Contents - Previous - Next


Waterlow: What was the effect on lactation of starvation in the Dutch famine?

Prentice: I don't think we know. We have some data from Macey in the Wuppertal study and they were amazed how good the milk production and composition was.

Shetty: What about the lactational performance of adolescent mothers with low BMI?

Prentice: I don't think we have any data.

Strickland: Is the maintenance of such excellent lactational performance at the expense of some other physiological function?

Prentice: In the papers that I reviewed, we don't have an answer to that. I think that in our own studies the mother's current intake must be enough but we are just not measuring intake properly: the sums just don't add up. In Susan Roberts' paper on baboons [Roberts SB, Cole TJ & Coward WA (1985) Lactational performance in relation to energy intake in the baboon. Am. J. Clin. Nutr. 41, 1270-1276] the baboons were restricted to 60% or 80% of requirements. At 80% there was no effect on milk production or quality, but at 60% of requirements milk output was reduced by 20%, and at both reduced intakes, the baboons were much less active.

Allen: What about the issue of fat loss or weight loss or gain. To what extent is body fat loss useful for fat and energy for milk production? Prentice: Weight is generally stable during lactation. The weight gain in pregnancy is not necessarily used for lactation, but serves as a buffer. In the Keneba region in the rainy season lactating women lose weight faster but in the harvest season, lactating women gain weight faster than the rest of the population.

Kusin: I agree that it is a marvel how undernourished women of reproductive age survive the reproductive years. But we seldom have data linking infant and maternal outcomes. Only below a BMI of 17 do you get a quantifiable effect. In poor mothers we see little breast engorgements and with 6h intervals between suckling we were never able to get samples of more than 40 ml. In well-nourished women you easily get 150 ml. So the infants suckle so much of the time because they get so little. We seem to be at the limit of production with substrate limitation. In Kenya data only 3% were undernourished and we did not really see the effect.

Prentice: All that data I reviewed was for 2 and 6 months' post partum or at peak lactation time. Duration is confounded by social class so the poorest and the thinnest mothers continue the longest with good milk production because they cannot afford to bring in weaning foods.

Waterlow: I think there is still a paradox because you say the milk production is protected and that it could be expanded. On the other hand, we know that as the infant gets to 4 months or 6 months, milk is not a satisfactory complete food and growth rates begin to fall off. I do not believe that milk production can continue to increase to cope with the needs of the healthily growing child.

Prentice: I invoke evolutionary theory to explain that. We have evolved in the best way for a mother to pass on her genes successfully. One way is by maintaining herself ready for the next reproduction. There has to be a balance between the child's needs and her own. The introduction of weaning foods is a powerful determinant of milk volumes. Women who don't introduce weaning foods early sustain a high milk production for longer.

Allen: I have two things to add that may explain growth failure. One is that just because the volume and fat content of milk is not affected by BMI does not mean that the composition of the micronutrients in milk is adequate. In Kenya 100% and in Mexico 70% of milk samples were deficient in vitamin B12. In Egypt B6 levels were extremely low.

Waterlow: I have never seen a B12-deficient infant.

Prentice: I agree with Lindsay, the water soluble vitamins reflect the maternal blood levels very closely. You do see differences in milk quality but it is not a BMI issue.

Allen: The store of micronutrients that an infant is born with is another factor in the growth rate phenomena.

Shetty: Is there any evidence of a compromise to the mother's health while she is lactating on a poor diet?

Prentice: In Keneba, or India, half the women's lives are spent lactating. When we supplied a lactation supplement in Keneba we did look for changes in maternal well-being. There was a reduction of reported illness when on supplements. Gastrointestinal complaints were most affected.

James: I wonder if there is a substrate requirement of a refined character relating to the flow of lymphocytes to the child. Could poorly nourished women be less able to sustain this thymocyte/Iymphocyte replication which seems to be sensitive to nutrient intake? What proportion of lymphocyte turnover is channelling into breast milk?

Prentice: Lymphocytes do pass over. The numbers are very small, e.g. ng/day, so they are insignificant in energy terms. The levels passing over are higher in Gambia than in Cambridge because they have a higher white cell count, because of the infectious load.

James: What about lactoferrin which may confer gut resistance to pathogens? Does this vary with communities in a nutrition related way?

Prentice: Ann Prentice found identical levels in Cambridge and Gambia, but IgA and lysozyme levels are higher in Keneba than in Cambridge.

James: But the Gambian women have a BMI of about 22, so there is a buffering system of N and amino acids which may be different from Prakash's really thin Indian women.

Prentice: Gambian women range from 19-21.5 BMI. The N content of milk does not seem to be affected however low the BMI, and the immune factors tend to be higher in a worse-off population to protect the mother and baby from infection.

Waterlow: In Ethiopia there was no decrease in IgA levels and the lymphocyte turnover is so vast in blood that I don't think the levels in milk could be limited.

Allen: It is clear in our three countries that there is an increase in energy intake in lactation of 400 kcal a day which is physiologically driven. BMI had an influence on that effect and perhaps prolactin drives appetite. So, in CED the mother may be driven to eat at the cost of other members of the family. Andrew Prentice says that none of his women were energy deficient, but Jane Kusin says that they were in her study. Did Jane see a difference in energy intake across the BMI groups and why didn't they respond to supplementation?

Kusin: We did not find a difference in intake between low and high BMI during the study. Food availability does not differ over the year because of marginal seasonality, but I maintain that they are CED, because if you combine the indicators of fetal growth and post-natal growth and weights of the mothers, they are not physiologically optimum.

Strickland: Are there any data on women who fail in lactation? Is there a level of BMI at which reproduction ceases altogether?

Prentice: Because we reproduce so slowly and there is so much investment from the time of conception to delivery, there is a high incentive to maintain that investment with successful lactation. In Gambia lactation failure is considered non-existent, but some are more successful than others and there is a genetic component to it. A weak baby will not drive a good lactation in the mother. Among Ethiopian refugees there must be some who are too thin to lactate successfully but we did not find any in Gambia.

I think we need to think more clearly about low birth weight. How meaningful is the association between BMI and birth weight? In nine different groups, including well-nourished European countries, the Philippines, Thailand and Gambia, there is a large difference (600-700g) in the birth weights of those populations, but if you express those relative to maternal weight in kg to the power of 0.75, then birth weight is the same for all of them. So, these women are producing babies appropriate to the mother's size. If we then follow through to neonatal mortality and outcome, there is no suggestion of any difference in smaller and larger babies as long as they are of the size appropriate to the mother's size. There is only a risk if there is some pathological reason for the child being small, such as small-for-dates or prematurity. To take this a little further, I have now added in Lindsay's data, the Indian data and Jane's data. The small babies that Jane was concerned about form part of this continuum - in fact are relatively big. The ones I worry about are the Indian babies from high BMI mothers; these are far too small for their mother's BMI.

Naidu: If you think small mothers should have small babies, are you then saying that it is wrong to give supplements to small mothers to improve the weight and health of the baby?

Allen: We found in our children with low birth weight, from mothers with low BMI, that the maternal BMI at the start of pregnancy was the strongest determinant of infant weight at six months. In communities like this where growth faltering is so severe it is important to get birth weight up as high as possible. The smallest babies at birth also have worse cognitive performances at 6 months. The intermediary effect of weight gain in pregnancy in developing countries being only half that in developed countries must also be considered because it too affects birth weight.

Shetty: I am stunned by Andrew Prentice's claims and don't know how to react. I would like to draw his attention to a large review which was published in the Bulletin of WHO by Kramer [Kramer MS (1987) Determinants of low birth weight: methodological assessment and meta-analysis. Bull. WHO 65, 663-787], based on a meta-analysis of anthropometric determinants of birth weight. The prime determinant is maternal body weight, so I am not surprised by his analysis. But we are here trying to decide what is an acceptable BMI and what is not, and what is an acceptable birth weight. There are several value judgments and moral issues involved.

Waterlow: The only person who has said anything definite about birth weight is Jane Kusin who says 2.4 is a cut-off point above which there was no increase in risk. That is the sort of information we need.

Weisell: I want to ask about the heavy Indian babies coming from short but heavy mothers. Is Andrew concerned about the babies or the mothers?

Prentice: I only wanted to get us thinking about this in a more sophisticated way. I am not saying that these small babies are not at risk, but we need the clinical measures of outcome and the sort of 2.4 kg cut-off evidence of increased mortality which Jane Kusin has. I am not saying that there is a problem with the fatter babies of short fatter mothers. There may be a problem with the mothers but it is a separate issue, not related to BMI. In the Gambia during the wet but not the dry season birth weights decrease by 250-300 g and supplementation is effective. A careful analysis shows that when the birth weights are less than you expect them to be then there is a positive effect of supplementation. Obstetricians would be very worried if we started to produce babies of 3.5 kg in Jane Kusin's mothers.

Contents - Previous - Next