by Rebecca Norton, Consultant, ACC/SCN.
In developing countries, many women are short and underweight and the number of low birthweight (LBW) babies is particularly high (more than 30% in South Asia, 10-20% in other regions, [l(a)]). LBW infants have less chance of I survival; when they do survive, they are more prone to disease, growth retardation and impaired mental development. A good start in life is important and maternal nutritional status during pregnancy has repeatedly been demonstrated to be associated with pregnancy outcomes for the infant (1).
Low birthweight (LBW) is defined by the World Health Organisation as birthweight less than 2500g. It is governed by two major processes: a short gestational period, i.e. the infant is born too soon and is qualified as premature (bwt<2500g and gestational age <37 weeks), or retarded intrauterine growth, i.e. the infant is small for gestational age (bwt<2500g and gestation age >37 weeks). In developing countries intrauterine growth retardation (IUGR) accounts for the majority of low birth weights whereas in developed countries most LBW babies are premature as opposed to growth retarded (see table 1 (2)).
Two different subtypes of IUGR can at least be distinguished: wasted or thin IUGR infants with a relatively normal length and head circumference, and stunted IUGR infants with proportional reductions in weight, length and head circumference. Such a distinction appears to be prognostically important as wasted IUGR infants appear to exhibit greater postnatal catch-up growth and less severe cognitive defects than stunted infants (1). Wasted infants however exhibit higher rates of neonatal morbidity and neonatal and perinatal mortality (3).
Preterm delivery is also a heterogeneous syndrome. Three subclasses can be distinguished: a) spontaneous pre-term delivery; b) medically induced pre-term delivery; and c) premature rupture of the membranes resulting in pre-term delivery. Only spontaneous preterm deliveries could be influenced by general interventions. Such interventions are however unlikely to have a major impact on the total preterm delivery rate among different populations as the contribution of the various subgroups to the total rate of preterm delivery appears to vary across populations (2).
Relatively few reports exist on the relationship between maternal nutrition and preterm birth. Although some studies have reported an association between the two it would seem that maternal nutrition during pregnancy is not an important determinant of prematurity (4). Poor maternal nutrition has however been shown to be one of the major causal determinants of IUGR in both developed and developing countries (2) (see figures 1 and 2).
The most sensitive measure of acute nutritional stresses during pregnancy is indeed maternal weight gain. There is strong epidemiological evidence of an association between maternal weight gain during pregnancy and LBW/IUGR, especially in undernourished women i.e. those who begin pregnancy in a nutritionally disadvantaged state. Women are at the greatest risk of having a LBW infant if low prepregnancy weight and low weight gain during pregnancy are combined.
Table 1. Incidence of Low Birth Weight (<2500 gm) in Developing and Developed Areas.
Populations from developing countries
Populations from developed countries
Number of populations studied
Total low birth weight (%, average)
Pre-term/low birth weight (%, average)
IUGR/low birth weight (%, average)
(Source: Table V in: Villar, J. et al (1994). See reference 2.)
Figure 1. Relative Importance of Established Factors with Direct Causal Impacts on Intrauterine Growth Retardation (IUGR) in Rural Developing Countries.
(Source: Figure 1, p.2 in: Kramer (1987). See reference 1)
Figure 2. Relative Importance of Established Factors with Direct Causal Impacts on Intrauterine Growth Retardation (IUGR) in Explaining the Difference in IUGR Rates in Rural Developing and Developed Countries.
(Source: Figure 2, p.2 in: Kramer (1987). See reference 1)
Many efforts have been made in order to identify a potential modifiable factor for LBW. Maternal nutrition is modifiable in the short term; if it is one of the major environmental causes of IUGR in the developing world, a substantial fraction of LBW could possibly be prevented. In turn this might reduce the prevalence of mortality, morbidity, physical and mental development, factors associated with LBW1.
1 It must however be born in mind that no matter how convincing the evidence that a given factor is causally related to LBW (IUGR or gestational duration), there is no guarantee that its elimination or reduction will lead to lower infant mortality or child morbidity. Other environmental factors may indeed directly affect these outcomes independently of birthweight.
Maternal nutrition has therefore been the focus of considerable research over the last few years. Due to the strong epidemiological evidence of a relation between maternal nutritional status and birthweight, a number of intervention studies of nutritional supplementation during pregnancy have been carried out both in developing and developed countries. Interest has however been largely restricted to birthweight as an outcome, especially in populations with high prevalences of maternal undernutrition, low birthweight and perinatal mortality.
During pregnancy, the foetus is solely dependent on maternal intake and nutritional stores, mostly fat, for its energy. Poor maternal nutrition during pregnancy in turn implies a risk of poor nutritional availability to the foetus. The best methodological approach for assessing the effect of this factor on birthweight and more specifically on IUGR or prematurity is thus supplementation. Will an increase in food intake increase birthweight? Will the prevalence of LBW and IUGR/premature infants thus be decreased?
Provision of food does not necessarily lead to its consumption. Even if it is consumed, it may replace some of the usual diet. Supplementation trials must take this into account in order to evaluate the actual extra amount ingested. Many such trials have been carried out. Until recently most of the evidence seemed to indicate that maternal caloric intake during pregnancy had no effect on prematurity; however supplementation had a positive effect on birthweight and IUGR. The effect was greater the more malnourished the mother was before pregnancy. Nutritional supplementation during pregnancy was also shown to be associated with a reduction in the incidence of LBW in developing and developed populations.
Surprisingly, and regardless of methodological and practical differences, the effect of nutritional supplementation during pregnancy on birthweight has generally been modest, with an average increase of about 100g (5).
One study showing a substantial effect was in the Gambia where daily supplements of groundnut based biscuits and vitamin fortified tea were distributed to pregnant women. The mean net increase of energy intake was 431 kcal per day. The resulting significant increase in birthweight was on average 120g and the overall prevalence of LBW babies decreased significantly from 20 to 6%. There were however marked seasonal differences. Supplementation during the wet season (hungry season) led to a significant increase in birthweight of about 200g and a decrease in the proportion of LBW from 23.7% to 7.5%; in the dry season supplementation had no effect (average increase of 2g only). (6)
From various data sets it was also established that for undernourished women an additional 100 kcal ingested per day throughout pregnancy would increase the birthweight by about 100g (provided kcal taken at any time during pregnancy have the same effect). In non malnourished mothers the effect was three times less important: on average 35g increase in birthweight for each additional 100 kcal ingested daily. Similarly a significant reduced risk of IUGR in women who received the supplements was shown. Collated data sets suggested that if 100 kcal per day were supplemented throughout pregnancy, the risk of IUGR would be halved in mothers undernourished prior to pregnancy, but only reduced by 1/5 in well-nourished mothers (1).
Most of the dietary intervention studies addressed only birthweight as the outcome variable. One study in East Java however showed that maternal nutrition during pregnancy influenced growth of the offspring beyond the intrauterine period. Supplements were distributed during the last trimester of pregnancy. Women received either a high energy (465 kcal) or a low energy (52 kcal) supplement. Children were subsequently followed up from birth until 5 years. Birthweight and length at one week were not significantly different between the two groups of women although there was a modest improvement in the children whose mothers had been in the high supplement group. These children were significantly heavier up to the age of 24 months and taller throughout the first 5 years. Stunting was less prevalent among children whose mothers had received the high energy supplement. Mothers may have had an improved breastmilk output and their better-nourished children were less likely to become sick. (7)
One study in Guatemala showed that continued supplementation during two pregnancies and the lactation period increased the average birthweight 3 times more (300g in the high supplement group) than the usual increment observed (100g). Thus improving the womens nutritional status prior to and during two consecutive pregnancies was more effective than improving nutritional status during one pregnancy only. (5)
As stated by Villar (5) women without overt malnutrition or in positive energy balance (Gambia dry season) obtain a limited benefit from nutritional supplementation during one pregnancy. Chronically malnourished mothers also supplemented during one pregnancy experience only a modest impact on birthweight of about 100g. When women are however in negative energy balance, food supplementation produces a significant increase in birthweight as in the Gambia (230g increase). The expectation of a dramatic recovery from generations of poverty and food scarcity in a short time is an overly optimistic proposition. It may be that extra, yet prolonged, intake during pregnancies and lactation, rather than large amounts of supplementation during short periods of a given gestation, produce the foetal growth effect.
A recent meta-analysis of controlled clinical trials on the effect of supplementation during pregnancy on the outcome of pregnancy confirmed that trials of nutritional advice to increase energy and protein intakes and of balanced energy and protein supplementation, have demonstrated only a modest increase in maternal weight gain and fetal growth, even in undernourished women, and no long term benefits to the child in terms of growth of neurocognitive development (8). That is, the clinical experimental evidence reviewed showed that modest increases in fetal growth in the absence of effects on gestational duration do not appear to confer long lasting benefits on infant and child survival, health and performance.
Only trials using controls or random or quasirandom methods of treatment allocation were included in the overview (the Guatemalan and Gambian trials were excluded). The author concluded that unless future trials of energy and protein supplementation demonstrate clear reductions in risk for preterm birth, stillbirth, or neonatal death, or improvements in maternal health, clinicians and politicians should avoid high expectations from this type of nutritional intervention and should perhaps shift their focus towards potentially more fruitful avenues for improving maternal and child health. (8).
The contrasts between the findings of this overview and the results of observational studies suggest that the latter may have overestimated the effects of supplementation on pregnancy outcome. The robust findings of a strong association between maternal weight gain and fetal growth and of an even stronger association in undernourished women may partly reflect a non-nutritional effect mediated by such factors as expanded maternal plasma volume and increased placental blood flow. (8)
Timing of Supplementation
Maternal weight gain does not differentiate between the weight of the mother, the foetus or the various other components such as fat stores, breast and uterine tissue, plasma volume, and the foetus. It provides only a general impression of foetal growth. Foetal weight however increases exponentially with high weight gains in the third trimester, while the overall rate of maternal weight gain is fairly constant after the first trimester.
An analysis of the components of weight gain during pregnancy revealed that 90% of the fat was deposited as maternal stores in the first two trimesters of pregnancy. Healthy women in urban Guatemala who delivered term infants were followed during pregnancy. Maternal nutritional status at the beginning of gestation and the rate of fat gain early in pregnancy were the two indicators most strongly associated with birthweight (10).
Such results lend support to the theory that fat deposited early in pregnancy acts as a reserve for the last trimesters caloric demands. As most nutritional interventions were implemented in the third trimester of pregnancy, this could explain the lack of a large effect on birthweight and IUGR, rather than the recently suggested lack of association (11). Supplementing in the last trimester (after 20 weeks) may indeed exclude from the benefits of stunting. Wasted infants, who have been shown to have diminished subcutaneous fat reserves (3), would on the other hand probably benefit from supplementation during the third trimester. Data on the prevalence of stunted versus wasted IUGR infants would therefore be most beneficial.
Energy Expenditure and Non-Nutritional Effects
Further studies are needed to attempt to distinguish between the effects of energy expenditure and non work-related factors such as posture, fatigue, and stress. Indeed maternal work could have an effect on pregnancy independently of its nutritional effect. Physical exertion or upright posture might diminish uterine blood and thus hinder the supply of nutrients and oxygen to the foetus. In Gambia for example the supplement effect in the wet season, when women are hard at work in the fields, was not mediated by a change in maternal nutritional status, i.e. there was no increase in weight gain. The timing and frequency of the supplementation, mediating an increased placental blood flow, may have been responsible for the increased birthweight in the wet season (6,8).
There is strong epidemiological evidence of an association between maternal nutritional status, both during and prior to pregnancy (prepregnancy weight and weight gain during pregnancy), and birthweight and intrauterine growth retardation.
Many trials of nutritional supplementation during pregnancy have been carried out, only to show a modest effect of supplementation on birthweight, even in undernourished women (about 100g), and, according to one review, no long term benefits to the child in terms of growth or neurocognitive development (8).
Whether the supplementation trials were carried out at the right time during pregnancy still remains doubtful. Future supplementation during pregnancy is probably best targeted at nutritionally disadvantaged populations during all three trimesters of pregnancy.
One of the major drawbacks of many of the studies carried out has been the failure to distinguish between IUGR and prematurity, and especially between stunted and wasted intrauterine growth retarded infants. Such distinctions would be most beneficial for future trials.
More research is also needed on the effect of maternal nutrition during pregnancy on preterm delivery, and during consecutive pregnancies and prior to pregnancy.
In the interim, maternal supplementation may be expected to have some benefit, but alone seems unlikely to make a major difference to low birth weight and child nutrition.
1. Kramer, M.S. (1987). Determinants of Low Birth Weight Methodological Assessment and Meta-Analysis. Bulletin of the World Health Organization, 65 (5), 663-737.
1a. ACC/SCN (1993). Second Report on the World Nutrition Situation. Volume I: Global and Regional Results. ACC/SCN, Geneva.
2. Villar, I, Gurtner de la Fuente, V., Ezcurra, E.J. & Campodonico, L. (1994) Pre-term delivery: Unmet need. In: Keirse, M. (ed). New Perspectives for the Effective Treatment of Pre-Term Labour. Wells Medical, UK. (in press)
3. Villar, J., de Onis, M., Kestler, E., Bolanos, F., Cerezo, R. & Berneded, H. (1990) The Differential Neonatal Morbidity of the Intrauterine Growth Retardation Syndrome. American Journal of Obstetrics and Gynecology, 163 (1), part 1, 151-582.
4. Kramer, M.S., McLean, F.H., Eason, E.L. & Usher, R.H. (1992). Maternal Nutrition and Spontaneous Preterm Birth. American Journal of Epidemiology, 136 (5), 574-582.
5. Villar, J. & Rivera, J. (1988). Nutritional Supplementation During Two Consecutive Pregnancies and the Interim Lactation Period: Effect on Birthweight. Pediatrics, 81(1), 51-57.
6. Prentice, A.M., Cole, T.J., Foord, F.A., Lamb, W.H. & Whitehead, R.G. (1987). Increases of Birthweight after Prenatal Dietary Supplementation of Rural African Women. American Journal of Clinical Nutrition, 46, 912-925.
7. Kusin, J.A., Kardjati, S., Houtkooper, J.M. & Renqvist, U.H. (1992). Energy Supplementation During Pregnancy and Postnatal Growth. Lancet, 340, 623-626.
8. Kramer, M.S. (1993). Effects of Energy and Protein Intake on Pregnancy Outcome: An Overview of the Research Evidence from Controlled Clinical Trials. American Journal of Clinical Nutrition, 58, 627-635.
9. Krasovec, K. & Anderson, M.A. (1991). Monitoring Pregnancy. Mothers and Children, 10(2), 4-6.
10. Villar, J., Cogswell, M., Kestler, E., Castillo, P., Menendez, R. & Repke, J.T. (1992). Effect of Fat and Fat-Free Mass Deposition During Pregnancy on Birth Weight. American Journal of Obstetrics and Gynecology, 167(5), 1344-1352.
11. Villar, J. (1994). Personal Communication, July 1994.