest drinkers to the heavy drinkers category would have skewed the data, giving the appearance that even modest increase above the recommended limits was highly risky behaviour. In addition, ‘translating’ the amount and type of alcohol consumed into standard ‘units’ was potentially flawed, especially if women chose to drink alcohol at home, when measures are inevitably more generous  than a pub measure.

“It appeared that women who drunk more than 2 units of alcohol a week had a much greater risk of a low birthweight baby, but this was not statistically significant until other factors that affect the risk of preterm baby, or a small for date baby, were taken into account. Women in the higher alcohol category also appeared to have a lower average calorie intake than alcohol abstainers. This is interesting in that it implies that the calories from alcohol were substituted as pregnancy progressed at the expense of food calories, and thus overall dietary quality could have been reduced.

“It’s also worth noting that the study originally tried to recruit 5959 women, but only 1374 consented to participate. From first to third trimester, whilst 81% of the non-drinking mums-to-be completed the third trimester alcohol assessment, only 27% of the <2 units/w completed, and a mere 17% of the >2units/w group completed the duration. Even allowing for a couple of pre-termers in the heavier drinkers in the third trimester, the loss of recruits from the study makes it extremely difficult to draw specific outcomes for alcohol. The authors highlight this in their discussion.

“Finally, there’s one reassuring outcome, in that 38% of women had a much higher alcohol intake in the weeks before pregnancy, so being less than delicate about it there’s no need to worry about what you drank around the time of conception, but that the current government recommendations are worth continuing to follow.”

Prof Andrew Whitelaw, Professor of Neonatal Medicine, University of Bristol, said:

“Maternal smoking has a definite effect on birth weight and a good feature of this study is that tobacco consumption was assessed by salivary cotinine (which comes from nicotine) as well as questionnaire. In addition, calorie intake was assessed by questionnaire and the analysis of birth weight was adjusted to take account of these factors as well as the mother’s height and ethnic group.

“Alcohol intake was measured by 3 questionnaires because there is no blood or saliva test that can measure alcohol intake over time, and is likely to be underestimated across the population, therefore still comparable.

“The finding that admitting drinking alcohol, even less than 2 units/week, in the first 12 weeks increased preterm birth by about 4 times is even more worrying than the reduction in birth weight for gestational age.

“It is also interesting that alcohol consumption was greatest in women from a strong educational and social background who should otherwise have the lowest risk of preterm birth and low birth weight.

“This is further evidence that even moderate amounts of alcohol are toxic to the growing fetus and direct toxicity is further worsened by the increased complications of premature birth.

“As 38% of the women admitted risky drinking (over 10 units/week) before pregnancy and pregnancy is sometimes discovered late, the advice has to be to avoid alcohol completely when trying to conceive.”

Dr Ron Gray, Senior Clinical Research Fellow, National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford, said:

“The authors of this new study correctly point out there is conflicting evidence on the fetal effects of consuming up to two units of alcohol per week during pregnancy, but their study does not really help to resolve this conflict.

“The authors investigated alcohol consumption during each trimester of pregnancy and tried to link it to five different birth outcomes (including birth weight and preterm birth).  They present data on 1135 women in the first trimester, 808 women in the second and 384 in the third. These numbers are relatively small for these kinds of studies. This means that the results are less precise than might be achieved with a much larger study. Also, given the high number of different statistical tests performed (around 40) one might expect a certain number of the results to be significant just by chance.

“In this study, women who drank alcohol during pregnancy differed in a number of respects from women who abstained (they were less likely to live in a poor area for example). Sensibly the authors used statistical adjustment of their results to take some of these differences into account. However, the adjustment process is only a partial solution: there may still be some left-over influence of the other differences between groups on the results.  This problem plagues these kinds of studies, making it very difficult to draw any sound conclusions from them: the results may be real but could be due to chance or other factors.

“Further research on whether there is any completely safe amount to drink during pregnancy remains necessary, but in the meantime pregnant women should follow the advice from the UK Chief Medical Officers to avoid alcohol during pregnancy.”

‘Maternal alcohol intake prior to and during pregnancy and risk of adverse birth outcomes: evidence from a British cohort’ by Camilla Nykjaer et al. published in the Journal of Epidemiology and Community Health on Monday 10 March