Intended for healthcare professionals

Letters

Maternal nutrition and birth weight

BMJ 2000; 320 doi: https://doi.org/10.1136/bmj.320.7239.941/a (Published 01 April 2000) Cite this as: BMJ 2000;320:941

These factors are related

  1. Wendy Doyle, senior research officer in nutrition (wendydoyle{at}nutrition.simplyonline.co.uk)a,
  2. Michael Crawford, professor of nutritional biochemistry,
  3. Kate Costeloe, professor of paediatrics
  1. Institute of Brain Chemistry and Human Nutrition, University of North London, London N7 8DB
  2. St Bartholomew's and the Royal London Hospital School of Medicine and Dentistry, Homerton Hospital, London E9 6SR
  3. Department of Zoology, University of Oxford, Oxford OX1 3PS
  4. Division of Public Health and Primary Health Care, Institute of Health Sciences, University of Oxford, Oxford OX3 7LF

    EDITOR—Mathews et al concluded that maternal nutrition had little effect on birth weight in industrialised countries.1 This was based on maternal nutrient intake during the second trimester of pregnancy in women in Portsmouth, an area where the incidence of low birth weight is relatively low (6.7%) compared with that in England and Wales (7.3%) or in east London and the City of London (8.9%). Furthermore, social classes IIIM, IV, and V were underrepresented when compared with those in the government's nationally representative dietary and nutritional survey of British adults.2

    The conclusions were not surprising for two reasons. Firstly, the mean daily intake from food of energy, fibre, and seven out of 10 micronutrients (reported elsewhere by the authors) was at or above the dietary reference values. If selenium is excluded because of missing information in the data sources used to calculate intake, nine out of 10 micronutrients were above the dietary reference values.

    We have shown a dose response relation between nutrient intake and birth weight up to 3270 g in a socially disadvantaged population in Hackney, east London.3 In the range where nutrient intakes were relevant to birth weight, these were below the dietary reference values.

    Secondly, dietary intakes were recorded between 9 and 20 weeks' gestation (mean 16.3 weeks). Our studies of nutritional intakes in east London have consistently shown that maternal nutrition before the end of the first trimester of pregnancy is related to birth weight. In a randomised controlled trial, although intakes of protein, six B vitamins, and four minerals recorded by 513 women during the first trimester, were highly correlated with birth weight, supplementation with a broad based nutritional supplement starting in the second trimester failed to show a reduction in the incidence of low birth weight.4 Programmes of nutritional intervention both pre-conception and during the first trimester with low income women in the United States have shown a reduction in the incidence of low birth weight.5

    To generalise the failure to find a relationship between low birth weight and maternal nutrition from this limited study is stretching the conclusion far beyond the evidence.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.

    Authors' reply

    1. Fiona Mathews, university research officer (fmathews{at}ermine.ox.ac.uk),
    2. Patricia Yudkin, university lecturer,
    3. Andrew Neil, university lecturer
    1. Institute of Brain Chemistry and Human Nutrition, University of North London, London N7 8DB
    2. St Bartholomew's and the Royal London Hospital School of Medicine and Dentistry, Homerton Hospital, London E9 6SR
    3. Department of Zoology, University of Oxford, Oxford OX1 3PS
    4. Division of Public Health and Primary Health Care, Institute of Health Sciences, University of Oxford, Oxford OX3 7LF

      EDITOR—Doyle et al argue that our failure to detect relations between birth weight and maternal diet was because our cohort was too high in social class and contained too few infants of low birth weight. However, the social class distribution of our subjects1 was similar to that of a nationally representative sample and to the cohort studied by Barker and colleagues.2 We based social class on the woman's occupation, or that of her partner if this gave a higher grouping. Using the partner's occupation alone would classify 62% of our cohort as manual/unemployed, compared with 66% of Doyle et al's cohort.3

      Doyle et al were more interested in low birth weights than in the whole range. We specifically included only term infants—excluding many babies of low birth weight— to permit comparisons with Barker's study. Nevertheless, at the time of recruitment our cohort was drawn from a population with an incidence of low birth weight (6.4%) similar to that of England and Wales (6.9%).4 Separate analyses using our entire cohort have found no associations between intake of any nutrient and poor outcomes of pregnancy, including preterm delivery and low weight for gestational age.5 Dietary data were available in our study for 51 mothers who delivered infants of low birth weight (2500 g), compared with 28 in Doyle et al's work.

      The suggestion that pregnancy outcome is influenced by maternal diet in the first but not in the second trimester contrasts with Barker and colleagues' work. Most women in Doyle et al's project were probably 9-12 weeks pregnant. In our study all women were between 9 and 20 weeks' gestation (mean 16 weeks). Even accounting for morning sickness, it seems improbable that the diets of our mothers would have been sufficiently different a few weeks earlier to have produced completely contrasting results.

      Finally, Doyle et al argue that the mothers in our study had diets insufficiently poor to permit us to see a relationship with birth weight. Unfortunately, the distribution of intakes in Doyle et al's study has not been published, so we cannot compare our study with theirs. More than 20% of our cohort had intakes below the reference nutrient intake for all nutrients examined except thiamin and vitamin B-12. We are now analysing the relations between pregnancy outcome and biochemical indices of nutritional status, paying particular attention to factors such as smoking and height, which complicate the interpretation of studies such as those presented by Doyle et al.

      References

      1. 1.
      2. 2.
      3. 3.
      4. 4.
      5. 5.
      View Abstract