Chapter 7 : Lipids in early development

Contents - Previous - Next

Maternal nutrition
Preconceptional nutrition
Infant nutrition

Maternal nutrition

During pregnancy, the acquisition of long-chain polyunsaturated fatty acids for placental and fetal development is required. In addition, there is firm evidence that maternal nutrition and health in the periconceptional period is of crucial significance. Events prior to conception influence the long term physiological process of fat storage and the nature of the fat store. This is the fat which is available during the critical period of cell commitment and division during embryonic and placental development in the first trimester. Substantial amounts of fat are also stored from conception to provide for fetal growth in the third trimester and the initial needs of lactation.

The evidence from animal experiments suggests that pre and post-natal nutrition have pronounced effects on brain lipid composition and learning (Gall) and Socini, 1983). Specific deficits of n-3 fatty acids influence neural integrity (Budowski, Leighfield and Crawford, 1987) and selectively affect learning and visual abilities (Wheeler, Benolken and Anderson, 1975; Lamptey and Walker, 1976; Bourre et al., 1989; Yamamoto et al., 1987). Studies with non-human primates confirm that n-3 deficiency depresses the development of retinal function and visual acuity (Neuringer, Anderson and Connor, 1988; Connor, Lin and Neuringer, 1990).

Recent studies replicated these findings in human infants indicating the essentiality of n-3 fatty acids and the need for the inclusion of docosahexaenoic acid (DHA) in food for infants (Birch et al., 1992, 1993 a, b; Carlson et al., 1993 a, b, Uauy et al., 1990; Uauy, Birch and Birch, 1992). Although there are not similar studies for arachidonic acid, the experimental evidence indicates that low arachidonic acid levels are associated with low pre-natal (Crawford et al., 1989; Leaf et al., 1992) and post-natal growth in preterm infants (Carlson et al., 1992). Conditionally, arachidonic acid should be considered as an essential nutrient during early development because: it is present in human milk (Sas et al., 1986; Koletzko, Thiel and Abiodun, 1992) along with DHA; the interplay between the n-3 and n-6 families; the specifie functions of araehidonic acid in neural and vascular function; and the role of its eicosanoids in cell regulation.

There have not been the same kind of controlled, randomized trials for term infants as those performed for preterm ones. However, the preliminary studies comparing term infants who were breastfed with those fed low a -linolenic acid formula suggest that the term infant's brain may also be responsive to external nutritional influences (Birch et al., 1993 b; Gibson et al., 1993). Data on brain composition provide a comparison of formula-fed versus breastmilk-fed babies and give further evidence that dietary fatty acids influence the developing brain of term infants (Farquharson et al., 1992).

Preconceptional nutrition

Embryology and clinical studies provide evidence that the mother's nutritional status around the time of conception has greater significance in terms of birth weight (Caan et al., 1978; Villar and Riveria, 1988; Wynn et al., 1991), prevention of neural tube defects (Wald et al., 1991) and non-genetic congenital disorders (Wynn and Wynn, 1981) than nutritional status in the latter part of pregnancy. Poor maternal nutrition or metabolic status at this early stage represents a significant risk of compromising embryonic development, cell commitment and the rate of DNA replication in a manner which cannot be compensated later.

During gestation, the placenta selects arachidonic acid and docosahexaenoic acid (DHA) at the expense of linoleic acid, a -linolenic acid and eicosapentaenoic acid (EPA) resulting in substantially higher proportions of arachidonic acid and DHA in the fetal circulation at mid-term (Crawford et al., 1976) and term (Olegard and Svennerholm, 1970). This is denied to babies born prematurely. Low birth weight and prematurity are associated with a high risk of neurodevelopmental disorders and disabilities (Wynn and Wynn, 1981; Dunn, 1986; Hack et al., 1991; Scottish Low Birth Weight Study Group, 1992 a, b). The incidence of neurodevelopmental disorder among premature and low birth weight babies in the UK and Sweden is reported as having increased threefold since 1967 (Pharoah et al., 1990; Hagberg, Hagberg and Zetterstrom, 1989) adding urgency to the need for a better understanding of the nutritional needs of these infants.

Low birth weight. In a number of developing countries, low birth weight is a particular problem associated with a high incidence of perinatal and matemal mortality and morbidity. It is also a problem within urban areas of developed nations, especially among lower socio-economic groups. Worldwide, WHO estimates that 17.4 percent of babies are born at low birth weights. This has important long term implications for the children's health and abilities (FAO/WHO, 1992).

Teenage pregnancies pose a special problem since the nutritional intake of the mother needs to support her own continued somatic growth as well as that of the fetus. The vulnerability of teenage mothers and their offspring is borne out by the high frequency of perinatal mortality, low birth weight and maternal morbidity and mortality. However, when pregnant teenagers eat adequately throughout their pregnancies, delivery outcomes may be successful.

While a relationship exists between maternal nutrition and birth weights (Caan et al., 1978), the causes of low birth weight are multifactorial, including: low caloric intake, low weight gain during pregnancy, low prepregnancy weight, short stature and disease (for example, malaria) (FAO/WHO, 1992). In developed countries, smoking presents an additional factor. Several nutrients, independent of smoking, are related to birth weight up to 3.0kg (Doyle et al., 1990). Retrospective evidence suggests that fetal nutrition is a determinant of risk of non-insulin-dependent diabetes and vascular disease in later life (Barker et al., 1993). This concept is supported by evidence of deficits of arachidonic acid, DHA and vitamin A in the circulation of low-birth weight newborns (Ongari et al., 1984; Crawford et al., 1989, 1990; Carlson et al., 1992; Leaf et al., 1992) and by evidence of vascular pathology in the placenta of low birth weight babies (Althabe, Laberre and Telenta, 1985; Winick, 1983).

Micronutrients andfat store. In many developing countries, vitamin A deficiency is still prevalent. An adequate intake of fat and fat soluble vitamins under these circumstances is important and there should be an effort to raise the percentage of dietary energy from fat to at least 20 percent for women of reproductive age. This strategy should also help to ensure an adequate supply of essential fatty acids and fat soluble vitamins.

Body fat plays a special role insofar as women on low fat and low-calorie diets and highly trained athletes may not conceive, or the embryonic and fetal development of their offspring may be jeopardized (Frisch, 1977). The mother's fat store at the time of conception will be relevant to maternal hormonal responses and to the nourishment of the embryo. Similarly, it will provide the basis for subsequent fat storage and utilization during the pregnancy (FAO/WHO, 1978). To ensure adequate nutritional preparation, women should not be on slimming, low calorie or low fat diets in the three months prior to conception. Calorie and nutrient intakes should be adequate to meet the general recommendations for the first trimester of pregnancy whilst essential fatty acid intakes should be similar to those mentioned below for the pregnancy itself to ensure an adequate quality of the fat store. The need for adequate nutritional status in terms of calories, vitamins, minerals and trace elements before conception should receive attention.

Pregnancy. An additional requirement for dietary fat is present throughout the nine months of pregnancy to provide for the fat storage in the early trimester and the growth in other compartments during the later trimesters. In the first trimester, the development of the embryo imposes a negligible quantitative demand for additional essential fatty acids, however, normal maternal fat deposition and uterine growth as well as the preparative development of the mammary gland do make significant demands. In the second and particularly the third trimester, the expansion of blood volume, placental and fetal growth add to the demand. Based on accretion data for these compartments, an average total of 600 g of essential fatty acids (approximately 2.2 g/day) will be acquired throughout a normal pregnancy of a well-nourished woman. Increased energy utilization may modify this need. This is consistent with the recommendation on increased energy requirements (WHO, 1985a) and will maintain the linoleic to a -linolenic balance with a ratio of 1:5 to 1:10.

This recommendation presumes adequate conversion of both parent essential fatty acids to their long-chain polyunsaturated fatty acid derivatives. However, recent studies suggest that a relative deficiency of long-chain polyunsaturated n-3 fatty acids develops during pregnancy (Holman, Johnson and Ogburn, 1991). Correlations between arachidonic acid and birth weight and between DHA and gestational age (Leaf et al., 1992) are consistent for either indicator of maturation and with DHA intakes. There is evidence that high fish intakes are associated with longer gestation, higher birth weights and a reduced incidence of premature births. An intervention study with fish oils indicates that long-chain polyunsaturated n-3 fatty acids play a role (Olsen et al., 1992). If these data are confirmed, this could indicate the need to consider DHA preformed to help prevent premature birth and pregnancy-related hypertension (WHO, 1985a).

Lactation. During lactation, there are increased nutrient and energy requirements. The additional energy requirement for lactation, assuming there has been an appropriate energy store laid down during pregnancy, is 500 kcals/day (WHO, 1985a). If the stores are limited, the energy need can be as high as 800 kcals/day. Fat output is the most variable component of milk and this depends, qualitatively and quantitatively, on maternal nutrition and prolactin secretion. The diet of a wellnourished mother provides the unsaturated fatty acids within a range that assures the essential fatty acid supply to the infant. The overall energy may be supplied by any source but the essential fatty acid component is fully dependent on fat in the diet and maternal stores. If a woman was well-nourished throughout pregnancy, her fat stores can provide approximately one-third of the energy and essential fatty acids needed for the first 3 months of lactation. Based on the composition of milk from omnivorous women, the mother's diet during lactation should provide an additional 3 to 4 g/day of essential fatty acids in the first trimester of lactation which should rise to 5 g/day as the fat stores become depleted (Koletzko, Thiel and Abiodun, 1992) (Table 7.1). It should be possible to do this by increasing normal food intake.

Infant nutrition

The premature infant. During the last decade, evidence on the accretion of parent essential fatty acids and derived long-chain polyunsaturated fatty acids for fetal and neonatal growth has become available (Martinez, 1988). Modern analytical techniques have provided substantial data on the fatty acid composition of human milk (Ibid. ). Babies born prematurely are denied the intra-uterine supply of arachidonic acid and DHA. In addition, they are born with little or no fat reserves making them fully dependent on the diet. Despite the provision of ample linoleic acid and rising linoleic acid levels in the plasma, it is evident that arachidonic acid and other polyunsaturated fatty acids (of at least 20 carbon atoms in length) drop rapidly (Leaf et al., 1992). Present assessment of essential fatty acid status is dependent on the measurement of plasma and red cell composition. While there are limited data on tissue pools, the primate data suggest that they do correlate (Neuringer, Anderson and Connor, 1988). Retinal and visual cortical function show the effects of tissue essential fatty acid status.

TABLE 7. 1
Average fatty acid values of mature human milk in Europe and Africa

  Medians and Ranges
Europe (14 studies) Africa (10 studies)
Fatty acid totals (% wt/wt)
Saturated 45.2 (39.0-51.3) 53.5 (35.5-62.3)
Monounsaturated 38.8 (34.2-44.9 28.2 (22.8-49.0)
n-6 + n-3 PUFA 13.6 (8.5-19.6) 16.6 (6.3-24.7)
n-6 PUFA (% wt/wt)
C18:2n-6 11.0 (6.9-16.4) 12.0 (5.7-17.2)
C20:2n-6 0.3 (0.2-O.S) 0.3 (0.34.8)
C20:3n-6 0.3 (0.2-0.7) 0.4 (0.2-0.5)
C20:4n-6 0.5 (0.2-1.2) 0.6 (0.3-1.0)
C22:4n-6 0.1 (0.1-0.2) 0.1 (0.0-0.1)
C22:5n-6 0.1 (0.0-0.2) 0.1 (0.1-0.3)
Total n-6 LCP 1.2 (0.4-2.2) 1.5 (0.9-2.0)
n-3 PUFA (%wt/wt)
C18:3n-3 0.9 (0.7-1.3) 0.8 (0.1-1.44)
C20:5n-3 0.2 (0.0-0.6) 0.1 (0.1-0.5)
C22:5n-3 0.2 (0.1-0.5) 0.2 (0.1-0.4)
C22:6n-3 0.3 (0.14.6) 0.3 (0.1-0.9)
Total n-3 LCP 0.6 (0.3-1.8) 0.6 (0.3-2.9)

Source: Adapted from Koletzko, Thiel and Abiodun, 1992.

Several studies on biochemical indices of essential fatty acid status documented the effect of adding long-chain polyunsaturated fatty acids to the diet (Koletzko et al., 1989; Carlson et al., 1992; Uauy, Birch and Birch, 1992; Makrides et al., 1993). If both arachidonic acid and DHA are added, the post-natal drop in blood plasma level can be ameliorated. If marine oil supplementation is used (EPAIDHA = 2: 1) this compromises the level of arachidonic acid which was found to adversely affect growth in one study (Carlson et al., 1992). There is convincing evidence that supplementation with DHA from marine oil improves the development of the rod photoreceptor and visual acuity, a measure of the receptor response to light, as well as the cortical, which is associated with cognitive ability to integrate information (Uauy et al., 1990; Uauy, Birch and Birch, 1992; Birch et al. 1992; Carlson 1993a). The initial concern (Carlson et al., 1992) that poor growth, associated with marine oil, was linked to depressed arachidonic acid status has been removed by using a low EPA marine oil with an EPA/DHA ratio of 1:10. This did not compromise weight gain and resulted in higher Bayley mental development scores at 12 months (Carlson et al., 1993 b). Eight year follow-up studies which compared premature babies who were fed human milk by tube with those fed formula showed an IQ which was eight points lower in those children who had been fed formula (Lucas et al., 1992).

There are limited data on the accretion of essential fatty acids of normal fetuses and healthy infants fed human milk and no quantitative human data on the metabolic conversions. Although there are variations in maternal diets, milk composition data obtained from omnivorous women were the basis for the consultation's recommendation. As a guide, formula for preterm babies should provide a mean of 700 mg linoleic acid, 50 mg a -linolenic acid, 60 mg of arachidonic acid and its associated long chain n-6 fatty acids, and 40 mg of DHA per kg body weight. That is, 5.6 percent of the energy as parent essential fatty acids and 0.8 percent as long-chain polyunsaturated fatty acids. Because of the known interference of linoleic acid with the metabolism of the long-chain polyunsaturated fatty acids, linoleic acid should not exceed 10 percent of the total energy. Until more data are available, particularly dose response data, the other fatty acids as mentioned above should remain within approximately 30 percent of the target recommendation. The remaining fatty acids should be selected on the basis of digestibility, lack of interference with essential fatty acid metabolism and other measures of function. In principle, a mixture of oleic and saturated fatty acids with a predominance of oleic appears to meet these conditions.

Term infants. Whenever possible, the preferred source of infant nutrition is human milk. In view of the evidence, education and health care programmes should actively promote breast feeding. In addition, because of the long term nutritional interactions with the mother even prior to conception and their profound implications for public health, similar programmes should be developed for women prior to conception, through pregnancy and lactation. Such programmes will improve maternal health, fetal development and neonatal nutrition.

When term infants who died from unexplained causes were compared, those infants who were fed milk formula without arachidonic acid and DHA had reduced levels of brain cortical DHA and higher n-6 docosapentaenoic acid (an index of DHA deficiency) than those fed breast milk (Farquharson et al., 1992). As the data on preterm infants now demonstrate, there are advantages for both rod photoreceptor and neural function to supplementing with n-3 long-chain polyunsaturated fatty acids; it would seem proper to provide both arachidonic acid and DHA preformed in term infant formula milks in similar proportions to breast milk from well-nourished, omnivorous mothers. For term infants, the provision, per kilogramme of body weight should amount to 600 mg of linoleic acid, 50 mg of a-linolenic acid, 40 mg of arachidonic acid and its associated n-6 fatty acids and 20 mg of docosahexanoic acid. Although there is not yet evidence from controlled, randomized trials in term infants, this is suggested to provide for the greatest possible release of the full genetic potential for neural and visual development.

As it has evolved, human milk provides long-chain polyunsaturated fatty acids preformed. In view of the evidence on the higher efficiency of long-chain polyunsaturated fatty acids for neural development (Sinclair, 1975; FAO/WHO, 1978; Leyton, Drury, and Crawford, 1987) and the data on premature infants discussed above, the long-chain polyunsaturated fatty acids should be included in infant formula. Providing parent essential fatty acid at a ratio which is claimed to optimize conversion to long-chain polyunsaturated fatty acids should be proven valid for physical, vascular and mental development and should provide for biochemical and functional normalcy. The burden of proof should be placed on those proposing that artificial formula should, in principle, be dissimilar to breast milk (Drury and Crawford, 1990).

These recommendations for term and preterm infants are broadly in agreement with those of the British Nutrition Task Force on Unsaturated Fatty Acids (BNF, 1992).

Weaning. Human milk provides 50 - 60 percent of its energy as lipid in which about 5 percent of the energy is essential fatty acid with 1 percent as long-chain polyunsaturated fatty acids. Six-month follow-up studies of lactation in a randomised trial showed that in wellnourished mothers, milk fat contribution increased from 40 - 50 g/litre at three weeks to 60 - 70 g/litre at 4-6 months (WHO, 1985b; Sas et al., 1986). In developing countries with lower energy intakes, the rise in milk fat content was less. It would, therefore, seem especially inappropriate to wean babies on to low fat diets.

The European Society of Paediatric Gastroenterology and Nutrition (1991) recommended that 40-55 percent of the dietary energy be provided as fat (4.4-6.0 g/100 kcal) for follow-up formula. During weaning, the fat component should provide 30-40 percent of the dietary energy and similar levels of essential fatty acids as are found in breast milk from appropriate foods until at least two years of age.

In practice, this means that complementary food used during the weaning period should include adequate amounts of fats and oils as the breast milk component of the diet declines. Commonly, infants in many developing countries are weaned on to cereal or tuber-based diets with a low energy density. In these countries, there is a need to reinforce educational messages about the use of vegetable oils or oil containing foods in the diet of weanling infants and young children.


It is acknowledged that human milk is the best and only time-proven source for fat and long-chain polyunsaturated fatty acids in neonate diets. Sources of long-chain polyunsaturated fatty acids that have been tested include egg phosphoglycerides, fish oils and fish oil fractions. Novel sources of fat which have been developed include phosphoglycerides from animal tissues, algal and microbial lipids. The safety and efficacy of novel sources should be adequately tested before they are added to infant formula.

Contents - Previous - Next